Change detection analysis of avalanche snow in Himalayan region using near infrared and active microwave images

Abstract

Snow avalanches claim lives and property year after year in Indian Himalaya. The major contributing factors to the avalanches are the changes in the characteristics of the snow cover of the high inaccessible slopes, which are difficult to assess through conventional techniques. Use of satellite imageries can greatly solve this problem. We have used classification matrix method for change detection studies using optical and microwave images. The relationship between temporal changes in snow and the avalanche occurrence in Indian Himalayas has been studied. From avalanche point of view, the snow has been classified in three categories, namely, dry, moist and wet. The study has been carried out for 141 km stretch of road Manali–Leh in Indian Himalaya in Pir Panjal range covering an area of 123,286 ha and altitude 2500–4500 m for the period October 2001 to May 2002. The analysis of multi-temporal optical and active microwave satellite data shows that visible bands of Indian Remote Sensing Satellite (IRS) LISS III sensors help in providing information about terrain features; NIR band is more useful for analysis of snow characteristics, and Radarsat image helps in extraction of snow wetness information. The fusion of optical and microwave data, whenever available, has been very useful for increasing the accuracy of the classification. The study confirms significant changes in snow characteristics and wetness due to large variations in temperature at these altitudes and their relevance with the avalanche activity. The study also shows that continuous monitoring of the changes in the snow using optical and microwave imageries can be a very useful tool to a forecaster for accurate assessment of avalanche danger.