CS(AR)-17/97-98 : Hydrological Modelling Using GIS

Abstract

In hydrological analysis and design, precipitation-runoff relations are useful to extrapolate or interpolate runoff records from the precipitation records and to estimate the runoff of ungauged catchments. Estimates of hourly, daily, monthly, seasonal or annual runoffs may be required for operational purposes or to have efficient flood forecasting or to provide a database for evaluating reservoir storage requirements. Precipitation-runoff models are classified as lumped and distributed models. Lumped models are developed to get rough estimate of runoff peaks. But the distributed models are developed to represent the complete catchment to get runoff hydrographs.

NASMO model is a rainfall - runoff model developed by Dr. Ing. A.Stodter, Leichtweiss Institute for Hydraulic Engineering, TU Braunschweig, Germany. In this model, the US SCS method is used to find effective rainfall, the linear reservoir method is used to route the runoff over land and the Modified Puls method is used to route the runoff in the stream.

The NASMO model is used to parameterise the following values: 1. The deviation from the soil moisture content 2. Interception 3. Ratio of overland flow (the ratio between effective precipitation to overland flow from direct runoff) 4. Factors to time of concentration for overland flow, interflow and base flow 5. Base flow 6. Factor to travel time in channel (Retention time) 7. Storage coefficient of base flow

G1S ARC/INFO and ILWIS were used to get characteristics of the sub catchment like size, slope, area and landuse details and the length of the stream.

The catchment of Malaprabha upto Khanapur (515.297 sq.km) in Karnataka State was selected and the toposheets were digitised to get the catchment characteristics. The catchment was divided into 39 sub catchments. Three storms 1987 ( 1st to 31st July), 1988 (9th to 26th July) and 1990 (1st to 31st July) and the corresponding hydrographs were selected for parameter fitting . The parameters were fitted systematically by trial and error process. Best-fitted parameters were arrived for all the storms. The parameters for storms are presented in tabular form. The averaged parameter values are presented in the conclusion. The averaged best-fitted parameters were evaluated using the storm 1991 (1st to 31st July) and the corresponding hydrograph. The output of the model for all the storms are presented in graphical form. It is observed from the results that this model can be used to predict runoff peaks from Indian catchment with more number of storms for parameter fitting.