Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

UAV remote sensing-based phenotyping to evaluate drought stress in turfgrass

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2021 ASABE Annual International Virtual Meeting  2100654.(doi:10.13031/aim.202100654)
Authors:   Tianyi Wang, Ambika Chandra, Meghyn Meeks, Dennis Genovesi
Keywords:   Drought stress, Machine learning, Remote sensing, Turfgrass, UAV

Abstract. Breeding turfgrasses using classical plant breeding methods is a long-term process which requires a breeder to sort through large segregating populations by phenotypically evaluating the plants across multiple environments and over several years. The quality and frequency of phenotypic data collection at a field-scale is currently the bottleneck limiting the efficiency and accuracy of classical phenotype-based breeding. The use of unmanned aerial vehicle (UAV) remote sensing has proven to be viable method to collect geospatial data rapidly and at fine spatial and high temporal scales in major agricultural crops. In this study, UAV remote sensing and machine learning algorithms were utlilzed to evaluate drought stress on zoysiagrass (Zoysia spp.) breeding nursery planted in 2017. Selections of the top 2% best-performing hybrids under drought stress were made using visual parameters (turfgrass quality under normal and drydown conditions) as well as UAV-derived NDVI (normalized difference vegetation index). The results suggest an agreement between the conventional selection methods and proposed UAV phenotyping approaches.

(Download PDF)    (Export to EndNotes)