Simulation-Optimization Modelling of Rainwater Harvesting from Karso Watershed of Damoder Catchment

Abstract

An evaluation was conducted to develop a stochastic time series model, capable of prediction of rainfall and runoff in Karso watershed. The Karso Watershed selected for hydrological studies is one of the sub watershed of the Damoder dam catchment of upper Damodar Valley, comprising a cover the area of 27.41 km 2 . The hydrologic sequences data of watershed collected from Soil Conservation Deptt., Damodar Valley Corporation, Hazaribagh, Jharkhand State were analysed. The watershed be capable of be divided into three main landscapes. The first one is the southern part which is highly undulating and rolling uplands, which drains from south to north which is parallel to the HazaribagPatna National Highway. The second is gently undulating and rolling uplands, that are dissected by narrow valley and depressions. The third is valley lands, which drains from south to north which is parallel to the Hazaribagh-Patna national highway. In this area sheet wash, rill erosion, shallow and medium gullies are prominent. The hilly area lies near the village Kundwa, Daurwa, Rola etc. The main objective of the study was to develop an autoregressive time series model for annual rainfall, runoff and sediment yield. The underlying stochastic process of annual rainfall, runoff and sediment yield is characterized by autoregressive time series model. The autoregressive time series (AR) model is applied to simulate water demand and supply year by year for each basin or aggregated basin used in impact water. The model assumes that non-agricultural water demand, including municipal and industrial water demand and committed flow for in stream uses, is satisfied as the first priority, followed by livestock water demand. The effective water supply for irrigation is the residual claimant, simulated by allowing a deficit between water supply and demand. This research is based upon identification and parameter estimation of model and evaluation of performance and adequacy of the model by statistical parameters and several other measures such as mean forecast error, mean absolute error, mean relative error, mean square error, root mean square error and integral square error. The assessment among the measured and predicted rainfall and runoff by AR(1) model clearly shows that the developed model can be used capably for the future prediction of rainfall and runoff in Karso watershed.