i
UNIDIR/2004/18
Open Skies
A Cooperative Approach to
Military Transparency
and Confidence Building
Pál Dunay, Márton Krasznai, Hartwig Spitzer,
Rafael Wiemker and William Wynne
UNIDIR
United Nations Institute for Disarmament Research
Geneva, Switzerland
NOTE
The designations employed and the presentation of the
material in this publication do not imply the expression of any
opinion whatsoever on the part of the Secretariat of the
United Nations concerning the legal status of any country,
territory, city or area, or of its authorities, or concerning the
delimitation of its frontiers or boundaries.
*
* *
The views expressed in this paper are those of the authors
and do not necessarily reflect the views of the United Nations
Secretariat.
Copyright © United Nations, 2004
All rights reserved
UNIDIR/2004/18
UNITED NATIONS PUBLICATION
Sales No. GV.E.04.0.18
ISBN 92-9045-164-5
The United Nations Institute for Disarmament Research (UNIDIR)—an
intergovernmental organization within the United Nations—conducts
research on disarmament and security. UNIDIR is based in Geneva,
Switzerland, the centre for bilateral and multilateral disarmament and non-
proliferation negotiations, and home of the Conference on Disarmament.
The Institute explores current issues pertaining to the variety of existing and
future armaments, as well as global diplomacy and local entrenched
tensions and conflicts. Working with researchers, diplomats, Government
officials, NGOs and other institutions since 1980, UNIDIR acts as a bridge
between the research community and Governments. UNIDIR’s activities
are funded by contributions from Governments and donors foundations.
The Institute’s web site can be found at URL:
http://www.unidir.org
Cover page: designed by Diego Oyarzún-Reyes (UNCTAD)
iv
v
CONTENTS
Page
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
About the Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix
Acronyms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Chapter 1
Arms Control in the Post-Cold War World
Pál Dunay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.1 The Legacy of the Past . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Prospects for the Future . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.3 Open Skies as a Bridge in the Structure of Arms Control . 13
Chapter 2
The Open Skies Negotiations and the Open Skies Treaty
Pál Dunay and Hartwig Spitzer . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1 Early History (1955-1989). . . . . . . . . . . . . . . . . . . . . . . . 17
2.2 The Ottawa, Budapest and Vienna Rounds of
Negotiations in the Light of Circumstances. . . . . . . . . 24
2.3 Negotiation of the Treaty Substance and
Analysis of the Results . . . . . . . . . . . . . . . . . . . . . . . . 34
Chapter 3
The Open Skies Treaty Post-Signature
Pál Dunay and Hartwig Spitzer . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.1 The Ratification Process . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.2 Establishment and Activity of the OSCC:
Decisions and Guidance Documents to the Treaty. . . 65
vi
Chapter 4
Technical Preparations for Treaty Implementation
Hartwig Spitzer and Rafael Wiemker. . . . . . . . . . . . . . . . . . . . . 69
4.1 Establishment of Operational Units. . . . . . . . . . . . . . . . . 69
4.2 Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.3 Sensors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
4.4 The Ground Resolution Limits and Their Verification . . . 74
4.5 Trial Flights and Lessons . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.6 Trial Certifications of Open Skies Aircraft . . . . . . . . . . . . 87
Chapter 5
Post-Ratification Phase and Entry into Force
Márton Krasznai . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.1 Post-Ratification Events. . . . . . . . . . . . . . . . . . . . . . . . . . 95
5.2 Entry into Force of the Treaty, 1 January 2002 . . . . . . . . 96
5.3 Certification of Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . 96
5.4 Decisions on Rules and Procedures. . . . . . . . . . . . . . . . . 99
5.5 Taxi Option . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
5.6 Accession of Additional States Parties . . . . . . . . . . . . . . . 100
Chapter 6
Image Analysis and Data Assessment:
What Can Be Learnt from Open Skies Image Data?
Hartwig Spitzer and Rafael Wiemker. . . . . . . . . . . . . . . . . . . . . 103
6.1 Imaging Targets and Assessment Potential. . . . . . . . . . . . 103
6.2 Potential of Different Sensors . . . . . . . . . . . . . . . . . . . . . 107
6.3 The Digital Revolution: Computer-Aided Image Analysis. 114
Chapter 7
Prospects for Extensions of the Multilateral Open Skies Treaty
Hartwig Spitzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
7.1 Inclusion of Additional States Parties. . . . . . . . . . . . . . . . 127
7.2 Conflict Prevention and Crisis Management . . . . . . . . . . 128
7.3 Potential of the Open Skies Regime and
Sensor Suite for Environmental Monitoring
and Other Non-Military Applications. . . . . . . . . . . . . 130
vii
7.4 Additional Sensors and Higher Resolution. . . . . . . . . . . . 139
7.5 Data Access for Non-Military Organizations . . . . . . . . . . 141
7.6 Issues and Challenges for the Review Conference
of States Parties 2005 . . . . . . . . . . . . . . . . . . . . . . . . 142
7.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Chapter 8
Regional Applications of the Open Skies Approach
Márton Krasznai, Hartwig Spitzer and William Wynne . . . . . . . 149
8.1 Detour: The Hungarian-Romanian Bilateral
Agreement on Open Skies. . . . . . . . . . . . . . . . . . . . . 149
8.2 Regional Application: The Bosnia Experiment
and the Bosnia Experience . . . . . . . . . . . . . . . . . . . . 156
8.3 Regional Open Skies Agreement in the Balkans:
A Missed Opportunity . . . . . . . . . . . . . . . . . . . . . . . . 165
8.4 Open Skies Outside of Europe:
Precedents and Prospects . . . . . . . . . . . . . . . . . . . . . 166
Chapter 9
The Improvement of Satellite Capabilities
and its Implications for the Open Skies Regime
Hartwig Spitzer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
9.1 The Advent of Commercial 1-metre Satellites . . . . . . . . . 183
9.2 Comparing Apples and Oranges . . . . . . . . . . . . . . . . . . . 189
9.3 Unmanned Aerial Vehicles . . . . . . . . . . . . . . . . . . . . . . . 191
Chapter 10
Outlook: Nothing to Hide?
Perspectives for the Open Skies Treaty
Pál Dunay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Appendix A Addresses of Open Skies Units . . . . . . . . . . . . . . . . . 201
Appendix B Open Skies Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Appendix C Sensor Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Appendix D Open Skies Test Missions and Quota Flights
(as of December 2002). . . . . . . . . . . . . . . . . . . . . 223
viii
Appendix E Verifying the Ground Resolution Limit
(Photo & Video) . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Appendix F Brief Glossary of Open Skies Treaty Terms . . . . . . . . 251
Appendix G Open Skies Basic Elements . . . . . . . . . . . . . . . . . . . . 259
Appendix H The Hungarian-Romanian Open Skies Agreement . . . 265
Appendix I Treaty on Open Skies—Preamble and Contents . . . . 291
Appendix J Decisions of the Open Skies Consultative
Commission (Titles only). . . . . . . . . . . . . . . . . . . . 295
Appendix K Sensor Guidance Document—Table of Contents. . . . 301
Selected Bibliography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Recent UNIDIR Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
LIST OF FIGURES
Figure 2.1 Open Skies observation flight line . . . . . . . . . . . . . . . 50
Figure 4.1 Photogrammetric and Treaty definition of resolution . 75
Figure 4.2 German calibration target . . . . . . . . . . . . . . . . . . . . . 79
Figure 4.3 Densiometric strip and D log E curve. . . . . . . . . . . . . 80
Figure 4.4 Open Skies test flights (31 December 2001) . . . . . . . 82
Figure 4.5 Flight route of a German-Russian test flight
over Siberia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Figure 6.1 Image fusion (examples) . . . . . . . . . . . . . . . . . . . . . . 119
Figure C.1 SAROS specifications and mission geometry . . . . . . . 220
Figure E.1 Resolution of bar groups for various ground sample
distances (pixel sizes) . . . . . . . . . . . . . . . . . . . . . . 236
Figure E.2 The calibration target. . . . . . . . . . . . . . . . . . . . . . . . . 240
Figure E.3 Model of the calibration target profile . . . . . . . . . . . . 242
Figure E.4 Point spread functions (PSF) and
modulation transfer function (MTF) . . . . . . . . . . . 243
ix
LIST OF PHOTOS
Photo 4.1 Detail of a military vehicle depot
at Tiraspol, Moldova. . . . . . . . . . . . . . . . . . . . . . . 76
Photo 4.2 Display of Open Skies aircraft during the test
certification in Fürstenfeldbruck, August 2001 . . . 91
Photo 4.3 Sensor operators, international experts and observers
on board of the Ukrainian Open Skies aircraft . . . 91
Photo 6.1 Open Skies image of a military airfield. . . . . . . . . . . . 108
Photo 6.2 Thermal infrared line scanner image of an airport
taken at night . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Photo 6.3 SAR image of the military airport at
Fürstenfeldbruck, Bavaria, Germany. . . . . . . . . . . 113
Photo 6.4 Image registration (example) . . . . . . . . . . . . . . . . . . . 117
Photo 7.1 Oder flooding near Wiesenau (Germany) . . . . . . . . . 133
Photo B.1 Boeing OC-135B. . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Photo B.2 Tupolev 154 M . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Photo B.3 Antonov 26 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Photo B.4 Andover twin turboprop . . . . . . . . . . . . . . . . . . . . . . 208
Photo B.5 Lockheed C-130 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Photo B.6 An-30 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
Photo B.7 CASA CN-235-M. . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Photo B.8 SAAB 340 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Photo C.1 Panoramic camera mounted in the
UK Open Skies aircraft. . . . . . . . . . . . . . . . . . . . . 215
Photo C.2 Camera control electronics in the
Ukrainian Open Skies aircraft. . . . . . . . . . . . . . . . 218
Photo E.1 Example of subpixel resolution . . . . . . . . . . . . . . . . . 238
Photo E.2 Example of imagery from a multispectral
line scanner . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Photo E.3 An example of inverse filtering, Wiener filtering
and simulated blur . . . . . . . . . . . . . . . . . . . . . . . . 246
x
LIST OF TABLES
Tab le 2.1 Flight Quotas and Maximum Distances . . . . . . . . . . . 45
Tab le 2.2 Initial Distribution of Active Quotas . . . . . . . . . . . . . . 48
Tab le 3.1 Status of Open Skies Treaty Ratification
as of 1 April 2004. . . . . . . . . . . . . . . . . . . . . . . . . 63
Tab le 4.1 Open Skies Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Tab le 4.2 Open Skies Sensors as of May 2004 . . . . . . . . . . . . . 74
Tab le 4.3 Open Skies Flights Involving Non-Signatory States
as of 31 December 2001 . . . . . . . . . . . . . . . . . . . 83
Tab le 6.1 Ground Resolution (Photogrammetric Definition)
Required for Target Detection, Identification,
Description and Analysis . . . . . . . . . . . . . . . . . . . 105
Tab le 7.1 Estimated Potential of Current Open Skies Sensors . . 131
Tab le 9.1 Performance Parameters of Commercial
1-Metre Optical Satellites. . . . . . . . . . . . . . . . . . . 185
Tab le 9.2 Future Commercial High-Resolution
Optical Satellites . . . . . . . . . . . . . . . . . . . . . . . . . 186
Tab le 9.3 High-Resolution Commercial and
Dual Use Radar Satellites . . . . . . . . . . . . . . . . . . . 187
Tab le 9.4 High-Resolution Military Optical Satellites
in Europe and Asia . . . . . . . . . . . . . . . . . . . . . . . . 188
Tab le 9.5 Resolution (GSD) of Imaging Sensors . . . . . . . . . . . . . 190
Tab le C.1 Sensor Specifications of Open Skies Aircraft. . . . . . . . 216
Tab le C.2 Sensor Parameters of the German AA/AAD-5
Thermal Infrared Line Scanner . . . . . . . . . . . . . . . 219
Tab le C.3 SAROS Operating Modes . . . . . . . . . . . . . . . . . . . . . 220
xi
PREFACE
When on 10 February 1990 the first round of negotiations on an Open
Skies regime opened in Ottawa it was the last time that the old bipolar
order, which dominated Europe for almost 45 years, was represented. The
Open Skies conference turned out to be the first step in the historical Two-
plus-Four negotiations that led to German reunification, the end of the
division of the European continent and of the confrontation between East
and West.
The Open Skies Treaty, signed at Helsinki on 24 March 1992 on the
occasion of the third follow-up meeting of the Conference for Security and
Cooperation in Europe, is the first agreement that fully reflects the changed
political conditions in Europe. It gives real meaning to the idea of a new
cooperative perception of security and stability in Europe. Conceived as a
broad multilateral confidence-building measure in arms control in order to
enhance mutual confidence by promoting openness and transparency it
revived the idea of mutual aerial observation flights for the purpose of
collecting information on military forces, installations and activities. The
Treaty introduces, for the first time, a verification process covering the
entire territory “from Vancouver to Vladivostok”. It is based on the placing
of all states on an equal footing, be they large or small. It helped to facilitate
the implementation of the 1990 Treaty on Conventional Armed Forces in
Europe (CFE) as well as of the Vienna Document on Confidence and
Security Building Measures first established in 1992, the cornerstones of
Europe’s new security architecture.
Almost ten years passed from the Treaty’s signing until its entry into
force in January 2002. During this time the functioning and practice of the
Open Skies regime was tested and refined by a number of bilateral trial
observation missions, which helped to establish and to deepen the
cooperation between states formerly belonging to opposing military
alliances. At the beginning of the 21st century it is time to emphasize the
provisions of the Open Skies Treaty that allow to extend the Open Skies
regime into additional fields. The Open Skies regime is not only a helpful
instrument in facing global challenges such as crisis prevention, post-crisis
management, the proliferation of weapons of mass destruction,
xii
environmental catastrophes or organized international terrorism—it also
could serve as a blueprint in the establishment of cooperative observation
regimes in other regions of the world.
This book co-authored by German, Hungarian and US experts is the
result of a research project carried out between 1995 and 2000. It is the first
publication in the English language that gives a detailed and extensive
account of the concept of the Opens Skies regime, its concretization and its
prospects. Published right in time to inspire future adaptations of the Open
Skies Treaty practice, which will be reviewed in 2005, it helps to
understand the Open Skies regime’s spirit and shows how to make use of
all its provisions in order to make the Treaty an element of a new
cooperative world order.
Bonn, 2 June 2003 Hans-Dietrich Genscher
xiii
EXECUTIVE SUMMARY
1. The origins of the idea of Open Skies date back to the early years of the
arms race between the United States and the Soviet Union. In 1955
US President Dwight D. Eisenhower suggested to the Soviet Union
aerial photography as a means to create mutual transparency of the
weapon arsenals on both sides in order to deter and lift suspicions of
surprise attacks. The idea was also conceived as a verification measure
to contribute to further disarmament. The proposal was strictly
bilateral. It was soon turned down by the Soviet Union (Chapter 1).
2. The Open Skies idea was introduced again in 1989 by US President
George Bush Sr in the final phase of the Cold War. It was an attempt
to overcome Cold War suspicion by mutually agreed openness. This
time the proposal was for multilateral participation. All member states
of NATO and the Warsaw Treaty Organization (WTO) were invited
and participated in the negotiations, which began in early 1990. The
Treaty was negotiated in parallel with the Treaty on Conventional
Forces in Europe (CFE) with the threefold objective of improving
openness and transparency, of supporting the verification of
existing of future arms control agreements and of strengthening the
capacity for conflict prevention and crisis management (see the
Preamble of the Treaty in Appendix I). The Treaty was finally signed in
March 1992 (sections 2.1 and 2.2).
3. The essence of the Treaty is the right to observe any point on the
territory of the states parties—from Vancouver to Vladivostok. The
legitimate interests of the observed state party are taken into account
by ensuring that the maximum ground resolution of the sensors to be
used allows for the reliable identification of major weapon systems, but
not for detailed analysis. The Treaty incorporates several innovations:
it has a strong cooperative element, since flight preparation, execution
and follow-up as well as aircraft certification are carried out by bilateral
or multilateral teams. The imagery taken during observation flights is
accessible to all states parties. Thus the Treaty places all states parties
on an equal footing. It prevents a monopoly on information and ensures
reciprocity of observation, in stark contrast to monitoring by
xiv
reconnaissance satellites owned and operated by individual states
(section 2.3).
4. Initially the Treaty had 26 signatories. By 1995 most member states
had completed their ratification processes and deposited their
instruments of ratification. These were Belgium, Bulgaria, Canada, the
Czech Republic, Denmark, France, Germany, Greece, Hungary,
Iceland, Italy, Luxembourg, Norway, Poland, Portugal, Romania,
Slovakia, Spain, the Netherlands, Turkey, the United Kingdom and the
United States. Georgia followed in 1998. Kyrgyzstan has signed the
Treaty but has not started its ratification process. Belarus, Russia and
Ukraine did not ratify until 2000-01. This delayed the Treaty’s entry
into force until 1 January 2002 (Chapter 3).
5. The long time period between signature and entry into force was used
in two ways. An Open Skies Consultative Commission (OSCC) was
established in Vienna. The Commission discussed and decided upon
many technical and procedural issues, which were left open by the
Treaty text, in particular procedures for verifying that the image sensors
would not underpass the agreed ground resolution (through a
certification procedure). In parallel an extended programme of test
flights was carried out under conditions very close to the Treaty
regulations. Most of the imaging sensor types, which are foreseen by
the Treaty were tested:
• Vertical and oblique optical framing cameras at 30 centimetres
resolution (ground sample distance);
• Panorama cameras at 30 centimetres ground resolution;
• Thermal infrared line scanners at 50 centimetres resolution;
• Sideward looking synthetic aperture radar (SAR) at 3 metres
resolution.
In the trial implementation phase 18 states provided aircraft for Open
Skies use. In the end nearly 400 bilateral or multilateral test flights were
performed. They provided results, of a quality which came close to
those from a full implementation of the Treaty (section 3.2 and
Chapter 4).
6. After entry into force, on 1 January 2002, the aircraft and sensors of 16
states parties successfully passed the certification procedure.
Observation flights under Treaty conditions began on 1 August 2002
xv
within an agreed framework of flight quota. Initially larger states like
the Russian Federation with Belarus have to accept 31 overflights
annually, whereas Germany has to receive four overflights.
In spite of the deep changes in European security relations since
signature, the Treaty has retained its relevance through established
cooperation and by attracting new members. The formerly neutral
states of Finland and Sweden as well as Bosnia and Herzegovina and
Latvia have acceded the Treaty recently. Croatia, Estonia, Lithuania and
Slovenia have applied for accession (Chapter 5).
7. Open Skies images may be used for monitoring all kinds of military
installations and activities, but also for assessing transport infrastructure
and industries. Crisis monitoring applications will include the detection
of illegal traffic in border zones, refugee camps, terrorist training
camps, freshly laid minefields and post-conflict damage assessment.
Photographic black and white images at 30 centimetres resolution will
allow for the detection and general identification of land vehicles and
military infrastructure. In addition, test missions have demonstrated an
excellent capacity for monitoring the effects of environmental disasters
such as floods and hurricanes. In the context of Open Skies, thermal
infrared imaging (not to be used until 2006) will be particularly useful
for monitoring military manoevres and production plants at day and
night. The operational status of vehicles and equipment can be
deduced from their heat profile. SAR images can be taken through
cloud cover and in darkness. The 3-metre resolution under Open
Skies, however, is quite crude: it will permit only the detection and
general identification of large structures such as buildings, airports and
ships.
Open Skies images have already been successfully used to support the
verification of several arms control agreements or arrangements. Once
the full sensor set is operative, its potential for such a contribution will
be significantly enhanced (Chapter 6).
8. Open Skies imagery competes well and is often superior in comparison
with commercial satellite images, which are now available to every
member state irrespective of their access to information derived from
military reconnaissance satellites. First, the resolution of the
photographic cameras (30 centimetres) used in Open Skies is
xvi
unmatched by existing commercial imaging satellites, which reach 60
centimetres at best. Moreover, Open Skies images are routinely taken
in stereo mode, which provides much enhanced power for object
identification through height determination. Second, it would be
extremely difficult to match from space the 50 centimetres resolution
of Open Skies thermal infrared images. It would require mirrors of 5
metres in diameter or more. No commercial satellite provides thermal
images even at 10 metres resolution, nor does any military satellite
provide thermal images at a resolution comparable to that of Open
Skies. In contrast, the 3 metres radar image resolution under Open
Skies will soon be overtaken by a commercial radar satellite of 1-metre
ground resolution developed by the German Aerospace Establishment
(Deutsches Luft- und Raumfahrtzentrum, DLR), which is due for
deployment in 2005 (Chapter 9).
9. Open Skies approaches have been applied successfully regionally. First
of all, Hungary and Romania have been the pioneers by concluding a
bilateral Open Skies Agreement already in 1991. The Open Skies idea
was also introduced to the war-torn entities of Bosnia and Herzegovina
through seven demonstration flights from 1997 to 2001. The United
States publicized the Open Skies concept in bilateral and multilateral
exchanges in Latin America and Asia. Both the (technically much
simpler) Hungarian Romanian Open Skies Agreement and the
(technically more demanding) multilateral Open Skies Treaty can serve
as models for other regions of the world (Chapter 8).
10. The multilateral Open Skies Treaty can and should be adapted to
current security needs and technological trends. Many of the
recommended adaptations can be arranged within the legal
framework of the existing Treaty, in particular:
• Inclusion of additional states parties in crisis-prone regions of
the application area (former Yugoslavia, Caucasus states,
Central Asian republics);
• Applications for conflict prevention, crisis management and
support of non-proliferation of weapons of mass destruction
within the Treaty area;
• Monitoring of environmental disasters and border crossing
environmental problems based on mutual voluntary agreement
of the states parties involved (within the Open Skies
framework).
xvii
The Open Skies Treaty provides a very flexible, modern and future-
oriented instrument. Most of the recommended technical adaptations
require a Decision of the OSCC but not any further ratification step, in
particular inclusion of additional sensors and readout media
(multispectral and electro-optical sensors, colour infrared film, Laser
Fluorescent Spectrometers, digital readout of photographic cameras).
It will be important to make arrangements that the Organization for
Security and Cooperation in Europe (OSCE), the United Nations and
other international security organizations can request Open Skies
support, as foreseen in the preamble of the Treaty (Chapters 7 and 10).
The review conference of 2005 will be an excellent opportunity to adapt
the Treaty implementation to current and future needs.
xviii
xix
ABOUT THE AUTHORS
Pál Dunay holds a Dr. Universitatis degree in international law from the
Loránd Eötvös University (1992) and a Ph.D. in international relations from
the Budapest University of Economics (2001), both “summa cum laude”.
Between 1982 and 1996 he worked at the International Law Department
of Loránd Eötvös University, in various capacities. From 1994 to 1996 he
was Deputy Director of the Hungarian Institute of International Affairs, and
from 1996 to 2004 he was Director of the International Training Course in
Security Policy at the Geneva Centre for Security Policy (GCSP). Since
August 2004 he is a senior researcher at the Stockholm International Peace
Research Institute (SIPRI). He was legal adviser of the Hungarian delegation
at the CFE talks in 1989-90 and at the final (Vienna) phase of the Open Skies
negotiations (1992).
Márton Krasznai joined the Foreign Ministry of Hungary in 1978. He
participated in the CFE and Open Skies negotiations. In 1991 he was chief
negotiator of the Hungarian-Romanian Open Skies Agreement. In 1995 he
was the Chairman of the Permanent Council of the Organization for
Security and Cooperation in Europe (OSCE). In 1996-97 he was Personal
Representative of the OSCE Chairman-in-Office in Bosnia and
Herzegovina; in that capacity he organized several Open Skies
demonstration flights in support of the implementation of the Dayton
Agreement. From 1998 to 2002 he served as Director of the Conflict
Prevention Centre of the OSCE.
Hartwig Spitzer received his Ph.D. in physics from the University of
Hamburg in 1967. From 1970 to 1971 he was a visiting scientist at the
Stanford Linear Accelerator Center, Stanford, USA. Since 1972 he is
Professor of Physics at the University of Hamburg, Germany. At the DESY-
Laboratory in Hamburg, he did research on elementary particle physics and
instrumentation, including questions of pattern recognition and
multidimensional computer aided data analysis. Since 1983 he is engaged
in arms control and disarmament studies at the University of Hamburg. He
is co-founder and spokesman of the Center for Science and International
xx
Security (CENSIS) at the University of Hamburg. Since 1990 he heads the
CENSIS project on “Physical Principles in Remote Sensing and Applications
for Arms Control Verification and Environmental Monitoring” and conducts
research on image analysis of multispectral aerial and satellite imagery.
Rafael Wiemker received a M.S. degree in physics and astronomy from the
Georgia State University, Atlanta, in 1992 and a Ph.D. in physics from the
University of Hamburg, Germany, in 1997. As a research assistant with the
CENSIS group at the Physics and Computer Science departments,
University of Hamburg, from 1993 to 1998, he has developed image
processing methods for multispectral aerial images in the 1-metre ground
resolution range. He is currently with Philips Research Labs, Hamburg,
working on medical image analysis.
William Wynne received his Ph.D. in political science from the Claremont
Graduate School, USA. Dr. Wynne served as a US Department of Defense
technical advisor to the US Open Skies Delegation in Vienna during the
Treaty's negotiation. In that capacity, Dr. Wynne also worked very closely
with the Chair of the Conflict Prevention Centre in helping to design an
Open Skies education programme for the members of the Dayton Contact
Group, and served as a technical advisor on the potential applications of an
aerial monitoring regime to the US Delegation to the Association of
Southeast Asian Nations' (ASEAN) Regional Security Forum. Dr. Wynne is
currently a foreign affairs specialist with the Advanced Systems and
Concepts Office of the US Defense Threat Reduction Agency.
xxi
ACRONYMS
CENSIS Center for Science and International Security (University of
Hamburg)
CCD Charge Coupled Device
CFE Conventional Armed Forces in Europe
CIR Colour infrared film
CPC Conflict Prevention Centre (of the OSCE)
CSBM Confidence- and security-building measures
CSCE Conference on Security and Co-operation in Europe
CTBT Comprehensive Test Ban Treaty
CTF Contrast Transfer Function
CWC Chemical Weapons Convention
DEM Digital Elevation Model
DTM Digital Terrain Model
DTRA Defense Threat Reduction Agency
FAA Federal Aviation Administration
FOV Field of View
GCP Ground Control Point
GDR German Democratic Republic
GIS Geographic Information System
GPS Global Positioning System
GRD Ground Resolved Distance
GSD Ground Sampled Distance
HVO Croatian component of the Army of the Bosnian-Croat
Federation in Bosnia and Herzegovina
IAEA International Atomic Energy Agency
ICAO International Civil Aviation Organization
ICBM Intercontinental Ballistic Missile
INF Intermediate Nuclear Forces
INS Inertial Navigation System
IRLS Infrared Line Scanner
IWG Informal Working Group
Lidar Laser Reflection Measurements
MBFR Mutual and Balanced Force Reductions
MTCR Missile Technology Control Regime
MTF Modulation Transfer Function
xxii
NATO North Atlantic Treaty Organization
NGA National Geospatial-Intelligence Agency (USA)
NIMA National Imagery and Mapping Agency (USA)
NPT Nuclear Non-Proliferation Treaty
OAS Organization of American States
OPCW Organization for the Prohibition of Chemical Weapons
OSCC Open Skies Consultative Commission
OSCE Organization for Security and Cooperation in Europe
POE Point of entry
PSF Point Spread Function
PTBT Partial Test Ban Treaty
RGB Red, green, blue
OPCW Organization for the Prohibition of Chemical Weapons
OSCE Organization for Security and Cooperation in Europe
SAR Synthetic aperture radar
SFOR Stabilisation Force in Bosnia and Herzegovina
START Strategic Arms Reductions Treaty
TCAS Anti-collision warning system (aircraft)
TLE Treaty Limited Equipment
UAV Unmanned Aerial Vehicle
WEU Western European Union
WMD Weapons of Mass Destruction
WTO Warsaw Treaty Organization
1
INTRODUCTION
Open Skies is the longest nurtured idea in the history of modern arms
control. It has bridged over different periods of the second half of the 20
th
century from the peak of the Cold War with its unrestricted confrontation
through its re-emergence at the end of the 1980s, when the Cold War was
about to close, to the post-Cold War era. Throughout the decades its
evolution followed the shifts in European history, albeit indirectly.
The end of the Cold War presents both a challenge and an opportunity
for the intentions implicit in Open Skies. On the one hand, the importance
of traditional arms control verification has diminished significantly. On the
other, new avenues have opened based on cooperation that go much
beyond the original scope of arms control. The changed European political
landscape has modified the previous military and security outlook. Instead
of bloc deterrence enhanced by arms control verification the different
nature of contemporary conflicts has made the monitoring of conflict
regions, particularly in post-conflict situations, necessary, while the
emergence of a range of other security concerns whose redress relies in part
on adequate observation has given monitoring an important role to play.
This book is the outcome of a research project on the “Extension of the
Open Skies Treaty as a Contribution to European Security—Political
Perspectives and Technical Options” which aimed to analyze the evolution
of the concept of Open Skies, its realization in the Treaty on Open Skies and
its prospects.
The project was carried out between 1995 and 2000 by German,
Hungarian and US experts who, beyond their common interest in Open
Skies, had different types of complementary expertise. Professor Hartwig
Spitzer, who has been in charge of the project, is a physicist with a long-time
background in experimental elementary particle physics. In 1989 he co-
founded the Center for Science and International Security at the University
of Hamburg (CENSIS), which specializes in the analysis of aerial images for
arms control and urban planning applications. Doctor Pál Dunay, who is
currently director of the International Training Course at the Geneva Centre
2
for Security Policy, has taught public international law at Loránd Eötvös
University in Budapest and was legal adviser to the Hungarian delegation to
the final (Vienna) phase of the Open Skies Treaty negotiations. Ambassador
Márton Krasznai was involved in the negotiation of the Open Skies Treaty
as a Hungarian diplomat from 1990. He also participated in the negotiation
of the Hungarian-Romanian bilateral Open Skies Agreement. He has been
head of the Conflict Prevention Centre of the Organization for Security and
Cooperation in Europe (OSCE) in Vienna. Doctor Rafael Wiemker is a
physicist. He has worked with CENSIS at Hamburg University for four years
on developing image analysis. At present he is with Philips Research Lab,
Hamburg. William Wynne has a doctoral degree in political science. He has
worked on Open Skies matters as a member of the United States On-Site
Inspection Agency since 1991 and as a technical advisor of the US Open
Skies Delegation in Vienna. He currently has other responsibilities in the US
Defense Threat Reduction Agency.
This book addresses practitioners, policy makers and scientists, who
are interested or involved in Open Skies. It examines the concept, practice
and future of the multilateral Open Skies Treaty from different perspectives:
political, technical and operational. As such, some of the chapters will be of
higher interest to one professional community than to another. Everyone
will find an overview of the substance of the Treaty in section 2.3 and a
general overview in the Executive Summary. The political context, historical
development and political outlook of the Treaty are covered in Chapters 1-
3, 5 and 10. All those interested in the practice of Open Skies flights and
their results are referred to Chapters 4 and 6, where Chapter 6 emphasizes
the information potential of Open Skies imagery. Chapter 7 addresses in
particular policy makers who are in charge of developing Open Skies
further. The chapter has been written in order to stimulate discussion in
preparation for the 2005 review conference of the Open Skies Treaty.
Chapter 9 demonstrates that Open Skies image data compare favourably to
imagery from commercial satellites. Finally, policy makers, security
practitioners and scientists from all over the world will find Chapter 8 a
useful guide on how bilateral or multilateral Open Skies agreements could
be developed outside of Europe and North America.
The project has been greatly facilitated by several interviews
conducted by the authors, who gratefully acknowledge the support of those
diplomats and military experts who were ready to share their invaluable
insights with them during the project. As the interviews were conducted
3
under conditions of anonymity we would like to thank them without
mentioning their names. We appreciate the assistance provided by the
Open Skies Division of the Verification Centre of the German Armed Forces
and of the Open Skies division of the German Ministry of Defence, in
particular by Col. Britting, Lt.Col. Fensch, Wing Commander Lohr, Lt.Col.
Saar and Col. Sperling. Special thanks go to the Volkswagen Foundation,
which financed the project and has patiently awaited this final report.
Geneva, Hamburg, June 2003
4
5
CHAPTER 1
ARMS CONTROL IN THE POST-COLD WAR WORLD
Pál Dunay
The concept of Open Skies emerged as an element of military
transparency in the mid-1950s and was the first, somewhat vaguely
formulated, proposal that sought to overcome the Cold War system
characterized by the militarily confrontation between the Soviet Union and
the United States. Even though the term arms control per se did not exist at
the time Open Skies was first proposed, the latter has retrospectively always
been considered to be an arms control policy. Open Skies was in fact the
first non-nuclear arms control initiative in the nuclear age. Since Open Skies
is thus essentially an arms control policy, a study of the Open Skies Treaty
must arguably commence with an inquiry into the arms control context in
which it emerged, and of which it is still part.
1.1 THE LEGACY OF THE PAST
By the beginning of the 1960s the generation-old idea of general and
complete disarmament had been replaced by the less ambitious, and more
realistic, goal of arms control. At about the same time, during the Cuban
Missile Crisis, the Soviet Union and the United States nearly went to war
with each other. The fact that the two leading military powers of the world
went to the brink of a nuclear disaster turned out to be conducive to arms
control and confidence building. The initial confirmation of this was the
Partial Test Ban Treaty (PTBT) and the first US-Soviet accord on the
establishment of a communication “hotline”.
It is important to note that these first arms control accords did not
oblige the parties to reduce their arsenals. Rather, they introduced
measures that sought to improve the political climate between them and
that encouraged communication even, or particularly, in times of difficulty.
6
Thus, by the time the PTBT was signed both the United States and the
Soviet Union were capable of carrying out underground test explosions,
and so the Treaty had no impact on their development of nuclear arms.
1
Likewise, the hotline agreement was clearly aimed at enhancing the
capacity of the parties to communicate in times of crises, and was not
meant to affect the state of their armaments.
Arms control, as an enterprise, is always premised on a certain level of
mutual confidence between the parties involved. To fully appreciate this
point it suffices to consider two alternative extremes: when states have full
confidence in each other, and when they have no confidence whatsoever
in one another. In case of the former arms control is unnecessary; in case
of the latter it is impossible. The Treaty on Conventional Armed Forces in
Europe (CFE), in its original form, offers a good illustration of the former. The
CFE Treaty regulated the military deployments of the Eastern and Western
alliances, but did not deem it necessary to address intra-alliance military
alignments.
2
The supervision of these was left up to the members of each
respective alliance. North Atlantic Treaty Organization (NATO) states, in a
gentleman’s agreement, have relinquished the right to carry out inspections
of each other’s armed forces. One may conclude from this that in the
absence of perceived threats countries have no need of arms control
agreements. On the other hand, when the relationship among states is
entirely hostile, arms control arrangements among countries become
impossible. This was the case of US-Soviet relations in the late 1940s and
the first half of the 1950s when each saw the other exclusively as an
adversary. Between 1946 and 1955 the two countries did not even attempt
to negotiate arms control provisions. This shows that a certain level of
mutual confidence among countries is necessary in order for arms control
to be feasible. This same observation is phrased by Thomas Schelling and
Morton Halperin in their classical book as follows: “The essential feature of
arms control is the recognition of the common interest, of the possibility of
reciprocation and cooperation even between potential enemies with
respect to their military establishments.”
3
Arms control does not function in a vacuum. Rather, its existence is
contingent on prevailing needs and opportunities. Experience with arms
control is fairly limited. Other than in the specific European context it is
largely confined to some global measures. Consequently, there is little that
can be inferred from the past about its eventual applicability to other
7
regions of the world. This counsels caution when such an application is
contemplated.
Historically two types of arms control may be distinguished: structural
and operational. Structural arms control limits the possession of the amount
or type of weapons by a country. Operational arms control restrains certain
military activities, such as troop movements and manoeuvres, or establishes
transparency measures, like the exchange of information on military
budgets or visits at military facilities.
The following measures are typical of structural arms control:
• The complete ban of a certain type of weapon;
• A quantitative limitation on the possession of a certain type of weapon;
• The prohibition to station a given type of arms in a particular
geographic area;
• The prevention of the proliferation of certain types of weapons.
Structural arms control has always entailed one or some combination
of the above measures.
Operational arms control involves a greater variety of measures than
the structural kind, and is hence less amenable to accurate definition. As
well, operational arms control has a more diffuse history than structural
arms control, which also complicates matters. Still, four types of operational
arms control measures are fairly discernible:
• Measures that facilitate communication in the event of a crisis or in
general;
• Measures that make force postures less offensive;
• Measures that indirectly decrease the reliability of armed forces and
their weapons;
• Measures that reduce secrecy and increase transparency in military
matters.
The difference between structural and operational arms control is not
absolute, however. For example, some instruments contain provisions that
are a mixture of both, and either type can work towards similar objectives.
Nevertheless, historically the differentiation has been relevant in the
European context, where the distinction of measures has been
8
accompanied by a differentiation of countries involved.
4
Thus, while
European operational arms control covers all participating states of the
OSCE,
5
structural (conventional) arms control is limited to the members of
NATO and of the former Warsaw Treaty (and its successor states). Today,
of course, such a geographic distinction makes little sense, yet calls to
harmonize arms control commitments, issued at the 1992 Helsinki summit
meeting of the CSCE, went nowhere.
6
The historicity of division is also
reflected in the adaptation of the CFE Treaty. Although the Treaty was
signed in late 1999, its ratification is still on going and will certainly not be
completed before 2005. The Treaty will not be opened for the accession to
the former non-aligned and neutral countries at least until its entry into
force. This will postpone an eventual harmonization of commitments well
into the middle of this decade. Harmonization, as far as the most important
exception to European structural arms control, the area of the former
Yugoslavia, may by then be less relevant as special arms control regimes
have been implemented there since the mid-1990s. On the other hand,
further regional arms control agreements beyond the confines of the former
Yugoslavia are unlikely, since eventual candidates prefer to accede to the
pan-European instruments rather than to local ones.
The legacy of the past of arms control embraces both the Cold War era
and the subsequent period of the 1990s through to the present. It is, of
course, much easier to draw conclusions about the former than about the
latter. The former was a time of rigid adversarial relations between two
superpowers that dominated the international system. Throughout most of
the era arms control played a substantial political role in terms of a means
of communication through which the two superpowers could discuss issues
of mutual interest irrespective of the prevailing level of tension. During the
Cold War, thus, arms control contributed mainly to bettering political
relations, while its military impact remained marginal. Although several
accords were achieved, these had symbolic political importance rather than
military significance. The militarily significant agreements, namely the ones
that affected force postures, were concluded only in the late 1980s when
the might of the Soviet Union had largely vanished and the Cold War was
on the wane.
It is important to note, however, that during the Cold War results in the
area of operational arms control preceded those in the area of structural
arms control. While this may be regarded as sheer coincidence, it may also
be an indication that when adversarial relations predominate, structural
9
arms control needs to be preceded by some sort of operational measures
that establish a minimum level of confidence. Although we know fairly little
about how confidence building actually works, it may well be that the entire
process begins with measures that promote transparency.
7
It is more difficult to draw conclusions as far as multilateral
conventional arms control in Europe in the broad sense is concerned. The
absence of any tangible results until the mid-1980s deprives us of any
measurable experience, and the legally or politically binding agreements
that were subsequently reached intervened at a time of transition from the
Cold War to the post-Cold War and are thus not representative of what
arms control could achieve during the times of confrontation. The drafting
of the CFE and Open Skies Treaties paralleled the demise of the Cold War.
1.2 PROSPECTS FOR THE FUTURE
Taking a closer look at the more recent situation that has emerged
since the collapse of the Soviet Union the following observations may be
noted. First, the bipolar structure of the international system of the Cold
War has dissipated and with it so has the all-out confrontation between its
main protagonists, although, as the second half of the 1990s illustrated,
residual tensions may have still remained. Second, the United States has
emerged as the sole global power whose worldwide reach and interests give
it a unique perception of international security, which has important
repercussions for its policies in strategic matters, including arms control.
Third, these changes in the structure of the international system have placed
a question mark over the role of arms control in the new European security
context. Thus, during the 1990s, whereas global arms control centred on
the non-proliferation of weapons of mass destruction (WMD) and of their
long-range delivery vehicles retained its importance, in Europe, most of the
main arms control regimes, like the CFE Treaty or the Vienna Documents,
maintained only a marginal relevance in relation to the main function of
arms control of decreasing the likelihood of war. Instead, with the
exception of the Dayton Agreement, European arms control since the end
of the Cold War has served mainly as a framework for dividing up the
conventional forces of the former Warsaw Treaty and of the Soviet Union
as a reassurance for some countries against some others, to marginally limit
arms transfers by dictating the destruction of demobilized weapons and to
introduce a heretofore unknown level of transparency.
10
From the above it is apparent why since the end of the Cold War
European arms control has unfolded in fits and starts. A precise overarching
motivation for it has been missing. A brief look at the course of European
arms control in the past decade illustrates the point.
In the period between late 1990 and mid-1992 European arms control
faced four groups of issues pending from the end of the Cold War. First,
Cold War era negotiations that were still on going had to be completed. The
Open Skies talks are an example of such discussions. Second, the CFE
Treaty, the single most important achievement of European arms control,
had to be brought into force. Bearing in mind the constant changes (mainly
in the form of the dissolution of states) unfolding in Europe at the time, this
proved to be a demanding task.
8
Third, when the Warsaw Treaty
Organization collapsed, the distinction between allied and non-aligned
countries disappeared and the way was opened for a harmonization of arms
control commitments by different countries. This idea launched at the CSCE
Helsinki summit of 1992 has failed badly, at least for the time being. Fourth,
it turned out that the existing treaties and institutions were not adequate to
prevent or mediate the upcoming military conflicts in the former Yugoslavia
and in the Caucasus area.
In the latter half of the 1990s European arms control was shaped by the
emergence of two major challenges. The first concerned the outbreak of
internecine and border armed conflicts as in the former Yugoslavia, which
indicated the need to establish arms control regime(s) directly relevant to
such zones of conflict. The Dayton Agreement and the two accords
adopted in January and June 1996 on confidence building and structural
arms control addressed this concern with respect to Bosnia and
Herzegovina. The second concerned the adaptation of those Cold War
arms control accords that had been sorely overtaken by events, primarily
the CFE Treaty.
9
The experience of the last few years shows how difficult it
is to adapt a treaty to fundamentally changed political conditions. In
addition to having to redraw some provisions, the adaptation process was
nearly blocked by some pending conflicts. Namely, Georgia and Moldova
insisted upon establishing a link between their agreement to the adaptation
of the Treaty and the withdrawal of Russian troops from their respective
territories.
10
It may still happen that ratification of the adapted Treaty will
be delayed should Russia fail to withdraw its troops. Here, it is worth noting
that an eventual withdrawal of Russian troops could be verified through
aerial monitoring.
11
In sum, the experience of European arms control over the last decade
can be summarized as follows. First, a clear overarching principle of post-
Cold War arms control has been lacking. In the absence of significant
residual threat perceptions arms control has become stagnant. Second, the
attempt to shift arms control from a confrontational to a cooperative basis
has proved laborious. European arms control in the 1990s has been a trial-
and-error process which has produced some results, but until now, no
definite outcome. Third, arms control has evolved as a reaction of the
community of European countries to the emergence of topical security
problems, like internecine conflict. This approach, though useful, carries
the risk of incoherence.
A quick look at global arms control yields similar conclusions, though
with a slightly different emphasis. In the industrialized countries improved
political relations since the end of the Cold War have given impetus to
factions that seek to reduce the threat of weapons that commonly create
extensive human suffering in conflicts as opposed to banning seldom-used
armaments. For example, non-governmental organizations motivated by
humanitarian concerns have lobbied extensively for initiatives that could
end up as “hard to verify” commitments. This is the case of the ban on anti-
personnel landmines, and it may yet be the case of certain categories of
small arms and light weapons. It is important to note to what extent the
international community has relinquished the stringent verification
requirements applied until the end of the Cold War. Unless the
international political climate changes considerably, the number of non-
verifiable commitments is certain to increase further still. This trend is
contrary to the one observed in the control of weapons of mass destruction,
where verification measures are increasingly stringent and their observance
is supported by international agencies.
It is difficult, or rather impossible, to offer specific comments about the
future of arms control. Instead it is more prudent to take a more cautious
approach and speak of general trends. With the end of the Cold War, the
importance of US-Russian arms control has declined sharply. There is a
residual risk stemming from the size of the arsenal of the Russian
Federation, addressed through cooperative measures, and the political
uncertainties in that country, although this risk is neither pronounced nor
imminent. Furthermore, the risk is most directly related to the state of the
post-Soviet arsenal of weapons of mass destruction, something that is better
addressed through non-traditional cooperative means, like the US
12
Cooperative Threat Reduction programme, than through traditional arms
control.
Beyond this there is the interest of the nuclear states, the five
permanent members of the United Nations Security Council, to maintain
their privileged status as far as their nuclear and long-range delivery
capabilities are concerned. This entails preventing other countries from
acquiring such capabilities, meaning that the preclusion of the emergence
of other states with nuclear, chemical or biological weapons or with long
range power projection, primarily in terms of missiles, capabilities will
remain the focus of arms control. It should be pointed out, however, that
the non-proliferation of weapons of mass destruction and long-range
missiles is not the exclusive domain of arms control, as recent US foreign
and defence policies have shown. It suffices to mention the so-called
Proliferation Security Initiative, which, although addressing primarily threats
stemming from weapons of mass destruction, is applicable to a broad range
of weapons and dual use equipment.
As far as European arms control is concerned, the following seems
likely. One, European arms control will be largely a corollary of global
efforts. Four of the five nuclear powers belong to the Euro-Atlantic area and
are participating states of the OSCE, and their interest often runs beyond
Europe. Two, arms control will also contribute to tackling the residual
military problems related to Russia, among others providing Russia with a
status in the international system resembling that of the former Soviet
Union, irrespective of its current weakness. Arms control, thus, has a
residual importance as far as the regulation of post-Cold War power
relations are concerned. Three, there seems to be broad-ranging interest in
regulating military relations through arms control following regional violent
conflicts. This has been shown in Bosnia and Herzegovina. It can be taken
for granted that arms control, mixed with other post-conflict settlement
measures, will be applied in other cases. The significance of this sort of
measures depends, of course, on the presence of localized conflicts in
Europe. In case the international community continues to apply such
conflict-specific arms control measures, there is a danger of European arms
control becoming fragmented. That would prevent the emergence of a
common European security space bound, among others, by a common web
of arms control commitments.
13
In the light of its recent past, it is tempting to conclude that the
prospect of European arms control is dim. The principal reason for this is
the absence of a readily identifiable threat that could be addressed by arms
control. This has led to several efforts being made to refocus attention on
cooperative arms control measures. These have achieved some success
noticeable to experts and defence establishments, but not to the wider
public. The current lack of political visibility of arms control makes it
doubtful whether arms control is domestically necessary any longer. This
impression is widely shared by decision makers in many countries, as well
as by the media, which does not attribute as much importance to arms
control as they did during the Cold War. Last, but not least, those security
issues that could be influenced by arms control are not high on the agenda
of public opinion in Europe and North America. Consequently, decision
makers perceive no public pressure to address them.
The shift from confrontation to cooperation based arms control has
been insufficient. Arms control has been able to maintain its relative
significance only in areas where the element of confrontation is at least
potentially present. Future arms control, thus, might be better oriented
towards those residual threats and shaky inter-state relations where the role
of arms control of decreasing the possibility of armed conflict still has value.
1.3 OPEN SKIES AS A BRIDGE IN THE
STRUCTURE OF ARMS CONTROL
As mentioned earlier, arms control may be divided into structural and
operational measures. This division, however, is especially rigid and need
not be particularly enduring. New arms control situations may appear
which require a reconsideration of it. Furthermore, the division of arms
control into structural and operational measures has been inferred from the
evidence gathered from a dozen or so treaties and politically binding
instruments from the European context, which scarcely amounts to a
plentiful basis for this type of exercise.
Hardly any other concept raises as many questions concerning the
differentiation of arms control as Open Skies. The notion of Open Skies
emerged as a means of verification in the 1950s when on-site inspection
was unthinkable and reconnaissance satellites were non-existent.
11
As such
14
conceived, it continued thereafter to supplement both structural and
operational European arms control. Nevertheless, verification measures can
be associated with confidence-building measures just as much as they are
with arms reduction ones, as the document on confidence- and security-
building measures (CSBMs) adopted at the 1980 Madrid OSCE summit,
which stipulates that the measures adopted “will be provided with
adequate forms of verification which correspond to their content”, clearly
suggests.
12
Consequently, the role of Open Skies is dependent upon the
measure with which it is associated.
Taking a closer look at the evolution of the notion of aerial inspection
it is apparent that over time its contribution as a confidence-building
measure has gradually increased, while its role as an associated measure of
either structural or operational arms control has diminished. This is due
primarily to intervening shifts in the context in which aerial inspection
operates. Since its inauguration in 1955, two major developments have
occurred, which have affected the role Open Skies could play. One of them
is technological advance. Namely, the appearance and spread of satellite
monitoring has dramatically altered the scope of aerial monitoring.
Currently, the leading military powers are far less dependent upon aerial
observation than they were in the mid-1950s when the Open Skies idea
was launched and satellite observation simply did not exist. The other is the
attitude of the great powers towards extensive on-site inspections. As long
as this was unacceptable and thus politically inconceivable, aerial
inspection could be assumed to provide the most fruitful means of
verification. When the Soviet Union dropped its opposition to on-site
inspections in the late 1980s,
13
the significance of Open Skies as an arms
control measure lessened considerably. Hence, over time, Open Skies as a
verification and transparency instrument has come under pressure from two
sides: improving satellite capabilities has tended to make it less necessary,
while at the same time the generalization of on-site inspections has tended
to make it less useful.
In sum, the importance of the categorization of Open Skies either as a
structural or as an operational arms control measure has diminished.
Although initially CFE negotiations contained the idea of an aerial
inspection regime, this notion did not reflect in the Treaty proper. Thus,
when Open Skies finally materialized in the 1990s, it was codified as an
element of post-Cold War European confidence building rather than as an
15
arms control means. As such, the Treaty has remained separate from other
European arms control instruments.
Notes
1
The United Kingdom was also capable of carrying out underground
test explosions, whereas France, the other nuclear power at the time,
which lacked such capabilities, did not enter the Treaty and continued
to test above ground.
2
It is of course a gross simplification to claim that no fears and concerns
exist between states that belong to the same alliance. In this respect it
suffices to mention intra-Warsaw Treaty relations at the end of the
1980s.
3
T. C. Schelling and M. H. Halperin, Strategy and Arms Control,
2
nd
edition, Washington: Pergamon/Brassey’s, 1975, p. 2.
4
The classical piece on the topic is still R. E. Darilek, “The future of
conventional arms control in Europe, A tale of two cities: Stockholm,
Vienna” in SIPRI Yearbook 1987: World Armaments and Disarmament,
Oxford: Oxford University Press, 1987, pp. 339-54.
5
Prior to 1993 named the Conference on Security and Co-operation in
Europe (CSCE).
6
The idea of harmonization aimed at addressing the “various existing
instruments concerning arms control, disarmament and confidence-
and security-building” so that the same commitments would apply to
each CSCE/OSCE state. See “CSCE Forum for Security Cooperation”,
para. 12, CSCE Helsinki Document 1992: The Challenges of Change,
Helsinki Summit Declarations, Final Decisions, Helsinki, 9-10 July
1992, http://www.osce.org/docs/english/1990-1999/summits/hels92e.
pdf.
7
J. Macintosh, “Open Skies as a Confidence-Building Process”, in
M. Slack and H. Chestnutt (eds), Open Skies—Technical,
Organizational, Operational, Legal and Political Aspects, To ro nto:
Centre for International and Strategic Studies, York University, 1990,
p. 49.
8
The successful bringing of the CFE Treaty into force nearly two years
after signature was not accompanied by the success of similar efforts
with respect to Open Skies.
16
9
The dissolution of the Warsaw Treaty had rendered the bloc premise
of the Treaty untenable.
10
The situation is easier in the case of Georgia as Russia has committed
itself to withdraw its forces. See Annex 13: “Statement on behalf of the
Republic of Moldova” and Annex 14: “Joint Statement of the Russian
Federation and Georgia”, of “Final Act of the Conference of the States
Parties to the Treaty on Conventional Armed Forces in Europe”, OSCE
Istanbul Document 1999, Istanbul, 17 November 1999, http://
www.osce.org/docs/english/1990-1999/summits/istan99e.pdf.
11
For more details see section 2.1 below.
12
“Questions Relating to Security in Europe”, CSCE Follow-up Meeting
1980-1983, Concluding Document, Madrid, 11 November 1980-9
September 1993, http://www.osce.org/docs/english/1973-1990/
follow_ups/madri83e.htm.
13
Then Soviet Foreign Minister Eduard Shevardnadze declared before
the beginning of the negotiations on conventional armed forces in
Europe in 1989: “We will insist on the strictest and most severe control,
including inspections without the right of refusal, aerial observation of
the situation and the verification of the communication under which
the transfer of troops and armaments takes place. There is no such
control measure that we would not be ready to examine and adopt on
the basis of reciprocity.” E. A. Shevardnadze, “Dan start Venskim
peregovoram: Vystuplenie E. A. Shevardnadze”, Pravda, 7 March
1989, p. 4.
17
CHAPTER 2
THE OPEN SKIES NEGOTIATIONS
AND THE OPEN SKIES TREATY
Pál Dunay and Hartwig Spitzer
2.1 EARLY HISTORY (1955-1989)
Open Skies is the single longest-lived idea of modern, post-World War
II, European arms control. It was the first arms control initiative presented
at the height of the Cold War in 1955 at the Geneva Conference of Heads
of Government.
1
The outlines of the proposal made by US President
Dwight Eisenhower were fairly vague, not surprisingly. This could be due to
the apparent lack of advance work that went into preparing the proposal.
2
It may well be, however, as in many cases with top-level initiatives, that it
was intentionally ill defined leaving the details to be worked out later in
lower level exchanges. As well, there was the importance of the other side’s
reaction. After all, why bother to outline a detailed proposal if it cannot be
assumed realistically that it will be accepted? President Eisenhower actually
stated the following:
Surprise attack has a capacity for destruction far beyond anything which
man has yet known. So each of us deems it vital that there should be
means to deter such attack. Perhaps, therefore we should consider
whether the problem of limitation of armament may not best be
approached by seeking—as a first step—dependable ways to supervise
and inspect military establishments, so that there can be no frightful
surprises, whether by sudden attack or by secret violation of agreed
restrictions. In this field nothing is more important than that we explore
together the challenging and central problem of effective mutual
inspection. Such a system is the foundation for real disarmament.
3
Looking closely at the idea set forth by President Eisenhower it is clear
that it was conceived as a verification measure intended to contribute to
future disarmament. Thus, it may be said that it aimed to provide the
18
transparency necessary for the verification of arms control measures to be
agreed upon later. This manner of approaching matters, that is,
transparency before arms control, is in fact the opposite of how arms
control initiatives were conceived later in the Cold War, when information
exchange and verification were conjured as supplementary to arms
reductions or limitations.
Aerial observation can, of course, serve multiple objectives. As
President Eisenhower noted not much after the Geneva meeting in his radio
and television address:
O
ur proposal suggested aerial photography, as between the Soviets and
ourselves by unarmed peaceful planes, and to make this inspection just
as thorough as this kind of reconnaissance can do. The principal
purpose, of course, is to convince every one of Western sincerity in
seeking peace. But another idea was this: if we could go ahead and
establish this kind of an inspection as initiation of an inspection system
we could possibly develop it into a broader one, and especially build on
it an effective and durable disarmament system.
4
The opportunity to use aerial photography for reconnaissance
purposes was there, as was the potential to apply it as part of an inspection
system to monitor disarmament. However, as at the time the disarmament
system to be monitored was non-existent, the two purported uses of aerial
monitoring could hardly assume equal status. A third possible justification
for Open Skies was the building of confidence. This aspect appeared only
on the margins of the initiative, as evident in the comment made by
Secretary of State John Foster Dulles at his after summit news conference:
P
resident Eisenhower’s dramatic proposal that the United States and the
Soviet Union should agree that peaceful planes would fly over each
other’s territory to take photographs so that each could be sure that the
other was not planning a massive surprise attack.
5
In the absence of any concrete measures to be monitored the notions
of disarmament and confidence building provided relatively weak
rationales for Open Skies. On the other hand, there was a lot to do on the
reconnaissance side. As it was noted:
19
… in 1955 the United States possessed all the necessary weapons for a
counter-force nuclear attack against the Soviet Union. The major
obstacle to confidence that such an attack could be carried out without
a massive Soviet counter-attack was the lack of accurate and complete
targeting data. The US Strategic Air Command was faced with a rapidly
expanding target list. … In this context the Open Skies plan can be seen
as a military intelligence measure of the highest importance, one which
would strengthen the weakest link in US nuclear war-fighting plans.
6
It is open to doubt whether any American politician planned a nuclear
attack, not to mention a first strike, against the Soviet Union. It is certain,
however, that a US-Soviet Open Skies could be used to acquire more
knowledge about the Soviet Union, in particular about its military. Unlike
the United States, the Soviet Union was a closed society, little accessible
from the outside. Open Skies was bound to be more beneficial to the US
than to the Soviet Union. As such, it was understandable for the United
States to propose the idea, and for the Soviet Union, which emphasized
disarmament rather than transparency, to reject it. Transparency,
moreover, could well be a “double-edged sword”. According to the
historian John Lewis Gaddis, President Eisenhower’s concern about surprise
attack was genuine, yet his Open Skies proposal, which he claimed would
help reduce the mutual fear of an attack, was actually a part of an American
political campaign to discredit Soviet peace overtures.
7
Arms control is
necessarily shaped by prevailing circumstances. Hence, Open Skies, was
not envisioned as a possible exit from the confrontation between the two
states, but rather was a reflection or manifestation of this competition.
8
In the light of the above, it certainly is plausible to assume that “the
Open Skies proposal was made with the knowledge that it would be
rejected by the Soviet Union.”
9
Nevertheless, even with such an
assumption, it is interesting to note how Premier Nikolai Bulganin reasoned
at the session of the Supreme Soviet:
At the Geneva meeting US President, Mr D. Eisenhower put forward a
proposal to organize an exchange of military information between the
Soviet Union and the United States and to carry out mutual aerial
photography of both countries’ territory. When giving the necessary
attention to the initiative that has tried to find a solution to the fairly
complex problem of international control, it has to be said at the same
time that the real effect of similar measures would not be great. In the
20
unofficial exchanges with the leaders of the US government we noted
directly that aerial photography cannot give the expected result as our
countries are both located on immense territory on which everything can
be hidden away as necessary. It has to be taken into account that the
initiated plan affects only the territory of the two countries leaving aside
military forces and their armaments located on the territory of other
states.
10
Bulganin’s remarks make it clear that for the Soviet Union the US
proposal was unacceptable. The territory of the two countries was too large
for Open Skies to be put into practice, and besides, in view of the
developing network of American bases abroad, limiting the regime to
national territory would have placed the USSR at a greater disadvantage
still. An alternative might have been to ask for a global Open Skies regime,
but that would have been even more impractical and, again, of little net
benefit to the Soviet Union. The shooting down of a spying American U-2
surveillance plane over Soviet territory in the spring of 1960,
11
only
increased Soviet hostility towards the idea of Open Skies, and the coming
to power of Leonid Brezhnev in 1964, who was even less inclined towards
military transparency than his immediate predecessor, spelled the end of
the idea at least until the latter’s demise.
In the 1960s and 1970s two major developments affected the scope of
monitoring arms control arrangements, respectively. First, was the
emergence of satellite technology. As it was put: “The information collected
by satellites ultimately became an essential element of bipolar stability, in
much the same way that Open Skies information could have done earlier,
had it been available.”
12
Second, the United States and the Soviet Union
concluded bilateral arms control agreements followed by several European
accords whose adequate verification had to be ensured. The appearance of
these two factors had a profound impact on Open Skies. The technology
that could, at least partially, replace aerial monitoring was available, at least
to the two superpowers. The arms control arrangements that made
verification necessary existed as well. It was open to question whether in the
light of these aerial monitoring, or more precisely Open Skies, would find
application. The rigidity of the Cold War order, which was dominated by
those states that had the most extensive and for some time nearly exclusive
access to national technical means, which alone could provide for the
necessary monitoring, did not bode well for Open Skies. It is not a
coincidence, thus, that serious consideration of Open Skies had to await the
21
beginning of the demise of the Cold War and the turn of the Soviet Union
toward a more favourable view of transparency effected by Gorbachev.
During the Brezhnev years the idea of aerial monitoring was
mentioned only once. In 1978, France proposed the “establishment of an
aerial or satellite surveillance system” in Europe, as a means of
stabilization.
13
It is interesting to note that the initiative was formulated
alternatively. Stabilization could either rely on aerial or satellite surveillance.
Of course, satellite monitoring on a European scale would have required a
significant expansion of access to satellite information both within NATO
and the Warsaw Treaty or an internationalization of monitoring by national
technical means on the continent. Even though the proposal was not taken
up it was the only reference to aerial surveillance on a multilateral basis
before the mid-1980s.
The issue of aerial monitoring next emerged at the Stockholm meeting
on CSBMs in 1984-86 in the context of monitoring limitations on ground
forces training exercises. The matter remained highly controversial
throughout most of the talks. The United States indicated that the most
effective way of checking compliance would be through on-site inspection.
Aerial inspection, involving suitable short take-off-and-landing aircraft (such
as the US C-130, Soviet AN-12, or Canadian DHC Dash-7) that permitted
to overfly the entire exercise area, “could be the optimum method of
checking compliance”.
14
The Soviet response until fairly late in the
discussions demonstrated continuity: “… we are categorically opposed to
such measures which, under the guise of verification, serve as a legalized
means of collecting intelligence data and constitute direct interference in
the internal affairs of another state…” (Tatarnikov). In other words, the
traditional Soviet military concern with secrecy still dominated the Soviet
diplomatic position. This stance prevailed until 20 days before the close of
the conference, when the newly arrived Marshal Akhromeyev, Chief of the
General Staff, announced that the Soviet Union agreed to aerial as well as
ground on-site inspection.
15
Although the exact sources of the shift in the
Soviet position are difficult to identify with accuracy, a likely explanation
may lay in the inner struggles taking place within the Soviet leadership at
the time and the eventual triumph of Gorbachev and his programme, which
aimed at a political rapprochement with the West.
Examining the first international document that introduced aerial
monitoring in a politically binding manner, the following can be noted.
22
Inspection was permitted on the ground, from the air, or both, meaning that
aerial inspection was regarded as a form of on-site inspection. As under
international law airspace is considered to be part of a country’s territory,
this is quite normal. The document, however, was vague insofar as the
equipment to be used for such purposes was concerned. Accordingly, aerial
monitoring could be conducted by airplane, helicopter, or both. This left
the question of the ownership of the plane or helicopter to be used during
inspection undecided. The ambiguity reflected the division of the
participating states on the matter. The Soviet Union had insisted that the
aircraft and aircrew of the inspected state would have to be used, whereas
NATO countries had preferred that the inspecting state’s own planes and
pilots be employed. The neutral and non-aligned states had presented their
compromise according to which the planes of states belonging to their
group would be used. The Soviet Union had rejected this on the grounds
that if the inspecting state’s aircraft were used this could “be equipped with
the appropriate intelligence gear that can check not only the actions of
troops in this region, but also be capable of reconnoitering any installation
that is not the object of monitoring. This would be unlawful intelligence
activity and a violation of a state’s sovereignty.”
16
In the light of the history
of aerial reconnaissance this complaint was understandable and remained
an issue for later Open Skies discussions. Eventually compromise was
reached so that aircraft “for inspection will be chosen by mutual agreement
between the inspecting and receiving states”.
17
It remained to be seen what
to do if the inspecting and the inspected state could not agree upon whose
aircraft to use.
Further measures guaranteed that the observation could not be used
for espionage. Among them, beyond the selection of the observation
aircraft, was the provision that directions “to the crew will be given through
a representative of the receiving State on board the aircraft involved in the
inspection.”
18
Moreover, one member of the receiving state’s inspection
team was to be permitted at any time to observe data on the navigational
equipment of the aircraft and to have access to the maps and charts used
by the flight crew for the purpose of determining the exact location of the
aircraft during the inspection flight.
19
The two provisions above clearly
indicate that the inspecting and the inspected (or, more precisely, the
receiving) party were obliged to cooperate throughout the inspection. This
was certainly a major advantage of the regime. Bearing in mind the limited
number of inspections carried out under the Stockholm regime and the
even more limited reliance on aerial inspections one could hardly speak
23
about an established practice before the end of the Cold War, however.
The regulation of the Stockholm regime proved nonetheless useful during
the drafting of the Open Skies Treaty in indicating both the limits of the
parties’ flexibility and their readiness to cooperate.
After the Stockholm conference the notion of aerial inspection
remained constantly on the agenda. When the Bush administration came
to office in the United States at the end of 1988 the National Security
Council staff was ordered to prepare a wide-ranging review of arms control
and confidence-building initiatives. Both for historical and actual reasons,
Open Skies was among the initiatives considered. It was a genuine
American idea whose time, as the Stockholm regime had demonstrated,
might have come. It was moreover a transparency measure which the
Soviet leadership with its claims of “glasnost’” could hardly find justification
to reject.
The re-launch of Open Skies was initially conceived as a bilateral US-
Soviet measure. This reflected the essentially status quo orientation of the
United States. Such an approach, however, overlooked many of the side
benefits, which a multilateralization of Open Skies could offer in the fluid
context of a faltering Soviet Union, like encouraging the emancipation of
the smaller members of the Warsaw Treaty or providing access to
monitoring information to states which had none. Such benefits, however,
were recognized by at least one US ally. In consultation with the US
President, the Canadian Prime Minister Brian Mulroney called attention to
the importance of a multilateral Open Skies.
20
Grasping the opportunities
to be seized, President Bush presented the new US Open Skies initiative in
his speech at Texas A&M University on 12 May 1989, as follows:
Now let us again explore that proposal, but on a broader, more intrusive
and radical basis—one which I hope would include allies on both sides.
We suggest that those countries that wish to examine this proposal meet
soon to work out the necessary operational details, separately from other
arms control negotiations. Such surveillance flights, complementing
satellites, would provide regular scrutiny for both sides. Such
unprecedented territorial access would show the world the true meaning
of the concept of openness. The very Soviet willingness to embrace such
a concept would reveal their commitment to change.
21
The Bush proposal differed from his predecessor’s in two important
respects. First, the proposal aimed to initiate multilateral negotiations with
24
the involvement of all the members of NATO and of the Warsaw Treaty.
