Remote Sensing of Water Status in Perennial Agroecosystems with Hyperspectral, Thermal, and SIF Imagery

Warming trends and variable precipitation increasingly pressure water resources in the Western U.S., yet growers have been slow to adopt manual techniques for understanding actual plant water relations in water intensive agroecosystems. However, image based digital sensors continually advance plant water status observations from the sky with the potential to revolutionize current irrigation practices. As a result, the overarching goal of this research was to develop and validate a remotely sensed cascade of crop water stress indicators for orchard systems with irrigation intervention opportunities and tradeoffs under aridification. We specifically leveraged drone and tower spectrometers to identify optical limits for detecting water stress responses in a California almond orchard. Of 225 nonpareil trees, 185 underwent delayed irrigation stress treatments, where first irrigation was delayed until stem water potential reached 6 bar below baseline. Trees were monitored before first irrigation and after each of three successive waterings. During observation, a tower-mounted hyperspectral system (400-950 nm) capable of extracting solar-induced chlorophyll fluorescence and RGB imagery scanned tree crowns while a remotely piloted drone captured wide-band multispectral and thermal imagery with 70% overlap. After photogrammetry, vegetation indices used as proxy for water status were mapped on all 3400 orchard trees. Both tower and drone data were ground validated with pressure chamber, gas exchange, and microtensiometer measurements. Early tower results indicate that immediately after delayed irrigation an increased SIF response occurs, while structural indices show little to no change. Results are corroborated by drone-based imagery and holistic water stress detection ground measurements. Observed methodological agreement indicates good potential for growers to adopt remote sensing techniques as a means to quickly detect subtle plant physiological changes with implications for efficient irrigation practices, especially during recurrent conditions of water scarcity. These developed imaging workflows are poised to inform resource management and cost effective conservation strategies that are increasingly essential to sustainable agricultural water management in a changing climate.