On the use of different approaches based on photochemical reflectance index and surface temperature to monitor the water status of winter wheat in semi-arid regions

Agriculture is considered to be the human activity that consumes the most mobilized water on a global scale. However, crops planted in semi-arid areas regularly face periods of moderate to extreme water stress. Such water stress periods have a considerable impact on the seasonal yield of these crops. In order to participate in a more rational irrigation water management, monitoring of the rapid changes in plant water status is necessary. For this purpose, the combination of two different wavelength ranges will be explored : an index based on Xanthophyll cycle (Photochemical Reflectance Index, PRI) and a commonly-used index from thermal infrared spectral range (LST). An experiment on winter wheat was carried out over two agricultural campaigns (2016 to 2018) in the Haouz basin, which is located in the Marrakech region, to better assimilate the temporal dynamics of PRI and surface temperature. In this study, four different approaches are proposed to study the functioning of wheat : 1- an approach based on solar angle to remove the structure effect (PRI₀) from the PRI signal and to derive a water stress index PRI_j, 2- an approach based on global radiation (R_g) to extrapolate a theoretical PRI (PRI_th) for R_g equal to zero and to calculate a water stress index PRI_lin, 3- an approach that determines an optimal PRI (PRI_pot) on the basis of the available water content (AWC) criterion in order to derive a stress index I-PRI and 4- an energy balance approach to extract dry and wet surface temperatures in order to establish a normalized surface temperature index (T_norm). The results of this work show a strong correlation between the PRI₀ and the Leaf Area Index with a coefficient of determination equal to 0.92, indicating that it is possible to isolate the structural effects of wheat on the PRI signal. In addition, over the range of variation in AWC, a significant correlation with PRI_j, PRI_jlin and I-PRI was observed with coefficients of determination of 0.71, 0.42 and 0.24, respectively. In contrast to the T_norm, which varies only for values of AWC below 30%, a coefficient of determination of 0.22 is obtained. Finally, the PRI allows us to acquire early and complete information on the response of wheat to change in AWC as opposed to the surface temperature index, revealing the potential of the PRI to monitor the water status of plants and their responses to changing environmental conditions.