Yield and climatic parameters in a common bean (Phaseolus vulgaris L.) regional trial

Understanding patterns of crop response to environmental factors is critical to simulating target environments in selection environments. In early stages of varietal development, breeding lines are usually developed in centralized facilities. Subsequently, testing continues on a broader array of research stations before dissemination to farmers, fields. Such a system is practiced in common bean (Phaseolus vulgaris L.) breeding in East Africa. A regional yield trial called the East Central Africa bean yield trial was distributed in three separate groups that aligned with three breeding pipelines: Andean bush beans (Group 1 or G1), Mesoamerican bush beans (G2), and climbing beans (G3). Forty-three trials met minimal standards of data quality. Trial environments composed of a site, planting date, and its climatic parameters were clustered based on 13 variables of temperature, rainfall, relative humidity, vapor pressure deficit, and altitude. Climatic data were derived from the NASA Prediction of Worldwide Energy Resources (POWER) database, which estimates day-by-day weather for each site. Four climate clusters emerged from this analysis. Climbing bean yield corresponded to climate clusters and heavily responded to high altitude and temperature. Most Andean bush bean environments occupied a centric cluster with few extreme variables. No patterns were observed in the Mesoamerican bush beans. This method should be extended to a larger set of trials and should be used to compare research station environments with farmer production environments.