parametric and non-parametric trend of reference evapotranspiration and its key influencing climatic variables (case study: southern iran)

evapotranspiration is one of the most important components of the hydrological cycle which is directly influenced by atmospheric conditions. this study investigated annual and seasonal trends in reference evapotranspiration (et0) and its key influencing climatic variables during 1966-2005 at 10 stations in southern iran (with centrality of fars province). first, multivariate regression analysis was performed to identify the major meteorological variables affecting et0. second, annual and seasonal trends in climatic variables as well as et0 were assessed using the mann-kendall test, spearman's rho, the pearson correlation and linear regression to evaluate their contribution to the temporal trend in et0. results suggested that the more effective variables for et0 were wind speed (u2), relative humidity (rh) and sunshine hours (n). also, the majority of trends in seasonal and annual et0 were non-significant and after that decreasing and increasing trends had higher frequencies. in addition, distributions of relative frequencies of trend types at all considered time-scales were similar for both parametric and non-parametric techniques. hence, the disagreement between parametric and non-parametric trend results did not depend on the degree of normality in the annual and seasonal et0 distributions in the study area.