Evaluating and improving crop growth models for simulating genotype-by-environment interactions in sugarcane

Abstract

In his thesis, entitled “Evaluating and improving crop growth models for simulating genotype-by-environment interactions in sugarcane”, Matthew Jones has enhanced our capacity to assist sugarcane breeding using crop growth simulation models. Matthew presents an analysis of genotype, environment and genotype-by-environment (GxE) effects in an international sugarcane multi-environment trial – the first study of its kind in sugarcane. As part of this analysis, a novel approach is used to assess the adequacy of established simulation concepts to account for genotypic control of plant process responses to environmental factors. This work is then expanded comprehensively to assess three sugarcane crop growth models for their abilities to simulate genotype performance in different environments. An important finding was that the duration of the germination phase strongly influenced subsequent canopy development and biomass growth. The thesis further describes the development of a new crop model, CaneGEM, to address weaknesses in existing models. Canopy development, biomass growth and biomass partitioning are simulated using a source-sink approach, enabling dynamic interaction between these processes – a necessity for realistic simulation of GxE interaction effects. CaneGEM showed improved capability for predicting GxE interaction effects at plant process level. A demonstration of the CaneGEM model revealed the potential to improve biomass yields via genotypic adaptations to cooler temperatures. Additionally, this study showed both the importance of the duration of germination phase in driving GxE interaction effects in canopy development and biomass yields, and some of the challenges involved in predicting this accurately.