A novel approach to accelerate ideotyping using model-aided envirotyping

Abstract

CONTEXT

Climate change threatens wheat production by intensifying drought, heat stress, and yield instability. Selecting optimal cultivars is crucial for mitigating climate change impacts. Crop model-assisted ideotyping, i.e., designing and/or selecting for traits that maximise yield or quality under defined conditions, requires exploring a large number of genotype-by-environment (GxE) interactions but is computationally demanding. This is where envirotyping, i.e., categorising environments into a few environment types (ETs), emerges as a promising solution. Integrating envirotyping with ideotyping enhances breeding efficiency and enables targeted trait optimisation. This scalable, data-driven approach supports the development of climate-resilient wheat cultivars suited to diverse and changing environments.

OBJECTIVE

Show how an innovative approach leveraging envirotyping can significantly cut down the computational demands of ideotyping, while still maintaining yield improvements. This approach offers a scalable framework for developing resilient crop cultivars tailored to diverse and changing environments.

METHODS

Using the next generation of Agricultural Production Systems sIMulator (APSIM Next Generation), wheat growth and development was simulated across diverse Australian environments. Four commercial cultivars were simulated under multiple sowing dates to determine optimal sowing windows and highest-yielding cultivars for each location. Cluster analysis of water supply/demand ratios identified six ETs with distinct seasonal drought patterns. A genetic algorithm was used to optimise 14 key cultivar parameters influencing phenology, morphology, resource use, and yield components. Three ideotyping strategies—global, targeted at high-stress ETs, and location-specific—were assessed for their impact on average yield and yield stability.

RESULTS AND CONCLUSIONS

The ideotyping strategies effectively reduced the occurrence frequency of late-season water stress. The identified ideotypes significantly improved average yield (∼18 %) and yield stability (up to 16 % reduction in coefficient of variation). Global and targeted ideotyping strategies outperformed location-specific approaches in enhancing broad adaptability. In these strategies, key traits influencing yield gains included low minimum leaf number, high grain potential size, high radiation use efficiency, low potential root water uptake rate, high stay-green, and high number of grains per gram of stem and spike biomass. Phenological traits and trait interactions were more influential in the location-specific strategy.

SIGNIFICANCE

This study demonstrates the potential of model-assisted envirotyping to improve wheat breeding efficiency by reducing computational demands while maximising average yield and yield stability. Incorporating envirotyping into breeding workflows provides a scalable, data-driven approach that complements traditional GxE testing. Our findings offer valuable insights for developing climate-resilient wheat cultivars and contribute to global food security in the face of increasing climatic variability.