Estimation of net short-wave radiation by the use of remote sensing and a digital elevation model-a case study of a high arctic mountainous area

Abstract

A fast method for estimating the net short-wave radiation in high arctic mountainous areas is developed by combining Landsat-5 Thematic Mapper (TM) satellite data and a digital elevation model (DEM). Methods to: (1) correct remote sensing data for the atmospheric effect; (2) perform a thorough and extensive topographic correction via the DEM; and (3) model the irradiance from a radiation model are described. By combining the above-mentioned methods, a model with great applicability-e.g. creation of albedo maps, terrain modelled short wave radiation and correction of satellite images for both topographic and atmospheric effects, is developed. A great advantage of this model is that it combines atmospheric and topographic correction. Most models only correct for one of the two. Another great advantage is that it has a limited number of inputs. Beside the satellite data and the DEM, the only inputs are visibility, latitude, longitude, atmospheric pressure and time of day and day of year (DOY). Furthermore the model can be implemented in any image processing software. The model was tested and validated on satellite data from April to September in the period 1995 to 1998 with a total of nine TM-images. To validate the model, the results were compared with ground measurements. The result of the validation showed that: (1) the modelled results show very good agreement with the measured data; (2) atmospheric and topographic correction of images is possible without using expensive software; and (3) the topographic correction is of great importance. Even a small change in the topography has a greater effect on the net short wave radiation than a lack of atmospheric correction. The future use of the model is to produce input to othermodels that estimate snowmelt, energy balance, biomass production and CO 2 -fluxes and to determine regional and seasonal variation.