Inne Vanderkelen

Large tropical lakes are highly sensitive to climate variability and anthropogenic warming, yet their future hydroclimatic responses remain insufficiently quantified. This study assesses future changes in Lake Victoria using the Lake Victoria Integrated Water Balance (LaVIWaB) model driven by bias-adjusted ISIMIP3b/CMIP6 climate projections under SSP1-2.6, SSP3-7.0, and SSP5-8.5 for 2015–2100. Ten global climate model simulations were integrated to project changes in precipitation, evaporation, inflows, outflows, lake levels, surface area, and storage dynamics.Results indicate an intensifying hydrological cycle, with increasing precipitation and moderate increases in lake evaporation. By 2100, mean annual precipitation is projected to increase by approximately 0.2 mm day⁻¹ under SSP1-2.6 and 0.6 mm day⁻¹ under SSP5-8.5, while evaporation increases by 0.1 and 0.4 mm day⁻¹, respectively. Assuming outflow management continues under the Agreed Curve, mean annual lake levels rise by approximately 0.5 m under SSP1-2.6 and up to 1.0 m under SSP5-8.5. Surface area expands nonlinearly, with annual maxima approaching 71,000–73,000 km², representing an increase of approximately 2% under high emissions. Projected lake levels range from 1135.3 to 1137.0 m above sea level, with peaks exceeding 1137.5 m, while lake area ranges from 67,000 to 73,000 km². Increasing interannual variability and model spread highlight substantial uncertainty and potential hydroclimatic extremes. These findings have important implications for flood risk, shoreline inundation, water resources management, and climate adaptation across the Lake Victoria Basin.

Global water models are essential tools for assessing water resource challenges in the context of climate change, land use changes, and human activities. The CoSWAT Global Water model is a global application of the Soil and Water Assessment Tool (SWAT+). It is a high-resolution tool designed to simulate water systems using a basin-oriented structure. The CoSWAT model currently lacks a realistic representation of reservoirs and lakes, limiting its ability to adequately represent basins where these water bodies play a significant hydrological role. The scarcity and limited accessibility of global reservoir operation or lake outflow data make it challenging to represent these water bodies with the current tools that SWAT+ supports, particularly at a global scale. In this study, we address this limitation by combining commonly used reservoir and lake modelling schemes from other global water models and the capabilities of SWAT+, and by implementing a topological approach to estimate irrigation water demands for reservoirs. We also introduce an improved workflow to integrate lakes and reservoirs into the river network, setting up the CoSWAT model at 0.01° resolution on selected regions worldwide, where validation of reservoir and lake storage, inflow, outflow, and the impact on streamflow performance was assessed. The results show that the model captures monthly storage dynamics with moderate performance. Overall, 34 % of evaluated reservoirs achieved a positive KGE, a performance comparable to other state-of-the-art global hydrological models, with CoSWAT surpassed by only two of five models in the comparison. Moreover, results demonstrate an improvement in performance in streamflow representation on 135 out of 192 (70 %) evaluated stations following the integration of lakes and reservoirs. Despite these improvements, persistent streamflow biases that consequently affect reservoir storage point to the need for hydrological calibration, which remains a priority for future developments. Overall, these methodological advancements represent a substantial improvement for the CoSWAT global model, enabling more robust and realistic assessments of inland water systems at the global scale.

National climate scenarios play a key role in translating global and regional climate projections into actionable information for impact assessment and policy-making. Within the framework of the CORDEX.be II project, a new generation of climate scenarios for Belgium is currently being developed, based on convection-permitting regional climate model simulations at kilometer scale. The ensemble combines three models - ALARO, MAR, and COSMO-CLM - and covers a reference period as well as two future 20-year time slices corresponding to global warming levels of +2°C and +3°C.Particular attention is given to the representation of climate extremes, such as heatwaves and intense precipitation, through the careful selection of boundary conditions from CMIP6 global climate models that demonstrate strong performance in simulating these phenomena. The credibility of the regional simulations is assessed through their ability to reproduce key features of the present-day climate. Moreover, the climate-change signals are considered against the range of the coarser-scale EURO-CORDEX projections over Belgium. A central component of this work is the close collaboration with stakeholders throughout the development process. End users from multiple sectors are actively involved in the scenario choice and the co-design of climate indicators, ensuring their relevance for practical applications, and shaping the format and usability of the final products. The climate indicators and bias-corrected time series will provide a robust basis for assessing climate impacts across sectors, including agriculture, health, and water management, and for supporting adaptation planning to future climate extremes in Belgium. The climate indicators will be provided to the stakeholders as well as through an online portal.

