1 of 15Global Change Biology, 2026; 32:e70720 https://doi.org/10.1111/gcb.70720 Global Change Biology RESEARCH ARTICLE Beyond Climate Change: The Role of Integrated Soil Fertility Management for Sustaining Future Maize Yield in Sub-Saharan Africa Antoine Couëdel1,2   | Gatien N. Falconnier1,3,4,5   | Myriam Adam6,7   | Rémi Cardinael1,3,5   | Johan Six8   | Moritz Laub8   | Alex C. Ruane9   | Kenneth Boote10   | Eric Justes11   | Ward N. Smith12   | Anthony M. Whitbread13   | François Affholder1,14,15   | Folorunso M. Akinseye16,17   | Juraj Balkovic18   | Bruno Basso19   | Arti Bhatia20   | Bidisha Chakrabarti20  | Regis Chikowo5,21   | Mathias Christina1,2,22   | Babacar Faye23   | Fabien Ferchaud24,25   | Christian Folberth18   | Thomas Gaiser26   | Marcelo Galdos27   | Sebastian Gayler28   | Aram Gorooei29   | Brian Grant12   | Hervé Guibert1,2   | Gerrit Hoogenboom10   | Bahareh Kamali29   | Fidel Maureira19   | Fasil Mequanint28,30   | Claas Nendel31,32,33   | Cheryl H. Porter10   | Dominique Ripoche34   | Leonard Rusinamhodzi35   | Shikha Sharma20  | Upendra Singh36   | Amit Srivastava26   | Bernard Vanlauwe37   | Antoine Versini38   | Murilo Vianna29   | Heidi Webber31,39   | Tobias Weber28,40  | Congmu Zhang10   | Marc Corbeels1,37,41 Correspondence: Antoine Couëdel (antoine.couedel@cirad.fr) Received: 13 May 2025  |  Revised: 15 December 2025  |  Accepted: 15 December 2025 Keywords: integrated soil fertility management | mineral fertilizers | model ensemble | organic amendment | soil organic carbon | soil-crop models ABSTRACT Climate change is projected to exacerbate food insecurity in sub-Saharan Africa (SSA) by reducing crop yields and soil fertility. Many climate change impact studies in SSA have overlooked long-term effects of soil fertility on crop yield. We evaluated maize yields under different scenarios of soil fertility (using soil organic carbon as a proxy) and climate change (considering changes in temperature, rainfall, and CO2) at four sites in SSA. Using an ensemble of 15 calibrated soil-crop models, we found a strong consensus that, without fertilization, soil fertility declines over time, impacting maize yields more strongly than changes in temperature, rainfall, or CO2. The model ensemble indicated that when accounting for soil fertility changes, the yield benefits of combined application of organic and mineral inputs increase over time, even under climate change. These findings highlight the importance of considering long-term change in soil fertility when assessing impacts of climate change and integrated nutrient management on crop production in SSA. 1   |   Introduction Sub-Saharan Africa (SSA) faces significant food security issues despite its considerable agricultural potential (Van Ittersum et al. 2016; Giller 2020). Food insecurity in the region has been exacerbated by recent geopolitical conflicts (Behnassi and El Haiba 2022), but is largely due to low soil fertility (Sanchez 2002; Vanlauwe et  al.  2015), inadequate nutrient inputs (Jayne and Sanchez  2021), insufficient or erratic rainfall (Kamali et al. 2022; Bouhenache et al. 2025), and pathogen and pest pres- sures (Savary et al. 2019), all leading to low crop yields. Maize productivity plays a crucial role in ensuring food security in SSA, as it is the most widely grown cereal and the main source of di- etary energy, providing at least 30% of the total caloric intake of the region's population (OECD 2016; Ekpa et al. 2019). Since the beginning of the 21st century, the area dedicated to maize cul- tivation in SSA has nearly doubled (FAOSTAT 2023), currently occupying about 40% of the total area cropped with cereals. Yet, average maize yields across the continent have largely stagnated at about 1.9 t ha−1 (FAOSTAT 2023), although some countries ex- hibit modest increases, such as Ethiopia, Angola, Mozambique, Ivory Coast, and parts of Nigeria, while others experience yield © 2026 John Wiley & Sons Ltd. https://doi.org/10.1111/gcb.70720 https://orcid.org/0000-0002-5786-0939 https://orcid.org/0000-0003-3291-650X https://orcid.org/0000-0002-8873-6762 https://orcid.org/0000-0002-9924-3269 https://orcid.org/0000-0001-9336-4185 https://orcid.org/0000-0003-2415-8067 https://orcid.org/0000-0002-5582-9217 https://orcid.org/0000-0002-1358-5496 https://orcid.org/0000-0001-7390-7058 https://orcid.org/0000-0001-7462-5247 https://orcid.org/0000-0003-4840-7670 https://orcid.org/0000-0002-3919-4805 https://orcid.org/0000-0002-8677-6306 https://orcid.org/0000-0003-2955-4931 https://orcid.org/0000-0003-2090-4616 https://orcid.org/0000-0001-7884-4658 https://orcid.org/0000-0003-3047-359X https://orcid.org/0000-0003-3618-756X https://orcid.org/0000-0002-8597-1679 https://orcid.org/0000-0002-2078-3570 https://orcid.org/0000-0002-6738-5238 https://orcid.org/0000-0002-5820-2364 https://orcid.org/0000-0002-6080-0726 https://orcid.org/0000-0002-8577-6486 https://orcid.org/0000-0002-0644-2032 https://orcid.org/0000-0001-5928-0391 https://orcid.org/0000-0001-5406-9204 https://orcid.org/0000-0002-1555-0537 https://orcid.org/0000-0002-8070-0175 https://orcid.org/0000-0003-1714-4759 https://orcid.org/0000-0001-6049-1794 https://orcid.org/0000-0001-7608-9097 https://orcid.org/0000-0001-7269-6543 https://orcid.org/0000-0003-0785-9953 https://orcid.org/0000-0002-5576-2040 https://orcid.org/0000-0001-8653-0333 https://orcid.org/0000-0001-8219-4854 https://orcid.org/0000-0001-6016-6027 https://orcid.org/0000-0002-5119-0567 https://orcid.org/0000-0003-1139-4589 https://orcid.org/0000-0001-8301-5424 https://orcid.org/0000-0002-1413-9834 https://orcid.org/0000-0002-8084-9287 mailto:antoine.couedel@cirad.fr http://crossmark.crossref.org/dialog/?doi=10.1111%2Fgcb.70720&domain=pdf&date_stamp=2026-02-03 2 of 15 Global Change Biology, 2026 declines (Ray et  al.  2013). The projected 2.3-fold increase in maize demand by 2050 (Robinson et al. 2015) is expected to put further pressure on food security and land resources. The urgent need for increased food production contrasts sharply with current trajectories of environmental factors affecting crop productivity in SSA. On one hand, rising temperatures across SSA and decreasing rainfall in West and Southern Africa (Trancoso et al. 2024), together with the increasing frequency of extreme weather events (Westra et al. 2014; Lesk et al. 2020; Ranasinghe et  al.  2021; Reed et  al.  2022), are expected to negatively impact maize yields (Sultan and Gaetani  2016; Jägermeyr et  al.  2021), thereby worsening food insecurity (Tadross et al. 2009; Falconnier et al. 2020; Carr et al. 2022; Ren et al. 2024). On the other hand, soil fertility continues to decline in many regions of SSA due to the low use of mineral fertilizers and organic inputs, along with the shortened fallow periods and widespread soil erosion (Vanlauwe and Giller 2006; Vanlauwe et al. 2015; Kihara et al. 2020; Falconnier et al. 2023). Evidence from long-term field experiments shows that low fertilizer in- puts lead to a progressive decline in both soil fertility and crop yields over time (Adams et al. 2020; Cardinael et al. 2022; Laub, Corbeels, Mathu Ndungu, et al. 2023; Laub, Corbeels, Couëdel, et al. 2023). Therefore, if soil fertility continues to decline, maize productivity in SSA is likely to decrease even further. Integrated soil fertility management (ISFM), which includes the combined use of mineral and organic inputs, offers the prospect of increasing crop yield and soil fertility (Vanlauwe et  al.  2010), thereby improving resilience to climate change (Gram et  al.  2020). However, due to the difficulty of conduct- ing climate manipulation experiments, the effectiveness of the combined use of mineral and organic inputs in the context of climate change remains insufficiently evaluated, leaving uncer- tainty about whether its benefits are maintained in changing cli- matic conditions. In this context, an ensemble of process-based soil-crop models can help assess the impacts of climate change (Falconnier et  al.  2020), as they can account for the complex, nonlinear interactions between soil organic carbon (SOC) dy- namics and crop growth (Couëdel et  al.  2024; Couëdel, Laub, et al. 2026). By reinitializing soil conditions at the start of each simulated year, most crop modeling studies assessing the impact of future climate change on crop productivity have focused on evaluating the direct effects of changing temperatures, rainfall, and atmo- spheric CO2 concentrations on crop growth and yield (Challinor et al. 2014; Falconnier et al. 2020; Jägermeyr et al. 2021; Rezaei et al. 2023). These studies ignored possible soil fertility decline over time, thus overlooking the combined effects of climate change and soil degradation on crop yields. Simulation studies that incorporate soil fertility dynamics are particularly import- ant in tropical low-input cropping systems, where soil organic matter mineralization is the primary source of nutrients for crop growth (Giller et al. 2011). The aim of our study was to assess how long-term changes in soil fertility and climate drivers affect maize yield in SSA, how these factors interact, and to what extent ISFM can alleviate the negative effects. SOC was used as a proxy for soil fertility, given its central role in determining soil chemical, physical, and biological properties (Lehmann et al. 2020). We simulated key components of climate change affecting crop growth, namely atmospheric CO2, temperature and precipitation, using an en- semble of 15 process-based soil-crop models. The models were calibrated using data from four long-term field experiments, covering diverse representative climates and soils in SSA (see Figure 1). Such long-term experiments remain scarce in the re- gion, making them particularly valuable for model calibration and scenario analysis. The study evaluated scenarios of soil fer- tility change, climate change impacts, and their combined ef- fects under two contrasting crop management options: (i) no use of mineral or organic inputs, representing current smallholder farmers' practices, and (ii) combined use of mineral (80 kg N ha−1 season−1) and organic (2 t C ha−1 yr−1) inputs, a core principle of ISFM, representing a plausible sustainable intensification scenario with nutrient levels required to achieve food security across SSA (Falconnier et  al.  2023). We hypothesized that (i) soil fertility decline has a stronger impact on maize yields than climate change alone, though with varying magnitudes across sites, and (ii) ISFM mitigates the negative effects of soil fertil- ity decline, although its effectiveness is reduced under climate change. 