The need for reinterpretation of seismic data

Part of the ERS-1 satellite mission included a precise laser altimetry of the Earth's surface. The orbital inclination of the satellite covered a sufficient part of the Antarctic Ice Sheet. Ridley et al. (1993) collated the laser altimetry data on the ice sheet's surface and mathematically summarized it from many different orbits. This gave them the ability to plot a very precise map of the Antarctic Ice Sheet's surface with an accuracy of 0.5 m within an area of 10 km2. Equal elevation lines on this map showed without doubt that there is a flat, nearly horizontal surface of a floating ice cover just above the area marked by Dr. Robin's bottom water reflections of 1971-1972 and 1974-1975. This ice cover differs distinctly from the slopes of a grounded ice sheet. The distinct shores of the lake are marked by a line of small depressions at the upward side and lines of gentle hills at the downward "banks" of the lake.

As a supplement to his work, Ridley et al. (1993) took the map of equal elevation lines for Antarctica and computed an isometric view, imitating a side look from a low-flying aircraft, with shading corresponding to the actual position of the low-angle Sun. It was suddenly clear (Figure 5.1) that there is a large oval-shaped lake more than 200 km long and about 50 km wide near Vostok Station, just south-south-west of it.

At about this time the Royal Society in London and the U.S.S.R. Academy of Sciences signed an agreement to start a program of grant awards. On the basis of this agreement, a Soviet scientist could apply for a grant from the Royal Society to cover expenses in the U.K. for a specified period of time on the stipulation that a prominent member of the Scott Polar Research Institute (SPRI) or any other British Institute would offer an invitation. I expressed my interest to Dr. Robin to work at SPRI on subglacial lake issues, and he replied positively, authorizing a 6-month stay.

The combination of Ridley's article and his map showing a distinct oval shape suggesting a "lake", my arrival at SPRI, and accumulated data provided a good

Figure 5.1. An isometric plot of Antarctica compiled from the ERS-1 satellite radar-altimeter data. Details of the surface topography are highlighted, including Lake Vostok (adapted from Ridley et al., 1993).

reason for Dr. Robin to organize a one-day workshop in Cambridge to discuss the scientific problems associated with the existence of the lake. There remained doubts as to whether the lake was real (i.e., whether it represented a reasonably deep layer of liquid water). Some related questions needed to be addressed. American scientists working on ice streams in West Antarctica found them to resemble "rivers" of relatively fast-moving ice compared with the rate of flow of the surrounding ice sheet. Reflections obtained by radio-echo sounding of the bottom of some of these ice streams showed flat features that resembled water. However, deep drilling to the bottom showed no open water existed at these sites, but instead they found a layer of till, fine moraine-type particles saturated with water melted at the bottom as a result of heat produced by fast-moving ice streams (Blankenship et al., 1986).

The lake below Vostok Station was located more than 100 km from the ice divide, with ice arriving at the lake at a rate of about 3myr_1 (an estimate arrived at from simple calculations). Experimental measurements at Vostok Station gave a similar rate. This is not particularly fast, but it means that the ice sheet continuously produces some kind of morainal debris at the ice/rock interface, which becomes trapped by moving ice, is brought to the lake, and is deposited (Figure 5.2). There should be no question that the ice sheet would have been the cause of this activity - the big unknown was, "How much material had been transported and for how long a time period?"

In preparation for the one-day workshop, we estimated how deep the lake water would be if the glacier did not bring debris into the lake. We took reels of bottom

Figure 5.2. Schematic representation of a subglacial lake in the central part of a thick, Antarctic-type ice sheet (adapted from Zotikov, 1986). (1) Depression in subglacial bedrock (lake reservoir); (2) bedrock; (3) thickness of ice sheet; (4) particles carried along by the ice sheet near the bedrock interface; (5) upper surface of the ice sheet; (6) "lake" boundary projected onto the glacier surface; (7) upper surface of the ice sheet above the lake (see Chapter 4); (8) ice flowlines; (9) subglacial lake; and (10) sedimentary layer at the lake bottom.

Figure 5.2. Schematic representation of a subglacial lake in the central part of a thick, Antarctic-type ice sheet (adapted from Zotikov, 1986). (1) Depression in subglacial bedrock (lake reservoir); (2) bedrock; (3) thickness of ice sheet; (4) particles carried along by the ice sheet near the bedrock interface; (5) upper surface of the ice sheet; (6) "lake" boundary projected onto the glacier surface; (7) upper surface of the ice sheet above the lake (see Chapter 4); (8) ice flowlines; (9) subglacial lake; and (10) sedimentary layer at the lake bottom.

reflections of Dr. Robin's flights from the 1971-1972 and 1974-1975 periods of sounding. The average roughness of a "dry" glacier bottom of East Antarctica for a 200-km section of ice (the approximate length of the lake) in terms of the elevation difference of the highest and lowest parts of the lake bottom was estimated to be about 500 m.

