Estimates of Pan Evaporation from Mean Maximum Temperature and Vapor Pressure

Abstract

An equation for estimating pan evaporation has been derived by regression analysis from climatic data of a number of widely separated stations in Australia. The required data are mean maximum temperature, mean vapor pressure, and day length. The equation resembles the bulk aerodynamic (mass transfer) expression, but does not contain a wind parameter. An empirical variate was postulated and compared with five other variates also available from limited data. Significance tests were applied to assess the likelihood that observed differences between correlations may have arisen through sampling. A number of cases are found with significantly (P<0.01) higher correlations in favor of the postulated empirical variate. For all the variates, the slopes and heights of individual reasons based upon data of specific stations often differed significantly (p<0.01) from those of a derived regression which best fitted the combined data from a number of stations. Regardless of the empirical variate considered for estimating the pan evaporation, no single regression is found which can be applied over a wide range of climatic conditions with an expectation that only random errors will remain. The degree to which the slopes and heights of the station regressions vary from those of the best-fitting combined regression nonetheless differs considerably between variates, and the postulated variate compares favorably in both respects. The method, although tentative, appears acceptable where estimates of pan evaporation over periods an short as one week are needed. There is evidence that maximum temperature and vapor pressure can be used more effectively in estimating pan evaporation, particularly under climatic conditions such as are characteristic in much of northern Australia.