Coupling spaceborne Solar-Induced Fluorescence and PRI to explain water stress variability in almond

Remote sensing plays an essential role in assessing crop water stress by providing insights to understand the spatial variability of water requirements. Traditional methods, such as the Crop Water Stress Index (CWSI), rely on canopy temperature due to its connection with transpiration rates and stomatal conductance. However, these methods face challenges, particularly when applied across large areas. Recent advances in spaceborne hyperspectral imaging offer alternative opportunities by quantifying physiological indicators, including solar-induced fluorescence (SIF) and narrow-band indices related to the xanthophyll dynamics, such as the Photochemical Reflectance Index (PRI). This study investigates the effectiveness of spectral indices derived from the DLR Earth Sensing Imaging Spectrometer (DESIS) hyperspectral imagery onboard the International Space Station for monitoring water stress in a large heterogeneous almond orchard. By comparing these spectral indices against conventional high-resolution thermal-based CWSI, we evaluated their performance in explaining crop water stress variability across a large almond orchard. Our findings show that SIF and PRI were sensitive indicators of water stress variability captured by thermal data, being the two most significant predictors of crop water stress among the indices tested. The coupled SIF and PRI model achieved an overall accuracy of 82% and a kappa coefficient of 0.77 when compared against the high-resolution CWSI data, significantly outperforming other combinations of spectral indicators. The model estimate was consistent over the range of thermal CWSI-based stress distribution throughout the orchard. This study highlights the potential of spaceborne hyperspectral data for large-scale operational assessment of crop water stress as an alternative to the lower sub-optimal resolution thermal imaging technologies currently available.