Journal of Agronomy and Field Innovations

Background: Heat stress is one of the crucial constraints affecting Gossypium hirsutum L. (upland cotton) yield since it can negatively impact the boll set, fiber quality, and lint on an increasing number of occasions. Identification of heat stress in the cotton foliage using conventional point detectors is complicated.
Objective: The goal of this research was to test an unmanned aerial vehicle (UAV)-based system for thermal imaging to detect heat stress early under field conditions and compare its efficiency with on-ground physiological and micrometeorological measurements.
Methods: A multirotor UAV equipped with a radiometric optical long-wave infrared sensor was flown at an altitude of 40 m over a cotton field in rain-fed conditions at three stages of plant development. Thermal orthomosaics were created using the Pix4D software and analyzed in QGIS and MATLAB in order to calculate the leaf temperature, the difference between the temperature of the canopy and the temperature of the air, and the Crop Water Stress Index. Data obtained from the ground included the temperature of the leaf, stomatal conductance, and volumetric soil water content obtained from 40 samples taken.
Results: The study found a strong correlation between UAV-derived CWSI values and in-situ stomatal conductance and leaf water potential, demonstrating an accuracy of 91.4% in distinguishing between stressed and non-stressed canopy zones, with a measured RMSE of 0.86 °C between UAV estimated canopy temperature and handheld measurement. The stressful zones were shown to appear 3-4 days before visible wilting take place.
Conclusion: Thermal imaging using UAV technology can be viewed as a non-destructive and fast way of detecting early heat stress in cotton crops, as it can help irrigation and management decisions to be taken in time, in addition to providing a scalable technology for sustainable cotton farming.