Comparative assessment of actual evapotranspiration partitioning models over an irrigated citrus orchard using remote sensing, agrometeorological, and micrometeorological data

Assessing actual evapotranspiration (ETa) and its partitioning into soil evaporation (Ea) and plant transpiration (Ta) can be used to derive crop water requirements, which are crucial for optimising water resource management in agriculture. In this study, different approaches were employed to monitor crop water use, based on a combination of in situ measurements and remote sensing observations. These approaches include: a Soil Water Balance (SWB) model, a Surface Energy Balance (SEB) model, and an Eddy Covariance (EC) tower flux partitioning method. This research aims to investigate the potential synergies between these approaches for an operational estimation of both Ta and Ea, in the context of future satellite missions. A SWB model was implemented to estimate daily soil water content (SWC), Ta, Ea and, therefore, ETa. A SEB model allowed estimating its instantaneous four main components and daily Ta and Ea. Finally, an EC-based partitioning method allowed the observed ETa partitioning, into Ta and Ea, at a daily scale. The approaches were applied in a citrus orchard (sparse crop), and their performance was evaluated via the comparison with field measurements (i.e., EC tower, SWC and sap flow probes). All three methods provided accurate daily estimates of Ta (RMSE ranging between 0.20 and 0.39 mm d−1). The outcomes of this research highlighted significant uncertainties in the estimation of soil-vegetation-atmosphere exchanges when weeds were present in the study area, suggesting that a joint use of different methodologies, based on different inputs, can help disclose these uncertainties.