Non-linear statistical model for the daily stream flow Prediction in the kalu river catchment in Sri Lanka

Abstract

Having a long record of stream flow is very valuable in planning water resources development projects. However, in many occasions, stream flow records are available for very short periods though very long rainfall records are available. Therefore, possibility to relate rainfall over a catchment to the stream flow at its outlet will enable having a long record of stream flow. Besides prediction of stream flow using already available predicted rainfall will permit taking precautionary measures in water related disaster situations such as floods and droughts. This paper presents a research carried out to find a model to predict daily stream flow of Kalu River at Ratnapura. The model, a non-linear regression model based on Marquardt’s procedure, was developed using measured daily stream flow at Kalu River at Ratnapura and daily rainfall at eight rainfall gauging stations within the catchment above Ratnapura. Data for the period 1987-1994 were used for the calibration of the model while data for the period 1995-2000 were used for verifying it. The model was validated using Nush-Sutcliffe efficiency and pseudo R2 . Nush-Sutcliffe efficiency (78%) and pseudo R2 (85%) show the possibility of the fitted model in predicting daily stream flow of Kalu River at Ratnapura.