Estimation of the spatial and temporal variability of evapotranspiration components in a vineyard with the two source energy balance model (TSEB) using both thermal airborne and the fusion of Sentinel 2 and 3 imagery

Accurate estimates of crop water status in heterogeneous crops by means of remote sensing thermal imagery has been mostly achieved through airborne imagery, but not so much with satellites. The lower spatial resolution in the thermal infrared bands in most of the currently available satellite platforms is, among others, one of the reasons that limited its practical implementation for irrigation purposes. In this context, the European Space Agency (ESA) is funding the project Sentinel for Evapotranspiration (SEN-ET), which has the goal to develop an optimal methodology for estimating actual crop evapotranspiration (ET_a) at a spatial resolution of 20 m based on synergistic use of Sentinel 2 and Sentinel 3 satellites’ observations. To achieve it, ET_a estimates were evaluated, among others, with the two-source energy balance (TSEB) model. This work aims to assess the vineyards’ spatial and temporal variability of the ET_a estimates and its components with the TSEB as well as to compare differences in ET_a and T (transpiration) using the land surface temperature (LST) measured by airborne thermal imagery (TSEB-2T) and with a data mining sharpening method capable to fuse images from Sentinel 2 and 3 imagery (TSEB-PT) and to increase the spatial resolution of LST from 1 km to 20 m. To achieve it, three flights were conducted throughout the growing season with an aircraft equipped with a thermal sensor. Concomitant with image acquisitions, Ψ_stem was measured in 32 vines uniformly distributed in the vineyard. The correlations between Ψ_stem with ET_a and T were significant in all cases, but coefficients of determination (R²) slightly improved when both ET_a and T were estimated with TSEB-2T. However, estimating time-series of ET throughout the growing season with the TSEB-PT, combined with crop models, could be used to assess vine water requirements and for scheduling irrigation.