and feedbacks and the importance of the Congo forest to its local hydrology and precipitation (Bell et al., 2015; Koster et al., 2004), could indeed induce considerable changes in hydrological regimes of the Congo Basin. Even though the physiographic characteristics of rivers connecting to the Congo River do have complex drainage systems that could create a non‐stationary relationship between surface water flow and rainfall (e.g., Ndehedehe et al., 2019), the terrestrial hydrology of the Congo Basin is directly regulated by the prolonged seasonal rainfall within the Congo Basin. For example, rainfall patterns over the Congo Basin are linearly correlated with the Congo River discharge (e.g., Conway et al., 2009). But from a sub‐regional analysis that included West Africa, the river discharge explained a considerable proportion of GRACE‐terrestrial hydrological signal in the Congo Basin (Ndehedehe et al., 2018b). Arguably, this relationship gives the notion that sink terms (runoff and evapotranspiration) in the basin are also key drivers of surface water hydrology other than rainfall. Locally recycled precipitation caused by the combined influence of the nearby ocean and evaporation from the Congo Basin (Dyer et al., 2017; Sorí et al., 2017) are further evidence supporting the argument of other hydrological drivers in the basin. Moreover, the observed change in hydrological response of the Congo river to strong deviations in rainfall suggests non‐linear interactions and complex hydrological processes in the basin. For instance, changes in the temporal series of discharge do not completely reflect those of observed land water storage. Although it is less debated that the waters of the Congo Basin are directly supplied by rainfall, changes in the surface water of the basin contribute significantly to variations in GRACE‐hydrological signals. Multi‐satellite assessments of the Congo terrestrial hydrology from recent studies (Becker et al., 2018; Ndehedehe et al., 2018b) agree that this is the case.
Figure 5.8 Surface water storage over the Congo Basin during the extreme drought period of 2004 (a, c, e, and g) and the 2007 wet episodes (b, d, f, and h).
Extreme negative anomalies in rainfall impacts surface water hydrology through a trickle‐down effect that culminates in soil moisture and hydrological droughts. While processes such as seasonality effects, catchment, and climate characteristics tend to influence drought propagation, strong precipitation deficits in tropical climates would normally result in reduced alimentation and temporary decrease in stream flows, storage reservoirs, and freshwater stocks (e.g., Kiem et al., 2016; Ndehedehe, 2019; Van Loon et al., 2014). However, it has recently been shown that this was not the case in the Congo Basin (1995–2010) as most drought episodes were inconsistent with discharge anomalies during the period (Ndehedehe et al., 2019). One wonders if there are known physical and ecological processes that play key roles in drought propagation in the Congo Basin. But the basin’s catchment stores (e.g., swamps, lakes, reservoirs, soil column, groundwater, etc.), which could create a prolonged reservoir memory in the hydrological system, could be a determinant in the delayed propagation of drought signals or even its absence in the discharge anomalies. It has been reported that the Congo Basin is the only river basin that seconds the Amazon River in terms of average yearly discharge (i.e., about 40,200 m3/s), and surface water storage (111 km3) (see Alsdorf et al., 2010; Lee et al., 2011). This storage capacity could increase catchment response time to drought events, and arguably create a non‐linear relationship that results in an asymmetric response of surface water dynamics to a drought signal (e.g., Loon, 2013; Ndehedehe et al., 2019). Although antecedent conditions could exist, this relationship can be disturbed or altered in the event of strong human footprints (e.g., deforestation), land surface conditions, and increased frequency in drought events triggered by changes in atmospheric circulation patterns. In other words, rainfall may not be the only driver of hydrological conditions and fluxes in the Congo Basin. Earlier studies have recognized rainfall as a key indicator regulating the hydrology of the region. However, river basin physiography and properties (e.g., topography, streamflow characteristics, etc.) and several ongoing human actions such as the effects of land use change and deforestation in the Congo Basin drive variability in river flows and surface water availability.
5.4.2. Surface Water Hydrology of the Congo Basin and the Role of Climate
Aerial averaged time series of TWS over the Congo basin between 2002 and 2017 showed no significant trend. But within the basin, the leading spatiotemporal mode of TWS, accounting for about 78% of the total variability, is dominated by annual signal, which coincides with annual fluctuations in rainfall. While the Congo River signal is also identified in the GRACE‐hydrological signal over the Congo Basin, there was a fall in TWS between 2003–2005 and a subsequent rise during the 2006–2017 period. These trends, though spatially explicit, are very consistent with both temporal drought patterns and the percentage of drought‐affected areas observed during the same periods. In fact, there was a relatively higher distribution of surface water inundation within the Cuvette centrale and floodplain corridor of the Congo Basin during wet years unlike dry years when rainfall was restricted. TWS variability are mostly characterized by strong annual changes and multi‐annual signals. There is also a significant surface mass variation emanating from the hydrology of the surrounding East African rivers and lakes (Lakes Tanganyika, Edward, and Kivu), which share boundaries with the Congo Basin. Considering the spatial patterns of observed GRACE‐hydrological signal over this area, there is a possible indication of significant exchange of fluxes within the various watersheds of the Congo Basin. These of fluxes among freshwater bodies may contribute to flow dynamics and lead to considerable amplitudes in surface storage of the Congo floodplain and the Cuvette centrale. This argument is consistent with an earlier insinuation by Tshimanga and Hughes (2014) that the hydrology of this region and other surrounding large floodplain wetlands are expected to contribute to downstream flow regimes of the Congo River. Furthermore, the surface water hydrology of the Congo Basin has considerable connections with the surrounding oceans. Predictive scheme based on a linear SVMR show that global climate through SST anomalies of the three oceans (Atlantic, Indian, and Pacific) have linear relationships with fluctuations in the Congo river discharge. The SST of the Atlantic and Pacific are relatively stronger predictors of river discharge compared to SST of the Indian ocean. Overall, the weight of coefficients of the predictands in the SVMR model confirm the importance of slow oceanic and climate signals from global SST anomaly on hydrological changes and surface water hydrology in the Congo Basin. Previous studies have reported the links between Congo discharge and SST of the surrounding oceans. The study by Materia et al. (2012), which confirmed the effect of freshwater on SST, suggests an interplay involving river discharge, sea surface salinity, and temperature. While these factors could be significant to the interannual variability observed in the region, a recent diagnostics study shows that ENSO‐related equatorial Pacific SST fluctuations have been identified as a key climate variability index associated with land water storage (Ndehedehe et al., 2018b). Additional evidence from a recent satellite‐based assessment of surface water dynamics in the Congo Basin confirm the influence of ENSO on its surface water hydrology (Becker et al., 2018).
Moreover, the implications of persistent droughts events on tropical rainforest systems was stressed by Zhou et al. (2014). They argued that the continued drying of the basin could lead to compositional and structural changes in the Congolese forest. Other than the well‐known influence of climate variability on fluxes and terrestrial hydrology of the Congo basin (Becker et al., 2018; Conway et al., 2009; Ndehedehe et al., 2018b), recent findings on drivers of TWS in the basin suggest the critical role of human actions (Ahmed & Wiese, 2019). The conversation around human‐induced changes in TWS of the Congo Basin is important and requires further details. This is because as home to the world’s second largest rainforest block (e.g., Oslisly et al., 2013), it is critical to advance knowledge on long‐term effects of intense human actions, such as deforestation, on TWS dynamics. This will build on existing compendium of knowledge highlighting the sensitivity of climate to the loss of the Congo Basin rainforest and other ecological disturbance in the region (e.g., Bell et al., 2015; Malhi et al., 2013; Verhegghen et
