Developing Monthly Hydrometeorological Timeseries Forecasts To Reservoir Operation in a Transboundary River Catchment

Forecasting of hydrometeorological timeseries data play vital role in flood forecasting and predicting the future water availability for various uses such as irrigation, hydropower generation, industrial, domestic, etc. Therefore, present study aims to forecast the hydrometeorological timeseries data, i.e., river inflows, precipitation, and evaporation for the improved reservoir operation of a transboundary Mangla catchment by using ARIMA (auto-regressive integrated moving average) model. Prior to applying the ARIMA model, stationarity of hydrometeorological timeseries data was checked. Moreover, ACF and PACF of timeseries were determined to determine the “ p ” and “ q ” terms of the ARIMA model. The best fitted structure of ARIMA model was used by evaluating the R 2 , MAE and RMSE to forecast the hydrometeorological timeseries. The seasonal ARIMA structure of (1,0,0)(2,1,2) 12 was found best fitted for the inflow timeseries whereas ARIMA structures of (14,1,15) and (9,1,19) were considered for forecasting the precipitation and evaporation timeseries, respectively. An average water shortage of 14% was detected by using these forecasted hydrometeorological timeseries in the reservoir operation during the period of 2016–2030. It was also observed that inflows into Mangla reservoir have seasonal effect more prominent compared to evaporation and precipitation. However, variations in the precipitation timeseries were found less smooth than the inflows timeseries. It is believed that the results of this study may support reservoir operators and managers for developing efficient real-rime reservoir operation policies and strategies based on the variations in the future water availability.