TR(BR)-8/96-97 : Catchment modelling in GIS environment

Abstract

Rainfall-runoff modelling helps in simulating and forecasting the flow from a catchment, and in determining the inflow series for the ungauged catchments on the basis of records of gauged catchments. Because of the scarcity of data, mostly lumped models have been used to model the rainfall-runoff process in a catchment. However, distributed models are more accurate and need to be adopted for modelling the complex processes at the scale of basins. With the development and wider availability of GIS tools, it is now easy to generate, store, manipulate, integrate and retrieve spatial data which can be used for distributed modelling of a basin. This has led to increased applications of distributed models for varied of water resources problems.

Top model is one such model having catchment representation in distributed functional form. Every such conceptualization also involves certain assumptions and the Top model is premised upon assumptions which lead to simple relationships between the catchment storage and local levels of the water table. The Top model is based on the concept of Topographic Index. This index represents as In(a/tang), where a is the area of the hillslope per unit contour length (m2) that drains through a point and tang is the local surface slope. This index was first proposed by Kirkby and developed as a complete hydrological model by Beven and Kirkby (1979). The Kirkby index represents a theoretical estimation of the accumulation of flow at any point or the propensity of any point in the catchment to develop saturated conditions. The computational requirements of the model and quite modest. However, the processing of topographic data requires huge computer resources are if the catchment size is large and the pixel size is small.

The Topmodel has been implemented and applied to Kolar catchment of size 820square km, located in Central India. The topographic data and DEM of the catchment were prepared from the Survey of India toposheets and the topographic data was processed through a GIS. The results of simulation show that a good match between the observed and simulated hydrographs was obtained despite handicaps in terms of data availability.