Keywords: Caspian Sea, rise in sea level, seismic survey planning, shoreline, base map

Abstract
This study focuses on the limitations and benefits of satellite images when scouting a near-shore area in the North Caspian Coastal Lowland of Kazakhstan for the practical purpose of seismic survey planning. In order to generate a base map for a seismic survey, the locations of recent transgressive and retrogressive shorelines were interpreted and extracted from multi-temporal LANDSAT MSS/TM, JERS-1 OVN/SAR and SPOT HRV images. The rise in sea level from 1977 to 1995 resulted in a northward migration of shorelines and reed zones in the study area near the Ural River mouth. The previous locations of these shorelines were examined in terms of the present depth of water using a published curve showing changes in Caspian Sea level. Thus a 1998 bathymetric map was successfully prepared by an integrated analysis. This bathymetric map was further interpreted and land cover units were also added as supplementary information. Finally, four zones are highlighted on the seismic base map as those to be the receiver and/or source lines for the seismic data acquisition of 1998 as well as at the time of the lowest (1977) and highest (1995) sea levels.

Introduction
When conducting a seismic survey in a very shallow water area for petroleum exploration, operational constraints are highly related to surface conditions. Therefore, during planning it is of crucial importance to investigate and analyze conditions such as depth of water, communities of aquatic plants (reed zone) and subsurface geology. However, those conditions are not classified or shown in detail on some published topographic maps. For this reason, a map with the above logistic information using satellite data should be of high value when selecting survey lines and methods. The final goal of this study is to create a base map for a survey design. The significant role of the satellite image interpretation is to obtain the present isobaths in the southern part of the study area.

Study Area and Satellite Data Used
The study area is situated in the western part of Kazakhstan and covers the North Caspian Coastal Lowland (Caspian Depression) as well as the near-shore to very shallow water area, with water depths of less than five meters (Figure 1). This area covers approximately 25,000 km 2 and is located in the southern part of the Pre-Caspian Basin where oil and gas potential is very high. Giant oil and gas fields, such as Tengiz oil field and Astrakhan gas field have been discovered and developed along the coastal lowland. The following multi-temporal satellite images, on a scale of 1:300,000, were used for the study; LANDSAT MSS/TM and SPOT HRV (Figure 2) as well as JERS-1 OVN/SAR.

Current State of Rise in Sea Level
From a long-term viewpoint, changes in sea level of several meters have repeatedly occurred since the 1600s. From 1977 up to 1995 the sea level has been continuously rising by up to more than two meters (Figure 3). Because of this recent rise in sea level, water has been gradually intruding into the coastal lowland, and a serious influence is taking place on harbor, petroleum facilities, roads etc. around the study area. On the other hand, in terms of seasonal changes, the sea level during summer rises by up to approximately 20 centimeters from the annual mean sea level, and, in turn, falls by approximately 10 centimeters in the winter (Figure 4).

Study Contents
This study involved the following contents.(1) The definition and extraction of shorelines from multi-temporal satellite images, (2) the interpretation of recent changes of shoreline, (3) a field verification survey, (4) an estimation of depth of water in 1998 and (5) the generation of a base map for seismic survey planning.

Definition and Extraction of Shorelines
During the period of annual mean sea level (in spring / autumn), the near-shore area is usually submerged up to the landward limit of the perennial dense reed zone which occupies the near-shore to very shallow water area (Figure 4). This phenomenon was confirmed during a field verification survey. It was defined that the shoreline at the annual mean sea level coincides with the landward limit of the reed zone. Thus, the shorelines in 1977, 1979 (equal to 1975), 1984, 1986, 1988, 1989 and 1998 (equal to 1993) were extracted by means of identification of those limits from multi-temporal satellite images (Figure 5).

Changes of Shoreline from 1977 to 1995
The most recent lowest sea level was recorded in 1977 (-29.04 meters above oceanic sea level), which was followed by a continuous rise in sea level (Figure 3). With this rise in sea level the shoreline in the study area has moved landward toward the north (Figure 6). This gradual rise in sea level extends as much as 25 kilometers, maximum, in the vicinity of the Ural River mouth, and 35 kilometers to the south-southeast of Atyrau from 1977 to 1995. Due to a 2.42-meter rise of sea level during the past 18 years (Figure 3), two near-shore terrace zones (slightly elevated) in 1977 have gradually been submerged and changed into very shallow water areas (Figure 5). An elongated peninsula near the Ural River mouth also has become narrower and narrower. Then the peninsula has changed into isolated islands from 1988 and finally disappeared in 1993 (Figure 5).