2   |   Materials and Methods 2.1   |   Multi-Model Ensemble and Model Calibration Methods An ensemble of 15 calibrated process-based soil-crop models was used in this study (see Table 1 and Couëdel et al. (2024) for detailed model descriptions). In brief, the models differ in how they simulate soil processes, crop development, crop growth, and grain yield formation. SOC dynamics are generally sim- ulated using first-order decay kinetics, but models vary in the number and type of SOM pools, with some models differenti- ating active, slow, and passive pools (e.g., DSSAT), while others use a single-pool approach (e.g., CELSIUS). Ten of the 15 mod- els include an explicit soil microbial biomass pool with first- order decay. The representation of fresh organic matter inputs (crop residues, organic amendments) ranges from a single pool to multiple specialized pools, as for example implemented in SIMPLACE. All models simulate soil nitrogen (N) cycling cou- pled with the SOM pools. Models also differ in their treatment of other soil processes such as soil water dynamics and nitrate leaching, while crop growth processes vary in their simulation of leaf area development, light interception, biomass accumula- tion, and grain formation. A single experienced modeling team conducted the simulations with each model. Different versions of a given model (i.e., DSSAT, Expert-N, and SIMPLACE) were conducted by the team working with that specific version of the model. Model calibration and evaluation under current conditions were deemed essential for ensuring plausible maize yield sim- ulations under the future climate change scenarios, while ac- counting for long-term SOC dynamics and their interactions with maize growth. The median of the model ensemble (15 soil-crop models) was used to evaluate how well the ensem- ble reproduced maize yields and SOC dynamics over time across the four long-term experiments. The modeling teams 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 3 of 15Global Change Biology, 2026 were free to use their own methods for calibration of the crop and soil modules of their respective model versions. Values for model input parameters and variables (observed or estimated), including soil characteristics, initial soil conditions, weather, soil and crop management, were provided to the modeling teams for conducting simulations of maize growth during the years of the experiments at the four sites. Observed data on maize phenology, grain yield, aboveground biomass, and SOC were used for model calibration. Couëdel et al. (2024) provide more detailed information about model structural differences FIGURE 1    |    Harvested areas of maize in sub-Saharan Africa showing the four study sites representativeness of the six CIMMYT maize mega en- vironments (a) and their baseline (1980–2010) mean annual temperature, mean annual precipitation, and clay content (b). Soil organic carbon (SOC) is the initial SOC concentration (%) (0–20 cm). Map lines delineate study areas and do not necessarily depict accepted national boundaries. TABLE 1    |    The 15 soil-crop process-based models used in the model intercomparison study. Model Model references APSIM v7.9 Holzworth et al. (2014) CELSIUS Ricome et al. (2017) DayCent Del Grosso et al. (2001) DNDC v.CAN Smith et al. (2020a, 2020b) DSSAT v4.8.0.19-CERES-Maize + Century Ritchie et al. (1998), Gijsman et al. (2002), Hoogenboom et al. (2021) DSSAT v4.8.0.19-CERES-Maize + Ceres-SOM Godwin and Jones (1991), Hoogenboom et al. (2021) EPIC Izaurralde et al. (2006) Expert-N v5.1-Gecros Biernath et al. (2011) Expert-N v5.1-Spass Biernath et al. (2011) Expert-N v5.1-Ceres Biernath et al. (2011) MONICA v3.3.1 Aiteew et al. (2024), Nendel et al. (2011) SALUS Basso et al. (2010) SIMPLACE-Lintul + Option 1 Enders et al. (2023), Gaiser et al. (2013) SIMPLACE-Lintul + Option 2 Faye et al. (2018) STICS v10 Brisson et al. (2003), Beaudoin et al. (2023) 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 4 of 15 Global Change Biology, 2026 and calibration procedures. In summary, model calibration consisted of adjusting maize phenology and key crop and soil parameters to reproduce observed yield, aboveground biomass, and SOC. The initialization and turnover of soil or- ganic matter pools were calibrated using observed SOC lev- els, spin-up model runs, or empirical relationships with soil texture, depending on the model, with each team applying its standard procedure to improve the agreement between simu- lated and observed SOC dynamics. Model calibration inputs and outputs are available in the CIRAD repository (Couëdel, Falconnier, et al. 2026). 2.2   |   Experiments and Data Data from long-term maize field experiments at four sites in SSA were used to calibrate 15 process-based soil-crop models (see Couëdel et al. (2024) for experimental details and model descrip- tions; in the present study InfoCrop was used only for the calibra- tion exercise and not for the sensitivity analysis to climate change factors). These sites cover contrasting climates, soil textures, and soil fertility levels, and are representative of three of the six maize mega-environments in SSA that are broad agroecologi- cal zones that share similar climatic, soil, and biotic conditions, which together determine how maize performs and adapts in a given region (Hartkamp et  al.  2000; Figure  1). The four experi- ments (Côte d'Ivoire-Gagnoa, Kenya-Embu, Kenya-Machanga, and Zimbabwe-Murewa) spanned between 9 and 18 years and were managed by researchers either on-station or on-farm. Such long-duration field experiments are scarce in SSA, making them particularly valuable for model calibration and scenario analy- sis. All sites, except Zimbabwe-Murewa, had two maize growing seasons per year. The experiments were rainfed and included the following two treatments: (i) a control without exogenous C and N inputs, and (ii) a combined application of mineral N (between 100 and 160 kg N ha−1 per growing season depending on the site) and organic inputs (between 3 and 4 t C ha−1 per year depending on the site). This latter treatment represents the input practice in ISFM. Organic input was compost made from crop residues at Côte d'Ivoire-Gagnoa and farmyard manure at the other sites. The input of crop residues to the soil was limited to roots, as abo- veground biomass residues were removed from the experimental plots. Biotic stresses (weeds, pests, and diseases) were controlled to minimize their impact on maize yield. General soil characteristics and initial soil conditions were determined at the beginning of the experiments. Data on crop management (tillage, maize cultivar, planting date and density, and fertilizer application) was available for all sites and years. Flowering and maturity dates of the maize cul- tivars were recorded for Côte d'Ivoire-Gagnoa and Kenya- Embu, while only cultivar maturity periods were available for Zimbabwe-Murewa and Kenya-Machanga. Grain yield data was available for all years and sites, and total aboveground biomass data was available for all sites except Zimbabwe- Murewa. SOC data were collected from measurements at multiple time intervals throughout the experiments on topsoil samples (0–15 cm at Kenya-Machanga and Kenya-Embu, and 0–20 cm at Zimbabwe-Murewa and Côte d'Ivoire-Gagnoa; see Figure 2). Weather data during experimental years were obtained pri- marily from onsite records (rainfall at all sites, minimum and maximum temperatures at all sites except Zimbabwe-Murewa). For daily incident solar radiation and relative humidity, and for temperature data at Zimbabwe-Murewa, records from NASA's Prediction Of Worldwide Energy Resources (POWER) database (Duarte and Sentelhas 2019) were used. More information about the experiments is provided in ear- lier publications, for Côte d'Ivoire-Gagnoa (Pichot et  al.  1977; Guibert 1999; Cardinael et al. 2022), for the two sites in Kenya (Chivenge et al. 2009, 2011; Gentile et al. 2009; Laub, Corbeels, Mathu Ndungu, et al. 2023; Laub, Corbeels, Couëdel, et al. 2023), and for the site in Zimbabwe (Zingore et al. 2007; Rusinamhodzi et al. 2013). 2.3   |   Model Simulation Scenarios The response of the model ensemble to the climate change factors of changing temperature, rainfall and atmospheric CO2 was as- sessed for two crop management scenarios: (i) without external C and N inputs and (ii) with combined mineral and organic in- puts, representing ISFM. The model simulations for the 30-year baseline climate (1980–2010) were done with uniform maize management practices across all years (Table S1). The 30-year baseline climate data (1980–2010) for rainfall and temperature were derived from onsite records, supplemented with bias- corrected NASA POWER data when necessary. NASA POWER also provided the data for daily incident solar radiation and rela- tive humidity for the 30-year baseline climate. The ISFM system combined application of mineral N fertilizer (80 kg N ha−1 per growing season) and organic input (2 t C ha−1 year−1), as manure at all sites except Côte d'Ivoire-Gagnoa where the addition of compost made from crop residues was simulated. This scenario represents a plausible sustainable intensification pathway with nutrient inputs sufficient to meet food security needs across sub- Saharan Africa (Falconnier et al. 2023). Following the AgMIP and C3MP protocols (Ruane et  al. 2014; Rosenzweig et al. 2013), we applied individual changes of tempera- ture, rainfall and atmospheric CO2 to assess model responses. This approach isolates crop model sensitivities to each climate driver, allowing for direct comparisons across models and sites while avoiding confounding effects from co-varying changes in climate scenarios. The response to increasing atmospheric CO2 was an- alyzed using concentrations of 360, 450, 540 and 720 ppm. The response to warmer temperatures was analyzed by uniformly in- creasing daily minimum and maximum temperatures by respec- tively 2°C, 4°C and 6°C, relative to the baseline climate. Finally, the response to change in rainfall was analyzed by decreasing daily rainfall by 25% and 50%, or by increasing it by 25% and 50%. The changes in CO2, temperature and rainfall were assessed in- dependently, and factorial combinations were not considered to isolate the individual effects of change in climate variables. This approach avoids additional uncertainty, as current models are not yet fully equipped to reliably simulate complex interactions among multiple concurrent climate stresses (Webber et al. 2022). In the scenario without C and N inputs, both reset and contin- uous model simulations were performed to isolate the effects 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 5 of 15Global Change Biology, 2026 FIGURE 2    |    Observed (in red) and simulated (in black) maize yield and soil organic carbon (SOC) change for no input (a) and mineral and organic fertilizers (b) treatments across four long-term experiments in sub-Saharan Africa. ‘Year’ represents years from the start of the experiments. Black data points are medians of 15 soil-crop models. Dark gray areas indicate the 25th to 75th percentile range of the values simulated. Red lines are linear (for observed yield) and/or exponential regression functions (for observed SOC). Only significant regressions at p < 0.05 are displayed. 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 6 of 15 Global Change Biology, 2026 of climate change factors and those of soil fertility dynamics. In the reset model simulations, SOC, soil total N, soil mineral N, and soil water were reinitialized each year with the same starting values, thereby eliminating the cumulative effects of changes in soil fertility over time. In contrast, in continuous simulations, the initial values of these variables for each year were set according to the simulated values at the end of the previous year. This allowed each model of the ensemble to simulate the full 30-year period continuously, accounting for soil-crop feedback effects on crop yield. In the ISFM scenario with mineral and organic inputs, only continuous simulations were conducted. In the main text and figures, we present results only for the scenarios of +2°C, −25% rainfall, +25% rainfall, and 540 ppm CO2, denoted as ‘warmer temperature’, ‘decreasing rain- fall’, ‘increasing rainfall’, and ‘higher CO2’, respectively. The +2°C and 540 ppm CO2 scenarios correspond to conditions projected under a high-emission pathway (SSP3-7.0) by 2050 (Meinshausen et  al.  2020; Thrasher et  al.  2022). The ±25% rainfall scenarios were chosen to represent the widespread uncertainty in rainfall projections across global climate mod- els, capturing plausible dry and wet future conditions (see Table S2). The simulation results from the other, more extreme scenarios (±50% rainfall, up to +6°C, up to 720 ppm CO2) are summarized briefly in the “Results” section and provided in detail in the “Supporting Information”. Model simulation inputs and outputs are available in the CIRAD repository (Couëdel, Falconnier, et al. 2026). 