Amundsen Sea Embayment Project (ASEP)
Science and Implementation Plan
MODIS image of Pine Island Bay (center) with Pine Island Glacier (upper center) and Thwaites Glacier (lower center)
The portion of West Antarctic ice sheet draining into the Amundsen Sea is melting, thinning, and retreating rapidly, contributing to global sea-level rise. These are among the largest documented changes in Antarctica. The pressing need to understand these changes and to predict their effect on global sea level and the surrounding environment can be addressed through the research in this Science and Implementation Plan, which was developed within open community science meetings over the last few years.
Satellite data have permitted reconnaissance of this area with unprecedented detail, but specific airborne, shipborne and surface field studies are essential to supply the necessary additional information to understand the cause of present changes and for accurate prediction. These field studies will be guided by the satellite data and can be conducted efficiently using research techniques refined in the West Antarctic Ice Sheet (WAIS) program.
The prediction will be made using advanced numerical models of ice flow now under development, that for the first time fully incorporate the critical flow dynamics identified during WAIS. Data collection will be guided by the needs of these models for an accurate specification of the character and history of the modern ice sheet and how it interacts with its surroundings. Elevation and surface-velocity data will be available from a number of existing and planned satellite missions, but additional data are required including: ice thickness and internal layer data from airborne radar; oceanic circulation on the continental shelf and under the numerous floating ice shelves; sea-floor morphology and sediments in the surrounding seas; distributed field measurements of accumulation, velocity and thickness change; seismic and radar probing of critical bed areas; direct sampling of the ice bed and ocean via boreholes; a geologic record of post-glacial thinning; a network of automatic weather stations; and the magnetic and gravitational potential fields of the region.
A field and funding schedule to achieve this goal in five years is attached. Three seasons of fieldwork are required followed by two years of continued data analysis and model assimilation. Total cost of the program is $16.6 million.
The West Antarctic Ice Sheet is the only marine ice sheet remaining from the last glacial period. The majority of the bed is below sea level and the majority slopes downward into the interior. It has been hypothesized that this configuration implies that the ice sheet may be susceptible to run-away grounding line retreat [Weertman, 1974] leading to rapid collapse. Were this to occur, the water would raise global sea level by 5–6 meters. The multidisciplinary West Antarctic Ice Sheet (WAIS) program has been addressing this concern for the past ten years through numerous focused studies on, within and underneath the ice sheet draining into the Ross Ice Shelf [Bindschadler et al., 1999], on surrounding mountain outcrops and along the floor of the adjacent Ross Sea. A clear picture of ice-stream behavior is emerging along with a detailed history of past flow in this sector. Unsteady flow has been identified in many areas, but a rapid collapse does not appear to be underway in the Ross Sea sector. Similarly, a European initiative, the Filchner-Ronne Ice Shelf Program (FRISP) has investigated the sector of the West Antarctic Ice Sheet that flows into Ronne Ice Shelf, but not found evidence for collapse in that area either.
The third and final sector of the West Antarctic Ice Sheet is that flowing into the Amundsen Sea. So far, this sector has been only rarely visited and is relatively poorly understood. The WAIS program has, however, fostered reconnaissance studies in this sector, especially over the largest glaciers, Pine Island Glacier (PIG) and Thwaites Glacier (TG). Remoteness and difficult weather conditions both onshore and offshore have limited the study of this area for decades, but interest has persisted since Hughes  identified the Amundsen Sea drainage as the most-likely site for the initiation of ice-sheet collapse. Recent satellite-based studies and research cruises of the N.B. Palmer showed that this area is undergoing major changes. These discoveries dictate that a coordinated, multidisciplinary effort be initiated in this region to assess the present trends and to predict the likely future of the ice sheet draining into the Amundsen Sea. Such a study will benefit from the successful conduct of multidisciplinary research of WAIS (http://igloo.gsfc.nasa.gov/wais), European research conducted by FRISP, coordinated research in Greenland (http://cires.colorado.edu/parca.html), as well as from the plethora of research techniques developed during these activities. A series of open community meetings coordinated with the annual WAIS meetings and extensive discussions among concerned researchers between those meetings led to this Science and Implementation Plan. This plan links directly with the field research planned for this area by the British Antarctic Survey (described in Section §6.2).
