Integration of remote sensing with GIS for grassland snow cover monitoring and snow disaster evaluating in Tibet

As an important pasture region, Tibet has about 82 million hectares of natural grassland, accounting for 68.11% of its total territory. Above 90% of Tibetan grassland belongs to the types of alpine meadow steppe and alpine steppe with highly nutritious forage plant. Animal husbandry constitutes a major part of agricultural economy in Tibet. It is believed that snow disaster become a significant threat to the development of animal husbandry in Tibet. The disaster often happens in winter and spring as a result of complicated mountainous features and mutable climatic conditions. Statistics indicates that, on average, there is a slight snow disaster for each 3-year, a medium disaster within 5 to 6 years, and a big disaster in 8-10 years. Large numbers of animals died of hungry and cold during the disaster period. Huge economic loss due to the disaster had brought giant difficulties to local herdsmen in Tibet. Accurate and timely monitoring of snow cover for snow disaster evaluating is very important to provide the required information for decision-making in anti-disaster campaigns. Remote sensing has many advantages in snow disaster monitoring hence been extensively applied as the main approach for snow cover monitoring. In this paper we present our study of snow cover monitoring and snow disaster evaluating in Tibet. An applicable approach has been developed in the study for the monitoring and evaluating. The approach is based on the normalize difference of snow index (NDSI) and DEM retrieved from MODIS and GIS data. Using the approach, we analyzed the snowstorm occurring in mid-March 2007 in southern Tibet. Results from our analysis indicated that the new approach is able to provide an accurate estimate of snow cover area and snow depth in southern Tibet. Thus we may conclude that the approach can be used as an efficient alternative for snow cover monitoring and snow disaster evaluating in Tibet. © 2008 SPIE.