CONTRIBUTION OF SNOW PARAMETERS IN A SEMI-EMPIRICAL MODEL

Snow cover has an immense value as a natural resource of water used particularly for irrigation purposes. The amount of snow accumulated in a (mountain) watershed determines the runoff after the onset of melt during spring. Hence, the measurements of snow cover extent and snow characteristics (snow depth and snow water equivalent) in real time are very important. Conventional methods of data collection (for example, snow surveys or isolated stations using in situ sensors) are time consuming and spatially limited. Consequently, the resultant snow water equivalent measurements are enormously different from the actual snow water equivalent. At high elevation and in remote areas of the globe where very little in situ data exists, remote sensing is the only mean by which to observe the snow cover distribution. Microwave radiation penetrating through clouds and snow covered area could provide snow depth and snow water equivalent information about a snow pack. However, due to the coarse spatial resolution of the passive microwave sensors, in a single footprint, the presence of snow surface area along with forest fraction and water bodies contribute considerable signal variation. The principal objective of this study is to analyze the degree to which SSM/I brightness temperature can be affected by the snow parameters. The sensitivity analyses of the snow surface on the brightness temperature estimation with semi-empirical model using Special Sensor Microwave Imager (SSM/I) frequencies are investigated. In the sensitivity analysis the parameters of interest were snow density, snow salinity, snow wetness, snow grain size, snow temperature and snow depth.