C32B-08
Satellite-based Observation of Arctic River Dynamics

Wednesday, 16 December 2015: 12:05
3007 (Moscone West)
Irina Overeem, Univ Colorado, Boulder, CO, United States, Robert Brakenridge, University of Colorado, CSDMS, Boulder, CO, United States and Benjamin Hudson, Applied Physics Laboratory University of Washington, Seattle, WA, United States
Abstract:
One of the indicators of a warming Arctic region is an intensification of the hydrological cycle, with increasing permafrost and glacial melt and possibly more precipitation resulting in higher river runoff. Indeed, a significant increase of nearly 10% in annual river flux has been observed in 13 major rivers throughout the entire Arctic region over the last 30 years. However, direct measurements are extremely sparse for 100’s of smaller-scale tundra river systems, as well as for proglacial rivers around the Greenland Ice Sheet margin. Observations at in-situ gauging stations are hampered by seasonal ice coverage, break-up and freeze-up dynamics, unstable banks, and difficult access. To overcome such difficulties, we develop remote-sensing based river discharge measurement techniques using a variety of satellite sensors, including reflectance in the near-infrared band of MODIS, LANDSAT, and brightness temperature from the passive microwave sensors AMSR-E and AMSR-2. We use varying inundation of the river channel and floodplain throughout a season to quantify the changing Arctic river flux. A new approach to detect river ice break up in spring has been developed, and is now undergoing validation. To calibrate the remote sensing signal to daily river discharge, we employ either in-situ short observation records, or a numerical distributed hydrological model driven by daily reanalysis climate data. Quantitative reconstructions of meltwater fluxes in rivers along the Greenland Ice Sheet margin obtained so far show a dampened response of these rivers to Greenland Ice Sheet melt. Techniques are now deployed to map river dynamics along the Chukchi Sea and Beaufort Sea coasts, and show shifts in break-up dynamics and flooding patterns. Once calibrated, satellite-based reconstructions have the potential to lengthen short observational records to a ~15 year timespan.