Second, Bush proposed to begin separate negotiations, thus de-linking
Open Skies from other forums, primarily the CFE talks, which had started
two months before the speech.
Unlike the Eisenhower initiative, Bush's proposal was positively
received both in the East and in the West. A conference was convened to
the capital of Canada between 12 and 28 February 1990 to discuss the
idea.
2.2 THE OTTAWA, BUDAPEST AND VIENNA ROUNDS OF
NEGOTIATIONS IN THE LIGHT OF CIRCUMSTANCES
The nine months that passed between President Bush’s address at
A&M University (12 May 1989) and the opening of the first round of
negotiations in Ottawa (12 February 1990) was one of the most tumultuous
periods of 20th century European history. Consequently, the underlying
conditions upon which Open Skies was premised, both in its original form
of 1955 and the renewed version of 1989, were no longer valid. This,
however, was not immediately apparent. Primarily, the United States still
perceived the Soviet Union as a hostile, though certainly more cooperative
than earlier, force. Forward deployed Soviet troops were still present in
Central Europe, the USSR still enjoyed some conventional superiority and
had far more armaments and military equipment than necessary under the
doctrine of reasonable sufficiency. Even though Soviet political intentions
seemed to be different and one strategic advantage, namely loyal allies
willing to host forward deployed troops, was gone, the military capabilities
of the Soviet Union remained comparable to those of earlier times. Western
strategic planners were in many cases reluctant to notice the depth of the
transpiring change. It is for this reason that the fundamental rearrangement
of European security conditions had no immediate bearing upon Open
Skies in that the expectations set for the regime did not change.
The goals of the Bush initiative thus remained: “to increase mutual
confidence through increased scrutiny of each other’s activities, and to test
President Gorbachev’s commitment to glasnost.”
22
Neither objective was
sufficiently specific to answer those questions that were pertinent to the
establishment of the Open Skies regime, however. It was not clear what
25
flights would have had to be carried out and what they would have had to
monitor in order for confidence building to occur. Moreover, it was fairly
difficult to discern exactly what Soviet concessions would have been
compatible with glasnost.
As in the case of its predecessors, the new multilateral Open Skies
proposal contained few details to speak of. Even Washington had little clue
as to what to expect. In fact, the US initiative hardly went beyond the words
pronounced by President Bush. Moreover, NATO allies had not been
briefed prior to the 12 May 1989 speech. Following the speech, the
surprised NATO heads of state and government issued the following
reserved reaction:
W
e consider as an important initiative President Bush’s call for an open
skies regime intended to improve confidence among States through
reconnaissance flights, and to contribute to the transparency of military
activity, to arms control and to public awareness. It will be the subject of
careful study and wide-ranging consultations.
23
The second sentence clearly indicated the dissatisfaction of the NATO
allies with the lack of consultation prior to the 12 May 1989 announcement.
This factor contributed to the ensuing hasty preparation of a more detailed
notion of the regime, primarily in Washington. For the United States,
though, the more pressing concern had to do with the fact that Open Skies
was going to subject the US homeland territory to extensive aerial
monitoring for the first time, and the attending strategic consequences of
this.
24
In view of the above, the main US concern in preparing for the
negotiations on Open Skies related to the intrusiveness of the future regime.
The United States intended to create a balance between “maximizing
openness and minimizing any harm to US national security that might result
from Warsaw Pact overflights”.
25
The cautious estimates of the Pentagon
and of the US intelligence community suggested that one Warsaw Treaty
overflight per week (carried out with sufficient advance notice) would not
pose an unacceptable security risk to sensitive US programmes and
installations. By August 1989, thus, the US position had settled on a
moderately intrusive Open Skies regime.
26
NATO consultations began later that month. The Senior Political
Committee of the organization established a drafting group that prepared a
paper laying out the basic elements of an Open Skies regime. It was
approved by the North Atlantic Council in December 1989 as an Annex to
the document of the Council meeting.
26
The draft played an important role
for two reasons. First, the drafting required that major internal differences
in NATO be sorted out. Second, it proved to be a document of lasting
relevance. Due to the severe difficulties of reaching a compromise within
the Alliance, NATO states were subsequently unwilling to depart much
from what had been achieved, and thus insisted that a final agreement on
Open Skies closely follow the principles set out in the document.
The fundamental difference within NATO occurred between the
United States and some European allies, most vocally represented by
France. It would be premature to enter into details concerning the points of
disagreement.
27
Broadly put, the United States emphasized the importance
of a bloc-to-bloc approach, which France rejected.
28
This time France had
persuasive arguments, however. Whereas Washington kept seeing the
Open Skies process through the framework of its strategic relationship with
Moscow, Paris—increasingly accompanied by other Western European
countries—intended to redefine that framework. France argued that the
regime should be organized strictly on the basis of individual states noting
that “the westward-leaning members of the Warsaw Treaty might be more
interested in overflying the Soviet Union than Western Europe”.
29
Likewise,
the United States, again seeing Open Skies through the prism of US-Soviet
relations, was reluctant to involve the non-aligned states in an eventual
arrangement. France, eager to break the preponderance of US-Soviet
relations in European security matters, was in favour.
The NATO consultations ended with compromise that reflected mostly
the US position. As it was stated following the first NATO-Soviet
consultations:
Because of the complexity of establishing such a regime, it would initially
be limited to interested members of NATO and the Warsaw Pact. …
Next steps: Further consultations are to be held within the Atlantic
Alliance. We anticipate a multilateral conference of interested NATO
and Warsaw Pact nations.
30
27
The US-Soviet consultations seem to have been easier than those
carried out inside NATO. For different reasons the positions of the two
superpowers were closer to each other than to those of their respective
allies. The Soviet Union wanted to maintain the bloc-to-bloc framework in
order to shore up its status within the Warsaw Treaty. In September 1989,
when the bilateral consultations took place, it was not at all evident that the
process of change active beneath the surface in Central and Eastern Europe
would result in the collapse of the Cold War order in such a short period of
time. The US felt it important not to challenge the Soviet Union
unnecessarily and prematurely. Furthermore, both countries felt more
comfortable with their well-established bilateral negotiation framework,
which they were understandably reluctant to discard in favour an untried
new one.
Before the fall of the Berlin wall, Canadian Prime Minister Brian
Mulroney offered to host an Open Skies conference in Ottawa in January
or February 1990. By the time the foreign ministers and the diplomats met
in the capital of Canada on 12 February 1990, the fundamental change in
European politics could no longer be ignored. The agenda of the meeting,
particularly in its initial foreign ministerial phase, was overshadowed by the
two major issues of German reunification and the stationing of foreign
troops in Europe under the CFE Treaty negotiated in Vienna, neither of
which had much to do with the professed purpose of the conference. In
terms of Open Skies discussions, the difference between the NATO and the
Soviet positions were and remained fundamental. In view of the wide gap
between the two sides it would be too fanciful to conclude which side
“won” in Ottawa.
Irrespective of their purported bloc-to-bloc arrangement, the Ottawa
talks signalled clearly for the first time that a new structure of international
relations was taking shape. On the one hand, the Soviet Union began to
represent strictly itself, and no longer even claimed to speak on behalf of its
Warsaw Treaty allies. Coming to the conference, the Soviets had sought to
present a unified Warsaw Treaty proposal. Such a proposal, however, had
been wrought only at the expense of concessions made in the course of one
full month of negotiations with its allies, something that the Soviets
feverishly resented. Hence, after presenting the common position, the
USSR reverted to its earlier stance on some of the substantive, controversial
issues. With the Warsaw Treaty clearly moribund, Moscow no longer felt
the need to pay heed to the views of its ostensible allies,
31
and instead
28
concentrated on countering the NATO position. On the other hand, some
Eastern European countries began to regard themselves more as mediators
between the Soviet Union and the US than as Soviet allies. Moreover, many
smaller members of the Warsaw Treaty started to emphasize the need to
draft an agreement of 23 sovereign states rather than two alliances. The
European political order of the Cold War had passed into history.
The Ottawa conference ended with the result that the parties agreed
to differ on all of the major issues. Talks at the conference had been divided
into working expert groups covering a range of topics: a) Aircraft and
sensors, inspection of aircraft and sensors, the role and status of inspectors
on board of observation aircraft; b) Quotas, geographical scope and
limitations; c) The conduct of observation flights, flight safety, transit above
third states; and d) The nature of the agreement, Consultative Commission,
liability, status of personnel, further measures. After the two weeks of
deliberations only the latter working group could present some results on
fairly simple issues such as depositories, entry into force and authentic texts,
while the others all faced major problems.
The Ottawa conference showed that it is extremely difficult to
negotiate an arms control arrangement when relations between the parties
are in great flux. By the time the conference was convened political
relations in Europe were well on their way to being radically revised. Under
the circumstances the parties could either continue their talks as if nothing
had changed at the risk of drawing an agreement that no longer
corresponded to any perceivable reality
32
or they could postpone decisions
until the contours of the emerging political relations had become more
clearly defined and seek to obtain a treaty that in fact reflected the changed
circumstances. In view of the results of the conference, the latter possibility
prevailed.
With little agreement achieved at the Ottawa conference, the
discussions continued at Budapest
33
two months later between 23 April
and 10 May 1990. In the light of the experience at Ottawa there was little
hope of arriving at an agreement in time for the close of the meeting,
although, foreign ministers had already been asked to reserve the dates of
11-12 May for a signing ceremony. Despite the subdued expectations the
meeting did witness some cautious advances and slight shifts of positions.
To begin with, the United States began to modify its stance and to apply
increasing pressure on the USSR. The revised US policy was a mixture of
29
sticks and carrots. The United States consolidated East-Central European
support behind NATO positions and made it plain to the Soviet Union that
it was isolated, while at the same time it showed itself accommodating
towards Soviet concerns over the technological superiority of the West and
the advantages which this might bestow.
34
On the Soviet side flexibility
began to creep in. The USSR accepted that synthetic aperture radar (SAR)
would be permitted on observation aircraft, though with strictly limited
resolution capability the exact definition of which remained undecided,
and differences were somewhat reduced over flight quotas, although the
issue remained unresolved especially that now it had become clear that
every party, including Warsaw Treaty members, was primarily interested in
overflying the Soviet Union.
35
The Budapest conference was a necessary if not particularly successful
stage in the evolution of Open Skies. Its most notable feature was the shift
in the US position away from a bloc-to-bloc perspective. The source of this
change is too complex to be untangled, although, two factors may be cited.
One, the European members of NATO had placed a lot of emphasis in
intra-alliance discussions on the irrevocably changed conditions of political
relations emerging in Europe and the consequent need for a commensurate
shift in the approach to relations between East and West. Two, direct
contact by US officials with Eastern European personalities such as Vaclav
Havel or Lech Walesa as well as the actual orientation favourable to NATO
evident in the policies of Eastern European countries at the negotiating
table, particularly in the concomitant CFE Treaty negotiations, combined
with their declared unwillingness to continue to host Soviet troops and
armaments on their territories, helped convince the Americans that the time
had come to push for the emancipation of the Warsaw Treaty countries
from the Soviet grip. Subsequently, the US position may also have been
influenced by the failure to include aerial observation in the monitoring
means specified in the CFE Treaty, which increased the importance of an
eventual Open Skies accord, and which may have induced the US to adopt
a somewhat more concessionary stance on certain issues.
36
When the Budapest conference came to an end it was clear that talks
would not resume any time soon. German unification dominated high
politics in Europe, and the CFE talks kept the arms control community
sufficiently busy. Still, aerial monitoring of conventional arms limitations,
one of the foreseen functions of Open Skies, constituted an important link
between CFE and Open Skies. As it turned out, progress on Open Skies
30
became inextricably linked with the proceedings and outcome of the CFE
negotiations, namely the attempt and ultimate inability to include aerial
monitoring under the CFE verification framework.
In the wake of the Budapest conference, attempts to make aerial
inspection part of the CFE Treaty intensified. On 14 June 1990 the so-called
group of 16, member states of NATO, tabled its aerial inspection protocol.
This move was urged by the United States and Canada, which considered
that aerial monitoring could contribute to a cost-effective verification
system that could reduce the number of on-site inspections by 25 to 30%.
Other Western delegations, notably Germany and the United Kingdom,
had doubts about the wisdom of starting discussions about such a
complicated matter as aerial inspection at such a late stage in the CFE
negotiations, and feared that the attempt could delay the reaching of an
accord. Their concerns were justified by the experience of two rounds of
the Open Skies negotiations. It would have been immensely difficult to
bridge the gap between the Western and the Soviet positions even if the
NATO proposal focused on technical details and made no mention of the
two issues where the most disagreement was likely, namely the ownership
of the aircraft used for carrying out inspections and the types of sensors
those aircraft were permitted to carry. The group of 7, the member states of
the Warsaw Treaty, tabled its own document on aerial inspection on 7
August 1990. The document did not address any of the fundamental issues,
like ownership of inspection aircraft, type of applicable sensors, inspection
quotas or information sharing.
37
With little scope for agreement, and with
growing concerns over the needless delay of reaching an accord or worse,
the attempt to incorporate aerial monitoring within the CFE framework was
abandoned.
The day before the signature of the CFE Treaty in Paris the states parties
exchanged information on their military deployments, as required. The
Soviet Union, which had redeployed thousands of pieces of armaments
outside of the treaty-covered area and had reassigned three army divisions
to naval infantry effectively placing them outside the scope of the Treaty,
notified a surprisingly small total number of treaty-limited armaments and
equipment and failed to give any account of the armaments and equipment
of the reassigned units. As the Dutch ambassador to the talks put it later: “…
on the eve of those euphoric days of the Paris Summit … the countries
which had participated in the Treaty negotiations were in for a cold
shower.”
38
31
The Soviets’ withdrawal of the more than 55,000 pieces of armaments
out of the geographical area covered by the CFE Treaty had placed 16,400
battle tanks, 15,900 armoured combat vehicles and 25,000 pieces of
artillery out of the reach of the Treaty’s inspection regime, and as such
susceptible only to monitoring by national technical means, which were
unavailable to most parties to the accord. For obvious reasons, this situation
was unacceptable, and an obstacle to the Treaty’s entry into force. More
than half a year was needed to straighten out the problem. Finally, the US
and the USSR struck a compromise whereby the Soviet Union pledged that
the withdrawn armaments would “not be used to create a strategic reserve
of operational groupings, and will not be stored in a way permitting their
rapid return to the area of application”
39
and committed itself to provide
information about the locations and quantity of withdrawn equipment as of
1 July 1991.
The Soviet antics of redeploying what should have been equipment
subject to limitations and inspection under the CFE Treaty and the
subsequent compromise on this issue had created a peculiar situation.
Although the withdrawn equipment was now subject to the exchange of
information, it still remained beyond the reach of verification. Open Skies,
however, with its broader territorial scope, held the potential to remedy this
condition, hence its increased importance from the point of view of both
NATO and Eastern European countries. But how about from the viewpoint
of the Soviet Union? Did the prevailing circumstances not militate against
opening the Soviet Far East to aerial overflight, and thus against Open Skies?
Ultimately, from the negotiating history of Open Skies it is clear that the
opening of the Soviet Far East to aerial monitoring was balanced against by
the opening to aerial monitoring of the territories of the United States and
Canada. Soviet strategists attributed great importance to the possibility of
overflying the United States, and Open Skies offered the only means of
achieving this. The Soviet Union also demanded that military bases of the
parties on the territory of third countries, like those of the US in the
Philippines, be subject to overflight as well, but the request proved both
unacceptable to the US and unworkable, and so had to be dropped at a late
stage of the talks.
40
Besides a reinforced strategic importance, Open Skies also benefited
from the exchanges around the ill-fated effort to endow the CFE Treaty with
an aerial monitoring regime. Namely, the positions of the parties on the
issue continued to soften. Unsurprisingly, given what had transpired, but
32
also due to a diminution of the estimated Soviet threat, the softening was
most evident on NATO’s side. As such, the draft inspection protocol tabled
by NATO now adopted the following vague formulation: “Within the area
of application, each Party concerned shall be obliged to receive, and have
the right to conduct a specified number of aerial inspection overflights as an
essential component of Treaty monitoring and verification.” The modalities
of these obligations remained to be worked out later: “(Details of the aerial
inspection regime will be developed.)”
41
Moreover, the Alliance revised its
earlier stance concerning the sharing of raw data, the performance of
sensors and the ownership of observation aircraft.
Although the summer of 1991 had not been the most conducive to
dealing with Open Skies, four events had reconfirmed the need to achieve
an agreement. One, there was the firm commitment of the parties to
continue the negotiations until the process was completed by the adoption
of a multilateral Open Skies agreement. Two, Hungary and Romania signed
a bilateral Open Skies agreement during the recess of the multilateral talks
in May 1991. This was the first breakthrough for an Open Skies approach
on the practical level.
42
Three, despite the failure to include aerial
monitoring in the CFE Treaty, the commitment to eventually agree on aerial
monitoring was clearly there. As such, the Treaty stipulated that after the
end of the completion of the 40 months reduction phase “each State Party
shall have the right to conduct, and each State Party with territory within the
area of application shall have the obligation to accept, an agreed number of
aerial inspections within the area of application. Such agreed numbers and
other applicable provisions shall be developed during” follow-on
negotiations.
43
Four, the already noted 57,300 pieces of heavy armaments
redeployed by the Soviet Union outside the zone of application of the CFE
Treaty, demanded some form of monitoring.
In sum, the critical mass for completing the Open Skies negotiations
was present by the summer of 1991. What remained to be seen was how
the deadlock was going to be broken. As it turned out, the impetus came
from Germany. Little after the Soviets had agreed to partially regularize the
situation of the pieces of equipment withdrawn from the CFE Treaty zone
of application, German Foreign Minister Hans-Dietrich Genscher, in taking
over the presidency of the Western European Union (WEU), sent a letter on
behalf of the WEU to his Soviet counterpart proposing that the Open Skies
talks recommence.
44
The initiative was skillfully conceived in two senses.
One, it managed Soviet sensitivities—formally, the proposal was coming
33
from an organization more palatable to the Soviets than NATO. Two, being
launched by Genscher, who, as a result of the German reunification
negotiations, had acquired a certain personal credibility in Moscow, the
initiative was bound to be better received than otherwise.
The renewed impetus toward Open Skies, notwithstanding, in some
areas, the positions of NATO and the Soviet Union continued to remain so
far apart that the differences between them seemed unbridgeable. In
situations such as this the skill of diplomats is of little use in overcoming the
blockage. This was the case here as well. It took an event of world political
importance to help move out of the stalemate. The abortive coup of August
1991 changed everything.
When the negotiating teams reassembled in Vienna in early September
for an exploratory session, the USSR delegation had no new instructions,
but was so confident of their imminent arrival that it predicated its entire
behaviour during the ensuing negotiations on the assumption that it would
soon have the flexibility needed to end the talks successfully. Informal
discussions focused almost exclusively on possible compromises in the
Soviet position. At the end of that week in September, the Soviet Union’s
representative did not demur in the proposal to set the beginning of the
Helsinki follow-up meeting in March 1992 as the target for the end of the
Open Skies negotiation.
45
It would be incorrect to assume, however, that the remaining months
that led to the signature of the Open Skies Treaty represented a simple
technical exercise. The on going shifts in political circumstances continued
to take their toll. Then there was also the fact that multilateral negotiations
are generally more complicated than bilateral ones. In all, by the time the
parties reconvened at Vienna in November to try to bring the discussions to
a close, a lot of work still remained to be done. A determined push did
result in the drafting of approximately one third of the Treaty text including
the appendices by Christmas, however. Issues such as the designation of
aircraft, transit procedures and deviations from the flight plan, among
others, were successfully resolved, whereas others such as sensors, their
resolution and processing, and the sharing of information, remained
outstanding.
The final months of the talks, that is, the first months of 1992, brought
a host of new problems that stemmed from the dissolution of the Soviet
34
Union. The question of succession created a situation where it was no
longer clear which successor states should rightfully assume the USSR’s
place in the discussions. Russia, as in other fora, such as the United Nations
and the CSCE, was immediately regarded as a successor state. Belarus and
Ukraine were accepted in a somewhat lesser status until they became
members of the CSCE. This raised the peculiar question of the participation
of the European neutrals, however. Countries like Switzerland, Sweden and
others, which had initially been left out of the negotiation process, now
tried to join the talks on the ground that with the dissolution of the bloc-to-
bloc approach engendered by the formal collapse of the Warsaw Treaty,
the justification for their initial exclusion had now vanished.
46
Still, no
difficulty could derail the process at this stage. With Russia willing to
cooperate and with little appetite on anyone’s part to reverse the process,
the successful outcome of the negotiations was assured.
The Open Skies Treaty was signed on 24 March 1992 by the foreign
ministers of the states parties. However, as the past decade has shown, the
signing of the Treaty was certainly not to be the end of the story. In order to
understand the main achievements and the pending problems of the Open
Skies Treaty it is necessary to analyze the main issues that dominated the
two years of negotiations, their resolution and the Treaty’s subsequent
evolution from signature to the completion of the ratification process.
2.3 NEGOTIATION OF THE TREATY SUBSTANCE
AND
ANALYSIS OF THE RESULTS
The radical political transformations of the 1980s and the 1990s
represented the background against which the Open Skies Treaty
negotiations took place. So far we have shown how those transformation
impacted upon the negotiations. Now we turn to how the negotiations
were actually conducted and how compromise was reached on some of the
major issues. We do not attempt a comprehensive analysis of the entirety
of the discussions for two reasons. Due to the huge complexity and often
the clearly technical character of the talks, it is neither possible nor
necessary to do so.
47
Nevertheless, in some of the instances presented here
the story of the talks is intriguing. It tells a lot about national interests, their
rearrangement and the cohesion of alliance(s) or the lack thereof. Whereas
heretofore we have focused on the broad picture and how structural
35
change in the political relations of the Cold War impacted upon Open
Skies, now we delve into the intricacies of the negotiations themselves,
which in turn also tell a lot about the large picture.
2.3.1 Definitions
Definitions are an inherent part of every major arms control
agreement. In case of the Open Skies Treaty, 35 terms are listed under
Article II.
48
They deal with the status of the parties, the area of application
of the Treaty, the equipment used in the observation flights, the status and
role of the personnel and so on. Some of them are linked with other
complex matters or mask the controversial issues behind the
compromise.
49
2.3.1.1 The Status of the States Parties
With the decision to shift Open Skies negotiations onto a multilateral
basis, the question of states parties was bound to come up. Initially the
multilateralization of the talks posed little difficulty. Although it was evident
that the negotiations would involve NATO and Warsaw Treaty members,
the eventual participation of neutral and non-aligned countries remained
open to question. Despite disagreement on granting them a full status at the
talks, a compromise was worked out to minimize the sharp distinction
between alliance members and non-aligned states so that the latter could
still take part in the discussions in one form or another. The participation of
non-aligned states in the Open Skies discussions did not translate into their
inclusion in the Treaty, however.
At the Ottawa conference, NATO’s position on the participation of
non-aligned states in the Open Skies discussions, outlined in the Basic
Elements paper,
50
was that “Open Skies is initially open to all members of
the Atlantic Alliance and the Warsaw Treaty Organization”, but that NATO
“will be ready to consider at an appropriate time the wish of any other
European country to participate in the Open Skies regime”. This guarded
position reflected the synthesis of different considerations. For the United
States, the extension of the talks to involve alliance members already
constituted a radical step. As well, though the Warsaw Treaty was clearly
faltering, the alliance divide in Europe was still in place, and its imminent
collapse was hardly foreseeable. Finally, NATO countries were of the view
36
that neutral countries would not be in favour of joining Open Skies, due to
the expenses of aerial inspections.
Despite further evidence of Warsaw Treaty decay, the alliance
framework of the talks remained intact at Budapest. As a prelude to events
still to come, many of the Eastern European countries sought to diminish
this constraint
51
so as to further lessen the sway of the Warsaw Treaty.
However, since the discussions ended without result and with little hope of
continuation, the matter of states parties lost its relevance together with
other issues.
The issue of Treaty participation next re-emerged in 1991 under
severely different circumstances. By the middle of that year the Warsaw
Treaty had been formally dissolved and the Soviet Union was coming
undone. The legitimacy of formal differentiation between allied and non-
aligned countries was thus gone, as the former members of the Warsaw
Treaty had become non-aligned. Furthermore, two other complications
now affected the states parties question. One had to do with how to handle
the matter of Soviet successor states, the other related to the quarrel
between Greece and Turkey that had erupted over the participation of
Cyprus in the Treaty. The ensuing debate made it clear that the issue of
Treaty membership had become a delicate matter.
The question of Soviet succession turned politically delicate the day
the Russian Federation took up the seat of the Soviet Union at the talks. At
the Almaty meeting of 21 December 1991 an agreement was achieved
which stipulated that Russia would take the seat of the Soviet Union in
international organizations, including the United Nations. This
arrangement, however, did not extend to on-going negotiations. The
problem stemmed furthermore from the fact that the involvement of
Belarus and Ukraine in the Treaty was seen as essential. Beyond important
political reasons, namely the early and symbolic recognition of Belarussian
and Ukrainian sovereignty, there were also geo-strategic considerations. It
suffices to look at the map of Europe to understand that the airspace of
Belarus and Ukraine represents a nearly indispensable connection between
Russia and the rest of Europe. The absence of these two countries from the
Treaty would have clearly left a crucial gap. The negotiating states were thus
agreed that Belarus and Ukraine too had to be involved in the talks;
however, a complication arose from the fact that Belarus and Ukraine,
contrary to Russia, due to the above-mentioned arrangement, were not
37
parties to the CSCE. As the Open Skies negotiations unfolded outside the
CSCE framework, it would have been possible to admit Belarus and Ukraine
without them being members of the CSCE, save for the opposition of one
party. A procedural way out of the impasse was found by the Canadian
delegation chairing the plenary session, which recommended that the
formal plenary session be suspended and that it continue informally until
the upcoming Prague meeting of the CSCE Council, at which time Belarus
and Ukraine could join the CSCE. Belarus and Ukraine, thus, formally
entered the Open Skies talks in January 1992.
52
The remaining nine Soviet successor states were granted preferential
status with respect to accession. As such, they could either sign the Treaty
before it entered into force and ratify it according to the provisions
provided, or “accede to it at any time by depositing an instrument of
accession” (Art. XVII, para. 3). This certainly represented a privileged
treatment, as the nine states could unilaterally decide their accession
without the need for approval by the other Treaty state parties. Since the
signature of the Treaty, only two of the nine successor states, Georgia and
Kyrgyzstan, have taken advantage of this privilege. This leaves the possibility
of a further seven countries acceding to the Treaty.
The Open Skies Treaty also makes it possible for states other than the
original signatories and the remaining Soviet successor states to join. The
Treaty stipulates that “For six months after entry into force of Treaty”
potential parties “may apply for accession by submitting a written
request”... “The matter shall be considered at the next regular meeting of
the Open Skies Consultative Commission and decided in due course.”
(Art. XVII, para. 4) Two constraints are placed on accession, however. One,
only states belonging to the OSCE may accede to the Treaty according to
this rule.
53
In view of the regional character of Open Skies, this is
understandable. Two, an application for accession to the Treaty is subject
to the approval of the OSCC. The Treaty does not set a deadline for such
an approval; it merely states that an accession request “shall be considered
at the next regular meeting” of the OSCC, which “shall take decisions or
make recommendations by consensus”. This implies both that the OSCC
may decide on an eventual application for accession on its own time, and
that any state already party to the Treaty may block the request. The latter
may not pose any difficulties in most cases, as has already been
demonstrated,
54
but it could conceivably lead to deadlock, as in the case
of the request for accession by Republic of Cyprus, which was vetoed by
38
Turkey. Needless to add that acceding countries may later on also block
further accessions, which perhaps suggests that in practice the sequence in
which new member states join the Treaty may have some consequence for
the possibility of other interested parties to accede at a later date.
In addition to OSCE states, the Open Skies Treaty also makes
allowance for accession by non-European countries. Non-European
countries may request to join the Treaty six months after its entry into force.
Once again, prospective applications require the endorsement of the
OSCC, although, the OSCC is not actually obliged to consider such
applications.
55
During the Open Skies negotiations, Japan expressed its
interest to join the emerging regime. As Open Skies is a particularly valuable
instrument in the regional context there is no reason to assume that there
would be requests for accession from other continents. It is also
conceivable, however, that countries of other regions would opt for
establishing their own regional regimes without attempting to accede the
1992 Treaty on Open Skies. Open Skies may in fact set an example for
other regions, which may follow suit and draft a similar agreement. Japan,
in view of its strategic challenge and technological capabilities, may well be
the initiator of a similar process in its neighbourhood.
In sum, the Open Skies Treaty established a semi-open accession
regime with three different categories of parties in mind: the NATO and
former Warsaw Treaty members including Belarus, Russia and Ukraine,
which had the right to sign the Treaty before it entered into force, the other
nine Soviet successors which enjoyed the possibilities outlined above, and
other states that may eventually want to join the Treaty, namely other OSCE
member states or other non-European countries, whose accession was to be
subject to approval by the OSCC.
In terms of the legal requirements of the Treaty, each member state
party is obliged to fulfill them on its own. To become a member party, each
country must sign and ratify the Treaty and deposit its instrument of
ratification with one of the depositaries. In addition, states parties can form
two sorts of “associations”. They can form groups of parties or they can—as
requested by the Benelux countries—establish a special entity, which to
some extent resembles a group.
The idea of setting up special rules for the Benelux countries was
present since an early stage of the Open Skies talks. Initially, the three
39
countries sought to form a so-called “combined party” under the Treaty.
The NATO position presented at Budapest reflected this intention, as it put
forth the concept of a combined party modeled on the Benelux case: no
more than three parties shall belong to a combined party and the territory
of neither party shall exceed 100,000 square kilometres. Here, a concern
to avoid the (re-)emergence of alliances under the guise of a combined
party is evident. Finally, the formulation gave a certain preference to small
countries, which as will be elaborated later, could be expected to meet
certain difficulties in the course of Treaty implementation.
Despite interest in forming an association of some sorts, the Benelux
countries initially disagreed on whether they preferred to be considered as
a case apart or to be treated as an ordinary “group of parties”. Belgium and
Luxembourg preferred the former, whereas the Netherlands was in favour
of the latter. Eventually, Dutch opinion prevailed so that “the three States
Parties have agreed to be considered a single state party for the purposes of
specified Articles and Annexes, in all other cases, they are considered to be
individual states parties.”
56
Later in the negotiations, the Benelux countries
also suggested that groups of countries might also be able to share the
equipment used for the purposes of the Treaty and to share the burden of
carrying out and receiving observation flights as a means of reducing
expenses.
Although the Benelux countries eventually decided against it, the
Open Skies Treaty does permit states parties to form groups either when
they enter the Treaty or at any time thereafter. During the negotiation of the
Treaty, the French delegation developed a concept according to which
each state party, which holds quotas, can form a group of parties. Group
members can redistribute their active quotas, but not their passive quotas
(Art. III, Section II, para. 2). The rationale for this is clear: since some parties
are more capable of carrying out inspections than others, the formation of
groups of states able to redistribute amongst themselves their active quotas
but not their passive quotas can result in significant economies without
impinging on the basic rights of each state party to be able to inspect any
other party. A group of parties has been formed by the member states of the
Western European Union.
The German delegation developed another approach to the issue of
groups of states during the Open Skies talks. Under the German formula
parties forming a group are allowed to hold active and passive quotas
40
together, thus resembling a single entity. Under the German method, for
instance, an observation flight over any member(s) of the group is counted
as one flight irrespective of the number of group members overflown. The
Russian Federation and Belarus formed a group according to this scheme.
They unified their passive quota and, by implication, their total active
quota, although the distribution of this remained unchanged. As such, all
observation flights conducted by either Russia or Belarus are conducted on
behalf of the group and charged against their total active group quota. On
the other hand, an observing party may decide to overfly one or the other
country or both of them, for that matter. An overflight over the airspace of
both countries would count as one flight against the active quota of the
observing party (Art. III, Section II, para. 3).
Beyond this fundamental difference in the two modalities of dealing
with groups of states parties, there are also some commonalities (see Art. III,
Section II, para. 6). Specifically, the Treaty grants permission to the parties
forming a group to shift their status from one format to another. This means
that parties, which formed a group under the looser format (Art. III, Section
II, para. 2), may decide to deepen cooperation and move to the tighter one
(Art. III, Section II, para. 3). The shift may also happen the other way
around. In addition, under both formats, parties are also granted the right
to withdraw from a group. Interestingly, the Treaty does not provide for the
right of accession to a group. It makes such accession conditional on the
consent of the group when it declares that a group of states parties “may
admit further States Parties which hold quotas”. Since there is reason to
assume that such decisions would be taken by consensus, it means that
admittance to a group is contingent on the agreement of all the members
of the group. All decisions taken under this regulation are to enter into force
no earlier than six months after all the other states parties have been
notified. This requirement guarantees that shifts in the membership of status
of group parties are clear to all other states, which may undertake
appropriate preparations to take account of the new developments.
It is premature to predict how these rules will be applied, although, it
seems unlikely that any additional groups will be formed in the near future.
Instead, it is far more probable that parties will engage in technical
cooperation without actually establishing groups of states parties.
41
2.3.1.2 The Area of Application
Although the term “area of application” is not part of the definitions
outlined in the Treaty, the issue was subject to debate during the Open
Skies negotiations. The most important decision with regard to the scope of
the Treaty’s area of application had in fact been taken prior to the beginning
of negotiations. The US decision to launch Open Skies as a bloc-to-bloc
initiative including all the members of NATO and the Warsaw Treaty to a
large extent predetermined the eventual area of application of the Treaty.
The starting positions of both NATO and the Warsaw Treaty shared the
view that “All territories of the participants in North America and Asia, as
well as in Europe will be included”.
57
The draft of the Soviet Union
prepared for the Ottawa conference stated that the “Open Skies regime will
extend to the territory of all states parties, including the island territories that
belong to them”.
58
This meant that unlike other European arrangements,
such as those on CSBMs or the CFE Treaty, the territory of the United States
and Canada on the one hand and that of the extra-European part of those
successor states of the Soviet Union, which are parties to the Treaty (most
importantly the Russian far East), were going to be subject to Open Skies
observation flights.
59
The importance of this is that the United States
mainland, Canada and Siberia all became subject to a regional arms control
regime for the first time.
Nevertheless, the broad agreement of the two sides on the Treaty area
of application contained one significant difference, namely the Soviet
desire that the Treaty zone of application include the territories of non-
parties hosting military bases of Open Skies states.
60
Evidently, the Soviet
position could hardly be acceptable to the United States, which unlike the
Soviet Union, was not contiguous to its allies and was hence dependent on
extra-territorial bases. Moreover, it would have been practically impossible
to guarantee the consent of some of the non-participating states for a variety
of reasons not necessarily related to European security matters. The
transparency acceptable in Europe at the end of the 1980s was in fact
unacceptable elsewhere. In view of the impracticability of this position,
which may well have been put forth by those inside the USSR who opposed
Open Skies, the idea was gradually abandoned.
Another issue related to the Treaty’s area of application was the
delimitation of observation flights. While the original NATO proposal
contained no provisions in this respect, the Soviet one went into details.
42
Most controversially, the Soviets proposed that areas of a country could be
excluded from observation on the grounds of “national security”.
Furthermore, the Soviets also advanced the notion of temporarily closed
areas, including training areas, launch areas of space objects, etc., for safety
reasons.
61
Later they also demanded that overflights keep a minimum
altitude of 10,000 metres above nuclear power plants, chemical plants and
densely populated areas.
62
Although the Soviet suggestion that certain
areas be excluded from observation found no support even among its allies,
a compromise on special regulation for the overflight of objects such as
nuclear power stations, large chemical factories and some others was
eventually agreed in the final stages of the negotiations at Vienna. The
compromise entailed referring the Open Skies Treaty to the “hazardous
airspace” regulation of the International Civil Aviation Organization (ICAO),
which made the demarcation of prohibited, restricted and dangerous areas
possible on the grounds of flight safety, public safety and environmental
protection. The reference to prohibiting overflight on the basis of national
security was given up de facto. This was apparently the single most
important matter for many.
63
2.3.2 Sensors
Initially the Western states tried to negotiate a sizeable and diverse set
of allowable sensors.
64
In contrast, the Soviet Union would accept only
optical framing and panoramic cameras and SAR with a coarse resolution
of 10 metres. A compromise reached by the US and Soviet Foreign
Ministers, James Baker and Eduard Shevardnadse, in August 1990
established the criterion of all-weather capability. This capability is provided
in principle by radar (SAR) sensors. Such sensors can take images through
cloud cover and at night. However, the finally agreed radar resolution of 3
metres is too coarse to allow the reliable detection of individual objects
such as land vehicles. This made it clear that the backbone of the sensor set
would be optical framing and panoramic cameras with an agreed resolution
of 30 centimetres. The images from such cameras, when analyzed in stereo
pairs, enable the recognition of different types of land vehicles, which are
relevant in a verification and military confidence-building context.
Upon the insistence of Western states, Russia also agreed to include
infrared line scanners with a resolution of 50 centimetres. These sensors can
provide temperature images at day and night. The scanners are obstructed
by clouds but can image through light haze. The sensors were introduced
43
in order to support data taking during winter in northern regions when
illumination by sunlight is short and faint.
65
Video cameras were introduced as a separate class since Russia
objected to the more general category of electro-optical sensors.
66
The
Western states accepted this solution since it kept the door open for a future
introduction of electro-optical sensors. This will become an issue in the
OSCC later on because electro-optical sensors with enhanced capabilities
are in the process of revolutionizing the civilian airborne remote sensing
market.
67
The OSCC can agree by consensus on the subsequent
introduction of additional sensor categories and on improvements of sensor
capabilities (Art. IV, para. 3).
In short, the agreed full sensor set comprised:
• one vertical and two oblique optical framing cameras at 30 centimetres
resolution;
• one optical panorama camera at 30 centimetres resolution;
• video cameras at 30 centimetres resolution;
• a thermal infrared line scanner at 50 centimetres resolution at a radiant
temperature differential of 3 degrees Celsius; and
• a sideward looking synthetic aperture radar (one side only) at 3 metres
resolution capable of recording a strip of up to 25 kilometres within a
sideward corridor of 50 kilometres.
The full sensor capability is to be introduced stepwise. If the observed
state provides the observation aircraft (taxi option), the full sensor set at the
nominal resolution has to be operative by the beginning of the fourth year
after the Treaty’s entry into force (i.e., 1 January 2006). If the observing state
provides the aircraft, the SAR resolution shall not be worse than 6 metres.
In the first three years after entry into force a reduced capability is allowed
(incomplete sensor set, higher than nominal resolution). Infrared line
scanners can only be used during the first three years if agreed by both the
observing and observed parties (Art. XVIII).
Resolution was defined in Article II as “the minimum distance on the
ground between two closely located objects distinguishable as separate
objects.” This is the traditional photogrammetric definition of ground
resolved distance.
68
The negotiators encountered considerable difficulties
when trying to specify procedures for verifying the sensor resolution.
44
Existing photogrammetric practices were discussed. However, the fear that
a sensor might underpass the nominal resolution complicated the
negotiations. In consequence, the discussions were not completed before
the Treaty’s signature in March 1992. The issue was handed over to the
OSCC and its sensor commission for completion.
The OSCC took two relevant decisions (Nr. 3 on 29 June 1992 and Nr.
7 on 10 December 1992). To the surprise of many observers the OSCC
adopted a different definition of sensor resolution, which—for optical
cameras—corresponds to approximately 60 centimetres ground resolved
distance.
69
This performance still allows detection of vehicles and general
identification of the vehicle type (e.g., car, truck, tank).
70
Two more regulations on sensors are worth mentioning. First, no data
taking is permitted during transit flights. Hence, each sensor has to have a
cover or other means which prevents data taking during transit flights.
Sensor covers or inhibiting devices must be removable or operable only
from outside the aircraft (Art. IV). Second, real-time data processing and
display is permitted for checking sensor functionality in flight, whereas real-
time data transmission via satellite or directly to a ground station is strictly
forbidden. This regulation is meant to prevent misuse of data taking for
targeting quasi real-time attacks. The time flow of an observation mission
leaves at least one day between the taking of data and the handing over of
the developed film or other image data to the observing party.
2.3.3 Flight Quotas
According to the Treaty each state party has the right to conduct a
certain number of observation flights using unarmed fixed-wing aircraft
(active quota) and is obliged to accept observation flights by other state
parties over its territory (passive quota).
71
No state party is obliged to accept
more than one observation flight within 96 hours (Art. II, para. 1). The total
active quota of a party can equal its passive quota but the possibilities for
overflight are limited by the passive quotas of parties to be overflown. The
allocation of passive quotas amongst the different state parties is given in
Table 2.1.
72
As an example, Germany and Italy have to receive up to 12
overflights each per year, while the Russia-Belarus group and the United
States are obliged to receive up to 42 overflights each. The active quota of a
state is usually equal to its passive quota. For the first three years of Treaty
implementation (i.e., 2002-2005) all quotas are capped at 75%.
45
Table 2.1: Flight Quotas and Maximum Distances
73
The table includes also states which acceded to the Treaty after its entry
into force (status as of 1 April 2004).
Notes:
a
These numbers apply to the period of full implementation. Signatories are only
obliged to receive 75% of their passive quota in the first three years of the
operation of the Treaty.
b
As of 1 April 2004. Some of the distance values set in the Treaty where changed
by Decisions of the OSCC.
c
Belgium, the Netherlands and Luxembourg.
d
Including Greenland.
e
At Treaty signature Czechoslovakia was still one state.
Country Quotas
a
Maximum Flight distance (km)
b
Russia-Belarus Group
United States
Canada
France
Germany
Italy
Turkey
Ukraine
United Kingdom
Norway
Sweden
Benelux Group
c
Denmark
d
Poland
Romania
Finland
Bosnia and Herzegovina
Bulgaria
Czech & Slovak Republic
e
Georgia
Greece
Hungary
Iceland
Latvia
Spain
Portugal
Kyrgyzstan
42
42
12
12
12
12
12
12
12
7
7
6
6
6
6
5
4
4
4
4
4
4
4
4
4
2
-
5,000-7,200
3,750-4,900
5,100-6,150
2,078-2,715
1,300
2,015
1,500
2,100
1,500
1,700
1,700
945
250-5,800
d
1,400
900
1,400
t.b.d.
660
800
1,255
910-1,170
860
1,500
800
750-2,000
1,030-1,700
-
46
The allocation of active quota entitlements proved difficult in the
negotiations of the Treaty because almost every party wanted to overfly
Russia and the Ukraine. Finally the parties agreed to an initial distribution
of active quotas, shown in Table 2.2, which is considerably below the 75%
mark. In case of Russia and Belarus only 28 out of 31 possible flights were
assigned. Two flights were reserved for Sweden and one for Finland in view
of their anticipated accession to the Treaty after its entry into force. This
quota could also be used for flights over newly acceding successor republics
of the Soviet Union.
74
The distribution shown in Table 2.2 is applicable for
the first year of the Treaty (1 August 2002-31 December 2003). The
distribution for 2004 and 2005 has to be negotiated (within the 75% limit).
If no new agreement can be reached the distribution of Table 2.2 will
continue to apply. The member states of the WEU have already agreed to
refrain from 5 (in 2004) and 4 (in 2005) flights over Russia, to which they
would be entitled.
2.3.4 Time Sequence of Overflights
The time sequence of overflights follows the procedures developed for
on-site inspections in the CFE Treaty. The party requesting an overflight must
inform the party to be overflown of its intention at least 72 hours before the
arrival of its aircraft at a designated point of entry; it shall make every effort
to avoid using the minimum pre-notification period over weekends
(Art. II, Section I). The party to be overflown must acknowledge receipt
within 24 hours and state whether it will allow the overflying country to
bring its own aircraft or would exercise its right to provide an aircraft. In
deviation from this short notice procedure, Appendix H allows for advance
notice of intended flights and for advance coordination of the sequence of
flights.
75
In that case the observed party can opt for the taxi option well in
advance. After arrival, the aircraft and sensors can be inspected, and the
proposed mission plan is to be handed to the host country.
76
The observed
party has very limited scope for requesting changes to the mission plan. After
acceptance or agreement on eventual changes, the mission plan forms the
basis for working out a flight course, which is prepared by the state that
provides the observation aircraft while following ICAO rules and taking into
account existing national flight regulations. Due to the time sequence, the
observed state has a minimum of 24 hours advance notice between learning
about the mission plan and the commencement of the observation flight.
This gives sufficient time for hide-away operations of moveable equipment,
but still comprises a certain surprise element.
47
Each overflight may vary in actual flight path and timing. Flight paths
may be unique for every overflight and may follow any course from a
straight line to a zigzag or a loop pattern. The observing party is, however,
restricted from loitering over one point, except on take-off and landing, and
from crossing its own flight path more than once at a certain point. The
amount of time the observing party allocates to execute the flight plan is
largely at its discretion. Observing parties have a total of 96 hours from the
earlier of notified arrival time or actual arrival time to complete their
observation overflight. Rest and refueling stops can be made as identified in
the mission plan (any airfield is eligible to be designated as a weather or
emergency alternative). Thus, the actual time spent collecting data during
an Open Skies observation overflight will vary with each occurrence.
77
In
practice, 72 hours or less will be available for the observation flight
including stopovers.
A typical observation event might proceed as follows:
Day 1: - arrival
- point of entry procedure
- pre-flight inspection
Day 2: - demonstration flight, if necessary
78
- handing over of the mission plan
- negotiation and agreement on the mission plan
Day 3: - observation flight
Day 4: - continuation of observation flight, if required
- processing of film
- duplication of film and magnetic tapes
Day 5: - continuation of film processing and duplication, if necessary
- completion of mission report
- departure
Figure 2.1 gives an accurate account of the time sequences to be
observed according to Article VI of the Treaty (Figure 1.2 of the Sensor
Guidance Document). Article VI of the Treaty rules that the observing state
can include the personnel of other states parties in its observation team.
This facilitates the formation of multinational observer teams—in the
cooperative spirit of the Treaty.
48
Table 2.2: Initial Distribution of Active Quotas
Observing
State Party
Observed State Party
Benelux v 1 1
Bulgaria v 1 1 1
Canada v 1 1 2
1
2
Czechoslovakia v 1 1
Denmark v 1 1
France v 1 3
Georgia
5
v
Germany v 3 1
Greece 1 v 1
Hungary v 1 1
Iceland v
Italy 1 v 2
1
4
Kyrgyzstan
5
v
Norway v 1 2
Benelux
Czechoslovakia
Hungary
Bulgaria
Canada
Denmark
France
Georgia
Germany
Greece
Iceland
Italy
Kyrgyzstan
Norway
Poland
Portugal
Romania
Russia*
Spain
Turkey
Ukraine
UK
USA
49
Note: The active quota of the former Czechoslovakia has been divided as follows: one flight by the Czech Republic over Ger-
many and one flight of Slovakia over Ukraine. It was also agreed that Canada, Spain and the Ukraine would have one quota
each over both the Czech Republic and the Slovak Republic. Georgia, Iceland and Portugal did not ask for assignment of an
active quota during the negotiations, nor was any quota for overflying Georgia, Iceland and Portugal assigned to any party.
Source: P. Jones, “Open Skies: Events in 1992”, in J. B. Poole and R. Guthrie (eds), Verification Report 1993, VERTIC, London
and New York: The Apex Press, 1993, p. 152.
Poland 1 v 1 1
Portugal v
Romania 1 1 1 v 1
Russia* 222331 22 v234
Spain 1 v
Turkey 1 2 v
2
3
Ukraine 1 1 1 1 2 v
UK 31v
USA 8
1
1
v
BX BU CA CZ DK FR GE GR HU IC IT NO PO PL RO RU SP TY UE UK US
Active quota 6 412 4 612 12 4 4 412 7 6 2 642 412121242
(75% quota) 439349 93339 541430399930
* Includes Belarus;
1
Shared with Canada;
2
Shared with USA;
3
One of which is shared with Italy;
4
Shared with Turkey;
5
No quota assigned.
50
A
BCD
EF
GH
I
J
KL
M
ON P
≥ 72 hrs
demo flight
≥ 24 hrs
≤ 96 hrs
≤ 8 hrs
≤ 4 hrs
≤ 24 hrs
≤ 7 days
≤ 10 days
≤ 3 days
≤ 24 hrs
≤ 4 hrs
observ. flight
Figure 2.1: Open Skies observation flight line
A: Notify intent to conduct the observation flight.
B: Confirm receipt of the notification of intent to conduct the
observation flight.
C: Arrive at point of entry (POE).
C-D: Arrival procedures, inspect covers, get latest weather forecast and
information on air traffic and air navigation safety.
D: Conduct pre-flight inspection (to be completed not later than 4 hours
before observation flight).
E-F: Time period to conduct demonstration flight. Duration of demo
flight not to exceed two hours over the calibration target. If demo flight
requested by observed party an additional 24 hours (beyond the 96 hour
period) is provided to the observing party to complete observation flight.
F-G: Transfer to the Open Skies airfield (if different from the POE).
G: Submit mission plan, at any time after arrival (C).
H: Complete examination of the mission plan and modifications
thereto.
I: Agree on mission plan or cancel mission.
J: Begin observation flight.
K: Complete observation flight.
K-L: Transfer to the point of exit (if different from the last Open
Skies airfield). Time for film processing by the observed party shall
not exceed 3 days after the arrival of the aircraft at the point of
exit.
K-M: Prepare and sign mission report.
M: Depart from the point of exit. Time for film processing by the
observing party shall not exceed 10 days after departure from the
territory of the observed party.
N: Distribute mission report to all states parties.
O: Film processing time if film is processed as arranged by the
observed state party.
P: Film processing time if film is processed as arranged by
observing state party.
51
2.3.5 Technical Requirements of Overflights: Whose Aircraft to Use?
Upon insistence by Russia, each state to be overflown has the choice
of either receiving the aircraft of the observing state or of providing an
aircraft with full sensor equipment of its own for the use of the observing
state. This provision reflects the initial Soviet hesitation about fully opening
its airspace to foreigners. A compromise worked out by James Baker and
Eduard Shevardnadze at Irkutsk in early October 1990 paved the way for
the future success of the negotiations. The Soviets conceded to open their
entire territory, including prohibited airspace, for observation flights. In
return, the observed state party was granted the right to exercise the so-
called taxi option, that is, the right to provide its own aircraft for an
observation flight.
79
The taxi option was decisive in obtaining the consent
of the Soviet military, which feared that other states would install hidden
prohibited sensors on their aircraft. Since Russia insisted on the taxi option,
the US and others insisted that a taxi aircraft would have to be equipped
with all permitted sensors operating at Treaty resolution. Demonstration of
that capability became an issue. Should the observed state decide not to
provide a taxi aircraft of its own, the observing state can also use the aircraft
of a third state party. This regulation is important for states which do not
operate an Open Skies aircraft of their own. The regulation is also
applicable for groups of states parties.
2.3.6 Territorial Restrictions
One important provision of the Treaty (Art. VI, Section II, para. 2) is
that the full territory of each state party may be overflown, except for a 10-
kilometre zone next to the state’s borders with non-state parties.
80
This
implies that the vast territories of North America and Siberia, which were
hitherto off limits to inspections under the CFE Treaty, are now accessible
to Open Skies flights. For example, sensitive sites like the White House in
Washington, the Space Shuttle launch sites at Cape Canaveral and Russian
missile silos have been overflown in trial flights. As another example, a test
flight over Frankfurt airport resulted in a shutdown of all other flight
operations for some one and a half hours. This created a major traffic jam
and probably will not be repeated. But it was a point in case. Only sites that
affect the safety of flight could be excluded. Each flight over a particular
country, however, will be restricted to a maximum flight distance, as
specified in Table 2.1.
52
Open Skies observation flights have priority over any regular civilian
and military air traffic. According to ICAO rules, three exceptions are
permitted: missions for aircraft emergency relief, air defence and medical
evacuation.
2.3.7 Dissemination of Information and Openness of Data
According to Article VI, Section I, paragraph 21 the mission report from
each observation flight shall be accessible to all member states. Based on
this report the states parties may decide to request a copy of the flight image
data (at an agreed cost). It was the Soviet Union that from the beginning
insisted on “equality in acquiring and in access to information”. Although
this proposal was not greeted with enthusiasm by all, after considerable
maneuvering the parties finally agreed on the sharing of image data. As
such, after an Open Skies observation flight any states party is entitled to
request to receive a first generation copy of the image data taken.
81
In the
case of photographic film, the image data can be easily analyzed through
inspection by image analysts. Digitalized and SAR images data, however,
require a greater technological effort and sensor knowledge to interpret.
The option and right of data sharing is one of the most innovative
features of the Treaty, which emphasizes its cooperative character.
However, although unclassified, data will be available only to state agencies
for purposes in accord with the intentions of the Treaty. Hence, there are
limits to openness. These limits, which date back to 1990, appear
somewhat outdated in view of the new generation of commercial high-
resolution photo-satellites, which provide black and white pictures with 1-
metre or even 0.6 metres ground (pixel) resolution worldwide.
82
Notes
1
Forty-five years after it had been tabled the proposal was still on its way
to becoming an active treaty.
2
There is no record of any major preparation of the proposal or any
indication that it had been thoroughly considered by the US
government prior to its announcement.
53
3
“Statement by President Eisenhower”, in The Geneva Conference of
Heads of Government, July 18-23, 1955, Washington, DC: US
Government Printing Office, 1955, p. 21.
4
“Radio-Television Address by President Eisenhower”, Washington, July
25, 1955, in The Geneva Conference of Heads of Government, July 18-
23, 1955, p. 86.
5
“News Conference Statement by Secretary of State Dulles”,
Washington, 26 July 1955, in The Geneva Conference of Heads of
Government, July 18-23, 1955, p. 87.
6
A. S. Krass, Verification: How Much Is Enough?, London and
Philadelphia: Taylor and Francis, 1985, p. 118.
7
J. L. Gaddis, Strategies of Containment: A Critical Appraisal of Postwar
American National Security Policy, Oxford: Oxford University Press,
1982, pp. 156-57.
8
The authors gratefully acknowledge the support of Mr James
Marquardt in developing this idea.
9
A. S. Krass, p. 118.
10
“Itogi Zhenevskovo Soobsheniya Glav Pravitelstv Cheteryekh Derzhav:
Doklad Predsedatelya Soveta Ministrov SSSR tovarishcha N. A.
Bulganina 4 avgusta 1955 na tretey sessii Verhovnovo Soveta SSSR”,
Pravda, 5 August 1955, p. 3.
11
US documents show that Eisenhower had been informed about the
coming breakthrough in surveillance capabilities by a high-altitude
aircraft (U-2) that would “open” skies with or without Soviet
acceptance, and gave approval to the U-2 programme. When the
Soviet Union shot down a U-2 aircraft in spring 1960 near Sverdlovsk
(now Ekaterinburg) the first phase of Open Skies history was over. The
US denied the existence of the U-2 until faced with solid Soviet
evidence to the contrary.
12
J. A. Hawes, Open Skies: Beyond “Vancouver to Vladivostok”,
Occasional Paper No. 10, Washington, DC: The Henry L. Stimson
Center, December 1992, p. 2.
13
J. Borawski, From the Atlantic to the Urals: Negotiating Arms Control at
the Stockholm Conference, Washington: Pergamon-Brassey’s, 1988,
p. 21.
14
Mentioned by J. Borawski, op. cit., p. 68.
15
It was learned much later that Akhromeyev adamantly opposed on-site
inspection.
16
Quoted by J. Borawski, op. cit., p. 98.
54
17
Document of the Stockholm Conference on Confidence- and Security-
Building Measures and Disarmament in Europe Convened in
Accordance with the relevant Provisions of the Concluding Document
of the Madrid Meeting of the Conference on Security and Co-operation
in Europe, para. 89, http://www.osce.org/docs/english/1973-1990/
csbms1/stock86e.htm.
18
Ibid., para. 90.
19
Ibid., para. 91.
20
J. Clark, “Foreword: Open Skies”, in Slack, M. and H. Chestnutt (eds),
Open Skies—Technical, Organizational, Operational, Legal and
Political Aspects, Toronto: Centre for International and Strategic
Studies, York University, 1990, pp. vi-vii.
21
G. Bush, “Notes for an Address to the Graduating Class of Texas A&M
University”, 12 May 1989, quoted in P. Jones, “Open Skies: A Review
of Events at Ottawa and Budapest”, J. B. Poole (ed.), Verification Report
1991, VERTIC, London and New York: The Apex Press, 1991, p. 73.
22
Ibid., p. 73.
23
“Declaration of the Heads of State and Government Participating in
the Meeting of the North Atlantic Council in Brussels, 29-30 May
1989”, para.18, in NATO Final Communiques 1986-1990, Brussels:
NATO Office of Information and Press, 1990, p. 35.
24
The US mainland had heretofore been a “sanctuary”. On-site
inspections under the Stockholm regime did not extend to extra-
European territories and the still pending CFE Treaty, which was being
negotiated concomitantly, would not allow them either. The only—not
multi-, but bilateral—regime that permitted on-site inspections, and
Soviet inspectors, to enter the territory of the United States was the
Intermediate Nuclear Forces (INF) Treaty. Under the INF Treaty access
by Soviet inspectors was confined to a very limited number of areas,
however.
25
J. B. Tucker, “Back to the Future: The Open Skies Talks”, Arms Control
Today, Vol. 20, October 1990, p. 21.
26
“Annex to the Communique of the North Atlantic Council meeting in
Ministerial Session on 14
th
and 15
th
December 1989—Open Skies
Basic Elements”, in NATO Final Communiques 1986-1990, pp. 128-
32. See Appendix G.
27
See section 2.3 below.
28
The French opposition to bloc-to-bloc arms control was well known.
France, as a “semi-member” of the Atlantic Alliance since 1966, had
expressed its disagreement with arms control negotiations based on
55
two blocs, NATO and the Warsaw Treaty, since the beginning of
conventional arms control in the early 1970s.
29
J. B. Tucker, op. cit., p. 21.
30
Arms Control Reporter, Cambridge, MA.: Institute for Defence &
Disarmament Studies, 1990, p. 409.B.3.
31
It is a fact that the Soviet Union was practically never obliged to seek
compromise in the Warsaw Treaty on arms control until the early
1980s. Major compromise efforts were necessary, however, at the end
of the 1980s and the beginning of the 1990s, the period between the
de facto and the de jure disappearance of the Warsaw Treaty.
32
According to our impression this was the case of the CFE Treaty signed
in November 1990.
33
Canada and Hungary, in particular, played a pioneering role as
facilitators of the process. Apart from hosting the first two Open Skies
conferences they performed two joint demonstration flights over
Hungary and Canada, using Canadian aircraft (4-6 January 1990 and
15-16 January 1992, respectively). Canada commissioned a number of
technical background papers for the benefit of all participating states
(see, e.g., A. V. Banner, A. J. Young, K. W. Hall, Aerial Reconnaissance
for Verification of Arms Limitation Agreements, United Nations Institute
for Disarmament Research (UNIDIR), Geneva: United Nations, 1990).
Canada, as was mentioned earlier, also played a “midwife” role at the
very beginning of the Open Skies process.
34
As the matter is specifically related to sensor performance it is
addressed in section 2.3 below.
35
S. Koulik and R. Kokoski, Conventional Arms Control—Perspectives on
Verification, SIPRI: Oxford University Press, 1994, pp. 174-75. This
latter issue, has persisted in a somewhat modified form ever since.
36
As one scholar has put it: “Events in Europe between May 1990 and
the summer of 1991 fundamentally changed the Open Skies dynamic
but in a very complex manner. While it was clear that NATO no longer
faced the same threat from the USSR, the failure to obtain an aerial
inspection regime in the CFE Treaty and the Soviet decision to move
large numbers of forces and CFE treaty-limited equipment out of the
‘Atlantic to the Urals’ zone, made an Open Skies agreement appear
more urgent to many in the Alliance. As a result, it became possible for
NATO countries to offer serious concessions …” See R. J. Lysyshyn,
“Open Skies Ahead”, NATO Review, Vol. 40, February 1992, p. 24.
37
For all these matters see section 2.3 below.
56
38
L. W. Veenendaal, “Conventional Stability in Europe in 1991:
Problems and Solutions”, NATO Review, Vol. 39, August 1991, p. 21.
39
“Statement of the Representative of the Union of Soviet Socialist
Republics in the Joint Consultative Group, Vienna, 14 June 1991”,
point 3, in Treaty on Conventional Armed Forces in Europe and Related
Documents, The Hague: The Netherlands Ministry of Foreign Affairs,
May 1996, pp. 182-83.
40
P. Jones, “Open Skies: A Review of Events at Ottawa and Budapest”,
op., cit., p. 80.
41
Group of 16 (NATO), “Draft Protocol on Inspections Relating to the
Treaty Between the Parties Art. XII”, in BASIC Reports from Vienna, No.
7, 11 April 1990, p. 11.
42
For more details see section 8.1 below.
43
Treaty on Conventional Armed Forces in Europe, Art. XIV, para. 6 in
Treaty on Conventional Armed Forces in Europe and Related
Documents, p. 35.
44
R. Hartmann and W. Heydrich, Der Vertrag über den Offenen Himmel,
Baden-Baden: Nomos Verlagsgesellschaft, 2000, p. 19.
45
R. J. Lysyshyn, op. cit., p. 24.
46
P. Jones, “Open Skies: Events in 1992”, op. cit., pp. 146-47.
47
For more details see the book (in German) of R. Hartmann and
W. Heydrich. R. Hartmann was the head of the German delegation at
the negotiations.
48
The full text of the Treaty can be found at http://
www.osmpf.wpafb.af.mil.
49
For this reason, several terms are clarified in the part where the
respective matters are addressed in detail.
50
See it as Appendix G in this volume.
51
See, for example, the opening speech of the Hungarian Foreign
Minister, which emphasized the importance of the “national
approach” in order to go beyond the “bloc approach”.
52
P. Jones, “Open Skies: Events in 1992”, op. cit., p. 146.
53
This constraint is no longer relevant for the remaining Soviet successor
states, since by 24 March 1992, the day the Open Skies Treaty was
signed at Helsinki, all Soviet successor states had joined the CSCE (now
OSCE).
54
See section 5.6 below for the details of the accession process.
55
According to the Treaty, the OSCC “may consider the accession”
(Art. XVII, para. 5) of non-European states. Of course, the possibility to
consider does not imply an obligation to do so.
57
56
Treaty on Open Skies: Message from the President of the United States
Transmitting the Treaty on Open Skies with Twelve Annexes, Signed at
Helsinki on March 24, 1992, Washington, DC: US Government
Printing Office, 1992, p. 165.
57
Open Skies “Basic Elements” Document, Appendix G in this volume.
58
Osnovnye polozheniya dogovorennosti o rezhime “otkrytovo nebo”,
Art. IV, para. 2.
59
The possibility that the Open Skies Treaty would be applicable to extra-
European territories of European states has not played any role. It was
evident that this would not happen and would not be requested.
France, a country that would have been affected by such a broad
interpretation did not even raise the matter. The only country that
reflected upon it at a plenary meeting a week before Treaty signature
was the United Kingdom. The UK declared that the Treaty would not
apply to Hong Kong in the light of its transfer to the People’s Republic
of China in 1997. See R. Hartmann and W. Heydrich, op. cit., p. 38.
For the German translation of the text of the UK declaration see ibid.,
p. 551.
60
The initial Soviet proposal stated the following: “Bearing in mind
military activity beyond national territory those states parties of the
regime, which have military bases abroad would if possible already
during the process of negotiation include the consent of those third
parties, which do not participate in the regime”. Osnovnye
polozhenia… Art. IV, para. 2.
61
Ibid., Art. XI.
62
J. B. Tucker, op. cit., p. 22.
63
The importance of this matter can best be illustrated by that the
“Article-by-Article analysis of the treaty on Open Skies” submitted by
the US Department of State to the Senate. It mentions four times that
“the term does not permit the closure of areas for national security
purposes”. See Treaty on Open Skies: Message from the President of the
United States Transmitting the Treaty on Open Skies, with Twelve
Annexes, Signed at Helsinki on March 24, 1992, Washington, DC: US
Government Printing Office, 1992, pp. 172-73.
64
For example, in the Budapest draft Treaty of May 1990 the Western
states had proposed the following sensor types: air sampling devices,
optical cameras, electro-optical cameras, infrared sensors including
video, gravity metres, magnetometres, forward looking infrared
sensors, infrared line scanners, multispectral imaging scanners,
spectrometres and SAR.
58
65
R. Hartmann and W. Heydrich, op. cit., p. 49.
66
Electro-optical sensors record radiation by many small radiation
sensitive solid-state elements (like photo cells). The image information
is rendered in digital format. The reason behind the Russian objection
was its inferiority in sensor technology vis-à-vis the West. While US
reconnaissance and civilian satellites were equipped with electro-
optical sensors already in the 1970s, the Soviet Union and then Russia
were still using optical film cameras on most of their imaging satellites
in the 1990s.
67
In 2000 new commercial electro-optical cameras were presented,
which feature both stereo and multispectral (colour) capability at
resolutions between 10-100 centimetres.
68
A ground resolved distance of 30 centimetres was envisaged at the
Vienna talks on Mutual and Balanced Force Reductions (MBFR) in
1984/85 as an adequate resolution limit for aerial inspections.
Lt.Col. Lars Olof Johansson, Stockholm, private communication.
69
The ground resolution of 30 centimetres according to Decision 3
corresponds to the width of an image element (pixel) of an electro-
optical sensor. This width is usually quoted as the resolution property
of electro-optic satellite sensors. Apparently, the negotiators—when
drafting the resolution definition of Article II—already had a coarser
resolution in mind (as specified in Decision III). F. Badstöber and
P. Harandt, IABG, D 85521-Ottobrunn, private communication, 2002.
F. Badstöber was a member of the German Open Skies delegation
1990-92. P. Harandt was technical advisor in the Sensor Working
Group from 1992 to 1994.
70
The issue of resolution and its certification is further discussed in
section 4.4.
71
This text reflects the wording of the Treaty. According to the Treaty, a
state is obliged to open its territory to overflights. This is reflected in the
passive quota. The number of passive quota for each state party is
specified in the Treaty. The active quota may well be the same but it
has to be set by the OSCC yearly (Art. III, Section I, para. 7).
72
P. Jones, “Open Skies: Events in 1992”, op. cit., pp. 146-47.
73
Table 2.1 gives the maximum flight distance of observation flights
counted from the agreed Open Skies airports (Appendix A of the
Treaty). After the Treaty’s entry into force in 2002 the Open Skies
airfields and the resulting maximum flight distances were revised and
some changes were introduced.
74
R. Hartmann and W. Heydrich, op. cit., p. 97.
59
75
The sequence of flights has to be notified by the observing state by
15 November, 15 February, 15 May and 15 August, respectively for the
following quarter of the year. The observed state has to state within
seven days for which of the flights it wants to exert its right of providing
a “taxi-aircraft”.
