UAV and Satellite-Based Images Correlation with Cotton Physiological Parameters and Their Use As a Tool for Variable Rate Irrigation in Cotton

Variable rate irrigation (VRI) is an important precision agriculture practice that aims to optimize water use efficiency in crops by applying varied rates of water in a timely manner. The irrigation is based on irrigation management zones (IMZs), which are areas within a field that present similar characteristics. These characteristics can be related to soil properties, crop performance, soil-plant relations, yield variability, soil water content, among others. Nevertheless, recent research has shown that plant water content is a great indicator of the plant’s water requirement variability across the field and throughout the growing season. Physiological parameters such as leaf water potential can be potentially used as a decision tool, but current sampling methods limit sampling in large scale. Remotely sensed platforms can provide quick data collection for large areas being a potential solution for the scalability problem. The current study aims to estimate cotton physiological parameters by correlating vegetation indices (VIs) calculated from UAV and satellite multispectral images with ground truth measurements collected throughout the 2019 growing season. The study was conducted in a 54ha cotton commercial field in southern Georgia. The field was divided in 4 alternating parallel strips, with 2 strips being irrigated with conventional method, and two irrigated based on a VRI system. IMZs were delineated based on soil properties, yield map and NDVI images. Each IMZ had at least one soil moisture sensor (continuously collecting moisture data) that was used to calculate application rates. LWP, plant height, nodes number, nodes above white flower, assimilation rate and stomatal conductance data were collected weekly in the same day images were acquired. To access results, regression analysis will be used to evaluate potential use of VIs, 1) in estimating plant water content, 2) in identifying water requirement spatial variability to the delineation of VI-based IMZs.