2.4   |   Data Analysis Comparing the reset and continuous model simulations in the scenario without external C and N inputs enabled to quantify the average relative change in maize yield (%) due to climate change alone (CC, Figure S1A), soil fertility change alone (SF, Figure S1B) and their combined (CCSF, Figure S1C) and inter- active effects (CC × SF), as follows: where BLreset and CCreset represent the average simulated maize yields from a given model within the 15-model ensemble across the 30-year simulation period under the reset simulation mode. BLreset corresponds to the baseline climate scenario, while CCreset refers to scenarios with changes in climate factors. BLcont and CCcont represent the average simulated maize yields from a given model within the 15-model ensemble across the 30-year simulation period under the continuous simulation mode. BLcont corresponds to the baseline climate scenario, while CCcont re- fers to scenarios with changes in climate factors. The interactive effect (CC × SF) represents the non-additive component of the combined effect of climate change and soil fertility change on maize yield, that is, the portion of the yield response that cannot be explained by the sum of their individual effects. A positive CC × SF value indicates a synergistic interaction, while a nega- tive value indicates that one factor partially offsets the effect of the other. The differences between maize yield simulations in the scenario without C and N inputs and that with mineral plus organic in- puts under continuous simulation mode allowed to compute the yield response (%) to organic and mineral input, considering soil-crop feedbacks, as follows: where Y0 and YMOF are the simulated maize yields from a given model within the 15-model ensemble across the 30-year simu- lation period without and with the mineral and organic inputs, respectively. Results are then shown using the median of the 15 models (Figure 4). We assessed the agreement between models by counting the number of models performing in the same direction for a given analysis. Following this approach, agreement between mod- els was categorized into low (< 50% models agreed), medium (50% < models agreed < 75%), and high (> 75% models agreed) confidence levels. The relationships between (i) observed or simulated maize yield versus time since the start of the experiment (Figure  2, Figure  S7) and (ii) simulated yield change with organic plus mineral inputs versus time (Figure 4) were fitted using the lm function in R (R core team 2023). For all statistical tests, the sig- nificance level, α, was set at 0.05. For the scenario without C and N inputs, observed SOC changes over time (Figure 2a) were fitted with an exponential decay function using the EXD.3 function of the ‘drc’ package in R (R core team 2023). For the scenario with C and N inputs, an asymptotic function using the AR.3() function was used (Figure 2b). 3   |   Results 3.1   |   The Calibrated Soil-Crop Model Ensemble Reproduced Trends in Maize Yield and SOC For the treatment without C and N inputs, the ensemble median accurately reproduced the observed decrease in SOC over time at the Côte d'Ivoire-Gagnoa and Kenya-Embu sites, but under- estimated SOC loss at the Kenya-Machanga and Zimbabwe- Murewa sites (Figure  2a). Although the models did not fully capture interannual maize yield variability and tended to over- estimate yields in the first years at all sites except Zimbabwe, they reproduced the observed temporal trends reasonably well, showing declining yields at Côte d'Ivoire-Gagnoa and Kenya- Machanga, and stable yields at Kenya-Embu and Zimbabwe- Murewa (Figure  2a). For the treatment with combined use of (1)CC (%) = CCreset − BLreset BLreset × 100 (2)SF (%) = BLcont − BLreset BLreset × 100 (3)CCSF (%) = CCcont − BLreset BLreset × 100 (4)CC × SF(%) = CCSF − (CC + SF) (5)ΔYMOF (%) = YMOF − Y0 Y0 × 100 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 7 of 15Global Change Biology, 2026 FIGURE 3    |    Boxplots of simulated maize yield change (%) by climate change only (CC only, see Equation (1) in Section 2.4), soil fertility declines only (SF only; i.e., difference between yearly reinitialization of soil conditions vs. continuous model simulation, see Equation (2) in Section 2.4) and the combination between climate change and soil fertility decline (CC and SF combined, see Equation (3) in Section 2.4) compared to no climate change and no soil fertility decline scenario (no CC and no SF decline). Simulations represent no fertilizer systems, averaged across 30 years for each of the 15 individual models. Results are averaged across sites (a) and per site (b). FIGURE 4    |    Simulated maize yield change (%) with mineral and organic inputs (80 kg N ha−1 season−1 and 2 t C ha−1 yr−1 organic amendment) compared to no fertilizer systems under climate change (CC) or not (no CC). Simulations are averaged across sites (a) and trends over years for the four sites (b). Data are from the median of 15 soil-crop models' simulations. Only significant linear regressions at p < 0.05 are displayed. All simula- tions included long-term soil-crop feedback (continuous model simulation without yearly reinitialization of soil conditions). 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 8 of 15 Global Change Biology, 2026 mineral and organic inputs, the ensemble median reproduced reasonably well the observed stability of SOC levels over time at Côte d'Ivoire-Gagnoa, but underestimated the SOC decrease at Kenya-Embu and the SOC increases observed at Zimbabwe- Murewa and Kenya-Machanga. Despite not fully capturing the interannual maize yield variability, the ensemble median repro- duced well the observed stability of yields over time at all four sites (Figure 2b). 3.2   |   Soil Fertility Decline Has a Larger Impact on Maize Yield Than Climate Change Factors In the maize systems without C and N inputs, the scenario simu- lations with the model ensemble over the 30-year period suggest that the decline in soil fertility had a larger impact on yield than climate change factors (Figure 3a, high confidence). The average simulated decrease in maize yield due to soil fertility decline was 24% across the four sites, far outweighing the effects of individ- ual climate factors: +2% for warmer temperature (+2°C), −4% for reduced rainfall (−25%), +8% for increased rainfall (+25%), and +6% for higher CO2 levels (540 ppm; Figure 3a). This greater impact of declining soil fertility over climate change factors was consistently simulated across a wider range of changes in tem- perature (from +2°C to +6°C), rainfall (from −50% to +50%), and CO2 levels (from 450 to 720 ppm; Figure S2). The differences in maize yield impacts between the soil fertility decline scenario and the climate change factors scenarios were most pronounced at Côte d'Ivoire-Gagnoa (for all climate change factors, me- dium confidence; Figure 3b), the site with the highest SOC loss (Figure S3). In contrast, the differences were smallest at Kenya- Embu (medium confidence), the site with the highest initial SOC and soil clay content. In the scenarios of climate change alone (i.e., without incor- porating soil fertility decline in the model simulations, see Methods), simulated maize yields without C and N inputs de- creased significantly (−10%) under a +2°C temperature increase at Côte d'Ivoire-Gagnoa, which is the warmest site (Figure 3b, high confidence). In contrast, simulated yields increased under warmer temperatures at Kenya-Embu (Figure 3b, medium con- fidence), the site with the coolest baseline climate. This was likely due to temperatures approaching the optimum thresholds for maize photosynthesis (Rezaei et al. 2023), as also reported for other high-altitude sites in Kenya (Waha et  al.  2013). At Kenya-Machanga and Zimbabwe-Murewa, a +2°C temperature increase had no effect on simulated maize yields, although with low confidence among the models. This uncertainty may arise from differences in how models simulate the processes driving yield reductions (shorter growth period and frequency of heat stresses on for example, photosynthesis, grain number develop- ment, and grain filling) or yield increases (when temperatures are nearing optimal thresholds for photosynthesis) (Webber et al. 2022). Changes in simulated maize yield due to increased or reduced rainfall were strongest at the two Kenyan sites (Figure 3b). At Zimbabwe-Murewa, simulated yields remained unchanged under higher rainfall but with low confidence among the models. This uncertainty could come from differences in how models simulate nitrate leaching and waterlogging, which can reduce yields under excessive rainfall (Webber et al. 2022; Kim et al. 2024). When comparing the combined effect of soil fertility decline and climate change factors on maize yield with the sum of their indi- vidual effects (Figure 3), it becomes clear that the models simulated only minor interactive effects, on average and per site. Indirect ef- fects of climate change factors resulted in yield changes of +2% for warmer temperatures, +3% for lower rainfall, −3% for higher rainfall, and −1% for higher CO2 levels (Figure S4a). These small interactive effects are largely explained by the minimal impacts of climate change factors on SOC dynamics in the model simulations (Figure S3, high confidence). 3.3   |   Maize Yield Responses to ISFM Increase Over Time Under the baseline climate scenario, the combined application of mineral and organic inputs increased average maize yield across the four sites by about 182% over the 30-year simulation period compared to no input use (Figure 4a, high confidence). The simu- lated yield response decreased slightly in the scenarios of warmer temperatures (170% yield increase) and lower rainfall (156%), but increased under higher rainfall (215%) and higher CO2 levels (186%; Figure 4a). The yield responses were even more pronounced in scenarios with more severe climate change (e.g., +6°C warm- ing and −50% rainfall resulted in 160% and 95% yield increases, respectively; Figure S5). The relatively small differences in sim- ulated yield responses between the baseline and climate change scenarios can be partly attributed to the simulation of small direct effects of the climate change factors on SOC dynamics, as influ- enced by input use (−5.5% SOC in the baseline versus −4.5% to −6% in the climate change scenarios), leading to only modest dif- ferences in maize growth over time (Figure S6, high confidence). At all sites, simulated maize yield responses to inputs increased over time both in the baseline climate (medium confidence) and climate change scenarios (except for the reduced rainfall scenario at Kenya-Embu and Kenya-Machanga; Figure  4b). Simulated yield responses to C and N inputs were larger at Côte d'Ivoire-Gagnoa, Zimbabwe-Murewa, and Kenya-Machanga, which are characterized by low initial soil fertility and high SOC loss without fertilization, than at Kenya-Embu, characterized by a high initial SOC content (medium confidence). With mineral and organic inputs, simulated SOC was maintained over time at Kenya-Machanga and Zimbabwe-Murewa, but slightly decreased at Côte d'Ivoire-Gagnoa and Kenya-Embu. Despite the SOC de- cline, absolute maize yield at Côte d'Ivoire-Gagnoa remained constant, in contrast to the scenario without C and N inputs, where both SOC and yield declined (Figure S7). At Kenya-Embu, the combined use of mineral and organic inputs was simulated to have the lowest yield benefit. It was the only site where simu- lated maize yields decreased significantly over time with C and N inputs (Figure S7). 