Uncertainties In Maize Crop Model Responses to Climate Factors.

Baron Christian, Basso Bruno, Biernath Christian, Boogaard Hendrik, Conijn Sjaak, Deryng Delphine, De Sanctis Giacomo, Gayler Sebastian, Grassini Patricio, Hoek Steven, Izaurralde Cesar, Jongschaap Raymond, Kemanian Armen, Kersebaum Kurt Christian, Müller Christoph, Nendel Claas, Priesack Eckart, Sau Federico, Shcherbak Iurii, Tao Fulu, Teixeira Edmar, Timlin Dennis, Waha Katharina, Hatfield Jerry, Corbeels Marc.

Simulation models can be used as strategic tools for evaluating the consequences of climate change on crop production and for evaluating possible adaptations of crop management practices. As many different crop models are available, it is important to compare several models in order to improve model functions and to assess uncertainties in the predicted crop response to climate change factors. Furthermore, the use of several models, already tested for maize, has a higher potential to capture the complexity of the multiple impacts of climate change on crops than using one single model. The Agricultural Model Intercomparison and Improvement Project (AgMIP) is a distributed simulation exercise for agricultural model intercomparison and future climate change assessments that has participation of multiple crop and economic modeling groups around the world. The goal of AgMIP is:

-                           To provide reasonable estimates of the impacts of climate change on crop yields of important crops of the world.

-                           To evaluate the uncertainty that comes from using different models in various climate and crop management situations and understand the causes of the different model responses to climate change factors.

-                           To improve our capacity for simulating the response of crops to yet unknown climates (combinations of extreme drought, temperatures and atmospheric CO₂ concentrations).

Our work conducted in the AgMIP project for the maize crop first consisted of testing model output sensitivity to climatic factors over a large ensemble of maize crop models and across four sentinel sites that are important contrasting pedoclimatic zones of maize production.

Multiple maize modeling groups are involved in the work and run their models for the four sites of Lusignan (France), Ames (Iowa), Rio Verde (Brazil) and Morogoro (Tanzania), using measured daily weather records from 1980 to 2010 from each location. Modelers simulated the crop performances for baseline (1980 to 2010) and one single A2-Global Climate Model generated End-of-Century scenario (including 734ppm CO₂) for each location after calibrating their models on the basis of 1-year experimental data at each location. In addition, models were run with the 30-year manipulated weather files in order to compare their sensitivity to temperature (-3, 0, +3, +6, +9 °C), CO₂ (360, 450, 540, 630, 720ppm), and rainfall (-30% rainfall) for each sentinel site.

The simulated maize crop yield simulated a limited response to CO₂ as expected, a clear trend in yield decrease with elevated temperatures (-3 up to +9 °C). Water appears to be a main source of variation in yields, but simulations indicate a possible strong interaction with nitrogen. With those general trends, results also showed that range of the simulated climate change effects on maize yield levels and interannual variations were highly dependent on the model and on the site.