Using UAV-based thermal imagery to detect crop water status variability in cotton

Plant-based measurements such as leaf water potential (LWP) are widely used for irrigation scheduling because they are accurate at indicating when irrigation is needed. Despite being a good indicator, scheduling irrigation with LWP is time consuming and scale-limited. The work reported in this study explored the potential of using thermal remote sensing to estimate cotton crop water status in the humid southeastern U.S.A. The study was conducted over two growing seasons (2018 and 2020) in southwestern Georgia, U.S.A using a complete randomized block design plot scheme with three irrigation treatments (0% ETc (crop evapotranspiration; rainfed), 100% ETc (well-irrigated), and 125% ETc (over-irrigated). To monitor the irrigation treatment effects on cotton physiological responses, predawn LWP (LWPPD), stomatal conductance (gs) and leaf area index (LAI) data were collected in both growing seasons. UAV-based images collected in the thermal infrared waveband were used to calculate crop water stress index (CWSI) based on three different methodologies and evaluated as predictors of LWPPD. Results in this study suggest that LWPPD values above-0.45 MPa indicate a non-stressed crop. No negative effects in leaf stomatal conductance and crop growth were observed in 2018. In 2020, the less and more irregular precipitation led to significant differences in LAI and gs, as well as in LWPPD. A moderate to strong relationship was observed for all dates in 2020, with the CWSI based on the Monteith approach (CWSIMonteith) showing the two highest R 2 values among the 3 dates (0.65 and 0.58) with low RMSE values of 0.02 and 0.04 MPa, respectively. Overall, the results showed that there is potential of using an affordable UAV-based thermal system to produce predicted LWP maps that are representative of the current field water status.