Geo-Spatial Modeling of Runoff of Large Land Mass: Analysis, Approach and Results for Major River Basins of India

Abstract

Soil Conservation Service (SCS) and overland flow models have been used for runoff modeling over major land mass of India. Remote sensing derived daily rainfall data (Climate Prediction Centre), high temporal NDVI data (SPOT VGT), DEM (GTOPO30) and soil texture maps were used as input for the runoff modeling. SCS model setup was done in GIS (Arc GIS) environment. In general the reported and model estimated runoff matched well for most of the basins. It was observed that there was a shift in the runoff pattern towards western region as compared to reported normal runoff. Total monsoon season (aggregating 1st June to 30th September) runoff was 126 Mha-m over mainland of India during 2004. About 44 percent of total rainfall was converted into surface runoff. Month-wise runoff contribution was of 17%, 35%, 36% and 12% during June, July, August and September months, respectively in 2004. SCS model doesn’t take into account flow of runoff. Therefore, another overland model was setup by taking SCS model as input along with DEM data. The overland flow model results indicate large difference in the spatial behaviour of runoff compare to SCS model estimated runoff pattern because of flow of runoff water to the down gradient. Remote sensing derived parameters facilitate spatial modeling due to its spatial format. High temporal resolution remote sensing data has been found useful in deriving the landuse/cover required for such study as it captures the variation both in spatial and temporal domain, thus, improving the model performance.