Other studies have shown that the ice sheet moving across the lake could bring about 100 m of debris at the upward bank of the lake with less at the opposite bank. Although the estimates are very rough, they indicate that the lake is probably not filled completely, leaving space for liquid water. These ideas formed the level of our understanding for our "Workshop on Geophysical Studies of 'Lake Vostok'''. This was also the first time that the lake was given that particular name - it was a convenient way to refer to it. No one objected to this name from the collective scientific body familiar with the ''lake'', and it was also considered appropriate because the lake was linked with Vostok Station. I also considered the name ''Lake Robinson'', after the pilot who discovered the surface indication of the lake and who died soon after while on another polar assignment. His father was English and his mother Russian, and in some way he symbolized close British-Russian cooperation regarding the study of the lake. Dr. Robin did not agree, however, and convinced me that ''Lake Vostok'' was more appropriate. As of the end of 2004 the name had not been approved by any official geographic body.

From the workshop report that was authored by Dr. Robin and me, the following statements describe Lake Vostok.

The workshop reviewed existing knowledge and suggested further studies to enlarge our understanding of 'the remotest lake on Earth' ... The lake has an area of around 10,000 km2, about one-third the area of Lake Baikal in Asia. It lies beneath about 4 km of ice to the north of Vostok Station. The depth of water layer is unknown but is likely to average about some tens to a few hundred meters.

The workshop was organized by the SPRI of Cambridge University and held at a building of the British Antarctic Survey (BAS). Attendance at the workshop included the following: G. Robin and P. Clarkson (SPRI); I. Zotikov and A. Kapitsa (Russians working at SPRI); J. Ridley (Millard Space Laboratory); and C. Doake, D. Vaughan, B. Storey, A. M. Smith, E. King, and D. J. Drewry (British Antarctic Survey). The agenda for the workshop contained five major items: (1) data review; (2) geophysical interpretation of data; (3) future programs; (4) long-term objectives; and (5) conclusions. Summaries of these items are as follows:

(1) Data review. Soviet aircraft navigators first reported in 1959 that oval depressions with gently inclined boundaries, or "shores", were among the natural landmarks on the ice sheet, although they were not located on maps (Robinson, 1960). Extensive studies of the sub-ice relief by airborne radioecho soundings made by the National Science Foundation-Scott Polar Research Institute-Technical University of Denmark (NSF-SPRI-TUD) program found occasional strong basal reflections over a few kilometers of the flight lines. These indicated specular reflection from a sub-ice-water interface, with a water depth of not less than one meter. The first was reported near Sovetskaia Station in the 1967-1968 season. Seventeen "sub-ice lakes'' were reported after the 1971-1972 season when their distributions and other properties were discussed in more detail (Oswald and Robin, 1973). During the 19741975 season, similar reflections, shown as heavy black bands in Figure 5.3, were noted along flight lines. The presence of a large "sub-ice lake'' was also reported in Robin et al. (1977). The overlying ice thicknesses ranged from 3,600-4,200 m. The relatively level ice surface over the sub-ice lake was noted. Digital data on the ice thickness and surface elevations from all NSF-SPRI-TUD flights are now stored on computer at the BAS. Copies of any of the relevant sections are supplied on request by D. Vaughan (BAS), with the original film records on flight data archived at the SPRI.

(2) Geophysical interpretation of data. Survey mapping of the surface elevation of the Antarctic Ice Sheet by satellite altimetry from ERS-1 has provided detailed information of surface contours as far south as 82°S. Initial mapping by preliminary or "fast-delivery" data, was used to provide the surface contours in Figure 5.3. About 2,700 surface elevations were used, with mean errors of ice sheet elevations at about 2 m. These errors drop to 0.2 m over the flat central area of the lake. The errors relate to local topography. In addition, satellite orbit errors could be plus or minus 15 m. The most interesting discussion at the workshop was on the possible depth of water in the lake. All our data from theoretical estimates and Robin's radio-echo sounding indicate that there could be a real lake there, representing one of the great geographical discoveries of the

Figure 5.3. Map showing the region of Lake Vostok as it was seen in 1993. Surface elevation lines are reconstructed by Ridley et al. (1993) on the basis of low bit rate (LBR) data from the ERS-1 satellite. Also shown are the radio-echo sounding (RES) flight paths and the location of strong bottom "lake-like" echoes from the radio-echo sounding by Robin et al. (1977) (adapted from Ridley et al., 1993).