4   |   Discussion 4.1   |   Soil Fertility Decline Outweighs Climate Change Effects on Maize Yield Our study has important implications for understanding the rel- ative impact of soil fertility and climate change on maize yields 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 9 of 15Global Change Biology, 2026 in smallholder cropping systems in SSA, which are typically characterized by maize monoculture with low input use. To our knowledge, this is the first study to use calibrated models that reproduce long-term trends in both soil and crop yield dynamics to investigate soil–crop feedbacks, apart from a few comparable modeling studies conducted in temperate conditions (e.g., Smith et al. 2020a, 2020b). Our model simulations show that if fertilizer inputs remain low, maize yields in SSA are expected to further decline by 20%–50% over the next three decades, as a result of soil nutri- ent depletion (notably N, as represented in our study). Soil nu- trient depletion driven by SOC decline is a major form of soil degradation in SSA (Kihara et al. 2020), reducing crop produc- tivity, biomass, and soil cover (Woomer and Muchena 1996). This, in turn, accelerates other degradation processes like erosion, acidification, and compaction, further reducing soil health (Lal et  al.  1997); these processes are not accounted for in the simulations of the study and would require further model improvement. Poor soil health is linked to low crop pro- ductivity, food insecurity, and persistent rural poverty across the continent (Sanchez 2002). Confirming our first hypothesis, climate change is expected to have a relatively smaller impact on maize yields than soil fer- tility decline, with the contrast in maize yield responses being strongest at sites with greater SOC loss and weakest at those with higher initial SOC and clay content. However, climate change may exacerbate these impacts, with the magnitude and direction of yield changes varying substantially across regions and maize mega-environments in SSA. Our model simulations indicate that rising temperatures and decreasing rainfall are likely to reduce maize yields in areas located in West Africa (Côte d'Ivoire-Gagnoa in our study), whereas in- creasing temperatures combined with higher rainfall are ex- pected to enhance yields in the highland regions of Eastern Africa (Kenya-Embu in our study). Lowland and dry mid- altitude areas may experience greater relative yield reductions compared to highland and mid-altitude environments (Waha et al. 2013; Dale et al. 2017). Besides, impacts will likely vary within regions, depending on differences in crop manage- ment and resource use (e.g., Falconnier et al. 2020; Siatwiinda et al. 2021). The interaction between climate change and soil fertility de- cline, defined as the non-additive component of their com- bined effects on maize yield, was relatively small. This finding is notable because soil organic matter is expected to decrease under warmer climate due to higher mineralization rates (e.g., Nissan et al. 2023), potentially leading to negative soil– crop feedback loops. A stronger influence of climate change on soil fertility might have been expected through its effects on mineralization dynamics, yet this was not evident in our simulations. 4.2   |   ISFM as a No-Regret Climate Change Adaptation Option African governments have endorsed the 2024 Nairobi Declaration on Fertilizer and Soil Health, emphasizing the need to promote agricultural practices that improve soil health, to in- crease fertilizer use, and to enhance food security across SSA (African Union 2024). In line with findings from long-term field experiments conducted in several countries in SSA (e.g., Adams et al. 2020; Cardinael et al. 2022; Sommer et al. 2018), our model simulations show the potential to maintain SOC or slow its loss and enhance maize productivity through the combined appli- cation of mineral fertilizer and organic inputs. The simulations further suggest an increasing yield benefit of mineral and or- ganic inputs over time due to their accumulating positive effects on SOC. Combined application of mineral fertilizer and organic inputs is one of the core principles of ISFM, recognizing that substantial improvements in crop productivity in SSA require addressing both the immediate nutrient needs of crops and the long-term maintenance of soil fertility (Vanlauwe et al. 2010). The combined application of mineral and organic inputs is often promoted as an effective climate change adaptation strategy in SSA (Nezomba et al. 2018; Gram et al. 2020; Carr et al. 2022; Markos et al. 2023). Our findings support this to a certain extent, showing that their combined use acts as a ‘true-adaptation’ measure (Lobell 2014) to increased rainfall, with yield benefits becoming more pronounced under cli- mate change in our model simulations. However, the model ensemble also indicates that the combined use of mineral and organic inputs, whilst increasing maize yields, does not mitigate the negative impacts of warmer temperatures and reduced rainfall. Therefore, application of mineral and or- ganic inputs can be seen as a ‘no-regret adaptation’, benefi- cial under current conditions, yet insufficient on its own to counteract future climatic stresses (Hallegatte 2009), thereby confirming the second hypothesis of this study. Finally, it is also important to note that the no-regret adaptation benefits of input use under warmer and drier climates could be further enhanced by reducing crop sensitivity to climate extremes, for example through the use of heat- and drought-tolerant culti- vars, longer-cycle cultivars, or adjusted sowing dates (Hunt et  al.  2019; Hasegawa et  al.  2022). These strategies form an integral part of the ISFM approach but were not simulated in our study. 4.3   |   Limitations of the Study and Future Research Directions In most climate change impact studies conducted to date, soil- crop models were reinitialized each year with the same initial soil water, soil N, and SOC levels (e.g., Dale et al. 2017; Falconnier et al. 2020; Waha et al. 2013). As a result, these studies fail to account for the shifts in baseline crop productivity caused by progressive soil fertility loss over time, which our study shows to be of high importance for low-input cropping systems in SSA. Only a few studies, all conducted in temperate regions, have included long-term soil-crop feedback and reported their significant impact on simulated crop yields (Basso et al. 2015, 2018; Smith et al. 2020a). Here we argue that to accurately assess the long-term impacts of climate change and the effectiveness of adaptation strategies, models must be run continuously, in- corporating soil-crop feedback mechanisms. This is essential to prevent underestimation of both yield losses due to soil fertility decline and yield gains from fertilizer application. 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 10 of 15 Global Change Biology, 2026 Our study calls for future research to reduce uncertainties in assessing the impacts of climate change on crop growth and soil fertility, as displayed by the variability of outputs from our modeling ensemble. Although our findings suggest that soil degradation currently exerts a stronger influence on maize yield than climate change in our study locations, they highlight the dual challenges of declining soil fertility and increasing climate threats over the coming decades. Over longer time horizons, cli- mate change effects are expected to become more pronounced and will likely emerge as the predominant driver of crop pro- ductivity (Waha et al. 2013; Dale et al. 2017). There are, how- ever, a number of important limitations to consider in our study and in most crop simulation modelling of climate change im- pacts in general (Corbeels et al. 2018; Kephe et al. 2021; Wang, Jägermeyr, et al. 2024). First, our simulations do not account for the projected increase in weather variability with more frequent extreme events. Extreme temperatures causing heat stress in crops and extreme rainfall leading to drought or flooding are likely to become more common with climate change (Taylor et  al.  2017; IPCC  2023). Some regions in SSA already experi- ence crop failure rates of 15% or more, and these are projected to increase significantly in the future (Parkes et al. 2018). Many current crop models do not adequately simulate the effects of ex- treme weather events and this requires further methodological development (Webber et al. 2022; Kim et al. 2024), coupled with innovative field experiments (for example manipulating rain- fall in SSA) to better calibrate these processes (Bouhenache et  al.  2025). Second, experiments show that the combined ef- fects of heat and drought stress have a more adverse impact on maize grain yield than the individual stresses alone (Cohen et al. 2021; Bheemanahalli et al. 2022), leading to larger yield penalties and increased yield variability. Models need further development and evaluation to account for these interactions (Webber et al. 2022). Using individual change in temperature, precipitation, and CO2, our study omits factorial combinations between climate factors (e.g., CO2 × temperature or CO2 × rain- fall), which may lead to under- or overestimation of combined effects under real-world conditions. However, this was a deliber- ate methodological choice, as most current crop–soil models are not yet well equipped to simulate multiple simultaneous stresses reliably, and including them would have introduced additional model uncertainty (Webber et  al.  2022). Recent work has fur- ther demonstrated the diagnostic value of this approach, show- ing that modeled yield sensitivities to individual climate factors differ widely across models (Müller et al. 2024). Third, climate change directly impacts soil fertility. Intensive rainfall events, expected to become more frequent with climate change (Donat et al. 2016; Reed et al. 2022), accelerate soil deg- radation and erosion (Van Oost et al. 2007; Borrelli et al. 2020). Most soil-crop models either oversimplify or omit soil erosion processes (Raza et al. 2021). Accurately assessing the impact of climate change on soil fertility processes requires incorporating the relationships between intense rainfall and soil erosion into models. Moreover, climate extremes, such as heatwaves, are also expected to have an exacerbated negative impact on SOC (Anjileli et al. 2021), further declining soil fertility (Lugato 2024; Wang, Zhang, et al. 2024). Our model-ensemble simulated rel- atively small impacts of climate change factors on SOC which likely led to an underestimation of the negative