76
A mission plan details the requested route, altitude and speed to be
flown by the observing party. Coordination with any other state is not
required, but it is recommended if the route of flight passes close to the
border of any other state. Flying within ten kilometres of the border of
a state non-party to the Treaty is prohibited. The final mission plan
must be accepted and signed by senior officials of both parties and this
is usually done after the mission has been coordinated with Air Traffic
Control Services. The general rule is that Open Skies aircraft fly the
agreed mission plan as long as safety of flight is not compromised.
77
M. Heric, C. Lucas and C. Devine, “The Open Skies Treaty: Qualitative
Utility, Evaluations of Aircraft Reconnaissance and Commercial
Satellite Imagery”, Photogrammetric Engineering & Remote Sensing,
Vol. LXII, No. 3, 1996, pp. 279-84.
78
A demonstration flight of maximum 2 hours over a calibration target
can be requested in order to verify the sensor resolution (Appendix F,
III). A demonstration flight will not occur very often; if the aircraft to be
inspected has successfully passed a certification and no modification to
the aircraft or to the sensors has taken place since, a demonstration
flight will not be necessary.
79
R. Hartmann, “Treaty on Open Skies, Historical Overview”, address to
an Information Seminar on the Open Skies Treaty, Vienna, OSCC DEL/
21/01, 1 October 2001.
80
Since the Treaty allows the use of side looking cameras, it is still
possible to observe areas of dangerous airspace and border areas,
without directly overflying them.
81
Radar data can be exchanged either as raw data or as processed image
data.
82
See section 9.1.
60
61
CHAPTER 3
THE OPEN SKIES TREATY POST-SIGNATURE
Pál Dunay and Hartwig Spitzer
3.1 THE RATIFICATION PROCESS
To enter into force, the Open Skies Treaty required that it be ratified
by at least 20 countries, including those with large passive (and hence
active) quotas and the two depositories (Art. XVII, para. 2). In practice this
meant that the Treaty could not enter into force unless ratified by the
following countries: Canada, France, Germany, Hungary, Italy, Russia and
Belarus (the latter two as a group of parties), Turkey, Ukraine, the United
Kingdom, and the United States.
1
Most signatories found ratification
unobjectionable. The number of instruments of ratification deposited
reached 22 by mid-1995 (see Table 3.1). Thus, just over three years after
the Treaty’s signature the only question remaining was whether the three
Slavic successor states of the Soviet Union, namely Belarus, Russia and
Ukraine, would ratify the Treaty—and if so, when. Ultimately, it took
another six years before the Open Skies Treaty could enter into force due
to problems with the ratification process in Russia and the Ukraine.
After two failed attempts, Ukraine eventually ratified the Treaty on 2
March 2000. Although some Ukrainian officials had reservations about
Open Skies out of traditional fears such as espionage, essentially these
reservations were minor. The two previous attempts to ratify the Treaty in
1996 and 1998 had failed due to the poor planning of the vote or to
particular concerns.
2
Ukraine was worried about the cost of preparing its
airfields to host observation flights and fretted over not being able to fully
use its active quota because of the high costs of observation flights. Whereas
the former concern was founded, the latter was not. No country is obliged
to exhaust its active quota, which is an entitlement. It is up to the country
to decide how many flights it intends to carry out depending on
circumstances such as the assessment of the military importance of
62
observation flights, the state of the international environment and, last but
not least, the resources available for the implementation of the Treaty.
Nevertheless, Ukraine’s delay in ratifying Open Skies was bound to remain
largely inconspicuous so long as Russia and Belarus had not done so either.
3
In case of Russia, important obstacles to the ratification of the Treaty
came from several quarters. To begin with, strong opposition to ratification
existed within the Russian military, which vividly remembered the US use
of aerial observation for espionage, and clearly feared more of the same.
4
Second, this resistance within the military was accompanied by the actually
much more important deadlock between the Russian Parliament, the
Duma, and President Boris Yeltsin. This institutional stalemate explains why
the Open Skies Treaty, submitted by President Yeltsin for ratification on 13
September 1994, was actually ratified only on 18 April 2001, well after
Yeltsin had resigned. Finally, these domestic tribulations surrounding
Russia’s ratification of the Treaty were complemented by the fact that the
United States gave little priority to the bringing of Open Skies into force,
Washington being rather more concerned with Russia’s ratification of the
Strategic Arms Reductions Treaty (START), which in turn served to diminish
the political importance of Open Skies in Moscow, and to further reduce its
urgency.
With ratification still pending, Russia’s attitude toward Open Skies
shifted to become far more cooperative after 1997. This was reflected in
Russia’s more frequent participation in trial inspections (see Figure 4.4).
Here, it is interesting to look at the analytical note on the Open Skies Treaty
prepared for the Duma.
5
The document takes up two important points.
First, the volume of information collected on other countries, both directly
through observation flights and indirectly due to access to information
gathered by other states parties and shared among others with Russia. On
this count the document states that the “Treaty entering into force … will
allow Russia to increase its volume of information on the USA and NATO”
… “the additional volume of information, just in the 0.3-0.6 micrometer
spectrum (information which Russia essentially does not possess) will
comprise 6-7% of the total Russian information volume and complement
space observation resources Russia is in a position to ‘obtain’…” In
“summary, we can conclude that the Treaty on Open Skies is advantageous
to Russia, and allows for some compensation of Western superiority in
obtaining information with minimal expenditures.”
6
63
Table 3.1: Status of Open Skies Treaty Ratification as of 1 April 2004
The table includes states that have successfully applied for accession after
entry into force of the Treaty, as discussed in section 5.6.
OPEN SKIES TREATY RATIFICATION
Country Signature Application
for Accession
Approved by
the OSCC
Ratification Deposition of
Instruments
of
Ratification
Entry into
Force
Belgium
Belarus
Bulgaria
Canada
*
Czech Republic
Denmark
France
Georgia
Germany
Greece
Hungary
*
Iceland
Italy
Kyrgyzstan
Luxembourg
Norway
Poland
Portugal
Romania
Russia
Slovak Republic
Spain
The Netherlands
Tur ke y
Ukraine
United Kingdom
United States
Bosnia and
Herzegovina
Croatia
Estonia
Finland
Latvia
Lithuania
Slovenia
Sweden
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
15.12.1998
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
24.03.1992
22.07.2002
22.07.2002
05.05.2003
04.02.2002
22.07.2002
22.07.2002
24.02.2003
04.02.2002
19.05.1995
20.05.2001
01.03.1994
04.06.1992
25.10.1993
19.12.1992
21.07.1993
12.06.1998
03.12.1993
25.08.1993
18.06.1993
15.08.1994
20.09.1994
20.12.1994
18.05.1993
22.03.1995
17.09.1994
16.05.1994
26.05.2001
26.11.1992
25.10.1993
15.01.1994
18.05.1994
02.03.2000
27.10.1993
02.11.1993
17.08.2003
30.11.2002
31.10.2002
04.06.2002
28.06.1995
02.11.2001
15.04.1994
21.07.1992
21.12.1992
21.01.1993
30.07.1993
31.08.1998
27.01.1994
09.09.1993
11.08.1993
25.08.1994
31.10.1994
28.06.1995
14.07.1993
29.05.1995
22.11.1994
27.06.1994
02.11.2001
21.12.1992
18.11.1993
28.06.1995
30.11.1994
20.04.2000
08.12.1993
03.12.1993
21.08.2003
12.12.2002
13.12.2002
28.06.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
01.01.2002
20.10.2003
10.02.2003
11.02.2003
27.08.2002
*
Depository State
64
Second, the attention paid to the issue of cost shows that the Russian
military was well aware of the constraints it faced. The costs of the
implementation of Open Skies, according to the document, could be
significantly reduced through the leasing of Russian observation aircraft to
countries, which did not possess such planes, and the selling of Open Skies
aeronautical, special and ground technical nomenclature externally.
7
Russia, thus, calculated the costs and benefits of the Treaty and ratified it
when political conditions were ripe. Whether Belarus’ ratification was
actually indispensable for the Treaty’s entry into force is open to debate.
Bearing in mind, however, that Russia had formed a group of states parties
with Belarus and that neither country had a passive quota of its own, in
retrospect, Belarus’ ratification of the Treaty was most likely also
indispensable.
8
On 1 January 2002, 60 days after Russia and Belarus
deposited their instruments of ratification the Open Skies Treaty finally
entered into force.
The nearly ten years that lapsed between the Treaty’s signature and its
entry into force did not pass in vain. While diplomatic officials were working
on bringing the Treaty into force military professionals prepared for its
implementation. Their activities encompassed the following areas: the
establishment of operational units dealing with the implementation of
Open Skies,
9
the selection and preparation of national or foreign
observation aircraft, trial certification of observation aircraft and trial
inspections (see Chapter 4).
Most signatories established Open Skies units under the auspices of
their Ministries of Defence, usually as part of their on-site inspection
agencies, which most parties already possessed to deal with verification
under the Vienna CSBM regime and the CFE Treaty.
In the ten years before the Treaty’s entry into force more than 350
international trial inspection missions were carried out. It is interesting to
note that all signatories, except for two (Iceland and Kyrgyzstan),
participated in such flights.
10
Furthermore, several demonstrations were
made in order to show the advantages of Open Skies to countries, which
are not parties to the Treaty, in particular, Bosnia and Herzegovina. These
demonstrations have shown that Open Skies can be used for post-conflict
monitoring and that its modification to encompass conflict and post-conflict
monitoring may well be a variant worth considering.
11
They also showed
that Open Skies could be used for non-military activities such as the
65
monitoring of floods, as on the Oder in 1997, or of the damages caused by
hurricane “Mitch” in Central America in late 1998.
12
In the latter case, the
US OC-135B aircraft took over 15,000 aerial photographs in Honduras,
Nicaragua and Guatemala to follow in detail the evolution of the storm.
In sum, the ten years period that passed between the Treaty’s signature
and its entry into force was used advantageously by the parties to prepare
for implementation and also to explore some new avenues where Open
Skies or the observation methods regulated and used by it, could be
applicable. It remains an open question, however, whether these are going
to be sufficient to sustain interest in Open Skies in the light of the
fundamentally changed security relations in the Euro-Atlantic area.
3.2 ESTABLISHMENT AND ACTIVITY OF THE OSCC:
DECISIONS AND GUIDANCE DOCUMENTS TO THE TREATY
The Open Skies Treaty mandated the establishment of a consultative
body, the Open Skies Consultative Committee (OSCC). This body is
responsible for the reallocation of active quotas on an annual basis. It will
also discuss any proposals for the upgrade of existing sensor types and the
introduction of new sensor categories. As called for in Article X, the OSCC
provides a forum within which disputes related to the Treaty may be
discussed if bilateral talks fail. The OSCC will discuss any technical
questions arising from the accession to the regime of new states. The OSCC
is also the forum to which bodies of the CSCE (now OSCE) or any other
relevant international organization would address requests for extraordinary
observation flights in times of tension.
13
The OSCC is mandated to meet at
least four times a year in Vienna. Its offices are next to the headquarters of
the Organization for Security and Cooperation in Europe in Vienna. The
OSCC has established four working groups around the following themes:
costs, sensors and calibration rules, notification procedures and formats,
flight rules and procedures. Based on the results of the working groups the
OSCC has established from 1992 to 2000 24 legally binding Decisions to
the Treaty. Some of the decisions were further elaborated by a Sensor
Guidance Document.
14
The OSCC and its working groups began their deliberations already on
2 April 1992. A sizeable number of questions related to sensor calibration,
66
aircraft certification and other procedures had to be addressed. The OSCC
also had to resolve the matter of Czech and Slovak flight quotas after the
dissolution of Czechoslovakia. As a result of these discussions several
decisions were taken concerning: (a) how to calculate the minimum
permissible flight altitude (H
min
) when using optical and video cameras; (b)
how to calculate the minimum flight height (H
max
) above ground level at
which each video camera with real-time display and each infrared line-
scanning device installed on an observation aircraft may be operated during
an observation flight; (c) calibration activities; (d) the format in which data
are to be recorded and exchanged on recording media other than
photographic film; and (e) the mandatory time period for sorting and
sharing data recorded during an observation flight. These decisions were
considered important milestones in the technical and procedural
elaboration of the Treaty provisions.
15
The OSCC also held two seminars on the possible use of the Open
Skies regime for environmental monitoring on 3-4 December 1992 and on
11-12 July 1994. The seminars underlined the potential of Open Skies in
the environmental area. In 1995 the work of the OSCC slowed down
somewhat due to outstanding Treaty ratifications, which prevented its
entering into force. Work on drafting a Guidance Document for aircraft and
sensor certification continued, however. After 1997 the activity of the
OSCC slowed down even more. The working groups stopped their
meetings, after settling many but not all pending questions. On the other
hand, after 1996 the negotiating climate within the OSCC improved,
partially due to a more accommodating Russian stance. This allowed the
OSCC to reach agreement on a more flexible certification procedure than
would have been possible theretofore. In particular—as practiced in the
trial certification of 2001—states parties were hence allowed to submit a set
of previously established H
min
-values for photographic cameras, which then
would have to be confirmed during the certification. This procedure
facilitated the certification activities of 2002 considerably. In 2001, after
Russia’s ratification, the OSCC resumed its more high profile activities by
addressing remaining certification questions and accession issues (for details
see Chapter 5).
In summary, the OSCC has proven to be a useful body when it comes
to resolving outstanding technical and procedural questions. It is also a
sounding board for potential future extensions of the Treaty.
67
Notes
1
Canada and Hungary being the two depositories. Canada was equally
a “large quota” country.
2
An account of the dates and the division of the votes in the Narodna
Rada of the two failed attempts to ratify the Treaty is given in Istoriya
Parlamentskikh slukhan’ Dogovoru z Vydkritovo Neba, 2 March 2000.
Interestingly, the report does not mention the reasons why these
ratification attempts failed.
3
As Belarus was widely expected to follow Russia’s example, all
international attention was in fact focused on Russia.
4
The Russian military could point out that Eisenhower’s initial Open
Skies proposal had been closely followed by the U-2 incident. On the
other hand, the advent of satellite technology had by this time largely
displaced the use of aerial surveillance in the collection of intelligence.
5
Council Meeting of the State Duma with Attachments, Excerpts from
the Minutes No. 5, Federal Assembly of the Russian Federation,
10 February 2001.
6
See ibid., pp. 4 and 7. It is interesting to note that Russia, due to the
lower resolution of its satellites, considers that Open Skies provides
particularly valuable information in the 30 to 60 centimetres resolution
range. An US assessment, due to its more developed reconnaissance
technology, would be different.
7
Ibid., p. 6.
8
See R. Hartmann and W. Heydrich, Der Vertrag über den Offenen
Himmel, Baden-Baden: Nomos Verlag, 2000, p. 123.
9
For further details see section 4.1.
10
See E. Britting and H. Spitzer, “The Open Skies Treaty”, in Poole,
J. B. and R. Guthrie (eds), Verification Report 1992, VERTIC, London
and New York: The Apex Press, 1992, pp. 223-38.
11
See sections 7.2 and 8.2.
12
R. Hartmann,“Inkrafttreten des Vertrags über den Offenen Himmel”,
SWP-Aktuell 25, Berlin, December 2001, p. 25. For more details see
F. Korkisch, “Open Skies: Die Entstehung eines Vertrages und die
Darstellung der Interessengegensätze zwischen Staaten”,
Referatsmanuskript, Europäisches Forum Alpbach 2000, p. 11.
13
P. Jones, “Open Skies: Events in 1993”, in Poole, J. B. and R. Guthrie
(eds), Verification Report 1993, VERTIC, London and New York:
Brassey’s, 1993, p. 155.
68
14
The text of Decisions 1-22 and of the full Sensor Guidance Document
can be found at http://www.osmpf.wpafb.mil. A German translation of
Decisions 1-22 and the English version of the Sensor Guidance
Document are reproduced in R. Hartmann and W. Heydrich, op. cit.,
pp. 247-550.
15
Appendix J gives a full list of the Decisions of the OSCC as of April
2003.
69
CHAPTER 4
TECHNICAL PREPARATIONS
FOR TREATY IMPLEMENTATION
Hartwig Spitzer and Rafael Wiemker
4.1 ESTABLISHMENT OF OPERATIONAL UNITS
Subsequent to the Treaty’s signature in March 1992, most states parties
established operational units which are in charge of technical preparations
and actual observation flights. It is the task of such units to prepare and carry
out Treaty-related activities, like observation mission planning, pre-flight
inspections of Open Skies aircraft and sensors, operating imaging sensors
during overflights and escorting foreign teams flying over national territory.
Aircraft, aircrews and mission specialists are usually provided by the armed
forces, to a large extent by the air force of the respective country, whereas
image analysis is carried out by intelligence units.
In most countries the Open Skies units operate under the Ministry of
Defence.
1
In some countries, like Turkey, the units are joint establishments
of the Ministry of Foreign Affairs and the Ministry of Defence. Appendix A
gives the addresses of the arms control and verification centres of the states
parties which are in charge of implementing the Open Skies Treaty. States
with dedicated Open-Skies aircraft and states with larger flight quota
employ sizeable staff for Treaty matters. As an example, the Open Skies
units of the German Verification Center and the Federal Armed Forces
Intelligence Office comprise 38 full time employees (as of March 2003), not
counting translators and other service personnel.
2
It can be said that through continued training and practice a
considerable amount of professional expertise and corps d’esprit has been
built up both within the national units and in areas of cross-national
cooperation.
70
4.2 AIRCRAFT
One important step of preparation for Treaty implementation was the
selection and retrofitting of suitable aircraft for Open Skies applications.
Several criteria had to be satisfied by the states parties in particular:
•type and size;
•range;
• ability to operate at quite different altitudes;
• availability.
Type: According to the Treaty, observation aircraft have to be
unarmed fixed wing types, with sufficient carrying capacity for sensors, as
well as flight crew, mission team and escort team. The aircraft need to have
down-looking windows (camera bays).
Range: The Treaty foresees (maximum) flight distances that range from
660 kilometres for Bulgaria to 7,200 kilometres for the Russian Federation
with Belarus (see Table 2.1). Although refuelling is allowed during an
observation flight, it is advantageous to execute the flight in one go. In
particular Germany opted for a medium to long-range aircraft, a Tupolev
154 M, which could be flown over Siberia with only one refuelling. The
German aircraft was lost in an accident in September 1997 and was not
replaced. The US chose Boeing OC-135 B aircraft of even longer range
(6,500 kilometres), which are well suited for transatlantic flights.
Ability to operate at different altitudes: Due to weather conditions in
the Northern hemisphere an ability to underfly clouds at altitudes of 800 to
1,000 metres is highly desirable. This is no problem for turboprop aircraft,
but extremely fuel-consuming for heavy jet aircraft. On the other hand,
operation altitudes above 5,000 metres and beyond are mandatory for
future flights over crisis areas, in order to stay out of the range of hand-held
anti-aircraft weapons.
Availability: Most States Parties decided to use existing observation
aircraft (Bulgaria, Czech Republic, Hungary, Romania, Russia, Ukraine, the
United Kingdom) or to retrofit existing aircraft for Open Skies use (Germany,
Sweden, the United States). The “Pod Group” (Belgium, Canada, France,
Greece, Italy, Luxembourg, Netherlands, Norway, Portugal, Spain) uses
existing Lockheed C-130 Hercules transport aircraft, which can carry a
sensor container (the “pod”) under one of the wings. Turkey has acquired
one CASA Airtech CN 235 aircraft for Open Skies purposes.
71
Table 4.1 shows a list of existing Open Skies aircraft. The aircraft are
adequate choices. In particular the Boeing, Lockheed and Tupolev types
have ample space for escorts and observers, at the expense of rather high
operation costs. Several states parties decided not to equip an Open Skies
aircraft of their own. Still, each of these countries can participate in the
Open Skies missions by leasing an aircraft from another state party or by
making arrangements with the state party to be overflown. Belgium,
Luxembourg and the Netherlands operate jointly and act as a single state
party (Art. XIV), as discussed in section 2.3.1.1. Kyrgyzstan has acceded to
the Treaty but has not yet ratified it. It did not participate in trial flights.
It should be kept in mind that most sensors are operated at
considerably lower altitudes than the maximum cruising altitude. Fuel
consumption is higher at lower altitudes resulting in reduced range. This is
illustrated by the range values given in Table 4.1 for flight altitudes of 1,000
and 6,000 metres. The table shows also the sensor operation altitudes and
the ground swath covered by existing vertical framing and panoramic
cameras. Some of the parties have more than two sensor configurations and
related operation altitudes. The table gives the minimum and maximum
altitude.
3
The values are subject to change. More details on aircraft are
given in Appendix B.
The costs of purchasing and equipping an Open Skies aircraft and
keeping it in service are considerable, particularly in view of the small active
quota most states parties have. Not to mention that one group of countries
with the highest passive quota, the Russia-Belarus group, is opting—with
some exceptions—to be flown over only by its own aircraft, which will
further limit the use of the observation aircraft of many other states parties.
Due to the rearrangement of security relations in Europe most states parties
are not interested in carrying out observation flights in the airspace of most
other states parties. In case the purposes for which Open Skies observation
flights could be used are not broadened the dilemma of whether or not to
purchase observation aircraft nationally to fly a small quota of missions
should grow further. Consequently, this is an area where cooperation
among the parties may result in all-around gains: reduced national expenses
without incurring any disadvantages. The EU, which has embarked upon the
building of a European Security and Defence Policy, has not addressed this
issue, yet. Addressing it would make perfect sense, however. The several EU
members which do not plan to purchase observation aircraft should
certainly be receptive to a pooling of resources.
4
72
Table 4.1: Open Skies Aircraft
State Type Range at different
flight altitudes (km)
Sensor operation
altitude (m)
Ground swath
(km)
Seats
including crew
1,000m 6,000m Vertical
camera
Panorama
camera
Vertical
camera
Panorama
camera
Bulgaria 1xAntonov 30 1,500 2,650 2,950
3,150
1,200 4.4
4.7
6.6 14
Czech Rep. 1xAntonov 30 1,500 2,650 1,803
2,047
-4.6
5.2
-16
Hungary 1xAntonov 26 1,500 2,650 1,972
2,993
-3.0
4.4
-30
Pod Group Hercules C-130 2,500 5,000 1,210
1,950
3,810
5,050
1.8
2.9
20.9
28.0
20
Romania 1xAntonov 30 1,500 2,650 t.b.d.
t.b.d.
-1.7
1.9
t.b.d.
t.b.d.
24
Russia 3xAntonov 30
1xTu 154
1,500
2,500
2,650
5,000
1,210
3,103
t.b.d.
-
t.b.d.
2.8
5.5
t.b.d.
-
t.b.d.
20
ca.30
Sweden 1xSaab 340 1,500 2,500 1,009
5,215
-1.5
7.9
-21
Turkey 1xCN 235 Casa 2,000 4,000 t.b.d. t.b.d. t.b.d. t.b.d. ca. 25
Ukraine 1xAntonov 30 1,500 2,650 1,073
2,993
-2.5
7.1
-16
UK 1xAndover MK1* 1,300 - - 854 - 12.2 16
USA 2xOC 135 3,000 6,000 1,456
2,104
4,698
11,101
2.2
3.1
9.9
23.4
39
* The UK aircraft was decommissioned in April 2003.
Source: German Verification Center, Geilenkirchen.
73
4.3 SENSORS
The Treaty foresees the use of four types of imaging sensors:
With regard to photographic cameras the Treaty allows for one
panoramic camera, one vertically mounted framing camera and two
obliquely mounted framing cameras. The ground coverage of these
cameras is limited to 50 kilometres on each side of the flight path.
6
Radar
coverage will be limited to a ground swath of 25 kilometres on one side of
the aircraft. The transverse ground distance of this swath from the flight
track can be chosen freely within a sideward corridor of 50 kilometres.
The recording media will be:
(a) black and white film for photographic cameras;
(b) magnetic tape for video cameras;
(c) black and white photographic film or magnetic tape for thermal
infrared sensors; and
(d) magnetic tape for radar.
The agreed-on tape format for the digital data recording is that of an
AMPEX tape machine. It should be noted that the specified resolution of 30
centimetres for photographic cameras is not defined as the standard
photographic resolution but rather as a kind of pixel resolution, as discussed
in section 4.4. Table 4.2 shows the sensors that have been installed on
Open Skies aircraft or in the sensor pod of the Pod Group. Most states are
using vertical framing cameras. The Pod Group, Turkey and the United
States are also using oblique framing cameras. Ukraine has oblique framing
cameras, which are not yet certified. The United Kingdom has relied solely
1. optical panoramic and framing cameras with a ground
resolution of 30 centimetres (panchromatic) at a modulation
(contrast) of the image of 0.4;
2. video cameras with real-time display and a ground resolution
of 30 centimetres (panchromatic);
3. thermal infrared imaging sensors with a ground resolution of
50 centimetres at ΔT=3°K (temperature differential)
5
; and
4. imaging radar (SAR) with ground resolution of 300 centime-
tres.
74
on a panoramic camera with a very wide field of view of 164 degrees. The
radar and infrared sensors of Russia and the US are under development.
More details on sensors and image examples are given in section 6.2 and
Appendix C.
Table 4.2: Open Skies Sensors as of May 2004
7
4.4 THE GROUND RESOLUTION LIMITS
AND
THEIR VERIFICATION
Article IV of the Treaty limits the ground resolution of optical cameras
to 30 centimetres. Ground resolution is defined in Article II of the Treaty as
“... the minimum distance on the ground between two closely located
objects distinguishable as separate objects.” This is a traditional definition.
However, when describing that resolution in Decision 3, the Treaty
Country Vertical
framing
camera
Oblique
framing
camera
Pano-
ramic
camera
Video
camera
Infra-
red line
scanner
Radar
(SAR)
Pod Group:
Belgium, Canada,
France, Greece,
Italy, Luxembourg,
The Netherlands,
Norway, Portugal,
Spain
xleft/rightx x – –
Bulgaria x – x – – –
Czech Republic x – – ? – –
Germany – – x x x –
Hungary x – – x – –
Romania x – – x – –
Russia 2 – planned planned planned 2006
Sweden x – – – planned –
Turkey x left/right x 3 planned ?
Ukraine x – – ? – –
United Kingdom – – x x – –
USA x left/right x x planned planned
75
deviates from traditional photogrammetric practice and specifies that the
“... value of the ground resolution shall be equal to the width of a single bar
in the smallest group of bars [in a calibration target as shown in Figure 4.2]
which can be distinguished as separate bars, in centimetres.” Since ground
resolution is most often explained in terms of Ground Resolved Distance
(GRD), or the sum of the width of a (black) bar and a (white) space in a
calibration target, 30 centimetres ground resolution as per Decision 3 is in
reality 60 centimetres GRD. Figure 4.1 illustrates this situation. Many
participants in Open Skies were disappointed to learn that the potential
image quality will be significantly less than what had been expected.
8
A
detailed discussion on resolution questions is presented in Appendix E.
However, this resolution will still allow for the detection of standard
military vehicles from their dimensions. Photo 4.1 shows as an example an
image of a military site at approximately 100-centimetre pixel resolution (a
factor 3 times worse than the ground resolution as defined in the Treaty).
Military trucks (some with towed artillery) and other vehicles can be
recognized.
Figure 4.1:
(a) Photogrammetric definition of resolution: Two objects of
diameter d can be resolved (given adequate contrast) at a
distance d.
(b) Treaty definition of resolution: Two black bars of width d and
centre distance 2d can be resolved.
76
Photo 4.1: Detail of a military vehicle depot at Tiraspol,
Moldova. The image was taken by the Ikonos-2 satellite with a
ground pixel size of 1 metre, corresponding to a GRD of about 2
metres. Open Skies pictures have superior resolution and
recognition potential. Source: H. Spitzer, University of Hamburg.
For optical cameras, Decision 3 to the Treaty is the foundation for
establishing the flight altitude H
min
at which the cameras achieve exactly a
30-centimetre resolution. H
min
is derived from analysis of aerial images
taken over a calibration target. The target consists of black and white bars
of different width as shown in Figure 4.2. Decision 3 defines H
min
as the
average of several measurements using at least five pictures (n ≥ 5) taken by
the sensor flying over the ground resolution calibration target according to
the following equation:
9
77
One of the main tasks of certification and demonstration flights is to
determine the minimum allowable flight altitude for each of the sensors. It
has been pointed out that the above formula might be insufficient since the
actual ground resolution depends also on atmospheric conditions (mean
visibility and aerosol content).
10
A more recent evaluation claims that the
Where n is the number of images being analyzed;
H
i
is the height of the aircraft, in meters, at the moment that
the target was photographed;
L
a
is the agreed ground resolution of 30 centimetres;
L
i
is the measured ground resolution, in centimetres from
image i;
K
a
is the agreed modulation contrast of 0.4 of the calibration
target at which the ground resolution is defined
corresponding to a contrast
ratio of 2.3:1;
K
i
is the effective modulation contrast (see definition below);
m is the agreed corrected exponent value of m =0.45 as
derived from laboratory examinations of several different
camera lenses;
and
K
i
(normalized contrast, as determined from
image i);
C
E exposure response of the film, or pixel intensity (gray
value), with E
1
and E
2
being measured on the white and
black contrast panel of the calibration target.
i.e.
Δlog E is the difference in the logarithm of the exposures between the
black and the white bars of the calibration target.
21
loglog
log
/1010
21
EE
EE
E
==
−
Δ
1
1
+
−
=
C
C
∑
=
⎥
⎦
⎤
⎢
⎣
⎡
⎥
⎦
⎤
⎢
⎣
⎡
=
n
i
m
i
a
i
a
i
K
K
L
L
H
n
H
1
min
1
78
above equation performs well under set conditions (clear weather). Data
taken by the United States Open Skies team shows that the above equation
does model US camera resolutions.
11
Furthermore, the determination of flight altitude based on air pressure
also is affected by uncertainties. In practice, it was agreed to tolerate a
resolution range of 25 to 35 centimetres.
The determination of H
min
for SAR, video and infrared sensors is
specified in Decisions 7, 14 and 15, respectively and in the Sensor
Guidance Document. Ongoing tests will probably lead to a simplification
and clarification of the certification procedure for infrared line scanners.
4.4.1 Determination of Contrast and Resolution for Optical Cameras
The formula provided by Decision 3 to the Treaty for determining the
proper minimum flight altitude requires two measurements: contrast ratio
and spatial resolution. The necessary measurements are carried out on test
photographs, by means of a ground calibration target. Its design is based on
Appendix D of the Treaty as well as on Decisions 12 and 14 (see also
section 2.6 of the OSCC Sensor Guidance Document). For example, the
German calibration target is spray-painted on collapsible aluminium sheets
(Figure 4.2).
Contrast
The calibration target contains two large rectangular panels, one black
and one white, for contrast measurement (arrows in Figure 4.2). By
means of a microdensitometer or a microphotometer, the density
value on film for both the black and the white panel is measured as the
average of several measurement squares, which are taken across the
black and the white calibration panel. Then the contrast is determined
as the ratio between the average “white” and “black”
values. Determination of the contrast allows the effective modulation
K
i
to be calculated and then inserted into the H
min
calculation. Prior to
the flight a so-called D log E-curve is produced for the film, using a
densitometer to measure the densities of a 21-step densitometric strip
exposed on the film.
12
Figure 4.3 shows an example.
79
Resolution
In terms of the Treaty, the resolution is defined as the smallest width
between a pair of black bars which can just be discerned on a white
background, where “black” and “white” are the same as used for the
contrast measurement. Nineteen pairs of black bars of decreasing
width and separation are grouped on the calibration target (Figure 4.2).
The decision as to which of the bar pairs are just resolved is made on
the basis of the eye appraisal of the analysts assisted by an optical
magnifier directly on the negative that is lit from below.
The calibration target contains 19 pairs of black bars. The separation
between each pair of bars decreases in the pattern of a geometric
series: each separation is smaller than the foregoing by a factor
1/
6
√2 = 0.891. Thus, after a succession of six pairs the separation is
down by one half, after 12 pairs down by one fourth, after 18 pairs
by one eighth. The largest bars are 50 centimetres wide, the smallest
6.25 centimetres.
Figure 4.2: Top: German calibration target from 900 metres. The image
was taken by the RGB video camera (green channel) on the former
German Open Skies aircraft. Bottom: top image after data processing.
The last resolved bar can clearly be recognized.
80
Figure 4.3: Top: Densitometric strip with 21 different grey
values. Bottom: Sample D log E curve for Agfa Pan 200 film.
Such curves are used to determine the optimum film processing
conditions. Source: German Verification Center, Geilenkirchen,
August 2000.
During the certification procedure, the photographic negatives of the
calibration target are visually analyzed by ten inspectors from different
countries. Each analyst casts an opinion on which pair of bars he can just
resolve. Then the highest and the lowest resolution estimate are eliminated
in order to neutralize outliers, and the final result is computed as the
average of the remaining eight resolution estimates.
Contrast and resolution results are then entered into the above H
min
-
formula for calculation of the minimum flight altitude.
81
4.5 TRIAL FLIGHTS AND LESSONS
After the signing of the Treaty in 1992 a growing number of training
and trial flights were performed. These flights had three purposes:
• Training of personnel and test of equipment and procedures on the
national level;
• Bilateral or multilateral (border crossing) trial flights for training and
testing under conditions that corresponded (in most cases) to the
provisions of the Treaty;
• Data gathering, which—although on a trial basis—in effect contributed
to capturing the aims of the Treaty.
National training is mandatory in order to prepare for full participation
and to maintain capabilities and skills. International trial flights are agreed
upon bilaterally or multilaterally on a voluntary basis. Partners of test flights
followed most of the Treaty provisions. Two important exceptions were
made. One, due to the voluntary character of the flights, image data were
just shared between the states involved in a particular flight and not among
all members states. Two, some airspace was excluded for various reasons.
Initial test flights aimed particularly at testing the procedures mandated
by the Treaty. As an example, in October and November 1992 three
aircraft, provided by Denmark, Russia and Canada, tested synthetic-
aperture radars over specially designed targets in Hungary in order to
demonstrate calibration of three very different SAR sensors. This successful
experiment was hailed as “a milestone in technical cooperation among
parties to the Open Skies Treaty” and it was noted that the “monumental
task of negotiating such complicated issues as SAR parameters was a vivid
example of the confidence-building intent of the Treaty at work”.
13
Based
on this experience the OSCC was able to agree on decisions, which
specified such procedures in more detail.
The early test flights revealed also some of the limitations of the Treaty.
One, that it is particularly difficult to underfly (low) cloud covers, when
adhering to the 30-centimetre resolution limit of optical cameras. An
operating flight altitude of 800 to 1,000 metres above ground turned out to
be a lower limit. Two, 3-metre resolution on SAR imagery produces a
limited amount of information, which may be insufficient to justify the high
costs involved.
14
82
In subsequent years the character of test flights changed from an
exploratory to a more routine nature. Figure 4.4 and Appendix D show the
development of the annual bilateral and multilateral trial flights. From 1992
to the end of 2001 some 350 missions were flown involving all member
states except Iceland and Kyrgyzstan. In addition, 42 test flights involving
both non-signatory and signatory states were undertaken as shown in
Table 4.3. Germany played a particularly active role taking part in 38% of
all flights. The lower bars in Figure 4.4 indicate the flights involving the
Russian Federation and Germany, respectively. Russia has shown an
increased involvement in test flights since 1996.
Figure 4.4: Open Skies test flights, including flights that involved
non-signatory states, as of 31 December 2001. Sources: Zentrum
für Verifikationsaufgaben der Bundeswehr, D-52503
Geilenkirchen; SIPRI Yearbooks 1993-1995.
Some of the test flights covered the vast territories in North America
and Siberia, which are not accessible to on-site inspections under the CFE
Treaty or the Vienna Documents. As an example, Figure 4.5 shows the flight
route of a German-Russian flight over Siberia using the German Open Skies
aircraft (9-17 October 1995). Similarly, a Russian test flight to the US made
83
its way from Washington, DC to Alaska and the US part of the Aleutian
Islands (22-30 July 2000).
Table 4.3: Open Skies Flights Involving Non-Signatory States
as of 31 December 2001
The flights were carried out upon bilateral or multilateral agreement
as demonstration or data gathering flights.
Date Mission Observing
State
Observed States Additional
Observers
Observation
Aircraft
14.08-23.08.96 Training
flight for
Finland
Germany Germany TU-154M
12.05-16.05.97 Test flight Germany Finland TU-154M
17.06-19.06.97 Observation
flight
Hungary/
Romania
Bosnia and
Herzegovina
CND, D, F,
GB, US,
KR, YU,
UKR
AN-26
18.08-22.08.97 Test flight Finland Germany TU-154M
25.08-29.08.97 Observation
flight
Germany Bosnia and
Herzegovina
Yes TU-154M
03.11-07.11.97 Observation
flight
Russia, USA Bosnia and
Herzegovina
Yes AN-30
13.07-17.07.98 Observation
flight
Germany,
Russia
Bosnia and
Herzegovina
Yes AN-30
14.09-18.09.98 Test flight Finland UK Andover (?)
04.12-19.12.98 Post-disaster
monitoring
USA Central America
(El Salvador,
Honduras,
Nicaragua)
OC-135B
14.03-20.03.99 Test flight Finland USA OC-135B
23.05-28.05.99 Test flight USA Finland OC-135B
05.07-16.07.99 Test flight Germany,
UK, USA
Estonia, Latvia,
Lithuania
OC-135B
July 1999 Test flight USA, 3 Bal-
tic States
UK OC-135B
12.07-16.07.99 Test flight Russia Finland AN-30
20.09-24.09.99 Test flight Finland Russia AN-30
21.09-23.09.99 Test flight France Bosnia and
Herzegovina
C-130H
25.10-29.10.99 Test flight Slovenia UK Andover
84
Source: Zentrum für Verifikationsaufgaben der Bundeswehr, OH, D-52503
Geilenkirchen.
22.11-26.11.99 Test flight UK Slovenia Andover
14.02-18.02.00 Test flight Finland France Germany C-130H
20.03-24.03.00 Test flight Slovenia USA Germany,
UK
OC-135B
25.03-03.04.00 Static display USA Chile Germany OC-135B
10.04-20.04.00 “Pre-disas-
ter” mission
USA Caribbean States Germany OC-135B
24.04-28.04.00 Test flight Sweden UK Andover
22.05-26.05.00 Test flight Finland Russian AN-30B
07.08-12.08.00 Test flight USA Slovenia Germany OC-135B
09.10-13.10.00 Test flight Sweden Russia Finland,
Latvia
AN-30
09.10-13.10.00 Test flight Czech
Republic
Bosnia and
Herzegovina
Denmark AN-30
13.11-17.11.00 Test flight UK Croatia OC-135B
09.04-13.04.01 Test flight Finland Russia AN-30
23.04-27.04.01 Test flight Norway/
Benelux
Finland C-130H
26.04-02.05.01 Test flight Norway/
Benelux
Sweden C-130H
14.05-18.05.01 Test flight Czech
Republic
Slovenia AN-30
28.05-01.06.01 Test flight Czech
Republic
Bosnia and
Herzegovina
Germany,
Denmark
AN-30
04.06-07.06.01 Test flight Russia Sweden AN-30B
11.06-15.06.01 Test flight Russia Finland AN-30B
11.06-15.06.01 Test flight UK Croatia Andover
CMK1
09.07-13.07.01 Test flight USA Latvia Germany,
UK
OC-135B
21.07-28.07.01 Test flight Sweden USA Germany OC-135B
04.08-07.08.01 Test flight USA Sweden OC-135B
07.08-10.08.01 Test flight USA Finland Germany OC-135B
04.09-07.09.01 Test flight Norway Latvia Greece C-130H
15.10-19.10.01 Test flight Sweden Russia AN-30B
85
Figure 4.5: Flight route of a German-Russian test flight over Siberia using the German Open Skies aircraft (9-17
October 1995). Source: Götz Sperling, Erfahrungen aus der Praxis der militärischen luftgestützten Verifikation,
Vortrag beim ISSC Seminar “Satellitengestützte Erdbeobachtung für Sicherheitspolitik, Wirtschaft und Ökologie”
(25-26.03.1996, Oberpfaffenhofen), unpublished, Zentrum für Verifikationsaufgaben der Bundeswehr, D-52503
Geilenkirchen.
86
It is remarkable to note that the total number of missions (64) flown in
1999 represents 39% of the initial quotas.
15
Thus, in the absence of entry
into force, the prolonged test phase that followed the signing of the Treaty
may well be seen as some form of quasi-implementation below the full
ratification level. Although part of the test operations were of purely
technical nature (including training), the test flights clearly demonstrated the
potential of the Treaty with regard to monitoring for enhancing military
transparency, and to confidence building through cooperative flights,
sharing of equipment and data sharing.
In particular, German flights over Russia and the Ukraine yielded
valuable insights from an arms control and transparency point of view, by
assessing sites which are not covered by other arms control treaties (e.g.,
CFE weapon reduction sites behind the Urals, Black Sea fleet). Russia has
permitted overflights of sensitive sites, such as Intercontinental Ballistic
Missiles (ICBM) silos.
16
Similarly, Russia performed a test flight over US
bases in Germany (7-11 June 1999) during the formation of the US Kosovo
task force. The flight followed on Russian inspections under the Vienna
Documents in Macedonia (7-9 May 1999) and Albania (16-19 May 1999)
as well as on an inspection of the US airbase at Aviano, Italy, under the CFE
Treaty (May 1999). This combined approach has demonstrated the value
and applicability of both treaties under crisis conditions.
The active test practice also led to opportunities for involving non-
signatory states. A total of 42 of such flights were carried out from mid-1996
to December 2001 as shown in Table 4.3. The partner states fell into three
categories:
• Candidates for accession to the Treaty after its entry into force. Here
Croatia, Finland, Slovenia, Sweden and the three Baltic States
participated in joint exercises with signatory states;
• Demonstration flights in areas where Open Skies agreements of their
own might be established (Bosnia and Herzegovina, Chile);
17
• Data gathering flights for pre- and post-disaster monitoring (Central
America, Caribbean Islands, Germany).
18
In summary, the trial implementation of the Treaty can be considered
as a success:
87
• It involved virtually all states parties to the Treaty (except for Iceland
and Kyrgyzstan);
• It proved the functionality of the equipment and of the Treaty
provisions;
• It demonstrated that Treaty objectives could be obtained through
cooperative observation flights;
• It showed that even small states with modest resources could play a
distinctive role;
• As example, Bulgaria has now—due to a prudent investment policy—
the most advanced camera equipment among all the parties to the
Open Skies Treaty;
19
• It created opportunities to demonstrate the Open Skies approach to
countries that are not parties to the Open Skies Treaty, yet.
The trial implementation also revealed a number of weaker points:
• Some of the Treaty provisions (like accurate H
min
determination)
require relatively complicated procedures and extensive training;
• The particular solution of the sensor pod of the Pod Group proved to
be less than optimal.
20
4.6 TRIAL CERTIFICATIONS OF OPEN SKIES AIRCRAFT
Aircraft certification is a major Open Skies event. It is intended to
assure participants that a designated aircraft and sensor combination, an
observation system, can and will perform within the parameters of the
Treaty specified limits. The certification event consists of a ground
examination and of data taking flights over calibration targets with
subsequent analysis of the collected in-flight data. The purpose of the flights
is to establish the minimum sensor operation altitude H
min
for each camera/
lens/film/filter combination, as discussed above (section 4.4).
Under Treaty conditions notification is given to all Open Skies states
parties at least 60 days prior to the scheduled date of the Certification Event.
Beyond the certifying state other states parties to the Treaty are allowed to
participate with a maximum of four persons per state party but the total
number of participants can be restricted by the certifying state to a total of
no more than 40. During the ground examination, the participants will be
briefed fully on the sensors and related equipment in the aircraft. They shall
88
also be briefed on the safety requirements around the aircraft and will
receive complete familiarization with the entire aircraft. During the in-flight
examination, the aircraft will be flown at the expected minimum height for
each sensor configuration for a minimum of five passes over the calibration
target to establish the verified minimum height for each sensor
configuration. The certification event is limited to a period totalling seven
days per aircraft to be certified.
As part of the trial implementation phase several test certifications have
been conducted. We refer here to the test certification of the former
German Open Skies aircraft, as well as to a joint test certification of six
Open Skies aircraft in August 2001.
The German Test Certification at Köln-Wahn, June 1997
The certification was held from 16-22 June 1997, at the Köln-Wahn
airfield and was organized by the German Verification Center,
Geilenkirchen. The certification was conducted under true Treaty
conditions, although the Treaty had not entered into force by then.
The response to the invitation for the test certification of the German
Open Skies aircraft was overwhelming; therefore, 40 observers—the
maximum number allowed by the Treaty text—had to be selected from 20
interested countries according to their respective flight quota.
21
Due to
their large number, the observers were split into three groups, which were
dispatched daily to the three locations of interest: the aircraft itself
(stationed at the airfield Köln-Wahn), the ground calibration target (at
Mendig airfield, approximately 80 kilometres from Köln-Wahn), and the
image processing and evaluation lab (then also at the airfield Köln-Wahn).
On two days the aircraft took off to the calibration target laid out at
Mendig airfield. For about five hours each the aircraft circled over the target
in order to take images using various sensor configurations at varying flight
altitudes. The different configurations were realized by changing
combinations of cameras (left oblique, vertical, right oblique), of
photographic films (a Kodak film for high altitudes, an Agfa film for low
altitudes), colour filters according to atmospheric conditions and resolution
degradation filters according to flight altitude. Fortunately, all necessary
image flights could be concluded before the weather conditions prohibited
further test flights.
89
The photographic films were processed at Köln-Wahn overnight
following the test flights, in the presence of the observers. Then, for each
sensor/altitude combination, the contrast and resolution of the images were
determined in the presence of the observers. The contrast was measured on
the photographic negatives as exhibited by the large rectangular contrast
fields of the calibration target. The resolution was determined on both the
negatives and on video data independently by ten observers, for each. From
these measurements the minimum flight altitude H
min
was determined and
compared to the expected and actual altitude of the respective image flight.
Numerous discussions between the observers helped to clarify even remote
aspects of the Treaty's certification regulations.
At the ground calibration target at Mendig airfield, the observers
inspected the make of the target and witnessed the ground measurement
of the photometric contrast. The measurements on the target panels were
sampled from a number of points by means of a hand-held photometer.
These measurements are meant to verify that the contrast on the ground
target is compatible with the agreed modulation contrast K
a
= 0.4 in the
H
min
formula. The aircraft itself could be inspected by all non-destructive
means. Representatives of the maintenance team and the refitting-
contractor company were present in order to answer all questions. All
technical reports required for the certification had been assembled
previously by the German Verification Center at Geilenkirchen, as
demanded by Treaty regulations. The course of the certification procedure
was certainly eased by the complementary English and Russian translation
of the documents, which were shipped to the participating observers four
weeks ahead of the certification.
The test certification was finally declared successful on all of the 24
tested sensor/film/filter/altitude configurations. Following two earlier test
certifications of US Open Skies aircraft, and one other test certification of
the Ukraine, this was probably the first really comprehensive certification of
the Open Skies test phase, which observed the Treaty regulations strictly.
The German aircraft was expected to be retrofitted again during 1998
in order to mount the video cameras on stabilized platforms and to
accommodate SAR and the infrared line scanner, requiring a new
certification afterwards. However, the aircraft crashed three months after its
test certification in September 1997.
90
Multilateral Data Gathering and Test Certification, Fürstenfeldbruck,
August 2000 and 2001
In August 2000 the German Verification Center organized a major data
gathering exercise at Fürstenfeldbruck near Munich. Two hundred and fifty
participants and observers from 31 countries were present. The Open Skies
aircraft of Bulgaria, the Czech Republic, Hungary, Romania and the
Ukraine participated in the exercise. The event demonstrated the high
professional standard and expertise of the Open Skies teams.
Based on the success of the event another multilateral certification
exercise was held at Fürstenfeldbruck from 30 July to 13 August 2001. The
aircraft of Canada (with sensor pod), Hungary, Russia, Ukraine and the
United Kingdom performed an extensive flight programme over two
calibration targets. Photo 4.2 shows a display of the five aircraft. For each
sensor configuration five overflights were made, at the nominal height
(expected H
min
).
22
All tested sensor configurations met the expectations.
Hence, it was to be expected that the real certifications after entry into force
of the Treaty would proceed without major problems. Nevertheless, it
turned out that a German and a French calibration target yielded H
min
values differing by 20%, due to different spacing of the bars. It is
recommendable to agree on one type of target in the future.
In addition, some standards in the area of film processing had to be
clarified. Some minor provisions of the Sensor Guidance Document turned
out to be impracticable. It is likely that the film developing facilities of some
states parties might not conform to the strict requirements of the Sensor
Guidance Document. However, the cooperative practice and the spirit of
the Treaty leave sufficient room for overcoming these problems. Photo 4.3
shows as an illustration international experts and observers on board of the
Ukrainian Open Skies aircraft during a calibration flight.
91
Photo 4.2: Display of Open Skies aircraft during the test
certification at Fürstenfeldbruck, August 2001. Photograph:
A. Rothkirch, University of Hamburg.
Photo 4.3: Sensor operators, international experts and
observers on board of the Ukrainian Open Skies aircraft,
August 2001. Photograph: A. Rothkirch, University of
Hamburg.
92
Notes
1
Treaty-relevant policies are coordinated with the Ministry of Foreign
Affairs, which is in charge of the diplomatic representation towards the
co-signatories.
2
For details on the US Open Skies Unit see http://
www.osmpf.wpafb.af.mil and www.dtra.mil/os/ops/os/os_os.html.
3
The data of Bulgaria, Hungary, the Pod Group, Russia, Ukraine, the
United Kingdom, and the United States correspond to their
certification in 2002. The sensor operation altitudes of the Swedish
aircraft are expected estimates, which have to be confirmed. Some of
the parties have more than two sensor configurations and related
operation altitudes. The table gives the minimum and maximum
altitude. The Pod Group panorama camera with reduction filter has an
H
min
of 786 metres at ground swath 4.3 kilometres. The data for
Romania are expected to be the same as for Hungary (same camera/
film).
4
Probably to the amazement of many, the list of countries which do not
intend to acquire their own observation aircraft include large countries
as well. The Federal Republic of Germany, after losing its observation
aircraft in September 1997, decided not to replace the plane. For more
details and an apparent argument for Germany having its own national
observation aircraft see Klaus Arnhold, “Der Vertrag über den Offenen
Himmel: Ein Konzept zur Aktualisierung des Vertrages”, SWP-Studie,
Berlin, June 2002, particularly pp. 15-16.
5
That is, an alternating pattern of warm and cold bars, with a width of
50 centimetres each, shall be resolved, if the (radiative) temperature
difference of two adjacent bars is 3°K.
6
In practice, the ground swath covered by photographic cameras will be
smaller. For example, the panoramic camera of the Pod Group covers
at present a maximum ground swath of 14.5 kilometres on either side,
when flying at sensor operation altitude (H
min
).
7
The German and British sensors are in storage.
8
D. G. Armstrong, “Technical Challenges Under Open Skies”,
Proceedings of the Second International Airborne Remote Sensing
Conference and Exhibition, Strasbourg, Environmental Research
Institute of Michigan, Ann Arbor, Vol. I, 1994, pp. 49-60.
9
Decision 3 uses a slightly different notation: L
2
for L
i
and K
2
for K
i
. We
prefer the running index i because the values L
i
and K
i
refer to results
from different overflights of the calibration target.
93
10
D. G. Armstrong, op. cit., pp. 49-60.
11
S. P. Simmons, “When Better Resolution Is Not Good: The Treaty on
Open Skies”, Proceedings of the Second International Airborne Remote
Sensing Conference and Exhibition, San Francisco, Environmental
Research Institute of Michigan, Ann Arbor, Vol. I, 1996, pp. 403-10.
This paper discusses also the limits of the above formula if non-linear
effects occur (resolution not proportional to flight altitude).
12
Here D is the density of the film and E is the exposure time. For more
details see ibid.
13
Disarmament Bulletin, No. 19, Winter 1992/93, p. 15. More details on
the test flights performed within years 1992-94 can be found in SIPRI
Yearbooks 1993-95.
14
Open Skies Backgrounder No. 5: Second Trial Overflight, Department
of External Affairs, Ottawa: Government of Canada, January 1992.
15
See Table 2.2.
16
R. Hartmann and W. Heydrich, Der Vertrag über den Offenen Himmel,
Baden-Baden: Nomos Verlag, 2000, p. 134.
17
See sections 8.2 and 8.4.
18
See section 7.3.2.
19
The cameras are also frequently used for civilian cartographic and
monitoring missions.
20
The present pod is shared between ten states, thus reducing the
flexibility of the members in scheduling flights. After each flight mission
the pod is dismounted from the aircraft and stored at Brussels. The
mounting procedure has impeded the flow of post-mission events.
Other shortcomings include: no possibility of film replenishment
during flight; in the event of camera malfunction, the camera failure
cannot be observed and corrected; film might have to be changed on
the airfield, i.e., in the open.
21
In the meantime the OSCC has decided that it can agree on admitting
more than 40 participants to a certification (see section 5.3).
22
The Hungarian and Ukrainian aircraft had to make also overflights
each at 0.85 x H
min
and 1.15 x H
min
. The Pod Group, Russia and the
United Kingdom had established the corresponding data prior to the
event, thus easing the workload during the certification exercise.
94
95
CHAPTER 5
POST-RATIFICATION PHASE AND ENTRY INTO FORCE
Márton Krasznai
5.1 POST-RATIFICATION EVENTS
In May 2001 the Russian Federation and Belarus informed the OSCC
that their respective national parliaments had ratified the Treaty on Open
Skies. Taking into consideration these statements, the OSCC started a
discussion on the resumption of activity in its informal working groups. This
was initiated by the joint proposal of the Ukrainian and Russian delegations.
The proposal had a dual aim: to begin the consideration of any remaining
unresolved questions before the initiation of observation flights, and to
commence preparations for the certification of observation aircraft.
On 25 June 2001, the OSCC adopted a Decision to resume the
informal working groups once the Russian Federation and Belarus had
deposited their instruments of ratification. Following this, the provisional
application of the Treaty was extended for the last time. The French
Chairmanship of the Commission put forward a suggestion to hold an
information seminar on the Treaty on Open Skies for all OSCE participating
states. It was agreed to hold the seminar at the beginning of October 2001
and the preparatory work started accordingly.
The seminar was held on 1-2 October 2001. The objectives of this
event were the following:
• revitalizing the spirit of the Treaty and reviewing the history preceding
its signing;
• achieving a deeper understanding of the main Treaty provisions;
• informing about the work that had been done by previous OSCC
working groups;
• recalling the Decisions taken by the OSCC;
96
• formulating some conclusions based on the provisional application of
the Treaty;
• exchanging views regarding further activities, in particular, certification
of the observation aircraft and sensors.
Once the seminar was over, the OSCC took a Decision, on 29 October
2001, to establish three Informal Working Groups (IWG) on Certification,
Sensors, and Rules and Procedures.
5.2 ENTRY INTO FORCE OF THE TREATY, 1 JANUARY 2002
On 2 November 2001, the Russian Federation and Belarus deposited
their instruments of ratification with the depository states Canada and
Hungary, thereby completing the ratification process by all states parties
with a passive quota of eight flights or more. The number of ratifications
now stood at 26 of the 27 signatory States (Kyrgyzstan had still not ratified
the Treaty). Hence, according to its provisions, the Treaty could enter into
force in 60 days, or in other words, on 1 January 2002.
At the OSCC plenary meeting of 5 November 2001, the delegations of
the depository states informed the delegations of the other states parties
that they had received the Russian and Belarussian instruments of
ratification. This was the signal to begin discussion in the IWGs on
Certification and Rules and Procedures.
5.3 CERTIFICATION OF AIRCRAFT
The main goal of the IWG on Certification was to enable states parties
to begin observation flights in a quick and comprehensive manner. To
complete this task, the OSCC began by clarifying the intention of states
parties with respect to the certification of their observation aircraft in 2002,
the desirable time and place of such certification, and finally their
willingness to conduct joint certification. To this end, the German
Delegation stated its government’s readiness to arrange a joint certification
on its territory.
97
The first results were apparent within one month, when the OSCC, at
its plenary meeting of 17 December 2001, adopted a Decision regarding
the provisions for the initial certification period and a Chairperson’s
Statement on issues related to the certification of observation aircraft and
sensors.
According to the above-mentioned Decision the initial certification
period was designated to last from 1 January to 31 July 2002. During this
period, the observation flights were to be conducted on an agreed bilateral
basis only, and in accordance with the Treaty provisions. The utilization of
states parties’ active quotas for the calendar year 2002 would take place
during the period of 1 August 2002 to 31 December 2003. The Decision
also established the initial certification schedule as follows:
• A joint certification of the Republic of Hungary, Ukraine and of the
group of states parties formed by the Republic of Belarus and the
Russian Federation from 15-29 April 2002 at the Naval Air Station
Nordholz, Germany;
• A separate certification of the US aircraft from 8-15 May 2002 at
Wright-Patterson Air Force Base in the United States;
• A unique certification of the Pod Group’s pod-system from 19-26 June
2002 at Orleans Bercy airbase in France;
• A joint certification of the United Kingdom and Bulgaria from 8-16 July
2002 at RAF Brize Norton airbase in the United Kingdom.
The OSCC Chairperson’s Statement established six arrangements to
facilitate the certification process. While doing so experts took into
consideration the previous discussions and agreements achieved in the
OSCC, as well as the experience gained during the period of provisional
application. The arrangements deal with both organizational and purely
technical matters, as follows:
• The methodology for the determination of the number of
individuals participating in a certification. It provides a common
interpretation of the relevant Treaty provisions aiming to find a
compromise between the limited number of participants (40) set forth
in the Treaty and the number of notified participation applications in
the event these exceed 40 persons;
• The principles for the conduct of the C-130H/pod-system
certification. This unique sensors system will be used by the Benelux
98
countries, Canada, France, Greece, Italy, Norway, Portugal and Spain,
referred to as the Pod Group, with their own aircraft C-130H.
According to the agreed principles, the certification results for one C-
130H and its pod-system will be valid for all the above-mentioned
states parties. In order to facilitate this, the Pod Group were obliged to
take additional steps while testing their pod-system prior to certification
and to provide information on their aircraft;
• The principles for joint certification. The definition of joint
certification, as well as its aims and the responsibilities of states parties
wishing to participate in a joint certification while preparing and
conducting this event, were spelt out in this document;
• A common understanding was reached on the relevant provisions of
the Treaty and the Guidance Document on Certification regarding the
determination of H
min
-cert in the course of a certification for sensors
whose ground resolution is dependent upon a height above ground
level;
• The use of CD-ROMs was accepted as the desired medium for the
distribution of data relevant to an observation aircraft and its associated
sensors certification. In order to enhance the effective use of CD-ROMs
containing this data, certain technical standards were agreed;
• The principle giving states parties a right to identify the calibration
target it will use in the course of flight examination during the
certification was established.
Meanwhile, the OSCC IWG on Certification continued its work on the
preparation of the certification of observation aircraft and observation
flights. In February and March 2002, the OSCC adopted Decision 4/02 on
“Provisions for the Use of a Standard Signature Page to the Certification
Report” and Decision 7/02 revising the maximum level of per diem for
states parties representatives during a certification. With the adoption of
these decisions, the IWG on Certification had addressed all the issues that
were to be resolved prior to certification and thus the initial certification
process was able to commence.
By July 2002 the scheduled certifications had been concluded
successfully. Hence, Bulgaria, Hungary, Russia, the Pod Group, Ukraine,
the United Kingdom and the United States had certified aircraft ready to
start regular observation flights. Table 4.1 shows the certified sensor
operation altitudes (H
min
) for vertical cameras. The certification of the
Czech and Romanian aircraft did not materialize and might take place in
99
2004. Sweden and Turkey have scheduled a joint certification at Nordholz
Naval Air Base, Germany, which will include the certification of a Russian
Tu-154 aircraft for April 2004.
5.4 DECISIONS ON RULES AND PROCEDURES
At the 17 December 2001 plenary, the OSCC adopted a Decision on
its rules of procedure and working methods that reflected the Treaty’s entry
into force. The Decision foresees three OSCC sessions per year from the
end of one OSCE recess to the end of the following recess, which would
amount to a period of about four months. Each session will be chaired by a
state party in rotation. Plenary meetings will be held at least once per
session, and not more than once a month. Informal meetings will be
convened by the Chairmanship when needed.
In summer 2002 many tasks listed on the agenda of the IWG on Rules
and Procedures were addressed, notably several difficult issues that may
only be resolved by bringing in appropriate expertise. The list covers a wide
range of topics, in particular those dealing with financial aspects of Open
Skies missions, data collected during observation flights and their copies,
mission planning in special situations and the provision of support for Treaty
implementation.
Many of these issues have emerged as a result of the experience of
states parties during the period of provisional application. Obviously, their
proposals reflect, to a certain extent, different national practices in the
conduct of observation flights. From time to time, during the discussion on
such proposals, the IWG on Rules and Procedures has faced difficulties in
reaching a common understanding of a particular problem and in finding
an appropriate solution. As a result, the consideration of these issues can be
time consuming. On occasion, consideration of such problems by states
parties had to be postponed until sufficient experience has been obtained
in the course of Treaty implementation.
Similarly, discussion on the suggestion of involving the OSCE
Secretariat’s Conflict Prevention Centre (CPC) in Treaty implementation, by
tasking it to maintain an Open Skies Central Data Bank, has also been a slow
process. Without a clear idea of the kinds of functions such a databank
would perform, states parties have not been able to take a common
decision on this issue. Some of the tasks which would be performed by the
100
databank, as well as those reflected in the OSCC Decision 21 of 23 October
1995, have been already carried out by the Conference Services Unit of the
OSCE Secretariat, and can often be performed by states parties themselves.
Therefore, the implementation of Decision 21 could be suspended until its
worth has been proven through experience gained during Treaty
implementation. Ultimately, on 24 February 2003, rather than establishing
a full databank, the OSCC requested the OSCE secretariat to provide
support in filing and distributing information relevant to Treaty matters
(Decision 2/03).
1
Nevertheless, a number of questions were successfully resolved by the
Group and subsequently approved by the OSCC as decisions or
Chairperson’s statements. The most important are the following:
• guidelines for accession to the Treaty on Open Skies;
• transit flights of observation or transport aircraft over territories of states
non-parties to the Treaty;
• coordination of observation flights for the year 2002;
2
• protection of data collected during observation flights.
5.5 TAXI OPTION
On 22 July 2002, Russia stated in the OSCC that it would not provide
its own observation aircraft to conduct flights from the Open Skies airfields
related to the point of entry Ulan-Ude (near Lake Baikal in Siberia) unless
otherwise established by means of formal notification. The inspecting states
parties would have to utilize their own aircraft or that of a third party from
this point of entry. Russia’s decision was motivated by financial, logistical
and technical considerations. Until March 2004 the pool of certified
Russian Open Skies aircraft consisted of medium range AN-30 solely. In
2002/03 Russia also refrained from exercising its taxi option on three
observation flights by France, the United Kingdom and the United States,
respectively, starting from the point of entry near Moscow. Thus, the
Russian position on the taxi option seems to have become more flexible.
3
5.6 ACCESSION OF ADDITIONAL STATES PARTIES
The OSCC dealt also with the guidelines for accession to the Treaty
and adopted Decision 8/02. The provisions laid down in Article XVII, which
101
deal with accession, were significantly developed and clarified. This
Decision is important because it provides the legal basis for accession to the
Treaty by non-states parties. Early in 2002 Finland and Sweden applied for
accession to the Treaty on Open Skies. The applications were deposited
with Canada on 4 January and with Hungary on 7 January. In their
applications, Finland and Sweden also asked to be allotted passive quotas
of five and seven observation flights, respectively.
4
The OSCC adopted
Decisions on 4 February 2002, which stated that the Swedish and Finnish
applications were accepted.
On 12 March 2002, Spain, in its capacity as Chair of the WEU Group
of States Parties, informed the other states parties that Finland and Sweden
had applied to join this Group. According to this annoncement, their
membership will take effect following their accession to the Treaty and not
earlier than six months after the circulation of the above-mentioned Spanish
notification. On 28 June 2002 and 12 December 2002, respectively,
Sweden and Finland deposited their instruments of ratification, meaning
that they became state parties 60 days thereafter.
Five additional states also applied for accession to the Treaty: Bosnia
and Herzegovina, Croatia, Cyprus, Latvia and Lithuania. All of them have
been accepted to join the Treaty by the OSCC on 22 July, except for
Cyprus, because of the veto of Turkey. The handling of this veto is a
challenge for OSCE diplomacy. Turkey had already strongly opposed that
Cyprus would become a state party in 1991-92 and even more so that
Cyprus would have a quota. Estonia and Slovenia applied for accession in
early 2003. Latvia’s accession entered into force on 12 February 2003.
The most recent applications have particular political relevance and
underline the future potential of the Treaty in areas of cross-border
tensions. Each of these states has some unresolved issues with one or more
of its neighbours. The relation of the Baltic states with Russia could be eased
by mutual Open Skies flights, in particular after the admission of the Baltic
states to NATO. Croatia strongly wants to integrate itself into the network of
European institutions. It has been involved in two wars from 1991 to 1995
and is a main actor in the future peaceful development on the Balkans.
Bosnia and Herzegovina still struggles with the wounds of the war and with
ethnic separation. It has been the scene of seven multilateral aerial
observation demonstration flights in 1997-2001. It is thus fully aware of the
potential of Open Skies. However, a number of implementation problems
102
have to be overcome by the different entities of Bosnia and Herzegovina.
The applications of Croatia and Bosnia and Herzegovina are also a reaction
to the failure of establishing a separate aerial monitoring regime under
Article V of the Annex to the Dayton Agreement of 1995, as discussed in
section 8.3.
Outlook
At the same time, for many OSCE participating states, as well as for
other countries interested in joining the Open Skies regime, the financial
burden regarding participation will be an important determining factor. As
the security situation in Europe improves, the question of costs moves to the
forefront. Another problem the states parties will probably encounter quite
soon is the constraints of the Treaty as it now stands. With this in mind, it
may well be worth taking another look at the additional areas of Treaty
application, namely conflict prevention and environmental monitoring,
which were envisaged, but not worked out before the Treaty entered into
force.
Notes
1
The support includes: maintaining a hard copy of all OSCC
documents; filing of circulated national data; maintaining a central
archive of Open Skies notifications; maintaining a point of contact list
of Open Skies experts; posting all OSCC documents on the internal
OSCE delegations web site.
2
The distribution of active quota for flights over the Russian Federation
and Belarus was agreed as follows: out of 28 such observation flights to
be carried out in the period 1 August 2002 to 31 December 2003 13
have been claimed for 2002 and 15 for 2003. The remaining three
flights are reserved for Finland (one) and Sweden (two).
