An increase in temperature under the shared socioeconomic scenarios in the Volta River Basin, West Africa: implications for economic development

The basic environmental parameter of climate change is the increase in the average temperature around the world due to global warming. Since 1900, the average surface air temperature has increased by around 1 °C due to global warming, with more than half of the rise occurring in the mid-1970s (Wolff et al. 2020). While global warming leads to a global-scale increase in the average temperature, which has become a global concern, there are varying increases in temperatures at spatial scales, giving rise to varying local concerns because of the broader set of effects on human society and ecosystems associated with rising temperatures.

The effects of rising temperatures on human society, economic development, and the environment across different spatial scales are diverse and well documented in literature (Hsiang et al. 2013; IPCC 2014; Pachauri et al. 2014; Watts et al. 2015; Diffenbaugh & Burke 2019). The conclusions drawn by existing studies are generally similar, especially for neighbouring regions or areas, that a global-scale increase in temperature and its associated effects have adverse impacts on national economies, social life, and the environment (Wartenburger et al. 2017; Sun et al. 2018). For instance, rising temperatures lead to a considerable decline in poor countries' economic growth because of their extensive adverse effects on agricultural and industrial productivity (Dell et al. 2012). Employing data from 174 countries over the period 1960–2014, it is observed that changes in temperature result in reduced per-capita real output growth (Kahn et al. 2019).

Lee et al. (2020) establish that in Asia the impact of rising temperatures extends beyond the direct and significant adverse effects on agricultural productivity as it considerably affects overall economic performance through declining industrial production and investment activities. Their study further reports a projected minimum decline of about 10% in the overall economic performance of developing Asian countries by the year 2100 under a business-as-usual scenario (Lee et al. 2020). Similar trends have been reported in Africa. A study in 46 African countries, including the West African sub-region, shows that average economic growth will decline by about 1.5 and 3.2% on the experience of a 1 °C increase in temperature and a temperature shock, respectively (Odusola & Abidoye 2015). The study further observes that the impact of temperature changes on the gross domestic product (GDP) across the 46 African countries resulted in a decline of about 1.2% to 1.8% (Odusola & Abidoye 2015). In Ghana and other economies in the sub-region, existing studies have also reported potential adverse climate impacts, including a reduction in crop and animal production and the associated decline in household income and food security, resulting in rising poverty, especially at the household level in a rural community (see Asare-Nuamah 2021).

As cautioned by the Intergovernmental Panel on Climate Change (IPCC), climate change impacts will be harsher and more severe in developing economies if robust adaptation strategies are not implemented. Studies show that on- and off-farm adaptation strategies such as the application of improved crop varieties and agrochemicals, as well as crop and livelihood diversification, have been employed among smallholder farmers in Africa to mitigate adverse climate change impacts (Kumasi et al. 2019; Asare-Nuamah & Amungwa 2021). However, considering these observed impacts of rising temperature increases on socioeconomic development, coupled with the high vulnerability of the economy and society in the Volta River Basin (VRB) due to their overdependence on rain-fed agriculture, understanding temperature variations at various spatial scales is crucial for informed adaptation and policymaking. This includes producing robust future temperature projections that will be useful in informing policy-makers on the type of adaptation strategies necessary to ensure sustainable economic growth. Such an understanding of temperature increase is particularly important at the river basin level, such as the VRB, where most of the economic activities of the inhabitants from riparian countries are hosted.

