Synthesis of incoming shortwave radiation for hydrological simulation

Over the past decade, measurements of incoming shortwave radiation and of its proxy, daily bright sunshine hours, have become scarce in western Canada and elsewhere. As these data are critical for computing net radiation for snowmelt, evaporation, soil thaw and other components of the hydrological cycle, other means to estimate incoming shortwave radiation are needed. National Centers for Environmental Prediction and North American Regional Reanalysis atmospheric model reanalysis estimates of daily incoming shortwave radiation (Q_si), as well as the results of simplified semi-empirical calculations, were compared with measurements to determine their usefulness for hydrological calculations. It was found that all of the daily estimates show considerable bias and scatter compared with measurements. The best estimated values were produced by the semi-empirical Annandale method, particularly for simulations over the period of spring snowmelt. As many models require hourly radiation data, a method is presented for rescaling simulated daily Q_si data to the hourly values required.