Spatio-temporal drought characteristics in the Limpopo basin from 1918 to 2018 - A case study based on analysis of the Standardized Precipitation Evaporation Index (SPEI)

Abstract: The identification of the characteristics of drought are of great importance in water resources planning and management. This study explored the characteristics of drought at Limpopo River Basin (LRB) from 1918 to 2018. Two time scales were used; they were the 3-month and 12-month time scales using the Standardized Precipitation Evaporation Index (SPEI), with the aim of studying the characteristics of drought over the past 100 years, analyzing its onset, termination, duration, severity, intensity and frequency. A python script based on the run theory method was used to determine the characteristics of drought in the study area. The results show that in the Limpopo River Basin (LRB) the onset and termination of drought are difficult to determine due to the complexity of the phenomena, however the beginning of drought tended not to start at a specific season while the cessation of the drought tended to end during spring or autumn. The number of droughts that last 3 months and 12 months in 100 years varies from 137 to 148 droughts and from 32 to 37 droughts respectively. The highest number of droughts was registered in South Africa and the lowest number of droughts was registered in Mozambique and Zimbabwe. The severity of the droughts in the region was also evaluated. The most severe droughts were registered in the years 1941-1942, 1946-1947, 1947-1948, 1963, 1972-1973, 1982-1983, 1990-1993, 2005-2006 and 2015-2016. Those droughts affected the entire basin or the major part of it. Further analysis about drought trends and correlation with NDVI is also analyzed in this study. The Mann-Kendall trend test analysis was used to analyze the trends of drought in the seasonal and annual scales. The results of the non-parametric Mann-Kendall trend test applied to SPEI 12-month scale indicate increasing in drought in the entire Limpopo River Basin region. The 3-month SPEI time scale was used to analyze seasonality of trends over the study area. Each country within the LRB experienced different drought transitions through the seasonal SPEI analysis, with the winter being the one with the greatest increase in drought trends. The response of vegetation to drought was evaluated by examining the correlation of the SPEI with the NDVI vegetation index from satellite images. The results suggest that the 12-month SPEI was found to have the best correlation with NDVI for the majority of the stations. However, the 3-month SPEI has also a high correlation with SPEI, but the correlation between NDVI and SPEI varies significantly between seasons. The highest correlations occurred during spring and autumn and lower correlations were observed during summer and winter for the majority of the stations.