Monitoring climate and man-made induced variations in terrestrial water storage (TWS) across Africa using GRACE data

It is common practice for researchers engaged in research related to climate change to examine the temporal variations in relevant climatic parameters (e.g., temperature, precipitation) and to extract and examine drought indices reproduced from one or more such parameters. Drought indices (meteorological, agricultural and hydrological) define departures from normal conditions and are used as proxies for monitoring water availability. Many of these indices exclude significant controlling factor(s), do not work well in specific settings and regions, and often require long (≥50 yr) calibration time periods and substantial meteorological data, limiting their application in areas lacking adequate observational networks. Additional uncertainties are introduced by the models used in computing model-dependent indices. Aside from these uncertainties, none of these indices measure the variability in terrestrial water storage (TWS), a term that refers to the total vertically integrated water content in an area regardless of the reservoir in which it resides. Inter-annual trends in TWS were extracted from monthly Gravity Recovery and Climate Experiment (GRACE) data acquired (04/2002 to 08/2011) over Africa and correlated (in a GIS environment) with relevant temporal remote sensing, geologic, hydrologic, climatic, and topographic datasets. Findings include the following: (1) large sectors of Africa are undergoing statistically significant variations (+36 mm/yr to -16 mm/yr) due to natural and man-made causes; (2) warming of the tropical Atlantic ocean apparently intensified Atlantic monsoons and increased precipitation and TWS over western and central Africa's coastal plains, proximal mountainous source areas, and inland areas as far as central Chad; (3) warming in the central Indian Ocean decreased precipitation and TWS over eastern and southern Africa; (4) the high frequency of negative phases of the North Atlantic Oscillation (NAO) increased precipitation and TWS over northwest Africa; (5) deforestation in the Congo Basin and southern Tanzania decreased TWS and (6) the construction of dams (e.g., Merowe High Dam, Tekezé, Amerti-Neshi, Beles, Gilgel Gibe I, Gilgel Gibe II, and Karadobi) throughout the GRACE period increased TWS in upstream Nile Valley countries. Given the 10-year monthly GRACE record of water availability data (represented by GRACE TWS) acquired on the sub-basin scale across the globe, and the plans underway for deployment of a GRACE follow-up (2016-2026), consideration should be given to using GRACE TWS data as an alternative, viable drought index, and for monitoring the impacts of human interventions on hydrologic systems.