Long-time global radiation for Central Europe derived from ISCCP Dx data
1TU Dresden, Fakultaet Forst-, Geo- und Hydrowissenschaften, Fachrichtung Wasserwesen, 01062 Dresden, Germany
2German Meteorological Service, Meterological Observatory Lindenberg (FELG), 15848 Tauche/Lindenberg, Germany
Abstract. The global Dx dataset of the International Satellite Cloud Climatology Project (ISCCP) with a spatial resolution of about 30×30 km² was analysed to produce spatially highly resolved long-time datasets to describe the radiation budget for Central Europe over the period of 1984–2000. The computation of shortwave and longwave radiant flux densities at top of atmosphere and at surface was based on 1D radiative transfer simulations. The simulations were carried out for all relevant atmospheric and surface conditions and the results were inserted into a look-up table. Thus, long-time calculations for all conditions and time slices of the Dx dataset could be realised. The study is focussed on the global radiation at surface.
The first examination was carried out for the ISCCP D1 and the ISCCP D2 dataset. These datasets, including cloud and surface information on a different spatial scale (280×280 km2), were applied to the produced look-up table analogue to the Dx data. The calculated global radiation of the D1 and D2 dataset were compared to the Dx dataset. The differences between these datasets mainly range from 5–15 Wm−2 (2–6%) with regional peaks up to 25 Wm−2 (10%).
The evaluation with the GEWEX Surface Radiation Budget (SRB) data emphasises differences between 5–25 Wm−2 (6–16%) over land areas. Deviations to an ISCCP provided flux data set vary from 0 Wm−2 in the North up to 35 Wm−2 (0–13%) in the South of Central Europe.
The global radiation datasets provided by the Global Energy Balance Archive (GEBA) and the German Meteorological Service (DWD) agree well, but they are 5–25 Wm−2 (7–10%) lower than the Dx results.
Annual analyses of global radiation of various regional climate models complete the study. It is figured out that the used models and methods reveal a couple of discrepancies. Especially in wintertime the results of our analysis differ to the considered models. Principally the uncertainties were caused by the determined range of values and simplifications for the computation of the radiative transfer simulation.