Green Above, Dry Below: Satellite Multi-Indices Reveal Contrasting Drought Trajectories Across the Tekeze Basin (2000–2025)

Abstract. Drought is one of the most pervasive and damaging climate-related hazards in the Horn of Africa, with profound impacts on agricultural productivity, water resources, ecosystem functioning, and socioeconomic development. The Tekeze Basin of northern Ethiopia is particularly vulnerable due to its high rainfall variability, recurrent dry spells, and dependence on rain-fed agriculture. Although numerous drought studies have been conducted in Ethiopia, few have simultaneously integrated vegetation, thermal, soil moisture, precipitation, and surface water indicators to provide a comprehensive assessment of drought dynamics. This study presents a multi-dimensional spatiotemporal analysis of drought conditions in the Tekeze Basin (2000–2025) using seven satellite-derived drought indices: the Vegetation Condition Index (VCI), Temperature Condition Index (TCI), Vegetation Health Index (VHI), Temperature Vegetation Dryness Index (TVDI), Soil Moisture Condition Index (SMCI), Precipitation Condition Index (PCI), and Normalized Difference Water Index (NDWI). The indices were derived from MODIS, soil moisture, and CHIRPS precipitation datasets. Long-term pixel-wise temporal trends were assessed using the non-parametric Mann-Kendall (MK) test and Sen's slope estimator across the growing season (April–September), while inter-index relationships were examined through monthly Pearson correlation analysis. Results reveal that VCI, TCI, VHI, and NDWI exhibited predominantly positive trends over the study period (71.5–86.4 % positive spatial coverage), indicating general improvement in vegetation and water conditions across most of the basin. In contrast, TVDI and SMCI showed dominant negative trends (82.2 % and 55.3 % negative coverage, respectively), indicating increasing land-surface dryness and declining soil moisture, particularly in the basin interior and lowlands. PCI displayed spatially heterogeneous patterns (54.7 % negative coverage), signaling progressive precipitation deficits in the northern sub-basin. Monthly correlation analysis revealed strong positive relationships between TCI and VCI (r = 0.48–0.97) and strong negative relationships between TVDI and VHI (r = −0.54 to −0.97) across all months. SMCI and PCI showed strengthening positive intercorrelations during the peak rainy season (June–September). The contrasting trajectories observed between vegetation-based indicators and soil moisture-related indices suggest a potential decoupling between vegetation response and underlying hydrological conditions. These findings demonstrate the value of multi-index remote sensing approaches for capturing the complex and multidimensional nature of drought and provide important scientific evidence for drought early warning, water resource management, and climate adaptation planning in northern Ethiopia.