Paleofluvial Mega-Canyon Beneath the Central Greenland Ice Sheet

See allHide authors and affiliations

Science  30 Aug 2013:
Vol. 341, Issue 6149, pp. 997-999
DOI: 10.1126/science.1239794

Ice Lubricant

The Greenland and Antarctic ice sheets both possess hydrological systems that allow water accumulating from the melting of surface ice to be transported to the base of the ice sheet. If that water, when it reaches the ice-bedrock interface, is distributed over large areas, it will lubricate rapid ice sheet flow toward the sea. Bamber et al. (p. 997) report the existence of a large, 750-km-long subglacial canyon in northern Greenland, which may act as a channel for the transport of basal meltwater to the margin of the ice sheet and thus influence overall ice sheet dynamics.


Subglacial topography plays an important role in modulating the distribution and flow of basal water. Where topography predates ice sheet inception, it can also reveal insights into former tectonic and geomorphological processes. Although such associations are known in Antarctica, little consideration has been given to them in Greenland, partly because much of the ice sheet bed is thought to be relatively flat and smooth. Here, we present evidence from ice-penetrating radar data for a 750-km-long subglacial canyon in northern Greenland that is likely to have influenced basal water flow from the ice sheet interior to the margin. We suggest that the mega-canyon predates ice sheet inception and will have influenced basal hydrology in Greenland over past glacial cycles.

Greenland is an ancient craton dominated by crystalline rocks of the Precambrian shield, composed largely of Archaean [3100 to 2600 million years (My) before the present (B.P.)] and Proterozoic (2000 to 1750 My B.P.) formations, with more limited sedimentary deposits along parts of the continental margins (1). These rocks are resistant to glacial erosion compared with the sedimentary deposits that are thought to underlie substantial sectors of Antarctica (2, 3). Although there is evidence for some glaciation dating back to 38 My B.P. (4), the island has only been extensively ice-covered for no more than ~3.5 My (5), as compared with the Antarctic ice sheet that first grew at ~34 My B.P., and has existed persistently since ~14 My B.P. There are, therefore, important geologic and glacio-morphological differences between the subglacial environments of the two ice sheets.

Ice-penetrating radar (IPR) is capable of imaging the ice-sheet base and, if collected as a series of transects, can reveal the landscape beneath the ice. In Greenland, such data have been acquired on numerous occasions over the past 40 years (6, 7). Using these data in combination with ice-surface topography measurements, a quasilinear feature was previously identified in northern Greenland from a combination of ice-surface topography and IPR data, but with limited knowledge on its extent and configuration (8). By inspecting IPR-derived basal topography from these and more recent data sources, we have discovered this bed feature to be a major subglacial canyon that extends from almost as far south as Summit in central Greenland to the northern margin of the ice sheet, terminating in the fjord that drains Petermann Gletscher (Fig. 1). The extent of the canyon is at least ~750 km, but its southern limit may continue further than we are able to identify because the density of IPR tracks below 74.3° N is limited. However, two tracks at 73.8° N, 42° W cross an over-deepening that appears likely to be the southern extension of the canyon. There is no obvious evidence of a geological boundary in this part of Greenland (9). The southern limit of the canyon, however, does lie close to a minimum in the free-air gravity anomaly and a maximum in the terrain-correlated gravity anomaly, which indicates crust that is isostatically thin and subject to compressive inflow of crustal material (9), which could form an explanation for its genesis.

Fig. 1 Bed elevation [between –900 and 900 m above sea level (asl)] for northern Greenland.

The area plotted is indicated by the red box in the inset. Airborne ice penetrating radar flight lines are shown in gray, and the three bed profiles plotted in Fig. 2 are shown by the solid blue (Fig. 2A), black (Fig. 2B), and red (Fig. 2C) lines. The mauve line shows the approximate location of the grounding line on Petermann Gletscher. Other places discussed in the text are also labeled: NGRIP, the onset area of the North East Greenland Ice Stream (NEGIS), and the camp near the highest point of the ice sheet, Summit. The red ellipse marks the approximate location of a marked minimum in crustal thickness, and hence maximum in geothermal heat flux, identified from gravity data (9).

The IPR data indicate that the canyon has a depth and width of up to ~800 m and ~10 km, respectively, toward the coast and a slightly reduced width and substantially shallower depth of ~200 m further south (Fig. 2 and figs. S1 and S2). The valley width-to-height ratio (used as a measure of river channel morphology and formation) for the profile in Fig. 2C is 30, which is indicative of a broad, shallow channel, subject to limited stream erosion (10, 11). The ratio for the profile in Fig. 2A is ~0.5, suggesting linear stream incision was important in the development of the channel at this location, with no evidence of glacial erosion being the dominant process (12). Indeed, none of the profiles are typical of glacially eroded valleys (fig. S2). The canyon follows a meandering path more typical of a large river system. The dimensions of the Greenland canyon are comparable with parts of the Grand Canyon in the United States in terms of width and length and, in its deepest region, are about half of the Grand Canyon’s depth.

Fig. 2 Ice-penetrating radargram profiles across the canyon at three locations.

The locations are indicated in Fig. 1 by (A) the solid blue line, (B) the black line, and (C) the red line. There is an exaggeration in the vertical by a factor of 26. The depth of the canyon in meters below sea level is indicated in each profile alongside internal reflections within the ice and the bed return. North is roughly into the page.

