Using machine learning and satellite data to analyse climate change in the Upper Awash Sub-basin, Ethiopia

Deme Betele Hirko, Jakobus Andries Du Plessis, Adele Bosman,
Using machine learning and satellite data to analyse climate change in the Upper Awash Sub-basin, Ethiopia,
Physics and Chemistry of the Earth, Parts A/B/C,
Volume 141, Part 2,
2025,
104137,
ISSN 1474-7065,
https://doi.org/10.1016/j.pce.2025.104137.
(https://www.sciencedirect.com/science/article/pii/S1474706525002876)
Abstract: Climate change presents a critical challenge for the Upper Awash Sub-basin in Ethiopia, exacerbated by rapid urbanisation and growing agricultural water demand. This study integrates satellite (Princeton) and observational datasets (1948–2010) with future projections (2025–2075), applying both machine learning (ML) and climate modelling approaches to assess long-term climate trends. A Random Forest model trained on satellite and ground-based data demonstrated high predictive accuracy (R2 = 0.97 for training, 0.96 for testing; RMSE = 0.12 °C), enabling robust estimation of temperature and precipitation patterns. Historical comparisons revealed that Princeton data showed 6.3 % more precipitation and 10.2 % lower temperature than observed records, while Coupled Model Intercomparison Project Phase 6 (CMIP6) data indicated 57.9 % less precipitation and an 18.2 % increase in temperature. For 2025 to 2075, the Princeton ML model projects a slight 0.4 % rise in precipitation and a 10.2 % decrease in temperature. In contrast, the Shared Socioeconomic Pathway (SSP5-8.5) scenario anticipates a 52.1 % decline in precipitation and a 26.1 % rise in temperature. These disparities reflect fundamental differences in model assumptions, spatial sensitivity, and data sources. While CMIP6 suggests intensified warming and drying, the ML-based results indicate possible localised climate buffering effects, potentially influenced by land use, vegetation cover, or data limitations. The findings underscore the importance of multi-model comparisons and region-specific analyses to support effective climate adaptation. Recommended strategies include early warning systems, drought-resilient agriculture, and integrated urban–rural water management tailored to the sub-basin's evolving climate conditions.
Keywords: Climate change; Climate model; Machine learning; Satellite data; Upper Awash sub-basin