资源环境科技发展态势分析平台

The socio-economic consequences posed by climate change in Africa are giving increasing emphasis to the need for trend analysis and detection of changes in hydro-climatic variables in data deficient areas. This study analyses rainfall data from 17 rain gauges unevenly distributed across the Logone catchment in the Lake Chad basin over a 50-year period (1951-2000). After quality control of the rainfall data using homogeneity tests, nonparametric Mann-Kendall and Spearman's rho tests were applied to detect the presence of trends. Trend magnitude was calculated using Sen's slope estimator. Results of the homogeneity test showed that rainfall was homogeneous across the catchment. Trend analysis revealed the presence of negative trends for annual rainfall at all the stations. Results of long-term trend analysis at a monthly time scale revealed the presence of statistically insignificant positive trends at 32% of the stations. Spatially, the analysis showed a clear distinction in rainfall magnitude between the semi-arid and Sudano zones. The slope of the trend lines for annual rainfall averaged over the respective zones was higher in the semi-arid zone (-4.37) compared to the Sudano zone (-4.02). However, the station with the greatest reduction in annual rainfall (-8.06 mm) was located in the Sudano zone.