Estimating Class A pan evaporation at locations with sparse climatic data

Abstract

Class A pan evaporation was found to correlate closely with daylength-adjusted maximum daily air temperature (ATM), on a bimonthly basis. ATM and altitude were used as variables in a regression analysis to develop a series of evaporation-estimation equations for various regions in southern Africa. These regions had been delineated on the basis of latitude, altitude and distance from the sea, which are all factors affecting evaporation rates. It was found that evaporation-estimation equations tested on an annual basis were preferable to seasonal expressions, with the exception of the Winter Rainfall Region. Standard errors of the estimates of pan evaporation were in the order of 1 mm per day on a bimonthly basis.