Modelling Wheat Production

Authors: JAMIESON, P - NZ INST CRP & FOOD RES,NZ; ASSENG, S - CSIRO, WEMBLEY AUSTRALIA; CHAPMAN, S - CSIRO, QUEENSLAND,AUSTRAL; DRECCER, M - CSIRO; White, Jeffrey; McMaster, Gregory; PORTER, J - U OF COPENHAGEN, DENMARK; SEMENOV, M - BIOMATH & BIOINFOMAT, UK

Submitted to: World Wheat Volume 2
Publication Type: Book / Chapter
Publication Acceptance Date: 1/15/2008
Publication Date: 6/18/2010
Citation: Jamieson, P.D., Asseng, S., Chapman, S.C., Dreccer, M.F., White, J.W., Mcmaster, G.S., Porter, J.R., and Semenov, M.A. (2010). Modelling wheat production. In: World Wheat Book (van Ginkel M., Bonjean A. and Angus W., eds.). Lavoisier, Paris France.

Interpretive Summary: Bread-making quality of wheat depends in part on the amount of protein in the grain, usually described as grain protein content (GPC). In wheat farming, GPC can be highly variable, depending on weather and soil conditions, as well as the type of wheat grown. Wheat cultivars can differ in many traits that affect GPC both directly and indirectly. Early flowering often increases grain protein content (GPC) but decreases grain yield. Cultivar differences in flowering are largely determined by two physiological responses, called vernalization and photoperiodism, plus basic differences in tendency to flower, termed earliness per se. This research examines whether cultivar differences in these traits affect GPC, especially whether the three traits can partially explain differences in GPC across different wheat production regions. This information might help breeders select for optimal flowering characteristics, including consideration of impacts of climate risk or even global warming. Twenty four winter wheat and five spring wheat cultivars selected from a large trial and twelve winter wheats tested over two years in Germany were characterized using a computer model called CSM-Cropsim-CERES-Wheat model. The model uses parameters to specify vernalization requirement, photoperiod sensitivity, and earliness per se of individual cultivars. For the multi-nation dataset, about 7% of variation in GPC was related to cultivar, with another 7% attributable to interactions of cultivar with major production region, location and year. The three model parameters all influenced GPC, but their effects varied with region, site and year. Analyses using the data from Germany confirmed that GPC increased with earlier anthesis, which was influenced by P1D and P123. The results indicate that efforts to improve GPC could benefit by targeting the three traits to specific populations of environments.

Technical Abstract: In wheat, a shorter pre-anthesis phase is often associated with increased grain protein content (GPC) but decreased grain yield. Cultivar differences in pre-anthesis development are mainly determined by vernalization requirement, photoperiod sensitivity and earliness per se. This research examines whether cultivar differences in these traits affect GPC, especially whether the three traits can partially explain G x E interactions for GPC. Twenty four winter wheat and five spring wheat cultivars selected from International Winter Wheat Performance Nursery (IWWPN) trials and twelve winter wheats tested over two years in Germany were characterized using the CSM-Cropsim-CERES-Wheat model. The model parameter P1V specifies the cultivar vernalization requirement, P1D, the photoperiod response, and P123, earliness per se. Covariance analyses of the IWWPN dataset indicated that about 7% of variation in GPC was explained by cultivar, with another 7% attributable to interactions of cultivar with region, site and year. P1V, P1D and P123 all influenced GPC, but their effects varied with region, site and year. Path analyses using the data from Germany confirmed that GPC increased with earlier anthesis, which was influenced by P1D and P123. The results indicate that efforts to improve GPC could target the three traits to specific populations of environments, which should reduce the large influence of environment on GPC.