In the VRB and its sub-basins, there has been a growing literature on temperature changes and trend analysis over the past decade (e.g., Neumann et al. 2007; Kabo-Bah et al. 2016; Aziz & Obuobie 2017; Annor et al. 2018; Okafor et al. 2019; Abungba et al. 2020; Larbi et al. 2021). Though most of these studies have focused on temperature projections under different climate scenarios, the data used were from the Coupled Model Intercomparison Project, Phase 5 (CMIP5) models, in which socioeconomic parameters, critical for understanding climate change impacts and making informed decisions, were not considered in the models. The few existing climate studies (e.g., Ajibola et al. 2020; Klutse et al. 2021) over West Africa and the VRB using data from CMIP6 models only evaluated the performance of the models without linking them to socioeconomic implications on the basin. Hence, using CMIP6 models for temperature projections over the VRB and linking them to socioeconomic development is imperative but lacking. This study, which aims to understand temperature variations at various spatial scales and their potential impact on economic activities in the West African VRB, fills the existing gap. The study offers new insights into the potential impacts of climate change on the VRB by specifically considering both the physical and economic implications. It is crucial for global readers interested in understanding regional climate change impacts and developing effective adaptation strategies, as well as for policymakers seeking high-quality research to inform decision-making and policy guidelines. Specifically, the study seeks to (i) evaluate the performance of CMIP6 models in reproducing the temperature patterns over the VRB and (ii) project temperatures across the basin for the future period (2021–2050) under two Shared Socioeconomic Pathway Scenarios (SSP2-4.5 and SSP5-8.5). The rest of the sections of the paper present the materials and methods used in conducting the study, the results and discussion, the implications of the findings on economic activities, and the conclusion.

Observed temperature data for the period 1985–2014 at the daily scale used in this study was obtained from the 50-km resolution NASA POWER satellite-climate product. The choice of the satellite-climate product is due to the poor spatial network of climate stations and the lack of continuous climate data on a temporal scale. NASA POWER data is frequently used as observational data, and its robustness has been demonstrated in many research studies across Africa (e.g., Larbi et al. 2018; Mwabumba et al. 2022). Nevertheless, validation of the NASA POWER temperature data over the Guinea Coast, Savanna, and Sahel regions within the VRB at the monthly scale (Supplementary material, Appendix 1) was further performed using station data before the NASA POWER data were used as observations. The validation of the NASA POWER temperature data against station data was necessary to ensure data accuracy, establish data reliability, assess regional applicability, and justify the use of the dataset for the study analysis.

The simulated temperature datasets at 100 km spatial resolution are from the CMIP6 simulations. The CMIP6 is a global research initiative that aims to enhance our understanding of Earth's changing climate and its future impacts (Eyring et al. 2016). It represents the latest phase of a long-standing collaborative effort among modelling groups worldwide. CMIP6 involves the utilisation of cutting-edge climate models to simulate the Earth's climate under various scenarios of future greenhouse gas emissions. CMIP6 incorporates advancements in climate science, including new data on past and present climate conditions, improved knowledge about the physical processes driving climate change, and enhancements in computing capabilities and modelling techniques. By conducting a range of experiments, CMIP6 simulates different aspects of the climate system, encompassing the atmosphere, oceans, land surface, and cryosphere (ice and snow). These experiments encompass diverse scenarios of future greenhouse gas emissions, ranging from optimistic ones with significant emission reductions to pessimistic scenarios with sustained high emissions. The outcomes of CMIP6 hold substantial value in numerous research domains, such as the examination of climate impacts, the formulation of climate policies, and the development of strategies for adaptation and mitigation. Furthermore, the CMIP6 results contribute to international policy decisions, including the United Nations Framework Convention on Climate Change (UNFCCC) negotiations (Eyring et al. 2016; O'Neill et al. 2016).

Within CMIP6, three notable global climate models (GCMs) were employed in the present research: BCC-CSM2-MR, NorESM2-MM, and MPI-ESM1-2-HR. They represent the most recent phase of a coordinated effort by modelling groups around the world.