impacts of cli- mate change on maize yield. Finally, climate change may also alter the dynamics of pathogens and pests, important yield- reducing factors not yet captured by current soil-crop models. While addressing these biotic interactions is beyond the scope of this study, it represents a valuable direction for future research and model development (Savary et al. 2018). A strong engage- ment of the modeling community is now critically required to address the above modeling issues on extreme climate events and conduct climate change impact studies that account for soil fertility changes, thereby enabling more realistic projections of climate change impacts on agriculture in SSA. 5   |   Conclusion Over the long term, soil fertility decline can reduce maize yields more strongly than climate change in SSA, making soil degrada- tion a priority to address alongside climate adaptation efforts. By explicitly accounting for soil fertility changes, our findings support the value of ISFM, through the combined application of organic and mineral inputs, as an effective long-term no-regret adaptation strategy to sustain and increase yields under both current and fu- ture climates. We recommend that assessments of long-term cli- mate impacts and adaptation strategies require continuous model simulations that capture soil–crop feedbacks to realistically rep- resent long-term system dynamics. Finally, improving model rep- resentation of climate extremes and interacting stressors on crops and soils will be key to avoid underestimating climate change im- pacts in smallholder farming systems in sub-Saharan Africa. Author Contributions Antoine Couëdel: conceptualization, data curation, formal analysis, investigation, methodology, project administration, software, supervi- sion, validation, visualization, writing – original draft, writing – review and editing. Gatien N. Falconnier: conceptualization, formal analysis, funding acquisition, investigation, methodology, project administra- tion, resources, supervision, validation, visualization, writing – original draft, writing – review and editing. Myriam Adam: conceptualization, investigation, methodology, supervision, validation, visualization, writ- ing – review and editing. Rémi Cardinael: conceptualization, funding acquisition, investigation, methodology, resources, supervision, valida- tion, visualization, writing – review and editing. Johan Six: conceptu- alization, data curation, funding acquisition, investigation, methodology, resources, supervision, validation, writing – review and editing. Moritz Laub: conceptualization, data curation, formal analysis, investigation, methodology, software, validation, visualization, writing – review and editing. Alex C. Ruane: conceptualization, funding acquisition, method- ology, project administration, resources, supervision, validation, writing – review and editing. Kenneth Boote: conceptualization, investigation, methodology, software, validation, visualization, writing – review and editing. Eric Justes: conceptualization, investigation, methodology, software, validation, writing – review and editing. Ward N. Smith: con- ceptualization, investigation, methodology, software, validation, writing – review and editing. Anthony M. Whitbread: conceptualization, in- vestigation, methodology, software, validation, writing – review and ed- iting. François Affholder: investigation, software, validation, writing – review and editing. Folorunso M. Akinseye: investigation, software, validation, writing – review and editing. Juraj Balkovic: investigation, software, validation, writing – review and editing. Bruno Basso: inves- tigation, software, validation, writing – review and editing. Arti Bhatia: investigation, software, validation, writing – review and editing. Bidisha Chakrabarti: investigation, software, validation, writing – review and editing. Regis Chikowo: data curation, investigation, resources, valida- tion, writing – review and editing. Mathias Christina: investigation, 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 11 of 15Global Change Biology, 2026 software, validation, writing – review and editing. Babacar Faye: in- vestigation, software, validation, writing – review and editing. Fabien Ferchaud: investigation, software, validation, writing – review and editing. Christian Folberth: investigation, software, validation, writ- ing – review and editing. Thomas Gaiser: investigation, software, val- idation, writing – review and editing. Marcelo Galdos: investigation, software, validation, writing – review and editing. Sebastian Gayler: investigation, software, validation, writing – review and editing. Aram Gorooei: investigation, software, visualization, writing – review and editing. Brian Grant: investigation, software, validation, writing – re- view and editing. Hervé Guibert: data curation, investigation, valida- tion, writing – review and editing. Gerrit Hoogenboom: investigation, methodology, software, validation, visualization, writing – review and editing. Bahareh Kamali: investigation, software, validation, writing – review and editing. Fidel Maureira: investigation, software, validation, writing – review and editing. Fasil Mequanint: investigation, software, visualization, writing – review and editing. Claas Nendel: investiga- tion, software, visualization, writing – review and editing. Cheryl H. Porter: investigation, software, validation, writing – review and edit- ing. Dominique Ripoche: investigation, software, validation, writing – review and editing. Leonard Rusinamhodzi: data curation, inves- tigation, validation, writing – review and editing. Shikha Sharma: in- vestigation, software, validation, writing – review and editing. Upendra Singh: investigation, software, validation, writing – review and ed- iting. Amit Srivastava: investigation, software, validation, writing – review and editing. Bernard Vanlauwe: funding acquisition, investi- gation, methodology, resources, validation, writing – review and editing. Antoine Versini: investigation, software, validation, writing – review and editing. Murilo Vianna: investigation, software, validation, writing – review and editing. Heidi Webber: investigation, software, validation, writing – review and editing. Tobias Weber: investigation, software, validation, writing – review and editing. Congmu Zhang: investigation, software, validation, writing – review and editing. Marc Corbeels: con- ceptualization, funding acquisition, investigation, methodology, project administration, resources, supervision, validation, visualization, writing – original draft, writing – review and editing. Affiliations 1AIDA, Univ Montpellier, CIRAD, Montpellier, France  |  2CIRAD, UPR Aida, Montpellier, France  |  3CIRAD, UPR Aida, Harare, Zimbabwe  |  4International Maize and Wheat Improvement Centre (CIMMYT)-Zimbabwe, Harare, Zimbabwe  |  5Department of Plant Production Sciences and Technologies, University of Zimbabwe, Harare, Zimbabwe  |  6AGAP Institut, Univ. Montpellier, INRAE, CIRAD, Inst. Agro., Montpellier, France  |  7Faculty of Agriculture and Food Processing, National University of Battambang, Krong Battambang, Battambang Province, Cambodia  |  8Department of Environmental Systems Science, ETH Zurich, Zurich, Switzerland  |  9NASA Goddard Institute for Space Studies, New York, New York, USA  |  10Department of Agricultural and Biological Engineering, University of Florida, Gainesville, Florida, USA  |  11PERSYST Department, Univ Montpellier, CIRAD, Montpellier, France  |  12Ottawa Research and Development Centre, Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada  |  13International Livestock Research Institute, Dar es Salaam, Tanzania  |  14CIRAD, UPR Aida, Maputo, Mozambique  |  15Faculdade de Agronomia e Engenharia Florestal, Universidade Eduardo Mondlane, Maputo, Mozambique  |  16International Crop Research Institute for the Semi-Arid Tropics (ICRISAT), Thiès, Senegal  |  17Centre d'Etude Régional pour l'Amélioration de l'Adaptation à la Sécheresse (CERAAS), Thiès Escale, Senegal  |  18Biodiversity and Natural Resources Program, International Institute for Applied Systems Analysis, Laxenburg, Austria  |  19Department of Earth and Environmental Sciences, Michigan State University, East Lansing, Michigan, USA  |  20ICAR, Indian Agricultural Research Institute, New Delhi, India  |  21Plant, Soil and Microbial Sciences Department, Michigan State University, East Lansing, Michigan, USA  |  22CIRAD, UPR Aida, Saint-Denis, La Réunion, France  |  23Université du Sine Saloum El Hadj Ibrahima Niass (USSEIN), Kaolack, Senegal  |  24BioEcoAgro Joint Research Unit, INRAE, Université de Liège, Université de Lille, Université de Picardie Jules Verne, Barenton-Bugny, France  |  25UMR Eco&Sols, Univ Montpellier, CIRAD, INRAE, IRD, InstitutAgro Montpellier, Montpellier, France  |  26Institute of Crop Science and Resource Conservation, University of Bonn, Bonn, Germany  |  27Sustainable Soils & Crops, Rothamsted Research, Harpenden, UK  |  28Biogeophysics, Institute of Soil Science and Land Evaluation, Hohenheim University, Stuttgart, Germany  |  29Institute of Bio- and Geosciences, Agrosphere (IBG-3), Forschungszentrum Jülich GmbH, Jülich, Germany  |  30CIMMYT, Addis Ababa, Ethiopia  |  31Leibniz Centre for Agricultural Landscape Research (ZALF), Müncheberg, Germany  |  32Institute of Biochemistry and Biology, University of Potsdam, Potsdam, Germany  |  33Global Change Research Institute CAS, Brno, Czech Republic  |  34AgroClim, INRAE, Avignon, France  |  35International Institute of Tropical Agriculture, Accra, Ghana  |  36International Fertilizer Development Center (IFDC), Muscle Shoals, Alabama, USA  |  37International Institute of Tropical Agriculture (IITA), Nairobi, Kenya  |  38CIRAD, UPR Recyclage & Risque, Saint-Denis, La Réunion, France  |  39Faculty of Environment and Natural Sciences, Brandenburg University of Technology, Cottbus, Germany  |  40Section of Soil Science and Agrohydrology, University of Kassel, Kassel, Germany  |  41CIRAD, UPR AIDA, Nairobi, Kenya Acknowledgements We are grateful to the support provided by the Agricultural Model Intercomparison and Improvement Project (AgMIP) leadership team and those who participated as part of the AgMIP Low Input Farming Systems Activity. Part of the study was funded by the DSCATT project “Agricultural Intensification and Dynamics of Soil Carbon Sequestration in Tropical and Temperate Farming Systems” (N° AF 1802–001, N° FT C002181), supported by the Agropolis Foundation (“Programme d' Investissement d' Avenir” Labex Agro, ANR-10-LABX0001–01) and by the TOTAL Foundation within a patronage agreement. This study was also supported by the RAIZ “Promoting agroecological intensification for resilience building” project FOOD/2021/424–933 (https://​raiz.​org.​ zw/​) funded by the European Union, and from the ALAMOD project (https://​www.​pepr-​fairc​arbon.​fr/​eng) of the exploratory research pro- gram FairCarboN and received government funding managed by the Agence Nationale de la Recherche under the France 2030 program, reference ANR-22-PEXF-0002. A.C.R. participation was made possible by NASA Earth Science Division support to the NASA GISS Climate Impacts Group. We thank the CGIAR Sustainable Farming Program for their support to this research. Funding This work was supported by Agropolis Fondation, ANR- 10-LABX0001-01. Conflicts of Interest The authors declare no conflicts of interest. Data Availability Statement The data that support the findings of this study are available at CIRAD repository: https://​doi.