Crop model improvement reduces the uncertainty of the response to temperature of multi-model ensembles

Authors: MAIORANO, ANDREA - Institut National De La Recherche Agronomique (INRA); MARTRE, PIERRE - Institut National De La Recherche Agronomique (INRA); ASSENG, SENTHOLD - University Of Florida; EWART, FRANK - University Of Bonn; MULLER, CHRISTOPH - Potsdam Institute; ROTTER, REIMUND - Natural Resources Institute Finland (LUKE); RUANE, ALEX - National Aeronautics And Space Administration (NASA); SEMENOV, MIKHAIL - Rothamsted Research; WALLACH, DANIEL - Institut National De La Recherche Agronomique (INRA); WANG, ENLI - Commonwealth Scientific And Industrial Research Organisation (CSIRO); ALDEMAN, PHILLIP - International Maize & Wheat Improvement Center (CIMMYT); KASSIE, BELAY - University Of Florida; BIERNATH, CHRISTIAN - Helmholtz Centre; BASSO, BRUNO - Michigan State University; CAMMARANO, DAVIDE - Institut National De La Recherche Agronomique (INRA); CHALLINOR, ANDREW - University Of Leeds; DOLTRA, JORDI - Center For Agricultural Research And Training, Cantabria Government (CIFA); DUMONT, BENJAMIN - Michigan State University; GAYLER, SEBASTIAN - Water & Earth System Science (WESS); KERSEBAUM, CHRISTIAN - Institute Of Landscape Systems Analysis, Leibniz Centre For Agricultural Landscape Research; Kimball, Bruce; KOEHLER, ANN-KRISTIN - University Of Leeds; LIU, BING - Nanjing Agricultural University; O'LEARY, GARRY - Department Of Environment And Primary Industries; OLESEN, JORGEN - Aarhus University; OTTMAN, MICHAEL - University Of Arizona; PRIESACK, ECKART - Helmholtz Centre; REYNOLDS, MATTHEW - International Maize & Wheat Improvement Center (CIMMYT); REZAEI, EHSAN - Institut National De La Recherche Agronomique (INRA); STRATONOVITCH, PIERRE - Rothamsted Research; STRECK, THILO - Stuttgart University; THORNBURN, PETER - Commonwealth Scientific And Industrial Research Organisation (CSIRO); WAHA, KATHARINA - Potsdam Institute; Wall, Gerard - Gary; White, Jeffrey; ZHAO, ZHIGAN - Commonwealth Scientific And Industrial Research Organisation (CSIRO); ZHU, YAN - Nanjing Agricultural University

Submitted to: Field Crops Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/4/2016
Publication Date: 1/30/2017
Publication URL: http://handle.nal.usda.gov/10113/5263060
Citation: Maiorano, A., Martre, P., Asseng, S., Ewart, F., Muller, C., Rotter, R.P., Ruane, A., Semenov, M.A., Wallach, D., Wang, E., Aldeman, P.D., Kassie, B.T., Biernath, C., Basso, B., Cammarano, D., Challinor, A.J., Doltra, J., Dumont, B., Gayler, S., Kersebaum, C.K., Kimball, B.A., Koehler, A., Liu, B., O'Leary, G.J., Olesen, J.E., Ottman, M., Priesack, E., Reynolds, M.P., Rezaei, E.E., Stratonovitch, P., Streck, T., Thornburn, P.J., Waha, K., Wall, G.W., White, J.W., Zhao, Z., Zhu, Y. 2017. Crop model improvement reduces the uncertainty of the response to temperature of multi-model ensembles. Field Crops Research. 202:5-20.

Interpretive Summary: Due to expected global warming, crop growth models must be improved to account for the future higher temperature effects on crop yields. To date, there are a huge variety of mathematical functions in use to simulate the response of various plant physiological processes to temperature. In this study, the formulations from 15 wheat growth models were modified to better fit data from wheat grown by ARS researchers in Maricopa, Arizona and a University of Arizona collaborator, who grew 16 crops with a variety of planting dates and infrared warming to produce a dataset covering a very wide range of temperatures. The "improved" models were tested against an independent dataset from several locations around the world. The simulations from the original models in this study varied widely, especially for higher temperatures. However, the authors were able to show that the improved models were indeed improved, both in accuracy and precision, which holds promise for greatly improving the accuracy of the projections of future wheat productivity. This research will benefit all consumers of food and fiber.

Technical Abstract: To improve climate change impact estimates, multi-model ensembles (MMEs) have been suggested. MMEs enable quantifying model uncertainty, and their medians are more accurate than that of any single model when compared with observations. However, multi-model ensembles are costly to execute, so model improvements have been suggested as a better approach for reduceing the uncertainty of climate change impact assessments. Herein, 15 wheat growth models of a larger MME were improved through re-parameterization and/or incorporating or modifying heat stress effects on phenology, leaf growth and senescence, biomass growth, and grain number and size using detailed field experimental data from the USDA Hot Serial Cereal (HSC) experiment (calibration dataset). Simulation results from before and after model improvement were then evaluated with independent field experiments from a CIMMYT world-wide field trial network (evaluation dataset). Model improvements decreased the variation (10th to 90th model ensemble percentile range) of grain yields simulated by the MME on average by 39% in the calibration dataset and by 26% in the independent evaluation dataset for crops grown in mean seasonal temperatures >24°C. MME mean squared error in simulating grain yield decreased by 37%. MME prediction skills increased by 47% with model improvement due to a reduction in MME uncertainty range by 27%. As a result, the number of required models for MMEs for impact assessments was halved. Improving crop models is therefore important to improve model-based impact assessments and allow more practical smaller MMEs to be used effectively.