The BCC-CSM2-MR, which stands for Beijing Climate Centre Climate System Model version 2, Medium Resolution, is a global climate model developed by the Beijing Climate Centre in China (Wu et al. 2019). It utilises the Community Earth System Model version 2 (CESM2) and operates at a horizontal resolution of 1.125° × 1.125° with 45 vertical levels. The model integrates five primary components: atmosphere, land surface, ocean, sea ice, and aerosols. The BCC-CSM2-MR employs the finite volume method for discretizing equations of motion and incorporates physical parameterisations for clouds, radiation, and turbulence. Additionally, it encompasses a carbon cycle that accounts for both terrestrial and oceanic carbon cycles. This model has been instrumental in simulating future climate change scenarios, such as the RCPs and SSPs (Wu et al., 2019).

The NorESM2-MM, representing Norwegian Earth System Model version 2, Medium Resolution, is a global climate model developed by the Norwegian Climate Centre in Norway (Ashfaq et al. 2022). It is based on the CESM2 and operates at a horizontal resolution of 1.9° × 2.5° with 32 vertical levels. Similar to the BCC-CSM2-MR, the NorESM2-MM comprises five components: atmosphere, land surface, ocean, sea ice, and aerosols. NorESM2-MM employs the spectral transform method to discretise equations of motion and incorporates physical parameterisations for clouds, radiation, and turbulence. It also includes a carbon cycle that considers both terrestrial and oceanic carbon cycles. The NorESM2-MM model has been widely used to simulate future climate change scenarios, such as the Representative Concentration Pathways (RCPs) and SSPs (Ashfaq et al. 2022).

The MPI-ESM1-2-HR, or Max Planck Institute Earth System Model version 1.2, High Resolution, is a global climate model developed by the Max Planck Institute for Meteorology in Germany (Mauritsen et al. 2019). It is based on the CESM2 and operates at a remarkable level of detail and resolution (Gutjahr et al. 2019).

The three selected GCMs are among 21 GCMs whose performance has been evaluated over West Africa by Klutse et al. (2021) and were found to perform well over the region. In addition, the models were selected by considering their equilibrium climate sensitivity (ECS) values, which are the anticipated long-term warming following a doubling of CO₂ concentrations in the atmosphere. The selected GCMs have relatively low ECS in the range of 1.9–3.0, which is consistent with the Fifth Assessment Report (AR5) range. The new CMIP6 scenario-based future projections, also known as ScenarioMIP, are based on five SSPs combined with different forcing levels for eight main scenarios. In this study, outputs from emission scenarios that combine intermediate societal vulnerability with an intermediate forcing level (SSP 2-4.5, i.e., an update of RCP4.5) and a fossil-based economy scenario (SSP5-8.5, i.e., an update of RCP8.5) were used.

SSP2-4.5 represents a future world where there is moderate socioeconomic development and gradual shifts towards lower greenhouse gas emissions. It assumes a world with increasing urbanisation, improved education, and economic development, leading to a decline in population growth rates. Energy sources gradually transition to lower-carbon options, such as natural gas and renewables, and there is a gradual increase in energy efficiency and sustainable practices (Riahi et al. 2017; IPCC 2021).

SSP5-8.5 represents a future world where there is high population growth, rapid economic expansion, and a heavy reliance on fossil fuels. It assumes a fragmented world with limited international cooperation and technological advancements, resulting in high energy demand and substantial greenhouse gas emissions. The scenario reflects a continuation of current trends and policies with limited emphasis on climate change mitigation or adaptation efforts (Riahi et al. 2017; IPCC 2019).

SSP 2-4.5 has a warming scenario in the range of 2.1 to 4.3 °C while SSP5-8.5 is in the range of 3.8 to 7.4 °C. The data used were for the historical (1985–2014) and future (2021–2050) periods.

To evaluate the capabilities of the CMIP6 models in simulating the observed temperature over the Guinea Coast, the Savannah, the Sahel, and the entire VRB, the simulated values were compared with the observed monthly and annual scales for the period 1985–2014. The annual cycle of the monthly mean analysis was used to assess how well the CMIP6 models reproduced the temperature patterns over the area. The Taylor diagram was used to further evaluate the performance of the CMIP 6 models at a monthly scale using indicators such as root mean square error (RMSE), normalised standard deviation (σ) and Pearson correlation coefficient (r) (Taylor 2001). The RMSE, σ, and r (Equations (1)–(3)) represent the temporal errors in the models, the temporal pattern, and the temporal variability, respectively. At the spatial scale, the biases between the models and the observations were also estimated, with positive and negative values indicating overestimation and underestimation of the observation data, respectively.