Figure 5.3. Map showing the region of Lake Vostok as it was seen in 1993. Surface elevation lines are reconstructed by Ridley et al. (1993) on the basis of low bit rate (LBR) data from the ERS-1 satellite. Also shown are the radio-echo sounding (RES) flight paths and the location of strong bottom "lake-like" echoes from the radio-echo sounding by Robin et al. (1977) (adapted from Ridley et al., 1993).

end of the 20th century. However, although Robin's data show a lake under Vostok Station, Kapitsa's seismic data shows no lake - with more confidence attributed to seismic data than radio-echo sounding data. Skeptics wondered if seismic data for other regions, which Robin interpreted as lakes from his radioecho sounding, would be negative. This uncertainty was also expressed by the editors of Nature, reflected in their rejection of a manuscript by Robin in which he first tried to publish a paper on the discovery of a lake using the results of

Ridley's glacial erosion/deposition estimates. Nature's editors said they had already published a paper on subglacial lakes by the same author several years previously, and they would publish new material only if it included at least one independent piece of evidence of the lake's existence (e.g., water thickness data from seismic work). However, this additional evidence appeared to be unobtainable. The answer appeared to lie with seismic work at Vostok Station by Kapitsa (1968), as outlined in the Workshop report:

Measurements of ice thickness by seismic reflection shooting were made at Vostok Station and three locations along and near the west edge of the lake by Kapitsa (1968). He interpreted a second, deeper echo at Vostok as coming from sedimentary layering, but the present information suggests that it could come from the base of a water layer around 70 m thick. Over some 50 km from where the other soundings were obtained, the snow surface was extremely soft and made travel difficult. This could result from the flat surface causing snow deposition in light air in contrast to almost continuous winds over sloping surfaces.

Kapitsa was present at the workshop and agreed with our interpretation and was eager to check the data, resulting in the following statement:

... but the present information suggests that it could come from the base of the water layer around 70 m thick. This record and that of the three other soundings near the lake margin will be re-examined in Moscow.

(3) Future programs. The first priorities for future study are as follows: (a) it was expected that assessment of the more accurate data would be completed within a month or two; (b) radio-echo data should be reassessed, especially in regard to their accurate location and the surrounding topography; (c) seismic soundings in the region of the lake would be re-examined by Dr. Kapitsa and his staff in Moscow.

Second priorities included obtaining additional field data to be by standard methods during the next few years. (a) Determination of surface movement by satellite position monitoring at selected points with equipment placed by surface parties or dropped from aircraft. Alternatively, if the recently developed new technique of radar (synthetic aperture radar, SAR) interferometry from ERS-1 could be used to measure velocities, as on the Rutford Ice Stream, this would provide valuable and comprehensive data at a limited cost. Dr. Doake (BAS) was asked to see if this was possible. (b) Ice thickness determination: additional airborne radio-echo sounding is needed to provide better definition of the basal topography over the lake and surrounding bedrock. This should be undertaken as soon as possible. It should also be used to determine whether the small lake segments near Vostok (Figure 5.3) are continuous with the large lake. (c) Water layer thickness: seismic reflection shooting measurements appear to be the best method available. Wide-angle reflection measurements may also provide information on the nature and thickness of basal sediments and bedrock. (d) Long-term monitoring from within the present hole at Vostok. When access to the hole is possible, its use for long-term monitoring (10-100 years) of geothermal heat flow and of microseismic and acoustic activity was suggested. This is of increasing value when the base of the borehole is approaching a sub-ice lake.

(4) Long-term objectives. The major long-term objective is to study the geophysics and biology of the lake system. While remote-sensing techniques already discussed can advance a general knowledge of the lake, direct sampling of chemical and biological properties and continuous monitoring of lake waters by instruments in situ must be a long-term objective. Extensive discussion of the best techniques of sampling is needed. Particular attention must be given to avoiding contamination of lake waters through the introduction of foreign material such as drilling fluids used in deep boring. Before the existing borehole at Vostok can be deepened to the lake level, a solution to this problem must be found, otherwise alternative drilling and sampling techniques need to be developed.

(5) Conclusion. The existence of the remotest lake on Earth, near Vostok Station, appears close to being confirmed by geophysical techniques. However, we require a better knowledge regarding the thickness of the water before we can be certain of our conclusions. The program recommended is first to ensure that the existing data is fully exploited (during the next year or two). Planning for further geophysical studies listed in (3) can commence during this period and may be expected to last between 2-5 years. At the same time consideration of the techniques to be used for sampling lake water and sediments, and other means of long-term monitoring, can take place with the aim of carrying out these investigations when associated engineering problems have been solved - possibly starting in 5-10 years.

(Excerpted from the Workshop Report by Robin and Zotikov, 29 November 1993.)

Was this article helpful?

0 0

Post a comment