To assess the canyon’s influence on the flow of subglacial water, and the extent to which it was a mega-channel for surface water, we calculated the hydraulic potential for both the present-day, ice-free, and last glacial maximum (LGM) configurations of the Greenland ice sheet (Fig. 3), accounting where necessary for isostatically compensated ice-free bedrock. We used the methods described in Shreve (1972) (13), which have been commonly applied to determine subglacial hydrological pathways (14). Because of its size, the canyon provides a hydraulic pathway in all cases tested, suggesting that it has influenced basal water flow in northern Greenland since ice sheet inception and, indeed, before this. For the ice-free (preglacial) state (Fig. 3A and fig. S3B), water is predicted to flow south-to-north along the majority of the canyon. The isostatically compensated channel gradient, without the ice sheet load, is 0.3 m km−1, which is similar to the southern 500 km of the Colorado River and around three times larger than the average for the Mississippi River. Regions where water diverts away from the canyon (such as at 79.3° N, 48° W) are coincident with limited IPR data coverage (Fig. 1), implying poor resolution of the bed topography in such places. Hence, we believe water will likely have routed continually along this feature before extensive ice sheet formation ~3.5 My B.P. (5).

Fig. 3 Subaerial and subglacial hydraulic potential flowpaths.

(A) An isostatically compensated bed without an ice sheet. (B) An ice sheet configuration at LGM. (C) The present-day ice sheet and bed configuration. H, Helheim; P, Petermann; R, Ryder. The bed elevations shown are all for present day. The isostatically compensated bed elevation is illustrated in fig. S3B.

In full glacial conditions [based on LGM (supplementary materials)], the flow of water is influenced by the ice overburden pressure, which forces water out of the canyon in parts of the upper Petermann catchment and along the direction of ice flow (Fig. 3B). Nonetheless, the canyon remains a conduit for the flow of basal water toward the coast even under LGM conditions, especially northward of ~79° N. For the present ice-sheet configuration, which is small and short-lived compared with its expanded full glacial state, steeper ice surface slopes (relative to LGM) force more water from the canyon to the west at certain locations (Fig. 3C and fig. S3A). In all cases, above ~76° N and within the entire length of the Petermann catchment, the canyon exerts a control on basal water flow. For ~200 km, it provides an uninterrupted hydraulic pathway (Fig. 3 and fig. S3A) that ends at the terminus of Petermann Gletscher. However, although extensive parts of northern Greenland have been identified as having a wet bed (fig. S4) (15), water currently may not be ubiquitous throughout its length. Under full glacial conditions, the ice sheet was larger and thicker than at present day (4), and therefore, a greater proportion of the bed was likely melting during the longer lasting LGM ice sheet configuration.

The lack of substantial subglacial lakes in the main Greenland ice sheet may be partly explained by generally steeper ice surface slopes as compared with those of Antarctica, but subglacial water must be evacuated by some means or it will pool. The subglacial canyon provides an efficient pathway for this water routing over an otherwise relatively flat bed (Fig. 3C). Given that water beneath former ice sheets in Greenland was likely to have been influenced by the canyon, it seems probable that conditions have never been well suited to substantial long-term basal water storage in this sector of Greenland.

The floating ice shelf in front of Petermann Gletscher has been found to possess sub–ice-shelf channels 1 to 2 km wide and 200 to 400 m deep incised upward into the ice (16). These channels are associated with high basal melt rates, exceeding 30 m year−1. Their existence has been attributed entirely to the action of the ocean (16). We believe the efficient routing of basal water via the canyon to the northern ice sheet margin is also important in explaining these features. Analysis of IPR echo-strength data has identified extensive regions of the northern half of the Greenland ice sheet that have water at the bed (15). Subglacial melt rates in the vicinity of the North Greenland Ice Core Project (NGRIP) drill site have been estimated to exceed 1 cm year−1 (17) and are as high as 15 cm year−1 near the onset of the Northeast Greenland Ice Stream (18). A study examining crustal thickness and geothermal heat flow in Greenland found a minimum in the former (and hence, a maximum in the latter) close to the northern limit of the canyon (Fig. 1) (9). Thus, substantial volumes of subglacial meltwater are generated at the bed, and additional surface meltwater may also penetrate to the bed near the coast (19). An effective means by which this basal water is routed to the ice-sheet margin must exist, else subglacial lakes would form. Basal water generated in the Petermann catchment is highly likely to be routed through the Canyon to the ice sheet terminus (Fig. 3C). Thus, basal water is likely to be delivered at a point source to the ice-shelf cavity, which is likely to influence the local subshelf water circulation (20). Similar sub-shelf channels have been identified underneath Pine Island glacier and elsewhere in Antarctica (21), and we suggest that well-organized subglacial meltwater flow is likely to play a similar role in the development of these features.

Supplementary Materials

Materials and Methods

Figs. S1 to S5

References (2230)

References and Notes

  1. Data set for (6) is available at
  2. The valley height-to-width ratio is Vf = 2Vfw/[(Eld – Esc) + (Erd – Esc)], where Vfw is the valley floor width, Esc is the valley floor elevation, Eld is the elevation of the left river divide, and Erd is elevation of the right river divide.
  3. Acknowledgments: J.L.B., J.A.G., and G.S. were supported by funding from the ice2sea program from the European Union 7th Framework Programme, grant 226375. This work is ice2sea contribution number 158. J.L.B. and M.J.S. were supported by Natural Environment Research Council grant NE/H021078/1. We acknowledge the use of data products from CReSIS generated with support from NSF grant ANT-0424589, NASA grant NNX10AT68G, and the NASA Operation IceBridge project. J.L.B. wrote the main text with substantial contributions from M.J.S. and input from all authors. J.A.G. carried out the hydraulic potential calculations. G.S. calculated the isostatically adjusted bed, and S.J.M. produced the LGM ice sheet configuration.
View Abstract

Navigate This Article