A simple cloud-based energy balance model to estimate dew

Two energy balance (EB) analyses, based on two different methods to estimate the downward long wave radiation, were used to model dew duration. Model I utilized cloud cover and cloud altitude to calculate sky temperatures whereas model II was based only on cloud cover to estimate sky apparent emissivity. The models were validated with hourly data from on-site weather station instrumentation for nights with and without occurrence of dew and cloud data from airport weather records, and compared with an empirical model, the CART/SLD model. The accuracy of predicting hourly dew occurrence with the EB models was greater than with the CART/SLD model. EB models I and II correctly predicted dew occurrence in 91 and 88% of hours, respectively, compared with 75% for the CART/SLD model. The majority of CART/SLD errors were due to late estimation of dew onset. It is proposed that an EB model can be used more effectively than the current CART/SLD model without a need for increased environmental observations in most localities.