Trend Analysis of Streamflow Records of Indian River Basins Accounting Long Term Persistence

Abstract

The long term hydro-climatic changes influences streamflow variability and the estimation of long term persistence (LTP) may help to improve the accuracy of hydrological modelling. This study first determined the strength of LTP of streamflow of stations located in five major river basins in India using the classical Rescaled range (R/S) analysis. Subsequently, the standard Mann – Kendall (MK1) is used to analyse the trends in streamflow data. Finally, the Mann-Kendall test with variations accounting for long term persistence (MK2) is employed for trend detection. It is found that standard MK test (MK1) showed statistical significance in the data of 33.8% stations, while MK2 test detected statistical significance in only 21.77 % stations. It is concluded that MK1 test overestimate the inherent trend in the datasets due to the presence of long term persistence in streamflow and the latter method is more reliable estimate in long term water resources planning and management.