3
A state party, which exercises the taxi option, has to cover the
operation cost of its taxi aircraft for the respective flight. In contrast, if
the observing party provides its own aircraft, it then has the obligation
to cover the aircraft operation cost for the overflight.
4
Also, a passive quota of four has been allocated to Georgia, but has not
been distributed yet.
103
CHAPTER 6
IMAGE ANALYSIS AND DATA ASSESSMENT:
WHAT CAN BE LEARNT FROM OPEN SKIES IMAGE DATA?
Hartwig Spitzer and Rafael Wiemker
Image analysis and interpretation is an art, which requires skills and
experience. In this chapter we will take a look at the detection and
identification potential of the Open Skies image data in different
application areas. We will then discuss shortly the traditional way of photo
interpretation and some of the challenges and opportunities of digital image
processing.
6.1 IMAGING TARGETS AND ASSESSMENT POTENTIAL
What kind of sites will be looked at in Open Skies missions? In
principle any site can be photographed in the spirit of enhancing openness
and confidence. In practice the application horizon will focus primarily on
the observation of military or military-industrial sites and activities, as well
as on the monitoring of crisis areas. The preamble of the Treaty also
mentions a possible future extension to other fields of application like the
protection of the environment.
Hence, potential imaging targets under Open Skies include a wide
choice of military sites and activities in particular:
• Headquarters and communication infrastructure;
• Barracks with vehicle depots and training grounds;
• Airfields;
• Naval facilities;
• Missile launch sites;
• Military production and repair facilities;
• Transport infrastructure (roads, bridges, railway lines etc.);
• Military activities, manoeuvres.
104
Conflict prevention and crisis management might call for the
monitoring of irregular forces, recently laid minefields, refugee movements
and camps, damage assessment of hostile actions, traffic anomalies.
Applications for the protection of the environment will likely focus on the
monitoring of environmental disasters and of emergency situations like
floods, heavy storm damage, forest fires, industrial disasters, etc. (see
section 7.3).
In addition, the Treaty allows for imaging of virtually all civilian, dual-
use or suspect sites without territorial restriction.
In practice the amount of data taking will be limited by the agreed
flight quota and the time constraints of the missions, as well as by weather
conditions. The Treaty aims at enhancing openness and confidence through
occasional checks rather than through regular and complete monitoring.
To what degree can Open Skies image data support the above
objectives? Access to raw image data does not by itself impart useable
knowledge. Image analysts need training and expertise as well as access to
contextual information and reference data.
1
Depending on the spatial
resolution, image analysts can use the images to address various imaging
interpretation tasks.
These tasks usually begin with detection (i.e., determining the presence
or absence of a particular feature like a vehicle) and general identification
(i.e., specifying a feature or an object within a particular class, like
identifying a vehicle as truck). Precise identification discriminates within a
target class of known types (e.g., by identifying a tank as a heavy battle tank
T 72). Description of an object requires much improved resolution in order
to provide precise dimensions, configuration, construction of components
(e.g., by specifying the calibre of the gun of a tank). Description is usually
seen as part of intelligence. A resolution capability that would enable
description of major land weapons systems (like tanks, artillery or combat
aircraft) are clearly excluded by the Treaty for all sensor categories.
Table 6.1 gives the approximate ground resolution in meters required
for target detection and identification (as well as for description and
technical analysis).
2
The origins of this table date back to the 1950s when
photographic film was used exclusively. Hence, we infer that the definition
of resolution used in Table 6.1 is the photogrammetric one, which differs
105
by approximately a factor of two from the definition used for Open Skies
Treaty implementation. Thus, the assessment potential of Open Skies images
at the 30 centimetres Treaty resolution corresponds to the 60 centimetres
resolution values in Table 6.1.
Table 6.1: Ground Resolution (Photogrammetric Definition) Required for
Target Detection, Identification, Description and Analysis
Approximate Ground Resolution in Meters for Target Detection,
Identification, Description and Analysis
Target
a
Detection
b
General
ID
c
Precise
ID
d
Descrip-
tion
e
Tec hnical
Analysis
f
Troop units 6.0 2.0 1.20 0.30 0.150
Vehicles 1.5 0.6 0.30 0.06 0.045
Aircraft 4.5 1.5 1.00 0.15 0.045
Airfield facilities 6.0 4.5 3.00 0.30 0.150
Nuclear weapons
components
2.5 1.5 0.30 0.03 0.015
Missile sites (SSM/SAM) 3.0 1.5 0.60 0.30 0.045
Rockets and artillery 1.0 0.6 0.15 0.05 0.045
Surface ships 7.5-15.0 4.5 0.60 0.30 0.045
Surfaced submarines 7.5-30.0 4.5-6.0 1.50 1.00 0.030
Roads 6.0-9.0 6.0 1.80 0.60 0.400
Bridges 6.0 4.5 1.50 1.00 0.300
Communications
radar 3.0 1.0 0.30 0.15 0.015
radio 3.0 1.5 0.30 0.15 0.015
Command and control
headquarters
3.0 1.5 1.00 0.15 0.090
Supply dumps 1.5-3.0 0.6 0.30 0.03 0.030
Land minefields 3.0-9.0 6.0 1.00 0.03 --
Urban areas 60.0 30.0 3.00 3.00 0.750
Coasts, landing beaches 15.0-30.0 4.5 3.00 1.50 0.150
Ports and harbors 30.0 15.0 6.00 3.00 0.300
106
Tab le 6.1 (continued)
Source: Y. A. Dehquanzada and A. M. Florini, Secrets for Sale: How Commercial
Satellite Imagery will Change the World, and sources quoted therein, Washington,
DC: Carnergie Endowment for International Peace, 2000, p. 45.
We can infer from Table 6.1 that Open Skies panchromatic images at
Treaty resolution will allow for the general identification of land vehicles,
rockets and artillery as well as for precise identification of troop units,
aircraft, airfield facilities, missile sites, surface ships and infrastructure, like
roads and headquarters. In addition, test missions have verified a very good
capability of monitoring the effects of environmental disasters like floods
and hurricane damage.
3
Approximate Ground Resolution in Meters for Target Detection,
Identification, Description and Analysis
Target
a
Detection
b
General
ID
c
Precise
ID
d
Descrip-
tion
e
Technical
Analysis
f
Railroad yards and shops 15.0-30.0 15.0 6.00 1.50 0.400
Terrai n -- 90.0 4.50 1.50 0.750
Notes:
a
The table indicates the minimum resolution in meters at which the target can be detected,
identified, described or analyzed. No source specifies which definition of resolution (pixel-
size or white-dot) is used, but the table is internally consistent.
b
Detection: location of a class of units, object, or activity of military interest.
c
General identification: determination of general target type.
d
Precise identification: discrimination within a target type of known types.
e
Description: size/dimension, configuration/layout, components construction, equipment
count, etc.
f
Technical analysis: detailed analysis of specific equipment.
Sources: US Senate, Committee on Commerce, Science, and Transportation, NASA
Authorization for Fiscal Year 1978, pp. 1642-43, and Reconnaissance Hand Book
(McDonnell Douglas Corporation 1982), p. 125. Table from Ann M. Florini, “The Opening
Skies: Third Party Imaging Satellites and US Security”, International Security, Vol. 13, No. 2
(Fall 1988), p. 98.
107
6.2 POTENTIAL OF DIFFERENT SENSORS
6.2.1 Photographic Cameras (stereo)
Photographic framing cameras equipped with panchromatic (black and
white) film have been—for a long time—the backbone of military aerial
surveillance and civilian photogrammetry. Usually the images are taken in
stereo mode. Here the images are taken in continuous succession so that
each two subsequent frames show a 60% overlap. Thus, each point in the
observed scene is mapped in at least two photographs. The observation
angle of a given landmark is different between the two photographs
(parallax) and its height can be determined from the relative displacement
between the two positions.
Hence, with a 60% frame overlap it becomes possible to construct a
three-dimensional model of the observed scene. The accuracy (
σ
z
) in
determining the height of a feature is about 50% worse than the position
accuracy (
σ
xy
) in the plane.
4
For example, with a photographic image scale
of 1:100,000 (ground swath 23 kilometres) one can expect a positional
accuracy of
σ
xy
≈ 90 centimetres and determine the height of building and
rooftops with an accuracy of
σ
z
≈ 135 centimetres.
Image analysts view two stereo images through stereoscopes, which
provide a three-dimensional impression. Looking at stereo images is like
overflying the observed space. The stereo method contributes significantly
to the identification potential of images, because objects can be
discriminated by their extension in three dimensions. Good image analysts
are trained for many months on known objects and in the use of context
knowledge in order to achieve optimum identification results. Photo 6.1
shows an image detail of a military airport taken for the US Open Skies
programme. The figure also indicates the annotated interpretation obtained
by stereoscopic viewing. Combat aircraft and munitions vehicles can be
identified.
Panoramic and oblique cameras provide the Treaty resolution only in a
small angular sector (in vertical direction for panoramic cameras), and give
a less sharp image in other directions. However, they allow for side views
of buildings, and give a wide overview of the underlying scenery. This
includes the capability to identify power transformers, and lead-ins adjacent
or attached to walls or under roof overhangs. However, the masking of
108
objects by terrain or trees can reduce the information value of oblique
imagery.
5
Photo 6.1: Open Skies image of a military airfield with magnified
details. Source: US Defense Threat Reduction Agency,
Washington, DC, Briefing to PPF 2000 Forum Innovations and
Technology Transfer, May 11, 1999.
6.2.2 Video Images
Video cameras on Open Skies aircraft are primarily used for providing
a real-time look at the scene under or ahead of the aircraft. The ground
swath covered is usually restricted to smaller values than those of aerial
cameras.
Video cameras can be distinguished into two different types:
• A conventional video camera forms a complete image frame in one
look. The camera’s lens focuses the image on a Charge Coupled Device
(CCD) array from where it is converted into either an analogue or
digital video signal. The full video frames come at a rate of 25 hertz
6
and can be viewed in real-time on a closed circuit TV monitor
7
or
recorded by a video tape recorder.
109
• A digital video line camera
8
sees only one line at a time. The line is
perpendicular to the flight track and the full image is formed only as the
aircraft moves along its flight track and image line after line is added
(so-called push broom scanner). The image lines can be recorded
digitally. The image can also be viewed immediately onboard, but it has
to be kept in mind that it does not show a real-time TV image but that
each image line depicts the state at its recording time, that is, a couple
of seconds earlier. The top lines on the display are current while the
bottom lines are the “oldest”.
Both camera types can be built to yield panchromatic or colour
images. Generally the video image is not as sharp as the imagery from the
framing cameras. While the Treaty allows for 30 centimetres ground
resolution of video cameras most systems give much coarser resolution
when flown at the H
min
of photographic cameras (850-11,000 metres). The
main advantages of video systems are the immediate availability of images
during the flight and the optional colour quality, which is permitted by the
Treaty and allows for easier image interpretation.
The silicon, which forms the CCD detector material for both types of
video cameras, is sensitive to visible light and to invisible near-infrared
radiation up to a wavelength of 1,050 nanometres. The former German
Open Skies video camera had a cut-off filter for wavelengths greater than
630 nanometres, since the Treaty does not permit near infrared imagery as
of yet.
9
The near infrared wavelength range, however, would be very useful
for all kinds of vegetation monitoring, and indirectly also for finding buried
items (such as mine fields) since it is capable of showing disturbances in
vegetated areas.
10
The line camera has a substantially higher resolution. A line of a
conventional video image has a resolution equivalent to 700-800 pixels
11
whereas a line camera typically contains 6,000 pixels.
12
This means that for
a fixed ground resolution the line camera can cover a seven times larger
ground swath, or if the ground swath is fixed the image is seven times better
resolved. On the other hand, the line camera images are very sensitive to
deviations from a straight flight path. Each turn or attitude change of the
aircraft will cause distortions in the image, which have to be corrected
during later geocoding (see section 6.3.1). Thus, the raw images resulting
from conventional and line cameras are not comparable. The conventional
video images are captured instantaneously and are much less distorted.
110
6.2.3 Thermal Infrared Images
Thermal infrared image detectors are sensitive to the thermal radiation,
which each body emits. Bodies at room temperature (20 degrees) emit
electromagnetic radiation primarily in the wavelength range of 8-13
micrometres, whereas the wavelength of visible light is at 0.4-0.68
micrometres. Hot objects like the afterburner of a jet fighter emit
dominantly at wavelengths between 1 and 5 micrometres (depending on
temperature). Thermal images can be taken during day and night,
independent of illumination by the sun.
Precision imaging with thermal infrared radiation presents bigger
technical problems than imaging in the visible range. Because of longer
wavelengths the diffraction limit
13
is higher and has to be compensated for
by larger apertures. Detectors need to be cooled in order to reduce thermal
noise. Since glass does not transmit thermal infrared radiation for most of
the wavelength range (≥ 3 micrometres), different lens materials such as
germanium have to be used. The intensity of infrared radiation is usually
smaller than that of visible-light radiation (during the day). In order to detect
enough radiation and to stay clear of the diffraction limit one has to use an
electro-optic image sensor with quite large pixels with dimensions of 20-
100 micrometres (as compared to typical pixel sizes of 10 micrometres in
CCD detectors for visible light).
The Open Skies Treaty foresees the future use of thermal infrared line
scanners in a second phase (final concept starting on 1 January 2006).
Infrared line scanners use a rotating mirror with optics to direct radiation
from a small ground-surface area to a detector or detector array. The mirror
rotates perpendicular to the line of flight so that, with each cycle, a strip of
ground perpendicular to the flight direction is covered. The ground swath
covered is quite wide, for example, 1.15 x h for the AA/AAD-5 line scanner
described in Appendix C, where h is the flight altitude. The resulting image
can be registered either on film or on magnetic tape.
In the context of Open Skies, thermal infrared imaging will be
particularly useful for monitoring military manoeuvres and production
plants at day and night. The operation state of vehicles or equipment can
be deduced from their heat profile. The fuel status of aircraft and storage
tanks can be determined as well as the thermal differences in effluents and
cooling ponds. Photo 6.2 shows a thermal infrared aerial image of an
111
airport. Warm objects appear white and cold objects dark. Aircraft and
buildings can be clearly recognized.
14
Photo 6.2: Thermal infrared line scanner image of an airport taken at
night. Notice the “hot” buried heating lines and trees along the
roadsides. Warm auxiliary power units can be seen near some of the
aircraft. One aircraft in the upper left corner appears anomalously warm.
It might have been parked in the nearby hangar not long before the
image was taken. Photo: Courtesy of Intera-Kenting, Ottawa.
112
The image was taken shortly before midnight. The quality of the
picture demonstrates the potential of thermal infrared imaging for night
vision independent of illumination. At the Treaty resolution of 50
centimetres thermal infrared images will be an important additional source
of information in addition to photographic cameras. Image analysis of
digitally stored infrared images will require computer processing and display
of the image data. The line scanners of type AA/AAD-5, which are prepared
for Open Skies use, record the thermal images on film. This eases the
analysis of the image data by human image analysts.
In summary, infrared imaging technology is available for aerial
monitoring of heat profiles of hardware and buildings. The available
resolution will be adequate to detect objects of interest, which are not
masked by roofs. Infrared radiation also can give indications of activities in
buildings of industrial production as derived from temperature profiles.
Infrared sensors will be particularly useful for crisis monitoring missions by
locating activities at night.
6.2.4 Synthetic Aperture Radar
Unlike other (“passive”) Open Skies sensors, SAR is a so-called “active”
sensor. It emits microwaves, which are scattered back from the ground and
are recorded by the sensor. The most prominent feature of SAR instruments
is their 24-hour/all-weather capability. Since SAR produces its own
microwave “illumination”, it does not need the sunlight to brighten the
scene. Moreover, SAR microwaves are not obstructed by haze and clouds.
In principle, the ground resolution of a SAR sensor is independent of
flight altitude. Better ground resolution requires higher microwave and
computing power, however. The ground resolution limit of 3 metres
harkens back to the fact that sub-meter resolution SAR has only very
recently become commercially available.
Unlike other sensors, the raw data recorded by the SAR antenna are
not in image-like form. The raw data, which comprise phase and amplitude
of the reflected waves as well as timing information, have first to undergo
sophisticated processing before the image can be seen. It has to be noted
that the final image quality depends to a certain degree on advanced post-
processing techniques. The Treaty foresees that copies of the original raw
data can be requested by other states parties.
113
SAR imagery looks quite different from optical imagery and is more
difficult to interpret. The signal strength of the ground reflected microwaves
depends on the roughness of the scanned surface and the orientation
relative to the antenna. Smooth surfaces like water bodies have a very low
reflection signal. The reflection signal becomes high when the surface
roughness scale is comparable to the radar wavelength (1-30 centimetres).
The reflection is most pronounced for conducting materials, particularly
metallic structures.
Photo 6.3 shows a SAR image of the military airport at
Fürstenfeldbruck, Bavaria, Germany as an example. The image was taken
by a commercial airborne SAR sensor, which operates at a nominal
resolution of 50 centimetres (both in flight direction and perpendicular to
the flight path). The airfield, various buildings, streets, trees and fields can
be recognized. The aircraft can be identified from their dimensions as
Transall transporters. A comparison with Photos 6.1 and 6.2 shows that the
contours of well defined objects appear fuzzier on SAR images than on
photographic or thermal infrared images of similar resolution.
Photo 6.3: SAR image
of the military airport
at Fürstenfeldbruck,
Bavaria, Germany at
nominal resolution of
50 centimetres.
Photo: Courtesy of
Dr. J. Moreira,
Aerosensing GmbH,
Wessling.
114
Insiders claim that a SAR resolution of 15-20 centimetres would be
required in order to provide the same sharpness of shapes as a photo of 30-
50 centimetres resolution. Still, the 3-metre resolution of SAR sensors under
Open Skies will provide detection and identification of infrastructure
facilities like airports, streets, bridges, buildings as well as of large vehicles
(e.g., aircraft, ships). Aircraft types can be determined from their
dimensions. The main advantage of SAR will be its all-weather day-and-
night operation capability.
6.3 THE DIGITAL REVOLUTION:
COMPUTER-AIDED IMAGE ANALYSIS
For many decades the analysis and interpretation of aerial images for
security purposes has been the task of well-trained human image analysts.
Optical equipment and computers have been used for supporting image
display and measurement. But the main information processor was the
human eye and brain, assisted by the memory of a long-term experience
and training.
With the advent of digital satellite and aerial image sensors the role of
computers in image processing and analysis is growing continuously. We
expect that a transition to the digital age will come sooner or later also in
the Open Skies context. Therefore, we will sketch shortly some of the
challenges and opportunities of both digital image processing and analysis.
Image processing technologies have reached a mature stage, in
particular in relation to:
• pre-processing (geometric and radiometric calibration, geocoding,
noise filtering, image restoration in the presence of sensor artifacts);
• visualisation (colour image display, formation of colour composite
images, stereoscopic display of terrain, virtual reality, that is, visual
display of a three dimensional view from a flight over a scene);
• segmentation and feature enhancement (edge detection, contrast
sharpening, etc.).
In the field of digital image analysis computerized land cover
classification has yielded a still growing variety of unsupervised and
115
supervised methods, which are widely used.
15
All approaches have to deal
with major validation problems:
• acquisition of reliable reference data (ground truth) in order to calibrate
the classifying algorithms or to validate the results;
• difficulties of quantifying the classification accuracy in an objective way,
even if some kind of ground truth is available.
Useful classification tools are available commercially. However,
research and development is still ongoing, in particular in the area of
enhancing the accuracy of classification.
Further steps towards a comprehensive computer assisted image
understanding are the fields of object recognition and knowledge-based
interpretation of scenes. These challenges are the concern of various
research and development programmes. Applications to a few rather simple
target configurations have reached the maturity required for commercial
applications.
Another important development is the trend towards combined
exploitation of image data from different sensors and different resolutions
(multi-sensor data fusion and multi-resolution analysis).
Below we elaborate on some of the techniques, which are most likely
to be applied in the Open Skies context, once the imagery is available in
digitalized form. These techniques can be applied already now, provided
the analogue imagery (photographs) is properly digitalized.
6.3.1 Geocoding/Image Registration
The imagery recorded during an Open Skies flight can of course be
analyzed in its original form as it comes from each sensor, for example,
frame-by-frame for photographic images and SAR, and track-by-track for
digital video and infrared line scanners.
16
However, for advanced archival
processing and computer aided analysis a crucial pre-requisite is—after
proper digitization—the geocoding or geometric registration of the imagery.
Ideally, the images are geocoded, that is, each image element (pixel) is
mapped onto a geographic world coordinate. Thus, the image can be
embedded into a Geographic Information System (GIS) for archival and
116
retrieval. From there all images available for a given coordinate range can
be retrieved and displayed in synopsis. A less demanding possibility than full
geocoding is the relative geometric registration of several images with
respect to each other for case based studies.
For illustration, Photo 6.4 shows two images of the same scene
(recorded in 1991 and 1995) registered to map. The images were recorded
by a multispectral line scanner from an altitude of 1,800 metres at
wavelengths of visible light. Note the distortions that stem from the slightly
bent flight path of the sensor-carrying aircraft (image recording time is
approximately 20 seconds). The images are resampled so that spatial
correspondence is achieved.
Operational geocoding/registration is still a tedious problem for
airborne imagery. Correct registration depends on the one hand on the
terrain of the imaged scene, and on the other hand on the flight path of the
aircraft. The position and attitude of the aircraft has to be computed for the
recording time of each pixel. Since the 1990s this can be achieved at the
necessary rate of about 100 hertz by combining different global positioning
system (GPS) and inertial navigation system (INS) data.
An alternative—if such an automated parametric registration is not
available—is the selection of landmarks or ground control points (GCPs) by
photo analysts. The same control point has to be identified in the recorded
image as well as on the topographic map or the other image in question.
Due to flight path ambiguity, airborne images need a rather high number of
hand selected GCPs.
17
By virtue of these landmarks, locally adaptive
coordinate transformation functions can be computed which are then used
to register the imagery.
18
A typical problem of imagery recorded with airborne line scanners is
that the registration inaccuracy can be as large as the size of one or several
pixels. The elevation of scene objects displaces their position in the image.
With a digital elevation model (DEM) or digital terrain model (DTM) this
effect can be corrected. However, small objects such as houses and trees
are usually not considered in a DEM with a common mesh width of say 50
metres. The residual displacements are much more of a problem in
airborne imagery than in spaceborne one, due to the larger scan angles of
sensors carried by aircraft.
117
The registration and geocoding problems are in principle well
understood and controlled. With the advent of differential GPS the
geocoding can theoretically be carried out by a fully automated procedure.
However, as of yet these procedures are still quite tedious. Often not all the
data is available: missing DTM, missing difference signal for the GPS, lacking
GPS accuracy due to atmospheric disturbance or mountainous terrain, high
frequency vibrations of the sensors relative to the platform, etc. are
Photo 6.4: Top: Images of the airport of Nürnberg, Germany and adjacent
areas taken by a line scanner in two time periods. Bottom: The same images
after registration to a map. Source: A. Rothkirch, University of Hamburg.
118
common problems. Large scale monitoring needs quite robust automated
procedures and precludes landmark selection by hand.
6.3.2 Computer Aided Fusion of Images from Multiple Sources
Data fusion allows the merging of imagery of the same scene from
multiple sources, such as images recorded by different sensors, or images
from the same sensor taken at different recording times (see section 6.3.3).
Often during Open Skies image flights images of the same scene are
recorded simultaneously by different sensors. In this case all sensors operate
from the same viewing angle and the mutual geometric registration is
relatively easy. Since the various sensors have different capabilities of
showing features of interest, the analysts can often benefit from fusing the
imagery.
In some cases the fusion may be a simple overlay of complementary
data sets represented in different colours (see Figure 6.1, Top). For
example, the image from the infrared line scanner may be coded in a red
colour with varying saturation and overlaid on a greyscale radar image for
better orientation. Then the resulting image is much more instructive as to
where strong thermal signatures occur and what they probably are.
Moreover, image fusion can sometimes produce a refined product of
a substantially improved quality. For example, the infrared line scanner is
restricted to a slightly coarser resolution (50 centimetres) than the
photographic imagery (30 centimetres); by computer aided digital image
fusion it is then possible to sharpen the coarser images by means of a finer
one (see Figure 6.1, Bottom). A digital colour video image of 1-metre
resolution, say, can be used with the registered 30 centimetres black and
white photographs of the same scene and yields a fused colour image which
appears almost as sharp as the 30 centimetres panchromatic photographs.
One can go even a step further by fusing Open Skies photos at 30
centimetres resolution with high-resolution commercial satellite images. A
new generation of commercial satellites provides images in 4 colour
channels (red, green, blue, near infrared) at 4-metre ground resolution.
Fusion of such images with Open Skies black and white photos will provide
fused multispectral or colour images, at approximately 50 centimetres
resolution, a product which is well suited for multispectral change
119
detection.
19
In order to ease the analysis both sets of images should be
taken in nadir mode (vertical).
Figure 6.1: Top: Overlay of images from two different sensors,
which are sensitive to different features. Bottom: Fusion
sharpening of a coarsely resolved colour image.
120
The resulting fusion-sharpened image is not only visually more
satisfactory, but it has been shown that subsequent digital image processing
applications such as automated ground cover classification,
20
change
detection, etc. show indeed clearly improved performance on fusion-
sharpened imagery.
21
We see that image fusion techniques are not restricted to mere
overlaying. They can rather produce imagery of an improved quality.
Certain objects may show features that are inconspicuous in the specific
images of a single detector, and only their co-occurrence in different sensor
modalities is meaningful. In these cases image fusion may allow detection
of objects, which would not be found in either of the single un-fused
detector specific images.
6.3.3 Computer Aided Change Detection
So far the imagery recorded during the Open Skies trial
implementation phase has been analyzed mostly by eye appraisal of
experienced photo analysts. Regardless as to whether the imagery is
reproduced on paper or digitalized or genuinely digital (such as digital
video, SAR), the analysis is done by human analysts seated at the light table
or before the computer monitor.
Once the Open Skies observation flights will be carried out in full, the
amount of both photographic and digital image data may become
substantial. As the comparison of aerial images of the same scene taken at
different recording times is a tedious and time consuming task, data
processing might become one of the central technical bottlenecks, in
particular if large territories are to be monitored on a regular basis. Hence,
it seems natural to investigate the possibility of computer aided change
detection. Can semi-automated digital image analysis give assistance to the
image analysts?
The Remote Sensing and Computer Vision literature provides two basic
approaches to computer aided change detection:
(1) Comparison of image pairs (data driven, image feature based)
Two images, I
1
and I
2
, recorded by overflights over the same scene at
two different time periods, T
1
and T
2
, are compared. An appropriate
121
algorithm must then check the two observed images against each other and
assist the analyst by designating those areas where the ground cover has
probably changed. The algorithm can base its comparison on so-called
“early” features such as brightness, texture, spectral properties (for colour/
multispectral imagery), temperature (for thermal imagery), polarization
properties (for SAR imagery). Other higher-level features can, for example,
be object signatures and land cover classes. These methods are based on a
pixel-to-pixel comparison. They can be extended to so-called scale space
approaches. The comparison is performed at different levels of increasing
scale, so that after single pixels have been compared local neighbourhoods
of increasing size are also contrasted.
Image-to-image comparisons have been tested, and used for a number
of different algorithms and geoscience applications.
22,23,24,25,26
They can
successfully designate the locations of probable change areas for closer
human inspection. They do not produce semantic interpretations of
changes, however.
(2) Comparison of image and scene models (map) (model driven,
knowledge based)
Another approach tries to match new image data with reference
information extracted from a topographic map, or better a geographic
information system, which represents the observed scene in a symbolic
form. The scene contents are described in terms of meaningful structures
(such as buildings, streets, vegetation areas, etc.). Changes should be
detected where structures appearing in the image cannot be explained by
the scene description and where expected scene objects cannot be verified
by corresponding image structures.
27,28,29
Although concepts and
description logic frameworks for this second approach have been
presented, practical application is still far away. A particular hindrance is the
requirement of a complete symbolic representation of all image scenes.
Between “Vancouver and Vladivostok” there are large areas which have
certainly never been depicted in 1:10000 topographic maps, not to
mention digital geographic information systems.
Problems related to the first approach are elaborated here below.
122
Limitations of Change Detection by Direct Image Comparison
A fundamental problem of comparing two images of the same scene is
that the recording conditions may have changed. Change detection
techniques should be robust in that they should be resistant to possible
variations in atmospheric parameters and imaging parameters that may
have occurred between the two recordings. In particular, the direct solar
illumination and the diffuse skylight, the path radiance, and the
transmittance of the atmosphere may have changed. For multispectral
sensors also the radiometric calibration may have changed individually for
each spectral band.
If simply the difference between the intensity values from images I
1
and I
2
is considered, then all these systematic errors would cause spurious
results and almost everything would appear “changed”. Therefore, robust
approaches find areas of probable change by their statistical “strangeness”.
The underlying idea is that all pairs of pixels which have not changed form
a certain kind of statistical cluster, and statistical outliers with “strange”
relations between their appearances in image I
1
and I
2
stick out.
Image-to-image change detection has, apart from the non-trivial
computing cost, a number of further problems and limitations:
• Shadows cause false indications of change, due to their shift depending
on the position of the sun at the time of recording;
• Elevated objects cause false indications of change whenever the
observation angle is different between the two compared images, due
to perspective displacement;
• Changes can only be detected reliably if the magnitude of their
expression is greater than the registration/geocoding margin of error. If
changes of smaller magnitude are not suppressed then spurious change
results appear which stem from mis-registration rather than from real
displacements of objects;
• Often the images contain also vast vegetational or agricultural areas.
Although ecological or seasonal changes in these areas are not of
interest, they will be indicated nonetheless as the algorithm cannot
automatically decide on the nature of the changes;
• Statistical change detection approaches work on the basis of identifying
areas that appear spectrally “strange”, that is, identifying statistical
outliers. If, however, changes make up for more than 50% of the whole
image area (e.g., seasonal vegetation changes) then statistical change
123
detection approaches are mislead into designating the constant features
as “changed” and vice versa;
• Direct comparison of the images (instead of statistical approaches)
need a very careful radiometric calibration of the imagery and are
easily deceived by changed atmospheric conditions between the two
recording times.
Some of these limitations could in principle be overcome by model
driven approaches (approach (2)); however, these require large databases
and enormous computing power and are not operational as of yet.
In summary, computer based change detection can be helpful to
indicate areas of potential change to the image analyst. Computerized
change detection can be applied already to digitalized single channel
images (e.g., black and white photos). However, the method is more potent
when applied to multichannel (multispectral) images.
Notes
1
J. C. Baker, “New Users and Established Experts: Bridging the
Knowledge Gap in Interpreting Commercial Satellite Imagery”, in
J. C. Baker, K. M. O’Connell and R. A. Williamson (eds), Commercial
Observation Satellites, Santa Monica and Bethesda: RAND and ASPRS,
2001, pp. 533-57.
2
See, for example, Table 1 of Y. A. Dehquanzada and A. M. Florini,
Secrets for Sale: How Commercial Satellite Imagery will Change the
World, and sources quoted therein, Washington, DC: Carnergie
Endowment for International Peace, 2000, p. 45.
3
See section 7.3.
4
See, for example, K. Kraus, Photogrammetrie, Vol.1, Bonn: Dümmler,
1994.
5
M. Heric, C. Lucas and C. Devine, “The Open Skies Treaty: Qualitative
Utility Evaluations of Aircraft Reconnaissance and Commercial Satellite
Imagery”, Photogrammetric Engineering and Remote Sensing, Vol. 62,
March 1996, pp. 279-84.
6
30 hertz for US and Japanese systems.
7
Like onboard the US Open Skies aircraft.
124
8
Like the Zeiss VOS-60 onboard the former German Open Skies
Tupolev-154.
9
B. Uhl, “High Resolution Digital Colour EO Camera System VOS”,
Proceedings of the Third International Airborne Remote Sensing
Conference and Exhibition, Copenhagen, Environmental Research
Institute of Michigan, Ann Arbor, Vol. II, 1997, pp. 21-28.
10
E. M. Winter, D. J. Fields, M. R. Carter, C. L. Benett, P. G. Lucey,
J.R.Hohnson, K. A. Horton and A. P. Bowman, “Assessment of
Techniques for Airborne Infrared Land Mine Detection”, Proceedings
of the Third International Airborne Remote Sensing Conference and
Exhibition, Copenhagen, Environmental Research Institute of Michigan,
Ann Arbor, Vol. II, 1997, pp. 44-51.
11
A conventional video image has 625 lines (US and Japan 525 lines) and
an aspect ratio of 4:3.
12
Like the Kodak detector KLI-6003 used in the Zeiss VOS-60 (see
B. Uhl).
13
The diffraction limit of the resolution of an optical system results from
the wave nature of the radiation. The larger the wavelength the larger
the aperture of the system that has to be chosen in order to provide a
specified resolution.
14
The profile of an aircraft remains visible for a while since the tarmac
under the aircraft stores the heat of the day longer than the metal
fuselage of the aircraft and hence remains at a higher temperature.
15
See, for example, X. Jia, J. A. Richards and D. E. Ricken (eds), Remote
Sensing Digital Image Analysis, Heidelberg and New York: Springer,
1999.
16
The various sensors may require an individual correction as pre-
processing. Due to their respective optical system some sensors, such
as line scanners, do not produce images at true scale. They can be
corrected by mathematical modelling of the image formation process.
17
The computer aided automatic selection of corresponding landmarks
(GCPs) is possible only for essentially undistorted photographic or
satellite imagery.
18
See M. Ehlers, “Geometric Registration of Airborne Scanner Data
Using Multiquadric Interpolation Techniques”, Proceedings of the First
International Airborne Remote Sensing Conference and Exhibition,
Strasbourg, Environmental Research Institute of Michigan, Ann Arbor,
Vol. II, 1994, pp. 492-502; R. Wiemker, K. Rohr, L. Binder, R. Sprengel,
and H. S. Stiehl, “Application of Elastic Registration to Imagery from
Airborne Scanners”, Proceedings of the XVIII Congress of ISPRS,
125
Commission VII, International Archives of Photogrammetry and
Remote Sensing, Vol. XXXI, No. B7, 1996, pp. 949-54; and
R. Wiemker, “Registration of Airborne Scanner Imagery Using Akima
Local Quintic Polynomial Interpolation”, Proceedings of the Second
International Airborne Remote Sensing Conference and Exhibition, San
Francisco, Environmental Research Institute of Michigan, Ann Arbor,
Vol. III, 1996, pp. 210-19.
19
See section 6.3.3.
20
Multispectral ground cover classification, or land use classification can
be performed on e.g., red, green, blue (RGB) video data. The
classification algorithms compute thematic maps from the imagery
based on the spectral appearance (colour) of each pixel.
21
See R. Wiemker, B. Prinz, G. Meister, R. Franck and H. Spitzer,
“Accuracy Assessment of Vegetation Monitoring with High Spatial
Resolution Satellite Imagery”, Proceedings of the ISPRS Commission VII
Symposium on Resource and Environmental Monitoring—Local,
Regional, Global, WG III, International Archives of Photogrammetry
and Remote Sensing, Vol. XXXII, No. 7, 1998, available at http://kogs-
www.informatik.uni-hamburg.de/PROJECTS/ censis/budapest98.
fusion.pdf; B. Prinz, R. Wiemker and H. Spitzer, “Simulation of High
Resolution Satellite Imagery for Accuracy Assessment of Fusion
Algorithms”, Proceedings of the Joint Workshop of ISPRS WG I/I, I/3 and
IV/4, Institute for Photogrammetry and Engineering Surveys, University
of Hannover, Vol. 17, pp. 223-31, 1997; and H. Spitzer, R. Franck, M.
Kollewe, N. Rega, A. Rothkirch and R. Wiemker, “Change Detection
with 1-Metre Resolution Satellite and Aerial Images in Urban Areas”,
CENSIS – Report – 37 – 01, Hamburg: Institut für Experimentalphysik,
University of Hamburg, 2002.
22
See, for example, T. M. Lillesand and R. W. Kiefer, Remote Sensing and
Image Interpretation, New York: John Wiley, 1994, pp. 621-23.
23
A. Singh, “Digital Change Detection Techniques Using Remotely-
Sensed Data”, International Journal of Remote Sensing, Vol. 10, No. 6,
1989, pp. 989-1003.
24
X. Jia, J. A. Richards and D. E. Ricken (eds), op. cit.
25
A. A. Nielsen, R. Larsen and H. Skriver, “Change Detection in Bi-
Temporal EMISAR Data From Kalr, Denmark, by Means of Canonical
Correlation Analysis”, Proceedings of the Third International Airborne
Remote Sensing Conference and Exhibition, Copenhagen,
Environmental Research Institute of Michigan, Ann Arbor, Vol. I, 1997,
pp. 281-87.
126
26
R. Wiemker, “An Iterative Spectral-Spatial Bayesian Labeling Approach
for Unsupervised Robust Change Detection on Remotely Sensed
Multispectral Imagery”, in G. Sommer, K. Daniilidis and J. Pauli (eds),
Proceedings of the 7
th
International Conference on Computer Analysis
of Images and Patterns, Kiel 1997, CAIP’97, Lecture Notes in Computer
Science, Vol. 1296, Heidelberg and New York: Springer, 1997,
pp. 263-70.
27
See, for example, L. Dreschler-Fischer, C. Drewniok, H. Lange and
C. Schröder, “A Knowledge-Based Approach to the Detection and
Interpretation of Changes in Aerial Images”, in S. Fujimura (ed.),
Proceedings of the International Geoscience and Remote Sensing
Symposium IGARSS’93, Tokyo, August 1993, IEEE, Vol. I, 1993, pp.
159-161.
28
H. Lange and C. Schröder, “Analysis and Interpretation of Changes in
Aerial Images”, in H. Ebner, C. Heipke and K. Eder (eds), Proceedings
of the ISPRS Commission III Symposium on Spatial Information from
Digital Photogrammetry and Computer Vision, International Archives of
Photogrammetry and Remote Sensing, Vol. XXX, No. 3, 1994, pp. 475-
82. SPIE, Vol. 2357, 1994.
29
C.-E. Liedtke, J. Bückner, O. Grau, S. Growe and R. Tönjes, “AIDA: A
System for the Knowledge Based Interpretation of Remote Sensing
Data”, Proceedings of the Third International Airborne Remote Sensing
Conference and Exhibition, Copenhagen, Environmental Research
Institute of Michigan, Ann Arbor, Vol. II, 1997, pp. 313-20.
127
CHAPTER 7
PROSPECTS FOR EXTENSIONS OF THE MULTILATERAL
OPEN SKIES TREATY
Hartwig Spitzer
In this chapter we will discuss prospects and avenues for extensions of
the Open Skies Treaty. We address the inclusion of additional states,
applications to additional fields and adoption of additional sensors. At the
conclusion of the chapter we discuss the issues and challenges facing the
review conference in 2005.
7.1 INCLUSION OF ADDITIONAL STATES PARTIES
Six months after entry into force, that is, after 1 July 2002 every state,
which intends to fulfil the objectives of the Treaty, may apply for
membership. It can be expected that additional OSCE participating states
may consider joining the Treaty, like Austria, Switzerland and some of the
successor states of the former Yugoslavia. Seen from the perspective of crisis
prevention and crisis management the inclusion of additional or all
Caucasus and Central Asian states would be extremely useful. This objective
will require a major diplomatic and political initiative. A first step would be
the initiation of the ratification process in Kyrgyzstan and technical support
to Georgia, in order to ease its full participation in the Open Skies flight
programme. Japan was an observer to the Open Skies negotiations in 1991-
92. In 2002 Japan probed informally the possibility of acceding to the
Treaty. Reportedly—for the time being—the US would object to the
admission of non-OSCE participating states to the Treaty. Decisions on
admission require unanimous consent of the states parties.
128
7.2 CONFLICT PREVENTION AND CRISIS MANAGEMENT
The preamble of the Treaty mentions the possibility to use the Open
Skies regime for conflict prevention and crisis management within the
framework of the OSCE or of other international institutions. So far—during
the trial implementation—Open Skies flights have been carried out in two
conflict or post-conflict situations:
a) A Russian flight over US bases and installations in Germany in June
1999 during the NATO air campaign in the former Yugoslavia and the
force build-up for the NATO intervention in Kosovo (see section 4.5.);
b) Seven demonstration flights over Bosnia and Herzegovina from 1997
to 2001 in support of post-conflict peace building, verification and
damage assessment. The flights were carried out under the auspices of
the OSCE in its capacity of monitoring arms control under Article II of
the Annex to the Dayton Accord (see sections 4.5 and 8.2).
Similar flights over member states could be performed in the future, in
case a crisis emerges in or between member states, as long as the parties
involved agree and flight safety is assured. In this context it is important to
note that some of the sensor configurations can be flown at altitudes above
5,000 meters outside the range of anti-aircraft artillery and shoulder-
launched surface-to-air missiles (e.g., Stinger missiles).
Open Skies flights could in particular become part of present or future
peacekeeping operations. Observation flights could usefully complement
ground monitoring, for example, in Georgia (Ossetia and Abkhazia), in
Moldova (in case a solution is found to the Transdniester conflict) or in
Azerbaijan and Armenia, in case an international peacekeeping operation
is deployed as part of a solution to the Nagorno-Karabakh conflict.
Open Skies flights could become useful tools of early warning and
conflict prevention. For example, high altitude observation flights along the
Russian-Georgian border (the Chechen, the Ingush and the Dagestani part)
could usefully complement a ground based monitoring operation by the
OSCE. Also, in Central Asia, Open Skies flights could become a highly
effective tool of early warning and confidence building along the borders of
Central Asian countries with Afghanistan and each other. In case of
reappearance of Islamic insurgents in the region, who often move across
129
state borders, observation flights could serve as a tool to dispel
misunderstandings, suspicion and accusations based on false perceptions.
Last but not least, highly cooperative joint observation flights (with the
addition to the sensor suite of infrared cameras with 50 centimetres
resolution) could become a key tool in the fight against other challenges of
the 21st century like terrorism, organized crime, drug smuggling, trafficking
in human beings, illegal transfers of small arms and light weapons, irregular/
illegal migration, etc. These illegal activities often take place in remote
regions with poor infrastructure, difficult terrain and high security risks for
ground-based observers, like in many parts of Central Asia, the Caucasus
and the Balkans. The states of these regions should be enabled to regularly
monitor these nefarious activities, agree on a desirable course of action to
fight them and request international assistance (by formulating projects
based on sound information). Joint flights (perhaps with the participation of
international experts) would enable the governments of these regions to
provide regular, well-documented information on these activities and base
their request for assistance on this information.
The sine qua non precondition for these new types of application
would be a cooperative arrangement, under which Western governments
would provide technical and operational assistance free of charge (or at low
cost) to the countries of Central Asia, the Caucasus and the Balkans. This
assistance would enable the less wealthy states of these regions to become
effective partners in the fight against the threats that mark the beginning of
the 21st century.
The advantages for Western Europe and North America would be
significant. Effective action against these new scourges would start not on
the Schengen border of the European Union, but already on the border
between Afghanistan and Central Asia, the Middle East and the states of the
Caucasus; or over Albania, rather than on the Italian coast.
The political arrangements for Open Skies flights in crisis regions
outside of the Treaty area are yet to be made. Of course, arrangements can
be made by mutual consent of one or several Open Skies states parties and
the states affected by a crisis. Alternatively, this could be negotiated under
the auspices of the OSCE or the United Nations. However, to date, neither
the OSCE nor the United Nations has actively pursued such options. It
would be helpful to develop diplomatic and political initiatives in order to
130
exploit the potential of the Treaty and the specialized and dedicated
resources of the various states parties for contributing to the management
of current or future crises.
7.3 POTENTIAL OF THE OPEN SKIES REGIME AND SENSOR SUITE
FOR
ENVIRONMENTAL MONITORING AND OTHER NON-
MILITARY APPLICATIONS
The preamble of the Treaty envisages “the extension of the Open Skies
regime into additional fields, such as the protection of the environment”.
One obvious field of application is the rapid monitoring of environmental
disasters with cross-border impact. This section addresses three questions:
1. To what extent is the present sensor set as specified by the Treaty suited
for monitoring of the environment?
2. Which modifications of the sensor set would strengthen the capacity
for environmental monitoring?
3. Which institutional provisions and operational procedures have to be
established in order to arrive at agreeable and cost-effective solutions?
Initially the interest and expectations in environmental applications of
the Open Skies Treaty were quite high. Consequently, the OSCC has held
two informal seminars on the possible use of the Open Skies regime for
environmental monitoring on 3-4 December 1992 and on 11-12 July 1994.
Due to the long delay of entry into force and technical developments in
non-military environmental monitoring the initial interest faded away to a
large degree. It also has turned out that the responsibility for Treaty
implementation and for bearing the costs, so far, has been assigned
exclusively to the military establishments of the states parties. Hence, in
order to arrive at a viable capacity for environmental monitoring under
Open Skies basic questions of institutional interest and responsibility have
to be clarified. This issue will be addressed below. Before, though, let us
take a closer look at the Open Skies sensor suite and its potential for
environmental monitoring.
7.3.1 Potential of Current Sensors for Environmental Monitoring
The current Open Skies sensor suite can be applied for a number of
monitoring tasks, which require spatial resolution between 30 centimetres
131
and 3 metres. Applications might exploit either one sensor type only (e.g.,
photographic cameras for mapping under fair weather conditions), or take
advantage of jointly exploiting different sensor types through sensor fusion
(e.g., photographic and thermal images for urban heat loss studies). The
study of vegetation is currently hampered by the lack of colour and
multispectral information. Table 7.1 gives our estimate of the usefulness of
current sensors for different monitoring tasks. SAR systems will be
particularly useful for situations where 3-metre resolution is sufficient but
all-weather capability is necessary.
1
Table 7.1: Estimated Potential of Current Open Skies Sensors at
Treaty Resolution for Different Monitoring Tasks
The estimated potential is indicated by the number of stars.
+ With sufficient image overlap for stereo viewing and panchromatic film.
The bottom part of Table 7.1 addresses the monitoring of emergencies.
Here thermal images will allow for a very good 24-hour coverage of fires
and other heat releases (e.g., from reactor accidents). SAR sensors can spot
the extent of flooding at day and night. Although rated lower, photographic
images will provide accurate baseline information on the spatial impact of
environmental disasters.
Photo Camera
+
Thermal
Imager
SAR
Mapping and urban planning
Urban heat losses
Vegetation, crops
Water supplies
Soils
Air Quality
Fires
Floods
Earthquake and hurricane
damage assessment
Nuclear reactor accidents
***
*
*
*
*
--
*
**
**
*
**
***
**
**
**
--
***
**
**
***
*
--
*
**
**
--
**
***
*
*
132
7.3.2 Past Environmental Flights of Open Skies Aircraft
So far, the United States, Germany and the Czech Republic have used
their Open Skies aircraft successfully in four extensive missions for pre-
disaster and post-disaster monitoring.
(a) Oder flood (1997)
In July and August 1997 a major flood of the river Oder affected the
Czech Republic, Poland and Germany. The Oder demarcates the border
between Poland and Germany.
To assist the regional Polish government and the German state of
Brandenburg in estimating the extent of the catastrophe the German Open
Skies aircraft was assigned for a training mission to take images of the
flooded areas. The aircraft mapped the full river area from the influx of the
river Neiße to the Baltic Sea twice within a ten-day period. The black and
white photographs were developed over night and were handed over to the
governments of Poland and Brandenburg on the next day. Photo 7.1 shows
an example of the images taken. It should be noted that environmental
satellites like ENVISAT can hardly provide imagery at such short notice due
to their long revisit times (typically 2-4 weeks) and data processing times
(between days and weeks).
However, even with a capacity to deliver data within 24 hours Open
Skies still faces competition in these sorts of missions as demonstrated by
the following example. During a major flood in the Czech Republic and in
Saxony (Germany) in August 2002 aerial monitoring was performed not by
Open Skies aircraft but by Tornado reconnaissance aircraft of the German
Air Force.
(b) Hurricane Mitch (1998)
In November 1998 one of the US Open Skies aircraft was sent to
Central America, shortly after the devastating impact of hurricane Mitch.
The territories of Honduras and Nicaragua and other areas were mapped in
five missions at a resolution of 20-30 centimetres. Copies of the images
were provided to the governments concerned and to a major US relief
organization.
2
The imagery was used for relief and preventive measures.
133
This underlines the potential of Open Skies assets for disaster monitoring on
short notice.
(c) Pre-disaster Monitoring in the Caribbean (2000)
Damage assessment through change detection depends equally on the
availability and quality of imagery taken prior to the disaster. Since the
Photo 7.1: Oder flooding near Wiesenau (Germany) as photographed from
the former German Open Skies aircraft. Courtesy: Zentrum für
Verifikationsaufgaben der Bundeswehr, Geilenkirchen.
134
Caribbean is hit by several hurricanes each year the US Southern Command
has asked for an extensive monitoring flight series by the US Open Skies
aircraft over most of the Caribbean islands (with the exception of Cuba and
Haiti which did not permit overflights). Images were taken with an overlap
of 56% (for stereo viewing) at a resolution of 70 centimetres. Image data
were recorded on magnetic tape by a digital module attached to the camera
KS-87.This allowed on-line display and eased data duplication. Copies
were distributed by the US Southern Command to the states overflown.
(d) Forest damage from Hurricane Lothar (2000)
Shortly after Christmas 1999 large parts of France and Southern
Germany were severely devastated by hurricane Lothar. In February 2000
Germany and the United States used a scheduled trial observation flight
with the US Open Skies aircraft for assessment of the damaged forest areas.
The target areas were selected after coordination with the ministries of
forestry of the states of Bavaria and Baden-Württemberg. Similarly forest
damage areas in Bavaria were photographed by a Czech-German Open
Skies trial flight using the Czech Open Skies aircraft during a scheduled trial
observation flight a few weeks later.
In summary, the four missions underlined the advantages of using
Open Skies resources for disaster monitoring (large area coverage at
adequate resolution, short reaction and image delivery times). In the future,
the acceptance of disaster monitoring by foreign Open Skies aircraft will be
eased by the fact that Open Skies aircraft have certified sensors with known
capabilities. States parties, which have an observation aircraft of their own,
should consider preparing the aircraft for quick modifications for such
flights beyond the restrictions of the Treaty, for example, by making
provisions for providing digitalized photographic image data or the
inclusion of additional sensor categories.
7.3.3 Additional Sensors in Support of Environmental Monitoring
The current sensor set was optimized for identification of major
weapon systems
3
and of military infrastructure rather than for
environmental monitoring. However, the Treaty contains the possibility of
including additional sensors at a later stage. According to Article IV “the
introduction of additional (sensor) categories and improvements to the
capabilities of existing categories of sensors provided for in this Article shall
135
be addressed by the Open Skies Consultative Commission….” In
accordance with Article X proposals on the inclusion of additional sensors
may be decided upon at any time. The first regular review conference in
2005 seems to be a good opportunity to do so.
For purposes of environmental monitoring the most cost-effective
upgrade would be the inclusion of colour infrared film (CIR) as a recording
medium of photographic cameras. Colour infrared film is sensitive to
radiation in the near infrared at wavelengths from 0.7 to 0.9 micrometres.
It has been and still is extensively used in civil monitoring of urban areas,
vegetation, soils, water supplies, etc. Such film provides also much
improved contrast and recognition potential for objects in built-up areas.
Although three times more expensive than panchromatic (black and white)
film, use of such film will raise the overall costs of an observation flight only
by a small fraction. Development of CIR film for a reliable product,
however, is a delicate and costly skill, which is mastered only by a few
specialized firms and agencies. Not every Open Skies state party will have
this capability.
As a next step multispectral digital imaging sensors (multispectral line
scanners) have to be considered for environmental monitoring.
4
Again, civil
multispectral sensors, like the Thematic Mapper on the LANDSAT satellites
(featuring seven spectral channels) and their airborne equivalents have
been used extensively in wide area multispectral monitoring of urban areas,
agricultural areas, nature reserves and geological sites for more than two
decades.
The US Defense Nuclear Agency has performed a comparative study,
which evaluated the potential benefits of multispectral and hyperspectral
sensor additions for Open Skies missions.
5
The study considered
multispectral imaging sensors to be the most beneficial addition both for
environmental monitoring and for military Open Skies objectives (e.g.,
camouflage detection). As a low risk approach, the study recommended the
Daedalus Thematic Mapper ATM. This sensor has 11 spectral channels at
wavelengths from 0.4 to 12.5 micrometres and an instantaneous field of
view of 1.25 milliradians in one of two possible operation modes, providing
a ground sampled distance of 2.5 metres at a flight altitude of 2,000 metres.
Multispectral sensors will enhance considerably the potential for monitoring
of urban areas, agricultural land, forests, rivers and special problem areas
like waste deposits. Hence, they open the door to detailed studies of land
136
use and the state of the environment. Digital multispectral imagery is also a
very useful basis for computer based semi-automatic change detection as
discussed in section 6.3. Needless to say analysis of multispectral digital
imagery requires some investments and expertise in computing
atmospheric corrections and geocoding. In addition, the quality of analysis
will improve a lot when airborne data are complemented with ground truth
data.
In a few years the role of multispectral line scanners will be largely
taken over by multispectral digital stereo cameras with resolution in the 0.1
to 1-metre range. Such cameras combine the virtues of panchromatic
stereo imaging with multispectral recording, however at the price of a more
demanding digital image processing.
A large number of non-imaging sensors have been discussed at the
informal seminars on the possible use of the Open Skies regime in the field
of environmental monitoring which were mentioned above. We want to
emphasize here only two types of devices:
a) Lidar (laser reflection measurements) for detection of atmospheric
composition and pollution;
b) Air samplers as a basis for the detection of radioactivity in the
atmosphere.
7.3.4 Application Scenarios and Institutional Questions
In spite of initial enthusiasm the interest in an extension of the Open
Skies regime into environmental monitoring has declined, at least among
the governments of major states parties like Germany and the United States.
The reason is quite obvious. Many states parties of the Treaty have
adequate airborne and spaceborne monitoring devices in the public and
commercial sectors for monitoring of the environment.
The respective monitoring aircraft are usually much smaller than
current Open Skies aircraft and hence can be operated at lower cost. There
is also an element of inter-agency competition. In consequence, several
basic questions have to be answered first, before environmental application
scenarios can be developed:
137
• What kind of environmental situation in state A could motivate state B
to perform a dedicated environmental monitoring flight in state A
under the Open Skies regime in spite of the cost?
• Which kind of environmental monitoring tasks would exploit and
require the special “virtues” of the Open Skies regime (unlimited
territorial access, short response time, priority over any other air traffic)?
• Are dual-use flights, which would cover both military and other target
areas in one go, negotiable? Such flights would be most cost-effective.
• Who would be responsible for requesting and analyzing data from
environmental Open Skies flights? Who would bear the costs?
• To what extent can data from environmental monitoring flights under
Open Skies be made fully open and accessible, for example, to
researchers and local users?
Let us discuss several application scenarios in the light of these
questions:
Environmental Emergencies
Certain emergency situations and disasters in state A could justify a
monitoring flight by state B under Open Skies, if:
• the impact is of border crossing nature, like the radioactive plume of
the reactor catastrophe in Chernobyl;
• humanitarian reasons require rapid response from the outside, like a
major earthquake or flood damage, which cannot be dealt with by
local/national resources.
Here, one can assume sufficient interest on the side of state B. Flight
costs would have to be covered by state B from funds for international
emergency situations. Alternatively, state B might dedicate one of its
obligatory national training flights to the disaster-monitoring mission.
Cross-border Environmental Problems
Certain environmental problems and management tasks are of a cross-
border nature (like pollution and flood control of border crossing rivers,
salination and wind erosion in arid areas, effects of acid rain, etc.). A
particular example is the state of nuclear reactors and waste sites of the
Russian Northern fleet, which are of concern to the Scandinavian countries.
Such flights under Open Skies will be only attractive if civilian monitoring
capacities are lacking, or civilian monitoring agencies of state B do not have
138
full territorial access to state A. In addition such flights might become
attractive if they can be arranged in a cost-effective, dual-use way serving
both military and civilian purposes.
Here, a mutual interest of two or more states parties affected by the
particular environmental problem can be assumed. Hence, it would be
natural to share the flight costs or to arrange flights on a reciprocal basis.
Within each state cost sharing between military and civilian users has to be
clarified also as well as mechanisms for mission request, mission planning,
shutter control and data distribution.
Verification of International Environmental Conventions
Several international environmental conventions have been concluded
or are in preparation (like the Montreal Ozone Protocol, the Climate
Convention, the Convention on Biodiversity, etc.). At present some of these
conventions lack agreed verification procedures based on satellite or
airborne monitoring. However, airborne multispectral monitoring under
Open Skies could make useful contributions in situations where good
spatial and spectral resolution matters.
7.3.5 Conclusions on Environmental Monitoring
The Open Skies regime opens interesting avenues for environmental
monitoring, in particular through:
• data fusion from different sensors (photographic cameras, thermal
imager, SAR);
• inclusion of colour infrared film and eventually also digital multispectral
imaging sensors;
• inclusion of non-imaging sensors.
In competition with other data sources (civil airborne and satellite
monitoring) application scenarios should concentrate on areas which relate
to the main intentions and virtues of the Open Skies Treaty, in particular
confidence building and management of (environmental) crises in a
cooperative way. This in turn means, applications for the monitoring of
• environmental emergencies and disasters;
• border crossing environmental problems; and
• verification of international environmental conventions
139
should be considered first and studied in more detail. Past flights have
demonstrated that Open Skies aircraft can make valuable and competitive
contributions to disaster monitoring. Challenging institutional questions
have to be solved in order to make best use of Open Skies in the other two
areas.
6
It should be emphasized, that the Open Skies Treaty allows for mutual
voluntary agreement between states parties on observation flights without
the restrictions imposed by the Treaty on sensor performance and mission
procedures. Thus, states parties have the opportunity to agree on flights for
environmental monitoring at performance conditions adapted to the
environmental problem. An initial first step in this direction would be flights
for environmental monitoring under the full rules of the Treaty. Apart from
such flights under an extended Open Skies regime, each state party, which
operates an OS aircraft, is free to use its capacity for national monitoring
and mapping flights. This route is followed, for instance, quite successfully
by Bulgaria and Hungary.
7.4 ADDITIONAL SENSORS AND HIGHER RESOLUTION
As already stated the OSCC can decide on the introduction of
additional sensor categories and on improvements of the capabilities of
existing sensor categories. This issue might be tabled at the review
conference of the states parties in 2005. The outcome will depend primarily
on the political will of the parties to develop and adapt the Treaty
implementation to current concerns and technical options. We, therefore,
will discuss the question of additional sensors and capabilities from the
perspective of these two criteria.
7.4.1 Demands from Current Security Concerns
K. Arnhold has argued vividly that in order to survive and flourish in the
longer term the Open Skies regime has to be adapted to current security
concerns:
7
• proliferation of weapons of mass destruction;
• crisis prevention and management;
• terrorism and illegal trafficking of humans and goods.
140
Let us assume for the time being that the states parties will agree on
such a reorientation. What kinds of sensor improvements are conceivable
and recommendable?
(a) Resolution
The current resolution limit of 30 centimetres of photographic cameras
has been set to allow for the identification of major weapons systems, but
not for detailed analysis. Given the objective of monitoring illegal trafficking
and proliferation a finer resolution of optical cameras in the order of 10
centimetres would be desirable. We doubt, however, that all the states
parties would agree to this. Still, a significant improvement of the
monitoring potential can be obtained by reducing the resolution values of
the other two sensor categories (thermal imaging and SAR) towards 30
centimetres.
This would respect the concerns of going beyond the identification of
major weapons systems, but would largely improve the ability of all-
weather, day-and-night monitoring. Such an improvement in resolution of
thermal infrared sensors from 0.5 to 0.3 metres could be achieved quite
easily by flying at lower altitudes. In contrast, the improvement of SAR
resolution from 3 to 0.5 metres would require a major investment in a
sophisticated technology, which, however, is commercially available.
8
(b) Sensors
At the Budapest negotiation round in 1990 the following additional
sensors were considered (but ultimately not retained): air samplers, electro-
optical (digital) cameras, gravitometers, magneto-meters, multispectral
scanners and spectrometers.
Seen from the above security concerns the first priority should be to
have operational thermal infrared line scanners in 2006, when the Treaty
allows for their use. Beyond that, the highest priority should be given to
multispectral scanners or multispectral digital cameras in order to provide
much improved identification of vegetation, camouflage and illegal
trafficking as discussed in section 7.3.3. Crisis prevention and management
would clearly benefit from the introduction of such multispectral sensors,
for example, for the monitoring and identification of freshly laid mine fields,
refugee movements and camps, and damage assessment.
141
Infrared line scanners, air samplers and fluorescent laser detection of
chemicals could support the monitoring of illegal production and storage of
chemical weapons and the monitoring of nuclear facilities. In contrast, the
production of biological weapons can hardly be identified from the air.
7.4.2 Technology Drive
The trend in commercial remote sensing is moving towards:
• higher resolution;
• replacement of film by digital sensors;
• provision of four colour channels (blue, green, red, near infrared) as a
standard.
At the same time GIS, which are built on digital maps and digital image
data are becoming the basis of planning and surveying processes as well as
of integrated data management. The military and the verification
communities will sooner or later have to embark on this leg of the digital
revolution.
9
Hence, the question of including digital cameras and of
exchanging all image data in digital form is bound to become an issue
among the Open Skies states parties.
In the area of SAR the trend runs towards higher resolution, multi-
polarization and multi-frequency sensors.
10
At present, non-military SAR
sensors are reaching a resolution in the range of 0.5 metres (air) to 1 metre
(satellite). An Open Skies SAR at 3-metre resolution is no longer
competitive.
7.5 DATA ACCESS FOR NON-MILITARY ORGANIZATIONS
Although the Treaty advocates openness and confidence building, the
access to data from Open Skies flights is strictly limited to the level of
governments. So far image data cannot even be transmitted to international
organizations concerned with confidence building and verification like the
OSCE or the Organization for the Prohibition of Chemical Weapons
(OPCW) without the consent of the states parties overflown.
Clearly this should be reconsidered in particular when applications for
crisis monitoring and crisis prevention will be on the agenda. For example,
142
the field missions of the OSCE would greatly profit from receiving Open
Skies images of their area of concern, which can be used as accurate maps.
The same holds for support of on-site inspections under the Chemical
Weapons Convention (CWC).
11
Should pre- and post-disaster monitoring be included in the Open
Skies agenda, data access by non-governmental relief organizations would
have to be agreed upon.
7.6 ISSUES AND CHALLENGES FOR THE REVIEW CONFERENCE OF
STATES PARTIES 2005
According to Article XVI of the Treaty the depository states are obliged
to convene the first regular review conference of states parties three years
after entry into force, that is, in 2005. The first review conference will be an
excellent and fitting opportunity to review and eventually adapt the Treaty
implementation. Unlike the CFE Treaty and the Vienna Documents the
Open Skies Treaty has not been adapted since its signature in 1992. On the
other hand, the Treaty is an extremely flexible legal instrument. The OSCC
has a wide range of possibilities to make any necessary changes or
modifications to the Treaty implementation, like deciding about new
sensors or other fields of application. Most of the foreseeable and desirable
adaptations can be decided upon without the need of another ratification
process. In addition, the Open Skies Treaty offers as a general rule in many
cases the principle of “unless otherwise agreed” (see Article VI, Section I,
paras. 9, 10, 11, 20), which means that states parties may deviate from
standard Treaty rules and regulations if they agree bi- or multilaterally. This
has resulted in a very pragmatic and effective implementation of the Treaty.
Still, the review conference itself offers an incentive to re-evaluate the
provisions of the Treaty and its implementation from a political and
technical point of view. The conference will be an occasion to adapt the
Treaty practice to new challenges and security needs. Needless to say, the
outcome of the conference will depend a lot on the conceptual work done
in the preparation process. Below we discuss as food for thought a number
of issues and challenges, which should be properly addressed in the
preparatory work. The following points result from the discussions of this
chapter and follow up also on some questions, which were touched in the
previous chapters.
143
Quota distribution
The initial allocation of active quota as shown in Table 2.2 was shaped
by the situation at the end of the Cold War: most of the Western states were
interested in overflying Russia, the Ukraine and other former members of
the Warsaw Treaty. NATO member states have agreed not to overfly each
other in the framework of regular quota flights. This created an imbalance
in the flight distribution, which will be sharpened by the recent NATO
extensions.
It is therefore appropriate to rethink the rationale behind the quota
distribution. In spite of the high degree of transparency obtained in Europe
and North America it is important to appreciate the Open Skies Treaty as
an insurance policy for rougher times. One has, after all, to be prepared for
the unexpected. The instrument of confidence building in security matters
through cooperative overflights should therefore be maintained and
cultivated on a basis of equity and reciprocity.
Hence, we suggest a baseline of a relatively small number of mandatory
quota flights on an equitable basis. This would include also mutual
overflights of NATO member states. Naturally, the geographic size and
military potential should be reflected in the number of allocations, too.
As a second tier a set of supplementary flights is needed, which would be
assigned at relatively short notice to support:
a) Conflict prevention and crisis management;
b) Non-proliferation and global arms control; and
c) Monitoring of environmental disasters.
There should be an upper ceiling on the number of such flights, which
could but need not be fully exploited.
Support of Conflict Prevention and Crisis Management
In order to make best use of Open Skies flights for support of conflict
prevention and crisis management a number of steps and institutional
arrangements need to be considered:
a) Accession of as many OSCE participating states as possible, in
particular through diplomatic efforts to support the accession of all
Central Asian republics as well as all the states in the Western Balkans
and the Caucasus;
144
b) Arrangements with the OSCE on the possibility of the OSCE to request
Open Skies support. Provided the state overflown agrees, the
requested image material should be made accessible to the OSCE
Conflict Prevention Centre and its respective field missions. Such
arrangements are logical, since all Open Skies states parties also
participate in the OSCE;
c) Apart from that, Open Skies states parties or groups of states should be
free to provide their Open Skies aircraft and assets for support of
conflict prevention and crisis management by other international
organizations like the United Nations, NATO, the European Union or
other international organizations with a security mandate;
d) Contributions of Open Skies flights to monitor illegal trafficking of
weapons, drugs and people across borders.
Open Skies in support of global arms control and non-proliferation of
weapons of mass destruction
As argued above Open Skies flights have the technical potential of
supporting the verification of global arms control treaties, which limit or ban
weapons of mass destruction, in particular nuclear and chemical weapons.
The main contribution in this respect comes from the 50 centimetres
resolution of thermal infrared image sensors. Already now the states parties
can exploit their active quota to monitor suspect nuclear and chemical
production sites or waste storage sites in the Treaty area.
