Tl. Mote et Mr. Anderson, VARIATIONS IN SNOWPACK MELT ON THE GREENLAND ICE-SHEET BASED ON PASSIVE-MICROWAVE MEASUREMENTS, Journal of Glaciology, 41(137), 1995, pp. 51-60
A simple microwave-emission model is used to simulate 37 GHz brightnes
s temperatures associated with snowpack-melt conditions for locations
across the Greenland ice sheet. The simulated values are utilized as t
hreshold values and compared to daily, gridded SMMR and SSM/I passive-
microwave data, in order to reveal regions experiencing melt. The spat
ial extent of the area classified as melting is examined on a daily, m
onthly and seasonal (May-August) basis for 1979-91. The typical season
al cycle of melt coverage shows melt beginning in late April; a rapid
increase in the melting area from mid-May to mid-July, a rapid decreas
e in melt extent from late July through mid-August, and cessation of m
elt in late September. Seasonal averages of the daily melt extents dem
onstrate an apparent increase in melt coverage over the 13 year period
of approximately 3.8% annually (significant at the 95% confidence int
erval). This increase is dominated by statistically significant positi
ve trends in melt coverage during July and August in the west and sout
hwest of the ice sheet. We find that a linear correlation between micr
owave-derived melt extent and a surface measure of ablation rate is si
gnificant in June and July but not August, so caution must be exercise
d in using the microwave-derived melt extents in August. Nevertheless,
knowledge of the variability of snowpack melt on the Greenland ice sh
eet as derived from microwave data should prove useful in detecting cl
imate change in the Arctic and examining the impact of climate change
on the ice sheet.