Skip to main content
U.S. flag

An official website of the United States government

Investigating the impacts of channelized subglacial hydrology on melt rates at the margin of Humboldt Glacier in northern Greenland

Presentation Date
Monday, December 14, 2020 at 4:00am



In the past two decades Humboldt Glacier (HG) in northern Greenland has experienced rapid retreat, the cause of which is not well understood. The recent ISMIP6 intercomparison project employed a submarine glacier front melting parameterization driven by ocean thermal forcing and glacier-averaged subglacial discharge that predicts mean melt rates of ≤0.5 m/day across the glacier front, indicating a limited role of submarine melting in the recent retreat. Here we assess the extent to which spatially-varying subglacial discharge, including from subglacial channels, could localize submarine melting in key areas. We model winter and summer subglacial discharge rates at HG using the subglacial hydrology component of the MPAS-Albany Land Ice (MALI) model, which is then used to calculate submarine melt rates using the ISMIP6 parameterization. Winter conditions are simulated by running the model to a steady state with only basal melting as an input to the subglacial hydrology model. From here, the summer configuration is run for 3 months, with MAR3.9 runoff (time-averaged between 1960 and 1989) representing summertime melt, assumed to drain directly to the bed. Submarine melt rates for summer and winter scenarios were calculated using ocean thermal forcing data from the HadGEM2 and MIROC5 models and the modeled water flux. End-of-summer submarine melt rates using our drainage model are ~10 ± 8 % lower than those calculated from ISMIP6 forcings, but our model’s melt rate is 24% higher than those calculated from ISMIP6 forcings at the location of maximum difference. Mean annual melt rates using our model are ~6 ± 6% lower than ISMIP6 melt rates, but our model is 16% higher than ISMIP6 melt rates at the location with maximum difference. The area of highest discharge and submarine melt appears along the northern section of the glacier terminus near the large trough where HG has been experiencing its fastest retreat since the 1990s. Although channels grow in the model to a maximum discharge of 34 m^3/s, inclusion of channels generally has a negligible effect on the spatial distribution and magnitude of submarine melt. Future work will couple our modeled submarine melt rates to glacier evolution to examine the potential impact of spatially variable melt rates on the recent and future retreat of HG.

Funding Program Area(s)