Temperature analysis for the past (1985–2014) and future (2021–2050) periods was performed at both monthly and annual scales using the ensemble mean of the three CMIP6 models. The selection of the ensemble mean is intended to minimise the uncertainties of future temperature projections (Yan et al. 2015). The projected absolute changes in the mean annual temperature at the temporal scale were estimated by determining the difference between the mean historical (1985–2014) climate and the future (2021–2050) climate under the SSP2-4.5 and SSP5-8.5 scenarios, and the obtained changes, whether significant or not, were assessed at a 95% confidence level using the t-test. The three CMIP 6 models were extracted to point scale using the R programming language. The Inverse Distance Weighting (IDW) interpolation technique was then used to map the spatial distributions of the changes in the past and future temperatures. IDW is performed with the assumption that the attribute value of an unknown location is the weighted average of a known location by assigning values to the unknown location using values from known neighbouring locations based on the concept of distance weighting.

To obtain an idea about the trends in the temperature in the entire VRB and the three zones for the period 1985–2050, the non-parametric Mann–Kendall (MK) test was computed at a 5% significant level. The MK test has been utilised in numerous studies (e.g., Okafor et al. 2017; Larbi et al. 2018; Nyembo et al. 2020) and shown to be appropriate for non-normally distributed hydro-meteorological data. The MK test compares the null hypothesis (Ho), which states that there is no trend, with the alternative hypothesis (H1), the existence of a trend (Önöz & Bayazit 2003). The MK test results that are positive or negative imply a rising or declining trend, respectively. Theil-Sen's estimator, as modified by Hirsch et al. (1982), was used to estimate the magnitude of the trend. Uncertainties associated with the projected changes in temperature by the three GCMs and their ensemble mean were assessed using the Violin plot. The interquartile range (IQR) of the box plot illustrates how dispersed the data are and is used as a measure of uncertainty. A wider box plot indicates a greater IQR, which suggests greater projection uncertainty.

The spatial distribution of the biases in the annual temperature at the basin is also shown in Figure 3. While BCC-CSM2-MR and MPI-ESM1-2-HR show overestimation in the northern part of the basin and underestimation in the southern part, the reverse situation was observed for NorESM2-MM. For example, the BCC-CSM2-MR overestimated temperature by up to 2.0 °C in the Savanah zone, 3.0 °C in the Sahel zone, and underestimated up to −0.5 °C in the Guinea Coast. The biases found in MPI-ESM1-2-HR range from −1.6 to 3.5 °C, and those found in NorESM2-MM range from −1.2 to 2.3 °C. In general, the biases found in the ensemble mean of the three models over the basin range from −0.4 to 2.0 °C, which are lower than those found in the individual models.

Table 2 presents the trend analysis results for the mean annual temperature over the VRB from 1985 to 2050 under the two scenarios of climate change. The results show a statistically increasing trend in temperature in both the three zones and the entire basin.

Under the SSP5-8.5 scenario as shown in Figure 5, the temperature change in the Sahel zone is within the range of 0.4 to 2.5 °C with an IQR of 0.7 °C for the ensemble mean, 0.9 to 3.2 °C with an IQR of 0.9 °C for MPI_ESM, 0.4 to 2.5 °C with an IQR of 0.9 °C for BCC_CSM2, and 1.0 to 3.4 °C with an IQR of 0.7 °C for NorESM. In the Savannah zone, the IQRs obtained for the ensemble mean, MPI_ESM, BCC_CSM2, and NorESM are 0.5, 0.5, 0.6, and 0.8 °C, respectively.

