Key Parameters for the Inconsistencies of the Incoming Solar Radiation Boundary Condition in Global Modeling

By a comparison of the insolation, computed by 19 different climate models for the International Panel on Climate Change Fourth Assessment Report (IPCC-AR4) test period from 1980 to 1999, it is shown that those models used different values for the solar constant and for its solar cycle variations. Meridional profiles for the monthly incoming radiation displayed diversities of up to ±10 Wm^-2, especially during the transient seasons. Sensitivity studies with minima and maxima for the assumed orbital parameters of the Earth show almost no change. However, the different temporal partitioning for onset and length of individual months based on different calendars (e.g. simplifications such as 30 days for each month) results in the difference in the insolation, which is strongly resemble in amount and in zonal pattern the observed diversity of the insolation in IPCC models. Contributing error sources are also different assumptions for cut-off angles at low sun-elevations and differences in increment-difference during spatial and temporal integrations. Possible impacts of these contributing errors in climate modeling are investigated within a coupled ocean-atmosphere model. It is found that monthly radiative fluxes, humidity, and temperature have a difference between the two vernal equinox experiments. Although it remained within the magnitude of the inter-model difference, the difference is systematic.