In order to give Open Skies a stronger role in this field it is desirable
that arrangements be made with the International Atomic Energy Agency
(IAEA) and the Organization for the Prohibition of Chemical Weapons
(OPCW). The IAEA has begun to use satellite images as a supplementary
information source in verifying the Non-Proliferation Treaty (NPT). In
contrast, aerial inspections or joint exploitation of satellite imagery is not yet
foreseen by the CWC. An appropriate adaptation is desirable.
Open Skies flights can support the verification of the Global Exchange
of Military Information. This is an additional data exchange under the
Vienna Documents. It covers all kinds of weapons systems including naval
vessels and naval aircraft of all OSCE participating states regardless of their
deployment site, worldwide. The exchange is not verified by on-site
inspections. Open Skies flights could be used to verify the notified forces in
most of the deployment sites, in particular naval forces, which are not
subject to inspection under the CFE Treaty.
145
Monitoring of environmental disasters
States parties should make arrangements for being prepared to launch
supplementary flights for the monitoring of environmental and industrial
disasters on short notice in coordination with the OSCC. To avoid
budgetary complications, such flights should be preferably performed as
national training flights.
Additional sensor types
The technological development and potential applications in the
above areas suggest the consideration of at least three types of additional
sensors and film:
• infrared sensitive film (false colour film), which is essential for the
monitoring and evaluation of vegetation;
• digital cameras, which are becoming state of the art in commercial
aerial photography;
• SAR at 1-metre (or better) resolution, in order to be competitive with
commercial radar satellites.
Trial operation of such devices by single states or a group of states
parties can help to promote their future inclusion.
New and jointly operated aircraft
Most of the existing Open Skies aircraft are older than 20 years and
have to be replaced within the next decade. In addition, major sensor
upgrades are due in 2006 when the full sensor set becomes mandatory for
states, which choose to apply the taxi option. This offers the opportunity to
consider the introduction of joint aircraft, which are equipped and
operated by several states, or a larger group of states. The Pod Group and
the Swedish-German aircraft cooperation are two examples along these
lines.
In the longer run it would be desirable to make confidence building
and conflict prevention through Open Skies flights part of a common
foreign and security policy of the European Union. A pool of two or three
aircraft operated by a “coalition of the willing” would be a first step into this
direction. Such groups of states parties would also coordinate their plans for
exploiting their active quota.
146
Cooperative data analysis
Although the ten years of trial implementation have been quite
successful, the exploitation of the image data obtained has been less than
optimal. Open Skies images have been analyzed by and large by analysts,
without the aid of a modern system of data integration. So far states process
and handle their imagery within classified systems and scarcely make data
and results available to each other. States still implement arms control as a
national prerogative and are reluctant to exercise real cooperation in data
analysis. One main future objective should be to achieve more real and
substantial cooperation and efficiency in the field of data analysis for arms
control verification, conflict prevention and transparency building.
Coalitions of the willing should instigate such joint action and set the
example.
Analysis centres should be run independently of the respective military
(strictly secret) intelligence services in a semi-open fashion.
12
This would
enable to share results with international organizations in charge of conflict
prevention, peacekeeping and post-conflict management, like the United
Nations or the OSCE. Modern data integration methods and the
combination of Open Skies imagery with image data from (open)
commercial satellites as well as with adequate ground-based data would be
the basis for producing first-class results at lower overall cost than the
present national operations.
Certification procedures
Finally it will be a major task for the OSCC and its IWG on Sensors and
Certification to finalize the certification procedures for infrared and SAR
sensors. The complex task of certifying infrared sensors can be eased by
adopting a procedure similar to one introduced for optical cameras: states
parties provide a so called MRT-curve, which characterizes the resolution
properties of their infrared sensors. This curve would then only have to be
checked and confirmed at the certification event. In addition, the
certification procedure for video cameras has to be re-discussed, because
the present procedure based on Decision 14 is valid only until 31
December 2005.
The preparation process for the 2005 review conference can build on
a crucial prerequisite: professional Open Skies capabilities, in particular
highly skilled specialists. The outstanding cooperation, which the Open
147
Skies experts of all participating nations have demonstrated frequently
during joint certifications and similar events, forms an excellent basis for
developing the Open Skies Treaty implementation further.
7.7 SUMMARY
In summary, the Open Skies regime can and should be adapted to
current security needs and technological trends. Many of the
recommended adaptations can be arranged within the legal framework of
the existing Treaty, in particular:
• Inclusion of additional states parties in crisis-prone regimes of the
application area (e.g., former Yugoslavia, Caucasus states, Central Asian
republics);
• Applications for conflict prevention, crisis management and support of
non-proliferation of weapons of mass destruction within the Treaty area
of application;
• Monitoring of environmental disasters and cross-border environmental
problems based on mutual voluntary agreement of the states parties
involved (within the Open Skies framework).
Most of the recommended technical adaptations require a Decision of
the OSCC but not any further ratification steps, in particular the inclusion
of additional sensors and readout media (multispectral, electro-optical,
colour infrared film, Laser Fluorescent Spectrometers, digital readout of
photographic cameras).
It is also desirable to adapt the resolution limit of thermal infrared line
scanners allowed under the Open Skies Treaty to 30 centimetres. The latter
change can be implemented without any additional cost simply by adapting
the required minimum flight altitude.
Notes
1
The usefulness of different SAR systems for environmental monitoring
is further discussed, for example, by Schmullius and Evans. See
C. Schmullius and D. L. Evans, “Synthetic Aperture Radar Frequency
148
and Polarization Requirements for Applications in Ecology, Geology,
Hydrology and Oceanography—A Tabular Status after SIR-C/X-SAR”,
International Journal of Remote Sensing, Vol. 18, 1997, pp. 2713-22.
2
B. F. Molnia and C. A. Hallam, “Open Skies Aerial Photography of
Selected Areas in Central America Affected by Hurricane Mitch”,
Internal Report, US Geological Survey, 1999.
3
In the CFE Treaty the respective systems are called Treaty Limited
Equipment (TLE).
4
During a workshop of Open Skies experts on 2 April 2003 at Ingolstadt,
Germany, Russian experts gave a presentation of a new Russian
infrared line scanner for Open Skies use, which would work both
within the 9-12 and 0.5 -1.1 micrometer ranges.
5
R. Ryan, P. Del Guidice, L. Smith, M. Soel, N. Fonneland, M. Pagnutti,
R. Irwin, and P. Saatzer, “U.S. Open Skies Follow-On Sensors
Evaluation Program, Multispectral Hyperspectral (MSHS) Sensor
Survey”, Proceedings of the Second International Airborne Remote
Sensing Conference and Exhibition, San Francisco, Environmental
Research Institute of Michigan, Ann Arbor, Vol. I, 1996, pp. 392-402.
6
For example, loitering is prohibited in regular Open Skies flights but
might be desirable for environmental monitoring. Exceptions are
possible by bilateral agreement according to Article VII, Section II,
para. 4e.
7
K. Arnhold, “Der Vertrag über den offenen Himmel—Ein Konzept zur
Aktualisierung des Vertrages”, SWP Studie, Berlin, June 2002, p. 21.
8
For example, the AER-II SAR currently operated from a C-160 of the
German Air Force flight test centre at Ingolstadt could be used as a test-
bed. See www.fhg.fgan.de.
9
The US military already has gone quite a way in this direction.
10
Multi-polarization and multi-frequency SAR sensors emphasize
different features of a scene similar to the multicolour displays of
multispectral sensors.
11
For the time being aerial inspections are not foreseen in the CWC.
However, governments can provide “supplementary information”
derived from their own sources in support of verification.
12
It is understood that sensitive information related to criminal or
terrorist activities would be handled with sufficiently high
confidentiality.
149
CHAPTER 8
REGIONAL APPLICATIONS
OF THE OPEN SKIES APPROACH
Márton Krasznai, Hartwig Spitzer and William Wynne
8.1 DETOUR: THE HUNGARIAN-ROMANIAN BILATERAL
AGREEMENT ON OPEN SKIES
Well before the signature of the multilateral Treaty on Open Skies on
11 May 1991, Hungary and Romania became the first states ever
successfully to negotiate and sign a bilateral Open Skies agreement.
The dramatic changes that had taken place in Europe since the revival
of Open Skies in May 1989 made this agreement possible. The collapse of
the Warsaw Treaty also brought about the end of the old European security
structure based on the existence of the two alliances and the approximate
balance between their armed forces. Instability became the most dramatic
negative side effect of the profound political changes in Central and Eastern
Europe. Earlier arms control agreements lost much of their relevance in the
light of the new political realities. Participants in the various arms control
fora acted cautiously in this new state of flux. The limitations set in the CFE
Treaty aimed at redressing the East-West balance but were not particularly
relevant to the new sources of military instability in Central and Eastern
Europe, which emerged with the disappearance of the East-West divide.
In this new political and military situation the states of Central and
Eastern Europe had a choice: either to renationalize their defence policies
with an emphasis on military force, which would further destabilize their
region, or to start building a new, cooperative security structure consisting
of trans-Atlantic, pan-European, regional, sub-regional and bilateral
arrangements. In view of the strong desire of the Central and Eastern
European states to align themselves with the West after the end of the Cold
War, their preference for building a cooperative security structure was
150
evident. Thus, many of the policies of these states at the time were a
reflection of their genuine readiness to share the values and security
concepts of the West and to demonstrate their ability to cooperate. The
bilateral Hungarian-Romanian Open Skies Agreement was an innovative
step in this respect. Furthermore, it proved helpful in promoting the idea of
Open Skies on a European scale.
Soon after the second Budapest round of the Open Skies conference,
Romania proposed to Hungary to commence negotiations on a bilateral
Open Skies agreement. Hungary did not at that time accept the Romanian
offer. As designated host of the third round of the Open Skies conference
(which was then actually held in Vienna, and, of course, not hosted by
Hungary), it made every effort possible to help bring about the continuation
of the multilateral negotiations. Hungary was well aware that a multilateral
regime would offer several obvious advantages: among others, it could
serve as an additional verification regime for existing and future arms
control agreements or as a framework for regional crisis-management
application of observation flights.
Only in January 1991 did Hungary agree to take up the renewed
Romanian proposal. By that time it was clear that an accord on a
multilateral Open Skies was impossible unless the Soviet Union was ready
to shift its position and stop blocking agreement. The decision to begin
bilateral Open Skies negotiations did not mean that Hungary had
abandoned the idea of a multilateral regime. In 1989, the Hungarian
delegation to the negotiations on confidence- and security-building
measures in Vienna introduced the concept of “amplified confidence-
building measures” in the relations of neighbouring countries. A bilateral
Open Skies regime, functioning in parallel with a multilateral regime, was
seen as a good example of the realization of this concept. Furthermore,
both countries believed that the creation of a bilateral regime could
demonstrate the viability and utility of Open Skies and help convince other
participants to continue discussions.
The signing of an Open Skies agreement between Hungary and
Romania had special political significance. The two countries had,
regrettably, a history of strained relations. For three and a half decades their
membership in the Warsaw Treaty had prevented Hungary from properly
addressing the problems, which the sizable Hungarian minority living in
Romania was facing. After the disbanding of the Warsaw Treaty, however,
151
tension between the two countries threatened to resurface. The willingness
to establish a regime which would ensure a high degree of openness and
transparency in their military activities and serve as an effective tool of
confidence building showed the determination of responsible politicians
both in Hungary and in Romania to solve their problems exclusively by
negotiations—either bilateral or within the framework of relevant
institutions such as the Council of Europe or within the CSCE (now OSCE).
Negotiation of the Agreement
The Romanian delegation arrived in Budapest mid-February 1991.
The first round of the negotiations was very intensive and fruitful. The fact
that both delegations presented drafts of the bilateral agreements on the
basis of the draft Treaty text of the second round of the Open Skies
conference facilitated matters. The main body of the agreement was
worked out in three days. The second round of the negotiations, held in
Bucharest in mid-March, was equally intensive and effective. In two and a
half days the delegations agreed on all eight annexes to the Agreement.
1
The Hungarian and Romanian negotiators agreed from the very
beginning that the regime to be created should be simple and yet effective
and should match the technical and financial resources of the two
countries. The provisions of the Agreement are therefore flexible and take
into account the requirement of cost-effectiveness to a great extent.
According to the quota annex, the two countries have a right to carry
out four flights a year in each other's airspace. This flight quota is quite
substantial if one takes into account the fact that the bilateral regime
continues to function even after the multilateral Open Skies agreement
enters into force. An observation flight is restricted by the following,
whichever applies first: a maximum duration of three hours or a maximum
distance of 1,200 kilometres. This was calculated on the average speed (400
kilometres/hour) and range (2,000 kilometres) of the designated
observation aircraft. A Hungarian observation aircraft using the nearest
point of entry (Timisoara) can carry out an observation flight in Romanian
airspace and return to Budapest without refuelling. However, if it uses the
more distant point of entry (Bucharest), refuelling is necessary.
Both countries use their existing observation aircraft for the purpose of
Open Skies flights: Romania uses the Soviet-made medium-size, two-
152
engine turboprop AN-30 (which is a specialized version of the AN-24
transport plane, modified for aerial photography), while Hungary uses the
two-engine turboprop AN-26 transport plane. The observing party may use
its own aircraft or an aircraft of the observed party. The right of choice
belongs to the observing party. As both countries have only a few aircraft
suitable for Open Skies purposes, the observing party must submit a request
seven days in advance if it intends to use an aircraft of the observed party.
The sensor annex specifies only the two sensors Hungary and Romania
had at the time of signature: aerial cameras and video camera. This annex
can, however, be updated if the parties wish to introduce further sensor
categories. In view of the possibility of upgrading the sensors, the parties
undertook the obligation to use similar sensors of comparable capability
and to facilitate access to such sensors for use by the other party. The
ground resolution of the sensors is not limited. This flexibility allows the pilot
of the observation aircraft to fly as low as flight safety requirements permit.
2
Low-level flights might be necessary in cloudy weather, when the only way
to take photographs is to fly under the cloud cover.
Request for an overflight is to be submitted 24 hours in advance and
shall be accepted promptly, unless force majeure prevents the party to be
overflown from receiving the observation aircraft. The Agreement
guarantees that overflights are as unrestricted as possible. Hazardous
airspace must be publicly announced in the aeronautical information
publication. If the observed party requests overflight of hazardous airspace,
such as airspace over nuclear-power stations, chemical plants or exercise
grounds where a firing exercise is taking place, the observed party may
specify the minimum safe altitude, may propose an alternate flight route as
near to the hazardous airspace as safety requirements permit, or may
propose a change in the timing of the overflight in the hazardous airspace.
The observed party has the right to inspect the aircraft and its sensors.
According to the Agreement, this pre-flight inspection may last no longer
than eight daylight hours and shall terminate no later than three hours prior
to the actual commencement of the observation flight. This timeframe had
been agreed with the intention of ensuring that the observation aircraft
would have to spend no more than two days on the territory of the observed
party. If the aircraft arrives in the observed country in the morning, the
inspection is finished by nightfall, the observation flight then takes place the
153
next morning, and the processing of observation materials (e.g., film) can be
completed in the afternoon of the second day.
Information sharing is ensured by the use of double cameras that
record two identical images of each scene. The two rolls of film are
developed by a joint team of the two parties at the end of the observation
flight in an established ground facility of the observed party. One negative
is taken by the observing party while the other remains with the party that
has been overflown. If a double camera is not available, the negative is
copied and the copy is taken by the party which carried out the observation
flight. The same applies to video cameras.
The Agreement establishes a Hungarian-Romanian Open Skies
Consultative Commission. The Commission’s task is to solve any dispute
that may arise in the course of the implementation of the Agreement. The
Commission is also responsible for updating the annexes on sensors, on the
flight quotas, and on entry and exit points. In the hypothetical case that a
party discovers, as a result of an overflight, disturbing or unusual military
activity on the territory of the other party, the issue may also be raised in
the Commission.
The Demonstration Flight
Hungary and Romania carried out a demonstration flight on 29 June
1991 not much after the signature of the bilateral Agreement.
Representatives of all countries participating in the multilateral Open Skies
discussions were invited to take part as observers in the trial flight. The
purpose of the flight was threefold: to demonstrate to other participants of
the Open Skies negotiations the viability and utility of Open Skies, to take
advantage of the confidence building potential of bilateral overflights even
before the Agreement entered into force,
3
and to test in practice the
solutions to various technical problems which were worked out in the
course of the negotiations.
Before the Agreement was signed, France had offered technical
assistance to both countries. France provided Hungary and Romania with
double cameras and automated film-developing machines. A French team
of technicians was sent to Budapest and Bucharest to install the cameras
and the apparatus and to train local personnel in their use.
154
The Romanian AN-30 aircraft arrived at Tököl military airport outside
Budapest on 28 June. Hungarian officials accompanied by Romanian
escorts inspected the aircraft to make certain there were no hidden sensors
on board of the aircraft. The inspection team divided itself into three sub-
teams: one sub-team inspected the fuselage, the second group the avionics,
the third group the sensor, a French-made OMERA-33 aerial camera.
4
The
team was able to check the aircraft thoroughly in less than three hours
because of the fact that the Hungarian air force had used this type of aircraft
for decades. After completion of the inspection the aircraft took off for a
short flight to calibrate and check the sensor. When the inspection and the
test flight were over, the Romanian crew filed a flight plan and the
Hungarian officials approved it. Before the aircraft took off the next
morning, Hungarian officials briefed the Romanian crew on expected
weather conditions and flight safety and navigation regulations.
The Romanian AN-30 aircraft took off at 8:00 a.m. on 29 June. The
observers followed the observation aircraft on board of an AN-24 transport
plane flown by the Hungarian air force. Following the flight plan, the
observation aircraft flew over various militarily significant objects: a civilian
airport, a military college with heavy armament openly displayed for this
occasion, an abandoned Soviet military airport and an exercise ground.
After a short technical landing on Romanian soil, the demonstration flight
was continued over Romania. There the observation plane flew over a
military airfield, a training ground, an ammunition depot and a railroad
junction.
Weather over Hungary was fair, with a cloud base at 2,000 to 3,000
metres. The plane therefore flew at an altitude of 1,500 metres.
Photographs taken from this altitude had a ground resolution of 10 to 15
centimetres. The weather over Romania was poor: it was raining and the
cloud base at some places was at 400 metres. The aircraft therefore had to
fly as low as 200 to 250 metres. The crew of the observation aircraft
decided to take the risk of low-level flight because of the desire on the part
of both countries to have a successful demonstration. Photographs taken at
this altitude had a ground resolution of 2 to 3 centimetres. Navigation
proved to be a much more serious problem than either side had expected.
The obsolete navigation equipment of the aircraft, the absence of detailed
maps of the sites to be overflown and the lack of an electronic pointing
device for the camera were responsible for the problems. The observation
aircraft had to fly over some sites twice or even three or four times to take
155
a good photograph. These repeated passes would have been impossible
during a “real” observation flight, as the Agreement prohibits loitering over
the same site.
After completion of the observation flight, the aircraft landed at
Otopeni Airport in Bucharest, the then civilian international airport of the
Romanian capital. A joint team of Hungarian and Romanian technicians
developed the films (for reasons of easy development, only black and white
film was used). They were able to produce a few prints as well. The quality
of some of the pictures was not satisfactory because of the adverse weather
conditions and a slight malfunctioning of the focusing mechanism of the
camera. Observers were briefed on the results of the demonstration flight
in Budapest.
The most important conclusion to be drawn from the more than 30
observation flights that have been carried out between the entry into force
of the Agreement and the end of 1998 is that two countries with modest
technical and financial resources can create and operate a relatively effective
Open Skies regime. Both the Hungarian and the Romanian air forces have
to perform regular training flights, and some of these flights are used for
Open Skies purposes. In this way the costs of observation flights are kept to
a minimum. Hungary spends about EUR 5,000 per flight for aircraft
operation costs (mostly fuel) and the per diem of two camera technicians.
The film was surplus material from France (close to expiration date) and
came for free.
Simple and not very expensive sensors—such as aerial cameras—can
be used very effectively under various meteorological conditions, provided
that the terms of the agreement are flexible enough (the so-called success
rate, that is, the ratio of good quality imagery has been about 80-85% in the
last six years). The mutual consent of setting no resolution limit makes the
agreement technically much simpler (and operationally less cumbersome)
than the multilateral Treaty. The agreement works well without a
complicated certification procedure (H
min
-determination) and leaves more
flexibility in film developing. At the same time, participation in a multilateral
Open Skies regime offers several (mainly political) advantages compared to
a bilateral regime.
The observation flights carried out so far have fully demonstrated the
excellent confidence building potential of Open Skies. The preparation and
156
carrying out of an observation flight requires a high degree of cooperation
and real teamwork on the part of the personnel involved. This in itself is a
confidence-building exercise. Short-notice overflights are tangible proof,
not only for the militaries of Hungary and Romania but also for the average
person, of the considerable degree of confidence between the two
countries. The confidence strengthened by these flights—no doubt—
contributed to the solution of bilateral problems and the more recent
dramatic improvement of bilateral relations.
The signing of the bilateral Agreement and the carrying out of the
demonstration flight facilitated efforts to bring about the continuation of the
multilateral Open Skies discussions. Technical experience gained during
the demonstration flight was used extensively by the Hungarian delegation
in preparation for the third round of the Open Skies conference. The
Hungarian-Romanian bilateral regime, functioning in parallel with the
multilateral Open Skies regime, could serve as an example for a sub-
regional or regional Open Skies structure—for example, within the
framework of a regional confidence- and security-building regime to be
negotiated under Article V Annex 1-B of the Dayton Agreement.
Last but not least, this cheap and effective bilateral Open Skies regime
can serve as an example for countries on other continents. Regions of
tension—such as the Middle East and the Korean peninsula—also could
take advantage of this new and highly effective confidence-building tool,
which in case of an arms limitation accord can also be used for verifying its
observance.
8.2 REGIONAL APPLICATION: THE BOSNIA EXPERIMENT
AND
THE BOSNIA EXPERIENCE
From proposal to reality
The possibility of a regional Open Skies regime—parallel with the
multilateral one or embedded within it—had been discussed informally
from the very beginning of the Open Skies conference. However, this
option was not taken up “officially” during the Ottawa and the Budapest
discussion rounds due to an agenda overburdened with more important
political and technical issues. The proposal to include some language on
regional application into the Treaty was first made by the French and
157
Hungarian delegations during the Vienna round of the conference. The
proposal was not welcomed by all delegations with enthusiasm. Some
expressed fear that the mere idea of a regional Open Skies regime
represented a dangerous deviation from the main thrust of the negotiations:
the creation of an East-West confidence- and security-building tool which
would further increase transparency between NATO members and (former)
member countries of the Warsaw Treaty. Some other delegations
questioned the feasibility of Open Skies flights in a crisis area. They argued
that a slow, unarmed observation aircraft could easily become victim to a
provocation by any of the conflicting parties.
After long negotiations in the working group that dealt with legal issues
(chaired by Ambassador M. de Brichambaud of France) a compromise was
found: delegations agreed to include a sentence into the Preamble of the
Treaty which took notice of the possibility of using Open Skies observation
flights for crisis management purposes.
This sentence found practical application in Bosnia and Herzegovina
well before the Treaty entered into force. Paragraph II of Annex 1-B of the
Dayton Agreement of November 1995 foresaw the conclusion of a
Confidence- and Security-Building Agreement by the two entities (the
Bosnian-Croat Federation and the Republika Srpska). This document (the
Agreement on Confidence- and Security-Building in Bosnia and
Herzegovina) was concluded on 26 January 1996 in Vienna. The record-
breaking speed and efficiency of the negotiations, chaired by Ambassador
István Gyarmati of Hungary, surprised even the most optimistic observers.
Implementation started a few days later and from May on was supervised
by one of the authors of this chapter—in his capacity of Personal
Representative of the Chairman in Office of the OSCE. The Agreement—
modeled on the pan-European Vienna Document of 1994—was
implemented much more successfully than other parts of the Dayton
Accord. The willingness of the parties to implement a rather intrusive CSBM
regime could be explained by war fatigue and a relative equilibrium of
forces, which did not make the option of renewed hostilities too tempting
to any, but a few of the military leaders.
As implementation proceeded with increasing efficiency and the
parties acquired considerable experience in arms control implementation,
the possibility and desirability of new, additional measures was mentioned
more and more frequently in informal contacts. Since the Agreement on
158
Confidence- and Security-Building Measures in Bosnia and Herzegovina
contained a full set of “classic” measures, a regional Open Skies regime, as
the logical continuation of the process of military stabilization, was
proposed by the Personal Representative. By the end of 1996 the parties
had signaled their readiness to examine the feasibility of a regional Open
Skies regime in and around Bosnia and Herzegovina.
5
On 12 and 13 February 1997 a seminar on Regional Confidence-
Building and Open Skies was organized in Sarajevo by the Personal
Representative. Keynote speakers from Bulgaria, Canada, France,
Germany, Greece, Hungary, the Russian Federation, Turkey, the United
Kingdom and the United States made presentations on both subjects. The
speakers briefed the audience on the Open Skies Treaty national
programmes of trial flights and also on possible other (non-military)
applications of unarmed observation flights. All the local participants—
civilian and military experts—expressed interest in Open Skies and
requested the Permanent Representative to continue with the programme.
All the foreign and local participants agreed that an Open Skies regime
for Bosnia and Herzegovina, which could also be extended to the
neighbouring countries or the whole sub-region of the former Yugoslavia,
would have several advantages:
• An aerial observation regime would be the cheapest and most efficient
way to improve transparency and openness in a country (or countries)
with an extremely rugged terrain;
• Short notice observation flights would be a convincing and cost-
effective way to deter any preparation for a surprise attack of a large-
scale offensive;
• Comparative analysis of imagery produced by regular Open Skies
flights over a longer period of time, covering most of the military
objects (military bases, airfields, storage facilities, training grounds, etc.)
would enable the parties to follow closely the development of military
infrastructure of the other sides;
• Open Skies flights are more visible than any other CSBM, and therefore
present a much higher public relation value. Between countries where
deep-rooted suspicion of one another is the rule rather than the
exception, the beneficial effect of extensive media coverage of arms
control and CSBM activities can hardly be overestimated;
159
• Implementation of a regional Open Skies agreement would require the
assistance of “third states”, that is, countries with an operational Open
Skies observation aircraft and enough expertise to act as lead nations.
The involvement of other European and North American states would
create a certain “trip-wire effect”: discovery of any unusual or
threatening military activity would be reported also by a neutral, third
party. This would increase the likelihood of a timely and appropriate
reaction by the international community;
• Finally, a regional Open Skies agreement could only be the result of
free and voluntary negotiations between the interested parties. These
negotiations would represent a significant step forward compared to
the Dayton Agreement, which was negotiated and concluded under
considerable external pressure.
The Hungarian-Romanian Demonstration Flight
The next step towards introducing Open Skies into Bosnia and
Herzegovina and the Balkans was a Hungarian-Romanian demonstration
flight between 17 and 19 June 1997. In spring 1997 Romania welcomed
the Hungarian initiative of a joint flight. This action undertaken a few weeks
before the Madrid NATO Summit was to demonstrate the determination
and readiness of the two countries to cooperate in spreading stability in the
Balkans. The two countries were well prepared for such a demonstration
flight: Hungary and Romania had performed dozens of observation flights
over each other’s territory within the framework of their bilateral Open
Skies regime. Their accumulated experience was truly exceptional at the
time of the demonstration.
A number of observers were invited from the member states of the
Contact Group and countries of the region. Participation of observers from
other Balkan states (including the Federal Republic of Yugoslavia, Croatia
and Slovenia) was to familiarize them with the idea of a regional Open Skies
regime.
Preparation of the demonstration flight was a politically complicated
exercise. Permission to use the airspace of Bosnia and Herzegovina had to
be given by four actors: the Stabilisation Force in Bosnia and Herzegovina
(SFOR), the state of Bosnia and Herzegovina, the (Bosnian-Croat)
Federation and the Republika Srpska. It was a complicated diplomatic
exercise to convince state authorities (particularly the Foreign Ministry of
Bosnia and Herzegovina) to previously consult Republika Srpska authorities
160
on the matter. According to the Dayton Agreement, the three-member
Presidency of the state of Bosnia and Herzegovina was to decide on matters
pertaining to sovereignty. Since the Presidency had not been convened by
that time, it was the Foreign Ministry that acted on behalf of the state. Any
action by the Foreign Ministry not cleared beforehand by the entities of
Bosnia and Herzegovina entailed the danger of obstruction by one or both
of the entities. An independent permission by the two entities, controlling
the two armed forces, the army of the Republika Srpska and the army of the
Federation, was also necessary to ensure safety of the aircraft. Finally,
permission by SFOR was required, since Bosnian airspace was effectively
controlled by NATO at that time (including air traffic control).
As a result of intensive preparations (mostly by the Office for Regional
Stabilisation, a section of the OSCE Mission in Sarajevo, tasked with
assisting the implementation of CSBM and arms control agreements in
Bosnia and Herzegovina), both entities and all three constituent nations of
Bosnia and Herzegovina cooperated with Hungary, Romania and the OSCE
in carrying out the demonstration flight. They had volunteered to play an
active role in the preparations and in carrying out the demonstration flight.
Hungarian and Romanian experts and the observers stayed in three
different locations (hosted by the Army of the Republika Srpska, the HVO
(Croatian component of the Army of the Federation) and the Army of
Bosnia and Herzegovina (the Bosnian component of the Army of the
Federation). Transportation, technical equipment and other services were
provided also by all three parties. The cooperative nature of the exercise
represented a new stage in the process of confidence building.
The Hungarian AN-26 with a joint Hungarian-Romanian crew landed
in Sarajevo on 17 June 1997. The aircraft was equipped with a French-
made Omera-33 double camera. This camera produces two identical sets
of films, which makes data sharing an easy exercise. The chief of the
Hungarian Open Skies Team briefed the representatives of the parties
(some 30 civilian and military experts) and the observers on the multilateral
Open Skies Treaty, the Hungarian-Romanian bilateral Agreement and
technical characteristics of the aircraft and sensors. Briefings on the flight
plan and safety regulations were made by other members of the team.
On 17 and 18 June the AN-26 carried out two flights with identical
flight plans. Since the small, two engine turboprop transport aircraft
modified for the purposes of Open Skies could carry only 25 observers, two
161
flights were required to enable all the experts and observers to participate
in the demonstration. The flights lasted some three hours and the distance
covered was about 1,000 kilometres. The flight plan included nine military
objects—six on the territory of the Federation and three on the territory of
the Republika Srpska.
There was considerable media coverage of the event. Central
Television (Sarajevo) as well as TV stations of the Republika Srpska
broadcast a short report about the demonstration flight—in addition to
coverage by local radio stations and newspapers.
The black and white aerial photographs were developed within 24
hours by Hungarian and Romanian technicians in Sarajevo. Negatives of the
two films remained with the local authorities. The imagery was displayed
and analyzed by a joint team of Hungarian, Romanian and local experts.
Quality and resolution of the imagery was rather poor due to changing
weather and a very high flight altitude (5,000-6,000 metres) required by
SFOR air traffic control.
6
SFOR in Sarajevo and air traffic control in Vicenza,
Italy were very helpful throughout the flight. The restrictions on flight
altitude were dictated by safety considerations (a few days earlier an anti-
aircraft battery of one of the parties had locked onto a NATO military
aircraft, so stricter regulations were introduced on minimum altitude).
Parties assessed the demonstration flight very positively. The
representative of the Republika Srpska stated that an Open Skies regime
was needed for the region. The representative of the HVO added that the
demonstration flight convinced them of the usefulness of such a regime.
The Bosnian representative also supported the idea of a regional regime and
promised further cooperation. He added that potential civilian uses of
Open Skies flights (e.g., for environmental monitoring or technical
assistance to reconstruction efforts) were of great interest.
The German Demonstration Flight
The next demonstration flight took place on 27 August 1997 with the
German Open Skies aircraft (this aircraft was lost a few weeks later in a
tragic accident near the Western shores of Africa).
Preparation of this second demonstration flight was hardly easier than
that of the first—both politically and technically. The original idea foresaw
162
a flight path covering both Bosnia and Herzegovina and the western part of
Croatia. This gradual broadening of participation was proposed in order to
explore the possibility of a regional Open Skies regime. A following joint
flight by the United States and the Russian Federation in turn would have
covered the Federal Republic of Yugoslavia and Bosnia and Herzegovina.
Croatia in the end did not allow the German aircraft to produce imagery
over its territory. Despite this refusal the German Open Skies aircraft took
off and landed in Split, Croatia, because the setup of the Sarajevo airport
was unsuitable.
The German aircraft flew at an altitude of around 6,000 metres with a
speed of around 660 kilometres/hour. Its 2,300 kilometres-long flight path
covered 122 military and civilian objects (proposed by the parties) on the
territory of both entities.
Representatives of the three parties and observers from the Contact
Group, Office of the High Representative, OSCE and journalists from both
entities were present on board. Observers were able to move freely on
board during the flight and observe the navigators and sensor operators.
The presence of TV crews on board helped in making this demonstration
flight a media event. A relatively good coverage by the Central TV Station
in Sarajevo and TV Banja Luka in the Republika Srpska helped significantly
to spread knowledge among the population on the military stabilization
process in Bosnia and Herzegovina. Earlier attempts to attract media
coverage for “regular” CSBMs had been less successful.
Photographs of the targets were taken by three framing and three
video cameras (a vertical framing camera LMK-2015 with 152 millimetres
lens, two oblique cameras placed at an angle of 33 degrees, one vertical
and two oblique video cameras VOS-60, also at an angle of 33 degrees).
The films were developed at the German laboratory in Cologne, monitored
by representatives of all three parties. Despite unfavourable weather
conditions during the last hour of the flight (growing cloud cover), the
German crew managed to photograph some 70% of the targets successfully.
Germany had planned another observation flight over Bosnia and
Herzegovina in October 1997. Unfortunately, due to the crash of the
German Open Skies aircraft that observation flight could not take place.
163
The American-Russian Demonstration Flight
Since the carrying out of demonstration flights also carried a political
message, it was important to involve as many of the Contact Group
countries as possible. The willingness of the United States (providing
military assistance to the Federation within the framework of the Train and
Equip programme) and the Russian Federation (a strong supporter of the
former Republic of Yugoslavia and the Republika Sprska throughout the
war) to undertake a joint flight had a special significance. These two powers
working together to facilitate the military stabilization of Bosnia and
Herzegovina had an unmistakable effect on the parties.
The demonstration flight took place between 3 and 7 November
1997—somewhat later than originally planned. Preparations were done
mainly by the Office for Regional Stabilisation—in a very short time, since
information on the planned flight had been conveyed to the Office only on
2 October.
The flight plan was put together in the German Verification Centre—
using the same object list that had been provided for the previous (German)
flight. Two flight profiles were developed, each covering 28 different—
military and civilian—objects. The two demonstration flights were carried
out with a Russian Air Force AN-30B Open Skies aircraft with eight Russian
and four American experts on board as mission team. The organizers
provided ten guest seats during each flight for the local parties and the guest
observers. This time the local media showed little interest in the event.
One set of film was developed in Hungary while the other set was
taken to Moscow for processing and copying. Due to very unfavourable
weather conditions the quality of the imagery was generally very poor.
By the conclusion of the third demonstration flight over 122 separate
military and civilian objects had been photographed and copies of the
entire imagery obtained were exchanged among the three entities, as well
as with the governments of the lead nations. More significantly, the flights
themselves over the year had transformed from being an extraordinary
event to merely a “routine” exercise, which in the context of Bosnia and
Herzegovina is a major success.
There is no doubt that the three demonstration exercises conducted in
1997 proved valuable in the confidence-building arena but it became
164
obvious that the process needed to be developed further, certainly by
engaging the parties of Bosnia and Herzegovina in a more professional way
and possibly by widening the group of participating states. Clearly, further
Open Skies activities would be dependent upon the cooperation, goodwill
and financial commitment of assisting states since the local parties have
neither the money, the equipment, nor the experience to maintain the
initiative unaided.
Further Flights
During the following years (1998-2001) four more flights were carried
out.
7
As a result of these missions military personnel from the Bosnian
Federation and from Republika Srpska were trained for aerial observation.
Conclusion of an Aerial Inspection Regime
Subsequent to the initial demonstration flights of 1997 the conclusion
of a regional Open Skies regime in Bosnia and Herzegovina was discussed
as a confidence- and security-building measure. One problem resulted
from the fact that none of the entities operates fixed wing aircraft suitable
for observation flights. Hence the continued use of existing foreign Open
Skies aircraft was considered. In the end a more modest solution was
favoured: the use of helicopters owned by the entities.
In April 2000 the Joint Consultative Commission of the three parties in
Bosnia and Herzegovina agreed to establish a regime of aerial inspections.
This project is taking place within the context of the Agreement on
Confidence- and Security-Building Measures in Bosnia and Herzegovina
(Annex 1-B of the Dayton Peace Accord), under risk reduction. Inspections
will be carried out by helicopters equipped with video cameras. Each
observed entity will provide a helicopter for flights within its boundaries.
Helicopters are not allowed to cross the borders of the entities. The
agreement was finally signed after Denmark provided technical assistance
(video cameras). By August 2000 two aerial inspections had been
performed. Observers on board included representatives of the three
entities as well as from OSCE and NATO.
Assessment and Future Prospects
Technically the conclusion of the aerial inspection regime can be seen
as a minimum solution. The quality and area coverage of video images
taken from helicopters is likely to be much lower than that of imagery taken
from fixed-wing aircraft fitted with cameras. It remains to be seen whether
165
the three entities will implement the accord appropriately. On the other
hand, the application of Bosnia and Herzegovina for membership in the
multilateral Open Skies Treaty is—ultimately—a positive result of the
extended foreign investment in demonstration flights.
8.3 REGIONAL OPEN SKIES AGREEMENT IN THE BALKANS:
A MISSED OPPORTUNITY
Article V of Annex 1-B of the Dayton Accord of 1995 foresaw regional
negotiations on confidence-building and stabilization measures in the
Balkans. The negotiations comprised all successor states of the Former
Republic of Yugoslavia (Bosnia and Herzegovina, Croatia, the Republic of
Macedonia, the Federal Republic of Yugoslavia, Slovenia) as well as
Albania, Austria, Greece, Hungary, Italy, The Netherlands, Romania, Spain,
Turkey and the members of the Contact Group (France, Germany, Russia,
the United Kingdom and the US). The negotiations lasted from 8 March
1999 to 18 July 2001.
Several attempts were made to include Open Skies elements in the
draft agreement. The last proposal for “aerial observation over the territory
of other participating States” was tabled by the French chairmanship in
spring of 2001 (Ambassador Jacolin). These flights were meant to be
conducted in addition to any rights or duties under the multilateral Open
Skies Treaty. Technical details of the flights including sensors and the
resolution as well as sharing of data would have to be agreed between the
observed and the observing states in an observation mission agreement.
Flights would be conducted by multinational observation teams from Article
V participating states. The draft also suggested that all participating states
which were not signatories of the multilateral Open Skies Treaty would
apply for accession after its entry into the force. In the end—upon insistence
by the US—any reference to aerial observation was dropped. The
participating states just reaffirmed the significance of the Open Skies Treaty.
Apparently the United States rather preferred to see the new states joining
the multilateral Open Skies Treaty.
8
This materialized to some extent, since
by February 2003 three of the seven potential applicants (Bosnia and
Herzegovina, Croatia and Slovenia) had applied for accession to the
multilateral Open Skies Treaty.
166
Thus, the idea of a regional Open Skies Regime in the Balkans under
Article V turned to naught. The concept could have been attractive as a
means of strengthening regional cooperation, confidence building and
identity building. The merit of the French proposal had been its flexibility in
fixing the details of flights; the weakness was the lack of binding
commitments. In the end the pull towards the established European
security systems and treaties like NATO and the adapted CFE Treaty
respectively, as well as the wish to join the European Union were stronger
than the interest in regional approaches. It remains to be seen to what
extent participation in the multilateral Open Skies Treaty can make up for
the missed opportunity. The Concluding Document of the Article V
negotiations allows for some confidence-building and security cooperation
measures on a voluntary basis.
9
It is thus a rather weak agreement of much
less weight than the Vienna Documents, the adapted CFE Treaty or even
the agreements reached on arms control and risk reduction in Bosnia and
Herzegovina itself under Article II of Annex 1-B of the Dayton Accord.
8.4 OPEN SKIES OUTSIDE OF EUROPE:
PRECEDENTS AND PROSPECTS
Understandably, much of the foregoing has been confined to
discussing Open Skies in the geographic context of Europe and the
historical political and military schisms that once—and in several cases,
still—divided it. Nor can one ignore the comparatively long experience
European nations have acquired from the negotiation and implementation
of numerous CSBM agreements dating back to the mid-1970s. When
coupled with the relatively large amount of resources European nations
have been able to devote to arms control, it is not unfair to ask whether
CSBMs such as Open Skies have any potential relevance outside the
isolated case of Europe. The answer, in a word, is most assuredly.
In order to understand this relevance, it is necessary to review the key
concepts underlying the Treaty on Open Skies. Only in this way it is possible
to elaborate on how something like an Open Skies regime could make a
significant contribution to enhancing regional security outside Europe. It
will also help later explain why proposals to create aerial observation
regimes in regions such as Latin America, South East Asia, and elsewhere
have found receptive audiences.
167
Before doing so, a caution is in order. It cannot be overly emphasized
that a regional aerial observation regime need not replicate either the
expense or the level of complexity of the European Open Skies Treaty.
Unique national and regional considerations such as geography and history
will dictate similarly unique bilateral and multilateral approaches to the
political and technological considerations associated with an aerial
observation regime. That is to say, an aerial observation regime negotiated
among neighbouring states in other regions of the world will undoubtedly
utilize different treaty provisions, as well as different categories of aircraft
and sensors that are better suited to those states’ unique operational
environments. It is a measure of the Open Skies concept's inherent
flexibility that it can be so easily adapted to meet these other requirements.
Open Skies: One Piece in a Broader Mosaic
The preceding sections have addressed many of the very technical
issues associated with implementation of the Open Skies Treaty. However,
the question to which we must now turn is why other non-European regions
of the world find some adaptation of the Open Skies Treaty to be
appropriate to their security concerns. In order to answer this question, it is
necessary to first place Open Skies in the broader context of “arms control”,
and thereby to explain what Open Skies does. In this way, it will be easier
to understand how Open Skies in one form or another could make a
significant contribution to enhancing national and regional security.
Arms control represents an agreement between two or more nations to
enter into a dialogue concerning the types of military activities each
perceives as potentially threatening. It reflects an interim step in the
transition from confrontation to cooperation by codifying the steps each will
take to allay the fears of the other. Arms control creates, if not trust, at least
a legal basis for acquiring some degree of confidence between nations. By
increasing transparency, arms control measures contribute to building
national confidence and reducing incentives to attack, thereby allowing
countries to direct scarce resources to safer, more productive activities.
Arms control agreements often share two common elements. The first
is an exchange of information on the weapons, forces, or military activities
that are viewed as potentially threatening. This information serves to define
the long-term military balance. The second element is the opportunity for
nations to, by some means, reassure themselves, that is, to verify, that the
information they were provided is accurate.
168
Each arms control agreement is unique in the opportunities that it
provides to confirm or verify information that has been exchanged by
touching, seeing, photographing, measuring, remotely monitoring,
overflying, or otherwise recording information. These variations can be
attributed to several considerations, the most important of which is the
military significance of what is being controlled. For example, Open Skies’
“loose” criteria of military significance are far less demanding than the
START Treaty provisions governing nuclear warheads. These variations are
also designed to prohibit access to sensitive areas and to protect legitimate
national security concerns, thus ensuring that information unrelated to a
particular arms control agreement cannot be gathered.
By defining the size, composition, and deployment of neighbouring
militaries, arms control agreements allow national security planners to tailor
the size, composition and disposition of their own military structure in order
to field only such forces as are necessary for self-defence. It also allows them
to reassure themselves that ambiguous military activities, such as
redeployments, exercises and the like, are not misinterpreted as a prelude
to surprise attack. By reducing the fear of aggression, arms control thus leads
to greater stability.
Why An Aerial Observation Regime?
At the political level, beginning with the initiation of the Helsinki
process in the mid-1970s, US officials have met with representatives from
throughout the international community concerning the potential
application of military confidence- and security-building measures to
regions of the world still characterized by tension and distrust. These CSBMs
span a wide range of activities, from the simple exchange of information, to
measures requiring extremely intrusive on-site inspection. Why, then, is
there specific international interest in the Open Skies Treaty?
Open Skies lies along the middle of this continuum of measures in that
it permits nations to confirm information they have acquired, but in a way
that does not require intrusive physical access. Because an aerial
observation flight does not arouse the same degree of political sensitivity as
having foreign military personnel conduct an on-site inspection, it is
generally thought to provide a more “neutral” means through which to
pursue monitoring as well as closer military-to-military contacts.
169
Closely related to the above point is the very nature of the contact
between the militaries taking part in this exercise. For example, ground-
based arms control inspections involve two clearly defined teams, the
inspected state party (us) and the inspecting team (them); they most often
take place under Treaty provisions permitting challenge inspections; and
result in a report jointly signed attesting to whether or not a nation is in
compliance with its treaty commitments. Not surprisingly, an on-site
inspection can assume an adversarial character.
In contrast, participants in an aerial observation mission, in addition to
generally being of the same military profession (i.e., pilots, navigators,
sensor operators), share a common interest in ensuring the safe operation
of the aircraft and that the flight conforms with regulations governing
international air traffic. This “commonality of purpose” tends to foster a
much more cooperative relationship throughout the mission between the
observed and observing states parties.
It is probably one of the unintended consequences of such individual
interactions that adversaries are not only forced to meet face-to-face, but to
actually cooperate. These opportunities allow each to learn something
about the other, to explore the perceptions and motivations of their
adversaries, and to learn what they share in common and what they do not.
At the national level, the term “military-to-military contacts” is the official
phrase used to encourage this behaviour.
Beyond the personal dynamics of an Open Skies mission another
fundamental difference distinguishes this from other treaty regimes. Key
among the central concepts underlying the Open Skies Treaty is the right of
states parties to jointly collect information over the entire territory of other
states parties without any restrictions for reasons of national security. In
other words, the judgement of compliance or non-compliance is not at
issue. Additionally, the Treaty allows these observation aircraft to be
equipped with a variety of sensors, such as panoramic cameras, infrared
line scanning devices and SAR. In concert, these sensors allow imagery to
be gathered during day and night, under any weather conditions.
By allowing one nation to overfly and image areas where it suspects
suspicious military activity to be taking place, Open Skies affords
reassurance against the fear of pre-emptive attack by a neighbour. Of equal
170
importance, aerial observation flights can be used to demonstrate the
absence of any activity that could be perceived as threatening.
The Open Skies Treaty is not a diplomatic European euphemism for
spying. As fundamental to the success of the Open Skies Treaty as territorial
access is, it is also the one provision most likely to raise the discomfort of
foreign audiences. This is particularly true in areas outside Europe where
there is not a comparatively long history of amicable military-to-military
contacts and where widespread reliance on national technical means is still
the rule. It is not difficult to understand why. Outside of Europe, having
one's national defence establishment routinely overflown and imaged by
potential adversaries is a novel experience for most nations.
On the positive side, the right to conduct airborne observation
missions over other countries provides most of these nations with their first
independent opportunity to confirm the location and activities of their
neighbour’s military forces. Furthermore, unlike more sensitive means of
gathering information, it importantly provides a source of information that
can be shared with foreign governments and the public owing to the
multinational composition of the observation team.
Central to then President Bush's proposal was that Open Skies sensors
only be capable of distinguishing major items of military equipment. For
example, negotiated limitations on sensor resolutions allow for the ability to
distinguish a tank from a truck, but not the ability to distinguish one type of
tank from another or its state of operational readiness.
Accordingly, a number of safeguards are built into the Treaty that
significantly minimize the possibility that Open Skies overflight missions can
be used to collect information in violation of the Treaty. For example, the
Treaty requires an intensive examination of the aircraft and sensors as part
of its certification as an Open Skies observation aircraft. Furthermore, the
observed nation has the right to conduct an extensive pre-flight inspection
of the aircraft upon its arrival in the country to satisfy itself that no
modifications have been made to the aircraft or sensors since its
certification. Finally, representatives of the observed nation fly onboard the
aircraft during the observation mission to ensure that the agreed flight route
is not deviated from and that the sensors are used only in compliance with
what has been agreed to by the observed and observing countries.
Protections like these are fundamental in allaying suspicions in regions
171
outside Europe that aerial observation flights are merely legalized aerial
reconnaissance missions, particularly in those regions where the memories
of conflict remain fresh.
For nations to benefit equally from the Open Skies Treaty, they
obviously must be capable of equally implementing the Treaty. This is one
reason why the Open Skies Treaty requires that all aircraft and sensors must
be commercially available. This requirement ensures also that no nation can
take technological or financial advantage to the detriment of its Treaty
partners. Further assurances are provided by Treaty provisions allowing
flights to be conducted with aircraft belonging to the observing nation, the
observed nation, or leased from another signatory nation.
From the perspective of the European arms control community, it is
sometimes easy to forget that such attention to ensuring the rights and
responsibilities of all states stems from a long-term, widely shared objective,
in this case the preservation of peace in Europe. In other regions where this
“collective sensitivity” is not so well developed, bilateral and multilateral
military-to-military relations on the basis of equality provides a novel
alternative to costly arms races and armed conflicts, particularly for those
nations too long accustomed to poverty, illiteracy and instability.
This is not to suggest that only benign intentions animate Open Skies
Treaty signatories. Otherwise, the level of complexity and expense of the
Open Skies Treaty would not have been necessary. It is, however, meant to
suggest that the nations of North America and Europe have come to accept
arms control as being in their individual and collective self-interest. This is
seen, too, in the common vocabulary of arms control words they have
developed over 30-plus years to describe that common experience. With
the increasing “internationalization” of arms control represented by
agreements such as the NPT, CTBT, MTCR and CWC,
10
it may be that this
lesson of the Cold War is being more widely accepted, although recent
events in the Balkans, Southeast Asia, and elsewhere provide scant basis for
optimism in the near-term.
Although perhaps obvious, one more observation is in order. Outside
of Europe, few nations possess active nuclear, biological or chemical
weapons programmes. All, however, possess conventional military forces
and equipment—albeit in differing quantities and qualities—and share the
same suspicions that have divided nations since time immemorial.
172
From a non-European perspective, there are a number of advantages
inherent in an aerial observation regime that are not to be found in other
potential arms control measures. This is particularly true in those regions
where there is not a long, well-established history of trust and cooperation,
and where military-to-military contacts remain largely in the tentative
stages. Three advantages are of particular note.
Traditionally, arms control treaties have restricted or limited military
forces and activities. For example, the CFE Treaty establishes precise
limitations on five categories of military equipment in each of the
geographic zones defined by that treaty. Moreover, under the verification
provisions of the CFE Treaty, any facility with a door two meters or greater
in width is subject to intrusive on-site inspection. The Open Skies Treaty
imposes no such constraints. Signatories are free to modernize their forces,
and to station and deploy them wherever they choose, subject only to the
possibility of being overflown and photographed.
Second, an aerial observation regime allows participating countries to
overfly and to photograph extremely large and in some cases rugged areas
rapidly and inexpensively. One need only briefly visualize the rain forests of
Latin America or the mountains of Central and Southeast Asia to grasp the
political and military complications posed by geography.
Third, beyond making a significant contribution to increased national
security, aircraft modified to conduct aerial observation missions may be
used for a wide variety of other purposes. These additional purposes help
to ensure the most cost-effective use of these aircraft. Among the more
traditional applications are: conducting environmental assessments;
documenting the consequences of natural disasters such as flooding and
forest fires; evaluating land usage, like tracing vegetation and deforestation;
mapping in order to establish territorial boundaries; and providing terrain
characteristics in support of infrastructure construction projects such as
roads, bridges, dams and distribution centres.
Other non-traditional uses could include recording the movements of
migratory peoples, such as those fleeing man-made and natural disasters,
and terrain exploration. Preliminary analysis of using airborne platforms to
identify and map minefields is encouraging.
11
173
These are not “rocket science” cutting edge programmes, and are in
fact being conducted daily throughout the world. In all likelihood,
requisitely equipped aircraft that could be rapidly and economically
modified to perform aerial observation missions are already in national or
commercial service, thus significantly reducing the costs associated with a
regional observation regime.
Must a Regional Aerial Observation Regime Look Like “Open Skies”?
If, in truth, nations outside Europe find these concepts and aircraft uses
so compelling, then why have no other regional observation regimes been
concluded? The answer, in large part, is that until recently the world of arms
control and CSBMs was defined by NATO and the Warsaw Treaty. With the
dissolution of the Warsaw Treaty and the emergence of a still ambiguous
new international security environment, Open Skies represents for many
their first introduction to the concept that arms control and CSBMs can
actually contribute to maintaining and enhancing their national security.
Regrettably, while Open Skies may be brilliant in the simplicity of the
concepts that underpin it, most—including many of its signatories—regard
it as technologically intimidating for all but the richest nations.
In this, there is a presumption that the very success of the Open Skies
Treaty is inseparably linked to the aircraft and sensors that are in use by its
US and European parties. This is not only unfortunate, but also completely
wrong. It is a mistake of the first order to assume that a regional aerial
observation regime must replicate the cost and complexity of the European
Open Skies Treaty. Each region of the world, whether in Asia, Latin
America, the Middle East, or elsewhere, presents distinctive political and
operational requirements.
For example, the sensors permitted under the Open Skies Treaty are
more than adequate to detect significant military concentrations or activities
that would be of concern to the nations of Central Europe. They are
completely inadequate to allow for the detection and identification of
mortars, a more immediate concern to the components of the former
Yugoslavia. Similarly, Open Skies sensors are of little utility in penetrating
the tropical canopy that covers much of South America or, more
specifically, in areas along the contested border between Ecuador and Peru.
It thus becomes readily apparent that nations must first agree amongst
themselves what types of forces or activities constitute a potential threat to
174
regional security and stability. For some, that may be heavily armed forces
stationed along border areas; for others, it may be more lightly armed forces
operating in remote or mountainous regions. In either case, clarity of
purpose regarding the objects to be imaged and in what detail is essential
to selecting the appropriate sensor(s). Open Skies can serve as a precedent
for this decision only to the extent the signatories are willing to share the
extensive operational experience they have acquired preparing for its
implementation.
For example, in hindsight one of the lessons the United States learned
is that the selection of an observation aircraft is closely related to the
question of sensor operating parameters. The selection of the KA-91 optical
camera forced the US to select an aircraft capable of flying up to 10,668
metres. Unfortunately, the selection of an OC-135 prohibited sensor
packages that operate at very low altitudes, which, of course, is the only
region where optical cameras have much utility given the prevailing cloud
cover across most of Europe throughout the year.
By the same token, many point to the size and sophistication of the
aircraft currently configured to perform Open Skies missions and
erroneously assume that the cost implications of an aerial observation
regime are prohibitive. Frequently lost in such considerations is that the
selection of an airframe must be based on operational requirements, and
that a wide variety of air platforms are readily available to perform such
missions. The United States is a case in point.
Under the terms of the Open Skies Treaty, the United States has the
right to conduct an un-refuelled 6,500 kilometres overflight of the Russian
Federation. At the operational level, exercise of this right means the US
aircraft must transit the Atlantic Ocean and refuel in Western Europe before
transiting to the Russian designated point-of-entry. Both requirements
dictated that the US select an airframe capable of flying extremely long
distances, in this case an OC-135, a predecessor of the civilian Boeing 707.
Other Open Skies countries such as the United Kingdom and Ukraine must
meet far less demanding requirements.
For these countries, propeller aircraft such as the Russian AN-30 or
American C-130 are more widely available, operationally simpler to use
(e.g., they can fly low and slow) and can far more easily operate from
austere locations. Indeed, austerity is another major determinant of what
175
aircraft and sensors are selected. For example, austere locations argue
against selection of extremely sophisticated sensor packages because of
size, weight and support requirements. Furthermore, austere airfields
infrequently are of the requisite length, width and strength to support large
jet aircraft; nor are sophisticated electronic maintenance facilities generally
available.
To take our example one step further, one of the options—which was
eventually retained—in support of an aerial observation regime in Bosnia
and Herzegovina would rely on a hand-held digital camera operated from
a helicopter. Excluding the operating costs of fuel, etc., digital cameras are
available in the US for about $600. Processing of the image data can be
accomplished on laptop computers, which are probably more widely
available than typewriters.
Returning to the precedent established by the Open Skies Treaty, at a
minimum an observation aircraft should be capable of carrying up to three
individuals in addition to the normal crew compliment. One representative
each from the observing and observed nations to ensure that questions do
not later arise concerning whether the imagery has been tampered with,
and that the observing nation has not gathered information in violation of
their agreement. In the case where the observing nation provides its own
observation aircraft, experience preparing to implement the Open Skies
Treaty has demonstrated the desirability of including a second individual
from the observed nation in order to facilitate coordination with the local
air traffic control authorities. At the operational level, observation aircraft
need only be capable of safely transiting the area of interest on the way to
the next refuelling field.
Regional Aerial Observation Regimes
Clearly, most nations already possess the technical capability to
implement some form of an aerial observation regime once the political
decision is made to do so. Have any states given indications of doing so?
The answer to date is mixed, but encouraging.
Latin America is a case in point. Since the late 1980s, the US, working
through the Organization of American States (OAS), has sought to establish
the relevance of CSBMs to the inter-American security system. Advances in
the establishment of CSBMs in the region have improved both military-to-
military and civil-military relations and thereby have served to decrease
176
tensions in the region. Since the OAS regional conference on CSBMs in
Santiago, Chile, US policy has aimed at fostering regional arms restraint and
control in Latin America through institutionalizing CSBMs, and expanding
as well as deepening security dialogue and transparency throughout the
hemisphere.
For example, the majority of Latin American states comply with annual
United Nations and OAS reporting for the Registers of Conventional Arms
and Military Expenditures; Argentina and Chile are producing the region's
first Defence White Papers; Argentina and Chile hosted the first OAS-
sponsored conference on identifying and codifying regional CSBMs; the
OAS adopted the Santiago Declaration on CSBMs as a formal reporting
requirement; in April 1997 the Brazilian and Argentine Presidents signed a
Military Cooperation Agreement to prevent arms races; in May 1997 the
Rio Group of South American countries agreed to establish a working group
to study the issue of arms races and to exchange information on the
acquisition of sophisticated weapons; and the US sponsored a resolution for
advance notification of major arms acquisitions at the June 1997 OAS
General Assembly and again at the March 1998 Summit of the Americas.
In support of these efforts, US representatives have sought to educate
civilian and military officials from throughout the inter-American region on
the concepts underlying the Open Skies Treaty, which have potential
application to this region. These efforts have included:
1) Beginning in May 1998, the Treaty on Open Skies was included as a
regularly scheduled agenda topic during bilateral Department of
Defense staff talks with Ministry of Defence representatives from
throughout Latin America. In March of this same year, the United
States Open Skies aircraft was displayed at the International Air and
Space Show (FIDAE ‘98), in Santiago, Chile. Beyond its simple
participation in this unique Latin American event, the presence of the
US Open Skies aircraft was hoped to underscore the importance the
United States attaches to this potential regional CSBM measure in
anticipation of the Second Summit of the Americas, which was also
held in Santiago the following month.
2) In August 1997 the United States and the Russian Federation jointly
sponsored an Open Skies programme for representatives of the OAS
and their Washington, DC embassies, including an exhibition of the US
177
and Russian Open Skies aircraft. This followed a similar programme in
April of 1997, when the American and Ukrainian Open Skies aircraft
were placed on exhibition for representatives of the Inter-American
Defence Board and the OAS. Beyond the broad national makeup of
the invitees, this later event was notable in two respects. First, among
the key presenters of information on the potential relevance of an
aerial observation regime to Latin America were the Head of the
Ukrainian Armed Forces Verification Center and the Hungarian
Ambassador to the OSCE. Of perhaps more historic importance,
certainly to the United States, the participation of the Ukrainian
observation aircraft marked the first time ever that an aircraft—
especially one from a former Warsaw Treaty country—had been
allowed unrestricted access to American airspace and to image the
areas it chose.
3) These events followed highly successful conversations held during the
first US- Chile Defense Consultative Committee session in July of 1996,
which included a briefing on the Open Skies Treaty and a tour of the
US Open Skies aircraft for the Chief of the National Defence Staff and
other delegation members. Also in 1996, Peru and Ecuador sent
delegations to the United States for an orientation workshop on the
contribution of regional CSBMs to Latin American security and to learn
more about the Open Skies Treaty.
Perhaps no greater practical demonstration exists of the high regard in
which the US holds the Open Skies Treaty than its willingness to conduct a
programme on regional CSBMs, to include the Treaty on Open Skies, at the
invitation of appropriate government officials throughout Latin America and
the Pacific region. As part of this programme, the United States has made
clear that it is prepared to also place its Open Skies aircraft on exhibition for
representatives of the government and the public at large.
The purpose of this programme is to allow the United States to share
the extensive experience it has gained during the Open Skies Treaty's
negotiation, and from the wide range of activities the US has participated in
during preparation for its implementation. These activities include
discussions concerning the creation of aerial observation regimes in the
Middle East, the former Yugoslavia, and in South East Asia. The theme of
this American policy is that the concepts underlying the Open Skies Treaty
could make a significant contribution to enhancing regional confidence and
178
security building, and that an unarmed aerial observation regime is relevant
to any region of the world on either a bilateral or multilateral basis.
Turning to East Asia, twice, once in 1996 and again in 1997, US Open
Skies Delegations travelled onboard the US Open Skies aircraft to Tokyo,
where extensive briefings were conducted for representatives of the
Japanese government on the regional security contribution of a measure
similar to Open Skies. Also in Eastern Asia, the US Open Skies aircraft was
placed on exhibition in the southern Chinese city of Zhuhai in November
of 1998. Elsewhere in the region, something akin to Open Skies has been
proposed as one possible means of monitoring a disengagement zone or
buffer area between North and South Korea. In early 1999, an aerial
observation CSBM formed the centerpiece of the US presentation during
the annual arms control conference sponsored by the United States Air
Force. Over 23 nations from throughout Southeast Asia attended this
unique regional programme.
Southern Asia: On a regular and now routine basis, the United States
has provided briefings and documentary information concerning the Open
Skies Treaty to representatives from both India and Pakistan, particularly as
it may relate to potential joint activities in and/or over the highly disputed
Kashmir region. These efforts have been supported by the Cooperative
Monitoring Center at Sandia National Laboratories, Albuquerque, New
Mexico. The Center assists experts from all over the world to acquire the
technology-based tools they need to assess cooperative security measures.
While India and Pakistan have negotiated some information exchange and
confidence-building measures, in particular on nuclear issues, cooperative
aerial observations are still missing. It was at the Cooperative Monitoring
Center where two retired air marshals from India and Pakistan drafted the
outline of an aerial monitoring regime for the border region between India
and Pakistan.
12
The proposal foresees the implementation of elements of
the Hungarian-Romanian and the multilateral Open Skies Treaty. A
resolution limit of 30 centimetres is envisioned, as well as initial training and
assistance from a third party, a party to the Open Skies Treaty. In the same
vein, the US is planning to charter an aircraft that will be used to conduct a
demonstration flight between India and Pakistan in the summer of 2004.
Middle East: Some beneficial, although limited use of aerial
inspections has been made in the Middle East by the United Nations
peacekeeping operations in Lebanon, along the Iran-Iraq border, and by
179
peacekeepers in the Sinai implementing disengagement and peace
between Israel and Egypt. Third party aerial inspections have also been
carried out by the United States to ease tensions over sensitive border areas
in the region, but the operation of these arrangements have not been widely
acknowledged.
13
M. Krepon and P. Constable have suggested a conceptual
framework for the expanded use of aerial inspections for four related tasks:
(1) enhancing border security; (2) observing areas where levels of military
personnel and equipment have been limited by agreement; (3) monitoring
exclusion or demilitarized zones; and (4) monitoring facilities of special
interest.
Conclusions
In the words of former US Defense Secretary Perry: “Our arms control
initiatives are an essential prevention measure that can yield
disproportionately significant results, often eliminating the need for a more
substantial response later.” In other words, arms control is a form of defence
by other means. Key among those measures has been the Treaty on Open
Skies.
Open Skies is not primarily about multispectral sensors, infrared
imagers, photographic resolutions, or large sophisticated aircraft. Open
Skies is about developing trust, cooperation, and good will between and
among neighbouring nations. Or, failing that, helping to diffuse the tensions
and distrust that otherwise could lead to an arms race or armed conflict. At
its most elementary level, the Open Skies Treaty does this by establishing
the precedents that observation missions will be conducted by
representatives of both the observing and observed nations, and that no
restrictions may be placed on the flight other than those related to the safety
of the crew and aircraft.
As a means of facilitating military transparency, Open Skies is unique
in the wide range of technologies and methodologies from which nations
may draw. The concept is flexible enough to be adapted to regional
conditions and concerns. What is needed is the mutual politics to establish
a regional agreement at one or several of the hotspots outside of Europe
where highly armed forces face each other and tensions are high. Clarity of
purpose regarding the type of objects to be imaged and in what detail is
absolutely essential. So too is the need to be aware of the operational
implications of those decisions.
180
Postscript
Since the coming into office of the Republican administration in
January 2001 concern has been expressed in some quarters that the
commitment of the United States to the Open Skies Treaty and its
adaptation to meeting other regional concerns has diminished. In fact, the
recent American “silence” concerning the Open Skies Treaty is as much
testimony to the Treaty's now routine acceptance by the American arms
control community as a vital component of the US's efforts to promote ever
greater transparency and trust among its Treaty partners. Nor has the United
States commitment to its possible adaptation to meeting other regional
challenges diminished. For example, the United States continues to work
with countries such as India and Pakistan in examining how an aerial
observation regime might be tailored to meet their unique security
concerns. These efforts are perhaps best illustrated by the recent joint
publication written by two former Indian and Pakistani Air Marshals
describing how an aerial observation regime might be put in place in that
region through a series of incremental steps. Elsewhere, the US continues
to work with countries from across Latin America, such as Peru and
Argentina, in examining how an aerial observation measure might
contribute to reducing concerns unique to South America. In Bosnia and
Herzegovina, US Open Skies representatives continue to routinely take part
in aerial observation missions designed to preserve the peace in that region
through the promotion of measures that enhance cooperation and trust.
Indeed, as this is being written, the Defense Department's National
Defense University is sponsoring a programme for visiting scholars and
government representatives from across Northeast and Southeast Asia
dedicated to a discussion of cooperative measures to reduce cross-border
security concerns in those regions. Foremost among those measures is the
adaptation of the Open Skies Treaty to meet their distinct challenges, which
range from interdicting proliferation networks (roads, airfields and
waterways) to resolving jurisdictional issues through joint mapping efforts,
to responding to environmental and humanitarian challenges, in addition to
the more traditional mission of promoting greater transparency and trust.