Owing to the need for accurate projections in climate change studies, the GCMs' ability to mimic the patterns of observed temperature data using statistical indicators such as RMSE, normalised standard deviation (σ), and Pearson correlation coefficient (r) was assessed in this study. Despite indications of overestimation in each of the zones and the entire Volta basin, the temperature patterns were well captured by all models, including the ensemble mean. These, together with the overall good figures of the statistical indicators used, lend credence to the robustness of the CMIP6 GCMs outputs. This is consistent with the observation by Fan et al. (2020) that there has been an improvement in the performance of the CMIP6 GCMs for modelling temperature in comparison with the CMIP5.

The determination of the biases between the models and the observations is also a key measure of the robustness of GCM outputs. The temperature biases were estimated at the annual timescale, where the relative systematic error associated with the CMIP6 GCMs data was described. The CMIP6 GCMs that showed overestimation and underestimation in the northern and southern parts of the VRB, respectively, and vice versa, could probably exaggerate the effects of global atmospheric CO₂ concentration (ACC), which has strong spatial variability as it depends on population growth and economic development among regions (Zhang et al. 2019). In their study, Zhang et al. (2019) argue that, though it is widely acknowledged that an increase in ACC is the primary cause of climate change, the implications of its spatial distribution on the climate system are still unclear, and therefore its effects over a particular region could be exaggerated by GCMs, especially when simulating temperature.

Understanding the current and future climate trends through climate projections is essential for informed adaptation and policymaking, particularly in vulnerable economies such as Ghana, where rain-fed agriculture, which is susceptible to climate change, is a major source of livelihood and contributes essentially to economic development. Specific to the VRB, apart from its contribution to agricultural activities, it also serves as a major source of hydropower in Ghana and other riparian states. Hence, understanding past, current, and future temperature variations across the basin is imperative. There is no denying that West African economies are agrarian, and achieving economic growth and sustainable development on the continent would require addressing the challenges facing the agriculture sector. One of the challenges confronting the agriculture sector in West Africa is climate change, partly due to the rain-fed, dependent nature of West African agricultural systems coupled with low adoption of technology and innovation and high poverty levels among smallholder farmers, who constitute the majority of the labour force employed in the sector. In the VRB, agriculture is the dominant economic activity and accounts for 40% of the basin's economic activities. The impact of high temperatures on agriculture has a high likelihood of reducing the sector's contribution to economic growth, particularly GDP and welfare (Arndt et al. 2015). Already, the agriculture sector's contribution to GDP in part of West Africa has dwindled in recent times, and climate change is likely to worsen it further (MOFA 2019; Adeosun et al. 2021). Agriculture is the sector that ensures food security, generates jobs, and provides raw materials to the manufacturing sector. According to predictions made by Arndt et al. (2015), maize crop yields in Ghana are predicted to fall by around 18% by 2050, and agriculture's share of the overall GDP decreases by about a third, from 35.1% in 2007 to 20.1% in 2050. Crop losses will lead to decreased productivity and higher food prices (Arora 2019), with the poor and women being the most negatively impacted. Such a trend would be disastrous to the economic development of the sub-region in general and of the riparian states in particular. Adusola & Abiyole (2015) also note that temperature and temperature shocks are associated with about a 1.58 and 3.22% decline in economic growth, respectively. Similarly, existing projections indicate that the West African economies are likely to observe a reduction in the contribution of agriculture to GDP and socioeconomic development due to climate change (Oxford Business Group 2021).

Consequently, high temperature patterns coupled with unpredictable and unreliable rainfall trends will have serious negative implications for the economic development of the agrarian economies along the VRB. This is because agriculture in these riparian states is highly sensitive to climate change, albeit contributing an average of about 14% to GDP (AGRA 2021). Any decline in agriculture productivity would first translate into a reduction in the contribution of agriculture to economic growth. Such a trend would be disastrous for development on the continent in general and in the riparian states in particular. This is because existing projections indicate that the West African economies are likely to observe a reduction in the contribution of agriculture to GDP and socioeconomic development due to climate change (Oxford Business Group 2021).

