Water resources management under climate change and anthropogenic pressure in the upper Bandama catchment in Northern Côte d'Ivoire

Water covers about 70% of the globe's surface, stored in various reservoirs, such as oceans, glaciers, atmosphere, lakes, rivers and aquifers. Around the world, its importance no longer needs to be demonstrated for human life as well as for plants and animals (Cosgrove & Loucks 2015; Mishra 2023). Nonetheless, water resources are under high pressure mainly because of climate change, population growth, groundwater depletion, energy demand rise and environmental flow requirements (Touch et al. 2020; Dembélé et al. 2022; Dembélé et al. 2023). The most remarkable consequence of climate change around the earth is both water scarcity and water-related hazards such as floods and droughts (Kopnina & Washington 2016; Payus et al. 2020; Dembélé et al. 2024).

In Côte d'Ivoire, past studies have highlighted temperature and rainfall variability, which has led to the scarcity of available water resources due to a decrease in streamflow of several rivers (Mahé et al. 2001; Soro et al. 2011). In contrast, during the recent decade, the country, particularly in the southern area, has experienced heavy rainfall with important floods events, with devastating implications including death, damage to properties and population exodus (Danumah et al. 2016; Konate et al. 2023). Moreover from 1960 to 2021, the number of inhabitants in Côte d'Ivoire grew from 3.5 million to 29.4 million, representing an increase of 739.69% (RGPH 2021). All these factors raise anthropogenic pressure over water resources thereby leading to crucial water scarcity challenges in the country (Gohourou et al. 2019; Konan et al. 2023). All the decisions taken both at the institutional and technical levels are not enough to sustainably resolve the thorny problem of the drinking water availability for rural and urban populations (M'bra et al. 2015; Koukougnon 2020; Kouamé et al. 2021).

The upper Bandama basin, located in the north of Côte d'Ivoire, is not an exception to this situation. Belonging to the Sudanian zone, the catchment is severely affected by the effects of climate change. The climate in the catchment is tropical and the rainfall is unimodal and concentrated over six months in the year. The catchment is subject to high temperatures with low humidity (60–70%) (Yapo et al. 2019; Timité et al. 2022). Its economy is based on rain-fed agriculture with a strong dependence on river flow for the city water supply, agriculture and livestock (Soro et al. 2017). Besides, the population has strongly increased with rapid urbanization, which raises the water stress. Moreover, conflicts between farmers and livestock breeders are frequent, particularly over the use of arable land and water resources (Sib et al. 2018; RGPH 2021). Addressing this challenge faced with the capacity of currently available water to meet people's needs under climate change and socio-economic pressure remains critical in the region. Many studies have been conducted in the upper Bandama basin related to hydrology, climate variability and groundwater potential (Dezetter 1991; Kouakou 2016; Ouedraogo 2016; Konate et al. 2023). So far, most studies focus on only climate change or anthropogenic impacts in the Bandama catchment. Hence, no study on the Bandama catchment evaluates the impacts of climate change and population growth on water supply by using the outputs of regional climate models (RCMs). The originality of this study consists in assessing the balance between water supply and demand under climate change and anthropogenic pressure by using the WEAP hydrological model coupled with RCMs. There are numerous examples of this model application in water resources management. Indeed, the WEAP model has been used in other areas of Côte d'Ivoire (Dao et al. 2018; Yao et al. 2021). But it has not been associated with the outputs of climate models.

The present study aims to assess the capacity of current water resources to meet the short and long-term needs of users by taking into account changes in climate and anthropogenic features. To achieve this goal, the Coordinated Regional Climate Downscaling Experiment (CORDEX) RCM data (CCLM 4-8-17 and RACMO 22T) were used to determine climate trends. Thus, the hydro-climatic, agricultural and socio-economic data have been incorporated into the WEAP hydrological model to compare water supply with the needs. Thereby, scenarios were conducted to suggest adaptive measures that can reduce the water shortages. The results of this work should help stakeholders and practitioners in making more reliable decisions for water resources management.

