The influence of oceanic and atmospheric drivers on Djibouti’s rainfall variability

Le résumé fourni par la source

Abstract Rainfall variability presents a major challenge for climate-sensitive sectors in arid and semi-arid regions such as Djibouti, where livelihoods predominantly rely on rain-fed systems. This study provides a comprehensive assessment of seasonal and interannual variability in rainfall and temperature, emphasizing spatial patterns and the underlying oceanic and atmospheric drivers. Using station observations, satellite-based precipitation datasets (CHIRPS), and global reanalysis products, we characterize the seasonal climatology, dominant rainfall modes, and associated linkage with global and local drivers over the period 1981–2024. The climatological analysis reveals a bimodal rainfall regime, with primary peaks during July–September (JAS) and secondary peaks in March–May (MAM), reflecting the seasonal migration of the Intertropical Convergence Zone (ITCZ). JAS accounts for ~ 45.8% of annual rainfall and, in some regions, extends into November, contributing over 60% of the annual total. Rainfall exhibits a westward gradient, with the western highlands receiving the highest amounts, while temperature extremes peak in June–September, exceeding 40°C in lowlands, and decline moderately during December–February (DJF). Rotated Empirical Orthogonal Function (REOF) and Rotated Principal Component (RPC) analyses identify consistent spatial zones of rainfall variability across seasons. During MAM, northeastern, western, and central regions dominate variability, explaining 33.6%, 31.8%, and 26.7% of variance, respectively. In JAS, western Djibouti contributes 54.2% of variance, highlighting strong monsoonal influence. The ON season is highly fragmented, with the eastern tip accounting for 40.6% of variance, followed by the northwestern (32.8%) and south-central zones (17.6%), illustrating erratic rainfall patterns along coastal regions and extension of JAS in other areas. DJF rainfall is more spatially coherent, with the east-central (36.6%), western (28%), and central (25.8%) regions contributing most of the seasonal variance, shaped by extratropical troughs, the Red Sea Convergence Zone (RSCZ), and orographic lifting. Correlation analyses between seasonal RPCs and global Sea Surface Temperature (SST) anomalies reveal strong connections with the western Indian Ocean, Red Sea, and Gulf of Aden, which serve as key moisture sources. Large-scale climate modes, particularly the Indian Ocean Dipole (IOD) and El Niño–Southern Oscillation (ENSO), significantly influence rainfall seasonality. Positive IOD phases enhance ON rainfall via increased westerly moisture transport, while La Niña events tend to amplify JAS precipitation through displacement of the Walker circulation, which facilitates lifting of moisture-laden air entering the country. The findings underscore the complex interactions between ocean-atmosphere dynamics and local topography that govern spatiotemporal rainfall variability. Identifying dominant rainfall zones and their associated climate drivers provide critical insights for improving seasonal forecasting, early warning systems, climate-resilient planning, and targeted adaptation strategies in Djibouti and the broader Horn of Africa.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.