Effect of the representativeness of meteorological input data on the estimation of reference evapotranspiration (FAO-56) and crop evapotranspiration in intensive agricultural systems under a subtropical climate

Reference evapotranspiration (ET₀) is a key variable for quantifying crop water requirements and for irrigation management. The FAO-56 Penman–Monteith equation is widely adopted as the standard method for estimating ET₀; however, its application depends strongly on the availability, quality, and spatial representativeness of the input meteorological variables. In agricultural regions with high microclimatic variability, the heterogeneous spatial distribution of meteorological stations can introduce substantial uncertainties into ET₀ estimates. Southern Brazil is characterized by a subtropical climate with relatively well-distributed precipitation throughout the year, thus allowing for rainfed cropland systems. In this region, agricultural systems exhibit a high intensity of land use, with croplands remaining under production almost continuously. As a consequence of this intensification, rainfed systems are dominated by crop rotations involving soybean, wheat, and maize, often combined with cover crops. In lowland areas, production systems are primarily based on flood-irrigated rice, alternated with soybean or pasture. This diversity of land uses, together with the resulting variability in surface conditions, poses additional challenges for the accurate estimation of crop evapotranspiration (ET_c), due to the strong influence of microclimatic conditions on soil–plant–atmosphere water and energy exchanges. Within this context, this study aims to evaluate the impact of using in situ versus regional meteorological data on the estimation of ET₀ and on the determination of actual crop evapotranspiration (ET_c,act) and therefore actual crop coefficients (K_c,act). To this end, the SIMDualKc model will be applied to estimate actual ETc following the FAO-56 approach, and the results will subsequently be evaluated against ET measurements derived from eddy covariance flux towers. These analyses will be conducted in two agricultural areas under crop rotation in the state of Rio Grande do Sul, Brazil: one under rainfed conditions and another located in a lowland system. For the application of the SIMDualKc model, meteorological data from the Instituto Nacional de Meteorologia (INMET) station located approximately 50 km from each field will be used, as well as data from a local meteorological station installed over a reference surface at a distance of about 1 km from the monitored areas. In addition to the difference in distance relative to the croplands, the local station provides direct measurements of net radiation and soil heat flux, variables that must be estimated when using INMET station data. Therefore, it is expected that the results will demonstrate that the use of local meteorological data allows a more robust quantification of the uncertainties associated with FAO-56-based ET_c estimates and, if necessary, supports adjustments of actual K_c derived from regional data, accounting for local microclimatic patterns. Finally, the study seeks to highlight that spatial discrepancies related to station distance may fail to represent the local climatic conditions relevant for ET₀ estimation and, consequently, may directly affect uncertainties in actual crop evapotranspiration estimates and irrigation management.