The database used (Table 1) includes meteorological and flow data (observed data) respectively from the Airport and Meteorological Operating and Development Company (SODEXAM) and the General Directorate of Human Hydraulic Infrastructures (DGIHH). All these data cover the period from 1962 to 2017. Socio-economics data, including population, livestock, water use rates and water consumption were provided by the water distribution company of Cote d'Ivoire (SODECI), and irrigation area and irrigation water volume were acquired from the African Company of Sugar (SUCAF). Future climate data are RCM outputs from CCLM 4-8-17 and RACMO 22T models, obtained from the CORDEX initiative and collected from the West African Science Service Center on Climate Change and Adapted Land Use (WASCAL).

The CORDEX RCM data (CCLM 4-8-17 and RACMO 22T) are obtained in gridded format stored in Network Common Data Format (NetCDF) file with a spatial resolution of 0.44° (∼50 km at the equator). These regional models have been derived from respectively the general climate models: MPI-ESM-LR and ICHEC-IC-EARTH. The variables considered from the climate model outputs are rainfall, minimum temperature and maximum temperature. The dataset contains three types of outputs. The first output is the historical simulation over the period 1951–2005. The two other outputs are future predictions under distinct greenhouse gas (GHG) emission scenarios (i.e. representative concentration pathways (RCPs)) over the period 2006–2100. In this study, the reference period is from 1986 to 2005. Whereas the future periods are 2021–2040, 2041–2060, 2061–2080 and 2081–2100. The first future prediction simulates the climate under a mid-level GHG emission scenario, called the optimistic scenario (i.e. RCP 4.5), and the second future prediction considers a high-level GHG emission, known as the pessimistic scenario (i.e. RCP 8.5).

The water evaluation and planning (WEAP) system model (Yates et al. 2005) is a user-friendly tool developed by the Stockholm Environmental Institute that can be applied to a single river basin or sub-catchments by using an integrated approach of balance between available water and various water components. For this study, the model runs by representing the hydrological system with nodes (catchment, demand sites, supply sites like rivers, groundwater and dams), which are linked through transmission links and return links. Then, the model assesses the scenarios of future water resource development under different demands.

The model has been calibrated over the period 1962–1968 and validated over the period from 1977 to 1986. The parameters considered for simulating streamflow are monthly rainfall, monthly mean temperature, crop coefficient (Kc), soil water capacity, root zone conductivity (Ks), deep conductivity, runoff resistance factor (RRF) and preferred flow direction (Table 2). The parameter values were selected from the Food and Agriculture Organization (FAO) of the United Nations database to simulate streamflow in the WEAP model.

NSE is considered satisfactory when it is greater than 0.6 (corresponding to 60%).

The water demand for domestic use depends on parameters such as the size of the sample and specific consumption. In the framework of this study, we have calculated three types of water demand for household uses:

• Water demand for domestic use of the population connected to the public drinking water supply network.

• Water demand for domestic use of the unconnected population to the public drinking water supply network.

• Water demand for domestic use of the rural population.

To run the model, the current account, which is assumed to be the starting year for all scenarios is 2012. Then, the reference period is from 2012 to 2016. Monthly data used are about climatic, hydrological, anthropogenic and physical characteristics of the study basin. Two main scenarios are developed, namely the trend scenario and the alternative scenario.

The reference or business-as-usual scenario projection is established based on trends in the basin features (demography, hydrology, water uses, agriculture, etc.).

Contrastingly, the alternative scenario is made with assumptions of how the watershed should evolve over time in case of major changes in the hydrological and socio-economics patterns. The result is a useful guide for stakeholders.

The RCMs RACMO 22T and CCLM 4-8-17 under RCP 4.5 and RCP 8.5 scenarios provide temperature and rainfall changes over the study area. The impacts of these changes on water availability in the basin are assessed by using the RCMs as input data to force the WEAP model.

The projections of climate made with the RACMO 22T and CCLM 4-8-17 models predict a temperature increase, greater in the RCP 8.5 scenario than the RCP 4.5 in the upper Bandama catchment for the future. Both models and scenarios forecast a rise of at least 1.5°C from 2030 to 2090.

The balance of the evolution of water needs and UWD every 10 years from 2020 for different scenarios is presented in Tables 3 and 4. It is noteworthy that the water requirements for the irrigation of sugarcane farms and their UWD are the highest.

This section presents measures to reduce the effects of climate hazards on the availability of water resources.

