Adaptive estimation of crop water stress in nectarine and peach orchards using high-resolution imagery from an unmanned aerial vehicle (UAV)

The capability to monitor water status from crops on a regular basis can enhance productivity and water use efficiency. In this paper, high-resolution thermal imagery acquired by an unmanned aerial vehicle (UAV) was used to map plant water stress and its spatial variability, including sectors under full irrigation and deficit irrigation over nectarine and peach orchards at 6.12 cm ground sample distance. The study site was classified into sub-regions based on crop properties, such as cultivars and tree training systems. In order to enhance the accuracy of the mapping, edge extraction and filtering were conducted prior to the probability modelling employed to obtain crop-property-specific ('adaptive' hereafter) lower and higher temperature references (T_wet and T_dry respectively). Direct measurements of stem water potential (SWP, ψ_stem ) and stomatal conductance (g_s ) were collected concurrently with UAV remote sensing and used to validate the thermal index as crop biophysical parameters. The adaptive crop water stress index (CWSI) presented a better agreement with both ψ_stem and g_s with determination coefficients (R² ) of 0.72 and 0.82, respectively, while the conventional CWSI applied by a single set of hot and cold references resulted in biased estimates with R² of 0.27 and 0.34, respectively. Using a small number of ground-based measurements of SWP, CWSI was converted to a high-resolution SWP map to visualize spatial distribution of the water status at field scale. The results have important implications for the optimal management of irrigation for crops.