Correlation of UAV and satellite-derived vegetation indices with cotton physiological parameters and their use as a tool for scheduling variable rate irrigation in cotton

Abstract Current irrigation management zones (IMZs) for variable rate (VRI) systems are static. They delineated in the beginning of season and used thereafter. However, recent research has shown that IMZ boundaries transient change with time during growing season. The primary goal this study was to explore potential using vegetation indices (VIs) developed from unmanned aerial vehicle (UAV) satellite images predict cotton physiological parameters can be delineate in-season IMZs. A 2 year conducted a 38 ha commercial field southwestern Georgia, USA. Throughout two seasons, VIs were calculated UAV, PlanetScope, Sentinel-2 images. Predawn leaf water (LWP PD ) plant height measured at 37 locations on same day as flights correlated UAV based-VIs. GNDVI (Green normalized difference index) best predictor correlation values 0.72 (p < .0001) 0.84 2019 2020, respectively. secondary compare performance dynamic VRI (DVRI) conventional irrigation. divided into alternating parallel conventional, DVRI strips scheduling methods. irrigated farmer’s standard method individual IMZs within based soil tension (SWT) wireless moisture sensor network. LWP SWT measurements well. initially texture, apparent electrical conductivity (ECa), yield maps previous years modified reflect patterns observed maps. In system prescribed an average amount 50.8 mm while applied 58.4 mm. Average yields 1248 1191 kg −1 , resulted 4.6% higher than irrigation, applying 14.0% less water. Despite lower application by DRVI system, comparison between not conclusive.