Resumen

CERVANTES-OSORNIO, Rocío et al. Comparison Models for calculating actual vapor pressure. Tecnol. cienc. agua [online]. 2013, vol.4, n.2, pp.37-54. ISSN 2007-2422.

Although real vapor pressure is a fundamental variable in the calculation of the evapotranspiration of crops, a component of the hydrological cycle, it is difficult and costly to measure directly. Consequently, in practice calculations are based on temperature data and psychometric relationships. The objective of the present work was to compare different conventional methods used to calculate real vapor pressure with calculations performed using two types of artificial neural networks-feedforward backpropagation and radial basis function. Meteorological data were used from four weather stations in Irrigation District 075, located in Valle del Fuerte, in northern Sinaloa State, Mexico. The results indicate that the radial basis function type of artificial neural network (Scenario E4) was the best method to calculate real vapor pressure.

Palabras llave : atmospheric humidity; vapor pressure deficit; artificial neural network.