Evaluación de dos modelos para la estimación de la evapotranspiración de referencia con datos CERES

Abstract

<p>Evapotranspiration is the most important variable in the Pampas plain. Information provided by sensors onboard satellite missions allows represent the spatial and temporal variability of evapotranspiration, which cannot be achieved using only measurements of weather stations. In this work, the Priestley and Taylor (PT) and FAO Penman Monteith (FAO PM) equations were adapted to estimate the reference evapotranspiration, ET<sub>0</sub> , using only CERES satellite products (SYN1 and CldTypHist). In order to evaluate the reference evapotranspiration from CERES, a comparison with in situ measurements was conducted. We used ET data provided by the Oficina de Riesgo Agropecuario, corresponding to 24 stations placed in the Pampean Region of Argentina (2001-2016). Results showed very good agreement between the estimates with CERES products and in situ values, with errors between ±0.8 and ±1.1 mm d–<sup>1 </sup>and r<sup>2</sup> greater than 0.75 at daily scale, and errors between ±14 and ±19 mm month<sup>–1</sup> and r<sup>2</sup> greater than 0.9, at monthly scale better results were obtained with adapted model FAO PM than PT. Finally, ET<sub>0</sub> monthly maps for the Pampean Region of Argentina were elaborated, which allowed knowing the temporal-spatial variation in the validation area. In conclusion, the methods presented here are a suitable alternative to estimate the reference evapotranspiration without requiring ground measurements.</p>