Retrieval of Snow Water Equivalent Using Passive Microwave Brightness Temperature Data Purushottam Raj Singh* and Thian Yew Gan* Existing algorithms for retrieving snow water equivalent water equivalent (SWE)] the most rapidly varying surface feature on Earth. Snow is a dominant source of water (SWE) from the Special Sensor Microwave/Imager (SSM/I) supply in Canada and some parts of the United States. In passive microwave brightness temperature data were as- the Canadian Prairies, the shallow snow cover generates sessed and new algorithms that include physiographic and as much as 80% of the annual surface runoff from some atmospheric data were developed for the Red River basin local areas (Granger and Gray, 1990). In the Colorado of North Dakota and Minnesota. The frequencies of SSM/I Rockies and Sierras of California, snowfall accounts for up data used are 19 GHz and 37 GHz in both horizontal and to 90% of the annual water supply. During spring, snow- vertical polarization. Encouraging calibration results are melt fills reservoirs and groundwater systems that provide obtained for the algorithms using multivariate regression water for agricultural and municipality use and hydropower technique and dry snow cases of the 1989 and 1988 SSM/I generation. Thus, knowing the seasonal variations of SWE data (from DMSP-F8). Similarly, validation results for data is critical for an effective management of water resources. not used in calibration [e.g., 1988 (1989 as calibration However, the only way to adequately estimate the spatial data), 1989 (1988 as calibration data), and 1997 (from coverage and temporal changes of snow cover in a regional DMSP-F10 and F13)] are also encouraging. The nonpara- scale is via remote sensing. metric, Projection Pursuit Regression (PPR) technique also Space-borne data have been utilized since the mid- gave good results in both stages. However, for the validation 1970s in water resource management. The advent of air- stage, adding a shift parameter to all retrieval algorithms borne gamma ray spectrometry and microwave remote sen- was always necessary, possibly because of different scatter- sors are keys to passive microwave snow research (e.g., induced darkening (caused by scattering albedo), which Goodison et al., 1986; Chang et al., 1987; Hallikainen, 1989). could arise even for snowpacks of the same thickness be- For example, the Office of Hydrology, National Weather cause snowpacks undergo different metamorphism in dif- Service of USA has been measuring SWE using airborne ferent winter years. Screening criteria are also proposed gamma radiation with as many as 1,578 flight lines distrib- to eliminate SSM/I footprints affected by large water bodies uted in 32 states/provinces of the United States and south- and depth-hoar—another key step toward reliable SWE ern Canada (see NWS, 1992). Unfortunately, very high estimation from passive microwave data. 2000 Elsevier operational cost involved with such an airborne survey Science Inc. restricts its application globally. England (1975), Chang et al. (1976), and others re- ported the scattering of microwave radiation by snow crys- INTRODUCTION tals. This scattering effect, which redistributes the upwell- In the northern hemisphere, the mean monthly land area ing radiation according to snow thickness and crystal size, covered by snow ranges from 7% to 40% during the annual provides the physical basis of microwave detection of snow. cycle, making snow cover [in terms of area extent and snow Despite of its coarse resolution (about 25 km), the ability of passive microwave to penetrate dry snow and clouds and to provide dual polarization information at different * Department of Civil and Environmental Engineering, University frequencies at night makes it attractive for snow studies of Alberta, Edmonton, Alberta, Canada T6G 2G7 on a global basis. Earlier studies (between 1978 and 1987) Address correspondence to Dr. T. Y. Gan, University of Alberta. were mainly based on the microwave brightness tempera- E-mail: tgan@civil.ualberta.ca Received 6 July 1999; revised 13 March 2000. ture (TB) data from the Scanning Multi-channel Micro- REMOTE SENS. ENVIRON. 74:275–286 (2000) Elsevier Science Inc., 2000 0034-4257/00/$–see front matter 655 Avenue of the Americas, New York, NY 10010 PII S0034-4257(00)00121-8