There has been quite a bit of misunderstanding in the media with respect to the paper
Zhou, Liming, Yuhong Tian, Somnath Baidya Roy, Chris Thorncroft, Lance F. Bosart and Yuanlong Hu 2012: Impacts of wind farms on land surface temperature. Nature Climate Chnage. doi:10.1038/nclimate1505
For example, there are these news reports which are accurate but understate the actual significance of the Zhou et al study; e.g. see
No, wind farms are not causing global warming by Brad Plumer
and those which overstate what the Zhou et al study means; e. g. see
Wind farms are warming the earth, researchers say by Eric Niiler
Thus, this weblog post is to summarize the implications of his study and to expand on what was posted in
First, as reported in the above post,
[I}n certain parameter spaces the nocturnal boundary layer can rapidly transition from a cold light wind solution to a warm windy solution. In these parameter spaces even slight changes in longwave radiative forcing or changes in surface heat capacity can cause large changes in surface temperatures as the boundary mixing changes. However, these temperature changes reflect changes in the vertical distribution of heat, not in the heat content of the deep atmosphere.
This, therefore, means that the vertical mixing of heat does not contribute to global warming as defined using heat measured in Joules. The kinetic energy dissipation due to the turbines, as an atmospheric heating term, is not a significant warming effect. Thus the headline that wind farms are warming the Earth is incorrect.
However, in terms of how global warming is measured by the IPCC and others in terms of a global average surface temperature anomaly, the Zhou et al paper has major implications which were not reported on in the Washington Post article.
First, other land use changes (which cover vast areas, see Pielke et al 2002) will similarly alter the vertical distribution of heat near the surface (both day and night). We have documented this effect during daytime for the conversion of short grass steppe to irrigated cropland as we show in Figure 1 in the paper
Pielke, R.A. and X. Zeng, 1989: Influence on severe storm development of irrigated land. Natl. Wea. Dig., 14, 16-17
based on aircraft and radiosonde flights in northeast Colorado described in depth in
Segal, M., W. Schreiber, G. Kallos, R.A. Pielke, J.R. Garratt, J. Weaver, A. Rodi, and J. Wilson, 1989: The impact of crop areas in northeast Colorado on midsummer mesoscale thermal circulations. Mon. Wea. Rev., 117, 809-825.
What makes the Zhou et al so important, that unlike other land use changes which also alter surface albedo, the partitioning between latent and sensible heat flux, ect, the Zhou et al study is able to isolate the effect of just altering the vertical distribution of temperature near the surface.
In terms of diagnosing global warming using the surface temperature anomalies, this is an important observational finding. The global average surface temperature anomalies are computed by the formula
ΔT (global average) = ΔT(ocean average) times fraction of the globe covered by ocean + ΔT (land average) times the fraction of the globe covered by land (including regions with ice sheets).
The value of ΔT (land average) is computed by
ΔT (land average) = ∑ Δ [Tmax + Tmin]/2
where Δ [Tmax + Tmin]/2 is the mean temperature anomaly, such as used in the BEST study, and by NCDC, CRU and GISS; focued on 2m above the surface [the Zhou al al values are at the surface which will have a larger difference than at 2m but of the same sign, and still significant].
As Zhou et al has shown, if the vertical mixing at night is increased, a warm bias occurs IF one is using Tmin in the construction of the land part of the global average surface temperature anomaly.
This change in vertical mixing will occur not only due to regional and mesoscale land use change but even for otherwise pristine locations (e.g. Siberia, northern Canada, Alaska etc) if the local siting where the surface air temperatures are measured has changed over time (such as due to buildings being constructed, adjacent roads). Such local siting issues has been very effectively described for the continental USA by Anthony Watts in his surface station project.
- Since the stable boundary layer occurs most nights (and occurs almost all winter at higher latitudes), this vertical mixing is contaminating the use of the surface land temperatures to diagnose global warming.
- It has introduced a warm bias in its construction that extends much beyond just where the wind farms occur.
The wind farms have provided a field study of opportunity which Zhou et al have been able to utilize for their important study.
I have suggested a follow on study to the authors. They should bin their data for days in which the wind turbines are turning and those in which they are not. For the days the turbines are not working, there should be little difference in trends with the surrounding area, but a larger amplitude signal when they are working. They should also bin by days with near calm large-scale winds and stronger winds of different thresholds to ascertain if, above a certain wind threshold, the natural vertical mixing of heat results in little difference between the wind farm and surrounding areas.
The Zhou et al 2012 paper, hopefully, will open a much needed discussion on the value of using surface temperature anomalies as the metric to define climate system heat changes (i.e. global warming and cooling).
Climate Science: Roger Pielke Sr. 
