Spatiotemporal rainfall variability and trend analysis over Mahaweli Basin, Sri Lanka

Abstract

The hydrometeorological characteristics of the Mahaweli Basin are infuenced by rainfall distribution. For that reason, it is signifcant to identify spatiotemporal rainfall fuctuations and trends over the Mahaweli Basin. Accordingly, rainfall data from 1990 to 2019 available for the 15 raingauge stations were analyzed for rainfall variability and trends. Serial autocorrelation was checked before applying rainfall time series data to Mann–Kendall (MK) test. The result exhibited no serial autocorrelation in the data. The MK test, Sen’s slope estimator (SSE), and innovative trend analysis (ITA) were applied to recognize rainfall trends. The inverse distance weighting (IDW) interpolation method was applied to show the spatial pattern of rainfall characteristics with the support of ArcGIS 10.1. Some fuctuations were observed in the rainfall over the 30 years with decreasing and increasing trends. Nevertheless, signifcant trends in the annual rainfall were noted for Bandarawela (+15.7 mm), Ledgerwatta (+40.3 mm), Duckwari (−36.3 mm), and Bakamuna (24.3 mm). At the basin scale, no signifcant trends were noted in rainfall of the Mahaweli Basin. The rainfall trend analysis results obtained by ITA have validated the results of the nonparametric test. Therefore, the analysis showed that despite the seasonal variations in rainfall over the Mahaweli Basin, rainfall is regular, and results acquired by MK test, SSE, and ITA methods are reliable.