There is a new paper that further assesses the issue of long term temperature trends as a function of height near the ground. This new study was motivated by our paper
Pielke Sr., R.A., and T. Matsui, 2005: Should light wind and windy nights have the same temperature trends at individual levels even if the boundary layer averaged heat content change is the same? Geophys. Res. Letts., 32, No. 21, L21813, 10.1029/2005GL024407.
The new paper is
Steeneveld, G.J., A.A.M. Holtslag, R.T. McNider, and R.A Pielke Sr, 2010: Screen level temperature increase due to higher atmospheric carbon dioxide in calm and windy nights revisited. J. Geophys. Res., in press.
The abstract reads
“Long-term surface observations over land have shown temperature increases during the last century, especially during nighttime. Observations analyzed by Parker  show similar long-term trends for calm and windy conditions at night, and on basis of this it was suggested that the possible effect of urban heat effects on long-term temperature trends are small. On the other hand, a simplified analytic model study by Pielke and Matsui [2005, henceforth PM05] suggests that at night the resultant long-term temperature trends over land should depend on height and strongly on wind speed (mostly due to alterations in the rate of nocturnal cooling in the stable boundary layer (SBL)). In this paper we expand the PM05 study by using a validated atmospheric boundary-layer model with elaborated atmospheric physics compared to PM05, in order to explore the response of the SBL over land to a change in radiative forcing. We find that the screen level temperature response is surprisingly constant for a rather broad range of both geostrophic wind speed (5-15 ms-1) and 10 meter wind (2-4.0 ms-1). This is mostly due to land surface-vegetation-atmosphere feedbacks taken into account in the present study which were not considered by PM05.”
Among the conclusions, we write
“Our study shows that the competing effects of boundary-layer height and wind speed dependent fluxes yield changes in shelter temperature that are largely independent of wind speed. PM05 only considered the role of wind speed and boundary-layer height. It is likely that urban heat island effects as observed, are due to the resident time of a parcel of air over a city and not due to the flux changes considered here or in PM05. Furthermore, the parameter spaces investigated in this paper are limited. For example, the CASES-99 observational site simulated here is quite smooth (the roughness length used in the simulations was 0.03 m). It is possible that larger roughnesses might provide more sensitivity to wind speed. This will be further explored in McNider et al. .
Finally, we agree with PM05 that additional work is needed to understand SBL responses to both land use change and radiative forcing. Klotzbach et al. , for example, which shows a statistically significant divergence between the long-term trends of the surface air and lower tropospheric temperatures at higher latitudes in the winter, indicate that the changes in the SBL over time remains an important climate change issue that has been not yet completely examined and understood.”
The answer to the question of whether long term temperature trends near the surface are a significant function of height is an important climate metric issue, as these trends are used in the construction of the annual average global surface temperature trend. From this new study, it appears that feedbacks mute temperature trends near the surface, however, this was for a specific situation and may not be general to other landscapes. The new McNider et al paper, that is in preparation, will examine this issue for other situations, and we will report on this weblog when this study is complete.