Assessing phenotypic diversity and sensor-based metrics for drought response in soybean

Abstract

Drought stress can be a significant yield-limiting factor in soybean (Glycine max [L.] Merr) production, necessitating improved drought-tolerant genetics to protect yield and promote resiliency under stress. A diverse panel of 450 maturity group 0–III soybean plant introgression accessions and checks was screened in a non-irrigated drought nursery for 3 years in 2020–2022 in Muscatine, IA. Visual- and sensor-based measurements were collected via red green blue, hyperspectral, and multispectral sensors via ground- and unmanned-aerial-vehicle-based platforms. A wide range of phenotypic diversity was observed and slow canopy wilting accessions were identified within three physiological stages for introduction into upper US Midwest breeding programs. Selections concentrating on canopy area production under drought held some overlap with slow wilting selections and also provide additional resources for breeders. Vegetation indices were able to mimic breeder selections for wilting, with red green blue vegetation index achieving up to 87.5% similarity in two years. Ratio analysis of reflectance spectra carotenoids also demonstrated potential in selecting accessions comparable to breeder choices. Accessions with consistently low wilting scores across multiple years provide valuable resources for drought-tolerance breeding.