The Texas A&M AgriLife Research Wheat Improvement Program for the Texas High Plains develops high yielding, drought tolerant, insect and disease resistant winter wheat (Triticum aestivum) varieties with high end use quality for rain-fed and irrigated production systems. Unoccupied aerial system (UAS) technology is revolutionizing precision agriculture, offering a cost-effective method for collecting high-spatiotemporal data in plant breeding nurseries.

Previous ground-based spectral reflectance research quantified genotypic variation in biomass for high-throughput phenotyping, which progressed to UAS-based visual and multispectral imagery data collection throughout our eight month long growing season. UAS-based phenomic data from these contrasting environments provide valuable insights on biomass production and grain yield that strengthens the breeding selection process. UAS-based red, green blue (RGB) and multispectral phenomic data was collected on 15,450 breeding plots in both rain-fed and irrigated nurseries during the 2018-2023 growing seasons, totaling 358,750 observations.

Geospatial data products are processed from raw UAS images by the Texas A&M Corpus Christi Digital Agriculture Hub. Spectral reflectance indices (SRIs) including Normalized Difference Vegetation Index (NDVI), Normalized Difference Red-Edge Index (NDRE), and Excess Green Index (ExG) are calculated and physical traits such as canopy cover and canopy height extracted.

Traditional breeding selection criteria of grain yield, test weight, heading date, plant height, forage biomass, early season vigor, and breeder’s ag score were compared to UAS-based predictions of grain yield.

Correlation between SRIs and grain yield varied by year based on timing and amount of precipitation received during the growing season. In general, irrigated grain yield correlated well with NDVI during boot stage through grainfill. Canopy volume predicted forage biomass and early season vigor, which are important traits for dual-purpose wheat in the Southern High Plains.