Surface Meltwater Drainage and Ponding on the Amery Ice Shelf, East Antarctica

Surface melting is widespread and persistent on many Antarctic ice shelves, and meltwater is theorized to be connected to ice-shelf collapse. Drainage systems move water significant distances across ice-shelf surfaces, potentially into areas vulnerable to hydrofracturing, so it is vital to understand how surface drainage systems will respond to potential future increases in melting caused by atmospheric warming. We use LANDSAT (4-8) and MODIS multispectral imagery and Sentinel-1 C-Band Synthetic Aperture Radar (SAR) to examine the seasonal evolution of a large drainage system that forms multiple times per decade in a bare-ice zone on the Amery Ice Shelf (AIS), East Antarctica. The drainage system consists of 100's of lakes connected by surface channels. Spatio-temporal changes in SAR backscatter strength suggest that most lakes freeze through in winter, while in 6-20% of the lakes (by area) remain liquid beneath a frozen lid through the winter. The drainage system terminates in several elongated 'terminal lakes', whose downglacier extent is highly variable inter-annually, often extending 10's of kilometers downstream into snow-covered areas. We estimate seasonal meltwater input m to one terminal lake using output from the regional climate model RACMO2.3, integrated over a catchment defined using the Reference Digital Elevation Model of Antarctica. Assuming uniform lake depth (as suggested by multi-spectral depth retrieval algorithms), large excursions from an approximately-linear relationship between m and annual maximum lake area may be due to (1) inaccurately modeled melt rates, (2) inter-annual changes to drainage catchments, or (3) evolution of the lake basin's hypsometry and/or permeability. Supporting the latter, we observe that consecutive years of significant melting lead to year-on-year expansion of the drainage network, potentially through a feedback between melt production, basin permeability and lateral flow. Such a feedback could result in downstream expansion of drainage systems if warmer summertime air temperatures cause more consecutive years of melt. Our findings suggest that understanding this feedback, and how catchment size responds to increased melt rates, will vital for determining where on Antarctica's ice shelves water accumulates and hence where hydrofracture is a risk.