In spite of concerns expressed above, the United States remains fully
committed to the realization of the benefits offered by the Open Skies
Treaty, whatever distinct or unique regional form that agreement might
take.
181
Notes
1
The complete text of the Agreement was issued as a United Nations
document, A/46/188-S/22638, Annex, and was published in
Disarmament, Vol. XIV, No. 4, 1991. It is reproduced in Appendix H.
2
The ground resolution of an aerial camera or a video camera depends
on its focal length and the altitude of the aircraft. If there is an upper
limit for the ground resolution, it might force the aircraft not to fly
lower than a certain minimum altitude.
3
There was extensive media coverage of the event in both countries.
4
This type of double camera has a focal length of 200 millimetres.
5
Observers from the regions were first introduced to Open Skies
practice during two joint Hungarian-US demonstration flights in
Hungary (28-31 October 1996).
6
The Hungarian-Romanian Open Skies Agreement does not contain
any limitation on flight altitude or resolution; therefore, it is possible to
fly as low as required by cloud cover and the technical characteristics
of the camera. The “ideal” imaging altitude for the Omera-33 camera
on board the Hungarian aircraft is 1,500-3,000 meters.
7
A joint Russian-German flight was conducted on 13-17 July 1998,
using a Russian AN-30 aircraft. A flight arranged by France on 21-23
September 1999 using a French Hercules C-130 H equipped with a
sensor pod. Two flights provided by the Czech Republic using its AN-
30 aircraft and co-organized by Denmark on 9-13 October 2000 and
25 May-1 June 2001. The former flight was accompanied by a
demonstration of aerial inspections via helicopter, organized by
Denmark. Film processing took place in Prague with participation of all
parties (13-18 October 2000). See Table 4.3.
8
The US was also concerned about the protection of its troops in Bosnia
and Herzegovina and Kosovo under such observation flights.
9
See Concluding Document of the Negotiations Under Article V of Annex
1-B of the General Framework Agreement for Peace in Bosnia and
Herzegovina, www.osce.org/representatives/arms/article5/article5.pdf
and Statement by Ambassador Henry Jacolin, Special Representative of
the OSCE for Article V (Regional Stability), at the Joint PC/FSC Meeting,
http://www.osce.org/representatives/arms/documents/files/
statement_19jul2001.pdf.
10
The Nuclear Non-Proliferation Treaty, the Comprehensive Test Ban
Treaty, the Missile Technology Control Regime and the Chemical
Weapons Convention.
182
11
Ben Maathuis, “Remote Sensing Based Detection of Landmine Suspect
Areas and Minefields”, dissertation, University of Hamburg, 2001,
available from ITC, P.O. Box 6, NL 7500 AA Enschede, The
Netherlands.
12
Air Marshal M. A. Chaudhry PAF (Ret.) and Air Marshal K. C. Cariappa
IAF (Ret.), “How Cooperative Aerial Monitoring Can Contribute to
Reducing Tensions Between India and Pakistan”, Report, SAND98-
0505/22, Dec 2001, http://www.cmc.sandia.gov/Links/about/about-
mainframe.htm.
13
M. Krepon and P. D. Constable, Confidence Building, Peace Making and
Aerial Inspections in the Middle East, Occasional Paper No. 6,
Washington, DC: The Henry L. Stimson Center, 1992.
183
CHAPTER 9
THE IMPROVEMENT OF SATELLITE CAPABILITIES
AND ITS IMPLICATIONS FOR THE OPEN SKIES REGIME
Hartwig Spitzer
In this chapter we will address the technological competition to Open
Skies brought by the advance of high-resolution commercial imaging
satellites and unmanned aerial vehicles (UAVs).
9.1 THE ADVENT OF COMMERCIAL 1-METRE SATELLITES
The digital revolution of information technology is conquering the
world in several ways. One of them is the unprecedented availability of
commercial satellite imagery at ground pixel resolution values of 1 meter or
better.
1,2
Such resolution performance had been achieved by US and
Soviet military reconnaissance satellites already in the early 1970s. It took
another 20 years and the collapse of the Soviet Union, before such imagery
was made accessible on the open market for research, business and “other
interests”. “Other interests” include the secret services and militaries of
countries around the world, which can acquire such imagery.
Commercial distribution of high-resolution images was authorized in
1987 through the Soviet trade association Soyuskarta, which offered
imagery with spatial resolution of approximately 5 meters. In 1992 two
Russian firms began to sell selected images with resolution values as low as
2 metres.
3
The availability and reliability of delivery was limited. It triggered,
however, a development and push towards the commercialization of such
imagery elsewhere, in particular in the US. Several industrial consortia were
formed, which could draw on know-how and manpower transferred from
military space technology.
In 1992 the US Congress passed the Land Remote Sensing Policy Act,
which was signed into law in October of that year. The law streamlined the
184
procedure for considering license applications for commercial satellites,
and eliminated many of the legal obstacles existing theretofore. In March
1994, in a further attempt to improve US commercial competitiveness vis-
à-vis Russia and France, President Bill Clinton issued a Presidential Decision
Directive (Nr. 23), which loosened the restrictions on the sale of high-
resolution imagery to foreign entities.
4
The US government, however,
maintains the option of temporarily blocking the collection and
dissemination of commercial satellite imagery. It is expected that such
collection and dissemination limitations will be exercised only in rare
instances.
5,6
The US Department of Commerce granted the first licence to operate
a high-resolution satellite to World View Inc. (now Digital Globe). A total of
14 commercial satellite systems were licensed between January 1993 and
March 1999.
7
Not every system was completed, however. The
development was and still is marked by technical delays and failed
launches. Eventually, on 24 September 1999, the first fully commercial 1-
metre satellite, IKONOS 2, was launched successfully by Space Imaging
Corporation, USA. On 18 October 2001 Digital Globe Inc., launched the
Quickbird satellite, which delivers panchromatic images even at 61
centimetres resolution. A third US consortium, Orbital Sciences Inc.,
succeeded in launching their ORBVIEW-3 1-metre satellite in June 2003.
All of these satellites have a stereo capability (for terrain recognition) and
four spectral channels (for colour displays and vegetation analysis in the
near infrared, see section 7.3.3). The Israeli owned company Image Sat
International launched a commercial satellite, EROS A, in December 2001,
which can provide images at 1.8 metres resolution (1 metre after re-
sampling). Finally, India launched an experimental defence 1-metre
satellite, TES, which will provide the technology base for a similar
commercial satellite (Cartosat-2).
Table 9.1 gives an overview of the commercial 1m satellite systems in
space.
8
Several more are to follow in the coming years, both with optical
and SAR sensors as shown in Tables 9.2 and 9.3. As a result a substantial
market for 1-metre satellite images with an estimated volume of $500
million per year was created.
9
Even at such market volume it will be difficult
for all four consortia to recover their substantial investment costs (e.g., $700
million for Space Imaging) and operation expenditures. Main customers are
industrial developers and the US National Imagery and Mapping Agency
(NIMA, now National Geospatial-Intelligence Agency, NGA). NIMA
185
awarded two five-year contracts of up to $500 million to both Space
Imaging Corporation and Digital Globe Inc.
10
A similar contract was
promised to Orbital Sciences.
11
In addition, several European and Asian
countries have developed or are developing military high-resolution
satellites, which might later on also supply images to commercial markets
(see Table 9.4). Given the success and quality of the new high-resolution
commercial satellite imagery it is thus appropriate to ask to what extent such
imagery can substitute for imagery from Open Skies flights.
Table 9.1: Performance Parameters of Commercial 1-Metre Optical Satellites
System/Corporation Mode Resolution
(nadir)
GSD (m)
Spectral
Bandwidth
(µm)
Swath
at nadir
(km)
Stereo
option
Revisit
Cycle
(days)
Repeat
Cycle
(days)
Launch
date
IKONOS-2
Space Imaging, USA
www.spaceimag-
ing.com
Pan
MS
1
4
0.45-0.90
0.45-0.52
0.52-0.60
0.63-0.69
0.76-0.90
11.3
11.3
in &
cross-
track
14
(max)
1-3
(max)
24 Sept.
1999
EROS-A
Image Sat International
Antilles/Israel
www.imagesatintl.com
Pan
Pan
(a) 1.8
(b) 1
0.5-0.9
0.5-0.9
13.5
9.5
in &
cross-
track
7
7
2-3
2-3
5 Dec.
2000
QUICKBIRD-2
Digital Globe, USA
www.digitalglobe.com
Pan
MS
0.61
2.44
0.45-0.90
0.45-0.52
0.52-0.60
0.63-0.69
0.76-0.90
16.5 in &
cross-
track
1-3 18 Oct.
2001
ORBVIEW 3
Orbital Sciences, USA
www.orbimage.com
Pan
MS
1
4
0.45-0.90
0.45-0.52
0.52-0.60
0.63-0.68
0.78-0.90
8
8
in &
cross-
track
26 June
2003
Pan = panchromatic, MS = multispectral, GSD = ground sampled distance.
Repeat cycle refers to the frequency of imaging opportunities of a particular site if sensor tilt is
exploited. Nadir refers to a direction perpendicular to the Earth’s surface.
186
Tab le 9.2 : Future Commercial High-Resolution Optical Satellites
Country Company, System Resolution at nadir (m) Swath at
nadir
(km)
Launch date Remark
panchrom. multispectral
USA Space Imaging
IKONOS 3/BLOCK 2
0.5 2 2007?
USA Digital Globe World View 0.5 2 (8 channels) 2006 Dual use
Israel ImageSatInternational
EROS B1
EROS B2
0.82
0.82
-
3.2 (4 channels)
Second half 2004
2005?
India Indian Space Corporation
Cartosat 2
1 2.5 9.6 2005?
France SPOT/Pleiades HR 0.7 2.8 (4 channels?) 21 2008 (Nr 1)
2009 (Nr 2)
Open access?
South Korea/
Malaysia
MACSAT, a project of the
company SatRec, Korea and a
Malaysian government agency
2.5 5 (4 channels) 20 2004 Dual use?
Near equato-
rial orbit
Sources: Y. A. Dequanzada and A. M. Florini, Secrets for Sale: How Commercial Satellite Imagery Will Change the World,
Washington, DC: Carnegie Endowment for International Peace, 2000; G. Steinberg, Dual Use Aspects of Commercial High-
Resolution Imaging Satellites, Mideast Security and Policy Studies, Bar-Ilan University, Israel, No. 37, Feb. 1998,
www.biu.ac.il/SOC/besa/books/37pub.html; J. C. Baker, K. M. O’Connell and R. A. Williamson (eds), Commercial Observa-
tion Satellites, Santa Monica and Bethesda: RAND and ASPRS, p. 643; G. Petrie, “Current Developments & Future Trends
in High Resolution Imaging & Mapping from Space”, ISPRS Workshop—High Resolution Mapping from Space, 6 October
2003, Hannover, Germany.
187
Table 9.3: High-Resolution Commercial and Dual Use Radar Satellites
Country System Wave-
length (cm)
Resolution
(m)
Swath (km) Launch date Remark
Canada RADARSAT-2
www.rsi.ca
5.6
(C Band)
3-100 20-500 2005
Germany Terra SAR-X
(DLR/Astrium)
3.3
(X Band)
(a) 1
(b) 2.5-3
(a) 5x10
(b) 15 (stripmode)
Oct. 2005
Italy/France Cosmo-Skymed
SAR X, 4 satellites
3.3
(X Band)
(a) 1
(b) few m
(a) 10x10
(b) stripmode
2005? Dual use
Japan IGS-1B
IGS-2B
3
?
23 March 2003
lost Nov. 2003
Mass 1.2t, dual
use?
USA RADAR-1 (Research
Development Laborato-
ries Space Corporation)*
1-5 4 ? Revisit time: 1
day
*RDL is the only US company which has received a license to launch a 1-metre resolution commercial radar satellite. RDL
was placed under stringent operating constraints and it was understood that the US government could insist on limiting the
resolution to 5 metres. According to the Federal Aviation Administration (endnote 12) the licence was subsequently can-
celled. Germany is also building five military radar satellites with resolution below 1 metre (scheduled to be launched
between 2005 and 2007).
Source: G. Petrie, “Current Developments & Future Trends in High Resolution Imaging & Mapping from Space”, ISPRS
Workshop—High Resolution Mapping from Space, 6 October 2003, Hannover, Germany.
188
Table 9.4: High-Resolution Military Optical Satellites in Europe and Asia
12
Little is known about Chinese high-resolution satellites. The Helios-2 satellites will have a medium-resolution
sensor for full night vision, probably in the mid-infrared (3.5-5 micrometres).
Source: G. Petrie, “Current Developments & Future Trends in High Resolution Imaging & Mapping from Space”, ISPRS Work-
shop—High Resolution Mapping from Space, 6 October 2003, Hannover, Germany.
Country System Resolution
at nadir
(Pan) [m]
Resolution at nadir
(MS) [m]
Launch date Remarks
France, Italy,
Spain
Helios-1A
Helios-1B
1
1
? July 1995
Dec. 1999
Share of France 79%
Mass of satellite 2.5 tons
France,
Belgium, Spain
Helios-2A
Helios-2B
0.5
0.5
(a) near infrared
(b) mid-infrared
2004 Share of France 95%
Mass of satellite 4.2 tons
Israel Ofeq 3 2 1997 Mass of satellite 225 kg, lasted
till 2000
Israel Ofeq 5 1 May 2003 Much heavier than Ofeq 3
Israel Ofeq 6, 7 below 1? 2004, 2008
India Technology Experi-
menal Satellite (TES)
1 22 Oct. 2002 Sensor from Israel (El Op Indus-
tries)
Japan IGS-1A
IGS-2A
1 28 March 2003
lost Nov. 2003
Supposed to be for dual use
South Korea Kompsat-2 1 4 (4 channels) Nov. 2004 Dual use (?)
Based on foreign technology
Taiwan ROCSAT-2 2 8 Feb. 2004 (?) Imager built by Alcatel Space
(France)
United
Kingdom
TopSat 1 2.5 2004 Enhanced microsatellite
Dual use
189
9.2 COMPARING APPLES AND ORANGES
In comparing high-resolution commercial satellite and Open Skies
imagery, first of all, one has to recall that both the Open Skies Treaty and
commercial satellite business have been designed with widely different
purposes in mind. Commercial companies sell pictures to almost any
customer irrespective of the end use of the images, be it research, media,
industrial development or data collection by foreign secret services. The
imagery may but does not necessarily increase transparency and openness.
The element of cooperation between the militaries of states in the data
taking process is non-existent. In contrast the Open Skies Treaty intends to
enhance transparency in military-security matters through joint overflights,
which embody the element of openness and cooperation. These qualities
and political achievements can never be replaced by the acquisition of
satellite images! The fact that a country agrees to be subjected to aerial
observation measures like Open Skies is in itself a visible sign of its readiness
for cooperation and openness.
On the other hand, if openness and cooperation between states have
developed sufficiently it may be fair to ask whether any remaining need for
image information could be met in a more cost-effective way by acquiring
commercial satellite images. It is only for the sake of this argument that we
try to compare the performance of Open Skies and commercial satellite
sensors and their images products.
The comparison is carried out according to five performance criteria:
•resolution;
• scheduling flexibility and access time;
• area coverage;
• hindrance by clouds;
•costs.
Resolution
Table 9.5 shows the pixel resolution (ground sampled distance) of
Open Skies sensors and commercial satellites. For completeness, we
include also the reported resolution of US reconnaissance satellites of type
KH11 (optical) and Lacrosse (radar).
13
Although commercial photo-
satellites reach resolution values of 0.6-1 metre in nadir mode the actual
resolution can be worse by factors 2 to 3 if the camera is tilted (for quick
190
access). Hence, Open Skies images are sharper by factors of at least 2 to 3
than commercial satellite images.
14
Even more important, the 0.5 metres
resolution of thermal infrared images under Open Skies is unrivalled by any
commercial or even military reconnaissance satellite!
15
In contrast, the 3-
metre resolution of Open Skies SAR sensors will be reached and surpassed
by commercial satellites soon. Hence, the strength of Open Skies is the
combination of good optical resolution combined with a unique thermal
infrared resolution.
Table 9.5: Resolution (Ground Sampled Distance) of Imaging Sensors on
Open Skies Aircraft, Commercial and Reconnaissance Satellites
Scheduling flexibility and access time
Commercial satellite operators offer priority data taking within a typical
timeframe of 7 to 14 days after order placement. Open Skies flights can be
scheduled and carried out within less than seven days (see section 2.3.4).
This would allow for reasonably quick responses in times of political
tension, assuming that quota were still available. Under normal
circumstances, however, Open Skies flights are planned in accordance with
the annual quota allocation and the quarterly sequencing. But in principle
the imaging access time of both systems is comparable. Yet, Open Skies is
vastly superior in allowing an overflight of many separate sites exactly at
nadir for optimum resolution.
Area coverage
Open Skies optical cameras cover ground swaths between about 2 and
28 kilometres (see Table 4.1), whereas the swath of optical 1-metre satellite
sensors is between 8 and 16 kilometres. Here, the capabilities are roughly
comparable.
Hindrance by clouds
Open Skies aircraft can underfly cloud ceilings at altitudes around
1,000 metres when using optical cameras. When using thermal infrared line
Sensor Optical Mid-infrared
at 3.5-5 µm (?)
Thermal
infrared
SAR
Open Skies
Commercial satellites
US reconnaissance satellites
0.3 m
0.6-1.0 m
0.1-0.5 m
-
-
0.6-0.9 m
0.5 m
(90 m)
-
3 m
1 m (2005)
0.6-0.9 m
191
scanners flight altitudes may even be slightly lower. Most importantly,
infrared line scanners can be operated irrespective of available daylight,
that is, 24-hours-a-day. In addition, thermal infrared radiation penetrates
haze, thus making Open Skies sensors all in all superior to satellites on
cloudy or hazy weather.
Costs
An Open Skies flight can image some 30 geographically separate sites
of interest within five flight hours at a total cost of about Euro 1,400 per
site.
16
Imaging of 30 geographically separate sites by commercial satellites
would require ordering 30 scenes of the mandatory minimum size (typically
10x10 kilometres) at a cost of Euro 20-40 per square kilometre and a total
cost of € 2,000-4,000 per site (standard geometric correction, no priority
tasking). Thus, Open Skies flights provide images of higher resolution at
approximately half of the cost. In addition, stereo coverage comes virtually
for free from the optical Open Skies cameras whereas stereo imagery from
commercial satellites is charged additionally.
In summary, Open Skies imaging is at present comparable or clearly
superior to the acquisition of commercial satellite imagery in all of the five
discussed technical categories. On the other hand, for objectives like
producing precision maps imagery from 1-metre satellites is the best choice.
9.3 UNMANNED AERIAL VEHICLES
UAVs are lightweight remotely controlled or autonomously navigated
aircraft, which can operate both in hazardous airspace and at very high
altitudes, without endangering a pilot. The performance and reliability of
UAVs has increased in the 1990s considerably. As a result, the US military
was able to use UAVs successfully for tactical reconnaissance in the recent
Kosovo, Afghanistan and Iraq wars. It is said that the US-made Predator can
operate at 12 kilometres altitude with an endurance of 20 hours and a
payload of 317 kilograms, whereas the Global Hawk flies at 20 kilometres
altitude with endurance up to 42 hours (range 25,000 kilometres) and a
payload of 481 kilograms.
17
Both systems can carry electro-optical, infrared
(probably in the wavelength band of 3.5-5 micrometres) and SAR sensors
with ground resolution in the range of 0.3-1 metre, similar to the resolution
of US reconnaissance satellites.
192
We have argued above that such systems are being operated in modes
(highly secret) and for purposes (strategic and battlefield reconnaissance)
which are completely different from those of Open Skies and thus cannot
substitute for Open Skies assets. Recently, however, UAVs have been tested
in a civilian experiment for the monitoring of wildfires. Such use of UAVs
might eventually compete with the use of Open Skies aircraft for disaster
monitoring. We therefore shortly elaborate on the potential and limitations
of UAVs in this context.
The above-mentioned experiment was performed by a collaboration
of NASA Ames Research Center with California State University and three
industrial firms.
18
An ALTUS II UAV built by General Atomics-Aeronautical
Systems, Inc., USA, was flown over test fires in a desert test field. By using
a 4-channel optical and infrared line scanner it was demonstrated that
georectified image and temperature data of wildfires could be obtained and
transmitted via a satellite data link and distributed via the Internet in nearly
real-time. As a next step an ALTAIR UAV (built also by General Atomics) will
be tested. This UAV flies at altitudes up to 16.7 kilometres with endurance
of up to 32 hours and a payload of up to 340 kilograms. At present, the
operational use of most UAV types is prohibited in regular airspace by rules
of the US Federal Aviation Administration (FAA) and by international
agreements.
19
The US UAVs industry is working closely with the FAA to
develop the regulatory framework to allow UAVs unrestricted access in
national airspace. These issues are expected to be resolved in the near
future allowing expanded use of UAVs for commercial and disaster support
activities.
In conclusion, although UAVs—similarly to satellites—exclude a basic
quality of Open Skies, the cooperative data taking by joint teams, they can
become a substitute means for Open Skies flights in the area of disaster
monitoring in the long run. It remains to be seen to what extent the use of
UAVs in peacetime will be agreed upon internationally.
Notes
1
Y. A. Dequanzada and A. M. Florini, Secrets for Sale: How Commercial
Satellite Imagery Will Change the World, Washington, DC: Carnegie
Endowment for International Peace, 2000; G. Steinberg, Dual Use
193
Aspects of Commercial High-Resolution Imaging Satellites, Mideast
Security and Policy Studies, Bar-Ilan University, Israel, No. 37,
Feb. 1998, www.biu.ac.il/SOC/besa/books/37pub.html.
2
J. C. Baker, K. M. O’Connell and R. A. Williamson (eds), Commercial
Observation Satellites, Santa Monica and Bethesda: RAND and ASPRS,
p. 643.
3
G. J. Tahu, “Russian Remote Sensing Programs and Policies”, in ibid.,
p. 166.
4
Y. A. Dequanzada and A. M. Florini, op. cit., pp. 18-19.
5
B. Preston, “Space Remote Sensing Regulatory Landscape”, in
J. C. Baker, K. M. O’Connell and R. A. Williamson (eds), op. cit.,
pp.501-31.
6
As an example, during the Afghanistan war of 2001 the US
government—instead of imposing shutter control—bought all the
imagery taken by Space Imaging over Afghanistan as a sole priority
customer. In contrast, no attempt to restrict the availability of imagery
was made during the 2003 Iraq war. John Pike, Global Security, Inc.,
Washington, DC, private communication.
7
Y. A. Dehquanzada and A. M. Florini, op. cit., p. 20.
8
Additionally, in June 2000 the Russian firm Sovinform Sputnik
announced that it was ready to distribute 1-metre resolution images
from Russian imagery archives.
9
John Pike, private communication, March 2003.
10
The new 1-metre satellite images form an excellent base for producing
high-resolution maps at scales of 1:10,000 or 1:25,000, depending on
the effort invested in applying geometric corrections. Prof. Heipke,
University of Hannover, Germany, private communication, 2003.
11
These contracts reflect a further shift in the US government’s image
acquisition policy. The National Reconnaissance Office has been faced
with a data supply crisis, since two of their three high-resolution KH11/
12 satellites have been in orbit since 1995/96 and need to be replaced
eventually. The operationalization of the subsequent generation
(Future Imagery Architecture, FIA) is much delayed. US President
George W. Bush has ordered in April 2003 federal agencies to rely to
the maximum practical extent on images from commercial satellites.
The new policy recognizes the improved quality and range of
commercial imagery. As a result of these developments, a contract in
excess of $500 million was awarded on 30 September 2003 to Digital
Globe to build a next generation of high-resolution commercial
satellites with NIMA/NGA as an anchor customer with privileged data
194
access. It is expected that Digital Globe and Space Imaging will operate
satellites of 0.5 metres resolution or better. Both Digital Globe and
Space Imaging were already been granted authorization to develop a
satellite capable of generating 0.5 metres imaging. Digital Globe and
Space Imaging have also applied for licenses to operate satellites at
0.25 metres resolution.
12
G. Petrie, “Current Developments & Future Trends in High Resolution
Imaging & Mapping from Space”, ISPRS Workshop—High Resolution
Mapping from Space, 6 October 2003, Hannover, Germany. For links
to the websites of remote sensing satellite operators see http://
www.weblinks.spakka.net/db/393. See also Federal Aviation
Administration and Commercial Space Transportation Advisory
Committee, 2002 Commercial Space Transportation Forecasts,
May 2002, http://www.futron.com/pdf/FAA%202002%20NGSO%20
Forecast.pdf. For information on Pleiades see http://www.space.se/
node2823.asp.
13
C. Couvault, “Secret NRO Recons Eye Iraqi Threats”, Aviation Week &
Space Technology, Vol. 16, 2002, p. 23.
14
This is illustrated by a comparison of Photo 4.1 (satellite image at
1-metre resolution) and Photo 6.1 (Open Skies image at
30 centimetres resolution).
15
The KH11 satellites carry infrared sensors. These operate probably—as
guessed by the author—in the spectral band 3.5-5 micrometers, not
8-13 micrometers, as can be deduced from the quoted resolution
values and the diameter of the mirrors on board (about 2.5 metres).
16
This cost estimate is based on a case study in which Germany leases an
AN-30 aircraft for a five-hour data taking flight as part of a five-day
mission in an European country:
(a) Aircraft and sensors leased at a cost of Euro 4,000 per flight hour
(price includes aircraft maintenance, depreciation, kerosene and
salaries of eight foreign flight and sensor operators). Total: € 20,000;
(b) Per diem of Euro 120 per day for eight foreign and ten German
personnel (four days). Arrival and departure day count as one day.
Total: € 8,640;
(c) Salary of ten German personnel for five days. Total: € 7,000;
(d) Cost for a group ticket on a commercial airline for ten German
personnel to and from point of entry. Total: about € 4,200;
(e) Cost of film and film development. Total: € 1,800.
Grand total: € 41,640, i.e., € 1,370 per site. Missions which require
trans-Atlantic transfer of Open-Skies aircraft will be more expensive.
195
17
Military Technology, Vol. 11, 2002, p. 55. See also A. R. Pustam,
“Strategic UAVs”, Military Technology, Vol. 12, 2003, pp. 60-70, which
quotes somewhat different performance values.
18
V. G. Ambrosia et al., “Demonstrating UAV—Acquired Real-Time
Thermal Data Over Fires”, Photogrammetric Engineering & Remote
Sensing, April 2003, pp. 391-402.
19
Because of its reliability, the Global Hawk was the first UAV to be
granted blanket clearance by the FAA to fly in US airspace. See
S. J. Zaloga, “UAVs Increase in Importance”, Aviation Week and Space
Technology, January 2004, p. 105.
196
197
CHAPTER 10
OUTLOOK: NOTHING TO HIDE?
PERSPECTIVES FOR THE OPEN SKIES TREATY
Pál Dunay
This concluding chapter briefly examines the political prospects for the
Open Skies Treaty. To recall, the Open Skies Treaty came into existence as
the Cold War was drawing to a close for the purpose of filling the double
task of European arms control verification and confidence building. The
future of the Open Skies Treaty is thereby contingent on its ability to
continue to fruitfully assume these two functions, as well as perhaps new
ones, in the face of two significant changes that have emerged since the
Treaty was signed: the reorganization of European security relations since
the end of the Cold War along largely non-conflictual lines, which has
severely diminished the scope for arms control and confidence building in
Europe, and the emergence of competing technical monitoring means in
the form of commercial high-resolution satellites and UAVs that threaten to
make Open Skies techniques obsolete.
Undoubtedly, the end of the Cold War and the reorganization of
security relations in Europe have largely diminished the scope for new, and
to some extent even existing, arms control measures on the continent. As a
result, the importance of the Open Skies Treaty for the verification of
structural arms control has practically vanished and there is no reason to
expect this trend to be reversed.
1
Although this observation appears to hold
grim implications for the future of Open Skies, in practice the situation is
somewhat different. As it happens, Western and Eastern European states
still want to inspect Belarus, Ukraine and especially the Russian Federation,
as in the days when there still was a Soviet Union. The current and
foreseeable practice of verification is thus in stark contrast to the general
perception of a decline in the utility of Open Skies due to the sharp
amelioration of European security relations. Moreover, it is also interesting
to note that formerly neutral countries such as Finland and Sweden have
198
acceded to the Open Skies Treaty and that many more from the Balkans
and all Baltic states have applied for accession. Quite curiously, many of
these states, which traditionally paid less attention to conventional arms
control, are still extremely keen to obtain as many flight quota allocated as
possible, and in fact insist on it on the diplomatic scene.
In terms of its other original purpose, military confidence building,
despite the vast improvement in European security relations since the end
of the Cold War, the Open Skies Treaty continues to retain a certain
amount of relevance. First, the Open Skies Treaty is still considered as a
useful implement in the management of relations between Western states
and Russia, in particular in the light of NATO expansion, by both sides. As
well, the fact that the Open Skies Treaty makes territories accessible for
overflights both in North America and in North Asia, neither of which was
theretofore accessible to such measures, makes it a valuable contributor to
enhancing confidence between Russia and the West. Second, the Open
Skies Treaty is widely seen and accepted as a key element of a pan-
European cooperative security structure, something that is regarded by
virtually all states in the region as extremely valuable of itself and which thus
lends the Treaty a large amount of political legitimacy. The recent
application of eight additional states for accession to the Treaty underscores
this point. Finally, although security relations in Europe have improved
tremendously over the past decade, new difficulties have emerged in
particular areas such as the former Yugoslavia, in the Caucasus and in
Central Asia. Here, the introduction of Open Skies procedures could make
a useful contribution in stabilizing relations between weary neighbours that
have either only recently been involved in open conflict or amongst whom
the spectre of war remains ever-present.
In addition to these two original purposes, arms control verification
and confidence building, the Open Skies Treaty could very well also take
on a number of additional functions in the future. As we have already seen,
the Open Skies Treaty is an extremely flexible and versatile tool that lends
itself well to the performance of valuable tasks not initially envisaged by its
founders but that are currently regarded as worthy. As such, Open Skies
measures could be deployed in support of international crisis prevention
and crisis management efforts within the framework of the OSCE or the
United Nations, it could be used as a model for the development of similar
arrangements in other regions of the world that would want to profit from
verification and confidence building and it could serve as an instrument for
199
environmental monitoring, particularly in terms of coping with natural
catastrophes, a role in which it has already proved itself. The inclusion of
some of these additional functions into the Open Skies regime will
necessitate an adjustment in the provisions of the Treaty, particularly with
respect to the type of sensors allowed and the modalities of their use, as we
have discussed in Chapter 7. However, the Open Skies Treaty being by its
design a very adaptable instrument, such adjustment could be
implemented without undue difficulties either through the OSCC or
through the accord of those parties directly concerned. The upcoming
review conference of 2005 provides a good opportunity to begin heading
in this direction.
Although the Open Skies Treaty still has a valuable role to play in the
new conditions of thaw of European security relations, both in terms of arms
control verification and confidence building, and increasingly in terms of
new tasks related to crisis management and environmental monitoring, the
advent of new technical means in the shape of easily accessible high-
resolution commercial imaging satellites and UAVs has been held by some
to imply the obsolescence of Open Skies. The argument made in support of
this claim is that high-resolution commercial imaging satellites and UAVs
can perform the same aerial monitoring functions as Open Skies, at a lower
cost. As we have shown above (see Chapter 9), however, this in fact is not
at all the case. When compared to high-resolution commercial imaging
satellites and UAVs, Open Skies retains decisive advantages or is at least
comparable in all decisive performance areas of resolution, scheduling
flexibility and access time, area coverage, hindrance by clouds and costs,
and is unmatched in its ability to foster cooperation and confidence by
virtue of the joint cooperative data taking that it implies, something that the
use of neither commercial high-resolution satellites nor of UAVs could
possibly accomplish. Moreover, an overhaul of the allowable sensor set and
operating procedures would further strengthen the advantages of Open
Skies over its technological competitors. Again, the review conference
scheduled for 2005 presents an excellent opportunity to address these
issues.
While the circumstance under which the Open Skies Treaty was
devised have radically changed since its signature, coping with change is
nothing new for the idea of Open Skies. The notion of Open Skies was born
at the height of the Cold War and has adapted itself successfully ever since
to suit the prevailing conditions. This characteristic of flexibility is in fact
200
embedded in every provision of the Open Skies Treaty, which remain open
to modification without the need for laborious and cumbersome
procedures of national ratification. The current trend of good neighbourly
security relations in most of Europe has diminished the role of arms control
and confidence building in the region, but has not eliminated it altogether.
The Open Skies Treaty, with the cooperative aerial monitoring that it
implies, continues to occupy an important place in the new cooperative
approach to security relations that has sprung up in Europe and its
contribution toward these is likely to remain appreciated by the states
parties for the foreseeable future. Moreover, the extension of the Treaty to
address other issues of rising interest such as conflict prevention, conflict
management and environmental monitoring can only strengthen its appeal,
especially in the light of its persisting advantages over other competing
technical means. Ever since its inception, the idea of Open Skies has been
carried by two related concerns: the project of cooperative security
relations within Europe and the need to adapt to circumstances in order to
bring this project into being and to sustain it. As of now, these two
imperatives look set to continue into the indefinite future.
Note
1
For example, the Adapted CFE Treaty no longer mentions aerial
inspections among its associated measures.
201
APPENDIX A
ADDRESSES OF OPEN SKIES UNITS
The table lists the units in charge of implementing the multilateral Open Skies
Treaty. The units are in most cases affiliated with the Verification Centre of the
Ministry of Defence (status April 2003).
Country Name of
Organisation
Fax Address
Belarus National Verification
Centre
+375 172 261
538
57 Kuibyshev Str.
Minsk 22 00 30
Belgium Verification Unit +32 2 7016671 Ministry of Defence
General Staff
JSO D/WV
Queen Elisabeth
Barracks
Everestraat
B-1140 Brussels
Bulgaria Arms Control Agency
(ACA), Open Skies
Dept.
+359 2 992 22
71
Totleben Bld. 34
1606 Sofia
Canada J3 Arms Control
Verification Directorate
National Defence-
Headquarters
Attn: J3 ACV 4
+1 613 9922348 MGen George R.
Pearkes Building
102 Colonel By Drive
Ottawa, ON K1A OK2
Croatia Croatian Verification
Centre
+38513784194 Ilica 242
10000 Zagreb
Czech
Republic
Arms Control Agency +420 2
20202161
160 00 Praha 6-
Dejvice
Tychonova 1
Denmark Tactical Air Command +45 99624955 Koluraa
DK-7470 Karup
Estonia Arms Control Verifica-
tion Section
+37 26661343 N.A.
Finland C/Arms Control Branch
Defence Staff
+358918182225
9
P.O. Box 919
00131 Helsinki
France Unité Française de
Vérification
+33 344 286292 BA-110
F-60314 Creil Cedex
202
Georgia Verification Center +995 32 990404 8 Gergeti Line
380008 Tbilisi
Germany Zentrum für
Verifikations-aufgaben
der Bundeswehr
Open Skies Division
+49 2451
992210
+49 2451
992230
Selfkant-Kaserne
Postfach 1391
52503 Geilenkirchen
Greece Arms Control Section +30 10 646 5037 156 Messogion
Athens
Great Britain Joint Arms Control
Implementation Group
(JACIG)
+44 1462
851515
x6317
+44 1462
813825
RAF Henlow
Bedfordshire SG16
6DN
Hungary Arms Control Agency
POC Open Skies
HDF JOC
+361 4741274 Balaton u. 7-11
1885 Budapest
Iceland Defence Dept.
Ministry of Foreign
Affairs
+35 4115 680 N.A.
Italy Verification Center
(CIVA)
+39 06
46915922
Aeroporto Roma
00040 Ciampino
Kyrgysztan Kyrgyz Verification
Section
+996 312228648 26 Logvirenko St.
720001 Bishkek
Latvia Latvian Air Force +3717207258 Int. Airport Riga
LV-1053 Riga
Lithuania Head of Supp. Div.
Lithuanian Air Force
+3707223650 Gedmino 25
3000 Kaunas
Luxembourg Groupe d’Inspections
Vérifications et
Observations Armée
Luxembourgeoise
+352 496306 38-44 rue Goethe
B.P. 1873
1018 Luxembourg
Netherlands Defence Staff,
Coordinator
Verification
Organization
Arms Control Agency,
Army Staff
Arms Control Agency,
Air Staff
+31 70 3187558
+31 70 3169258
+31 70 3397286
P.O. Box 20701
2500 ES The Hague
P.O. Box 90824
2509 LV The Hague
203
Norway Arms Control Branch
Coordinator Open
Skies
+47 23098319 HQ Defence
Command
Oslo Mil/Huseby
0016 Oslo
Poland Verification Unit
Department of Military
Foreign Affairs
+48 22 6826030 Ulica Krolewska
100-909 Warsaw 60
Portugal Portuguese Verification
Unit
Open Skies Section
+351 2
13013471
National Verification
Unit Portugal -
UNA VE/EMGFA-
Av. Ilha da Madeira
1449-004 Lisbon
Romania Verification Section
Regional Cooperation
& OS Office
+40 1 3122648 13-15 Izvor Street,
70462 Bucharest
Russia National Nuclear Risk
Reduction Center
(NRCC)
OS Section
+7 095 2004261 Znamenka Str.
19103160 Moscow,
K-160
Slovakia Slovak Verification
Centre
OS Department
+421 244250694 Ministerstvo obrany
SR, 132/23
Slovenské Verifikacné
Centrum
Kutuzovova 8
83247 Bratislava
Sweden Chief, Open Skies
Swedish Air Force
Command
+4618281579 P.O. Box 660
71528 Uppsala
Spain Spanish Verification
Unit (JUVE)
+34 91 4651942
+3491 5616322
Camino Ingenieros 6
28047 Madrid
Turkey Genelkurmay
GN.P.P.BSK.LIGI
SUGI.D.SKUD.S.
Disarmament Division
+90 312
4250813
+90 312
4183047
Genelkurmay
06100 Bakanliklar
Ankara
204
Ukraine Verification Directorate
of the GS of Ukrainian
Armed Forces
+380 44
2955429
+380 44
2440813
+380 44
2440828
+380 43
2212608
Vozdukhoflotsky Str. 6,
Kiev 252049
USA Open Skies, DTRA
European Operation
DTRA
+1 703 810 4893
+49 69 693482
Washington, DC
20166
45045 Aviation Drive
P.O. Box 5000
Rhein Main Air Base
60549 Frankfurt a. M.,
Flugplatz
205
APPENDIX B
OPEN SKIES AIRCRAFT
Photo B.1: One of the two US Open Skies aircraft, a Boeing OC-
135B (San Francisco, 1996). Photo: H. Spitzer.
The US Open Skies Aircraft
The United States uses two OC-135B observation aircraft to
implement the Open Skies Treaty. The Open Skies aircraft was
reconfigured from a WC-135B weather research and atmospheric sampling
plane (see Photo B.1). The OC-135B aircraft is assigned to Air Combat
Command’s 24th Reconnaissance Squadron at Offutt Air Force Base,
Nebraska. In modifying the WC-135B, the United States installed the full
complement of optical sensors permitted by the Treaty, including four
cameras in the aircraft fuselage and a variety of other equipment designed
to support the OC-135B’s observation mission. One vertical and two
206
oblique KS-87 framing cameras are used for low altitude photography
(1,500 metres above ground). Each camera has a field of view of 73
degrees. The axes of the oblique cameras are tilted by 38 degrees from the
vertical direction. One KA-91 panoramic camera provides a wide sweep of
93 degrees for high altitude photography between 5,000 and 11,000
metres above ground. Other modifications included installing an auxiliary
power unit, work stations for the observation flight representative
(inspector) team chief and the flight monitor (escort and interpreter) team
chief, crew luggage compartment, sensor operator console, flight-following
console, upgraded avionics and compartments to store and maintain film.
The aircraft has seating for 38 people, including the flight crew, aircraft
maintenance crew, foreign country representatives and crew members
from the Defense Threat Reduction Agency (DTRA). While the US Air Force
will provide the flight crew for the OC-135B, DTRA is responsible for
providing observation inspectors and escorts, linguists and other operations
support personnel for Open Skies overflight and escort missions.
Specifications:
1
• Power Plant Mfr: Four Pratt & Whitney TF33-P-5 Turbofans with thrust
reversers;
• Thrust: 16,050 pounds (7,222 kg) each engine;
• Speed: 460 knots (850 kmph);
• Unrefueled range: 3,900 miles (6,500 km);
• Dimensions: Wing Span 131 feet (39.9 m); Length 135 feet (41.1 m);
Height 42 feet (12.8 m);
• Max Takeoff Weight: 300,500 lbs (136,281 kg).
207
Photo B.2: The former German Open Skies aircraft, a Tupolev 154 M,
during its test certification in June 1997 (airfield Köln-Wahn). The aircraft
crashed two months later on 13 September 1997, leaving 33 dead. Photo:
R. Wiemker.
The former German Open Skies Aircraft
Germany retrofitted a Tupolev 154 M for Open Skies use (see
Photo B.2). This aircraft had been originally purchased by the German
Democratic Republic (GDR) to serve as the official aircraft of then GDR-
president Honecker. The aircraft was fully equipped with vertical and
oblique framing cameras and three color RGB video cameras.
2
In a tragic accident on 13 September 1997 the Tupolev collided with
a US “Starlifter” C-141, 100 kilometres west off the Namibian coast. Both
aircraft crashed at 5:10 p.m. from 12,000 metres altitude, leaving all 24
crew and passengers on board the Tupolev and the 9 on board the Starlifter
dead. The German Tupolev was flying southward from Niamey en route to
Cape Town. The flight was not related to Open Skies matters but carried
German officers on the invitation of the South African Naval Forces. The US
Starlifter was bound northwest flying en route from Windhoek to Ascension.
Neither aircraft was equipped with anti-collision warning systems (TCAS),
and the Starlifter did not feature a transponder. The German Ministry of
Defence decided not to retrofit another Tupolev 154 M for Open Skies use,
which left Germany without an Open Skies aircraft.
208
Photo B.3: The Hungarian Open Skies aircraft, an Antonov 26
at Tököl airfield, October 1996. Photo: R. Wiemker.
The Hungarian Open Skies Aircraft
Hungary uses a two-engine turboprop transport plane AN-26, made in
the former Soviet Union (see Photo B.3). The cruising speed is 390
kilometres/hour. One camera window was cut in the fuselage. After using a
French-made Omera-33 aerial camera for some years, Hungary recently
acquired a modern Leica Wild RC 30 framing camera.
Photo B.4: The UK Open Skies aircraft, an Andover twin turboprop
(United Kingdom).
209
The British Open Skies Aircraft
The United Kingdom’s Open Skies aircraft (see Photo B.4), a former
Royal Air Force Andover, was operated by the Aircraft and Armament
Evaluation Establishment at Boscombe Down. It was decommissioned in
April 2003. The aircraft has two turboprop engines. The cruising speed is
about 360 kilometres/hour at altitudes of about 2,500 metres. Cruising at
much higher altitudes is prevented by lack of a pressurized cabin. The
aircraft was equipped with one KA-95 B panoramic camera (made by
Recon/Optical Inc., USA), which scans a field of view of up to 164 degrees
with a rotating prism. The sensor operation altitude of the camera is above
2,500 metres. However, the camera was equipped with a forward motion
compensation mount and a special degrading filter for Treaty-compatible
operation at altitudes as low as 850 metres. The extremely wide field of
view provides for a ground swath of 12.2 kilometres at such low altitudes.
However, the Treaty resolution of 30 centimetres is obtained only in a
narrow swath around the vertical axis.
The scanning system is capable of six different scan angles:
40 V scanning 20° either side of vertical;
90 V scanning 45° either side of vertical;
90 L/R scanning 85° from the vertical left or right 5° past nadir;
140 V scanning 70° either side of vertical;
165 V scanning 82.5° either side of vertical.
The United Kingdom will have to lease foreign aircraft for future active
quota flights. It has expressed interest in leasing the Swedish aircraft.
210
Photo B.5: Image of a Lockheed C-130 Hercules (operated by the
Pod Group). Source: German Verification Center, Geilenkirchen.
The insert shows the sensor pod mounted under a wing.
The Open Skies aircraft of the Pod Group
The group consisting of Belgium, Canada, France, Greece, Italy,
Luxembourg, the Netherlands, Norway, Portugal and Spain uses C-130
Hercules aircraft (see Photo B.5) equipped with a “SAMSON” sensor pod
to conduct observation flights. The aircraft—a standard transporter—has
four turboprop engines. The cruising speed is 430 kilometres/hour
(maximum). The pod is a converted C-130 fuel tank that has been modified
to carry the permitted sensors. The costs of purchasing and maintaining this
pod are shared, based on each nation’s flight quota and actual use. The pod
was produced using existing photographic cameras. It is a unique piece.
The mounting time is about 6 hours. The cycle time for one mission and
maintenance at the Brussels base is about 10-14 days. The camera set
includes one KS-116 A panoramic camera, one vertical and two oblique KS-
87 B framing cameras and two SEKAI RSC-100 video cameras (forward
looking and vertical). The field of view of the photo cameras is:
KS-116 A along track: 20°;
across-track selectable: (1) 70° left to 70° right;
(2) 45° left to 45° right;
(3) 0° to 90° right;
(4) 0° to 90° left;
KS-87 B along and across track 74° x 74°;
angle of deviation of oblique sensor axis from vertical
direction: 32°.
211
The Antonov 30 aircraft operated by Bulgaria, the Czech Republic,
Romania, the Russian Federation and Ukraine
An-30 is a high-wing aircraft, powered by two AI-24 VT turboprop
engines with 2,072 kilowatts output each and one RU-19A-300 additional
turbojet unit with 8 kilonewton trust (see Photo B.6). The additional unit
supplies the necessary power for starting up the main engines and it is used
for taking-off, climbing and, should the main power pack fail, for flying.
The An-30 flies at an operation speed of 430 kilometres/hour at an
altitude of 6,000 meters. The fuel consumption is approximately from 850
up to 1000 kilograms/hour at that altitude with an un-refuelled range of
2,600 kilometres. The range for image taking flights at low altitudes (1,000
metres) is about 1,500 kilometres. The An-30 aircraft distinguishes itself by:
• high operational reliability;
•easy maintenance;
• the ability to take-off from and land on non-asphalted runways.
Photo B.6: The An-30 aircraft of Ukraine. Photo: A. Rothkirch, University
Hamburg.
212
One characteristic feature is the multi-window nose, which houses the
navigators cabin. It is situated under the elevated cockpit. The fuselage has
five camera hatches, which are closed by built-in blinds when not
operating. Russia and the Ukraine have equipped their aircraft with existing
cameras and navigation systems. Bulgaria, Romania and the Czech
Republic have installed new framing cameras of type Zeiss LMK 1000/9
(Czech Rep.) and Leica Wild RC30 (Bulgaria, Romania), as well as modern
navigation systems. Bulgaria operates also a panoramic camera Vinten 900
B with along track field of view of 41degrees and across track field of view
of 140 degrees. Details of all cameras which are currently in use on board
of Open Skies aircraft are given in Table C.1 (Appendix C).
The Tu 154-M aircraft of Russia
Russia is using a Tupolev 154-M aircraft for Open Skies. This aircraft
has an operational range between 2,500 and 5,000 kilometres depending
on flight altitude. It is thus much better suited for transatlantic flights than
the An-30. Flight tests and data gatherings took place in 2003. Certification
took place for spring 2004.
213
The Open Skies Aircraft of Turkey
Turkey has retrofitted one CASA CN-235M aircraft for Open Skies use
(see Photo B.7). The CN-235M is a light medium-range military transporter
with two turboprop engines. It was jointly developed by Spain and
Indonesia. The aircraft can take-off from short non-asphalted runways
independent of ground service. It cruises at an altitude of 6,800 metres with
a speed of 460 kilometres/hour. Physical characteristics of the aircraft
include:
• Length of freight bay: 9.65 metres;
• Width of freight bay: 2.36 metres;
• Height of freight bay: 1.84 metres;
• Max weight of load: 6 tons.
Turkey is equipping its aircraft with existing US-made vertical and
oblique framing cameras (KS-87), with a panoramic camera (KS-116) and
an infrared line scanner (AA/AAD-5). There will be one vertical and two
oblique video cameras, as well. The panoramic camera has six options for
the field of view: 40 degrees, 60 degrees, 90 degrees, 120 degrees, 140
degrees, 160 degrees. Certification took place in April 2004.
Photo B.7: The CASA CN-235-M transporter. Source: German
Verification Center, Geilenkirchen.
214
Photo B.8: The SAAB 340 aircraft of Sweden in its former capacity
as aircraft of the Royal Family. Source: German Verification Center,
Geilenkirchen.
The Swedish Open Skies Aircraft
Sweden has dedicated a Saab 340 two-engine turboprop aircraft for
Open Skies use (see Photo B.8). The aircraft will be equipped with a ZEISS
RMK framing camera and probably also an infrared line scanner AA/AAD-
5, both of which have been provided by Germany. Close cooperation
between Sweden and Germany in using the aircraft has been agreed upon
in a Memorandum of Understanding, which was signed on 14 May 2003.
Certification took place in April 2004.
Notes
1
See www.dtra.mil/news/fact/nw_oc135b.html (January 2003).
2
B. Uhl, “High Resolution Digital Colour EO Camera System VOS”,
Proceedings of the Third International Airborne Remote Sensing
Conference and Exhibition, Copenhagen, Environmental Research
Institute of Michigan, Ann Arbor, Vol. II, 1997, pp. 21-28.
215
APPENDIX C
SENSOR PROPERTIES
The focus of the trial implementation of the Open Skies Treaty from
1992 to 2001 and of the certification events in 2002 has been on testing
and operating optical cameras. The cameras themselves take little space as
shown in Photo C.1, the control electronics are more bulky, as shown in
Photo C.2. Different combinations of cameras, film and filters were used
and the resulting Treaty-compatible sensor operation altitude H
min
was
determined. The finer the grain and resolution of film and camera the
higher the flight altitude H
min
at which the Treaty-compatible minimum
resolution of 30 centimetres is obtained. Table C.1 shows from left to right
the state party, sensor (camera) type, camera field of view (FOV), film type
and width, filter type, focal length of camera objective, H
min
and ground
swath covered while operating at H
min
. This list contains all configurations,
which passed certification by August 2002.
Photo C.1: Panoramic camera mounted on board the United
Kingdom Open Skies aircraft. Photo: A. Rothkirch, University
Hamburg.
216
Table C.1 : Sensor Specifications of Open Skies Aircraft
State Sensor FOV
degree
Film Width
mm
Filter
(factor)
Objective
mm
H
min
m
Swath
km
1 BUL Leica Wild RC 30 74 PAN 200 240 Dark Yellow (2) 152.9 3,149 4.7
2 Leica Wild RC 30 74 KODAK 3404 240 Dark Yellow (2) 152.9 2,724 4.1
3 Vinten 900B 140 KODAK 3404 70 Minus BLUE (2) 76.2 1,174 6.5
4 RUS AFA 41-7.5 99 TYP 42 L 190 Zhs-18 (Y_2.62) 75 1,210 2.8
5 AFA 41-10 (2121) 84 TYP 42 L 190 Zhs-18 (Y_2.7) 100 1,711 3.1
6 AFA 41-10 (2123) 84 TYP 42 L 190 Zhs-18 (Y_2.7) 100 1,711 3.1
7 AFA 41-10 (2221) 84 TYP 38 190 Zhs-18 (Y_2.7) 100 3,103 5.6
8 AFA 41-10 (2233) 84 TYP 38 190 Zhs-18 (Y_2.7) 100 3,103 5.6
9 ROM Leica Wild RC 30 74 KODAK 2403 240 Yellow (2) 152.9 2,000 3.0
10 Leica Wild RC 30 74 PAN 200 240 Yellow (2) 152.9 2,800 4.2
11 HUN Leica Wild RC 30 74 KODAK 2403 240 Yellow (2) 152.9 1,972 3.0
12 Leica Wild RC 30 74 KODAK Plus-X 2402 240 Yellow (2) 152.9 2,993 4.5
13 CZ LMK 1000/9 104 FOMA AIR 200 240 Yellow 89 1,803 4.6
14 LMK 1000/9 104 FOMA AIR 200 240 Orange 89 1,803 4.6
15 LMK 1000/9 104 PAN 200 240 Yellow 89 2,047 5.2
16 USA KA-91 C 93 KODAK 3412 127 Yellow (2) 427.2 10,814 22.8
17 KA-91 C 93 KODAK 3404 127 Yellow (2) 427.2 6,169 14.0
18 KA-91 C 93 KODAK 3404 127 Red (4) 427.2 6,624 14.0
19 KA-91 C 93 SO-050 KODAK TRI X 127 Yellow (4) 427.2 4,834 10.2
20 KS-87 E (left) 145 SO-050 KODAK TRI X 127 Yellow (4) 152.4 2,099 13.3
21 KS-87 E (left) 145 SO-050 KODAK TRI X 127 Red (8) 152.4 2,224 14.1
22 KS-87 E (left) 145 KODAK 3404 127 Yellow (2) 152.4 3,004 19.1
23 KS-87 E (left) 145 KODAK 3404 127 Red (4) 152.4 2,801 17.8
24 KS-87 E (vert.) 73.7 KODAK 3404 127 Yellow (2) 76.2 1,506 2.3
25 KS-87 E (vert.) 73.7 KODAK 3404 127 Red (4) 76.2 1,440 2.2
26 KS-87 E (vert.) 73.7 KODAK 3412 127 Yellow (2) 76.2 2,172 3.3
217
27 KS-87 E (vert.) 73.7 KODAK 3412 127 Red (4) 76.2 1,954 3.0
28 KS-87 E (right) 145 SO-050 KODAK TRI X 127 Yellow (4) 152.4 2,099 13.3
29 KS-87 E (right) 145 SO-050 KODAK TRI X 127 Red (8) 152.4 2,224 14.1
30 KS-87 E (right) 145 KODAK 3404 127 Yellow (2) 152.4 3,004 19.1
31 KS-87 E (right) 145 KODAK 3404 127 Red (4) 152.4 2,801 17.8
32 UKR AFA 41-7.5 (V) 99 TYP 42 L 190 ZhS-18 (Y_1.5) 75 1,073 2.5
33 AFA 41-20 (V) 48.5 TYP 42 190 ZhS-18 (Y_2.0) 200 2,308 2.1
34 POD KS 116 140 KODAK 3404 127 Yellow (1.8) 305 5,290 29.1
35 KS 116 140 KODAK 3404 127 Yellow (1.8) Red.F. 305 766 4.3
36 KS 116 140 Agfa PAN 200 127 Yellow (1.8) 305 4,987 27.4
37 KS 116 140 Agfa PAN 200 127 Yellow (1.8) Red.F. 305 824 4.5
38 KS 116 140 SO-050 KODAK TRI X 127 Yellow (1.8) 305 3,999 22.0
39 KS-87 B (left) 150 KODAK 3404 127 Yellow (1.8) 76.2 1,622 12.1
40 KS-87 B (vert.) 74 KODAK 3404 127 Yellow (1.8) 76.2 1,684 2.5
41 KS-87 B (right) 150 KODAK 3404 127 Yellow (1.8) 76.2 1,996 14.9
42 KS-87 B (left) 150 Agfa PAN 200 127 Yellow (1.8) 76.2 1,758 13.1
43 KS-87 B (vert.) 74 Agfa PAN 200 127 Yellow (1.8) 76.2 1,965 3.0
44 KS-87 B (right) 150 Agfa PAN 200 127 Yellow (1.8) 76.2 1,903 14.2
45 KS-87 B (left) 150 SO-050 KODAK TRI X 127 Yellow (1.8) 76.2 1,308 9.8
46 KS-87 B (vert.) 74 SO-050 KODAK TRI X 127 Yellow (1.8) 76.2 1,354 2.0
47 KS-87 B (right) 150 SO-050 KODAK TRI X 127 Yellow (1.8) 76.2 1,545 11.5
48 UK KA-95 B 164 KODAK SO-50 127 Yellow IDF Red.F. 305 750 10.7
Table C.1 shows from left to right the states parties, sensor (camera) type, camera field of view, film type and width, filter
type, focal length of camera objective, H
min
and ground swath covered while operating at H
min
. Red.F. denotes the use of
a reduction filter. This list contains all configurations that had passed certification in 2002. The configurations of the Czech
Republic and Romania are tentative, and have still to be certified.
218
As mentioned before, the United Kingdom has used a special
degrading filter in order to operate its high-resolution panoramic camera at
altitudes as low as 850 metres. Similarly the framing camera on the Swedish
aircraft and the panoramic camera of the Pod Group can be operated with
different kinds of degrading filters in order to allow for a wide range of flight
altitudes between 800 and about 5,000 metres.
Photo C.2: Camera control electronics on board the Ukrainian
Open Skies aircraft. Photo: A. Rothkirch, University Hamburg.
Thermal infrared line scanners: The United States tested in 1995-97
a thermal infrared line scanner of type AA/AAD-5, which was formerly used
for reconnaissance missions.
1,2,3
Germany has set aside similar sensors of
type AA/AAD-5 for Open Skies use. Germany has performed several
demonstration flights of that sensor mounted in a Transall C-160
transporter. One of the sensors will probably be installed as a German-
Swedish cooperation project on the Swedish Open Skies aircraft. The
sensor readout has been reconverted from digital to wet film readout. Some
of the relevant parameters are shown in Table C.2.
219
Turkey is also preparing for the installation of an AA/AAD-5 infrared
line scanner on its Open Skies aircraft. Russia is planning to mount an
infrared line scanner with an angular resolution of 0.3 milliradians, field of
view of 120 degrees and digital readout.
Tab le C.2 : Sensor Parameters of the
German AA/AAD-5 Thermal Infrared Line Scanner
Radar sensors: The US has modified an existing US AN/APD-12
analog radar system for Open Skies use (SAROS). The SAROS system has
digital recording of radar, motion and annotation data. SAROS is an X-band
SAR operating at 9.6 gigahertz. The ground coverage is a constant 18.5
kilometres swath, located to the side of the aircraft ground track at a range
dependent on the mode of operation selected. SAROS performance and
image quality specifications, an illustration of the baseline mission
geometry, and a table listing SAROS modes of operation is shown below in
Figure C.1 and Table C.3.
4,5
SAROS has been further modified since 1996,
however system tests have been halted for the time being. Russia, in
cooperation with Germany, has developed a SAR-system ROSSAR for
Open Skies use. Russia is also considering another existing system. Russia
already installed one of the SAR systems on its TU-154 aircraft for testing.
Spectral sensitivity
Ground swath
Angular resolution (instantaneous field of
view)
Scan angle (Field of view)
Ground swath at flight altitude h = 1 km
H
min
Temperature resolution
Film type
Film width
Film length/cassette
Film coverage/cassette at h = 3,500 m
8-13 micrometers
1.15 x h (h = flight altitude)
(a) 0.25 mrad
(b) 0.50 mrad
(a) ± 30 °
(b) ± 60 °
(a) 1,150 m
(b) 3,460 m
1,500-2,000 m (to be verified)
0.2 °C
Kodak RA R2494
5 inch
350 ft
3,500 km
220
Figure C.1: SAROS specifications and mission geometry
Ta ble C.3 : SAROS Operating Modes
SAROS S
PECIFICATIONS
Parameter
Radar Center Frequency
Value
9.6 GHz ± 10 MHz
Aircraft Velocity
Squint Angle
Azimuth Resolution
Slant Range Resolution
Image Dynamic Range
Peak Sidelobe Ratio
Integrated Sidelobe Ratio
Ambiguity Level
Geometric Distortion
85 - 278 m/s
90°
3.3 ± 0.3 m
3.3 ± 0.3 m
³ 50 dB
£ -25 dB
£ -10 dB
£ -20 dB
< 1 IPR or 2%
221
Notes
1
L. Lesyna, A. Grillo, G. Gilchrist, M. Pagnutti, P. Saatzer, K. Simco and
R. Yurman, “Characterization of Infrared Line Scanners for the Treaty
on Open Skies”, Proceedings of the Third International Airborne
Remote Sensing Conference and Exhibition, Copenhagen, Denmark,
Environmental Institute of Michigan, Ann Arbor, Vol. II, 1997, p. 2.
2
V. Kumar, P. Saatzer, W. Goede, Maj. Rhett Ferguson and Ken Fortner,
“Film vs. Magnetic Tape Recording IRLS AN/AAD-5 for Open Skies
Imaging”, Proceedings of the Second International Airborne Remote
Sensing Conference and Exhibition, San Francisco, California,
Environmental Institute of Michigan, Ann Arbor, Vol. III, 1996,
pp. 200-9.
3
R. S. Bird and C. S. Kaufmann, “Digital Conversion of Infrared Camera
for ‘Open Skies’ Application”, Proceedings of the Third International
Airborne Remote Sensing Conference and Exhibition, Copenhagen,
Denmark, Environmental Institute of Michigan, Ann Arbor, Vol. II,
1997, pp. 3-11.
4
For further details on the SAROS system see K.R. Fortner and
P. L. Hezeltine, “The Open Skies Synthetic Aperture Radar (SAROS)”,
Proceedings of the Second International Airborne Remote Sensing
Conference and Exhibition, San Francisco, California, Environmental
Research Institute of Michigan, Ann Arbor, Vol. III, 1996, pp. 359-65.
5
P. L. Hezeltine, K. R. Fortner and J. B. Floyd, “Portable SAR Data
Processor for System Resolution Determination”, ibid., pp. 368-76.
222
223
APPENDIX D
OPEN SKIES TEST MISSIONS AND QUOTA FLIGHTS
(AS OF DECEMBER 2002)
Date Mission Observing
State
Observed
States
Additional
Observers
Observation
Aircraft
02.04.92 Test flight Benelux Poland C-130
07.04-08.04.92 Test flight Poland Benelux C-130
02.09-07.09.92 Trial observa-
tion flight
UK Russia,
Belarus
Andover
Oct./Nov. 92 SAR Test flights CND, DK, RUS Hungary 3 aircraft
April 93 Training USA Canada CV-580
16.06-19.06.93 Test flight Russia UK Belarus AN-30
15.06-17.06.93 Test flight Hungary Romania AN-26
10.07-12.07.93 Test flight USA Hungary CV-580
26.07-30.07.93 Test flight Germany Russia AN-30
23.08-26.08.93 Test flight Russia Germany AN-30
05.12-10.12.93 Test flight Germany USA CV-580
07.02-11.02.94 Test flight USA Germany OC-135
28.02-04.03.94 Test flight
WEU-Group
realized by UK B, D, F, UK,
NL
Andover
08.03-10.03.94 Test flight Russia France TU-154 M
14.03-17.03.94 Test flight UK Ukraine Andover
27.03-31.03.94 Test flight
WEU-Group
realized by F D, UK, NL Andover
10.04-16.04.94 Test flight Romania Hungary AN-30
18.04-22.04.94 Test flight USA Greece OC-135
25.04-29.04.94 Test flight Ukraine UK
23.05-24.05.94 Test flight UK Ukraine
24.08-31.08.94 Test flight Ukraine USA OC-135
05.09-09.09.94 Test flight Slovakia Ukraine
26.09-30.09.94 Test flight Slovakia Ukraine AN-30
10.10-15.10.94 Test flight Ukraine Slovakia AN-30
24.10-28.10.94 Test flight Ukraine Germany AN-30
14.11-18.11.94 Test flight Germany Ukraine AN-30
05.12-09.12.94 Test flight USA Canada
16.01-20.01.95 Exercise Fly-
catcher
UK Norway Andover
28.02-03.03.95 Exercise Spar-
row Hawk
USA UK OC-135 B
224
20.03-24.03.95 WEU-Test
flight
Romania D, Benelux AN-30
24.04-28.04.95 Test flight Germany Spain TU-154 M
08.05-12.05.95 Test flight Germany Portugal TU-154 M
29.05-02.06.95 Test flight Ukraine Germany TU-154 M
06.06-07.06.95 Exercise Spar-
row Hawk
UK Slovakia Andover
19.06.95 Test flight France Czech Rep. AN-30
26.06-02.07.95 Test flight Germany Canada TU-154 M
26.06-30.06.95 WEU-Test
flight
UK I, UK, E Andover
17.07-21.07.95 Test flight USA Germany OC-135 B
07.08-11.08.95 Test flight Germany Ukraine TU-154 M
11.09-15.09.95 Test flight Russia Germany AN-30
24.09-28.09.95 Test flight Germany Poland TU-154 M
09.10-17.10.95 Test flight Germany Russia TU-154 M
23.10-27.10.95 Test flight UK Ukraine Andover
23.10-27.10.95 Test flight Czech Rep. France AN-30
23.10-27.10.95 Test flight Poland Germany TU-154 M
27.10-03.11.95 Test flight Italy Romania AN-30
06.11-11.11.95 Test flight Germany Romania AN-30
12.02-16.02.96 Test flight Germany Italy TU-154 M
10.03-21.03.96 Test flight UK Georgia
March 96 Test flight UK Czech Rep.
19.03.96 Exercise Adv.
Express
TU-154 M
April 96 Test flight USA Canada
09.04-24.04.96 Reference
data gathering
Germany Spain TU-154 M
06.05-10.05.96 Test flight Hungary Germany AN-26 T
13.05-17.05.96 Test flight UK Slovakia
20.05-24.05.96 Test flight Germany Turkey TU-154 M
10.06-14.06.96 Test flight Italy Czech Rep.
01.07-05.07.96 Test flight UK Hungary
08.07-12.07.96 Test flight USA Czech Rep.
