A method for compositing MODIS images to remove cloud cover over Himalayas for snow cover mapping

Snow plays an important role in the global energy and water budgets, as a result of its high albedo, thermal and water storage properties. Snow is also the largest varying landscape feature of the Earth's surface. Snow cover influences several processes that occur at or near earth surfaces. The snow cover extent of the snowpack is an extremely useful snow hydrology and climate variable for a number of reasons. The major problem in snow cover mapping is cloud because of high altitude position. Due to similar reflectance between snow and clouds it can be difficult to distinguish the snow information with spectral information alone. NASA's satellites Aqua and Terra MODIS have 36 channels onboard which significantly contributes to study the snow parameters. However, persistent cloud cover over Himalayan region limits the applicability of individual MODIS image with cloud cover. One solution to minimize the effect of cloud cover is to produce a composite image of cloud free pixels selected from several satellite passes of the same area and separated by time using image compositing algorithm. In this study to identify the cloud region MODIS cloud mask product MOD35 is used. This cloud mask products ability to mask the cloudy pixels over snow cover region gave better results than other methods. To check the accuracy of this method snow cover area has been calculated by different existing image compositing method. Further, the results were compared with snow cover area value derived from a cloud free dated image. This comparative study shows accuracy of the new compositing method using MODIS cloud mask product MOD 35.