Further, an increase in temperature in the VRB can affect agriculture productivity, leading to a drastic reduction in agriculture yields and income (Asare-Nuamah 2021; Asante & Amuakwa-Mensah, 2015). This is problematic for the riparian states in general and the VRB communities in particular. A decline in agriculture productivity will have serious implications for food insecurity, which is important for sustainable development. Several studies have shown strong associations between agriculture productivity and food security (Mozumdar 2012; Muzari 2016). Changing climatic conditions tend to alter the optimal suitability of major food crops, posing challenges for food insecurity (Chemura et al. 2020). As the main source of livelihood and food security for many poor households, any reduction in agriculture yields would have rippling effects on livelihoods and food security at both the household and national levels of the riparian states, thereby worsening hunger and food insecurity. Existing studies show that food insecurity and its associated nutritional challenges are high among many smallholder farmers in West Africa (WFP & FAO 2015). Amidst the low adaptive capacity of African farmers, smallholder farmers contribute substantially to food security on the continent by producing about 80% of the food supply (FAO 2012). However, climate change increases the vulnerability and risk of smallholder farmers, thereby affecting their potential to contribute to food security, which is critical for human development and achieving sustainable development.

Asare-Nuamah (2021) argues that poor and rural communities turn to skipping meals or eating less-preferred foods, which, according to Coates et al. (2007), are dominant among food insecure populations. High temperatures may exacerbate food insecurity through the incidence of pests and diseases in crops and livestock, coupled with poor yields. A study in Ghana by Asare-Nuamah (2020) noted that rising temperatures and poor rainfall trends have facilitated the activities of fall armyworms (FAWs) on maize farms, which concurs with the position of the African Union (2017) that the FAW invasion has resulted in an annual loss of $3 billion to maize production on the continent. Intuitively, poor crop and livestock yields spell doom for the majority of smallholder farmers and rural communities whose livelihoods depend on agriculture. There is no doubt that the rippling effect of the reduction in agriculture yields due to high temperatures will affect poverty as a result of the nexus between agriculture and household income. Gurib-Fakim (2015) argues that sustainable and mechanised agriculture can substantially cushion households out of poverty in Africa. However, a setback in the sector through reduced productivity will escalate poverty, especially at the household level.

The possibility of climate change worsening migration and conflict has been highlighted in literature (Kartiki 2011; Klepp 2017). This is largely a result of high job losses (unemployment) in the agricultural sector coupled with the displacement of communities after chronic environmental shocks such as pest invasion and drought. Rising temperatures facilitate evaporation and droughts, and evidence also points to pest invasions, such as FAWs, as a result of changing temperature conditions. The effects of such shocks push poor households out of employment in the agricultural sector, leading to their migration to urban communities. Although migration is used by smallholder farmers as a temporary coping strategy (Asare-Nuamah 2020), the continuous occurrence of climate shocks and their associated displacement of communities may facilitate permanent migration, which exerts serious consequences on resources at urban centres. In extreme cases, there is a tendency for climate shocks to result in resource conflicts due to the competing demand for the available but scarce resources. Studies have reported water and land resource conflicts among smallholder farmers and herders due to climate change, especially in West Africa (Bukari 2021). Such a situation, be it migration or conflict, poses a serious threat to economies.

In addition, rising temperatures are expected to influence the water ecosystem. Fishing is the major economic activity in communities around the VRB. Climate change is expected to cause water stress and competition for limited water resources. Increasing water temperatures will result in the extinction of some freshwater fish species. These have implications for the loss of livelihood for fishing communities within the VRB and the resultant deepening of poverty (Manful & Opoku-Ankomah 2021). Besides, changes in hydrological processes that also affect water quality tend to threaten water security, weakening the water-energy-food security nexus (Dembélé et al. 2022) and impeding access to energy, water, and food, which are important promoters of economic development. Rising temperatures in the VRB may also pose serious threats to energy supply in the region in general and especially in Ghana, where the Akosombo dam in the basin serves as the major source of hydropower in the country. This could tend to disrupt economic activities that depend largely on hydropower, thereby affecting economic growth at both household and national levels.

