Modelling climate change impacts on maize yield under low N input conditions. Crop modelling for Agriculture and Food Security under Global Change
GN Falconnier, M Corbeels , KJ Boote , F Affholder , M Adam, DS MacCarthy, Alexander C Ruane, C Nendel, AM Whitbread, E Justes , LR Ahuja, F.M. Akinseye , IN Alou, KA Amouzou , S.S. Anapalli , C Baron, B Basso, F Baudron , P Bertuzzi, AJ Challinor , Y Chen, D Deryng , ML Elsayed , B Faye, T Gaiser , M Galdos , S Gayler , E Gerardeaux , M Giner , B Grant, G Hoogenboom , ES Ibrahim, B Kamali , KC Kersebaum , SH Kim, M van der Laan , L Leroux , JI Lizaso , B Maestrini , EA Meier, F Mequanint , A Ndoli , CH Porter, E Priesack , D Ripoche , T Sida , U Singh, W Smith, A Srivastava, S Sumit , F Tao, PJ Thorburn , D Timlin , B Traore , T Twine, H Webber. Modelling climate change impacts on maize yield under low N input conditions in sub Saharan Africa. Crop modelling for Agriculture and Food Security under Global Change. ICROPM 2020, Montpellier, France, Feb 3-5, 2020.
Smallholder farmers in sub-Saharan Africa (SSA) currently grow rainfed maize with limited inputs such as fertilizer. Maize is crucial for food security in SSA, and climate change may exacerbate current production constraints. Crop models can help quantify the potential impact of climate change on maize yields in these low-input systems. We evaluated the impact of changing [CO2], temperature and rainfall conditions on maize yield, for different nitrogen (N) inputs (0, 80, 160 kg N ha-1) in cool sub-humid Ethiopia, cool semi-arid Rwanda, hot sub-humid Ghana and hot semi-arid Mali and Benin. For that, we performed a model intercomparison including 25 maize models, evaluated their ability to predict observed yields and tested their responses to climate change. Models were calibrated with measured grain yield, plant biomass and plant N, Leaf Area Index, harvest index and in-season soil water content from two-year experiments in Benin, Mali, Ghana, Rwanda and Ethiopia. Calibrated models reproduced grain yield variations across sites and experiments well with average rRMSE of 26%. We show that without N fertilizer input (i.e., 0 kg ha-1), maize (i) benefits less from an increase in atmospheric [CO2], (ii) is less affected by higher temperature (+4°C) or decreased rainfall (-50% of current), and (iii) is more affected by increased rainfall (+150% of current) compared with N fertilizer inputs (80, 160 kg N ha-1). This model inter-comparison analysis reveals that simulation of daily N supply and N leaching plays a crucial role in simulating climate change impacts in low-input systems. More detailed crop field experiments are needed to better understand how the interactions between N input and climate change operate in the real world. A robust characterization of these interactions are critical for supporting the design of successful adaptation practices to climate change across SSA, as farmers who intensify maize production with increased use of mineral fertilizer are likely to be affected differently by climate change.