Tracking Snowmelt Events in Remote High Asia Using Passive Microwave Data

Abstract:

While snowfall can comprise a significant percentage of the yearly water budget in High Asia, Snow-Water Equivalent (SWE) is poorly constrained due to lack of in-situ measurements and complex terrain that limits the efficacy of modeling and observations. Over the past few decades, SWE has been estimated with the SSMI/S and AMSR passive microwave (PM) sensors, with low reliability in High Asia. Despite problematic SWE volume estimation, PM data contains information on the buildup and melt of snowpack, which is difficult to measure in-situ, particularly in remote areas. We present a new methodology for tracking the timing, frequency, and relative intensity of melt events across High Asia.

To measure SWE, we use raw swath data from the SSMI/S (1987-2015, F08, F11, F13, F17), AMSR (2002-2011), and GPM (2014-2015) satellites. This allows us to improve both spatial and temporal resolution over daily gridded products by leveraging multiple overpasses per day in an imperfectly overlapping grid pattern. We then examine SWE estimates, intra-day PM variance, and the interacting impacts of satellite look angles and topography on measured PM at arbitrary point locations. We develop a more thorough understanding of the uncertainties in our SWE estimates by examining the impacts of aspect, relief, slope, and elevation across the Tibetan Plateau on Tb and SWE estimates. High Asia, with its large topographic gradients and low relief at high elevations provides an excellent context to examine a wide range of topographic settings and terrain complexities to better constrain our analysis of sensor bias. We find that slopes above ~10° have a strong impact on SWE variability. We also find a consistent intra- and inter-day variability within constant-SWE periods, as defined as periods without precipitation and with constant temperatures below 0°C.

Using this measure of native sensor variability, we filter our SWE time series to identify events of snowmelt which are outside of the expected SWE variability envelope. We find that these events correlate with abnormally warm periods, as well as increases in variability in the high frequency (85-92 GHz) Tb channel.

This work will improve SWE storage forecasting, as well as help understand the drivers of, and predict the timing of, major snow melt events in remote catchments throughout High Asia.