The Nilsen Plateau ponds

In the process of updating the geology of the Nilsen Plateau for submission to the digital geological map of Antarctica (Cox et al. Reference Cox, Smith Lyttle, Elkind, Smith Siddoway, Morin and Capponi2023), a frozen pond surface was observed on one of the United States Geological Survey (USGS) airphotos (TMA 2190 33R 0075; Fig. 3). The field notebook of W.E. Long (compiled during the 1963–1964 field season) has recently become available at the US Polar Rock Repository, and in that notebook the presence of the frozen pond was noted at the base of section 7 (‘top’ in the field notebook; 7B in the stratigraphic columns in Elliot Reference Elliot2025), but no details were provided.

Figure 3. Location of the principal pond on United States Geological Survey (USGS) airphoto TMA 2190 33R 0075 (acquired on 30 January 1969).

The pond (Fig. 4), at 86°24´S, 159°30´W (86.396°S, 159.493°W), lies at an elevation of 2350 m based on a 25 m contour interval on the Reference Elevation Model of Antarctica (REMA; Polar Geospatial Center (PGC), University of Minnesota; pgc.umn.edu; Howat et al. Reference Howat, Porter, Smith, Noh and Morin2019). When observed by W.E. Long, the pond (Fig. 4) appeared to be shallow and permanently frozen. The pond, which must be fed by seasonal snowmelt, overlies surficial or morainal debris.

Figure 4. Photograph of the higher-elevation pond taken during the 1963–1964 field season by W.E. Long. Mount Hassel (2350 m), 35 km to the west across the Amundsen Glacier, forms the summit of the distant hills.

Within Fig. 5 (satellite images), the red dotted oval in Fig. 5’s bottom-left panel marks the approximate extent of the pond as seen in Fig. 4. The scale allows an approximate estimate of the dimensions of the pond, which is no more than ~20 × 50 m. The northern (left in Fig. 3) small, snow-covered area might also be a frozen pond. Examination of the available satellite imagery, which spans from December 2010 to December 2024, yielded only a limited number of images clearly showing the pond, mainly because of cloud cover or deep shadows. The 20 images showing the pond suggest that the apparent ‘shoreline’ lying to the right (south) in Fig. 3 is an artefact of the snow cover of the hollow in which it lies, with the apparent margins varying between one image and the next (Fig. 5).

Figure 5. Satellite images of the higher-elevation pond. Dates acquired and scales are on each image. Red arrows point to the frozen pond.

How long the pond has existed remains speculative. Its apparent shallowness suggests that this probably depends on the climate trends over time. The pond is located on a west-facing bedrock platform covered by surficial debris, and it lies at ~450 m above the current ice level. For the Dry Valleys region of south Victoria Land, several hundred kilometres to the north, Dickinson et al. (Reference Dickinson, Schiller, Ditchburn, Graham and Zondervan2012) estimated a rate of erosion/denudation of 1–6 cm/My for the last 6 my. The platform on which the pond resides may have existed since ca. 12–14 Ma, when conditions changed from warm-based glaciation to cold-based glaciation (Barrett Reference Barrett1996, McKay et al. Reference McKay, Barrett, Levy, Naish, Golledge and Pyne2015), and this suggests that the pond could be extremely old.

The photographs taken by W.E. Long include another small, frozen pond (Figs 6 & 7). Its location cannot be identified with certainty, although it must lie south-east of ‘Windy Gap’ (unofficial name). Judging from the photographs, it lies at approximately the same elevation as the base of ‘Windy Gap’ itself (~2000 m, REMA elevation).

Regional context

In the more sheltered places along the Transantarctic Mountains, surface melting and associated running water are not uncommon, with the Onyx River in the McMurdo Dry Valleys being the largest example (Chinn & Mason Reference Chinn and Mason2016). Other streams occur in south Victoria Land (USGS topographic map Ross Island and Vicinity). Evidence of running water was observed at elevations up to 1800 m at Allan Hills in December 2002 (D.H. Elliot, unpublished data 2002). Farther south, in December 1991, at Mount Falla (84°22´S), water was observed trickling down a north-facing slope of sandstone and finer-grained sedimentary rocks at ~3000 m elevation (D.H. Elliot, unpublished data 1991). Local weather conditions (cloud cover, temperature, wind) and topographic setting are significant factors in whether snow is lost by melting rather than ablation.

The nearest surficial lakes and ponds to those at the Nilsen Plateau occur in Moraine Canyon (Fig. 2) a short distance to the north at an elevation of ~1400 m (REMA elevation), and these appear to overlie glacial debris. Three lakes or ponds are indicated on the USGS Nilsen Plateau topographic sheet, at ~86°9´S, 157°35´W, but these are difficult to distinguish on the USGS airphotos (TMA 789 F33 282; TMA 1200 F31 176).

In the La Gorce Mountains, ~90 km to the south-east in the Scott Glacier region, numerous small, shallow ponds were found on an ice-cored moraine (Broady & Weinstein Reference Broady and Weinstein1998). This locality is slightly farther south (~25 km) and at a lower elevation (less than 2000 m).

To the north-west, the USGS Liv Glacier topographic sheet shows many lakes along the foothills adjacent to the Ross Ice Shelf and between the Shackleton and Liv glaciers at ~84°35´S. Most are probably meltwater lakes on ice (e.g. at Sage Nunatak, 84.533°S, 172.917°W; prr.osu.edu/collection/object/158347). Aerial photography provides no definitive information on this matter. The nearest recorded and examined ponds that occupy a glacially carved landscape occur in the Mount Heekin region at ~85°05´S (Elliot et al. Reference Elliot, Collinson and Green1996) and adjacent to the lower Shackleton Glacier at ~84°55´S (Diaz et al. Reference Diaz, Gardner, Elliot, Adams and Lyons2023). Both are ~200 km to the north-west of the Nilsen Plateau. The elevations are much lower (below ~1000 m) except for one frozen pond north-west of Mount Heekin at 1250 m (REMA elevation).

Figure 6. Photograph of the lower-elevation pond south-east of ‘Windy Gap’ taken by W.E. Long during the 1963–1964 field season.

Figure 7. Satellite image (acquired on 20 November 2020) of the area within which the lower-elevation pond occurs. The blue arrow points in the direction of ‘Windy Gap’ (Fig. 2). The red arrow points to the best estimate of where the pond lies.

Documented lakes and ponds occur in the Transantarctic Mountains in the Dufek Massif (Hodgson et al. Reference Hodgson, Convey, Verleyen, Vyverman, McInnes and Sands2010) and from the Scott Glacier to the Terra Nova Bay region of north Victoria Land (e.g. Priscu 1988, Andreoli et al. Reference Andreoli, Scarabel, Spini and Grassi1992, Vincent & Howard-Williams Reference Vincent and Howard-Williams1994, Broady & Weinstein Reference Broady and Weinstein1998).

These frozen ponds at the Nilsen Plateau and other small ponds, particularly if they are perennially frozen and too small to have been recorded by Hawes et al. (Reference Hawes, Howard-Williams, Gilbert, Hughes, Convey and Quesada2023) and Tóth et al. (Reference Tóth, Terauds, Chown, Hughes, Convey and Hodgson2025), may be important in the future because of the ongoing and well-documented effects of climate change in Antarctica.