Polar Radiation Budgets of the NCAR CCM3*,1

Present-day Arctic and Antarctic radiation budgets of the National Center for Atmospheric Research Community Climate Model version 3 (CCM3) are presented. The CCM3 simulation is from a prescribed and interannually varying sea surface temperature integration from January 1979 through August 1993. Earth Radiation Budget Experiment (ERBE) data from 1985 through 1989 are used for validation of top-of-atmosphere (TOA) absorbed shortwave radiation (ASR) and outgoing longwave radiation (OLR). Summer ASR in both polar regions is less than the observations by about 20 W m22. While the annual mean OLR in both polar regions is only 2– 3 W m22 less than the ERBE data, the seasonal amplitude in OLR of 40 W m22 is smaller than the observed of 55–60 W m22. The annual polar TOA radiation balance is smaller than observations by 5–10 W m22. Compared to selected model and observational surface data, downward shortwave (SW) is too small by 50–70 W m22 and downward longwave (LW) too large by 10–30 W m22. Surface downward LW in clear atmospheres is too small by 10–20 W m22. The absence of sea-ice melt ponds results in 10–20 W m22 too much SW absorption during early summer and from 20 to 40 W m22 too little during late summer. Summer cloud covers are reasonably well simulated, but winter low cloud cover is too high by 0.5–0.7 compared to surface cloud observations. Comparison with limited satellite and in situ observations indicates cloud water path (CWP) is too high by about a factor of 2. While cloud particle sizes are approximately in the range of observed values, regional variation between maritime and continental droplet sizes is too strong over coastlines. Despite several improvements in CCM3 radiation physics, the accuracy of polar TOA annual radiation balance is degraded against the ERBE data compared to CCM2. Improvement in CCM3 polar radiation budgets will require improved simulation of CWP, clear sky LW, and sea ice albedo.