Linking reported drought impacts with drought indices, water scarcity, and aridity: the case of Kenya
Marleen R. Lam1,Alessia Matanó2,Anne F. Van Loon2,Rhoda Odongo2,Aklilu D. Teklesadik3,Charles N. Wamucii1,Marc J. C. van den Homberg3,Shamton Waruru4,and Adriaan J. Teuling1Marleen R. Lam et al.Marleen R. Lam1,Alessia Matanó2,Anne F. Van Loon2,Rhoda Odongo2,Aklilu D. Teklesadik3,Charles N. Wamucii1,Marc J. C. van den Homberg3,Shamton Waruru4,and Adriaan J. Teuling1
Received: 07 Jun 2022 – Discussion started: 17 Jun 2022
Abstract. The relation between drought severity, as expressed through widely used drought indices, and drought impacts is complex. In particular in water-limited regions where water scarcity is prevalent, the attribution of drought impacts is difficult. This study assesses the relation between reported drought impacts, drought indices, water scarcity, and aridity across several counties in Kenya. The monthly bulletins of the National Drought Management Authority in Kenya have been used to gather drought impact data. A Random Forest (RF) model was used to explore which set of drought indices best explains drought impacts on: pasture, livestock deaths, milk production, crop losses, food insecurity, trekking distance for water, and malnutrition. The findings of this study suggest a relation between drought severity and the frequency of drought impacts, whereby the latter also showed a relation with aridity, whilst water scarcity did not. The results of the RF model reveal that drought impacts can be explained by a range of drought indices across regions with different aridity. While the findings strongly depend on the availability of drought impact data and the socio-economic circumstances within a region, this study highlights the potential of linking drought indices with text-based impact reports. In doing so, however, spatial differences in aridity and water scarcity conditions have to be taken into account.
There is still no full understanding of the relation between drought impacts and drought indices in the Horn of Africa where water scarcity and arid regions are also present. This study assesses their relation in Kenya. A Random Forest model reveals that each region, aggregated by aridity, has their own set of predictors for every impact category. Water scarcity was not found to be related to aridity. Understanding these relations contributes to the development of drought early warning systems.
There is still no full understanding of the relation between drought impacts and drought indices...