​org/​10.​18167/​​DVN1/​EGJXVL. References Adams, A. M., A. W. Gillespie, G. S. Dhillon, et al. 2020. “Long-Term Effects of Integrated Soil Fertility Management Practices on Soil Chemical Properties in the Sahel.” Geoderma 366: 114207. https://​doi.​ org/​10.​1016/j.​geode​rma.​2020.​114207. African Union. 2024. Africa Fertilizer and Soil Health Summit 7–9 th May 2024. African Union. Aiteew, K., J. Rouhiainen, C. Nendel, and R. Dechow. 2024. “Evaluation and Optimisation of the Soil Carbon Turnover Routine in the MONICA Model (Version 3.3.1).” 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://raiz.org.zw/ https://raiz.org.zw/ https://www.pepr-faircarbon.fr/eng https://doi.org/10.18167/DVN1/EGJXVL https://doi.org/10.1016/j.geoderma.2020.114207 https://doi.org/10.1016/j.geoderma.2020.114207 12 of 15 Global Change Biology, 2026 Anjileli, H., L. S. Huning, H. Moftakhari, et  al. 2021. “Extreme Heat Events Heighten Soil Respiration.” Scientific Reports 11: 1–9. https://​doi.​ org/​10.​1038/​s4159​8-​021-​85764​-​8. Basso, B., D. Cammarano, A. Troccoli, D. Chen, and J. T. Ritchie. 2010. “Long-Term Wheat Response to Nitrogen in a Rainfed Mediterranean Environment: Field Data and Simulation Analysis.” European Journal of Agronomy 33: 132–138. https://​doi.​org/​10.​1016/j.​eja.​2010.​04.​004. Basso, B., B. Dumont, B. Maestrini, et al. 2018. “Soil Organic Carbon and Nitrogen Feedbacks on Crop Yields Under Climate Change.” Agricultural & Environmental Letters 3: 180026. https://​doi.​org/​10.​2134/​ ael20​18.​05.​0026. Basso, B., D. W. Hyndman, A. D. Kendall, et al. 2015. “Can Impacts of Climate Change and Agricultural Adaptation Strategies Be Accurately Quantified if Crop Models Are Annually Re-Initialized?” PLoS One 10: 1–12. https://​doi.​org/​10.​1371/​journ​al.​pone.​0127333. Beaudoin, N., P. Lecharpentier, D. Ripoche, et al. 2023. STICS Soil-Crop Model, Conceptual Framework, Equations and Uses. Éditions Quæ. Behnassi, M., and M. El Haiba. 2022. “Implications of the Russia– Ukraine War for Global Food Security.” Nature Human Behaviour 6: 754–755. https://​doi.​org/​10.​1038/​s4156​2-​022-​01391​-​x. Bheemanahalli, R., R. R. Vennam, P. Ramamoorthy, and K. R. Reddy. 2022. “Effects of Post-Flowering Heat and Drought Stresses on Physiology, Yield, and Quality in Maize (Zea Mays L.).” Plant Stress 6: 100106. https://​doi.​org/​10.​1016/j.​stress.​2022.​100106. Biernath, C., S. Gayler, S. Bittner, et  al. 2011. “Evaluating the Ability of Four Crop Models to Predict Different Environmental Impacts on Spring Wheat Grown in Open-Top Chambers.” European Journal of Agronomy 35: 71–82. https://​doi.​org/​10.​1016/j.​eja.​2011.​04.​001. Borrelli, P., D. A. Robinson, P. Panagos, et al. 2020. “Land Use and Climate Change Impacts on Global Soil Erosion by Water (2015- 2070).” Proceedings of the National Academy of Sciences of the United States of America 117: 21994–22001. https://​doi.​org/​10.​1073/​pnas.​ 20014​03117​. Bouhenache, A., R. Chikowo, A. Shumba, et al. 2025. “Intra-Seasonal Rainfall Patterns and Extremes Drive Maize Productivity and Nitrogen Use in Sub-Humid Zimbabwe.” Field Crops Research 334: 110126. https://​doi.​org/​10.​1016/j.​fcr.​2025.​110126. Brisson, N., C. Gary, E. Justes, et al. 2003. “An Overview of the Crop Model STICS.” European Journal of Agronomy 18: 309–332. https://​doi.​ org/​10.​1016/​S0223​-​5234(03)​00100​-​4. Cardinael, R., H. Guibert, S. T. K. Brédoumy, et al. 2022. “Sustaining Maize Yields and Soil Carbon Following Land Clearing in the Forest– Savannah Transition Zone of West Africa: Results From a 20-Year Experiment.” Field Crops Research 275: 108335. https://​doi.​org/​10.​ 1016/j.​fcr.​2021.​108335. Carr, T. W., S. Mkuhlani, A. C. Segnon, et al. 2022. “Climate Change Impacts and Adaptation Strategies for Crops in West Africa: A Systematic Review.” Environmental Research Letters 17: 053001. https://​ doi.​org/​10.​1088/​1748-​9326/​ac61c8. Challinor, A. J., J. Watson, D. B. Lobell, S. M. Howden, D. R. Smith, and N. Chhetri. 2014. “A Meta-Analysis of Crop Yield Under Climate Change and Adaptation.” Nature Climate Change 4: 287–291. https://​ doi.​org/​10.​1038/​nclim​ate2153. Chivenge, P., B. Vanlauwe, R. Gentile, et al. 2009. “Organic and Mineral Input Management to Enhance Crop Productivity in Central Kenya.” Agronomy Journal 101: 1266–1275. https://​doi.​org/​10.​2134/​agron​j2008.​ 0188x​. Chivenge, P., B. Vanlauwe, R. Gentile, and J. Six. 2011. “Comparison of Organic Versus Mineral Resource Effects on Short-Term Aggregate Carbon and Nitrogen Dynamics in a Sandy Soil Versus a Fine Textured Soil.” Agriculture, Ecosystems & Environment 140: 361–371. https://​doi.​ org/​10.​1016/j.​agee.​2010.​12.​004. Cohen, I., S. I. Zandalinas, C. Huck, F. B. Fritschi, and R. Mittler. 2021. “Meta-Analysis of Drought and Heat Stress Combination Impact on Crop Yield and Yield Components.” Physiologia Plantarum 171: 66–76. https://​doi.​org/​10.​1111/​ppl.​13203​. Corbeels, M., D. Berre, L. Rusinamhodzi, and S. Lopez-Ridaura. 2018. “Can We Use Crop Modelling for Identifying Climate Change Adaptation Options?” Agricultural and Forest Meteorology 256: 46–52. https://​doi.​org/​10.​1016/j.​agrfo​rmet.​2018.​02.​026. Couëdel, A., G. N. Falconnier, M. Adam, et al. 2024. “Long Term Soil Organic Carbon and Crop Yield Feedbacks Differ Between 16 Soil-Crop Models in Sub-Saharan Africa.” European Journal of Agronomy 155: 127109. https://​doi.​org/​10.​1016/j.​eja.​2024.​127109. Couëdel, A., G. N. Falconnier, M. Adam, et al. 2026. “Data for “Beyond Climate Change: The Role of Integrated Soil Fertility Management for Sustaining Future Maize Yield in Sub-Saharan Africa”.” Global Change Biology. https://​doi.​org/​10.​18167/​​DVN1/​EGJXVL. Couëdel, A., M. Laub, R. Ranaivomanana, et  al. 2026. “Evaluating DayCent and STICS in Simulating the Long-Term Impact of Contrasting Organic Resource Amendments on Soil Organic Carbon and Maize Yields in Sub-Saharan Africa.” Field Crops Research 335: 110169. https://​ doi.​org/​10.​1016/j.​fcr.​2025.​110169. Dale, A., C. Fant, K. Strzepek, M. Lickley, and S. Solomon. 2017. “Climate Model Uncertainty in Impact Assessments for Agriculture: A Multi-Ensemble Case Study on Maize in Sub-Saharan Africa.” Earth's Future 5: 337–353. https://​doi.​org/​10.​1002/​2017E​F000539. Del Grosso, S., W. Parton, A. Mosier, et al. 2001. “Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas FLewis Using the DAYCENT Model.” In Modeling Carbon and Nitrogen Dynamics for Soil Management, edited by M. J. Shaffer, L. Ma, and S. Hansen. Lewis Publishers. Donat, M. G., A. L. Lowry, L. V. Alexander, P. A. O'Gorman, and N. Maher. 2016. “More Extreme Precipitation in the World's Dry and Wet Regions.” Nature Climate Change 6: 508–513. https://​doi.​org/​10.​1038/​ nclim​ate2941. Duarte, Y. C. N., and P. C. Sentelhas. 2019. “NASA/POWER and DailyGridded Weather Datasets — How Good They Are for Estimating Maize Yields in Brazil?” International Journal of Biometeorology 64: 319–329. Ekpa, O., N. Palacios-Rojas, G. Kruseman, V. Fogliano, and A. R. Linnemann. 2019. “Sub-Saharan African Maize-Based Foods - Processing Practices, Challenges and Opportunities.” Food Reviews International 35: 609–639. https://​doi.​org/​10.​1080/​87559​129.​2019.​ 1588290. Enders, A., M. Vianna, T. Gaiser, et al. 2023. “SIMPLACE—A Versatile Modelling and Simulation Framework for Sustainable Crops and Agroecosystems.” In Silico Plants 5: 1–18. https://​doi.​org/​10.​1093/​insil​ icopl​ants/​diad006. Falconnier, G. N., R. Cardinael, M. Corbeels, et  al. 2023. “The Input Reduction Principle of Agroecology Is Wrong When It Comes to Mineral Fertilizer Use in Sub-Saharan Africa.” Outlook on Agriculture 52: 311–326. https://​doi.​org/​10.​1177/​00307​27023​1199795. Falconnier, G. N., M. Corbeels, K. J. Boote, et  al. 2020. “Modelling Climate Change Impacts on Maize Yields Under Low Nitrogen Input Conditions in Sub-Saharan Africa.” Global Change Biology 26: 1–23. https://​doi.​org/​10.​1111/​gcb.​15261​. FAOSTAT. 2023. Food and Agriculture Organization of the United Nations. FAO. Faye, B., H. Webber, M. Diop, et al. 2018. “Potential Impact of Climate Change on Peanut Yield in Senegal, West Africa.” Field Crops Research 219: 148–159. https://​doi.​org/​10.​1016/j.​fcr.​2018.​01.​034. Gaiser, T., U. Perkons, P. M. Küpper, et  al. 2013. “Modeling Biopore Effects on Root Growth and Biomass Production on Soils With 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.1038/s41598-021-85764-8 https://doi.org/10.1038/s41598-021-85764-8 https://doi.org/10.1016/j.eja.2010.04.004 https://doi.org/10.2134/ael2018.05.0026 https://doi.org/10.2134/ael2018.05.0026 https://doi.org/10.1371/journal.pone.0127333 https://doi.org/10.1038/s41562-022-01391-x https://doi.org/10.1016/j.stress.2022.100106 https://doi.org/10.1016/j.eja.2011.04.001 https://doi.org/10.1073/pnas.2001403117 https://doi.org/10.1073/pnas.2001403117 https://doi.org/10.1016/j.fcr.2025.110126 https://doi.org/10.1016/S0223-5234(03)00100-4 https://doi.org/10.1016/S0223-5234(03)00100-4 https://doi.org/10.1016/j.fcr.2021.108335 https://doi.org/10.1016/j.fcr.2021.108335 https://doi.org/10.1088/1748-9326/ac61c8 https://doi.org/10.1088/1748-9326/ac61c8 https://doi.org/10.1038/nclimate2153 https://doi.org/10.1038/nclimate2153 https://doi.org/10.2134/agronj2008.0188x https://doi.org/10.2134/agronj2008.0188x https://doi.org/10.1016/j.agee.2010.12.004 https://doi.org/10.1016/j.agee.2010.12.004 https://doi.org/10.1111/ppl.13203 https://doi.org/10.1016/j.agrformet.2018.02.026 https://doi.org/10.1016/j.eja.2024.127109 https://doi.org/10.18167/DVN1/EGJXVL https://doi.org/10.1016/j.fcr.2025.110169 https://doi.org/10.1016/j.fcr.2025.110169 https://doi.org/10.1002/2017EF000539 https://doi.org/10.1038/nclimate2941 https://doi.org/10.1038/nclimate2941 https://doi.org/10.1080/87559129.2019.1588290 https://doi.org/10.1080/87559129.2019.1588290 https://doi.org/10.1093/insilicoplants/diad006 https://doi.org/10.1093/insilicoplants/diad006 https://doi.org/10.1177/00307270231199795 https://doi.org/10.1111/gcb.15261 https://doi.org/10.1016/j.fcr.2018.01.034 13 of 15Global Change Biology, 2026 Pronounced Sub-Soil Clay Accumulation.” Ecological Modelling 256: 6–15. https://​doi.​org/​10.​1016/j.​ecolm​odel.​2013.​02.​016. Gentile, R., B. Vanlauwe, C. van Kessel, and J. Six. 2009. “Managing N Availability and Losses by Combining Fertilizer-N With Different Quality Residues in Kenya.” Agriculture, Ecosystems & Environment 131: 308–314. https://​doi.​org/​10.​1016/j.​agee.​2009.​02.​003. Gijsman, A. J., G. Hoogenboom, W. J. Parton, and P. C. Kerridge. 2002. “Modifying DSSAT Crop Models for Low-Input Agricultural Systems Using a Soil Organic Matter–Residue Module From CENTURY.” Agronomy Journal 94: 462–474. https://​doi.​org/​10.​2134/​agron​ j2002.​4620. Giller, K. E. 2020. “The Food Security Conundrum of Sub-Saharan Africa.” Global Food Security 26: 100431. https://​doi.​org/​10.​1016/j.​gfs.​ 2020.​100431. Giller, K. E., P. Tittonell, M. C. Rufino, et  al. 2011. “Communicating Complexity: Integrated Assessment of Trade-Offs Concerning Soil Fertility Management Within African Farming Systems to Support Innovation and Development.” Agricultural Systems 104: 191–203. https://​doi.​org/​10.​1016/j.​agsy.​2010.​07.​002. Godwin, D. C., and C. A. Jones. 