Unmanned Aerial System (UAS) Platforms for Cotton Breeding: Findings and Challenges

Thursday, January 5, 2017: 9:15 AM
Reunion A (Hyatt Regency Dallas)
Murilo Maeda , Texas A&M AgriLife Research
Juan Landivar , Texas A&M AgriLife Research
Josh McGinty , Texas A&M AgriLife Extension Service
Andrea Maeda , Texas A&M AgriLife Research
Jinha Jung , Texas A&M University - Corpus Christi
Anjin Chang , Texas A&M University - Corpus Christi
Junho Yeom , Texas A&M University - Corpus Christi
C. Wayne Smith , Texas A&M University
Steve Hague , Texas A&M University
David M. Stelly , Texas A&M AgriLife Research
Jane Dever , Texas A&M AgriLife Research
Juan Enciso , Texas A&M AgriLife Research
Unmanned Aerial System (UAS) and advanced computational technology are rapidly evolving. The development of methodologies to extract meaningful biological and physiological crop information from UAS imagery presents a unique opportunity for agriculture researchers and plant breeders alike.  The possibility to obtain accurate growth, development, health, and productivity estimates for every square meter of a field does not come without its challenges, however. In 2016 researchers from Texas A&M AgriLife Research & Extension and Texas A&M University at Corpus Christi teamed-up with Texas A&M AgriLife cotton breeders to conduct a genotype evaluation trial and further explore the potential of UAS technology for plant breeders. The field trial consisted of 31 genotypes which included 23 breeding lines and 8 cultivars. Plots were 2 rows by 10m in a paired plot design. UAS platforms equipped with natural color (RGB), multispectral, and thermal infrared sensors were flown over the test field on a weekly basis during the growing season. From these sensors parameters such as plant height, growth rate, canopy cover, canopy cover progression, and boll count and size estimates were extracted. A genotype ranking (i.e. selection) based on UAS-derived variables was performed and results contrasted with actual field harvest data ranking of the same genotypes. Challenges in ranking/separating genotypes utilizing the UAS variables will be discussed.