Optimizing soybean variety selection for the Pan-African Trial network using factor analytic models and envirotyping.

Area of Science:

• Agricultural Science
• Plant Breeding
• Climate Change Adaptation

Background:

• Climate change poses a significant threat to global soybean production, particularly impacting grain yield in Sub-Saharan Africa.
• Developing climate-resilient soybean varieties adapted to diverse environments is crucial for food security and agricultural sustainability in the region.

Purpose of the Study:

• To identify soybean varieties with superior performance and stability across multiple environments in South-Eastern Africa.
• To determine the key environmental factors influencing soybean grain yield and understand genotype-by-environment interactions.
• To delineate mega-environments for targeted soybean breeding and variety selection.

Main Methods:

• Evaluated 169 soybean varieties across 83 environments in Malawi and Zambia using randomized complete block designs.
• Utilized factor analytic (FA) models to estimate genotype adaptation and environmental kernel/XGBoost methods to identify mega-environments.
• Incorporated 37 environmental features from NASA POWER and SoilGrids for envirotyping.

Main Results:

• A four-factor FA model best explained genotype-by-environment interactions, with 59.6% being crossover interactions.
• Identified three distinct mega-environments, with growing degree days, mean temperature, and photosynthetically active radiation use efficiency strongly associated with yield.
• Varieties V025, V035, and V158 showed high yield potential and reliability, though with moderate stability.

Conclusions:

• Soybean yield is significantly influenced by specific environmental factors, necessitating tailored breeding approaches for different mega-environments.
• Integrating machine learning with crop modeling is essential for accurately assessing environmental influences and enhancing variety adaptation strategies.
• Targeted selection of soybean varieties based on identified mega-environments and environmental drivers can improve crop performance in climate-vulnerable regions.