IAHR Document Library

« Back to Library Homepage « Proceedings of the 20th IAHR APD Congress (Colombo, 2016)

Seasonal Stochastic Model for Long Term Reservoir Inflow Forecasting for Ukai Reservoir, India

Author(s): Priyank J. Sharma; P. L. Patel; V. Jothiprakash

Linked Author(s): Priyank J. Sharma, Prem Lal Patel, V. Jothiprakash

Keywords: Box-jenkins model; Calibration-validation; Streamflow forecasting; Tapi basin

Abstract: This study presents application of seasonal streamflow forecasting model of Box and Jenkins family for inflow prediction into Ukai reservoir in Tapi Basin, India. The analysis of historical inflow data at the reservoir from year 1975 to 2014 reflected that almost 98%of the inflow volume is reported in the monsoon months, i.e. June to October. There is negligible inflows in the remaining months, and most of the periods had zero values which result in more skewed distribution when visualized on annual scale. Hence, the seasonal inflow model, capturing the inherent periodicities, has been selected for present study. The model has been developed considering different proportions of calibration and validation periods of the entire data length, viz. 70:30 and 50:50, and its effect on the parameter estimation is also investigated. Different candidate models are formulated, and the best model is selected using Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). The selected models were parsimonious and further validated by diagnostic check of residuals. The seasonal forecasts are made using the calibrated model, and the performance of model has been assessed using various statistical indices like Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and coefficient of correlation (R). The models yielded better correlation for low and moderate inflow scenarios, while for high inflows the correlations were low. The model forecasts would be helpful in taking decisions pertaining to the management of water resources for irrigation planning, hydropower operation as well as optimal reservoir releases.


Year: 2016

Copyright © 2023 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions