An Analysis of Models for Estimating Surface Net Radiation from Incoming Solar Radiation

Abstract

<p>The importance of surface net radiation to several branches of physical environmental research is stressed and the dependence of this flux on surface controls is explored. An analysis of a group of models, employing incoming solar radiation to estimate net radiation, is presented, including a new approach which utilities parameters of greater specificity than appear in previous models. Potentially, this model is applicable in both clear and cloudy conditions. An expression for the parameter representing the intensity of the thermal response of a surface to radiant energy input is obtained analytically for some surfaces and the factors in this expression are discussed for all surfaces.</p> <p>The diurnal regimes of the radiation budget components for three surfaces - grass, bare soil and a cornfield - are presented in both graphical and tabular form for several days over a growing season. Differences in net radiation totals are shown to be the results of both short- and longwave factors. An analysis of this data reveals that the "heating coefficient" concept is unlikely to be a useful one for estimating net radiation on a routine basis, but that the resolution of the net longwave coefficients into unidirectional components holds some promise. The effect of surface dessication on the thermal response of a surface (and, hence, on net radiation) is apparent in the seasonal behaviour of the outgoing longwave parameters. This behaviour is different for each surface and, for the soil, is analysed in terms of the thermal and radiative properties specified in the theoretical analysis. Reflection coefficient differences were marked and the seasonal regimes of this parameter differed for each surface. Difficultly was encountered, however, in adequately predicting the parameters representing the influence of variations in atmospheric longwave emission on net radiation totals.</p> <p>The data collected were used to assess the predictive accuracy of various net radiation equation, over the period of a growing season. It was found that simple linear regression equations with insolation performed as well, in this respect, as the more sophisticated models. The potentially large errors associated with the use equations derived for different surfaces or locations is demonstrated. The model introduced in this study, however, represents more explicitly the controls on net radiations and could conceivably provide superior estimate of the flux when daily totals are required over periods of time longer than a growing season. Some suggestions for future studies of radiations exchange at the earth-atmosphere interface are presented.</p>