Modeling climate and land use land cover changes impact on water balance in the Rift Valley Lake Basin, Ethiopia
摘要
Water resources in the Rift Valley Lakes Basin (RVLB) of Ethiopia are increasingly threatened by human-induced climate and LULC changes. This study evaluates the impacts from climate and LULC changes on water balance components (WBCs) within the Bilate, Gidabo, and Gelana watersheds. CMIP6-GCMs were applied to simulate historical (1985–2014) and future projections under SSP2-4.5, SSP3-7.0, and SSP5-8.5 for the 2050s (2035–2064) and 2080s (2065–2094). LULC classification for 2002, 2013, and 2024 was conducted using a Random Forest (RF) classifier on Google Earth Engine, while future LULC for 2050 and 2080 was predicted using RF integrated with Cellular Automata–Markov Chain. Agriculture, agroforestry, and built-up areas are predicted to expand by 1.99%, 3.52%, and 0.98%, respectively, while pasture, rangeland, and barren land are expected to decline between 2024 and 2080. SWAT+ demonstrated strong performance and was used to simulate WBCs. Under SSP5-8.5 in the 2050s (2080s), precipitation, Tmax, and Tmin are projected to increase by 1.36–3.36 (4.29–5.11) mm/year, 0.044–0.047 (0.066–0.072) °C/year, and 0.077–0.087 (0.124–0.132) °C/year, respectively. Under the climate change scenario, surface runoff (SURQ) is expected to rise by 0.45–1.13 (1.36–1.74) mm/year, and evapotranspiration (ET) by 0.94–1.93 (2.3–2.98) mm/year. Under the combined LULC and SSP5-8.5 scenario, SURQ may change by –0.14–1.3 (–0.55–1.72) mm/year, and ET may increase by 0.24–3.32 (1.95–5.07) mm/year. These hydrological responses suggest water availability but also increased risks of flooding and water scarcity, emphasizing climate-resilient and LULC-sensitive water management strategies.