15.07-19.07.96 Test flight Bulgaria Germany AN-30
12.08-16.08.96 Test flight Hungary UK
26.08-30.08.96 Test flight Germany Bulgaria TU-154 M
02.09-06.09.96 Test flight Poland UK
02.09-06.09.96 Test flight Germany Hungary TU-154 M
23.09-27.09.96 Test flight Czech Rep. Germany AN-30
30.09-04.10.96 Test flight UK Romania
07.10-11.10.96 Test flight Romania UK AN-30
14.10-18.10.96 Test flight Romania Germany AN-30
14.10-18.10.96 Test flight USA Ukraine
21.10-25.10.96 Test flight Germany Czech Rep. TU-154 M
21.10-25.10.96 Test flight UK Poland
225
28.10-01.11.96 Test flight USA Hungary
11.11-22.11.96 Reference
data gathering
Germany Spain TU-154 M
03.02-08.02.97 Test flight Poland USA OC-135 B
10.02-15.02.97 Test flight Germany Greece TU-154 M
09.03-14.03.97 Test flight Slovakia USA OC-135 B
10.03-21.03.97 Test flight UK Georgia Andover
17.03-21.03.97 Test flight Hungary Romania AN-26
07.04-11.04.97 Test flight Germany Slovakia TU-154 M
16.04-24.04.97 Test flight Ukraine USA AN-30
21.04-25.04.97 Test flight Germany Czech Rep. TU-154 M
25.04-05.05.97 Test flight Georgia UK Andover
May 97 Test flight Hungary USA OC-135 B
12.05-17.05.97 Test flight USA Poland OC-135 B
02.06-06.06.97 Test flight Ukraine Poland AN-30
16.06-23.06.97 Test certifica-
tion
multilateral Germany TU-154 M
23.06-27.06.97 Test flight Poland Ukraine AN-30
30.06-04.07.97 Test flight Czech Rep. Germany AN-30
30.06-04.07.97 Test flight Russia UK AN-30
14.07-18.07.97 Test flight Russia Germany AN-30
14.07-19.07.97 Test flight USA UK OC-135 B
21.07-25.07.97 Test flight Hungary Germany AN-26 T
22.07-24.07.97 Desaster mon-
itoring (Oder)
Germany Germany/
Poland
TU-154 M
28.07-01.08.97 Desaster mon-
itoring (Oder)
Germany Germany/
Poland
TU-154 M
28.07-01.08.97 Test flight Russia USA AN-30
04.08-09.08.97 Test flight Russia Canada AN-30
18.08-23.08.97 Test flight USA Russia OC-135 B
25.08-30.08.97 Test flight Bulgaria UK AN-30
01.09-05.09.97 Test flight Germany Hungary TU-154 M
01.09-05.09.97 Test flight Turkey Russia AN-30
08.09-12.09.97 Test flight Germany Russia TU-154 M
08.09-12.09.97 Test flight Ukraine France AN-30
14.09-20.09.97 Test flight Italy Russia AN-30
22.09-26.09.97 Test flight Russia France AN-30
22.09-26.09.97 Test flight Turkey USA OC-135 B
29.09-03.10.97 Test flight Norway Russia C-130-Pod
06.10-10.10.97 Test flight Germany Romania AN-30
06.10-10.10.97 Test flight Bulgaria Germany AN-30
20.10-24.10.97 Test flight France Russia C-130
03.11-07.11.97 Test flight Germany Bulgaria AN-30
03.11-07.11.97 Test flight France Ukraine AN-30
10.11-14.11.97 Test flight Russia Italy AN-30
Not known Test flight France Romania AN-30
Not known Test flight Ukraine Romania AN-30
226
Not known Test flight Russia Turkey AN-30
12.01-16.01.98 Test flight USA Turkey OC-135 B
02.02-06.02.98 Test flight Czech Rep. USA OC-135 B
02.03-06.03.98 Test flight Czech Rep. UK AN-30
16.03-20.03.98 Test flight UK Czech Rep. AN-30
06.04-10.04.98 Test flight Russia Bulgaria AN-30
06.04-10.04.98 Test flight Hungary France AN-26
06.04-10.04.98 Test flight Romania Germany AN-30
Not known Test flight Ukraine Italy AN-30
13.04-17.04.98 Test flight Bulgaria Romania AN-30
20.04-24.04.98 Test flight USA Czech Rep. AN-30
20.04-24.04.98 Test flight Hungary Germany AN-26
20.04-24.04.98 Test flight UK Russia
25.04-30.04.98 Test flight Italy Ukraine AN-30
27.04-08.05.98 Reference
data gathering
Germany Bulgaria AN-30
04.05-08.05.98 Test flight Germany Hungary AN-26
11.05-15.05.98 Test flight Turkey Romania
11.05-15.05.98 Test flight Russia Germany AN-30
18.05-22.05.98 Test flight Czech Rep. Turkey AN-30
18.05-22.05.98 Test flight Romania UK AN-30
18.05-22.05.98 Test flight Bulgaria Russia RUS AN-30
25.05-28.05.98 Test flight
USA, UK
Georgia OC-135 B
01.06-05.06.98 Test flight France Russia AN-30
01.06-05.06.98 Test flight USA Bulgaria OC-135 B
08.06-12.06.98 Test flight UK Romania AN-30
08.06-12.06.98 Test flight Russia Norway AN-30
08.06-12.06.98 Test flight Germany Czech Rep. AN-30
15.06-19.06.98 Test flight Germany Russia AN-30
16.06-19.06.98 Test flight Turkey Bulgaria AN-30
22.06-26.06.98 Test flight Bulgaria Turkey AN-30
22.06-26.06.98 Test flight USA Ukraine OC-135 B
23.06-26.06.98 Test flight Hungary Romania AN-26
29.06-03.07.98 Test flight Turkey Czech Rep. AN-30
06.07-10.07.98 Test flight Romania Hungary AN-30
06.07-10.07.98 Test flight Russia Germany AN-30
13.07-19.07.98 Test flight Slovakia UK CZE AN-30
20.07-24.07.98 Test flight Romania Turkey AN-30
26.07-01.08.98 Test flight Bulgaria USA AN-30
27.07-31.07.98 Test flight Germany Ukraine AN-30
03.07-07.08.98 Test flight Romania Bulgaria AN-30
07.08-14.08.98 Test flight Turkey Ukraine AN-30
10.08-14.08.98 Test flight Germany Russia AN-30
10.08-14.08.98 Test flight Canada USA C-130 H
24.08-28.08.98 Test flight Germany Romania AN-30
227
24.08-28.08.98 Test flight Czech Rep. Germany AN-30
24.08-28.08.98 Test flight Russia UK AN-30
24.08-28.08.98 Test flight Turkey Hungary AN-26
31.08-04.09.98 Test flight Poland Germany UKR AN-30
31.08-04.09.98 Test flight Russia France AN-30
21.09-25.09.98 Test flight Ukraine UK AN-30
21.09-25.09.98 Test flight Bulgaria Germany AN-30
05.10-09.10.98 Test/ Training
flight
USA Germany OC-135 B
12.10-16.10.98 Test flight Germany Poland OC-135 B
12.10-16.10.98 Test/ Training
flight
UK Slovakia Andover
26.10-30.10.98 Test flight Romania Germany AN-30
25.01-29.01.99 Test flight France Germany C-130 H
08.02-12.02.99 Test flight Ukraine USA OC-135 B
20.02-26.02.99 Test flight France USA C-130 H
22.02-26.02.99 Test flight Romania Hungary AN-30
01.03-05.03.99 Test flight USA Romania Germany OC-135 B
22.03-26.03.99 Test flight Germany Czech Rep. France,
Tur key
US OC-135 B
30.03-02.04.99 Test flight Hungary Romania AN-26
12.04-16.04.99 Test flight Italy Hungary C-130 H
17.04-23.04.99 Test flight Czech Rep. USA OC-135 B
25.04-30.04.99 Test flight UK Georgia Andover
26.04-30.04.99 Test flight Russia Norway AN-30
03.05-07.05.99 Test flight Czech Rep. Germany AN-30
03.05-07.05.99 Test flight Italy USA Germany OC-135 B
17.05-21.05.99 Test flight Poland Russia Germany AN-30
17.05-21.05.99 Test flight Bulgaria France AN-30
07.06-11.06.99 Test flight Russia Germany AN-30
02.06-10.06.99
Tes t fligh t /
Data gathering
Bulgaria UK AN-30
07.06-11.06.99 Test flight USA Czech Rep. OC-135 B
21.06-25.06.99 Test flight UK Turkey Finland Andover
21.06-25.06.99 Test flight Russia Poland Belarus AN-30
21.06-02.07.99
Tes t fligh t /
Data gathering
Germany Bulgaria Denmark AN-30
28.06-02.07.99 Test flight Hungary Italy AN-26
05.07-09.07.99 Test flight Russia UK AN-30
06.07-10.07.99 Test flight Ukraine Germany AN-30
12.07-16.07.99 Test flight Italy Hungary AN-26
19.07-30.07.99
Tes t fligh t /
Data gathering
Bulgaria Germany AN-30
19.07-22.07.99 Test flight USA UK OC-135 B
23.07-27.07.99 Test flight USA Italy OC-135 B
228
26.07-30.07.99 Test flight Canada Russia C-130 H
26.07-30.07.99 Test flight Germany Hungary AN-26
02.08-06.08.99 Test flight Ukraine Canada C-130 H
02.08-06.08.99 Test flight Turkey UK Andover
08.08-13.08.99 Test flight Canada Ukraine C-130 H
09.08-13.08.99 Test flight Germany Russia AN-30
16.08-20.08.99 Test flight Poland UK AN-30
16.08-20.08.99 Test flight Russia Turkey AN-30
23.08-27.08.99 Test flight USA Italy OC-135 B/
C-130 H
23.08-27.08.99 Test flight Norway Russia AN-30
30.08-03.09.99 Test flight USA Bulgaria Germany OC-135 B
30.08-03.09.99 Test flight Turkey Ukraine AN-30 B
31.08-03.09.99 Test flight Hungary Romania AN-26
06.09-10.09.99 Test flight Germany Ukraine AN-30
06.09-10.09.99 Test flight Czech Rep. Russia RUS AN-30
13.09-17.09.99 WEU test flight USA Germany,
France, Spain
OC-135 B
13.09-17.09.99 Test flight Turkey Russia AN-30
20.09-24.09.99 Test flight UK Bulgaria Andover
27.09-01.10.99 Test flight Hungary Germany AN-26
27.09-01.10.99 Test flight Ukraine UK AN-30
04.10-08.10.99 Test flight Romania Germany AN-30
04.10-08.10.99 Test flight Russia Czech Rep. AN-30
11.10-15.10.99 Test flight UK Russia Andover
12.10-15.10.99 Test flight Hungary Romania AN-26
18.10-22.10.99 Test flight Turkey Bulgaria CASA CN-235
25.10-29.10.99
Tes t fligh t /
Data gathering
Germany Romania AN-30
25.10-29.10.99 Test flight Hungary Denmark Germany AN-26
01.11-05.11.99 Test flight Greece USA OC-135 B
08.11-12.11.99 Test flight Czech Rep. Slovakia AN-30
08.11-12.11.99 Test flight UK Ukraine Andover
08.11-12.11.99 Test flight Bulgaria Italy AN-30
13.11-18.11.99 Test flight Slovakia Czech Rep. AN-30
15.11-19.11.99 Test flight USA Greece C-130 H
29.11-03.12.99 Test flight Italy Bulgaria C-130 H
06.12-10.12.99 Test flight Germany Germany Slovenia BR-Atlantic
13.12-17.12.99 Test flight France Bulgaria C-130 H
17.01-21.01.00 Test flight USA France Germany OC-135 B
07.02-11.02.00 Test flight/
Desaster
monitoring
USA Germany OC-135 B
28.02-03.03.00 Test flight Italy Czech Rep. Germany C-130 H
Feb./ March 00 Test flight Romania Hungary AN-26
07.03-10.03.00 Test flight Hungary Romania AN-26
229
12.03-18.03.00 Data gathering Benelux Benelux Germany C-130 H
13.03-17.03.00 Test flight Ukraine UK AN-30
20.02-24.03.00 Test flight Czech Rep. Germany AN-30
27.03-31.03.00 Test flight Germany Romania AN-30
27.03-31.03.00 Test flight Georgia UK Andover
27.03-31.03.00 Test flight France Poland NL, Austria C-130 H
02.04-07.04.00 Test flight Greece Russia AN-30
03.04-07.04.00 Test flight Italy Ukraine AN-30
10.04-14.04.00 Test flight Germany Russia/
Belarus
AN-30 B
10.04-14.04.00 Test flight UK Ukraine Andover
25.04-28.04.00 Test flight France Czech Rep.
08.05-12.05.00 Test flight Romania UK AN-30
08.05-12.05.00 Test flight Benelux Hungary C-130 H
15.05-19.05.00 Test flight Czech Rep. France Slovakia AN-30
15.05-19.05.00 Test flight Germany Ukraine AN-30
15.05-19.05.00 Test flight Bulgaria Greece AN-30
15.05-19.05.00 Test flight Turkey Russia AN-30 B
22.05-25.05.00 Test flight USA Romania Germany OC-135 B
22.05-26.05.00 Test flight Russia WEU Bulgaria,
USA
TU-154 M
22.05-26.05.00 Test flight Ukraine Norway AN-30
26.05-29.05.00 Test flight USA Hungary Germany OC-135 B
29.05-02.06.00 Test flight Italy Romania C-130 H
05.06-09.06.00 Test flight Poland France AN-30
05.06-09.06.00 Test flight Germany Czech Rep. AN-30
05.06-09.06.00 Test flight Denmark Hungary AN-26
05.06-09.06.00 Test flight UK Russia AN-30 B
13.06-17.06.00 Test flight Russia Poland Bulgaria AN-30 B
17.06-24.06.00 Test flight Slovakia UK Czech Rep. Andover CMK 1
25.06-30.06.00 Test flight Hungary Benelux AN-26
21.07-31.07.00 Test flight Russia USA Germany TU-154 M
25.07-29.07.00 Test flight Bulgaria Germany AN-30
31.07-04.08.00 Certification
exercise
Bg, Cz, Hun,
Rom, Ukr
Germany 25 states five aircraft
07.08-11.08.00 Test flight UK Romania Andover
13.08-19.08.00 Test flight Germany Russia USA TU-154 M
04.09-08.09.00 Test flight Germany Hungary Slovenia AN-26
04.09-08.09.00 Test flight USA Ukraine OC-135 B
18.09-22.09.00 Test flight Germany Ukraine AN-30 B
18.09-22.09.00 Test flight Czech Rep. Russia Germany AN-30
25.09-29.09.00 Test flight Russia Germany AN-30 B
25.09-30.09.00 Test flight USA Russia Germany TU-154 M
01.10-06.10.00 Test flight Greece Hungary AN-26
02.10-06.10.00 Test flight Ukraine Germany USA AN-30 B
02.10-06.10.00 Test flight Czech Rep. Italy AN-30
09.10-13.10.00 Test flight UK Poland Andover
09.10-13.10.00 Test flight USA Benelux Germany OC-135 B
230
16.10-20.10.00 Test flight Russia Greece AN-30
23.10-27.10.00 Test flight Hungary Greece AN-26
30.10-10.11.00 Test flight Germany Bulgaria AN-30
04.11-11.11.00 Test flight Hungary USA Germany OC-135 B
12.11-17.11.00 Test flight Spain France C-130 H
27.11-01.12.00 Test flight Benelux USA Germany OC-135 B
22.01-26.01.01 Test flight Germany USA Bulgaria OC-135 B
04.02-10.02.01 Test flight Ukraine USA Germany OC-135 B
11.02-15.02.01 Test flight Spain Hungary C-130 H
26.02-02.03.01 Test flight USA Poland Germany OC-135 B
26.02-02.03.01 Test flight USA Ukraine OC-135 B
05.03-09.03.01 Test flight Romania Greece AN-30
12.03-16.03.01 Test flight Italy Poland C-130 H
19.03-23.03.01 Test flight Ukraine Germany AN-30 B
20.03-23.03.01 Test flight Hungary Romania AN-26
26.03-30.03.01 Test flight Russia Germany USA AN-30
02.04-06.04.01 Test flight Russia Norway Germany,
Latvia
AN-30
02.04-06.04.01 Test flight Romania France AN-30
21.04-27.04.01 Test flight Poland USA OC-135 B
23.04-28.04.01 Test flight Germany Russia/Belarus Poland AN-30 B
23.04-27.04.01 Test flight Ukraine UK AN-30 B
29.04-04.05.01 Test flight USA Canada OC-135 B
14.05-18.05.01 Test flight Bulgaria Spain Italy AN-30
12.05-19.05.01 Test flight Romania USA
14.05-18.05.01 Test flight France Ukraine
17.05-30.05.01 Test flight Russia WEU
(Germany,
Benelux, UK)
AN-30
28.05-01.06.01 Test flight Turkey Italy
28.05-01.06.01 Test flight Hungary Spain AN-26
05.06-12.06.01 Test flight Germany Hungary AN-26
07.06-11.06.01 Test flight Russia Denmark Germany,
Norway,
Sweden
AN-30 B
10.06-16.06.01 Test flight Greece Romania Norway
11.06-15.06.01 Test flight USA Norway OC-135 B
11.06-22.06.01 Test flight Germany Bulgaria Germany AN-30
11.06-15.06.01 Test flight Ukraine France AN-30 B
18.06-22.06.01 Test flight Germany Russia Czech Rep. AN-30
25.06-29.06.01 Test flight UK Romania ANDOVER
CMK 1
02.07-06.07.01 Test flight Germany Ukraine AN-30 B
10.07-13.07.01 Test flight Hungary Romania AN-26
16.07-20.07.01 Test flight Bulgaria UK AN-30
22.07-26.07.01 Test flight Italy Turkey C-130 H
231
30.07-13.08.01 Test certifica-
tion
CND, HUN,
RUS, UKR, UK
Germany 31 states five aircraft
13.08-17.08.01 Test flight USA Ukraine Germany OC-135 B
13.08-17.08.01 Test flight Russia Germany USA AN-30
20.08-24.08.01 Test flight UK Russia AN-30 B
03.09-07.09.01 Test flight Germany Hungary AN-26
03.09-07.09.01 Test flight Italy Russia AN 30
03.09-07.09.01 Test flight UK Bulgaria Andover CMK 1
10.09-14.09.01 Test flight Germany Russia Belgium AN-30
17.09-21.09.01 Test flight Romania Italy AN-30
17.09-21.09.01 Test flight Norway Russia Latvia AN 30 B
17.09-21.09.01 Test flight UK Ukraine Andover CMK 1
24.09-28.09.01 Test flight Denmark Russia AN-30 B
24.09-28.09.01 Test flight Germany Ukraine Kyrgyzstan AN-30 B
01.10-05.10.01 Test flight Spain Bulgaria C-130 H
02.10-05.10.01 Test flight Hungary Romania AN-26
08.10-12.10.01 Test flight UK Georgia Andover CMK 1
08.10-12.10.01 Test flight France Russia Belgium AN-30 B
15.10-19.10.01 Test flight USA Bulgaria Germany AN-30
15.10-19.10.01 Test flight Germany Czech Rep. USA AN-30
16.10-19.10.01 Test flight Romania Hungary AN-30
22.10-26.10.01 Test flight Russia Italy/France Belarus AN-30 B
29.10-02.11.01 Test flight Germany Romania AN-30
12.11-16.11.01 Test flight Benelux Czech Rep. C-130 H
12.11-16.11.01 Test flight Romania UK AN-30
27.11-30.11.01 Test flight USA Slovakia/
Czech Rep.
AN-30
19.02-22.02.02 Test flight Hungary Romania Germany AN-26
04.03-08.03.02 Test flight Germany Hungary USA AN-26
05.03-08.03.02 Test flight Romania Hungary AN-30
11.03-15.03.02 Test flight Ukraine Germany USA,
Ukraine
AN-30 B
18.03-22.03.02 Test flight Russia Germany Belarus AN-30 B
19.03-22.03.02 Test flight Romania Hungary AN-30
29.04-03.05.02 Test flight UK Bulgaria Andover CMK 1
30.04-04.05.02 Test flight Canada Portugal C-130 H
13.05-17.05.02 Test flight Russia Sweden AN-30
27.05-31.05.02 Test flight Czech Rep. Benelux AN-30
27.05-31.05.02 Test flight Germany Russia Ukraine,
Sweden,
Germany
AN-30
03.06-07.06.02 Test flight Sweden Russia Germany AN-30
232
17.06-21.06.02 Data gathering Romania Germany Sweden,
Finland,
Denmark,
Lithania,
Spain,
Latvia,
Estonia
AN-30
02.07-05.07.02 Test flight Bulgaria UK AN-30
15.07-
19.07.02
Test flight Germany Ukraine Denmark AN-30 B
20.07-26.07.02 Test flight Slovakia USA OC-135 B
05.08-09.08.02 Test flight USA Romania OC-135 B
05.08-09.08.02 Quota flight Russia UK AN-30 B
05.08-09.08.02 Quota flight Ukraine Hungary AN-30 B
12.08-16.08.02 Quota flight Germany Russia Norway AN-30 B
12.08-16.08.02 Quota flight Russia Benelux/
Germany
AN-30 B
19.08-23.08.02 Quota flight Ukraine Romania AN-30 B
26.08-30.08.02 Quota flight Belarus/Russia Norway AN-30 B
01.09-06.09.02 Test flight Finland Norway C-130 H
02.09-06.09.02 Quota flight Turkey Russia AN-30 B
02.09-06.09.02 Quota flight Ukraine Turkey AN-30 B
09.09-13.09.02 Quota flight Russia UK AN-30 B
09.09-13.09.02 Quota flight Poland Ukraine AN-30 B
09.09-13.09.02 Test flight Germany Bulgaria AN-30
16.09-20.09.02 Test flight USA Italy Germany OC-135 B
16.09-20.09.02 Quota flight Norway Russia Germany AN-30 B
16.09-20.09.02 Quota flight Hungary Ukraine AN-26
16.09-20.09.02 Quota flight Greece Bulgaria C-130 H
23.09-27.09.02 Quota flight Russia Germany Belarus AN-30 B
23.09-27.09.02 Test flight Denmark Bulgaria Finland,
Norway,
Sweden
23.09-27.09.02 Test flight Italy USA Germany OC-135 B
30.09-04.10.02 Quota flight France Russia AN-30 B
30.09-04.10.02 Quota flight Ukraine Poland AN-30 B
07.10-11.10.02 Quota flight UK Russia Andover CMK 1
07.10-11.10.02 Quota flight Russia Italy AN-30 B
14.10-18.10.02 Test flight Germany Ukraine AN-30 B
14.10-18.10.02 Quota flight Italy Russia
15.10-18.10.02 Quota flight Hungary Romania AN-26
21.10-25.10.02 Quota flight Russia France AN-30 B
21.10-25.10.02 Quota flight Ukraine Slovakia AN-30 B
28.10-01.11.02 Quota flight UK Ukraine Andover CMK 1
28.10-01.11.02 Test flight Norway Estonia,
Latvia, Lithua-
nia
C-130 H
04.11-08.11.02 Quota flight Russia Turkey AN-30 B
233
11.11-15.11.02 Test flight Benelux Bulgaria C-130 H
11.11-15.11.02 Test flight USA/Romania Greece Germany OC-135 B
AN-30
18.11-22.11.02 Quota flight Russia Greece AN-30 B
25.11-29.11.02 Quota flight Spain Slovakia France C-130 H
26.11-29.11.02 Test flight Romania Hungary
02.12-06.12.02 Test flight Romania Germany Bulgaria,
Finland,
Sweden
AN-30
07.12-13.12.02 Quota flight USA Russia OC-135 B
09.12-13.12.02 Quota flight Russia Turkey AN-30 B
234
235
APPENDIX E
VERIFYING THE GROUND RESOLUTION LIMIT
(PHOTO & VIDEO)
E.1 WHAT IS “RESOLUTION”?
“Resolution” is one of the central terms of the Treaty. Its definition,
measurement, and observance are responsible for the vast amount of
technical provisions of the Treaty, and constitute a notable obstacle in its
implementation.
1
Therefore, we want to probe a little further into the
question what “resolution” means, and clarify under which conditions the
resolution of an image can be enhanced (i.e., sharpened).
Considering digital data, we have to distinguish between the ground
sampled distance (GSD), which is also called pixel size or pixel footprint, on
the one hand, and the resolution (ground resolved distance, GRD) on the
other hand. The pixels are considered as samples taken from the true
continuous image at regular intervals, the ground sampled distance.
Let us consider a group of black bars of equal width on a white
background, in other words, black bars which are separated by white bars
of the same width (Figure E.1). We want to be able to resolve single bars
within this pattern. How fine a pixel size (GSD) is necessary? Obviously, the
absolute resolution limit is reached when the separation (SEP) is SEP = 1/2
GSD (Figure E.1), i.e., when each pixel covers equal areas of black and
white.
The modulation transfer function (MTF) describes how the contrast of
a sinusoidal wave pattern is diminished by the sensor with increasing spatial
frequency, i.e., with decreasing separation. Here, the spatial frequency is
the inverse of the separation.
2
The MTF of an ideal pixel-sampling sensor is
known to be the “sinc-function” (sin
w t/w t). In Figure E.4 we see that the
MTF (dashed line) vanishes for a frequency of 1/pixel, i.e., one wavelength
of 1 pixel contains a white and a black bar, and thus the bar width and the
Open Skies definition of separation is SEP = 1/2 pixel. With SEP = 0.5 GSD
being the theoretical resolution limit, the practical resolution limit is
236
commonly
3
taken to be at SEP = 0.7 GSD (see Figure E.1b). There, the MTF
has already dropped to 37 per cent of the original contrast (Figure E.4, top).
Figure E.1: Resolution of bar groups for various ground sampled
distances (pixel sizes).
Source: K. Kraus and W. Schneider,
Fernerkundung, Vol. 1—Physikalische Grundlagen, Bonn:
Dümmler, 1994.
The fundamental Sampling Theorem
4
tells us that if we consider one
black bar together with one white bar as one wavelength, we need at least
two samples for reproduction, i.e., a pixel size which is equal to the bar
width: SEP = 1.0 GSD. In Figure E.1c it can be seen, however, that such a
sampling can fully retain the contrast (CTF = 1), but also fully destroy the
contrast (CTF = 0), depending on the phase displacement between bar
pattern and sampling pattern. Thus, with SEP = 1.0 GSD, the detection of
the bar pattern still depends on accident.
237
In order to guarantee a reliable reproduction of the bar pattern, we
need a pixel size as fine as GSD = 0.7 SEP (Figure E.1d). In other words, we
can only reliably detect bar patterns with a separation of SEP = 1.4 GSD. In
this case, a resolution of 30 centimetres requires a pixel size of GSD = 0.21
centimetres. Then we have an MTF-value of 80 per cent as the optimal
case.
We have seen that “resolution” is not a sharply fixed quantity; rather,
contrast gradually vanishes with decreasing separation.
In accordance with the Treaty, when we talk about bar separation SEP
in the following, we always mean the width of the white gap between two
black bars (of the same width as the gap); not the distance between the
centres of the two black bars (common “ground resolved distance”).
For the example of GSD = 30-centimetre-pixels we can summarize
our considerations into the following rules of thumb:
These are theoretical best values assuming the absence of noise,
optical distortions and atmospheric effects. The values cannot be bettered
by any method of image “sharpening”.
Note that so far we have discussed the detectability of single bars in
continuous bar patterns. We have to keep in mind, though, that it may
always be possible to detect objects which are much smaller than the pixel
size (GSD) if only they exhibit a strong contrast to their background and if
the background is uniform. For an example see Photo E.1.
It should be mentioned that the geometric resolution is in principle
uniform across the image area for photographic film and images from CCD
With
30 cm GSD
(pixel size)
• reliable detection of 42 cm separated bar patterns
SEP = 1.4 GSD, ideal MTF = 81%
• probable detection of 30 cm separated bar patterns
SEP = 1.0 GSD, ideal MTF = 64%
• possible detection of 21 cm separated bar patterns
SEP = 0.7 GSD, ideal MTF = 37%
• absolute resolution limit of 15 cm separated bar patterns
SEP =0.5 GSD, ideal MTF = 0%
238
arrays or CCD lines (staring array or line camera). In contrast, the geometric
resolution of line scanners (such as for infrared) degrades with increasing
scan angle (~1/cos
2
q).
Photo E.1: Example for subpixel resolution: Left: Sportsfield as
imaged by an airborne line scanner with GSD (pixel size) ³ 70
cm. The white marking lines are still partially visible on the lawn
of the sports field, although the pixel size is ³ 70 cm and the
marking lines are of only » 10 cm width. Right: Image of the same
scene taken by a camera of about 10 cm resolution.
Blur/The Point Spread Function
The MTF of a real sensor system will always perform significantly worse
than the theoretical optimum of the “sinc-function”.
5
The less than optimal
modulation transfer is due to degradation by the detector, the optical
system, the atmospheric effects, etc. The MTF of the camera itself can of
course be determined by laboratory calibration measurements. However,
the in-flight recorded images will also be blurred by stray radiation from
neighboring pixels. The severeness of this adjacency effect depends on the
atmospheric conditions such as aerosol content, haze and visibility. The
image blur is described by means of the point spread function (PSF). A
common mathematical model for the PSF is given by equation (E.1)
where the halfwidth
s is a measure of the spread, and the exponent a
describes the steepness of the PSF. The ideal PSF is a box function with
s =1/2 pixel and a®¥, so that the pixel intensity is just the mean
intensity of the pixel’s ground footprint. In Figure E.4, we see three different
a
sa
as
x
ex
1
,
~)(PSF
-
239
PSFs (left-hand) and their respective MTFs (right-hand, which are the
absolute magnitude of the Fourier-transform of the PSF). Considering the
bars of the Open Skies calibration target, we see that the contrast at given
bar separations depends very sensitively on the form of the PSF. For a
halfwidth of
s = 2 pixel (Figure. E.4, bottom) we see that the contrast
vanishes completely even for separations as large as SEP = 1.4 GSD. In other
words, with a strong blur (i.e., a wide PSF) the resolution is degraded and
can be worse than what would be expected based on the GSD. Therefore,
it seems advisable to determine the GSD as well as the actual PSF effective
in the recording of a specific image. In the next section we will reconstruct
the actual PSF/MTF of an Open Skies test image by means of the calibration
target.
Computer Aided Automatic Resolution Measurement
The calibration target shown in Figure 4.2 (section 4.4) was recorded
from 900 metres altitude. Eye appraisal shows clearly that the 10
th
bar pair
(the last bar group left of the second wedge) is the smallest resolved one.
Thus, the resolution would—according to the Treaty—be estimated as
17.7 centimetres.
How could the resolution be determined in an automated fashion by
a computer algorithm? Research in the field of Computer Vision provides
the powerful tool for fitting a parametric model to an observed image. In
our case, we begin by building an exact representation of the calibration
target used in the calibration test. Then we simulate the image formation
process by merging the model of the calibration target with the analytic
model of the point spread function (equation E.1), and then sampling at the
stepwidth of the GSD. Since we do not know the true parameters of the
GSD, the PSF-halfwidth
s and its steepness a, we start with first-guess
values. Next we repeat the process with varying parameter guesses, and
each such simulated image of the calibration target is compared with the
truly observed image. A numerical minimization scheme can thus
determine the optimum parameter set, which can best resolve the observed
image.
6
This algorithm has been applied to the image in Figure 4.2, which was
recorded with a digital video camera from 900 metres altitude. The best
fitting parameters provide a simulation, which fits the observed intensity
values almost perfectly (Figure E.3). The computer aided analysis tells us
three things:
240
1. The GSD (pixel size) for this image (which was recorded for camera
calibration purposes during a test certification procedure) is 17.3
centimetres. The GSD can be determined to a precision
7
of 1
millimetre. The GSD is the most important parameter since it
determines the ultimately possible resolution. According to the above
definition of probable detection, we take the resolution to be
equivalent to a bar width of 1pixel = 1 GSD = 17.3 ± 0.1 centimetres.
2. The PSF halfwidth is
s = 0.48 ± 0.1 pixel, with a steepness a = 2.2
± 1, i.e., close to a Gaussian distribution. The determined PSF is
plotted in Figure E.4b. It is very close to the ideal PSF-halfwidth of
s =
1/2. We see that the MTF of this point spread function retains the
contrast quite well: we still obtain an MTF-value of 33 per cent for a
bar width of one pixel.
3. The deviation between the simulation and observation is only 3 per
cent, i.e., we have a very good signal-to-noise-ratio of SNR
» 30.
Figure E.2: The calibration target in various processing stages. From top
to bottom: the model, the target as recorded from 900 m altitude, same
data after magnification, and intensity profile.
241
It may seem surprising that the PSF of this illustrating image (900
metres altitude) can be determined to a precision of 1 millimetre. This is
made possible by fact the that the observed target intensity profile image is
made up of approximately 80 pixels; hence a misestimation of the GSD
value by 1 millimetre yields a final displacement of 8 centimetres, which is
almost half a pixel size and appears clearly in the comparison of the
simulated and the observed intensity profile.
Conclusion of Section E.1
From the previously discussed example, it is apparent that an
operational software code could be employed which would automatically
determine the image resolution from the calibration target with a very high
precision. Fitting the model of the calibration target, with known ground
dimensions, to its digital image yields the pixel size (ground sampled
distance, GSD). The GSD is the central parameter that determines the lower
boundary for the ultimately possible resolution.
Furthermore, an appropriate algorithm can automatically estimate the
point spread function (PSF, and thus the MTF) from the comparison of the
digital image of the calibration target with its model. If the width of the PSF
turns out to be much larger than 1 pixel (
s » 0.5 pixel) then the apparent
resolution of the image will be worse than expected from the GSD value,
meaning that the image is blurred. In that case, the noise level is of interest,
since for low noise it is sometimes possible to restore the image to a better
resolution (see section E.2).
Even though a computer code for automatic resolution determination
could be easily provided to the image analysts, it remains to be discussed if
automatic resolution measurement would serve the spirit of the Treaty.
Naturally, computer aided automation of resolution determination would
ease the technical procedures required by the Treaty and yield more
objective and more reproducible results. On the other hand, it might be just
the simplicity of the eye appraisal by the analysts of the observing and
observed state parties, which lends the procedure its credibility. Thus, the
automated determination of GSD and PSF should probably be
recommended as a complementary procedure, without replacing the eye
appraisal.
242
Figure E.3: Model of the calibration target profile (top), and simulated
versus observed calibration target (bottom). The best-fitting PSF function
is found by numerical optimization. The simulated profile (thin line) fits
the observed intensity profile (thick line) almost perfectly (deviation 3%).
The observed profile is extracted from Figure E.2.
243
Figure E.4: Left row: The point spread functions (PSFs) modelled according to
Equation E.1 for (a) top a near ideal box function, (b) centre one from 900 m
imagery, and (c) bottom a blurred case. Right row: The corresponding modulation
transfer functions (MTFs). The dotted line indicates the limiting case of the optimal
MTFs, the “sinc-function”. The contrast values for 0.7, 1.0 and 1.4-pixel separation
(i.e., bar width) can be determined at the positions of the light dotted vertical lines.
244
E.2 CAN THE IMAGES BE “SHARPENED”?
Since so many of the technical provisions of the Treaty are motivated
by the aim to limit the ground resolution of the images, the obvious
question arises of whether the images can be “sharpened” afterwards in
order to circumvent the Treaty limitations. This question—whether the
resolution can be pushed beyond the limitation of the original data by
means of digital image post-processing—will be discussed in this section.
Computer Aided Magnification and Enhancement
The images can of course be magnified in order to try to facilitate the
eye appraisal. Often, however, the analysts work directly on photographic
negatives, because the magnification does not necessarily improve the
analysis. Magnification of photographic film will soon expose the grain
structure of the emulsion, and magnification of digital data (video or
digitalized negatives) will only show the pixel grid structure (see Photo E.2,
top). However, using digital image processing, it is possible to resample the
image to a larger number of pixels and to interpolate between the known
grey values of the truly sampled pixels.
8
On the one hand, this certainly
seems to enhance the image quality and to ease interpretation (as
demonstrated in Photo E.2, bottom). On the other hand, it is evident that
there is no new feature to be seen in the enhanced magnification, which
was not already present in the coarser original. In other words, the
interpolation between grey values is pure speculation and does not add any
new information to the image content. It cannot disclose so far undetected
features but it can aid the human eye.
Computer Aided Image “Sharpening”(Inverse/Wiener Filtering)
In textbooks on image processing we find astonishing examples of
image restoration. A typical case is the removal of blur by inverse filtering in
the (Fourier-transformed) frequency domain (Photo E.3, top row).
Why can this not be done with Open Skies imagery on a regular basis?
Let us consider the example of Photo E.3. The original image (top left) is
digitally blurred, using a PSF with
s = 2 and a steepness exponent of a =2
(Gaussian). The PSF can be inspected in Figure E.4, bottom row. The
blurring convolution was performed by multiplication in the Fourier-
transformed space (Convolution Theorem). With the perfect knowledge of
the blurring PSF and in the absence of noise, the blurred image (top centre)
can be restored to full resolution by inverse filtering in the Fourier space.
245
However, if the blur process is combined with even very low noise (the
superimposed noise of 1 per cent is almost invisible to the eye), the simple
inverse filtering fails completely, in that it mainly amplifies the noise
(Photo E.3, second from top). Thus, in practice, Wiener filtering (or iterative
restoration) or other more sophisticated methods of image restoration have
to be employed. These techniques can restore some of the resolution, but
the results remain clearly far from the original quality—depending on the
noise level. Still, note that the four engines of the airliner are not
distinguishable in the blurred image but are clearly visible after Wiener
filtering. The crucial feature of this example is that the blur function was
considerably wider than one pixel.
Photo E.2: Example of
imagery from a
multispectral line scanner:
Trucks in approximately 1-
metre resolution (at
wavelength
l=980 nm).
Source: R.Wiemker,
University Hamburg.
Top: Original.
Bottom: The same image
after resampling and
interpolation between the
known grey values of the
truly sampled pixels.
original (126 x 107 pixels)
trucks
Nürnberg 1991 300m 980 nm CENSIS
magnified x 5 & smoothed
246
In other words, the information of each pixel was smeared into the
neighbouring pixels, and can be “re-concentrated” by digital image
processing methods. The corresponding modular transfer function
9
in
Figure E.4, bottom right, shows that the contrast vanishes for all three bar
separations (0.7, 1.0, 1.4-pixel), which could be resolved as far as the GSD
is concerned. Digital de-blurring can at best restore the contrast to the
Photo E.3: An example of inverse filtering and Wiener filtering of simulated
blur (Gaussian,
s = 2 pixel): In absence of noise (top); and with noise
(below). Note that in this simulation the blur-PSF is perfectly known. (The
original image of an airliner was recorded by an airborne multispectral
scanner from 300 m altitude with a pixel size of
» 70 cm, Nuremberg
airport 1995.) Source: R. Wiemker, University Hamburg.
civil airliner
pixel resolution ca. 70 cm
Ideal inverse filtering in the
Fourier domain works for the
noise-free case only.
In practice, iterative methods
or Wiener filtering with
experimental parameters can
restore the image only partially.
original blurred inverse filtered
+ noise
(1%)
iterative restauration
Wiener filtered
interpolated
inverse filtered
CENSIS 1997 (R.W.)
247
limiting optimal case (dashed line, the “sinc-function”). Thus the resolution
of bar pairs at 0.7, 1.0, 1.4-pixel separation could be restored.
Now we come back to the PSF, which was determined by computer
aid for the imagery of Figure 4.2. The PSF is shown in Figure E.4, centre left.
The PSF is not wider than 1 pixel, and thus it is clear that even digital image
processing means will not be able to restore the imagery to a better
resolution. When looking at the corresponding MTF, centre right, it is
apparent that the contrast of frequencies corresponding to 0.7, 1.0, 1.4-
pixel bar separation is not zero. Image processing could only enhance the
contrast (with the dotted line as the optimal limit), but no new details would
become visible.
By eye appraisal it is not easy to decide whether an image can be
sharpened or not. The safe criterion is given by the PSF and the noise level.
In practice, both are not well known, and thus attempts of image restoration
are often fruitless and rather speculative. With the ground calibration
target,
10
however, we can determine the PSF to subpixel precision, assess
the noise level, and make a well-founded statement about the possibility or
impossibility of digital image sharpening.
Degradation filters
If the flight altitude is for some reason (weather conditions,
requirements of other sensors, etc.) so low that the actually achieved
ground resolution would be better than allowed by the Treaty, then an
option is the use of degradation filters.
From the above-discussed image enhancement possibilities it becomes
clear that it is not sufficient to simply blur (optically or digitally) the image
with a blur characteristic (point spread function) that is uniform over the
entire image. From the literature it is well known that such a uniform blur
can be removed if the point spread function is known (provided a high
signal-to-noise ratio).
Hence, it has to be made certain that:
(a) for digital degradation the image is not only blurred but resampled to
a smaller number of pixels (undersampling), or
248
(b) for optical degradation filters that their influence cannot be removed,
e.g., in the Fourier-transformed frequency domain. This can be
achieved if the optical filter transparencies have local PSFs which vary
over the image area, or enough noise is induced by the filtering.
By reconstructing the actual PSF from images of the ground calibration
target, it can be determined in the same fashion as described in Section E.1
that degradation filters are indeed effective. For example, the German-
manufactured optical degradation filters, which were used on the German
Open Skies framing camera, were requested by the Russian side for
extensive laboratory testing. The irreversibility of the degradation was
confirmed.
Conclusion of section E.2
“Magnification”
Images can be magnified by some appropriate interpolation scheme.
However, this will not reveal any new details to the subjective impression of
the analyst, though it may permit a better eye appraisal.
“Sharpening”
The images can be sharpened only if:
• the PSF has considerable spread of more than 1 pixel and is spatially
constant over the entire image area;
• the PSF is known or can be determined with good accuracy;
• the noise level is low enough.
The PSF of the camera/sensor can be optimized such that it covers only
1 pixel and thus no digital image sharpening is possible. This can be verified
during the certification. The atmospheric conditions could superimpose a
wider PSF during the actual recording flight, but such an atmospherically
induced PSF will usually not be uniform across the image and thus preclude
later sharpening. In any case, it would be technically possible to reconstruct
the actual PSF from the calibration targets. If the spread of the PSF is not
much larger than 1 pixel then it is verified that the imagery cannot be
restored to better resolution later on.
249
E.3 THE TRADE-OFF BETWEEN RESOLUTION/ALTITUDE/GROUND
SWATH
The Treaty’s definition of the minimum flight altitude H
min
in function
of the actual resolution clearly implies variances in the achievable data
taking results. For example, flying on a clear day at high altitude with a fine-
grained photographic film allows for the required ground resolution and for
a large ground swath at the same time. In contrast, a cloudy day will require
a low flight altitude, and—as a result—a coarsely-grained film in order not
to violate the resolution limit, and a small ground swath. Thus, in the second
scenario, the observing party is able to obtain a much smaller amount of
data. The first scenario allows the inspection of much larger areas of the
overflown state.
With regard to photographic cameras, the Treaty allows for a ground
coverage of up to 50 kilometres on each side of the flight path.
11
In
practice, the ground swath covered by photographic cameras will be
smaller. For example, the panoramic camera KA-91C on board of the US
Open Skies aircraft covers a ground swath of 10-23 kilometres at flight
altitudes of 4.7-11.1 kilometres.
These considerations suggest that the Treaty was not designed
primarily with large-area monitoring purposes in mind. It seems to be
understood that the Treaty’s objective is not the detection of unknown
activities somewhere, but rather the recognition of specific activities and
localized objects.
12
Otherwise, the Treaty would probably have been
designed to guarantee equal territorial coverage rather than equal ground
resolution. An alternative formulation of the Treaty provisions would fix the
flight altitude such that with a given camera focal length and film format a
certain ground swath size would not be surpassed.
Notes
1
Note that the bilateral Hungarian-Romanian Open Skies Agreement
(see section 8.1 and Appendix H) works completely without either a
minimum flight altitude or a ground resolution limit.
250
2
With the discrete black and white bars of the calibration target (the
sharp contours cause high frequencies in the Fourier-transform) it is
more appropriate to talk about the contrast transfer function (CTF). See
K. Kraus, Photogrammetrie, Vol. 1—Grundlagen und
Standardverfahren, Bonn: Dümmler, 1994.
3
Ibid.
4
R. C. Gonzalez and R. A. Wintz, Digital Image Processing, Boston, MA:
Addison-Wesley, 1987.
5
As stated in section 4.4, a GSD of 30 centimetres corresponds in
practice to a photogrammetric resolution (GRD) of approximately
60 centimetres.
6
In practice, a number of other parameters have to be fitted as well,
such as the observed contrast, the exact (subpixel) location of the
target, etc. Simultaneous multi-parameter minimization is achieved by
means of a Marquardt-Levenberg gradient descent method. See, e.g.,
S. Brandt, Datenanalyse, Mannheim, Leipzig, Wien, Zürich: BI-
Wissenschaftsverlag, 1992.
7
Precision means a high statistical confidence, i.e., low statistical error
margins; whereas accuracy also implies the absence of systematic
errors (see, e.g., B. Jähne, Practical Handbook on Image Processing for
Scientific Applications, Boca Raton FL: CRC Press LLC, 1997).
8
See, e.g., B. Jähne, op. cit.
9
The modulation transfer function is related to the point spread function
in that the MTF is the absolute magnitude of the Fourier-transform of
the PSF.
10
It may sometimes be possible to determine the PSF even without
ground calibration target, but these are rather special or speculative
cases. See I. Keller, “Extraction of the Point Spread Function with
Subpixel Accuracy from Natural Scenes with Nonlinear Optimisation
and Simulated Annealing Methods”, Proceedings of the Third
International Airborne Remote Sensing Conference and Exhibition,
Copenhagen, Denmark, Environmental Research Institute of Michigan,
Ann Arbor, Vol. I, 1997, pp. 499-506.
11
Radar coverage is limited to a ground swath of 25 kilometres on one
side of the aircraft within a 50 kilometres strip.
12
Positions of locations of interest would be provided by other sources
such as CFE on-site inspections, intelligence, etc.
251
APPENDIX F
BRIEF GLOSSARY OF OPEN SKIES TREATY TERMS
This glossary was compiled by the United States Defense Threat
Reduction Agency.
1
Active Quota
The number of observation flights that each State Party has the right to
conduct as an observing Party. A Party's active quota may not exceed its
passive quota. The initial distribution of such numbers is contained in
Section II of Annex A of the Treaty and is subject to annual review in the
Open Skies Consultative Commission (OSCC).
Alternate Airfield
An airfield to which an observation aircraft or transport aircraft may proceed
when it becomes inadvisable to land at the airfield of intended landing. This
category includes emergency airfields and airfields that may be used as a
weather alternates if weather at the planned destination is unsuitable for
landing.
Danger Area
An airspace of defined dimensions, within which activities dangerous to the
flight of aircraft may exist at specified times. This is one of the three types of
areas that are covered under the term "hazardous airspace," on the basis of
the International Civil Aviation Organization (ICAO) Convention.
Escort
An individual from any State Party who accompanies the inspectors of
another State Party.
1
See www.dtra.mil/news/fact/nw_os_glsry.html (January 2003).
252
Flight Crew
Individuals who perform duties associated with the operation or servicing
of an observation aircraft or transport aircraft, including interpreters.
Flight Monitor
An individual who, on behalf of the observed Party, is on board an
observation aircraft provided by the observing Party during the observation
flight.
Flight Plan
A document elaborated on the basis of the mission plan agreed between
the observing and observed Parties that is presented to air traffic control
authorities. It must be in the format and contain the content specified by
the ICAO.
Flight Representative
An individual who, on behalf of the observing Party, is on board an
observation aircraft provided by the observed Party during an observation
flight. The procedures for designating such individuals, as well as their
privileges and immunities, are contained in Article XIII of the Treaty.
Ground Resolution
The minimum distance on the ground between two closely located objects
at which they are distinguishable as separate objects. The methodologies for
determining the capability of individual sensors to achieve a specified
ground resolution, including the minimum altitude from which such a
resolution can be achieved, were developed by the OSCC during the
Treaty’s period of provisional application.
Group of States Parties
Two or more States Parties that have agreed to form a group for purposes of
the Treaty.
Hazardous Airspace
The prohibited areas, restricted areas and danger areas, defined on the
basis of Annex 2 to the Convention on International Civil Aviation, that are
established in accordance with Annex 15 to that Convention in the interests
of flight safety, public safety and environmental protection and about which
information is provided in accordance with ICAO provisions.
253
Infrared Line-Scanning Device
A sensor capable of receiving and visualizing thermal radiation emitted in
the invisible infrared part of the electromagnetic spectrum by objects due
to their temperatures and in the absence of artificial illumination. Line-
scanning devices were selected instead of other types of infrared sensors to
avoid technology transfer problems that might arise with more sophisticated
systems.
Individual Active Quota
The number of observation flights that a State Party has the right to conduct
each year over the territory of another State Party. These numbers are
subject to annual review.
Inspector
An individual from any State Party who conducts an inspection of sensors
or observation aircraft of another State Party.
Maximum Flight Distance
The maximum distance over the territory of the observed Party from the
point at which the observation flight may commence to the point at which
that flight may terminate. The maximum flight distance includes those route
segments for intermediate stops, including refueling airfields.
Mission Plan
A document that is presented by the observing Party to the observed Party
and forms the basis for the elaboration of the flight plan. The mission plan
contains the route, profile (altitudes), order of execution and support
required to conduct the observation flight.
Mission Report
A document describing an observation flight, which is completed after the
termination of the observation flight by the observing Party and which is
signed by both the observing and observed Parties.
Observation Aircraft
An unarmed, fixed wing aircraft designated and certified to make
observation flights, registered by the relevant authorities of a State Party and
equipped with agreed sensors.
254
Observation Flight
The flight of the observation aircraft conducted by an observing Party over
the territory of an observed Party from the point of entry or Open Skies
airfield to the point of exit or Open Skies airfield, as the flight of that aircraft
is provided in the flight plan.
Observation Period
A period of time during an observation flight, specified by the observing
Party, when a particular sensor installed on the observation aircraft is
operating. There is no limit on the number or duration of such observation
periods during an observation flight. Sensors may be operated for the entire
duration of a flight, as long as the observation aircraft does not deviate from
the agreed flight path and the flight altitude is appropriate for each sensor.
Information on such sensor operating periods is required because the film
or other recording medium has to be annotated with such information.
Such information permits all Parties to know when and where data was
collected, and permits third Parties to request data taken during specified
periods.
Observed Party
The State Party or group of States Parties over whose territory an observation
flight is conducted or over whose territory an observation flight is intended
to be conducted.
Observing Party
The State Party or group of States Parties that intends to conduct an
observation flight, or conducts an observation flight, over the territory of
another State Party or group of States Parties.
Open Skies Airfield
An airfield designated by the observed Party as a point where an
observation flight may commence or terminate. An Open Skies airfield may
be a point of entry or a point of exit, or it may be a separate airfield
designated solely as an Open Skies airfield.
Open Skies Consultative Commission (OSCC)
The body established to facilitate the provisional application and the
implementation of the Open Skies regime and for handling any compliance
issues that may arise.
255
Passive Quota
The number of observation flights that each State Party is obligated to
accept as an observed Party. This number, unlike the distribution of active
quotas, is not subject to annual review by the OSCC.
Pilot in Command
The pilot on board the observation aircraft who is responsible for the
operation and safety of the observation aircraft and the execution of the
flight plan.
Point of Entry
A point designated by the observed Party for the arrival of personnel of the
observing Party on the territory of the observed Party. The observation flight
may also commence at this point if it is also designated an Open Skies
airfield.
Point of Exit
A point designated by the observed Party for the departure of personnel of
the observing Party from the territory of the observed Party. The observation
flight may also terminate at this point if it is designated an Open Skies
airfield.
Prohibited Area
An airspace of defined dimensions, above the territory of a State Party,
within which the flight of aircraft is prohibited, in accordance with specified
conditions. This is one of the three types of areas that are covered under the
term “hazardous airspace”, on the basis of the ICAO Convention.
Refueling Airfield
An airfield designated by the observed Party to be used for fueling and
functioning, operation and maintenance of the servicing of observation
aircraft and transport aircraft.
Representative
An individual who has been designated by the observing Party and who
performs activities on behalf of the observing Party during an observation
flight on an observation aircraft designated by a State Party other than the
observing Party or the observed Party.
256
Restricted Area
An airspace of defined dimensions, above the territory of a State Party,
within which the flight of aircraft is restricted in accordance with specified
conditions. This is one of the three types of areas that are covered under the
term “hazardous airspace”, on the basis of the ICAO Convention.
Sensor
Equipment of a category specified in Article IV of the Treaty that is installed
on an observation aircraft for use during the conduct of observation flights.
The categories of equipment specified in Article IV are: optical and framing
cameras; video cameras with real-time display; infrared line-scanning
devices; and sideways-looking synthetic aperture radar.
Sensor operator
An individual from any State Party who performs duties associated with the
functioning of the sensors on an observation aircraft. Such personnel could
also be considered to be representatives, flight representatives or flight
crew, depending on their duties.
Synthetic Aperture Radar (SAR)
Observation aircraft may be equipped with sideways looking synthetic
aperture radar. SAR is a high resolution ground mapping technique which
takes advantage of the forward motion of a vehicle that is carrying a pulsed
radar to synthesize the effect of a large antenna aperture. In other words,
the larger the radar antenna (aperture), the higher the radar picture's
resolution. SAR uses a radar with a very small antenna (such as can be
carried in an aircraft) that synthesizes a series of recurring radar pulse
returns to simulate the effect of a much larger antenna aperture; thus, the
name "synthetic" aperture. SAR is effective in detecting large objects day or
night, even through cloud cover.
Tax i Option
A treaty provision that gives the observed Party the right to demand that its
observation aircraft, along with its flight crew and sensors, be used to carry
out an observation flight over its territory requested by an observing Party.
Territory
The land, including islands, and internal and territorial waters, over which
a State Party exercises sovereignty.
257
Tot al Pass ive Quota
The total number of observation flights that a State Party is obliged to accept
over its territory. Passive quotas are allocated by Annex A, Section II of the
Trea ty.
Transit Flight
The flight of an observation aircraft or transport aircraft over the territory of
a third State Party en route to or from the territory of the observed Party.
Transport Aircraft
An aircraft other than an observation aircraft that conducts flights to or from
the territory of the observed Party exclusively for the purposes of the treaty.
258
259
APPENDIX G
OPEN SKIES BASIC ELEMENTS
NATO Communiqué 14-15 December 1989
NATO Office of Information and Press
1110 Brussels
Annex to the Communiqué of the North Atlantic Council meeting
in Ministerial Session on 14
th
and 15
th
December 1989
‘OPEN SKIES’
BASIC ELEMENTS
I. INTRODUCTION
1. On 12
th
May 1989, President Bush proposed the creation of a so-
called ‘Open Skies’ regime, in which the participants would voluntarily
open their airspace on a reciprocal basis, permitting the overflight of their
territory in order to strengthen confidence and transparency with respect to
their military activities.
This proposal expanded on a concept that had already been proposed
during the 1950s but had failed to reach fruition because of the
unfavourable international political climate prevailing at the time.
Today, this new initiative has been made in a very different context as
openness becomes a central theme of East-West relations and the past few
years have been marked by important advances in the areas of confidence-
building and arms control.
2. The provisions for notification and observation of military activities
specified in the Helsinki Final Act were strengthened and made obligatory
by the Stockholm Document concluded by the CDE in 1986.
With respect to arms control, in 1987, the INF Treaty, apart from is
immediate goals, represented a very important precedent because of the
extent of its verification provisions.
260
All this leads one to expect today even more spectacular advances will
be achieved in the near future. In particular, a two-pronged effort is under
way in Vienna: on the one hand, to deepen the measures for confidence-
building and transparency among the 35 countries of the CSCE, and on the
other, to reach an unprecedented agreement between the countries of the
Atantic Alliance and the Warsaw Treaty Organization on the elimination of
large numbers of conventional arms.
Furthermore, one awaits important developments in other sectors of
disarmament such as chemical weapons and the Soviet-American strategic
arms negotiations.
3. All of these agreements will naturally require their own verification
regimes, often of a highly intrusive nature. Moreover, the specific provisions
of each verification treaty will be supplemented by the habitual means by
which countries verify compliance with agreements (national technical
means).
It seems useful, however, particularly in the prevailing context of
improved East-West relations, to reflect on other ways of creating a broadly
favourable context for confidence-building and disarmament efforts.
In this context, the Open Skies concept has a very special value. The
willingness of a country to be overflown is, in itself, a highly significant
political act in that it demonstrates its availability to openness; aerial
inspection also represents a particularly effective means of verification,
along with the general transparency in military activities discussed above.
This double characteristic of an Open Skies regime should make it a
valuable complement to current East-West endeavours, mainly in the
context of the Vienna negotiations but also in relation to the other
disarmament efforts (START, chemical weapons).
It would seem desirable to focus now on the European region, while
also including the entire territories of the Soviet Union, the United States,
and Canada. Accordingly, we will be ready to consider at an appropriate
time the wish of any other European country to participate in the Open
Skies regime. This element could be complementary to their efforts at
confidence-building and conventional arms control and would conform to
the objectives of those negotiations.
4. To this end, the Open Skies Regime should be based on the following
guidelines:
- The commitment of the parties to greater transparency through aerial
overflights of their entire national territory, in principle without other
261
limitations than those imposed by flight safety or rules of international
law.
- The possibility for the participants to carry out such observation flights
on a national basis or jointly with their allies.
- The commitment of all parties to conduct and to receive such
observation flights on the basis of national quotas.
- The establishment of agreed procedures designed to ensure both
transparency and flight safety.
- The possibility for the parties to employ the result of such overflights to
improve openness and transparency of military activities as well as
ensuring compliance with current or future arms control measures.
Il. PURPOSE
The basic purpose of Open Skies is to encourage reciprocal openness
on the part of the participating states and to allow the observation of military
activities and installations on their territories, thus enhancing confidence
and security. Open Skies can serve these ends as a complement both to
national technical means of data collection and to information exchange
and verification arrangements established by current and future arms
control agreements.
Ill. PARTICIPATION AND SCOPE
Participation in Open Skies is initially open to all members of the
Atlantic Alliance and the Warsaw Treaty Organization. All territories of the
participants in North America and Asia, as well as in Europe, will be
included.
IV. QUOTAS
1. Open Skies ‘accounting’ will be based on quotas which limit the
number of overflights. The quotas will be derived from the geographic size
of the participating countries. The duration of flights can also be limited in
relation to geographic size. For larger countries, the quota should permit
several flights a month over their territory. All of the parties will be entitled
to participate in such observation flights on a national basis, either
individually or jointly in co-operation with their allies.
262
2. Effective implementation of a quota system requires agreement that a
country will not undertake flights over the territory of any other country
belonging to the same alliance.
3. Quota totals for participating states should be established in such a
manner that there is a rough correspondence between totals for NATO and
the Warsaw Treaty Organization and, within that total, for the USSR and the
North American members of NATO.
4. Every participant, regardless of size, would be obligated to accept a
quota of at least one overflight per quarter.
5. Smaller nations, that is, those subject to the minimum quota, may
group themselves into one unit for the purposes of hosting Open Skies
overflights and jointly accept the quota that would apply to the total land
mass of the larger unit.
V. AIRCRAFT
The country or countries conducting an observation flight would use
unarmed, fixed-wing civilian or military aircraft capable of carrying host
country observers.
VI. SENSORS
A wide variety of sensors would be allowed, with one significant
limitation—devices used for the collection and recording of signals
intelligence would be prohibited. A list of prohibited categories and types
of sensors will be agreed among the participating states which will be
updated every year.
VII. TECHNICAL CO-OPERATION AMONG ALLIES
Multilateral or bilateral arrangements concerning the sharing of aircraft
or sensors, as well as the conduct of joint overflights, will be possible among
members of the same alliance.
VIII. MISSION OPERATION
1. Aircraft will begin observation flights from agreed, pre-designated
points of entry and terminate at pre-designated points of exit; such entry
263
and exit points for each participating state will be designated by that state
and listed in an annex to the agreement.
2. The host country will make available the kind of support equipment,
servicing and facilities normally provided to commercial air carriers.
Provision will be made for refuelling stops during the overflight.
3. An observing state will provide 16 hours notification of arrival at a point
of entry. However, if the point of entry is on a coast or at a border and no
territory of the receiving state will be overflown prior to arrival at the point
of entry, this pre-arrival period could be abbreviated.
4. The crew of the observation aircraft shall file a flight plan within six
hours of its arrival at the point of entry.
5. After arrival and the filing of a flight plan, a 24 hour pre-flight period
will begin. This period is to allow time to determine that there are no flight
safety problems associated with the planned flight route and to provide
necessary servicing for the aircraft. During this pre-flight period the aircraft
will also be subject to intrusive but non-destructive inspection to prohibited
sensors and recorders.
6. Prior to the flight, host-country monitors will be able to board the
observation aircraft. During the flight they would ensure that the aircraft is
operated in accordance with the flight plan and would monitor operation
of the sensors. There would be no restrictions on the movement of the
monitors within the aircraft during flight.
7. The flight will be from the agreed point of entry to an agreed point of
exit, where the host country observers would depart the aircraft. The points
of entry and exit could be the same. Loitering over a single location will not
be permitted. Aircraft will not be limited to commercial air corridors.
Observation aircraft may in principle only be prohibited from flying through
airspace that is publicly announced as closed to other aircraft for valid air
safety reasons. Such reasons would include specific hazards posing extreme
danger to the aircraft and its occupants. Each country will make
arrangements to ensure that public announcements of such hazardous
airspace are widely and promptly disseminated; each country will produce
for an annex to the agreement a list of where these public announcements
can be found. The minimum altitudes for such flights may vary depending
upon air safety considerations. The extent of ground control over aircraft
will be determined in advance by agreement among the parties on
compatible rules such as those recognized by ICAO. In the application of
264
these considerations and procedures, the presumption shall be on behalf of
encouraging the greatest degree of openness consistent with air safety.
8. The operation of the Open Skies regime will be without prejudice to
states not participating in it.
IX. MISSION RESULTS
The members of the same alliance will determine among themselves
how information acquired through Open Skies is to be shared. Each party
may decide how it wishes to use this information.
X. TRANSITS
A transit flight over a participating state on the way to the participating
state over which an observation flight is to be conducted shall not be
counted against the quota of the transited state, provided the transit flight is
conducted exclusively within civilian flight corridors.
XI. TYPE OF AGREEMENT
The Open Skies regime will be established through a multilateral treaty
among the parties.
XII. OPEN SKIES CONSULTATIVE BODY
To promote the objectives and implementation of the Open Skies
regime, the participating states will establish a body to resolve questions of
compliance with the terms of the treaty and to agree upon such measures
as may be necessary to improve the effectiveness of the regime.
265
APPENDIX H
THE HUNGARIAN-ROMANIAN OPEN SKIES AGREEMENT
Agreement Between the Government of the Republic of Hungary
and the Government of Romania on the Establishment of
an Open Skies Regime
1
The Government of the Republic of Hungary and the Government of
Romania, hereinafter referred to as the Parties;
Recalling their commitments in the Conference on Security and Co-
operation in Europe to promoting greater openness and transparency of
their military activities and to enhancing security by means of confidence
and security building measures;
Seeking to implement in their bilateral relations in addition to the
provisions of the 1990 Vienna Document of the Negotiations on
Confidence- and Security-Building Measures, further cooperative
confidence and security building measures;
Reaffirming their desire to further contribute to the successful
conclusion of the negotiations of the Open Skies Conference, as expressed
in the Charter of Paris for a New Europe;
Convinced that a successful bilateral Open Skies regime provides
valuable experience for the elaboration of an Open Skies Treaty, and the
simultaneous functioning of the regimes will lead to enhanced confidence
and security;
Noting that an Open Skies regime and its successful implementation
would encourage reciprocal openness on the part of the States Parties,
enhance the predictability of their military activities and strengthen
confidence between them;
1
The agreement was signed on 11 May 1991 at Bucharest. It was submitted to
the General Assembly/Security Council of the United Nations as document A/
46/188, S/22638, 24 May 1991.
266
Convinced that the Open Skies regime will be implemented on a
reciprocal and equitable basis which will protect the interest of each State
Party;
Noting the possibility of employing the results of such overflights to
improve openness and transparency, to enhance confidence and security
building, and to improve the monitoring of, and thus promote compliance
with, current or future arms control measures;
Noting that the operation of an Open Skies regime will be without
prejudice to States not parties to this Agreement;
Believing that an effective Open Skies regime would serve to
consolidate improved good neighbourly relations between the States
Parties.
Have agreed as follows:
Article I. Definitions
For the purposes of this Agreement and its Annexes:
(1) The Term “Aircrew Member” means an individual from any of the two
Parties who has been designated and accepted in accordance with Article
XIX of this Agreement, and who performs duties associated with the
operation or maintenance of the Observation Aircraft or its sensors, and
participates as a member of the aircrew of the Observation Aircraft during
the Observation Flight, or who is an Inspector Escort.
(2) The term “Observation Crew Member” means an individual from the
Observing Party who has been designated and accepted in accordance with
Article XIX of this Agreement, and who performs duties associated with the
operation of the sensors of the Observation Aircraft of the Observed Party
and who participates as an Aircrew Member of the Observation Aircraft of
the Observed Party during the Observation Flight.
(3) The term “Flight Monitor” means an individual designated by the
Observed Party to be on board the observation Aircraft during the
Observation Flight and who performs duties in accordance with Annex D.
(4) The term “Flight Plan” means a flight plan of the Observing Party
meeting the requirements of Article VI.
267
(5) The term “Hazardous Airspace” means areas of an Observed Party in
which there are invisible or unusual dangers to the safety of the aircraft.
Hazardous airspace include prohibited areas, restricted areas and danger
areas, established in the interest of flight safety, public safety and
environmental protection and published by the Observed Party in
accordance with ICAO rules in the Aeronautical Information Publication
(AIP).
(6) The term “Inspector” means an individual who is designated by the
Observed Party or Observing Party to conduct inspections of the
Observation Aircraft, its equipment, its sensors in accordance with Article
IX and Annex C.
(7) The term “Inspection Team” means the group of Inspectors designated
by the Observed Party or Observing Party to conduct the inspection of the
Observation Aircraft, its equipment and its sensors in accordance with
Article IX. and Annex C.
(8) The term “Inspector Escort” means a designated representative of the
Observing Party or the Observed Party who has been authorized to monitor
all activities of Inspectors and Inspection Team during inspections and
perform other specified duties in accordance whit Article IX and Annex C.
(9) The term “Inspection” means activity described in and performed
pursuant to Article IX and Annex C.
(10) The term “Period of Inspection” means the period of time during which
the Inspection Team inspects the Observation Aircraft, its equipment and its
sensors in accordance with Article IX and Annex C.
(11) The term “Observation Aircraft” means an unarmed, fixed wing
aircraft, capable of carrying two Observed Party Flight Monitors in addition
to its Aircrew Members. An aircraft is considered unarmed when it is not
carrying any armament (munitions) of any type or equipment dedicated to
armament operations.
(12) The term “Observation Flight” means a flight and any accompanying
refuelling stops, conducted in accordance with the provisions and
restrictions of this Agreement by an Observation Aircraft over the Territory
of an Observed Party.
(13) The term “Observed Party” means a Party over whose Territory an
Observation Flight is conducted.
268
(14) The term “Observing Party” means a Party conducting an Observation
Flight.
(15) The term “Point of Entry” means the Airfield (s) in the Territory of each
party that are designated in Annex B for the arrival of the Observation
Aircraft at the Observed Party s Territory.
(16) The term “Point of Exit” means the Airfields(s) in the Territory of each
Party that are designated in Annex B for the departure of the Observation
Aircraft from the Observed Party’s Territory.
(17) The term “Permitted Observation Equipment” means on-board
observation equipment of the Observation Aircraft as described in Annex E.
(18) The term “Quota” means the number of Observation Flights that each
Party undertakes to accept annually (“Passive Quota”) and also the number
of Observation Flights each Party shall have the right to conduct annually
(“Active Quota”), as set forth in Annex A.
