Statistical analysis of recent rainfall variability and trend using a merged gauge and satellite time series data for the cotton zone of Mali

Abstract

The variability of rainfall patterns has a significant impact on people's livelihoods especially in areas where rain-fed agriculture predominates. The temporal variability and trend of rainfall was analysed using a newly rainfall time series created from the fusion with ground measurement and satellite data for ten met stations for the period 1983-2021. Mann Kendall's (MK) non-parametric test was used to verify the rainfall trend, the variability was analysed using the Coefficient of Variation (CoV), the annual distribution of precipitation was checked using the precipitation concentration index (PCI) and finally the synchronicity between stations was tested with an Kruskal Wallis (KW) test. The MK-test shows an increasing trend in precipitation for all stations even if this trend is not significant. This increase resulted in a general increase in the amount of rain received in the area. The analysis of rainfall data revealed a moderate intra and inter annual variability (CoV± 20%). The PCI also revealed an irregularity of rainfall due to its high seasonality. The analysis of variance highlighted an asynchrony of the stations. These results provide an in-depth understanding of the recent rainfall variation and trend in the cotton zone of Mali.