Feasibility of various remote sensing data for mapping snow cover area around Gangotri Glacier

Abstract

The Himalayan region comprises extensive snow cover areas having various degrees of glaciations that act as a huge fresh water reservoir. Mapping the extent of these snow covered areas and deciphering the characteristics of these snow pack help in the estimation of snow melt runoff. This study involves the mapping of snow cover area around Gangotri glacier using remote sensing data analysis. The data products include IRS LISS-III, IRS P6 AWiFS and TERRA MODIS. The methodology involves conversion of digital numbers (DN) into actual reflectance values. Snow index derived on the basis of reflectance value will be more accurate because different objects having the same DN value may likely to correspond to different reflectance and therefore the use of reflectance improves the identification of objects. The final Snow cover area (SCA) maps are prepared based on the Normalised Difference Snow Index (NDSI) values. The Snow cover maps were classified into two classes: Snow cover and Non-snow cover. The threshold value greater than 0.4 is used in distinguishing snow from soil, rock and cloud. In this investigation it has been found out that from the remote sensing data products used, LISS-III and AWiFS sensors are far superior compared to MODIS and MERIS. Hence the snow cover maps for different months of 2003 and 2004 were produced using only LISS-III and AWiFS data products. The analysis of these maps indicates that the snowfall in 2002-03 season was higher than in 2003-04. The melting of snow started as early as March in 2004, thus resulting in depletion of snow cover extent. These observations could be corroborated with the field weather data collected by Automatic Weather Station (AWS).