The Accuracy of CHIRPS rainfall data and its utilization in determining the onset of the wet and dry seasons in North Sumatra

Abstract

Rainfall is a weather and climate parameter with high variability in space and time. Geographical conditions and atmospheric dynamics cause high rainfall intensity in the Indonesian Maritime Continent (IMC). The availability of rainfall data is crucial to determine the onset of the wet and dry seasons that affect community activities and the formulation of water resources policies. The limited number of rain gauges is a major obstacle in monitoring rainfall. Overcoming this, satellites can be used for rainfall estimation data, one of which is Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS). However, the use of CHIRPS data needs to be tested for accuracy. This study aims to calculate the accuracy of CHIRPS data and its utilization in determining the onset of the wet season and the onset of the dry season. Validation is conducted using in-situ observation data at four BMKG stations in the North Sumatra region using the Contingency Table method. The results show that the four stations have an Equatorial rain pattern, with rain peaks in the September-October-November (SON) and March-April-May (MAM) periods. The monthly CHIRPS data has the highest performance based on the Pearson Correlation value. While the Proportion Correct (PC) of daily rainfall averages 62%, the highest value is at Binaka Meteorological Station. Calculating decade (10-day) and monthly rainfall results in the West Coast region is more accurate than the East Coast of North Sumatra. Determining dry season onset with CHIRPS data tends to be more advanced than dry season onset using the observation data. Meanwhile, the determination of wet season onset shows that most wet season onset occurs simultaneously on the west coast of North Sumatra