Effect of Management Factors and Climate Change on Mixed Crop-Livestock Systems in The Sudan Savanna of Burkina Faso and Ghana

Abstract

Diversified farming systems integrating crop and livestock operations are vulnerable to climate change and need adaptation strategies to improve the system’s productivity. The main objective of this study was to assess the effect of climate change on the mixed crop-livestock systems (MCLS) productivity in the Sudan Savannah of West Africa. A systematic review and meta-analysis were first conducted to determine the contribution of MCLS to soil greenhouse gas (GHG) emissions. The effect of fertilizer application practices on grain and vegetable yields of MCLS in Ghana and Burkina Faso was also quantify using detailed information on fertilizer management practices and yield collected from 317 MCLS farms distributed across three (3) districts/provinces in the Upper East region of Ghana (Kassena Nankana, Builsa North, and Bongo) and over the Plateau central of Burkina Faso (Loumbila, Ziniaré, and Zitenga). Data was summarized, based on their grain and vegetable yields under (1) major fertilizer sources, (2) N fertilizer rate (kg ha−1), (3) fertilizer application timing, and (4) fertilizer placement methods. The SIMPLACE modelling framework was also used to simulate the output of typical MCLS farms. Applying such a framework allowed us to estimate the effect of future climate change on crop (maize, millet, sorghum, and peanut) yield and livestock numbers in the selected locations in Ghana and Burkina Faso. The findings of the meta-analysis showed that moderating variables such as drainage class, stocking density, climate, N fertilizer amount, fertilizer type, and soil texture exerted greater control over the GHG emissions in MCLS. Similarly, results showed that fertilizer source and country-specific recommended N fertilizer rates significantly increased (P < 0.05) yields of most grain crops in both Burkina Faso and Ghana, and all vegetable crops in Ghana. On the other hand, the SIMPLACE model was calibrated and validated for grain yield prediction of MCLS. Bias quantity, mean residual error and root mean squared error was used to determine the agreement between observed and model-simulated grain yield. Model evaluation indicated that maize and sorghum yields were slightly and moderately overestimated, respectively. Conversely, grain yields of millet and peanut were moderately underestimated. Overall the mean residual error (MR) was between -12% to 18% which was judged satisfactorily. For livestock numbers, the model deviated from observed values by recording an MR of 27%. Furthermore, the validated model was used to estimate grain yields and livestock numbers for the past (2000-2024) and project their future changes using downscaled Global Circulation Model (GCM) projections (2025–2050). The results demonstrated that in the future, the study locations will likely experience a significant decline in maize (up to -34%), millet (up to -57%), sorghum (up to -53%), peanut (up to -23%), and a significant increase in livestock numbers (up to 181%).