There is a new paper which is directly related to the ability of models to skillfully simulate temperatures in the lowest levels of the atmosphere. This includes, of course, the 2m level which was discussed in several recent Climate Science weblogs (see, see and see).
The new paper is
Mölders, N., and G. Kramm, 2009: A case study on wintertime inversions in Interior Alaska with WRF, Atmos. Res., doi:10.1016/j.atmosres.2009.06.002, in press.
The abstract reads
“The Weather Research and Forecasting (WRF) model is run in various configurations for a five day cold weather period with multi-day inversions over Interior Alaska. Comparison of the simulations with radiosonde data and surface observations shows that WRF’s performance for these inversions strongly depends on the physical packages chosen. Simulated near-surface air temperatures as well as dew-point temperatures differ about 4 K on average depending on the physical packages used. All simulations have difficulties in capturing the full strength of the surface temperature inversion and in simulating strong variations of dew-point temperature profiles. The greatest discrepancies between simulated and observed vertical profiles of temperature and dew-point temperature occur around the levels of great wind shear. Out of the configurations tested the radiation schemes of the Community Atmosphere Model combined with the Rapid Update Cycle land surface model and modified versions of the Medium Range Forecast model’s surface layer and atmospheric boundary layer schemes capture the inversion situation best most of the time.”
This paper confirms that the accurate paramterization of the temperatures at 2m is a challenge. The abstract writes
“Simulated near-surface air temperatures as well as dew-point temperatures differ about 4 K on average depending on the physical packages used. All simulations have difficulties in capturing the full strength of the surface temperature inversion and in simulating strong variations of dew-point temperature profiles.”
The same type of inaccurate paramterizations is used in the multi-decadal global climate models that were used in the 2007 IPCC report.
Since the errors are several degrees within stable atmospheric boundary layers (which are typical at night over land almost everywhere, and in the higher latitude winters all day), there should no confidence in the ability of these IPCC modes to skillfully predict the change in 2m temperatures for these conditions decades into the future [the 2m temperatures are used in the construction of the global average surface temperature trends].
This paper further illustrates major problems with using surface temperature trends to diagnose and predict global warming and cooling, as we have discussed, for example, in
Pielke Sr., R.A., C. Davey, D. Niyogi, S. Fall, J. Steinweg-Woods, K. Hubbard, X. Lin, M. Cai, Y.-K. Lim, H. Li, J. Nielsen-Gammon, K. Gallo, R. Hale, R. Mahmood, S. Foster, R.T. McNider, and P. Blanken, 2007:Unresolved issues with the assessment of multi-decadal global land surface temperature trends. J. Geophys. Res., 112, D24S08, doi:10.1029/2006JD008229,
and, see also, the excellent guest weblog by Professor McNider,