Depth of Water in 1998
Recognizing the rise of sea level from previous shoreline positions in 1977, 1979, 1984, 1986, 1988 and 1989, respectively, the depth of water in 1998 can be calculated. This can be done on the assumption that there were no significant changes through erosion or sedimentation in submarine topography of the submerged area during the 21 years since 1977. Based on this principle, depth of water in 1998 was obtained using a curve showing recent changes in sea levels. Thus depth of water in 1998 was recognized through image interpretation by up to approximately 35 kilometers from the shoreline.

1998 Bathymetric Map
Prior to the generation of a base map covering the entire study area, six submarine topographic profile lines were set up with almost an orthogonal direction to the shoreline in 1998. Subsequently, profiles were drawn considering partial changes of sea bottom gradient created by two terrace zones (slightly elevated). A 1998 bathymetric map (Figure 7) was generated, expressed in isobaths of 50-centimeter intervals, using these profiles as well as actual depths of water measured during the field verification survey conducted in June, 1998. As a result, a submarine topography was obtained in the present offshore area including two terrace zones mentioned above. The sea bottom gradients from 2.0-meter isobath southward are steeper than shallow areas of less than 2.0 meters deep. The previous peninsula near the Ural River mouth has also steep submarine topography (Figure 7).

Generation of a Base Map for Seismic Survey Planning
The above bathymetric map was further interpreted and land cover units were also added as supplementary information. Finally, four zones are highlighted on the seismic base map (Figure 8) as those to be sources and receivers for seismic data acquisition. The following methods of measurement were designed respectively in three zones of the very shallow water area among four zones shown in Figure 8.

(1)Zone A: Approximately less than one meter deep. Explosives and geophones should be applied respectively as source and receiver. A cable or radio telemetry system is preferable as a receiving system.

(2)Zone B: Depth of water exceeds one to six meters, maximum. Air guns and hydrophones should be applied respectively as source and receiver. A bottom cable or radio telemetry system is preferable as a receiving system.

(3)Zone R: This is a reed zone along a shoreline. Although depth of water is approximately one meter, a seismic survey could not be conducted by any kind of method during 1998 because deforestation was prohibited within this zone.

Conclusions
(1)When extracting “the shoreline at annual mean sea level” from satellite images, the landward limits of the reed zone along the shoreline can be applied for the definition of the shoreline. The previous land area of more than 35 kilometers wide has been submerged by a rise in sea level during the period from 1977 to 1995 (18 years) to the south-southeast of Atyrau. (2)The bathymetric map in the very shallow water area was generated through drawing submarine topographic profiles, and thus the complicated submarine topography was obtained in detail. The land cover units were added to this bathymetric map and finally four zones are highlighted on the seismic base map as those to be the source and receiver. (3)It is very difficult to conduct a seismic survey in a reed zone along the shoreline without deforestation. On the other hand, a shallow water-seismic survey can be conducted in the offshore area, on the basis that an appropriate method of measurement is selected according to the depth of water.

Acknowledgements
The authors are grateful to Mr. Yuichi MARUYAMA, General Manager of the Department of Research and Development of the Earth Remote Sensing Data Analysis Center, for permission to present this paper. They also are much indebted to Dr. Oleg A. FEDORENKO, Yuggeo Co. Ltd. who kindly provided technical support during the field verification survey and Dr. Shunji SATO, General Manager of the Exploration Department of Japan Energy Development Co., Ltd., for critical reading of the manuscript.

Selected References

• Nakayama, Y., 1997: Monitoring Changes of Lakes in Central Asia by Satellite Data. Proceedings of the Nihon University International Symposium on Global Environmental and Human Living, p.271-283.
• Sydykov, J. S., Golubsov, V. V. and Kuandykov, B. M., 1995: Caspian Sea and its Coastal Zone. National Scientific Academy of Republic of Kazakhstan, Institute of Hydrogeology and Hydrophysics, Ministry of Science and Technology, Government Company “Kazakhstan-Caspi-Shelf”, 211 pp. (in Russian).