1991. “Nitrogen Dynamics in Soil-Plant Systems.” In Modeling Plant and Soil Systems-Agronomy Monograph no 31, edited by J. Hanks and J. T. Ritchie. American Society of Agronomy. Gram, G., D. Roobroeck, P. Pypers, J. Six, R. Merckx, and B. Vanlauwe. 2020. “Combining Organic and Mineral Fertilizers as a Climate-Smart Integrated Soil Fertility Management Practice in Sub-Saharan Africa: A Meta-Analysis.” PLoS One 15: e0239552. https://​doi.​org/​10.​1371/​journ​ al.​pone.​0239552. Guibert, H. 1999. Évolution de la Matière Organique et de la Capacité D'échange Cationique des Alfisols Tropicaux Cultivés. Institut National Polytechnique de Lorraine. Hallegatte, S. 2009. “Strategies to Adapt to an Uncertain Climate Change.” Global Environmental Change 19: 240–247. https://​doi.​org/​10.​ 1016/j.​gloen​vcha.​2008.​12.​003. Hartkamp, A. D., J. W. White, A. R. Aguilar, et  al. 2000. Maize Production Environments Revisited: A GIS-Based Approach. Natural Resources Group. Hasegawa, T., H. Wakatsuki, H. Ju, et al. 2022. “A Global Dataset for the Projected Impacts of Climate Change on Four Major Crops.” Scientific Data 9: 1–11. Holzworth, D. P., N. I. Huth, P. G. deVoil, et al. 2014. “APSIM - Evolution Towards a New Generation of Agricultural Systems Simulation.” Environmental Modelling & Software 62: 327–350. https://​doi.​org/​10.​ 1016/j.​envso​ft.​2014.​07.​009. Hoogenboom, G., C. H. Porter, V. Shelia, et al. 2021. “Decision Support System for Agrotechnology Transfer (DSSAT) Version 4.8.” DSSAT Foundation, Gainesville, Florida, USA. Hunt, J. R., J. M. Lilley, B. Trevaskis, et al. 2019. “Early Sowing Systems Can Boost Australian Wheat Yields Despite Recent Climate Change.” Nature Climate Change 9: 244–247. IPCC. 2023. Summary for Policymakers: Synthesis Report. IPCC. Izaurralde, R. C., J. R. Williams, W. B. McGill, N. J. Rosenberg, and M. C. Q. Jakas. 2006. “Simulating Soil C Dynamics With EPIC: Model Description and Testing Against Long-Term Data.” Ecological Modelling 192: 362–384. https://​doi.​org/​10.​1016/j.​ecolm​odel.​2005.​07.​010. Jägermeyr, J., C. Müller, A. C. Ruane, et al. 2021. “Climate Impacts on Global Agriculture Emerge Earlier in New Generation of Climate and Crop Models.” Nature Food 2: 873–885. https://​doi.​org/​10.​1038/​s4301​6-​ 021-​00400​-​y. Jayne, T. S., and P. A. Sanchez. 2021. “Agricultural Productivity Must Improve in Sub-Saharan Africa.” Science (80-) 372: 1045–1048. Kamali, B., F. Jahanbakhshi, D. Dogaru, J. Dietrich, C. Nendel, and A. AghaKouchak. 2022. “Probabilistic Modeling of Crop-Yield Loss Risk Under Drought: A Spatial Showcase for Sub-Saharan Africa.” Environmental Research Letters 17: 024028. https://​doi.​org/​10.​1088/​ 1748-​9326/​ac4ec1. Kephe, P. N., K. K. Ayisi, and B. M. Petja. 2021. “Challenges and Opportunities in Crop Simulation Modelling Under Seasonal and Projected Climate Change Scenarios for Crop Production in South Africa.” Agriculture & Food Security 10: 1–24. https://​doi.​org/​10.​1186/​ s4006​6-​020-​00283​-​5. Kihara, J., P. Bolo, M. Kinyua, S. S. Nyawira, and R. Sommer. 2020. “Soil Health and Ecosystem Services: Lessons From Sub-Sahara Africa (SSA).” Geoderma 370: 114342. https://​doi.​org/​10.​1016/j.​geode​rma.​ 2020.​114342. Kim, Y. U., H. Webber, S. G. K. Adiku, et  al. 2024. “Mechanisms and Modelling Approaches for Excessive Rainfall Stress on Cereals: Waterlogging, Submergence, Lodging, Pests and Diseases.” Agricultural and Forest Meteorology 344: 109819. https://​doi.​org/​10.​1016/j.​agrfo​rmet.​ 2023.​109819. Lal, R., A. Wagner, D. J. Greenland, et  al. 1997. “Degradation and Resilience of Soils [and Discussion].” Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 352: 997–1010. Laub, M., M. Corbeels, A. Couëdel, et al. 2023. “Managing Soil Organic Carbon in Tropical Agroecosystems : Evidence From Four Long-Term Experiments in Kenya.” Soil 9: 301–323. Laub, M., M. Corbeels, S. Mathu Ndungu, et  al. 2023. “Combining Manure With Mineral N Fertilizer Maintains Maize Yields: Evidence From Four Long-Term Experiments in Kenya.” Field Crops Research 291: 108788. https://​doi.​org/​10.​1016/j.​fcr.​2022.​108788. Lehmann, J., D. A. Bossio, I. Kögel-Knabner, and M. C. Rillig. 2020. “The Concept and Future Prospects of Soil Health.” Nature Reviews Earth and Environment 1: 544–553. https://​doi.​org/​10.​1038/​s4301​ 7-​020-​0080-​8. Lesk, C., E. Coffel, and R. Horton. 2020. “Net Benefits to US Soy and Maize Yields From Intensifying Hourly Rainfall.” Nature Climate Change 10: 819–822. https://​doi.​org/​10.​1038/​s4155​8-​020-​0830-​0. Lobell, D. B. 2014. “Climate Change Adaptation in Crop Production: Beware of Illusions.” Global Food Security 3: 72–76. https://​doi.​org/​10.​ 1016/j.​gfs.​2014.​05.​002. Lugato, E. 2024. “Soil Organic Carbon Losses Exacerbated by Climate Extremes.” Nature Climate Change 14: 17–18. https://​doi.​org/​10.​1038/​ s4155​8-​023-​01873​-​4. Markos, D., W. Worku, and G. Mamo. 2023. “Exploring Adaptation Responses of Maize to Climate Change Scenarios in Southern Central Rift Valley of Ethiopia.” Scientific Reports 13: 1–18. https://​doi.​org/​10.​ 1038/​s4159​8-​023-​39795​-​y. Meinshausen, M., Z. R. J. Nicholls, J. Lewis, et al. 2020. “The Shared Socio-Economic Pathway (SSP) Greenhouse Gas Concentrations and Their Extensions to 2500.” Geoscientific Model Development 13: 3571– 3605. https://​doi.​org/​10.​5194/​gmd-​13-​3571-​2020. Müller, C., J. Jägermeyr, J. A. Franke, et al. 2024. “Substantial Differences in Crop Yield Sensitivities Between Models Call for Functionality – Based Model Evaluation.” Earth's Future 12: 1–21. https://​doi.​org/​10.​ 1029/​2023E​F003773. Nendel, C., M. Berg, K. C. Kersebaum, et  al. 2011. “The MONICA Model: Testing Predictability for Crop Growth, Soil Moisture and Nitrogen Dynamics.” Ecological Modelling 222: 1614–1625. https://​doi.​ org/​10.​1016/j.​ecolm​odel.​2011.​02.​018. Nezomba, H., F. Mtambanengwe, J. Rurinda, and P. Mapfumo. 2018. “Integrated Soil Fertility Management Sequences for Reducing Climate Risk in Smallholder Crop Production Systems in Southern 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.1016/j.ecolmodel.2013.02.016 https://doi.org/10.1016/j.agee.2009.02.003 https://doi.org/10.2134/agronj2002.4620 https://doi.org/10.2134/agronj2002.4620 https://doi.org/10.1016/j.gfs.2020.100431 https://doi.org/10.1016/j.gfs.2020.100431 https://doi.org/10.1016/j.agsy.2010.07.002 https://doi.org/10.1371/journal.pone.0239552 https://doi.org/10.1371/journal.pone.0239552 https://doi.org/10.1016/j.gloenvcha.2008.12.003 https://doi.org/10.1016/j.gloenvcha.2008.12.003 https://doi.org/10.1016/j.envsoft.2014.07.009 https://doi.org/10.1016/j.envsoft.2014.07.009 https://doi.org/10.1016/j.ecolmodel.2005.07.010 https://doi.org/10.1038/s43016-021-00400-y https://doi.org/10.1038/s43016-021-00400-y https://doi.org/10.1088/1748-9326/ac4ec1 https://doi.org/10.1088/1748-9326/ac4ec1 https://doi.org/10.1186/s40066-020-00283-5 https://doi.org/10.1186/s40066-020-00283-5 https://doi.org/10.1016/j.geoderma.2020.114342 https://doi.org/10.1016/j.geoderma.2020.114342 https://doi.org/10.1016/j.agrformet.2023.109819 https://doi.org/10.1016/j.agrformet.2023.109819 https://doi.org/10.1016/j.fcr.2022.108788 https://doi.org/10.1038/s43017-020-0080-8 https://doi.org/10.1038/s43017-020-0080-8 https://doi.org/10.1038/s41558-020-0830-0 https://doi.org/10.1016/j.gfs.2014.05.002 https://doi.org/10.1016/j.gfs.2014.05.002 https://doi.org/10.1038/s41558-023-01873-4 https://doi.org/10.1038/s41558-023-01873-4 https://doi.org/10.1038/s41598-023-39795-y https://doi.org/10.1038/s41598-023-39795-y https://doi.org/10.5194/gmd-13-3571-2020 https://doi.org/10.1029/2023EF003773 https://doi.org/10.1029/2023EF003773 https://doi.org/10.1016/j.ecolmodel.2011.02.018 https://doi.org/10.1016/j.ecolmodel.2011.02.018 14 of 15 Global Change Biology, 2026 Africa.” Field Crops Research 224: 102–114. https://​doi.​org/​10.​1016/j.​ fcr.​2018.​05.​003. Nissan, A., U. Alcolombri, N. Peleg, et  al. 2023. “Global Warming Accelerates Soil Heterotrophic Respiration.” Nature Communications 14: 3452. https://​doi.​org/​10.​1038/​s41467-​023-​38981-​w. OECD. 2016. “Agriculture in Sub-Saharan Africa: Prospects and Challenges for the Next Decade.” In OECD-FAO Agricultural Outlook 2016–2025. OECD Publishing. Parkes, B., D. Defrance, B. Sultan, P. Ciais, and X. Wang. 2018. “Projected Changes in Crop Yield Mean and Variability Over West Africa in a World 1.5K Warmer Than the Pre-Industrial Era.” Earth System Dynamics 9: 119–134. https://​doi.​org/​10.​5194/​esd-​9-​119-​2018. Pichot, J.-P., F. Al Zahawe, and P. F. Chabalier. 1977. “Evolution d'un Sol Ferralitique de Côte D'ivoire Après la Mise en Culture. Effet des Apports D'engrais Azoté et de Compost.” In Soil Organic Matter Studies, 83–96. IAEA. R Core Team. 2023. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Ranasinghe, R., A. C. Ruane, R. Vautard, et al. 2021. “Climate Change Information for Regional Impact and for Risk Assessment.” In Climate Change 2021 – The Physical Science Basis: Working Group I Contribution to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, edited by V. P. Masson-Delmotte, A. Zhai, S. L. Pirani, C. Connors, S. Péan, and N. Berger, 1767–1926. Cambridge University Press. Ray, D. K., N. D. Mueller, P. C. West, and J. A. Foley. 2013. “Yield Trends Are Insufficient to Double Global Crop Production by 2050.” PLoS One 8: e66428. https://​doi.​org/​10.​1371/​journ​al.​pone.​0066428. Raza, A., H. Ahrends, M. Habib-Ur-rahman, and T. Gaiser. 2021. “Modeling Approaches to Assess Soil Erosion by Water at the Field Scale With Special Emphasis on Heterogeneity of Soils and Crops.” Land 10: 422. https://​doi.​org/​10.​3390/​land1​0040422. Reed, C., W. Anderson, A. Kruczkiewicz, et al. 2022. “The Impact of Flooding on Food Security Across Africa.” Proceedings of the National Academy of Sciences of the United States of America 119: e2119399119. https://​doi.​org/​10.​1073/​pnas.​21193​99119​. Ren, S., T. Wang, B. Guenet, et  al. 2024. “Projected Soil Carbon Loss With Warming in Constrained Earth System Models.” Nature Communications 15: 1–10. https://​doi.​org/​10.​1038/​s4146​7-​023-​ 44433​-​2. Rezaei, E. E., H. Webber, S. Asseng, et  al. 2023. “Climate Change Impacts on Crop Yields.” Nature Reviews Earth and Environment 4: 831–846. https://​doi.​org/​10.​1038/​s4301​7-​023-​00491​-​0. Ricome, A., F. Affholder, F. Gérard, et  al. 2017. “Are Subsidies to Weather-Index Insurance the Best Use of Public Funds? A Bio-Economic Farm Model Applied to the Senegalese Groundnut Basin.” Agricultural Systems 156: 149–176. https://​doi.​org/​10.​1016/j.​agsy.​2017.​05.​015. Ritchie, J. T., U. Singh, D. C. Godwin, and W. T. Bowen. 1998. “Cereal Growth, Development and Yield.” In Understanding Options for Agricultural Production, edited by G. Y. Tsuji, G. Hoogenboom, and P. K. Thornton, 79–98. Springer. Robinson, S., D. Mason-D'croz, S. Islam, et al. 2015. “The International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT) Model Description for Version 3.” IFPRI Discuss Pap 01483. Rosenzweig, C., J. W. Jones, J. L. Hatfield, et al. 2013. “The Agricultural Model Intercomparison and Improvement Project (AgMIP): Protocols and Pilot Studies.” Agricultural and Forest Meteorology 170: 166–182. https://​doi.​org/​10.​1016/j.​agrfo​rmet.​2012.​09.​011. Ruane, A. C., S. McDermid, C. Rosenzweig, et  al. 2014. “Carbon– Temperature–Water Change Analysis for Peanut Production Under cli- mate Change: A Prototype for the AgMIP Coordinated Climate-Crop Modeling Project (C3MP).” Global Change Biology 20: 394–407. https://​ doi.​org/​10.​1111/​gcb.