(19) The term “Arrival Fix” means the compulsory reporting point specified
and promulgated by the Observed Party in Annex B through which the
Observation Aircraft shall enter the territorial airspace of the Observed
Party.
(20) The term “Departure Fix” means the compulsory reporting point
specified and promulgated by the Observed party in Annex B through
which the Observation Aircraft shall depart the territorial airspace of the
Observed Party.
(21) The “ATS” Route means a specified route designed for channelling the
flow of traffic as necessary for the provisions of Air Traffic Services.
Article II. Basic Rights and Obligations of the Parties
(1) Each Party shall have the right to conduct Observation Flights in
accordance with the provisions of this Agreement.
(2) Each Party undertakes to permit Observation Flights over its Territory in
accordance with the provisions of this Agreement.
(3) Each Party may conduct Observation Flights with its own Observation
Aircraft or the Observation Aircraft of the other Party.
(4) Areas with Hazardous Airspace are excepted in accordance with the
provisions of Articles I, VIII and Annex G.
269
Article III. Quotas of Observation Flights
(1) For the purposes of fulfilling objectives of this Agreement, each Party
shall have the right to conduct and undertakes the obligation to accept an
agreed number of Observation Flights in accordance with Annex A.
(2) The number of Observation Flights a Party shall be allowed to conduct
shall be equal to the number of overflights it shall be required to accept.
Article IV. Observation Aircraft
While conducting flights under this Agreement the Observation
Aircraft shall comply with the provisions of this Agreement.
Unless inconsistent with the provisions of this Agreement, the
Observation Aircraft shall also comply with:
(a) the published standards and recommended practice of ICAO;
(b) published national air traffic control rules, procedures and
guidelines on flight safety of the Observed Party;
(c) the instructions of the ATC authorities and the ground control
services.
Article V. Pre- and Post-Observation Flight Procedures
(1) Upon entry into force of this Agreement, each Party shall provide the
other Party with the following information:
(a) emergency airfields between its Arrival Fixes and Points of Entry and
between its Points of Exit and its Departure Fixes;
(b) Instrument arrival and departure procedures:
- for its Points of Entry and Exit;
- for its alternate airfields near its Points of Entry and Exit;
- for suitable airfields along route of flight which may be used in
an emergency.
(2) Each Party shall promptly notify the other Party of any updates and
amendments to such information.
(3) A Party may change the location of its Points of Entry and/or Exit upon
three months prior notification to the other Party.
270
(4) In order to conduct an Observation Flight, the Observing Party shall
notify the Observed Party of the estimated time of arrival of its Observation
Aircraft at the Observed Party’s Point of Entry. Such notice shall be given less
than 24 hours in advance of the estimated time arrival.
(5) The notification to the Observed Party shall also indicate the type and
model of the incoming aircraft, its registration number and call sign, as well
as the names, passport types and numbers and functions of each Aircrew
Member.
(6) In case the Observing Party intends to use the Observation Aircraft of
the Observed Party, it shall submit its request to do so 7 days in advance of
the proposed time of the commencement of the Observation Flight.
(7) Upon completion of the Observation Flight, the Observation Aircraft
shall depart the Territory of the Observed Party from the Point of Exit. The
departure flight from the Point of Exit shall commence not later than 24
hours following the completion of the Observation Flight, unless weather
conditions or the airworthiness of the Observation Aircraft do not permit.
Article VI. Flight Plans and Conduct of Observation Flights
(1) Within six hours following the arrival of the Observation Aircraft or the
Observation Crew at the Point of Entry, the Observing Party shall submit a
Flight Plan for the Proposed Observation Flight to the Observed Party. The
Observed Party shall as soon as possible review and approve or amend and
approve the proposed Flight Plan in accordance with the provisions of this
Agreement.
(2) The Observation Flight shall be conducted in accordance with the
approved Flight Plan and in accordance with clearances and instructions
from the Observed Party’s air traffic controllers.
(3) The Flight Plan shall have the content according to Annex 2 to the
Convention on International Civil Aviation, signed in Chicago, 1944, and be
in the format specified by ICAO Document 4444-RAC/501, Rules of the Air
and Air Traffic Services, as amended or revised.
(4) The Flight Plan shall provide and require that:
(a) the planned duration of the Observation Flight shall not exceed the
duration of Observation Flights that is set forth in Annex A;
271
(b) the Observation Flight commences not earlier than 16 hours and
not later than 48 hours after delivery of the Flight Plan to the Observed
Party;
(c) the observation Aircraft shall fly a direct route between the
coordinates or navigation fixes designated in the Flight Plan, and shall visit
each coordinate or navigation fix in the declared sequence set forth in the
Flight Plan; and
(d) the Observation Aircraft shall not hold over, delay departure from
or otherwise loiter at any point its approved Flight Plan route nor otherwise
unreasonably disrupt the normal flow of air traffic except:
- as allowed for in the approved flight plan;
- as necessary for the purposes of arrival or departure at
designated airfields when executing published procedures or
the instructions of air traffic control;
- as instructed by air traffic control;
- as required for reasons of flight safety;
- flight tracks shall be permitted to intersect provided that no
point of intersection is crossed more than once on any
Observation Flight.
(5) The Observed Party shall ensure that Aircrew Members are given the
Observed Party’s most recent weather and safety information pertaining to
the Flight Plan for each Observation Flight, including Notices to Airmen, IFR
procedures and information about alternate and emergency airfields along
the flight route stated in the approved Flight Plan.
(6) All Observation Flights shall be carried out in compliance with the
provisions of this Agreement and ICAO standards and recommended
practice, and with due regard for differences existing in national rules and
regulations, published in AIP or in accordance with national flight and air
traffic control requirements of which the Observation Aircraft’s Aircrew
shall be informed.
(7) In the event that the Observation Aircraft makes a deviation from the
Flight Plan, as permitted under Article XIII of this Agreement, the additional
flight time arising from such deviation shall not count against the duration
specified in Annex A.
272
Article VII. Sensors
(1) Each Party may use during Observation Flights any sensor that is
necessary for reaching the objectives of this Agreement listed in Annex E.
Sensors not listed in Annex E are prohibited and shall not be on board of
the Observation Aircraft.
(2) The Parties undertake to use the same types of sensors of comparable
capability and to this end to facilitate access to such sensors for use by the
other Party.
(3) The Observation Aircraft shall be equipped with the same sensors,
when used upon request by the other Party.
(4) Data acquired by sensors during Observation Flights will remain
encapsulated on board the Observation Aircraft until the termination of the
Observation Flight. Sensor data link operations of any kind are prohibited.
(5) As provided in Paragraph 4 of Article XVI of this Agreement, a Party
may utilize a type or model of sensor not listed in Annex E for or in
connection with an observation Flight upon:
(a) receiving the approval of Hungarian-Romanian Open Skies
Consultative Commission (HROSCC), and;
(b) making a representative type or model of such sensor available for
pre–flight examination by the other Party in accordance with the provision
of Annex E.
(6) Any Party operating an Observation Aircraft will ensure that the sensors
function to specifications and also that their specifications conform with
agreed requirements.
Article VIII. Hazardous Airspace
(1) Observation Aircraft may conduct Observation Flights anywhere over
the Territory of the Observed Party in accordance with Article II and
Article VI.
(2) Hazardous Airspace must be publicly announced. Such public
announcements must specify the dangers to the Observation Aircraft and
Aircrew Members. Each Party shall ensure that such public announcements
of Hazardous Airspace are promptly provided to the Other Party by the
source designated by the Party in Annex H.
273
(3) Particular Hazardous Airspace announced in Annex H must be taken
into account by the Observing Party when preparing an Observation Flight
Plan.
(4) Each Party may introduce amendments and additions to Annex H,
giving notice thereof to the other Party.
(5) In case of need, the Observed Party shall inform the Aircrew Members
during preparations for the Observation Flight of the new particular
Hazardous Airspace, indicating the causes for the restrictions introduced.
(6) In the event that the Flight Plan of the Observing Party requests
overflight of Hazardous Airspace of the Observed Party, the Observed Party
shall approve the Flight Plan if it conforms with Article VI, but may amend
it to specify the minimum safe altitude over the Hazardous Airspace. This
minimum safe altitude shall be made part of the Flight Plan. If there is no
minimum safe altitude available consistent with air safety requirements, the
Observed Party shall propose an alternative flight routing as near to the
Hazardous Airspace as is permitted by air safety requirements. Alternatively
the Observed Party may propose that the time of arrival of the Observation
Aircraft over the Hazardous Airspace be amended to a time consistent with
flight safety requirements. Such alternative flight routing or timing shall be
incorporated in a revised Flight Plan and approved the Observed Party.
(7) The Observing Party may elect either to conduct Observation Flight on
the basis of an amended Flight Plan, avoiding the particular Hazardous
Airspace, or to cancel the Observation Flight. In that latter event, the
Observation Aircraft or the Observation Crew shall depart the Territory of
the Observed Party in accordance with Article V and no overflight shall be
recorded against the Quota of either Party.
(8) In the event the Observing Party informs the Observed Party that denial
of access to any portion of the Hazardous Airspace of the observed Party
was not justified on the basis of air safety considerations and in a further
event that the matter is not resolved through diplomatic channels, the
Observing Party may raise the matter for consideration in the Hungarian-
Romanian Open Skies Consultative Commission pursuant to Article XVI of
this Agreement.
Article IX. Aircraft and Sensors Inspections
When an Observation Flight is conducted using an Observation
Aircraft of the Observing Party, upon delivery of the Flight Plan, unless
274
otherwise mutually agreed to by the Observed and the Observing Party, the
Inspection Team of the Observed Party may inspect the Observation
Aircraft, accompanied by Inspector Escorts of the Observing Party, to
determine whether there is any Prohibited Equipment on the Observation
Aircraft. Such inspection shall terminate no later than three hours prior to
the scheduled commencement of the Observation Flight set forth in the
Flight Plan. All such inspections shall be conducted in accordance with
Annex C.
Article X. Flight Monitors on Observation Aircraft
The Observed Party shall have the right to have two Flight Monitors on
board the Observation Aircraft during each Observation Flight in
accordance with Annex D. Such Flight Monitors shall have the right of
access to all areas of the Observation Aircraft during the Observation Flight.
Flight Monitors have the rights and obligations specified in Annex D. In
discharging their functions, Flight Monitors shall not interfere with the
activities of the Aircrew Members.
Article XI. Observation Aircraft Servicing and Maintenance
(1) The Observed Party shall, upon request, provide
(a) customary commercial aircraft fuelling, servicing, and maintenance
for the Observation Aircraft at the Point of Entry or Exit and at any
predesignated refuelling point specified in the Flight Plan; and
(b) meals and the use of rest facilities for Observation Aircraft Aircrew
Members.
(2) On request of the Observing Party, further services will be agreed upon
between the Parties in order to guarantee the effective realisation of the
Observation Flight. Should unscheduled technical demand arise for the
Observation Aircraft, the necessary support will be provided without delay
by the Observed Party. A protocol about the obtained services will be
established between the Inspector Escort of the Observing Party and a
responsible officer of the Observed Party at the Point of Entry or Exit.
(3) The Observing Party shall reimburse the Observed Party for the
ordinary and reasonable costs of such fuelling, maintenance, servicing,
meals and use of rest facilities. The amount of reimbursement will be agreed
upon by the Parties on a case-by–case basis and will represent a fair
275
estimate of the cost of such services at the time rendered, exclusive of taxes,
fees, duties or other similar charges.
(4) The Observing Party shall reimburse the Observed Party for the use of
the Observation Aircraft of the Observed Party. The Observed Party shall
inform in advance the Observing Party of estimated cost of one flight hour
by the Observation Aircraft.
(5) Such charges shall not be greater than that which the Observed Party
would charge itself for the same service.
Article XII. Prohibition, Correction or Curtailment of Observation
Flights
(1) The Observed Party, by notifying the Observing Party, may prohibit
prior to its commencement, or correct or curtail in a non-harmful manner
subsequent to its commencement, any Observation Flight:
(a) that is not permitted by the terms of Annex A;
(b) for which a Flight Plan has not been filed in accordance with this
Agreement;
(c) that arrives at the Point of Entry less than 24 hours after the
notification required by Article V. of this Agreement;
(d) that fail to arrive at the Point of Entry within 6 hours the estimated
time of arrival set forth in said notification;
(e) that deviates from the Flight Plan, except as permitted by Article XIII
of this Agreement;
(f) that is conducted by aircraft other than an Observation Aircraft; or
(g) that is otherwise in non-compliance with the terms, conditions,
provisions and restrictions of this Agreement.
(2) The Observed Party may correct or curtail in its territorial airspace a
flight to a Point of Entry or from a Point of Exit that deviates from the direct
route required by Article VI.
(3) When an Observed Party prohibits, corrects or curtails an Observation
Flight in accordance with this Article, it must provide in writing to the
Observing Party through routine diplomatic channels an explanation for its
action.
276
(4) An Observation Flight that has been prohibited shall not be recorded
against the Quota of the Observed Party. A proposed Observation Flight
that has been corrected or curtailed shall not be recorded against the Quota
of the Observed Party.
(5) Disputes bearing on this Article may be submitted to the Hungarian-
Romanian Open Skies Consultative Commission for resolution as stipulated
in Article XVI of this Agreement.
Article XIII. Deviations and Emergencies
(1) Notwithstanding any other provisions of this Agreement deviations by
an Observation Aircraft from a Flight Plan or from the routes to and from
the Points of Entry and Exit, that are necessitated by: (a) adverse weather
conditions, (b) air traffic control instructions related to flight safety, or (c)
aircraft mechanical difficulty or other event beyond the control of the
Observing party, shall not be deemed a violation of this Agreement and shall
not be grounds for correction, curtailment or prohibition by the Observed
Party of an Observation Flight, a flight arriving at a Point of Entry or a flight
departing from a Point of Exit.
(2) Any Observation Aircraft declaring an emergency shall be accorded
the Observed Party’s full range of distress and diversion facilities in order to
ensure the most expeditious recovery to the nearest suitable airfield. A full
investigation of the declaration shall be conducted in accordance with the
regulations of the Observed Party, with the participation of the Observing
Party, at a place of the Observed Party’s choosing.
(3) In the case of an accident involving the Observation Aircraft in the
Territory of the Observed Party, search and rescue operations will be
conducted by the Observed Party in accordance with its own regulations
and procedures for such operations. A full investigation of the accident by
the Observed Party shall be conducted in accordance with the regulations
of the Observed Party, with the participation of the Observing Party at a
place of the Observed Party’s choosing. At the conclusion of the
investigation, all wreckage and debris of the Observation Aircraft,
equipment, and sensors if found and recovered will be returned to the
Observing Party if so requested.
277
Article XIV. Non-Interference
No Party shall use any device or equipment to interfere with the
operation of the Observation Aircraft, with the functioning of the sensors,
or with the safe conduct of any Observation Flight.
Article XV. Use of Information
(1) Information acquired through Observation Flights shall be used
exclusively for the attainment of the purpose of this Treaty.
(2) Both the Observing and the Observed Parties shall receive complete
set of the data obtained as a result of processing of observation materials.
(3) Observation materials obtained as a result of an Observation Flight
shall be processed in accordance with Annex H.
(4) Information obtained by a Party as a result of Observation Flights must
not be used to the detriment of the other Party’s security or other interests
and must transferred to any third State.
Article XVI. Hungarian-Romanian Open Skies Consultative Commission
(1) To promote the objectives and implementation of the provisions of this
Agreement, the Parties hereby establish the Hungarian-Romanian Open
Skies Consultative Commission (hereinafter referred to as “the
Commission“).
(2) The Commission shall make decisions and undertake actions on the
basis of agreement of the Parties.
(3) Each Party may raise before the Commission any issues concerning
compliance with the obligations of this Agreement.
(4) The Parties shall meet within the framework of the Commission to:
(a) agree upon such technical and administrative measures, consistent
with this Agreement, as may be necessary to ensure the viability and
effectiveness of this Agreement;
(b) consider questions relating to compliance with the obligations
assumed under this Agreement;
(c) agree on updates to the Annexes that so provide; and
278
(d) consider and act upon all matters referred to it by a Party pursuant
to this Agreement.
(5) General provisions for the operation of the Commission are set forth in
Annex F.
Article XVII. Notifications
Except as otherwise stipulated, the Parties shall provide the
notifications required by this Treaty through diplomatic channels.
Article XVIII. Liability
A Party shall, in accordance with international law and practice, be
liable to pay compensation for damage to the other Party, or to its natural
or juridical persons or their property, caused by it in the course of the
implementation of this Agreement.
Article XIX. Aircrew Members and Inspection Crew Members
(1) Aircrew Members and Inspection Crew Members shall be designated
by each Party in the following manner:
(a) Within 30 days after signature of this Agreement each Party shall
provide to the other Party for its review a list of proposed Aircrew Members
and Inspection Crew Members who will conduct Observation Flights for
that Party. This list shall not exceed 30 persons and shall contain the name,
birth date, rank, function and passport type for each person on the list. Each
Party shall have the right to amend its list of Aircrew Members and
Inspection Crew Members. Each Party shall have to provide to the other
Party its amended list of Aircrew Members and Inspection Crew Members.
(b) If any person on the original or amended list is unacceptable to the
other Party, it shall, within 14 days, notify the Party providing the list that
such persons will not be accepted as Aircrew Members and Inspection
Crew Members. Persons not declared unacceptable within 14 days are
deemed accepted as Aircrew Members and Inspection Crew Members. In
the event that a Party subsequently determines that an Aircrew Member or
an Inspection Crew Member is unacceptable, the Party shall so notify the
Party that designated the Aircrew Member or Inspection Crew Member,
which shall, not later than two working days thereafter, strike such person
from its Aircrew Member and Inspection Crew Member list.
279
(2) In order to exercise their functions effectively, for the purpose of
implementing the Agreement, Aircrew Members and Inspection Crew
Members shall be accorded the inviolability and immunities as specified in
Articles 29, 30, paragraph 2 with respect to papers and correspondence
and 31 of the Convention on Diplomatic Relations done in Vienna on 18
April 1961. Such inviolability and immunities shall be accorded for the
entire period from the arrival of the Aircrew Members or Inspection Crew
Members to the Territory of the Observed Party until their departure from
it, and thereafter with respect to acts previously performed in the exercise
of their official functions as Aircrew Members or Inspection Crew Members.
The immunity from jurisdiction may be waived by the Observing Party in
those cases when it is of the opinion that immunity would impede the
course of justice and that it can be waived without prejudice to the
Agreement. Such waiver must always be express. Without prejudice to their
inviolability and immunities or to the rights of the Observing Party under this
Agreement, it is the duty of Aircrew Members and Inspection Crew
Members to respect the laws and regulations of the Observed Party.
(3) Aircrew Members and Inspection Crew Members of a Party shall be
permitted to bring into the Territory of the Observed Party, without payment
of any customs duties or related charges, articles for their personal use, with
the exception of articles the import or export of which is prohibited by law
or controlled by quarantine regulations.
(4) In the event that either the Observing Party or the Observed Party
considers that there has been a violation or an abuse of the inviolability or
immunities accorded under this Article, that Party may forward a report
specifying the nature of the issue to the Commission for consideration.
Article XX. Ratification, Entry into Force
(1) The present Agreement is subject to ratification in accordance with
constitutional procedures of each Party.
(2) This Agreement shall enter into force upon the exchange of the
instruments of ratification.
Article XXI. Amendments; Implementing Measures; periodic Review
(1) Each Party may propose amendments to this Agreement. Agreed
amendments shall enter into force in accordance with the procedures set
forth in Article XX governing the entry into force of this Agreement.
280
(2) Any decision taken by the Commission pursuant to subparagraphs (a)
or (c) of Paragraph 4 of Article XVI shall be deemed not to be amendments
to this Agreement.
(3) Within 60 days of the signature of a multilateral Open Skies Treaty a
session of the Commission is to be convened to consider matters related to
the further implementation of this Agreement.
Article XXII. Duration; Denunciation
(1) This Agreement shall be of unlimited duration.
(2) Each Party may denunciate this Agreement if it decides that
extraordinary events related to the subject matter of this Agreement have
jeopardised its supreme interests. A Party intending to denunciate the
Agreement shall give notice of its decision to the other Party at least six
months in advance of its denunciation.
(3) In the event that a Party gives notice of its decision to denunciate this
Agreement in accordance with paragraph 2 of this Article, a meeting of the
Commission shall be convened by the Parties within 30 days after such a
notification has been received in order to consider practical matters related
to the denunciation of the Agreement.
Article XXIII. Registration
This Agreement shall be registered pursuant to Article 102 of the
Charter of the United Nations.
Article XXIV
This Agreement contains XXIV Articles and Annexes A-H, all of which
form an integral part of this Agreement.
Done at ……………………….., this …………………… day of
……………………. 19……., in two copies, each in Hungarian and
Romanian languages, all two texts being equally authentic.
281
Annex A
This Annex may be updated by the Commission. This update shall not
be considered an amendment of the Agreement.
Annex B
This Annex may be updated by the Commission. This update shall be
considered amendment of the Agreement.
Annex C – Inspections
The following procedures shall govern the inspection of the
Observation Aircraft by the Inspection Team conducted to determine
whether there is any prohibited equipment on the Observation Aircraft
pursuant to Article IX of the Agreement.
Hungary Romania
Number of Observation Flights per Year
44
Maximum Length of Observation
Flights
3 hours 3 hours
Maximum Distance of Observation
Flights
1200 km 1200 km
Hungary Romania
Points of Entry and Exit: Budapest-Ferihgy
Szolnok
Bucharest-Otopeni
Timisoara
Arrival and Departure Fixes:
All Arrival and Departure Fixes along the
Hungarian-Romanian border published
in the AIP.
Air Routes to and from Points
of Entry and Exit:
The International airways.
Language to be used during
briefings:
Hungarian Romanian
282
(1) Upon arrival of the Observation Aircraft at the Point of Entry, the
Inspection Team shall, if requested by the Inspector Escorts, provide the
Inspector Escorts a briefing on how the Inspection Team intends to inspect
the Observation Aircraft, including, but not limited to, any safety
precautions pertaining to Inspection Team activities, and shall undertake
the following measures:
(a) deliver to the Inspector Escorts a list of members of the Inspection
Team, which shall not exceed 10 members, unless otherwise agreed to by
the Observing Party and the Observed Party, and a statement of the general
function during the inspection of each member of the Inspection Team; and
(b) deliver to the Inspector Escorts a list of each item of inspection
equipment to be used by the Inspection Team in conducting the inspection,
which shall be limited to the following items:
(i) flashlights;
(ii) still and video cameras;
(iii) notepads, inspection records, rulers, pens, and pencils;
(iv) hand-held audio recorders, the use of which shall be limited
to recording inspection activities;
(v) passive infrared sensors;
(vi) ultrasonic equipment;
(vii) lens measuring devices;
(viii) borescopes;
(ix) other specialised measurement equipment approved by the
Inspector Escorts and appropriate for inspection of the type
of Observation Aircraft, equipment and sensors being
inspected; and
(x) other equipment as approved in writing by the Inspector
Escorts for that inspection; and
(c) with the participation of Inspector Escorts, conduct an inventory of
each item inspection equipment set forth on the list delivered by the
Inspection Team pursuant to subparagraph 1(b) of this Annex, and review
with the Inspector Escorts the accounting procedures the Inspector Escorts
shall follow pursuant to Paragraph 9 of this Annex to confirm that each item
of inspection equipment brought aboard the Observation Aircraft by the
283
Inspection Team has been removed from the Observation Aircraft upon
conclusion of the inspection.
(2) Upon delivery of the Flight Plan, unless otherwise mutually agreed to
by the Observed Party and Observing Party, the Inspection Team of the
Observed Party may inspect the Observation Aircraft, accompanied by
Inspector Escorts of the Observing Party, to determine whether there is any
Prohibited Equipment on the Observation Aircraft. All such inspections shall
be conducted in accordance with Article IX and the Annex C.
(3) The Inspection Team shall be accompanied throughout the entire
inspection of the Observation Aircraft by the Inspector Escorts to confirm
that inspection is being conducted in accrdance with the provisions of this
Annex. The Inspection Team shall facilitate the execution of this duty by the
Inspector Escorts. The Inspector Escorts shall facilitate the inspection of the
Observation Aircraft, its equipment, and its sensors by the Inspection Team.
(4) In conducting its inspection, the Inspection Team shall have full access
to the entire exterior and interior of the Observation Aircraft and its
equipment. Such access shall be provided to, but not limited to, the
following:
(a) cockpit;
(b) cabin area;
(c) tail section;
(d) nose;
(e) wings;
(f) engines;
(g) fuselage; and
(h) cargo and storage areas.
(5) n conducting its inspection, the Inspection team shall have full access
to sensors. All access to sensors and electronic equipment associated with
such sensors connected to or protruding from the exterior or located within
the interior of the Observation Aircraft shall be obtained through access
panels, where such access panels are designed to be opened, removed, and
re-emplaced.
(6) Notwithstanding the provisions of paragraphs 4 and 5 of this Annex,
the inspection shall be conducted in a manner that does not:
284
(a) degrade or damage, or prevent subsequent operation of the
Observation Aircraft, its equipment, or its sensors;
(b) alter the electrical or mechanical structure of the Observation
Aircraft, its sensors, or its equipment; or
(c) impair the air-worthiness of the Observation Aircraft.
The Inspection Team may not open compartments on board the
Observation Aircraft, remove aircraft, sensor, or equipment panels, or
remove physical barriers to access to the Observation Aircraft, its
equipment, or its sensors; provided, however, that the Inspector Escorts
shall, upon request, do all such opening or removal, to the extent that
compartments, panels and barriers in question are designed to be opened,
removed, and re-emplaced. The Inspector Escorts shall equip themselves
with necessary tools to fulfil all such requests promptly. The Inspector
Escorts shall be provided sufficient time during the inspection to re-emplace
and secure all components, panels and barriers that are opened or
removed, so that at the end of the inspection all such components, panels,
and barriers are re-emplaced and secured.
(7) Equipment not on the inspection equipment list delivered by the
Inspection Team pursuant to subparagraph 1 (b) of this Annex may not be
brought on board the Observation Aircraft by the Inspection Team, nor may
the Inspection Team bring weapons of any kind on board the Observation
Aircraft.
(8) The Inspection Team may make notes, photographs, video and voice
recordings, sketches and similar records of the Observation Aircraft and
sensors during the inspection, none of which shall be subject to any review
or examination by the Observing Party.
(9) Upon completion of the inspection, which shall terminate no later than
three hours prior to the scheduled commencement of the Observation
Flight, and shall have a duration of not more than 8 daylight hours, unless
otherwise agreed by the Parties, the Inspection Team shall:
(a) withdraw from the Observation Aircraft and its immediate area to
a location not closer than 25 meters from any part of the observation
Aircraft; and
(b) demonstrate to the Inspector Escorts that all inspection equipment
on the list delivered pursuant to subparagraph 1(b) of this Annex has been
removed from the Observation Aircraft.
285
The Inspector Escorts may use their own accounting procedures to
confirm compliance with subparagraph (b) of this paragraph. If the
Inspector Escorts are unable to confirm compliance with subparagraph (b)
of this Paragraph, the Observed Party may prohibit the Observation Flight,
and no Observation Flight shall be recorded against the Quota of either
Party.
(10) The Inspection Team shall immediately inform the Inspector Escorts of
any equipment suspected to be Prohibited Equipment located by the
Inspection Team on board the Observation Aircraft. If the Observating Party
is unable to demonstrate that the items in question are not Prohibited
Equipment, the Observed Party may prohibit the Observation Flight
pursuant to subparagraph 1(g) of Article XII of the Agreement, and the
Observation Aircraft shall thereupon depart the Territory of the Observed
Party.
(11) Information and briefings furnished by a Party pursuant to this Annex
shall be provided in the language that is designated for that Party in the
Annex B, unless the Party receiving the information or briefing otherwise
agrees.
(12) The Observed Party shall, upon request, provide a suitable briefing
room for briefings provided for by this Annex and for use by Inspector
Escorts in preparing information in connection with inspections. The
Observed Party shall also provide the assistance of clerical personnel to
Inspector Escorts in connection with the performance of their
responsibilities under this Annex.
(13) The Observed Party shall not disclose to non-Parties information about
the Observation Aircraft, its equipment, or its sensors obtained pursuant to
Article IX or this Annex without the express permission of the Observing
Party.
(14) Upon entry into force of this Agreement, each Party shall notify to the
other Party of each type and model of Observation Aircraft and sensor it
intends to use for Observation Flights. Each time a Party intends to use for
Observation Flights a new model of Aircraft or a new model of sensor of
agreed types, it shall notify to the other Party the model of the Aircraft or
sensor. Functional description and generic diagrams of the Aircraft, its
equipment and sensors, to include all sensor components, shall be provided
upon request.
286
(15) Within a period of 30 days after notification of each type and model of
Observation Aircraft and sensor pursuant to Paragraph 14, each Party shall
notify to the other Party of a 7 day period during which a representative
type and model of each such Observation Aircraft and/or sensor shall be
available for examination. The Party whose Observation Aircraft and/or
sensors are being examined shall provide adequate facilities in which to
conduct the examination.
(16) Examinations shall not exceed 48 hours in length without the consent
of the Party whose Observation Aircraft and/or sensors are being examined.
(17) The representatives of Party conducting the examination shall be:
(a) identified to the Party whose Observation Aircraft and sensors are
being examined in advance of the examination;
(b) nationals of the Party;
(c) accorded the inviolability and immunities enjoyed by diplomatic
agents pursuant to Articles 29 and 31 of the Vienna Convention on
Diplomatic Relations for the entire period of their presence in the Territory
of the Party whose Observation Aircraft and sensors are being examined,
and thereafter with respect to acts previously performed in exercise of their
official functions;
(d) accorded the same treatment as is accorded to Aircrew Members
and Inspection Crew Members under paragraph 2 of Article XIX of the
Agreement regarding waiver of immunity, and under paragraphs 3 and 4 of
Article XIX of the Agreement;
(e) governed by the provisions of paragraphs 1, 3-8, and 12 of this
Annex to the extent that those paragraphs are applicable to Inspections
Team members;
(f) accompanied during the examination by representatives of the
Party whose Observation Aircraft and sensors are being examined; and
(g) required to identify specific inspection equipment, and, if
requested by the Party whose Observation Aircraft and sensors are being
examined, shall demonstrate that such equipment will not degrade,
damage, alter or impair the normal operation of the Observation Aircraft
and its sensors.
(18) The Party whose Observation Aircraft and sensors are being examined
shall, prior to commencement of such examination, undertake the
following measures:
287
(a) brief the Party conducting the examination on all necessary safety
precautions for the examination of the Observation Aircraft;
(b) brief the Party conducting the examination on the procedures the
Party whose Observation Aircraft and sensors are being examined intends
to use to allow a thorough examination;
c) brief the Party conducting the examination on the configuration of
the Observation Aircraft and on the location of sensors and associated
equipment on the Observation Aircraft; and
d) use best efforts to answer questions of the Party conducting the
examination pertaining to the examination.
(19) Pursuant to paragraph 17(e) of this Annex, the Party conducting the
examination may not open compartments on board the Observation
Aircraft, remove aircraft, sensor, or equipment panels, or remove physical
barriers to access to the Observation Aircraft, its equipment, or its sensors;
provided, however, that the Party whose Observation Aircraft and sensors
are being examined shall, upon request, do all such opening or removal, to
the extent that the compartments, panels, and barriers in question are
designed to be opened, removed, and re-emplaced.
Annex D – Flight Monitors
(1) Obligation of the Parties
Each Party shall facilitate the mission of the Flight Monitors.
(2) Purposes of the Flight Monitors
The purposes of having Flight Monitors aboard the Observation Aircraft
during the Observation Flight are:
(a) To represent the Observed Party;
(b) To monitor compliance by the Observing Party with the provisions
of the Agreement;
(c) To ensure compliance with the Flight Plan;
(d) To monitor the operation of sensors and other equipment of the
Observation Aircraft;
(e) To advise on national rules of the Observed Party (e.g., rules on
flight safety) as requested by the Observing Party;
288
(f) In the event of an emergency, to facilitate communications as
directed by the pilot in command of the Observation Aircraft.
(3) General Rules for the Conduct of Flight Monitors
(a) Two Flight Monitors shall have the right to board the Observation
Aircraft at the Point of Entry and to remain aboard during the Observation
Flight, including any stops for refuelling or emergencies.
(b) The Flight Monitors shall have the right to bring aboard the
Observation Aircraft maps, flight charts, publications, equipment operating
manuals, and other equipment, such as tape voice recorders.
(c) Except for flight safety reasons, the Flight Monitors shall have the
right to move unencumbered about the Observation Aircraft, including the
flight deck. In exercising their rights, the flight Monitors shall not interfere
with the activities of the Aircrew Members.
(d) The Flight Monitors shall have the right to view the operation of the
sensors by the Observing Party as well as all activities on the flight deck
during the Observation Flight. This includes the right to listen to the
communication of the Observation Aircraft (internal and external) and to
monitor the flight and navigation instrument of the Observation Aircraft.
(e) The Flight Monitors are the representatives of Observed Party
during the conduct of the Observation Flight. Flight Monitors may offer
advice, communicate with air traffic controllers as appropriate, and may
help relay and interpret communications, from the air traffic controllers to
the Aircrew Members, about the conduct of the observation Flight. For this
purpose, the Flight Monitors shall be given access to the radio equipment
of the Observation Aircraft.
(f) Flight Monitors are responsible for knowing the position of the
Observation Aircraft and the location of Hazardous Airspace along and near
the route of the Observation Flight. If a Flight Monitor or air traffic control
personnel of the Observed Party believes that the Observation Aircraft is
deviating from its Flight Plan, the Aircrew Members shall be advised.
(g) Should the Flight Monitors determine that they are not being
permitted to exercise their rights under the Agreement, the Observed Party
shall forward a report specifying the nature of the issue to the Joint Open
Skies Consultative Commission for consideration.
289
Annex E – Sensors
(1) The sensor package for Open Skies purposes may comprise any of the
following types of sensors in any number and combination:
(a) Camera, Optical;
(b) Video Camera.
(2) The Hungarian-Romanian Open Skies Consultative Commission shall
annually consider updates to this Annex.
(3) Signals intelligence collection from the Observation Aircraft is
prohibited. Any device that can collect, process, retransmit, and/or record
electronic signals related to communications, instrumentation, telemetry,
and electronic non-communication signals is prohibited, except: (a) that
equipment required for navigation and flight operations, and (b) those
devices that are components of other sensors (e.g. recording equipment for
onboard non-prohibited sensors). Such excepted equipment and devices
shall not be used to perform any prohibited function.
(4) Data link (encrypted/unencrypted) equipment, such as that which
could be used to transmit sensor data from the Observation Aircraft to a
ground station, to other aircraft or to satellites, is prohibited.
Annex F – Commission
(1) The Commission shall undertake such action as is provided for in
Article XVI of the Agreement.
(2) Each party shall appoint a Representative, assisted by such staff as that
Party deems necessary, to the Commission.
(3) The Commission shall hold one regular session per calendar year
unless it decides otherwise. Special sessions may be convened upon the
request by a Party. Such a Party shall inform the other Party in advance of
the matters to be submitted for consideration.
(4) The initial session of the Commission shall be held within sixty days of
the entry into force of the Agreement. Thereafter, sessions of the
Commission shall be held at the capitals of the Parties, and shall alternate
between the two capitals every year. The Party at whose capital a session is
held shall provide administrative support for that session. Sessions may also
be held at such other places as the Parties may agree.
290
(5) At its initial session, the Commission shall establish its Rules of
Procedure.
(6) The proceeding of the Commission shall be confidential. The
Commission may agree to make its decisions public.
(7) Each Party shall bear the expenses incurred from its participation in the
Commission. Expenses incurred by the Commission as a whole shall be
shared equally by the Parties.
Annex G – Hazardous Airspace
The Hazardous Airspaces of the Parties are those that are published in
the AIP.
Annex H – Processing of Materials of the Observation Flights
(1) Obligations of the Parties
(a) Each Party will in every possible way facilitate the timely and high-
quality processing of the observation materials and their provision to the
Observing Party.
(b) The Party carrying out the processing shall be responsible for the
quality of the processing of the Observation Flight materials.
(2) (a) The initial processing (development) of Observation Flight materials
shall be carried out in established ground facilities to be notified by the
Parties upon entry into force of the Agreement, by mixed groups of
specialists of the Observed and the Observing Parties and with the aid of
agreed equipment.
(b) Whenever it is possible to install dual sensors on board of the
Observation Aircraft, the Observing Party shall take home one set of
observation materials while the other original set of observation material
shall be retained by the Observed Party. If it is not possible to install dual
sensors on board of the Observation Aircraft, the observation material shall
remain with the Observed Party while the copy shall be taken home by the
Observing Party.
291
APPENDIX I
TREATY ON OPEN SKIES—PREAMBLE AND CONTENTS
References, Preamble, Table of Contents
I.1 References to full text
The full text of the Treaty on Open Skies as signed on 24 March 1992 can
be found, for example, at the following sources:
a) World Wide Web:
http://www.state.gov/www/global/arms/treaties/openski1.html,
http://www.osmpf.wpafb.af.mil (22 March 2002);
b) Printed version:
- Bundesgesetzblatt 2045 Teil II Z 1998 A, ausgegeben zu Bonn am
3. Dezember 1993, Nr. 43, Gesetz zu dem Vertrag vom 24. März
1992 über den offenen Himmel, Bundesanzeiger
Verlagsgesellschaft mbH
- SIPRI, Yearbook 1993, pp. 653-71 (without Annexes), (in English)
- R. Hartmann and W. Heydrich, Der Vertrag über den offenen
Himmel, Nomos: Baden-Baden, 2000 (in German).
I.2 Preamble and Article I of the Treaty
TREATY ON OPEN SKIES
Preamble
The States concluding this Treaty, hereinafter referred to collectively as the
States Parties or individually as a State Party,
Recalling the commitments they have made in the Conference on
Security and Co-operation in Europe to promoting greater openness and
transparency in their military activities and to enhancing security by means
of confidence- and security-building measures,
292
Welcoming the historic events in Europe which have transformed the
security situation from Vancouver to Vladivostok,
Wishing to contribute to the further development and strengthening of
peace, stability and co-operative security in that area by the creation of an
Open Skies regime for aerial observation,
Recognizing the potential contribution which an aerial observation
regime of this type could make to security and stability in other regions as
well,
Noting the possibility of employing such a regime to improve openness
and transparency, to facilitate the monitoring of compliance with existing or
future arms control agreements and to strengthen the capacity for conflict
prevention and crisis management in the framework of the Conference on
Security and Co-operation in Europe and in other relevant international
institutions,
Envisaging the possible extension of the Open Skies regime into
additional fields, such as the protection of the environment,
Seeking to establish agreed procedures to provide for aerial
observation of all the territories of States Parties, with the intent of observing
a single State Party or groups of States Parties, on the basis of equity and
effectiveness while maintaining flight safety,
Noting that the operation of such an Open Skies regime will be without
prejudice to States not participating in it,
Have agreed as follows:
Article I
GENERAL PROVISIONS
1. This Treaty establishes the regime, to be known as the Open Skies
regime, for the conduct of observation flights by States Parties over the
territories of other States Parties, and sets forth the rights and obligations of
the States Parties relating thereto.
2. Each of the Annexes and their related Appendices constitutes an
integral part of this Treaty.
293
I.3 Table of Contents of the Treaty on Open Skies
Preamble
Article I General Provisions
Article II Definitions
Article III Quotas
Article IV Sensors
Article V Aircraft Design
Article VI Choice of Observation Aircraft, General Provisions for
the Conduct of Observation Flights, and Requirements
for Mission Planning
Article VII Transit Flights
Article VIII Prohibitions, Deviations from Flight Plans and
Emergency Situations
Article IX Sensor Output from Observation Flights
Article X Open Skies Consultative Commission
Article XI Notifications and Reports
Article XII Liability
Article XIII Designation of Personnel and Privileges and Immunities
Article XIV Benelux
Article XV Duration and Withdrawal
Article XVI Amendments and Periodic Review
Article XVII Depositaries, Entry into Force and Accession
Article XVIII Provisional Application and Phasing of Implementation
of the Treaty
Article XIX Authentic Texts
Annex A Quotas and Maximum Flight Distances
Annex B Information on Sensors
Annex C Information on Observation Aircraft
Annex D Certification of Observation Aircraft and Sensors
Annex E Procedures for Arrivals and Departures
Annex F Pre-Flight Inspections and Demonstration Flights
Annex G Flight Monitors, Flight Representatives, and
Representatives
Annex H Co-ordination of Planned Observation Flights
Annex I Information on Airspace and Flights in Hazardous
Airspace
294
Annex J Montreux Convention
Annex K Information on Film Processors, Duplicators and
Photographic Films, and Procedures for Monitoring the
Processing of Photographic Film
Annex L Open Skies Consultative Commission
295
APPENDIX J
DECISIONS OF THE OPEN SKIES CONSULTATIVE
COMMISSION (TITLES ONLY)
The full text of decisions 1-22 can be found at http://
www.osmpf.wpafb.af.mil (22 March 2002).
1992-1997
DECISION ONE: Distribution of Costs Arising Under the Treaty on Open
Skies as Decided in the Open Skies Consultative Commission on 29
June 1992 and Adopted by the Commission's Decision of 10
December 1992
DECISION TWO: Additional Non-Destructive Testing Equipment, 29 June
1992
DECISION THREE: Methodology for Calculating the Minimum Height
Above Ground Level at Which Each Optical Camera Installed on an
Observation Aircraft May Be Operated During an Observation Flight,
29 June 1992
DECISION FOUR: Minimum Camera Specification for an Observation
Aircraft of an Observed Party Exercising its Right to Provide an
Observation Aircraft for an Observation Flight, 29 June 1992
DECISION FIVE: Responsibility for the Processing of Film Used During an
Observation Flight, 29 June 1992
DECISION SIX: Rules of Procedure and Working Methods of the Open
Skies Consultative Commission, 1992
DECISION SEVEN: Methodology for Determining the Ground Resolution of
Synthetic Aperture Radar (SAR), 10 December 1992
DECISION EIGHT: Intervals at Which Data Shall Be Annotated With
Information, 16 July 1993
DECISION NINE: Codes Other than Alphanumeric Values to Be Used for
the Annotation of Data, 16 July 1993
DECISION TEN: Scale of Distribution for the Common Expenses Associated
With the Operation of the Open Skies Consultative Commission, 1993
296
DECISION ELEVEN: Financial and Administrative Questions, Relating to
Point VI of Decision OSCC/I/Dec. 6 of June 1992, 1993
DECISION TWELVE: Information to Be Provided Together With Calibration
Target Diagrams, 6 December 1993
DECISION THIRTEEN: Methodology for Calculating the Minimum
Permissible Flight Altitude When Using Optical and Video Cameras, 18
April 1994
DECISION FOURTEEN: Methodology for Calculating the Minimum Height
Above Ground Level at Which Each Video Camera With Real Time
Display Installed on an Observation Aircraft May Be Operated During
an Observation Flight, 12 October 1994
DECISION FIFTEEN: Methodology for Calculating the Minimum Height
Above Ground Level at Which Each Infrared Line-Scanning Device
Installed on an Observation Aircraft May Be Operated During an
Observation Flight, 12 October 1994
DECISION SIXTEEN: Calibration of Ground Processing Equipment, Used
for the Determination of H
min
from Video Cameras or Infrared Line-
Scanning Devices and for Calibrating Ground-Based Tape Reproducers
Used to Replay Data from SAR Sensors, 12 October 1994
DECISION SEVENTEEN: The Format in Which Data is to Be Recorded and
Exchanged on Recording Media Other Than Photographic Film, 12
October 1994
DECISION EIGHTEEN: Mandatory Time Period for Storing and Sharing
Data Recorded During an Observation Flight, 12 October 1994
DECISION NINETEEN: Supplementary Provisions for the Completion of the
Mission Plan and for the Conduct of an Observation Flight, 23 January
1995
DECISION TWENTY: Provisions for a Three-Letter and Telephony
Designator for the Open Skies Consultative Commission and Aircraft
Identification for Open Skies Flights, 12 June1995
DECISION TWENTY-ONE: Establishment of an Open Skies Central Data
Bank, 23 October 1995
DECISION TWENTY-TWO: Provisions for the Use of a Standard “Pre-Flight
Inspection Report”, 18 March 1996
DECISION TWENTY-THREE: was not issued.
DECISION NUMBER TWENTY FOUR: Extension of the Period of
Provisional Application of the Open Skies Treaty Working Modalities of
the OSCC, 24 December 1997
297
2001 (Decisions were issued in 2001 without numbering)
OSCC Decision on the Working Modalities of the OSCC, 25 June 2001
OSCC Decision on Information Seminar on the Treaty on Open Skies, 17
September 2001
OSCC Decision on the Informal Working Groups of the Open Skies
Consultative Commission, 29 October 2001
OSCC Decision on Rules of Procedure and Working Methods of the Open
Skies Consultative Commission, 17 December 2001
OSCC Decision on Provisions for the Initial Certification Period, 17
December 2001
2002
DECISION No. 01/02 Fulfilment of the Requirements for the Czech
Republic and Slovakia to Exercise the Rights and
Fulfil the Obligations, 21 January 2002
DECISION No. 02/02 Accession of the Republic of Finland to the Treaty
on Open Skies, 21 January 2002
DECISION No. 03/02 Accession of the Kingdom of Sweden to the Treaty
on Open Skies, 21 January 2002
DECISION No. 04/02 Provisions for the Use of a Standard “Signature
Page to the Certification Report”, 18 February
2002
DECISION No. 05/02 Provision for the Use of “PG” as a Unique
Identifier for Pod Group Sensor Configurations, 1
February 2002
DECISION No. 06/02 Procedures for Allocation of Observation Flight
Reference Numbers, 18 February 2002
DECISION No. 07/02 Revision 1 of DECISION NUMBER ONE (Travel
Expenses), 18 March 2002
DECISION No. 08/02 Guidelines for Accession to the Treaty on Open
(REV1/Corr 1) Skies, 22 April 2002
DECISION No. 09/02 Protection of Data Collected During Observation
Flights and Transfer of Recording Media
Containing this Data, 13 May 2002
DECISION No. 10/02 Revision 1 of DECISION NUMBER EIGHTEEN, 10
June 2002
298
DECISION No. 11/02 Revision 2 of DECISION NUMBER ONE, 10 June
2002
DECISION No. 12/02 Allocation of a Passive Quota to Sweden, 22 July
2002
DECISION No. 13/02 Allocation of Observation Flight Reference
Numbers, 22 July 2002
DECISION No. 14/02 Revision 1 of DECISION NUMBER TWENTY, 22
July 2002
DECISION No. 15/02 Provision on Calibration Targets, 22 July 2002
DECISION No. 16/02 Mission Plan Submission and Review, 22 July
2002
DECISION No. 17/02 Accession of the Republic of Lithuania to the
Treaty on Open Skies, 22 July 2002
DECISION No. 18/02 Accession of the Republic of Croatia to the Treaty
on Open Skies, 22 July 2002
DECISION No. 19/02 Accession of Bosnia and Herzegovina to the Treaty
on Open Skies, 22 July 2002
DECISION No. 20/02 Accession of the Republic of Latvia to the Treaty
on Open Skies, 22 July 2002
DECISION No. 21/02 Scale of Distribution for the Common Expenses
Associated with the Operation of the Open Skies
Consultative Commission, 9 September 2002
DECISION No. 22/02 Procedures for Transit Necessary During a
Segment of an Open Skies Observation Flight, 16
December 2002
DECISION No. 23/02 Allocation of a Passive Quota to Finland, 16
December 2002
DECISION No. 24/02 Revision 2 of Decision Number Twenty to the
Treaty on Open Skies, 16 December 2002
DECISION No. 25/02 Revision 1 of Decision No. 6/02 to the Treaty on
Open Skies, 16 December 2002
2003 (Until June 2001)
DECISION No. 01/03 Revision of DECISION NUMBER TWENTY, 27
January 2003
DECISION No. 02/03 Amendment 1 to DECISION NUMBER SIX, 24
February 2003
DECISION No. 03/03 Open Skies Aircraft Status, 24 February 2003
299
DECISION No. 04/03 Correction to Annex G, Section I, Paragraph 7, 24
February 2003
DECISION No. 05/03 Accession of the Republic of Slovenia, 24
February 2003
DECISION No. 06/03 Scale of distribution for the common expenses, 24
March 2003
DECISION No. 07/03 Scale of distribution for the common expenses, 24
March 2003
DECISION No. 08/03 Allocation of a passive quota to Georgia, 24 April
2003
DECISION No. 09/03 Revision 2 of Decision No. 6/02, 5 May 2003
DECISION No. 10/03 Accession of the Republic of Estonia to the Treaty
on Open Skies, 5 May 2003
DECISION No. 11/03 Supplement 1 to Decision Number One to the
Treaty on Open Skies, 16 June 2003
300
301
APPENDIX K
SENSOR GUIDANCE DOCUMENT—TABLE OF CONTENTS
The full text of the document is reprinted in R. Hartmann and W. Heydrich,
Der Vertrag über den offenen Himmel, Baden-Baden: Nomos, 2000, pp.
359-550. It can also be found at http://www.osmpf.wpafb.af.mil (22 March
2002).
No. 500/A
Open Skies Consultative Commission 26 May, 1997
Informal Working Group on Sensors GTOCL.DOC
GUIDANCE FOR CERTIFICATION OF OPEN SKIES SENSORS INSTALLED
ON
OBSERVATION AIRCRAFT AND THEIR ASSOCIATED PROCESSING,
DUPLICATING, AND ANALYSIS EQUIPMENT AND THE CONDUCT OF
DEMONSTRATION FLIGHTS
TABLE OF CONTENTS
SECTION 1 GENERAL
1. 1 INTRODUCTION
1.1.1 Purpose And Use Of Guidance Document For Certification And
Demonstration
1.1.2 Organization Of The Guidance Document
1.1.3 Definitions
1.2 GENERAL PROVISIONS
1.2.1 Purpose Of The Certification; Rights And Obligations Of State
Parties
1.2.2 Purpose Of A Demonstration Flight
1.2.3 Timing
1.2.3.1 Timing For Certification
1.2.3.2 Timing For Observation Flights
1.2.4 People Participating
302
SECTION 2 PREPARATION FOR CERTIFICATION AND/OR
DEMONSTRATION FLIGHT
2.1 NOTIFICATIONS AND PRE-ARRIVAL EXCHANGES OF
INFORMATION FOR CERTIFICATION
2.2 AIRCRAFT AND SENSOR CONFIGURATIONS
2.2.1 Identification Of The Certifying State Party's Observation
Aircraft
2.2.2 Identification Of Certifying State Party's Observation Aircraft
Sensors
2.2.2.1 Sensor Configuration And Installation
2.2.2.2 Information To Be Provided On Resolution Degrading
Devices
2.2.3 Identification Of Certifying State Party's Sensor Recording
Media
2.2.4 Identification Of Certifying State Party's Data Annotation
Technique And Capabilities
2.2.5 Timing of Information To Be Provided by Certifying State Party
2.2.6 Preparation For Ground And In-Flight Examination
2.3 AIRBORNE AND GROUND TEST EQUIPMENT
2.3.1 Equipment Permitted To Be Brought On Board The Observation
Aircraft During The Ground Examination And In-Flight
Examination Of A Certification
2.3.2 Equipment Permitted To Be Brought Into The Processing And
Duplication Facilities
During Certification and Demonstration Flights
2.3.3 Equipment To Be Provided To Personn el A t Th e Cal ibrati on
Tar g et Site
2.3.4 Atmospheric Measurement Equipment W1ich May Be Supplied
By The State Party Conducting The Certification Or By The
Observed State Party When A Demonstration Is Requested By
Any State Party
2.3.5 Analysis Equipment Necessary To Conduct A Demonstration
Flight
2.4 CALIBRATION OF EQUIPMENT PRIOR TO ARRIVAL OF OTHER
STATES PARTIES
2.4.1 Target Measurement Equipment
2.4.2 Atmospheric And Altitude Measurement Equipment
2.4.3 Metrological Verification And Ground Equipment Calibration
2.5 SUPPORT TO BE SUPPLIED DURING CERTIFICATION
303
2.6 CALIBRATION TARGET SPECIFICATIONS, LAYOUT AND DATA TO
BE PROVIDED PRIOR TO TESTS
2.6.1 General
2.6.2 Optical And Video Calibration Targets
2.6.3 Infrared Targets And Measurements
2.6.3.1 Target Construction
2.6.3.2 Temperature Control Specifications
2.6.3.3 Target Layout
2.6.3.4 Target Background Considerations
2.6.3.5 Target Temperature Measurements
2.6.4 SAR Targets
2.6.4.1 Trihedral Corner Reflectors
2.6.4.2 SAR Impulse Response (IPR) Calibration Target Array
2.6.4.3 Dynamic Range And Amplitude Linearity Array
2.6.4.4 Multipath Considerations For Calibration Array Deployment
2.6.4.5 SAR Target Tolerance
2.7 PROCESSING AND DUPLICATION INFORMATION
2.7.1 Processing And Duplication Information For Sensors That
Record On Photographic Film
2.7.2 Processing And Duplication Information For Sensors That
Record On MagneticTape
SECTION 3 CONDUCT OF GROUND ACTIVITIES
3.1 GENERAL CONSIDERATIONS
3.2 NOTIFICATION OF THE AIRCRAFT SENSORS AND ASSOCIATED
EQUIPMENT
3.3 INFORMATION TO BE EXCHANGED
3.4 VERIFICATION OF CALIBRATION AND PROCESSING PARAMETERS
OF EQUIPMENT FOR THE PROCESSING, DUPLICATION, AND
CONVERSION OF. DATA
3.4.1 General Considerations
3.4.2 Verification Of Calibration And Preparation Of Equipment For
The Processing And Duplication Of Film
3.4.3 Calibration Of Processing Equipment And Conversion Facilities
Used For Recording And Exchanging Media Other Than
Photographic Film
3.4.3.1 Ground Processing Equipment For The Determination Of
H
min
304
3.5 CONDUCT OF THE GROUND EXAMINATION OF THE AIRCRAFT
AND SENSORS
3.5.1 General Considerations
3.5.2 Ground Examination
3.5.2.1 Inventory Of Equipment
3.5.2.2 Information To Be Collected
3.6 ATMOSPHERIC DATA
3.6.1 Definitions And Measurement Conditions
3.6.1.1 Atmospheric Data Required To Support Certification Flights
3.6.1.2 Atmospheric Data Required To Support Demonstration
Flights
3.7 TARGET MEASUREMENTS
3.7.1 General Considerations
3.7.2 Target Measurements For Optical And Video Sensors
3.7.3 Target Measurements For Infrared Sensors
3.7.4 Target Measurements For SAR Sensors
3.8 TECHNICAL ASPECTS OF FLIGHT ORGANIZATION
3.8.1 General Considerations
3.8.2 Request For And Organization Of A Demonstration Flight
3.8.3 Briefings On Flight Conduct
3.8.4 Flight Path Planning
3.8.4.1 Optical Camera Considerations
3.8.4.2 Video Camera Considerations
3.8.4.3 Infrared Linescanning Device Considerations
3.8.4.4 Sideways Looking SAP. Considerations
3.8.5 Weather, Visibility And Time Of Day Considerations (Rules,
Limitations, Etc.)
3.8.6 Selection of Height Above Ground Of The Flight
3.8.6.1 Certification: Selection Of The Height Above Ground Of
The Flight
3.8.6.2 Demonstration: Selection Of The Height Above Ground Of
The Flight
3.9 TRAINING OF PERSONNEL PARTICIPATING IN CERTIFICATIONS
AND DEMONSTRATION FLIGHTS
3.9.1 General
3.9.2 Media Processing
3.9.3 Training Of Observers Used For Visual Analysis
305
SECTION 4 CONDUCT 01E IN-FLIGHT EXAMINATIONS & THE
COLLECTION OF IMAGERY FOR BOTH CERTIFICATION
& DEMONSTRATION FLIGHTS
4.1 GENERAL CONSIDERATIONS
4.2 IN-FLIGHT EXAMINATION OF THE AIRCRAFT AND SENSORS
4.2.1 General Considerations
4.2.2 Information To Be Provided
4.2.3 Parameters Monitored During The Flight
4.2.4 Collection Of The Imagery And Flight Data
4.2.4.1 Flight Data
4.2.4.2 Flight Path
4.2.4.3 Additional Conditions Of Flight
4.2.5 Sensor Operating Modes
SECTION 5 MEDIA PROCESSING, ANALYSIS AND THE GENERATION
OF REPORTS
5.1 PROCESSING AND ANALYSIS OF DATA COLLECTED&
CALCULATION OF H
min
5.1.1 General Considerations
5.1.2 Media Processing
5.1.2.1 Black & White Film Processing
5.1.2.2 Analog Tape Data Processing
5.1.2.2.1 Phase Correction
5.1.2.3 Digital Tape Data Processing
5.1.3 Data Annotation Acceptability Criteria
5.1.4 Imagery Acceptability Criteria And Preparation For Imagery And
Data Analysis
5.1.4.1 Optical Cameras
5.1.4.2 Video Cameras
5.1.4.3 Infrared Linescanning Devices
5.1.4.4 Sideways Looking Synthetic Aperture Radar Sensors
5.1.5 Imagery And Data Analysis For Ground Resolution
Determination
5.1.5.1 Analysis Of Data And Imagery From Optical Cameras
5.1.5.1.1 Determination Of The Ground Resolution (L
2
)
306
5.1.5.1.2 Determination Of The Effective Target Contrast
Modulation (K
2
)
5.1.5.1.3 Determination Of H
min
5.1.5.2 Analysis Of Imagery And Data From Video Cameras
5.1.5.2.1 Determination Of The Ground Resolution (L
i,J
)
5.1.5.2.2 Determination Of The Modulation Of The Calibration
Target As Measured In The Image (K
i
)
5.1.5.2.3 Colour Video Camera Considerations
5.1.5.2.4 Procedure For Calculating The Minimum Height Above
Ground Level At Which A Video Camera May Be
Operated During An Observation Flight
5.1.5.3 Analysis Of Imagery And Data From Infrared Linescanning
Devices
5.1.5.3.1 LOWTRAN Input Parameters
5.1.5.3.2 Determination Of The Maximum Heights, (H
i
), At.Which
The Bar Groups Are Resolved
5.1.5.3.3 Determination Of The Exponent ‘m’
5.1.5.3.4 Radiant Temperature Calculation (Dt
i
)
5.1.5.3.5 H
min
Determination
5.1.5.4 Analysis Of Imagery And Data From Sideways Looking
Synthetic Aperture Radar (SAR) Sensors
5.1.5.4.1 Data Analysis Procedures
5.1.5.4.2 Image Production From Initial Phase Information
5.1.5.4.3 Conversion Of Pixel Resolution To Ground Resolution
5.1.5.4.4 Comparison Of The Two Methods Of Resolution
Measurement Performed During The SAR Certification
5.1.5.4.5 Amplitude Linearity Of The Radar
5.2 MEDIA DUPLICATION
5.2.1 Film Duplication Facilities First Generation Duplicates
5.2.2 Calibration Of Duplication Facilities For Media Recorded On 0-
ther Than Film
5.3 CERTIFICATION REPORT AND OTHER DOCUMENTATION
5.3.1 General Considerations And Treaty Citations
5.3.2 Certification Report Format And Contents
FIGURES AND TABLES
Figure 1-1 Open Skies Certification and Designation Timeline
Figure 1-2 Open Skies Observation Flight Timeline
307
Figure 2-1 Target Types
Figure 2-2 Brightness Panels
Figure 2-3 Taper on Edge of Trihedral Corner Reflector
Figure 2-4 Measurementof Trihedral Corner Reflectors
Figure 2-5 Experimental Indications of Diffusion for Corner Reflectors
in the Horizontal Plane
Figure 2-6 Experimental Curves of the Corner Reflector Scattering Cross
Section in the Resonance Area in Relation to the Design
Scattering Cross Section Value
Figure 2-7 Calibration Target Array to Measure Impulse Response (IPR)
Figure 2-8 Optimal Elevation Angles for Maximum Radar Cross Section
(RCS)
Figure 2-9 Calibration Target Array to Measure Dynamic Range or
Linearity of Amplitude Response
Figure 2-10 Amplitude Linearity Response Plot
Figure 2-11 No-Return Region Using Available Target Area
Figure 2-12 No-Return Region Using Marimade Target
Figure 3-1 Example of Open Skies Calibration Target Arrays for
Certification Flights at an Airport Facility
Figure 5-1 K
p
Diagram - Horizontal View
Figure 5-2 K
p
Diagram - Vertical View
Figure 5-3 Signal Magnitude vs. Radians/Sample
Figure 5-4 Measured vs. Actual Corner Reflector Cross Section dB
Table 2-1 Calibration Target - Example 1
Table 2-2 Calibration Target - Example 2
Table 2-3 Radar Cross Section and Physical Dimensions of Individual
Trihedral Corner Reflectors for Two Wavelengths
Table 5-1 LOWTRAN Variables
Table 5-2 Resolution vs. Height - Example
Table 5-3 Resolution vs. Height - Summary Example
Table 5-4 Resolution vs. Height - Miscalculated Height Example
Table 5-5 Resolution vs. Height - Summary Example 2
Table 5-6 Examples of Weighting Functions
308
309
SELECTED BIBLIOGRAPHY
Armstrong, D. G., “Technical Challenges Under Open Skies” in
Proceedings of the First International Airborne Remote Sensing
Conference and Exhibition, Strasbourg, Ann Arbor: Environmental
Research Institute of Michigan, Vol. I, 1994, pp. 49-60.
Arnold, K., Der Vertrag über den Offenen Himmel: Ein Konzept zur
Aktualisierung des Vertrages, SWP Studie, Stiftung Wissenschaft und
Politik, Berlin, June 2002.
Baker, J. C., K. M. O’Connell and R. A. Williamson (eds), Commercial
Observation Satellites, Santa Monica and Bethesda: Rand and ASPRS,
2001.
Britting, E., Rüstungskontrolle im "Offenen Himmel" Europaische Sicherheit,
No. 6, 2000, pp. 15-17.
Britting, E. and H. Spitzer, “The Open Skies Treaty”, in Poole, J. B. and
R. Guthrie (eds), Verification Report 1992, VERTIC, London and New
York: The Apex Press, 1992, pp. 223-38.
Clark, J., “Open Skies”, in M. Slack and H. Chestnutt (eds), Open Skies—
Technical, Organizational, Operational, Legal and Political Aspects,
Toronto: Centre for International and Strategic Studies, York
University, 1990.
Gabriele, M. D., The Treaty on Open Skies and its Practical Applications
and Implications, Santa Monica, CA: RAND, 1998.
Hartmann, R., Inkrafttreten des Vertrags über den Offenen Himmel, SWP
Aktuell 25, Berlin, Stiftung Wissenschaft und Politik, December 2001,
p. 4.
Hartmann, R. and W. Heydrich, Der Vertrag über den Offenen Himmel,
Baden-Baden: Nomos Verlag, 2000.
Hawes, J. A., Open Skies: Beyond “Vancouver to Vladivostok”, Occasional
Paper No. 10, Washington, DC: The Henry L. Stimson Center,
December 1992.
Jones, P. and M. Krasznai, “Open Skies: Achievements and Prospects”, in
Poole, J. B. and R. Guthrie (eds), Verification Report 1992, VERTIC,
London and New York: The Apex Press, 1992.
Jones, P., “Open Skies: A Review of Events at Ottawa and Budapest”, in
J.B. Poole (ed.), Verification Report 1991, VERTIC, London and New
York: The Apex Press, 1991.
310
Jones, P., “Open Skies: Events in 1993”, in Poole, J. B. and R. Guthrie (eds),
Verification Report 1993, VERTIC, London and New York: Brassey’s,
1993.
Koulik, S. and R. Kokoski, Conventional Arms Control—Perspectives on
Verification, SIPRI: Oxford University Press, 1994.
Krepon, M., P. Zimmerman, L. Spector and M. Umberger (eds),
Commercial Observation Satellites and International Security, New
York: St. Martin's Press, 1990.
Krepon, M., and A. E. Smithson (eds), Open Skies, Arms Control, and
Cooperative Security, New York: St. Martin’s Press, 1992.
Krepon, M. (ed.), A Handbook of Confidence-Building Measures for
Regional Security, Washington, DC: The Henry L: Stimson Center,
1995.
Lillesand, T. M. and R. W. Kiefer, Remote Sensing and Image Interpretation,
New York: John Wiley, 1987.
Lysyshyn, R. J., “Open Skies Ahead”, NATO Review, Vol. 40, No. 1,
February 1992, pp. 23-26.
Jia, X, J. A. Richards and D. E. Ricken (eds), Remote Sensing Digital Image
Analysis, Heidelberg and New: York Springer, 1993.
Rostow, W. W., Open Skies, Austin: University of Texas Press, 1982.
Simmons, S. P., “When Better Resolution Is Not Good: The Treaty on
Open Skies”, in Proceedings of the Second International Airborne
Remote Sensing Conference and Exhibition, San Francisco, Ann
Arbor: Environmental Research Institute of Michigan, Vol. I, 1996,
pp. 403-10.
Slack, M. and H. Chestnutt (eds), Open Skies: Technical, Organizational,
Operational Legal and Political Aspects, Toronto: Centre for
International and Strategic Studies, York University, 1990.
Spector, L. S., “The Not-So-Open Skies,” in Krepon et al. (eds),
Commercial Observation Satellites and International Security, New
York: Macmillan, 1990, pp. 165-66.
Spitzer, H., “Technical Potential for Monitoring from Air and Space”, in
Altmann, J., H. J. van der Graaf, P. Markl, and P. Lewis (eds),
Verification at Vienna—Monitoring Reductions of Conventional Armed
Forces, New York: Gordon & Breach, 1992, pp. 125-139.
Spitzer, H., “The Open-Skies Treaty as a Tool for Confidence Building and
Arms Control Verification”, extended contribution to the 18th
ISODARCO Summer Course, Siena, July 1996, Technical Report,
CENSIS-Report-23-96, University of Hamburg, Luruper Chaussee
149, D-22761 Hamburg, 1996.
311
Treaty on Open Skies: Message from the President of the United States
Transmitting the Treaty on Open Skies with Twelve Annexes, Signed at
Helsinki on March 24, 1992, Washington, DC: US Government
Printing Office, 1992, pp. 267.
Treaty on Open Skies, Report to Accompany Treaty Doc. 102-37,
Washington, DC: US Government Printing Office, 1993, pp. 174.
Tucker, J. B., “Back to the Future: The Open Skies Talks”, Arms Control
Today, Vol. 20, No. 8, October 1990, pp. 20-24.
312