Soil carbon sequestration is an important strategy for climate change mitigation with several co-benefits, including increased water holding capacity and infiltration. However, a global-scale quantification of hydrological co-benefits for water availability to plants is still lacking. In this study, we investigate the effect of soil carbon sequestration on hydrology and water resources by conducting experiments with the Community Terrestrial Systems Model (CTSM). Using global experiments with spatially explicit soil organic carbon (SOC), we apply various soil carbon sequestration scenarios, including one aligned with the “4 per 1000” initiative, to investigate the effect on soil moisture and soil water balance variables with a focus on cropland regions. Our results show that soil organic carbon redistributes water within the soil profile, retaining moisture in the rooting zone and limiting percolation into deeper layers, which is particularly pronounced in relatively arid regions with sandy soils. Global average soil water content increases by 4 mm in the first 30 cm under a scenario with a uniform SOC increase of 5.5 gC kg−1 soil. Carbon sequestration also redistributes the mean annual soil water balance, with global mean reductions in surface runoff (−1 mm yr−1), subsurface runoff (−0.6 mm yr−1), and an increase in evapotranspiration (+2 mm yr−1), contributing to improved vegetation productivity. Water stress is overall reduced across most regions. Although the hydrological impacts of soil carbon sequestration are generally small in magnitude, they are consistent and systematic. The relative changes following realistic and policy-relevant SOC enhancement scenarios, such as those under the 4 per 1000 initiative, are limited due to the modest carbon additions involved. Nevertheless, these changes offer measurable eco-hydrological co-benefits that may support both climate mitigation and ecosystem resilience, particularly in water-limited environments.

Global water models are essential tools for assessing water resource challenges in the context of climate change, land use changes, and human activities. The CoSWAT Global Water model is a global application of the Soil and Water Assessment Tool (SWAT +). It is a high-resolution tool designed to simulate water systems using a basin-oriented structure. The CoSWAT model currently lacks a realistic representation of reservoirs and lakes, limiting its ability to adequately represent basins where these water bodies play a significant hydrological role. The scarcity and limited accessibility of global reservoir operation or lake outflow data make it challenging to represent these water bodies with the current tools that SWAT+ supports, particularly at a global scale. In this study, we address this limitation by combining commonly used reservoir and lake modelling schemes from other global water models and the capabilities of SWAT+. Moreover, to model irrigation reservoirs, we implemented an approach that combines global datasets with a topological method to estimate irrigation demand for each reservoir. With these new implementations, the CoSWAT model was restructured for selected regions worldwide, where validation of reservoir or lake storage, inflow, and outflow was performed, and the impact of these implementations on streamflow performance was assessed. The results show that the model captures storage dynamics with reasonable performance, comparable to other state-of-the-art global models, and demonstrate a general improvement (70 % of evaluated stations) in streamflow representation following the integration of these water bodies. The new methodological advancements represent a substantial improvement for the CoSWAT global model, enabling more robust and realistic assessments of inland water systems at the global scale.

National climate scenarios provide a consistent translation of global and regional climate projections into information relevant for impact modelling and decision-making. Here, we present the development of a new set of national climate scenarios for Belgium based on regional climate model (RCM) simulations at convection-permitting scale. The ensemble comprises three RCMs (ALARO, MAR, and COSMO-CLM) that simulate the present-day climate and two future 20-year periods corresponding to global warming levels of +2 °C and +3 °C. The choice of global climate models and the downscaling approach specifically target climate extremes, including heatwaves and heavy precipitation events. Accordingly, boundary conditions are provided by CMIP6 models selected for their demonstrated skill in simulating these extremes. The performance of the three RCMs is evaluated for the present-day period based on their ability to simulate key climate variables. From the raw model output, we co-develop a set of climate indicators with key stakeholders, who also contribute to defining the format of the final products. The resulting national climate scenarios provide a robust basis for assessing climate impacts across multiple sectors, including agriculture, health, and water management, and for supporting adaptation planning to future climate extremes in Belgium.

Climate extremes such as heatwaves, river floods, droughts, crop failures, including aspects of wildfires and tropical cyclones, are increasingly attributable to anthropogenic climate change. Yet how this translates into unprecedented levels of extreme event exposure in one’s lifetime remains unclear. Here we show that, neglecting adaptation, many of today’s youth will experience unprecedented exposure to extremes during their lifetimes. For the events above, the share of people facing unprecedented lifetime exposure is projected to at least double from 1960 to 2020 birth cohorts under current mitigation policies aligned with a global warming pathway reaching 2.7 °C above pre-industrial temperatures by 2100. In a 1.5 °C pathway, ∼50% of people born in 2020 will experience unprecedented lifetime exposure to heatwaves. If global warming reaches 3.5 °C by 2100, this rises 30 to ∼90% of this birth cohort. For the same cohort and warming pathway, ∼30% will live with unprecedented exposure to crop failures and ∼10% to river floods. Further, under current policies, two indicators of vulnerability show that the most vulnerable experience significantly more unprecedented exposure to heatwaves than the least vulnerable. Our results call for sustained greenhouse gas emissions reductions to lower the burden of climate change on young generations