​12412​. Rusinamhodzi, L., M. Corbeels, S. Zingore, J. Nyamangara, and K. E. Giller. 2013. “Pushing the Envelope? Maize Production Intensification and the Role of Cattle Manure in Recovery of Degraded Soils in Smallholder Farming Areas of Zimbabwe.” Field Crops Research 147: 40–53. https://​doi.​org/​10.​1016/j.​fcr.​2013.​03.​014. Sanchez, P. A. 2002. “Soil Fertility and Hunger in Africa.” Science 295: 2019–2020. https://​doi.​org/​10.​1126/​scien​ce.​1065256. Savary, S., A. D. Nelson, A. Djurle, et al. 2018. “Concepts, Approaches, and Avenues for Modelling Crop Health and Crop Losses.” European Journal of Agronomy 100: 4–18. https://​doi.​org/​10.​1016/j.​eja.​2018.​04.​003. Savary, S., L. Willocquet, S. J. Pethybridge, P. Esker, N. McRoberts, and A. Nelson. 2019. “The Global Burden of Pathogens and Pests on Major Food Crops.” Nature Ecology & Evolution 3: 430–439. https://​doi.​org/​10.​ 1038/​s4155​9-​018-​0793-​y. Siatwiinda, S. M., I. Supit, B. van Hove, et al. 2021. “Climate Change Impacts on Rainfed Maize Yields in Zambia Under Conventional and Optimized Crop Management.” Climatic Change 167: 39. https://​doi.​ org/​10.​1007/​s1058​4-​021-​03191​-​0. Smith, W., B. Grant, Z. Qi, et  al. 2020a. “Towards an Improved Methodology for Modelling Climate Change Impacts on Cropping Systems in Cool Climates.” Sci Total Environ 728: 138845. https://​doi.​ org/​10.​1016/j.​scito​tenv.​2020.​138845. Smith, W., B. Grant, Z. Qi, et  al. 2020b. “Development of the DNDC Model to Improve Soil Hydrology and Incorporate Mechanistic Tile Drainage: A Comparative Analysis With RZWQM2.” Environmental Modelling & Software 123: 104577. https://​doi.​org/​10.​1016/j.​envso​ft.​ 2019.​104577. Sommer, R., B. K. Paul, J. Mukalama, and J. Kihara. 2018. “Reducing Losses but Failing to Sequester Carbon in Soils – The Case of Conservation Agriculture and Integrated Soil Fertility Management in the Humid Tropical Agro-Ecosystem of Western Kenya.” Agriculture, Ecosystems & Environment 254: 82–91. https://​doi.​org/​10.​1016/j.​agee.​ 2017.​11.​004. Sultan, B., and M. Gaetani. 2016. “Agriculture in West Africa in the Twenty-First Century: Climate Change and Impacts Scenarios, and Potential for Adaptation.” Frontiers in Plant Science 7: 1–20. https://​doi.​ org/​10.​3389/​fpls.​2016.​01262​. Tadross, M., P. Suarez, A. Lotsch, et al. 2009. “Growing-Season Rainfall and Scenarios of Future Change in Southeast Africa: Implications for Cultivating Maize.” Climate Research 40: 147–161. https://​doi.​org/​10.​ 3354/​cr00821. Taylor, C. M., D. Belusic, F. Guichard, et al. 2017. “Frequency of Extreme Sahelian Storms Tripled Since 1982 in Satellite Observations.” Nature 544: 475–478. https://​doi.​org/​10.​1038/​natur​e22069. Thrasher, B., W. Wang, A. Michaelis, et al. 2022. “NASA Global Daily Downscaled Projections, CMIP6.” Scientific Data 9: 1–7. Trancoso, R., J. Syktus, R. P. Allan, et al. 2024. “Significantly Wetter or Drier Future Conditions for One to Two Thirds of the World's Population.” Nature Communications 15: 1–11. https://​doi.​org/​10.​1038/​ s4146​7-​023-​44513​-​3. Van Ittersum, M. K., L. G. J. Van Bussel, J. Wolf, et al. 2016. “Can Sub- Saharan Africa Feed Itself?” Proceedings of the National Academy of Sciences of the United States of America 113: 14964–14969. https://​doi.​ org/​10.​1073/​pnas.​16103​59113​. Van Oost, K., T. A. Quine, G. Govers, et  al. 2007. “The Impact of Agricultural Soil Erosion on the Global Carbon Cycle.” Science 318: 626–629. https://​doi.​org/​10.​1126/​scien​ce.​1145724. Vanlauwe, B., A. Bationo, J. Chianu, et  al. 2010. “Integrated Soil Fertility Management: Operational Definition and Consequences for 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.1016/j.fcr.2018.05.003 https://doi.org/10.1016/j.fcr.2018.05.003 https://doi.org/10.1038/s41467-023-38981-w https://doi.org/10.5194/esd-9-119-2018 https://doi.org/10.1371/journal.pone.0066428 https://doi.org/10.3390/land10040422 https://doi.org/10.1073/pnas.2119399119 https://doi.org/10.1038/s41467-023-44433-2 https://doi.org/10.1038/s41467-023-44433-2 https://doi.org/10.1038/s43017-023-00491-0 https://doi.org/10.1016/j.agsy.2017.05.015 https://doi.org/10.1016/j.agrformet.2012.09.011 https://doi.org/10.1111/gcb.12412 https://doi.org/10.1111/gcb.12412 https://doi.org/10.1016/j.fcr.2013.03.014 https://doi.org/10.1126/science.1065256 https://doi.org/10.1016/j.eja.2018.04.003 https://doi.org/10.1038/s41559-018-0793-y https://doi.org/10.1038/s41559-018-0793-y https://doi.org/10.1007/s10584-021-03191-0 https://doi.org/10.1007/s10584-021-03191-0 https://doi.org/10.1016/j.scitotenv.2020.138845 https://doi.org/10.1016/j.scitotenv.2020.138845 https://doi.org/10.1016/j.envsoft.2019.104577 https://doi.org/10.1016/j.envsoft.2019.104577 https://doi.org/10.1016/j.agee.2017.11.004 https://doi.org/10.1016/j.agee.2017.11.004 https://doi.org/10.3389/fpls.2016.01262 https://doi.org/10.3389/fpls.2016.01262 https://doi.org/10.3354/cr00821 https://doi.org/10.3354/cr00821 https://doi.org/10.1038/nature22069 https://doi.org/10.1038/s41467-023-44513-3 https://doi.org/10.1038/s41467-023-44513-3 https://doi.org/10.1073/pnas.1610359113 https://doi.org/10.1073/pnas.1610359113 https://doi.org/10.1126/science.1145724 15 of 15Global Change Biology, 2026 Implementation and Dissemination.” Outlook on Agriculture 39: 17–24. https://​doi.​org/​10.​5367/​00000​00107​91169998. Vanlauwe, B., and K. E. Giller. 2006. “Popular Myths Around Soil Fertility Management in Sub-Saharan Africa.” Agriculture, Ecosystems & Environment 116: 34–46. https://​doi.​org/​10.​1016/j.​agee.​2006.​03.​016. Vanlauwe, B., J. Six, N. Sanginga, and A. A. Adesina. 2015. “Soil Fertility Decline at the Base of Rural Poverty in Sub-Saharan Africa.” Nature Plants 1: 15101. https://​doi.​org/​10.​1038/​nplan​ts.​2015.​101. Waha, K., C. Müller, and S. Rolinski. 2013. “Separate and Combined Effects of Temperature and Precipitation Change on Maize Yields in Sub-Saharan Africa for Mid- to Late-21st Century.” Global and Planetary Change 106: 1–12. https://​doi.​org/​10.​1016/j.​glopl​acha.​2013.​ 02.​009. Wang, B., J. Jägermeyr, G. J. O'Leary, et al. 2024. “Pathways to Identify and Reduce Uncertainties in Agricultural Climate Impact Assessments.” Nature Food 5: 550–556. https://​doi.​org/​10.​1038/​s4301​6-​024-​01014​-​w. Wang, M., S. Zhang, X. Guo, et  al. 2024. “Responses of Soil Organic Carbon to Climate Extremes Under Warming Across Global Biomes.” Nature Climate Change 14: 98–105. https://​doi.​org/​10.​1038/​s4155​8-​023-​ 01874​-​3. Webber, H., E. E. Rezaei, M. Ryo, and F. Ewert. 2022. “Framework to Guide Modeling Single and Multiple Abiotic Stresses in Arable Crops.” Agriculture, Ecosystems & Environment 340: 108179. https://​doi.​org/​10.​ 1016/j.​agee.​2022.​108179. Westra, S., H. J. Fowler, J. P. Evans, et al. 2014. “Future Changes to the Intensity and Frequency of Short-Duration Extreme Rainfall.” Reviews of Geophysics 52: 522–555. https://​doi.​org/​10.​1029/​88EO0​1108. Woomer, P. L., and F. N. Muchena. 1996. “Recognising and Overcoming Soil Constraints to Crop Production in Tropical Africa.” African Crop Science Journal 4: 503–518. Zingore, S., H. K. Murwira, R. J. Delve, and K. E. Giller. 2007. “Influence of Nutrient Management Strategies on Variability of Soil Fertility, Crop Yields and Nutrient Balances on Smallholder Farms in Zimbabwe.” Agriculture, Ecosystems & Environment 119: 112–126. https://​doi.​org/​ 10.​1016/j.​agee.​2006.​06.​019. Supporting Information Additional supporting information can be found online in the Supporting Information section. Figure S1: Summary of the methods used to calculate the impact of climate change only (A), soil fertility only (B) and the combined impact of climate change and soil fertility on maize yield (C). Figure S2: Boxplots of simulated maize yield change (%) by climate change only (CC only) and soil fertility declines only (SF only; i.e., difference between yearly reinitialization of soil conditions vs. continuous model simulation) compared to no climate change and no soil fertility decline scenario. Figure S3: Simulated soil organic carbon (SOC) change (%) compared to initial SOC under climate change (CC) or not (no CC). Figure S4: Boxplots of simulated maize yield change (%) due to the interaction between climate change (CC) and soil fertility (SF) compared to no climate change and no soil fertility decline scenario (no CC and no SF decline). Figure S5: Boxplots of simulated maize yield change (%) across years due to climate change for four sites with organic plus mineral fertilizers (160 kg N ha−1 season−1 and 4 t C ha−1 yr−1 organic amendment). Figure S6: Simulated soil organic carbon (SOC) change (%) with mineral and organic fertilizers (80 kg N ha−1 season−1 and 2 t C ha−1 yr−1 organic amendment) compared to initial SOC under climate change (CC) or not (no CC). Figure S7: Simulated maize yield trends (t ha−1) and SOC change (%) for the four sites with organic plus mineral fertilizers (80 kg N ha−1 season−1 and 2 t C organic amendment ha−1 yr−1, in red) or without fertilizer (in blue) under baseline climate. Data are from the median of 15 soil-crop models simulations. All simulations included long-term soil-crop feedback (continuous model simulation without yearly reinitialization of soil conditions). For yields, significant linear regressions at p < 0.05 are displayed. Table  S1: Management practices used in the simulation design for the four sites in Sub-Saharan Africa. DOY = day of the year. Table S2: Future climate scenarios from the Phase 6 of the Climate Model Intercomparison Project (CMIP6) for each site using the low (SSP1-2.6) and high (SSP3-7) emission scenar- ios from the 2040 to 2069 period compared to the 1980–2010 period. Results correspond to the median of 21 climate models. 13652486, 2026, 2, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/gcb.70720 by A ntoine C ouëdel - C IR A D , W iley O nline L ibrary on [03/02/2026]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://doi.org/10.5367/000000010791169998 https://doi.org/10.1016/j.agee.2006.03.016 https://doi.org/10.1038/nplants.2015.101 https://doi.org/10.1016/j.gloplacha.2013.02.009 https://doi.org/10.1016/j.gloplacha.2013.02.009 https://doi.org/10.1038/s43016-024-01014-w https://doi.org/10.1038/s41558-023-01874-3 https://doi.org/10.1038/s41558-023-01874-3 https://doi.org/10.1016/j.agee.2022.108179 https://doi.org/10.1016/j.agee.2022.108179 https://doi.org/10.1029/88EO01108 https://doi.org/10.1016/j.agee.2006.06.019 https://doi.org/10.1016/j.agee.2006.06.019 Beyond Climate Change: The Role of Integrated Soil Fertility Management for Sustaining Future Maize Yield in Sub-Saharan Africa ABSTRACT 1   |   Introduction 2   |   Materials and Methods 2.1   |   Multi-Model Ensemble and Model Calibration Methods 2.2   |   Experiments and Data 2.3   |   Model Simulation Scenarios 2.4   |   Data Analysis 3   |   Results 3.1   |   The Calibrated Soil-Crop Model Ensemble Reproduced Trends in Maize Yield and SOC 3.2   |   Soil Fertility Decline Has a Larger Impact on Maize Yield Than Climate Change Factors 3.3   |   Maize Yield Responses to ISFM Increase Over Time 4   |   Discussion 4.1   |   Soil Fertility Decline Outweighs Climate Change Effects on Maize Yield 4.2   |   ISFM as a No-Regret Climate Change Adaptation Option 4.3   |   Limitations of the Study and Future Research Directions 5   |   Conclusion Author Contributions Acknowledgements Funding Conflicts of Interest Data Availability Statement References