Northeastern Section - 40th Annual Meeting (March 14–16, 2005)

Paper No. 8
Presentation Time: 8:00 AM-12:00 PM

CAN PASSIVE MICROWAVE-DERIVED SNOWMELT PREDICT RUNOFF AND CHANNEL STABILITY?


RAMAGE, Joan M.1, MCKENNEY, Rose2, THORSON, Blair3, MALTAIS, Pat4, FONTAINE, Abigail5 and PAYNE, Jon2, (1)Earth and Environmental Sciences, Lehigh Univ, 31 Williams Hall, Bethlehem, PA 18015, (2)Environmental Studies Program/Geosciences, Pacific Lutheran Univ, Tacoma, WA 98447, (3)Water Survey of Canada, Environ Canada, 91782 Alaska Highway, Whitehorse, YT Y1A 5B7, Canada, (4)Water Survey of Canada, Environ Canada, 91782 Alaska Highway, Whitehorse, YT Y1A 5B7, (5)Environmental and Atmospheric Sciences, Creighton Unviversity, 2500 California Plaza, Omaha, NE 68178, ramage@lehigh.edu

Snow volume and melt timing are major factors influencing the water cycle at northern high altitudes and latitudes, yet both are hard to quantify or monitor in remote, mountainous regions. Twice-daily Special Sensor Microwave Imager (SSM/I) passive microwave observations of seasonal snow melt onset in the Wheaton River basin, Yukon Territory, Canada (~60º 08’ 05’’N, ~134º 53’ 45’’W) are used to test the idea that melt onset timing and duration of snowpack melt-refreeze fluctuations control the timing of the early hydrograph peaks with predictable lags. This work uses the SSM/I satellite period of record (1988 - 2004) and daily hydrograph data to evaluate the variability in melt and runoff patterns in the upper Yukon River basin. The Wheaton River is a small (875 km2) tributary to the Yukon, a subarctic, partly glacierized, heterogeneous basin with near-continuous hydrographic records dating back to 1966. SSM/I pixels are sensitive to melt onset due to the strong increase in snow emissivity, and have a robust signal, in spite of coarse (>25 x 25 km2) pixel resolution. Results show that peak flows closely follow the end of large diurnal variations in the pixels representing the Wheaton River, but the magnitude of flow is highly variable, as might be expected from interannual snow mass variability. Preliminary data also show that larger discharges deepen the channel but do not cause lateral migration at the gage, located near the river mouth. Additional study of aerial photos will be used to determine whether this is typical basin-wide. Subsequent work will apply these techniques to a larger (7250 km2), unglacierized tributary to the Yukon River, the Ross River, which is farther North (~61º 59’ 40” N, ~132º 22’ 40” W) in the Yukon Territory. These techniques will also be used to try to determine the improvement in melt detection and runoff prediction from the higher resolution (~ 15 x 15 km2) Advanced Microwave Scanning Radiometer for EOS (AMSR-E) sensor (AMSR-E) sensor.