Potential for hydrologic characterization of deep mountain snowpack via passive microwave remote sensing in the Kern River basin, Sierra Nevada, USA
Li, D., M. Durand, S.A. Margulis, 2012: “Potential for hydrologic characterization of deep mountain snowpack via passive microwave remote sensing in the Kern River basin, Sierra Nevada, USA.” Remote Sensing of Environment, v. 125, pp. 34-48, doi: 10.1016/j.rse.2012.06.027.
Snow plays a critical role in hydrology and water resources, but snow properties in mountainous areas vary dramatically in space, making characterization difficult: in situ measurements represent a single point of spatially-variable snow properties. Spaceborne passive microwave (PM) remote sensing has spatially continuous coverage, but coarse spatial resolution. PM spatial footprints are nominally elliptical, with the spatial orientation changing relative to a given basin on the ground from one satellite pass to the next. The widely used Equal-Area Scalable Earth Grid (EASE-Grid) resamples the raw PM observations to a 25 km × 25 km grid; this is far coarser than the Advanced Microwave Scanning Radiometer for EOS (AMSR-E) 37 GHz Level 2A (L2A) footprints with an area of 87.9 km2. This paper presents methods for processing the L2A data in order to illustrate that PM measurements contain information about snow accumulation and ablation cycles in mountainous regions. Methods are presented 1) to calculate the average Tb over a hydrologic basin for monitoring SWE accumulation amount, and 2) to interpolate Tb to a point for monitoring SWE ablation timing. Across the six-year study period, the range of the Tb was 50 K, while the range of the air temperature was only 30 K, indicating significant surface emissivity variations. The minimum Tb of each water year (WY), which starts from October 1 of a calendar year and ends at September 30 of the next, showed a strong inverse relationship to SWE measured in situ; the correlation coefficient between Tb and an in situ basin average SWE was −0.94. The L2A data is three times more sensitive than the EASE-Grid data to in situ SWE. The diurnal amplitude variability (DAV, or difference between daytime and nighttime Tb) was used to identify the day of melt onset and compared with in situ estimates of onset derived from daily SWE measurements. L2A data corresponded well with the in situ onset dates with a correlation coefficient of 0.94 and an RMSE of 5.04 days. When using EASE-Grid data, the RMSE in onset date was 11.7 days. In addition, Tb DAV was found to be correlated with Tair DAV during winter, with an average correlation coefficient of 0.72. Future work will explore methods to extract this information in order to improve estimates of snow accumulation and ablation patterns.