Validation of Satellite Rainfall Estimates against Rain Gauge Data for Enhanced Drought Monitoring and Early Warning Systems in Botswana

Satellite rainfall estimation products (SREs) play a crucial role in overcoming the absence and scarcity of rain gauge data to monitor rainfall-related risks and provide early warnings. However, SREs can be subject to several sources of errors and need to be validated before use. This work provides the validation of three (3) SREs: Climate Hazards Group Infrared Precipitation with Stations (CHIRPS v2.0), Tropical Applications of Meteorology using Satellite and Ground-based Observation (TAMSAT) v3.0, and TerraClimate with respect to their performance in detecting rainfall patterns in Botswana at monthly scales from 1984 to 2016. Rain gauge data from the Botswana Department of Meteorological Services (BDMS) was used as reference data. Botswana is highly vulnerable to droughts; thus, understanding rainfall products strengths and weaknesses is important. A comparative analysis was conducted using pairwise comparison statistics and categorical statistics. CHIRPS and TAMSAT strongly correlate with low rainfall, while TerraClimate overestimates it. All three products underestimate high rainfall amounts. CHIRPS has the best efficiency in estimating rainfall, as evidenced by its high Kling-Gupta efficiency (KGE) score of 0.94. All products show their best results during the wet season, while CHIRPS typically surpasses the other two in the dry season, evidenced by a False Alarm Ratio (FAR) of 0.12, indicating effective detection under low rainfall conditions. All products demonstrate the best performance in Northern Botswana, where rainfall is dominated by the incursion of tropical depression. The study determined that the CHIRPS dataset can serve as an effective alternative to rain-gauge rainfall data in semiarid Botswana.