Similarities in basal sliding between Greenland and Alpine Glaciers

Basal sliding speed has been observed to vary on synoptic time-scales in marginal portions of the Greenland Ice Sheet. In this study, we document seasonal variations in basal sliding velocity in Western Greenland, along the terminal portion of the flowline extending from the terminus of Dead Glacier (69.37°N, 50.28°W) to the main ice divide (71.54°N, 37.81°W). GPS observations from four stations along the terminal 100 km of this flowline show a strong annual cycle in ice velocities. We interpret the annual velocity cycle to represent seasonal variations in basal sliding speed. Observations reveal that the glacier moves at winter speeds until a summer acceleration that coincides with the onset of melt (around Julian Day 120). The summer period of ~30-80% enhanced velocities lasts ~50 days and is immediately followed by deceleration to an annual minimum, ~10-20% below winter velocities, reached around JD 250. Ice velocities increase again to winter speeds by JD 300. We speculate that this Greenland Ice Sheet flowline responds to the seasonal melt input history in a manner akin to the “spring speedup” events on alpine glaciers, where meltwater-induced pressurization of the subglacial system is followed by deceleration due to the establishment of efficient drainage and decreasing meltwater input. We can efficiently characterize the annual velocity cycle using two Gaussian curves superposed on the mean winter velocity (one representing summer high speeds and the other representing fall low speeds). This representation of the motion history is dependent on five observable parameters that vary from site to site (maximum, minimum and winter velocities, as well as Julian Day of maximum and minimum velocity). Based upon a projection of the spatial pattern of dates of maximum and minimum speeds, and hence of the duration of summer anomalous behavior, seasonal variations in basal sliding should not exist above approximately 87 km upstream from the terminus of Dead Glacier. The three stations downglacier of this point exhibit annual velocity variations, while the one station upglacier of this point does not. Our curve fits to this dataset allow seasonal basal sliding velocity to be parameterized as a function of distance along flowline and Julian Date. The sensitivity of both the magnitude of seasonal basal sliding, and the duration of the sliding season to distance along flowline suggests that a small inland migration in the location of basal sliding onset should result in a non-linear increase in annual total ice discharge to the ice sheet margin.