How much error is there? A slight amount of error in the data (likely or at least not complete data) or a slight difference between theory and reality (very likely) means that the results are meaningless. A very minor difference means that their results don't support Global warming. There will be people who go back and redo this study to show that global warming doesn't exist.
There aren't enough temperature sensors. Of the ones there are (i.e. Fig 1 from the study) many are in urban areas that are likely to have higher temps due to being in an area that absorbs the sunlight. That skews the data. I think the study showed exactly what the they expected it to show because that was how they interpreted their data.
Totally crap research, quote Direct measurements of
Rs cannot be quanti-tatively related to such variability because they have been limited
in their geographical coverage. The approach used here is to
examine a global land dataset of diurnal temperature range(DTR). This concept is not new, indeed, Wild et al. (7) noted
Another **** proxy research paper, not worth the paper it is printed on.
Since you posted the link to it, it is unlikely to be objective, and thus not worth the time it would take to read it.
In a recent study, approved earlier this month, in PNAS uses diurnal temperature range to establish how much solar radiation is hitting the surface. They show that solar input reached a peak in the 1940s and 1950 and has steadily fallen since then. since the mid-1980s solar input has been relatively unchanged. Do you see any problems with this paper after reading it through? Note: this is different than data from sunspot cycles in that these measurements also take into account such things as aerosol activity.
http://www.pnas.org/content/early/2013/08/21/1311433110.full.pdf
