Abstract : 3N.20
Shortwave radiation balance in complex terrain and the Radiosity method

Nora Helbig, Henning Löwe, Michael Lehning
helbig@slf.ch
Swiss Federal Institute for Snow and Avalanche Research, SLF / WSL

The accurate prediction of the spatial distribution of local surface temperatures in mountainous terrain is an important and challenging task. In particular it is of interest for a winter surface energy balance for avalanche formation and road surface condition prediction. On principle, for a shortwave radiation balance in complex terrain with a spatial resolution of tens of meters topographic shading, single and multiple reflections at surrounding terrain surfaces and atmospheric backscattering should be taken into account. The influence of multiple reflections on the netto incoming shortwave radiation becomes most important in the presence of large fluctuations in surface albedo values (e.g. for partly snow covered domains). Therefore, a common approach which adopts the concept of sky view factor and a single terrain view factor is generalized to heterogeneous terrain with varying albedos. This leads to the known Radiosity equation for the incoming shortwave radiative fluxes which neglects the feedback of the atmosphere but accounts for shading and diffuse reflections at terrain surfaces. For the iterative solution of the linear system of equations the Progressive Refinement algorithm is applied which best meets the high memory requirements for large model domains with local resolutions on single processor computers. The performance of the model is discussed by simulating radiation transfer on artificial digital elevation models. Due to lacking experimental data for the individual simulated radiation components (direct, diffuse sky, reflected terrain) we are restricted to available high quality measurements of direct and diffuse (here diffuse means diffuse sky AND reflected terrain) components from a measurement site at Weissfluhjoch(CH). Although the instruments only see about 4% terrain we find significant differences for the incoming fluxes when taking terrain reflections into account. Simulations were conducted with a model system for alpine surface processes.