Skip to Main content Skip to Navigation
Conference papers

Using a crop model to evaluate and design combinations of genotypes X management X environments that improve sunflower crop performance

Abstract : Crop modelling might help at the different steps of genotype design and assessment, by completing and widening the factorial experiments. A model of the sunflower crop (SUNFLO) where complex traits (yield and oil content) are sensitive to main abiotic stresses and to genotype-specific traits has been developed in France. The model parameters representing the genetic variability were measured as phenotypic traits. The aim of this study is to exploit the integrative properties of crop modelling to identify genotype x environment x management (GEM) combinations that are interesting for crop yield performance or stability.The space defined by all the possible GEM combinations can be referred as the fitness landscape (Hammer and Jordan, 2007). This landscape is partly sampled during the breeding steps, when multienvironment trials (MET) and various models of variance analysis are used mainly to identify the most performing genotypes and, additionally, to check that GxE interactions are not too excessive. This study focus on the inclusion of management actions and target environments in the otherwise genotype-centered view to crop improvement. A virtual factorial plan was built from seven genotypic traits (phenology, architectural and response traits), two management actions (sowing date and nitrogen fertilization) and an environmental variability representative of the soils and cultivation area. These GEM combinations were simulated and variance analysis methods were used to identify the pertinent combinations. Simulation results were analyzed from two viewpoints: (i) by identifying favourable links between phenotypic trait combinations ("virtual genotypes"), target environments and management actions and (ii) by assessing the climatic variability impact on these favorable combinations. Some traits impact on performance were trivial (precocity), others were foreseeable but has been quantified (plant leaf area). Some traits whom impact is difficult to assess in field trials (shape of leaf profile, stomatal regulation) showed positive properties on drought-prone conditions. It was also shown that annual variations in climate could greatly affect the apparent benefit of a yield-increasing trait. The same way that a conciliation of genetics and crop physiology is necessary to bridge the genotype-phenotype gap, the inclusion of computer science seems essential to navigate into this additional layer of complexity created by modeling approaches.
Complete list of metadata

https://hal.inrae.fr/hal-02747666
Contributor : Migration Prodinra Connect in order to contact the contributor
Submitted on : Wednesday, June 3, 2020 - 12:11:52 PM
Last modification on : Thursday, October 28, 2021 - 10:40:03 AM

Identifiers

  • HAL Id : hal-02747666, version 1
  • PRODINRA : 270754

Collections

Citation

Pierre Casadebaig, Philippe Debaeke. Using a crop model to evaluate and design combinations of genotypes X management X environments that improve sunflower crop performance. 18. International Sunflower Conference, Feb 2012, Mar del Plata, Argentina. ⟨hal-02747666⟩

Share

Metrics

Record views

35