Analysis the Long-Term Rainfall Trends in the Tropical Regions, Using Satellite-Based Rainfall Estimates: A Case Study from Sabaragamuwa Province - Sri Lanka

Abstract

The rainfall is a critical climatic factor for successful agricultural activities anywhere in the world. An analysis of long-term rainfall behavior and trends contribute significantly to effective crop planning, water resource management and mitigation of the impact on hydro-meteorological hazards. This study reveals the rainfall variability and trends in the Sabaragamuwa Province, Sri Lanka from 1990 – 2020. The uniqueness of this study is the use of daily raster rainfall data generated by mixing gauge rainfall with satellite estimates known as Climate Hazards Group InfraRed Precipitation corrected with Stations (CHIRPS). As the first step, the correlation between gauge precipitation and CHIRPS data was analyzed and it showed a high accuracy of 81%. The Geographic Information System (GIS) was primarily employed to identify spatial variability in long-term rainfall and to extract data at the district level for rainfall trend analysis. Then, the Sen’s slope estimator and the Mann–Kendall (M–K) test were used to calculate and evaluate the trends in annual and seasonal rainfall within two districts, Rathnapura and Kegalle in the Sabaragamuwa Province. The results proved a statistically significant (0.05) positive trend in annual rainfall from 1990 to 2020 in both the districts of Sabaragamuwa Province. The highest trend 31.85mm/year is recorded in Rathnapura district and the lowest trend in Kegalle district which is 24.54 mm/year. Furthermore, all four monsoons: the First inter-monsoon, the Second inter-monsoon, and the Southwest monsoon and Northeast monsoon showed a statistically significant upward trend. If these trends continue in the next 10-20 years, the annual rainfall in Sabaragamuwa Province could be increased by 320 mm to 740 mm.