Interannual TWS Trends in the East-African Rift Region 

Abstract

<p class="western">The German-American satellite missions GRACE (Gravity Recovery and Climate Experiment, 2002-2017) and its successor GRACE-Follow-On (GRACE-FO, since 2018) observed terrestrial water storage (TWS) variations over the continents. With 20 years of data, we can now study interannual trends and variations in water storage beyond the strong declining trends of ice sheets or glaciers. Africa is the only continent which exhibits an overall positive trend in TWS for the GRACE/GRACE-FO period. In this contribution, we analyse the interannual TWS variations in Africa and focus on the East-African Rift region around Lake Victoria, Lake Tanganyika, and Lake Turkana, where the long-term TWS increase is most pronounced.</p> <p class="western">As TWS trends are not monotonous over time, a signal decomposition into linear trend and sinusoidal annual and semiannual seasonality is insufficient to investigate interannual variability. Hence, we employ the STL method (Seasonal Trend decomposition based on Loess) to separate the TWS signals into an interannual trend signal, which is not a linear trend, a seasonal signal, and residuals. These interannual trend signals are used in a subsequent cluster algorithm to identify regions with similar interannual variability. We found complex interannual TWS signals in East Africa and many African regions. In the East African Rift region, we can observe a decrease in TWS until around 2006, after which an increasing trend started. Finally, in the last few years, the trend has further accelerated.</p> <p class="western">To better understand the origin of the observed interannual signal, we compare the TWS time series with precipitation and evaporation data and SWS data derived from satellite altimetry. The interannual variations of precipitation are insufficient to explain the strong interannual variations visible in the TWS data in Eastern Africa. SWS variations, in contrast, are highly correlated with TWS, explaining nearly 50% of the TWS variations. Among the surface water bodies, we study the influence of Lake Victoria in particular, as it is the largest lake in the region, and its water balance is also governed by a dam at its outlet. SWS of this lake is heavily affected by man-made decisions and is visible in the GRACE/GRACE-FO observations.</p>