With the advent of optical sensors, thermal-based indicators can be retrieved at multiscale levels from handheld devices to satellite platforms, providing a low-cost method to mirror plant water status. Here, we measured the canopy temperature of Orange trees subjected to different irrigation levels (100%, 75%, and 50% of crop water requirement) and strategies (regulated deficit irrigation (DI) and partial root drying (PRD)) to determine the crop water stress index (CWSI). Additionally, the CWSI was estimated based on Land Remote-Sensing Satellite (Landsat) thermal data using hot-cold patches (approach 1) and a novel mechanistic method combined with Sentinel-2 data (approach 2). Based on the in-situ measurements, the CWSI non-water stressed baseline was estimated as Tc – Ta = −0.57 × (VPD) + 2.31 (N = 370, R2 = 0.82), defining ‘VPD’ as ‘vapour pressure deficit’, and the upper limit was found to be relatively constant (Tc – Ta = 3.43°C). The in-field water stress variability among the different irrigation levels was effectively captured using the CWSI; however, the difference between the DI and PRD irrigated trees was only significant at the 50% irrigation level. Considering the remotely-sensed approach, the CWSI from our proposed method (approach 2) resulted in higher accuracy (root mean square error, RMSE = 0.03; mean bias error, MBE = −0.02) compared to approach 1 (RMSE = 0.10, MBE = −0.08). The improved accuracy from our proposed method was attributed to accounting for VPD and net radiation, applying an iterative method to calculate and calibrate aerodynamic resistance, and the use of high-resolution imagery from Sentinel-2 for reducing the soil background impact on canopy temperature.
A simple Landsat-MODIS (Moderate Resolution Imaging Spectroradiometer) fusion model was used to generate 30-m resolution evapotranspiration (ET) maps for the 2010 growing season over a heterogeneous sparse vegetation, agricultural region using the METRIC (mapping evapotranspiration with internalized calibration) algorithm. The fusion model performance was evaluated, and experiments were undertaken to investigate the frequency for updating Landsat-MODIS data into the fusion model during the growing season, to maintain model accuracy and reduce computation. Initial evaluation of the fusion model resulted in high bias stemming from the landscape heterogeneity and small landholdings. To reduce the bias, the fusion model was modified to be applicable pixel-wise (i.e., implementing specific pixels for generating outputs), and an NDVI-based (Normalized Difference Vegetation Index) coefficient was added to capture crop phenology. A good agreement that resulted from the comparison of the fused and non-fused maps with root mean square error (RMSE) of 0.15 mm day−1 with coefficient of determination (R2) of 0.83 indicated successful implementation of the modifications. Additionally, the fusion model performance was evaluated against in-situ observation at the pixel level as well as the watershed level to estimate seasonal ET for the growing season. The default METRIC model (Landsat only) yielded relative error (RE) of 31% and RMSE of 2.44 mm day−1, while using the modified fusion model improved the accuracy resulting in RE of 3.5% with RMSE of 0.37 mm day−1. Considering different data frequency update, the optimal fusion experiment (RMSE of 0.61 mm day−1, and RE of 6.5%) required the consideration of the crop phenology and weekly updates in the early growing stage and harvest time, and bi-weekly for the rest of the season. The resulting fusion model for ET output is planned to be a part of ET mapping and irrigation scheduling systems.