Oats have become a major forage crop for livestock during fall and winter months in the southern US when other high-protein feeds are scarce, due to high quality, palatability, early grazing and…

The Department of Agriculture National Institute of Food and Agriculture awarded the University of Florida $760,000 on April 1, 2026, under the Agriculture and Food Research Initiative (AFRI, CFDA 10.310) to develop high-throughput phenotyping tools and genomic selection methods for improving biomass yield in winter oats grown for livestock forage across the southern United States.

The research characterizes 480-line training and 200-line validation oat panels across four states (Florida, Louisiana, South Carolina, and Texas) over three years through March 31, 2029. The project collects UAV-based multispectral sensor data, molecular markers, and biomass measurements to develop multivariate and machine-learning prediction models that integrate genomic data with unmanned-aircraft phenotyping to forecast genotypic performance under multiple environmental conditions. Two workshops train plant-breeding graduate students within the SunGrains program on accessible state-of-the-art technology combining UAV-based high-throughput phenotyping and genomic selection. The work produces breeding tools and easy-to-use protocols to accelerate development of high-biomass-yielding oat varieties.

Place of performance is Gainesville, Florida, with field trials distributed across the recipient's sub-recipient locations. Texas A&M AgriLife Research (sub-award dated May 3, 2026) collects forage data at College Station, Texas, and participates in model development and breeder training. Louisiana State University Agricultural Center (sub-award dated May 1, 2026) grows forage and grain trials at two Louisiana locations. Clemson University (sub-award dated April 14, 2026) generates field data, maintains oat plots, collects multispectral imagery through seasonal UAV flights, and coordinates stakeholder outreach including field days and webinars.

Oats have become a major forage crop for livestock during fall and winter months in the southern US when other high-protein feeds are scarce, due to high quality, palatability, early grazing and regrowth capacity, and silage production potential. Systematic genetic improvement of biomass yield in winter OAT is limited, as breeding programs in the southern US have predominantly focused on grain yield improvement and a lack of breeding tools to enable efficient and accurate selection for biomass. Increased biomass yield is a crucial goal that must be accomplished to meet the increasing demand for quality livestock feed, but the lack of in-season plot biomass data is a major limiting factor to genetic gain. High-throughput and precise in-field phenotyping techniques are needed to maximize genetic gain for biomass yield. The proposed research aims to combine state-of-the-art high-throughput phenotyping tools with genomic selection to predict genotypic performance for biomass yield, to train future plant breeders, and to implement biomass selection protocols based on results. This will be accomplished by characterizing a 480-line training and a 200-line validation OAT panels in four states over three years in the southern US. UAV-based multispectral sensor, molecular markers, and biomass data will be collected and used to develop different multivariate and machine-learning prediction models. Two workshops will be organized to train plant breeding graduate students from sungrains universities at the UAS-wheatcap hub. Results will provide clarity and direction for breeders, and easy-to-use tools to breed high biomass-yielding varieties.