The city of Ferkessédougou could be under strong pressure on its water resources by 2050. This is highlighted by the scenario of high population growth. Indeed, the scenario shows an increase in the UWD from 471,500 m³/year in 2020 to about 5.64 million m³/year in 2050. However, by reinforcing the capacity of the drinking water network, the drinking water deficit could be reduced to 1.2 million m³/year (Figure 10). Strengthening the production of drinking water in Ouangolodougou will reduce the city's drinking water deficit. In fact, according to the scenario of strong growth in the subscribed population, the UWD in 2050 will be 18.2 million m³/year. The scenario about the decrease of the water deficit by strengthening the drinking water supply shows that the UWD in 2050 could be reduced to 1.4 million m³/year (Figure 10).

The hydrological modeling with WEAP shows a good agreement between the simulated and observed flows during calibration and validation periods, with NSEs around 70% (Figure 4). It denotes a reliable WEAP model used for this research in the upper Bandama basin (Ritter & Muñoz-Carpena 2013). Even though, some mismatches are observed between observed and simulated flows during calibration and validation. This could be explained by the limited length of the time series of gauged data which did not take into account all characteristics of particular hydrometeorological events during both the calibration and validation periods (Montecelos-Zamora et al. 2018). The projection of precipitation and temperature has been performed with RACMO 22T and CCLM 4-8-17 RCMs selected CORDEX for Africa under the RCPs 4.5 and 8.5 scenarios. Indeed, the RCMs are developed with higher spatial resolution and are capable of providing more detailed and accurate information about local and regional climate conditions (Giorgi et al. 2009; Endris et al. 2013; Dosio & Panitz 2016). This is due to their ability to capture localized features such as land-sea interactions, complex terrain and mesoscale atmospheric processes (Endris et al. 2013; Kassahun et al. 2023). The higher spatial resolution of RCMs makes them more suitable for end-users to conduct climate change impact assessments on water resources and agriculture at regional and local scales (Kassie et al. 2014; Matiu et al. 2020; Ilori & Balogun 2022). Moreover, RCMs were used by Dembélé et al. (2023) to propose a comprehensive framework that integrates remotely sensed data, a spatially calibrated hydrological model, climate change scenarios for water accounting in the transboundary Volta River Basin, which is close to the upper Bandama watershed. In addition, the moderate scenario emission (RCP 4.5) and the highest scenario emission (RCP 8.5) were selected for this study. Due to their ability to align with historical and future GHG emissions, both scenarios were employed in several studies for West African climate studies (Coulibaly et al. 2018; Ogega et al. 2020; Adeneyi 2021; Diouf et al. 2022; Kwawuvi et al. 2023). The use of RCMs in our study area has concluded that the temperature could increase at least by 1.5°C over the period 2030–2090 compared to the 1986–2005 reference period. For precipitation, an average decrease of at least 3% is predicted over the period 2030–2090. This could be linked to the considerable reduction in soil moisture and vegetation cover (Soro et al. 2017; Yapo et al. 2019). Also, the declining trends concerning precipitation (Figure 5) are in accordance with several studies predicting a slight decrease of 10% for rainfall in northern Côte d'Ivoire (Kouman et al. 2024). It is imposed by interrelated dynamic structures at the synoptic scale with the Intertropical Convergence Zone and the East African Jet, which originate in the West African monsoon (Kouakou 2016; Yapo et al. 2020). Moreover, this result is in agreement with the trends found by other authors, since they highlighted, by using innovative statistical tests, a downward trend in precipitation around Iran and Serbia (Amiri & Mesgari 2019; Amiri & Gocic 2021, 2023). For both scenarios, reference (Figure 6) and high growth (Figure 7), the water demand for sugar cane plantation irrigation remains the highest, and accounts for about 80–90% of the global water consumption in the Bandama catchment, which is in agreement with the findings that irrigated agricultural accounts for 85–90% of global water consumption, making it the largest consumer of water globally (Qin et al. 2019). This is explained by large plantations of sugarcane that are irrigated by the main river in the Bandama basin. Hence, this result is in accordance with the global future trend for water demand in the Mediterranean basin (Bakken et al. 2016; Fader et al. 2016; Eekhout et al. 2024). Also, Mana et al. (2023) investigated the effect of climate change on reservoir water balance and irrigation water demand in the Rift Valley basin in Ethiopia. They concluded that the required irrigation water might rise by 26–36% in the future. The UWD of urban sites is lower than agricultural sites (Figure 8). Mounir et al. (2011) have obtained similar results by using the WEAP model to estimate future water demand in the Niger River basin, in West Africa. Moreover, whatever the type of water needs (domestic, livestock, agricultural), water demand is difficult to meet (Table 4). This result is confirmed by an UWD between 70,000 m³ and 34 million m³ by 2050. It is comparable to the estimated water deficit of 90 million m³ for 2020 in the Awara basin of Nigeria (Nwabuogo & Yahaya 2018). A similar study conducted over the Mbaghati sub-basin in Kenya also revealed an estimated water deficit of 60 million m³ by 2050 (Nyika et al. 2017). This situation of insufficient water resources has several reasons, including the limited resource availability and inadequacy of water infrastructure. In fact, climate change is contributing to the drying up of rivers, slowing down the recharge of groundwater and decreasing the water availability in the upper Bandama catchment (Savané et al. 2001; Yao et al. 2012). For instance, villages and big cities in the northern Côte d'Ivoire, mainly, Korhogo and Boundiali, have experienced a drop in water levels in their drinking water dams.

