Evidence of Lake Ellsworth
The following eight lines of evidence collectively prove the existence of Lake Ellsworth. [Detailed explanation of this evidence is available in Siegert et al. 2004].
(1) Radio-echo sounding (Figure 1a) reveals a strong, flat, specula return that could only come from an interface with a significant dielectric contrast, which has no roughness at the scale of the radio-wave (Figure 1b,c). Such an interface is best explained by the ice-water surface of a subglacial lake (Siegert et al., 1996).
(2) Small scale changes in the slope of the reflector are apparent, which are similar to those observed in Lake Vostok and are symptomatic of a lake beneath (Figure 1d,e).
(3) The ratio of the slopes of the ice-water and ice-surface interfaces, averaged over the lake, is at the amount required for the lake to be in hydrostatic equilibrium with the ice sheet. This, in combination with the point above, means that our target lake passes every criterion used previously to identify subglacial lakes.
(4) The slope of the ice-water interface is also noticeably less than the surrounding topography, inferring strongly that the water depth of the lake is several tens, possibly hundreds, of metres.
(5) Internal ice sheet layers, observed above the lake, converge in vertical section with the lake surface (Figure 1a). Ice flow modelling that we have undertaken shows this is only possible if the ice base above the lake is melting at a rate of several cm per year. Hence, the internal layer pattern is consistent with the production of subglacial meltwater.
(6) Thermo-mechanical ice-sheet models of West Antarctica predict that the bulk of the ice base is at the pressure melting point. Such models certainly expect melting beneath 3.4 km of ice, which is the ice thickness above our target lake.
(7) Thermodynamic calculations reveal the heat required for subglacial melting above Lake Ellsworth now and at the last glacial maximum (LGM) is 54 and 46 mW m-2, respectively. The basal heat flux calculated in West Antarctica from temperature gradients in boreholes is ~70 mW m-2. Hence, basal heating at the level expected is sufficient for the subglacial lake to exist both now and at the LGM.
(8) Regional radio-echo sounding shows that our lake occupies a distinct subglacial bed depression; a topographic requirement for subglacial water to pond.
"Figure 1. (a) Raw RES transect. The subglacial lake is located between 28 and 38 km. These RES data were acquired using a depth-differentiated system, in which the signal was amplified in proportion to two-way travel time. Note that, despite this amplification, the basal signal is either very weak or absent either side of the subglacial lake. (b) Magnified section of RES data over the central region of Lake Ellsworth (located by the box in (a)). These data are in non-differentiated form. (c) Further magnification of RES data shown in (b), which demonstrates the specula, flat and smooth nature of the reflector, and its steady along-track strength. (d and e) The slopes of the ice-water interfaces over the upstream and downstream sides of (d) the Ellsworth subglacial lake, (e) the southern end of Lake Vostok (Vostok Station is denoted by the surface parabola). These transects demonstrate the expected shape of the lake-surface, which is diagnostic of the lake beneath."