Tropical rivers provide essential services and ecological functions for society and ecosystems such as regulating nutrient cycle, maintaining fishery production, water supply, recreation and tourism, generation of hydropower, and support for a range of terrestrial and aquatic biodiversity (e.g., Bunn et al., 2006; Gidley, 2009; Keddy et al., 2009; Kennard et al., 2010; Ndehedehe et al., 2020b, c; Tockner et al., 2010; Zhao et al., 2012). Process‐based knowledge of the cascading impacts of extreme events such as drought on hydrology is crucial and can directly feed into management and policy frameworks. Because large‐scale hydro‐climatic fluctuations and decadal‐scale droughts impact hydrological regimes, a key focus of this chapter is to improve understanding on the response of the freshwater ecosystem to extreme drought and the role of climate variability on the terrestrial hydrology of the Congo Basin. This knowledge is important to help highlight the contributions of human activities such as deforestation and land cover change on surface water hydrology.
In other large watersheds and river basins, multiple lines of evidence confirm significant large‐scale alteration of hydrological processes caused by several human activities, including surface water developments for agriculture and hydropower and water diversion (e.g., Ndehedehe et al., 2019; Wada et al., 2017). For instance, Lake Volta, the largest man‐made lake, contributed 41.6% to the observed increase in GRACE‐derived TWS over the Volta basin during the 2002–2014 period when there was an apparent fall in precipitation (see, Ndehedehe et al., 2016, 2017a). Lake Victoria is the largest lake in Africa and as recently demonstrated, its water storage variability is dam controlled, contributing about 64% of TWS variability to its basin (Getirana et al., 2020). Arguably, the water resources in several river basins in Africa are generally being disturbed by natural variability, large‐scale ocean‐atmosphere phenomena, and a combined human‐induced factors, e.g., land use changes and surface water schemes (e.g., Descroix et al., 2009; Moore and Williams, 2014; Ngom et al., 2016; Redelsperger & Lebel, 2009). The impacts of these interventions have always altered surface water hydrology, culminating in complex hydrological processes and or increased variability in these regions (e.g., Gal et al., 2017; Li et al., 2007; Mahé and Olivry, 1999; Mahé & Paturel, 2009).
Apparently, the Congo Basin contains some of the largest areas of the world’s tropical forests and wetlands, which are considerably important to global carbon and methane cycle (Achard et al., 2002; O’Loughlin et al., 2013). And within the context of global environmental change triggered by various human actions and climate variability, the Congo Basin, which is home to the largest river in Africa and contains about 18% of the world’s tropical forests (e.g., Achard et al., 2002; Becker et al., 2018; Ndehedehe et al., 2018b; Verhegghen et al., 2012) are also vulnerable to multiple influences of human actions and climate change. The main contribution of this study therefore is to improve contemporary understanding on the influence of climate variability on surface water hydrology in the Congo Basin. Specifically, this study (i) investigates the characteristics of extreme events and land water storage using GRACE observations and multi‐scaled indicators and (ii) predicts the influence of global climate on surface water hydrology by integrating multivariate analysis with support vector machine regression. Although in this era of the Anthropocene where combined climate and human actions are leading drivers of environmental change, global hydrological hotspots such as the Congo Basin will experience more climatic disturbance due to the influence of the tropical oceans, physical mechanisms, and climate teleconnections. These factors regulate precipitation and the transport of moisture and will be the vehicle by which climatic extremes will be delivered across the basin and its environs. This chapter will therefore focus on exploring the interactions and links between land water storage (surface water hydrology) and global climate using sea surface temperature, GRACE‐derived TWS, and standardized precipitation evapotranspiration index (SPEI) data. Further details on data, statistical analysis, and modeling employed in this chapter are highlighted in subsequent sections.
5.2. MATERIALS AND METHOD
5.2.1. Terrestrial Water Storage
This study used three GRACE mascon (mass concentration) solutions from JPL, CSR, and GSFC and was accessed from the Center for Space Research (CSR) at The University of Texas through its data portal (http://www2.csr.utexas.edu/grace/RL05_mascons.html). Generally, Mascons solves for monthly gravity field variations in terms of a 120‐km wide mascon block (Save et al., 2016, Wiese et al., 2016, Watkins et al., 2015). GRACE solutions based on the so‐called mascon from different processing centers at the Center for Space Research (CSR), the Goddard Space Flight Center (GSFC), and Jet Propulsion Laboratory (JPL) were considered for estimating TWS fields. The CSR solution describes the global mass changes expressed in TWS solved for 40,962 cells in which each has an approximately 12,400 km2 with the average distance of about 120 km between the cells and finally resampled into 0.5°×0.5° (Save et al., 2016). The GRACE GFSC mascon solution is solved for 1°×1° equal‐area grid blocks, in which there are 41,168 mascon blocks covering the entire globe with mean area of 12,389 km (Luthcke et al., 2013). The JPL mascon solution solves for monthly gravity field variations in terms of 4,551 equal‐area 3‐degree spherical cap mascons covering the time of April 2002 to June 2017 and are also resampled into a fine resolution of 0.5°×0.5° (Watkins et al., 2015).
5.2.2. Surface Water Storage Hydrology
Surface Water Storage
Using hydrological models, Getirana et al. (2017a) decomposed the TWS variability into its four major components: surface water storage (SWS), groundwater storage (GWS), soil moisture (SM), and snow water equivalent (SWE). Two state of‐the‐art models, the Noah land surface model (LSM) with multi‐parameterization options (Noah‐MP Niu et al., 2011) and the Hydrological Modeling and Analysis Platform (HyMAP) river routing scheme (Getirana et al., 2012, 2017b), are combined in order to represent the physical processes controlling TWS dynamics. Noah‐MP is a multi‐physics version of the community Noah LSM (Ek et al., 2003). As in most LSMs, Noah‐MP maintains surface energy and water balances while simulating direct evaporation from soil, transpiration from vegetation, evaporation of interception and snow sublimation, and estimating key surface energy and moisture prognostics such as land surface temperature, snowpack, soil moisture and soil temperature. In addition, Noah‐MP incorporates a three‐layer snow physics component and a groundwater module with a prognostic water table (Niu et al., 2011). HyMAP is a state‐of‐the‐art global scale river routing scheme capable of simulating surface water dynamics in both rivers and floodplains using the local inertia formulation (Bates et al., 2010; Getirana et al., 2017b), derived from the full hydrodynamic equations. The local inertia formulation accounts for a more stable and computationally efficient representation of river flow diffusiveness, essential for a physically based representation of wetlands, floodplains, and backwater effects. Noah‐MP and HyMAP are one‐way coupled. This means that, at each time step, gridded surface runoff and baseflow output from Noah‐MP are transferred to HyMAP and used to simulate spatially continuous surface water dynamics. No information is returned from HyMAP to Noah‐MP. Several meteorological and precipitation datasets were used as model inputs, resulting in a 12‐member ensemble model output. Here, the ensemble mean is used as the reference. The output from this model is used in this study as a surrogate for the
SWS over the Congo Basin.