Avaliação de métodos de estimativa da evapotranspiração de referência para a região do vale do São Francisco

Abstract

Information related to reference evapotranspiration (ET0) is very important for irrigation management, especially in regions dependent on agriculture. To determine it several methods can be used, among which are the alternative methods, which are used of few data. This work hypothesized that new models for estimating evapotranspiration and those developed locally better represent ET0 of the Submedio São Francisco. In order to investigate the adequacy of historical and contemporary models for the estimation of ET0 and to develop a new method for its estimation in the region of the Vale do São Francisco. The work was carried out in Juazeiro, Ba, using data referring to a meteorological station belonging to EMBRAPA Semiarid, using data of temperature, relative humidity, wind speed, precipitation and global solar radiation, for the year 2017 and 2018, and for validation of the poposed odel, data from 2016. The data were used in the determination of evapotranspiration using the Penman-Monteith methods and the alternative ones, subdivided into two groups: historical,those developed before theyear 2000with the methods of Hargreaves-Samani, Camargo, Jensen-Haise, Makkink, Solar radiation FAO 24 and Benevides-Lopes, and contemporary, developed afterthe year 2000, with Irmak, Castaneda-Rao, Moretti-Jerszurki-Silva, Silva-Souza, additionally, the information was used in the development of a simple model for ET0 estimation. This information was compared using the indices for which Pearson's correlation coefficient (r), coefficient of determination (r²) and Willmont's concordance index were considered. Based on these results, it was possible to observe that contemporary methods, with the exception of Moretti-Jerszurki-Silva, tended to reduce errors in the estimation of reference evapotranspiration.The most appropriate method to estimate ET0 was the FAO-24 Solar Radiation. The alternative model obtained in the present work was based only on temperature and humidity data and was classified as very good.