The first scenario implemented shows that water reuse after its treatment would reduce the UWD in the catchment area by 92%. Similar work was carried out to reduce the gap between water demand and supply in the megalopolis of Chennai, India, up to 2050 by considering several scenarios and using the WEAP model. The results indicate a 30% increase in meeting water demand through wastewater reuse (Gao et al. 2017; Nabaprabhat & Elango 2018). In the last scenario (Figure 10), the impact of high population growth on water resources was elucidated. It would require the construction of water mobilization infrastructure (e.g. boreholes, dams, etc.) to meet the water demand. In big cities such as Korhogo and Ferkessédougou, the population size is projected to increase by 50% in 2050. This might lead, like in other countries, to strong socio-economic pressures on water resources (Reynard et al. 2014; Polpanich et al. 2017). The results obtained in this study show that improving the supply of drinking water could gradually reduce the UWD by 80% in 2050 because the construction of water storage and supply facilities with an efficient management system largely contributes to meeting water demand. Identical findings have also been observed in Algeria (Aoun-Sebaiti et al. 2013), Morocco (Johannsen et al. 2016), Kenya (Okungu et al. 2017), Zambia (Tena et al. 2019) and India (Agarwal et al. 2019).

However, this study presents some limitations. Firstly, it would be interesting to explore the implications of using more than two climate models for the assessment of climate change's impact on water resources. In addition, the use of the WEAP model was justified but it would be interesting to investigate the use of a longer time series of in-situ data for modeling, which is not available currently. Future studies should try linking the WEAP model with a hydrogeological model to explore the impacts of anthropogenic pressure and climate change on groundwater.

This study uses two RCMs, namely RACMO 22T and CCLM 4-8-17, associated with the WEAP hydrological model to simulate water supply and demand over the upper Bandama catchment in the North of Côte d'Ivoire under climate change and anthropogenic pressures. According to findings, the growth of population and sugarcane plantations with climate change remain the main drivers of water availability. Also, the highest water demands are respectively from the irrigated sugarcane farms and domestic use mainly in Korhogo, Ferkessédougou and Ouangolodougou cities. This situation could have serious implications for various sectors in the region. Therefore, it is vital to implement innovative measures for solving the water supply challenges. Different scenarios conducted by the WEAP model have shown the ways forward for sustainable water management. The reuse of treated wastewater for sugarcane farm irrigation coupled with the construction of boreholes could improve water availability.

Finally, this study shows the challenges of water availability and the impacts of measures for water stress reduction on decision-makers. The major recommendation is the rehabilitation of hydrometric stations to improve water resources monitoring and water availability estimation for various users' satisfaction.

This research has been funded by the Swiss Confederation through the excellence scholarship for foreign students obtained by Franck Zokou YAO (Ref: 2017.0757/Côte d'ivoire/OP). The first author thanks Prof. Emmanuel Reynard at the University of Lausanne for his great support of the accomplishment of this work.

All authors brought contributions to the production of this manuscript. F.Z.Y. conducted the conceptualization, data collection, data analysis, and writing of the original draft. M. D. and Y.A.N. reviewed, edited, and improved the manuscript. Y.E.K. supervised the overall research work of this study. All authors read and approved the final manuscript.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.