Water scarcity is a growing concern in many regions worldwide, as demand for clean water increases and supply becomes increasingly uncertain under climate change. Developing socio-economic conditions and growing population increase water demands, while climate change leads to changes in freshwater availability. Water scarcity assessments typically rely on static biophysical measures within discrete time windows, using fixed population and climate change projections, while overlooking demographic dynamics, lifetime evolution, and cumulative deficits across generations.Here, we calculate monthly water deficits based on sectoral, population-driven demand and water availability worldwide by combining demographic data with an ensemble of global climate and hydrological models from the InterSectoral Impact Model Intercomparison Project (ISIMIP2b). By linking these deficits with gridded population projections and life expectancy, we estimate the proportion of lifetime water demand that remains unmet per individual. Thereby we capture how shifting hydro-climatic and demographic conditions shape water scarcity across generations.Our analysis shows that younger generations will bear a significantly greater share of lifetime water scarcity. Across all regions, younger generations will face higher lifetime water deficits compared to older generations. Without adaptation, a child born in 2020 is projected to experience 45% of their lifetime water demand unmet. Approximately 706 million children are expected to encounter deficits exceeding half of their lifetime needs—1.5 times more than individuals aged 50–59. This intergenerational disparity is primarily driven by population growth and rising life expectancy in areas with limited adaptive capacity. These findings underscore the urgent need for accelerated adaptation strategies to safeguard water security for future generations.

While increasing heat is a direct impact of climate change on health, the contribution of climate change to temperature-related neonatal deaths in Low- and Middle-income Countries (LMICs), including Kenya, is unknown. We aim to estimate the temperature-related burden of neonatal deaths (children less than 28 days of age) in Kenya between 2022 and 2024 that is attributable to climate change. We use daily neonatal mortality counts for the period ranging from January 1, 2022, to December 31, 2024, from the Kenya Health Information System (KHIS) tracker database. For heat exposure, we use daily reanalysis mean temperature data from the third simulation round of the Inter-Sectoral Impact Model Intercomparison Project (ISIMIP3a), including both the obsclim (factual) and counterclim (counterfactual) scenarios at a 0.5° x 0.5° spatial scale. We perform an extended two-stage design for small geographical areas to estimate temperature-neonatal mortality associations and temperature-related burden of neonatal deaths in Kenya between 2022 and 2024 that is attributable to climate change. The KHIS data includes all hospital-based neonatal deaths (~ 29,000) recorded across all the 47 counties in Kenya between 2022 and 2024. We find that across all counties in Kenya, exposure to extreme heat (99th percentile temperature) relative to the minimum mortality temperature for a period of seven days increases the relative risk of neonatal mortality by 1.517 (95% C.I 1.129 - 2.037), although with important geographical differences. Moreover, we found a larger effect in regions with a smaller ratio of health workers per 100,000 population than in those with a higher ratio, and in areas with poor access to insurance compared to those with higher access. Overall, climate change was responsible for 3.6% (95% C.I. 0.9% – 6.5%) of heat-related neonatal deaths in Kenya between 2022-2024. This study underscores the negative impacts of extreme temperatures on neonatal health. Future increases in global mean temperature will likely amplify heat-related health risks, highlighting the urgent need for climate-informed neonatal health mitigation and adaptation measures to protect newborns' health in the face of a changing climate.

Recent decades have seen an increase in the frequency and intensity of extreme heat events. Such events have a profound impact on society, as they increase human exposure to heat stress and can negatively affect energy consumption and agriculture. These effects are further amplified in urban environments due to climate change. Belgium, among the most urbanised countries worldwide, has experienced an increasing number of heatwaves in recent years, with several years recording multiple events. However, the future extent and characteristics of heatwaves remain unclear, making it difficult to assess whether current suggested adaptation strategies and heat action plans for today's climate will remain effective under future warming. With this research, we aim to develop a storyline-based selection method to quantify single future heat waves based on the characteristics of past events and thereby enable the identification of relevant case studies for assessing future risks.With the new Belgian climate projections from the CORDEX.be II project, different global warming levels are being explored (1.5°C, 2°C, 3°C and 4°C), to gain insights into different future extreme events. In this study, we analyse high-resolution data from the regional climate model COSMO-CLM, in which urban areas are explicitly modelled, forced by the global climate model EC-Earth3-Veg (SSP5-8.5), at two horizontal resolutions: 12.5 km (all warming levels) and 2.8 km (2°C and 3°C). By identifying historical extreme events and mapping events with similar characteristics in higher warming levels using different heat metrics, we quantify changes in event duration, temperature characteristics, and intensity relative to the recent past to obtain interesting future cases.