The expected temperature rise in the VRB requires all stakeholders, including the communities in the three zones, to take steps to build resilience, improve adaptive capacities, and implement early warning systems. This study proposes the adoption and scaling up of climate-smart agriculture (CSA) practices among smallholder farmers in the VRB through collaborative state interventions. This is necessary to improve agriculture yields and income for farmer households while minimising the contribution of African agriculture systems to the emission of greenhouse gases (Mwongera et al. 2017; Sain et al. 2017; Lopez-Ridaura et al. 2018). Studies in many agriculture communities in Africa have demonstrated the essential role of climate-smart agriculture in driving food security and poverty reduction on the continent. In Zimbabwe, for instance, the adoption and promotion of CSA, such as drought-tolerant maize, has resulted in increases in agriculture production and household food security (Makate et al. 2017). The findings of Lunduka et al. (2017) also point to a significant increase in agriculture income among smallholder farmers as a result of the promotion of drought-tolerant crops. Crop adaptation can be improved in a variety of ways, including crop rotation, irrigation techniques, crop type modifications, and the cultivation of new crops (Arndt et al., 2015). Environmentally smart agriculture practices, such as conservation agriculture, which minimises soil disturbance, improves the activities of microorganisms, and promotes crop diversification, are all essential for the sustainability of the agriculture system in the basin. Given the susceptibility of the basin to climate change, it is imperative to intensify climate information services (Antwi-Agyei et al. 2021), particularly in poor communities that depend on the basin for agriculture and other livelihood activities. This will enable communities to take proactive and reactive measures and minimise adverse climate effects. Achieving this requires a multistakeholder approach across the riparian states.

Climate change is a major challenge to achieving economic growth. This study used CMIP6 models to analyse temperature variations over the VRB in West Africa to advance our understanding of climate change and how it affects economic growth. The findings suggest that temperatures will rise across the basin under the SSP2-4.5 and SSP5-8.5 scenarios. This indicates that steps must be taken to reduce the adverse effects of rising temperatures on economic growth in the area. Most importantly, efficient and effective water and soil management practices are paramount for agriculture and other economic activities across the basin. Without a doubt, different stakeholders across the VRB will be affected by climate change, especially high temperatures, which, if unchecked, can erupt into resource conflict among the competing interest groups. Hence, adopting a multistakeholder approach across the riparian states will ensure effective needs identification and allocation of resources to meet the needs of the competing interest groups that depend on the basin for their daily livelihood strategies. To ensure the effective and efficient use of the water resources in the basin, an integrated water resources management framework should also strengthen the capacities of competing interest groups, particularly smallholder farmers and fishing communities. Through education and the provision of sufficient socioeconomic resources, this can be accomplished. It is also important to strengthen climate information services among stakeholders in the basin to enable vulnerable groups to respond effectively to adverse climate shocks and stressors such as floods and droughts in the basin.

The authors are grateful to AERC and to also the Climate Computing Community for making freely available the CMIP 6 models.

This research was funded by African Economic Research Consortium (AERC) under the AERC collaborative research project on Climate Change and Economic Development (CCED), grant number RC21585.

A.M.L, I.L., and S.-Q.D. acquired funds, did project administration, conceptualised the study, prepared the methodology, wrote and prepared the original draft. G.C.O., P.A.-N., L.K.F., A.-R.M.A., E.N.A, J.P.J.P, G.L., T.A.A., T.A.-D, N.K., and N.A.P. wrote, reviewed, and edited the article. All authors have read and agreed to the published version of the manuscript.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.