2. Overview of the Amundsen Sea Embayment
That portion of the West Antarctic ice sheet discharging into the Amundsen Sea, termed here the Amundsen Sea Embayment (ASE), covers 320,000 km2, roughly 20% of the area and 20% of the total volume of the West Antarctic ice sheet (Figure 1). The ice discharge rate is much higher than in either the Ross ice streams or those feeding the Ronne Ice Shelf—the two other major West Antarctic basins. The high ASE discharge is driven by a more-vigorous snow accumulation, more rapid ice flow and more intense sub-ice-shelf melting than is usual for Antarctica. Storms circling the Antarctic continent are steered landward across the Amundsen and Bellingshausen Seas by the blocking effect of the Antarctic Peninsula. The relatively low elevations of inland West Antarctica compared with interior East Antarctica allow larger amounts of precipitation to spread farther from the coast, reaching the entire ASE catchment. These accumulation rates are some of the highest in the continent (comparable to those of the Antarctic Peninsula) suggesting that the glaciers would respond more quickly to changes in precipitation, storm tracks, and sea-ice cover.
Figure 1. Surface elevation and outlines of catchment basins in the Amundsen Sea Embayment sector of West Antarctica (from Vaughan et al., 2001).
This snowfall supplies two of the fastest glaciers in the world, Pine Island Glacier (PIG) and Thwaites Glacier (TG). The PIG and TG catchments extend 600 km inland from the coast and consist of a number of smaller tributaries that coalesce 70–100km from the coast into the well-defined outlet glaciers (Figure 2). The catchment basins are deep, approaching 2000 meters below sea level in the Byrd Basin (Figure 3). Together, these two glaciers account for approximately 5% of the ice discharge of the entire Antarctic Ice Sheet [Vaughan et al., 1999].
Figure 2. Flow convergence in the Amundsen Sea sector of West Antarctica represented by the darkness of individual cells. Tributary structures is evident for PIG and TG (from Vaughan et al., 2001).
The mechanism controlling the flow of these glaciers probably differs from that controlling the Ross ice streams and, perhaps, the Ronne-Filchner ice streams. Flow speeds at the transition from tributary to glacier are approximately 1 km/a rising to 2 km/a at the grounding line and rising still higher to 2.5 km/a where the floating ice shelf meets relatively warm ocean water. The driving stress in the tributaries is approximately 50 kPa, rising to more than 100 kPa near the grounding line, leading Vaughan et al.  to suggest that the tributaries are akin to ice streams and the main glacier akin to an East Antarctic outlet glacier.
Figure 3. Bed elevation of the Amundsen Sea sector of West Antarctica (from Vaughan et al., 2001).
When the glaciers begin to float, warm ocean water is able to reach the undersides of the floating glacier tongues and cause melt at rates larger than anywhere else yet found around the Antarctic coast. The reasons for this distinction are not clear, but the close proximity of the Antarctic Circumpolar Current to the continental shelf and diminished rates of sea ice production may be responsible. The result is that exceptionally intense subglacial melting occurs throughout this area, suggesting the glaciers may be strongly influenced by oceanic processes. Large icebergs are created at the ends of the rapidly moving glaciers.
Adjacent to the PIG and TG catchments is the broad coastal catchment that feeds Getz Ice Shelf. It, too, is likely fed by high accumulation rates, but it is much thinner, rests on several large islands and has many openings to the ocean complicating the sub-shelf circulation. Discharge of Getz Ice Shelf is less defined by fast ice flow due to the lack of convergent flow within the catchment. A significant recent retreat of this ice shelf suggests it may be experiencing a similar incursion of warmer water as the floating tongues of PIG and TG. Thus, the temporal response of Getz Ice Shelf may be important in understanding the current behavior of the ASE sector.
The subglacial topography and character of the Amundsen Sea Embayment are poorly known. Traverse data, using either seismic or radar methods, and widely spaced radar-sounding flights provide the only information. Figure 4 shows the radar-souding coverage. The most recent survey from Siple Station showed complex subglacial character (roughness, internal layering, and topography) along a line that crosses from inland ice onto a tributary and then onto the main part of PIG [Vaughan et al., 2001]. This tributary appears to sit in a well-defined trough that extends down into the main trunk of PIG and has a substantially smoother bed than that of the inland ice. However, a large sill or shelf at the grounding line forces a large increase in driving stress as the glacier approaches the coast. Clearly, the geology of this region plays an important role in the configuration of the PIG/TG trunks and tributaries.
Figure 4. Radio-echo sounding coverage of the Amundsen Sea sector of West Antarctica (from Vaughan et al., 2001). Widely spaced dots are surface traverse data.