Identification of abiotic stress environments in maize ( Zea mays L.) across Argentina

Maize ( Zea mays L.) production is challenged by climate change, particularly increasing drought stress conditions. Enhancing drought tolerance is essential to obtain yield stability across environments. This study aimed to generate an environmental classification from multi‐environment trials in Argentina and to compare classification methods for identifying abiotic stress environments relevant to maize breeding. The effectiveness of three environmental classification methods—visual, combined biplot and clustering, and self‐organizing map (SOM)—was evaluated to identify abiotic stress environments in maize multi‐environment trials (METs) conducted in Argentina. Fifteen hybrids were assessed across 221 environments, including early and late planting dates, over four growing seasons. Classifications were based on grain yield and weather variables and compared based on the variance explained by genotypic‐related effects. The SOM method explained the highest proportion of genotype + genotype × environment group (EG) variance and more effectively identified abiotic stress environments while minimizing confounded effects from disease or agronomic issues. Environments classified as affected by abiotic stress showed yield reductions of approximately 50% compared with non‐stress environments. The wide geographical distribution of stress‐affected environments highlighted the non‐predictable nature of repeatable genotype × environment interaction caused by abiotic stress. Two hybrids showed broad adaptation across all EGs, while others showed specific adaptation to stress or non‐stress conditions. These findings support SOM as a robust tool for environmental classification and hybrid characterization. Its application in early breeding stages may enhance the selection of broadly adapted germplasm under increasing climate variability.