New research shows that local and regional socioeconomic influences on temperature play a significant role in “observed” global warming.
Factors such as economic activity and data quality—which are not included in climate models—are closely tied to the temperature increases observed during the past two decades, according to a study published May 25 in Climate Research.
This finding has important ramifications: 1) That human-induced greenhouse gas emissions and changes in the sun are not the only causes of the observed warming. 2) The degree of warming associated with an enhanced greenhouse effect has apparently been overestimated.
The research was a cooperative effort by University of Guelph economist Ross McKitrick and University of Virginia climatologist Patrick Michaels. They developed a mathematical model that related local temperature trends to measures of local climate (e.g., atmospheric pressure, nearness to large bodies of water, latitude) as well as economic factors (e.g., population, GDP growth rate, per capita income, national coal use) and data quality factors (e.g., amount of missing data, national literacy rate). What they found was that the climate factors, as well as socioeconomic factors, were significantly related to the observed warming trends.
The net effects of non-climatic signals in the data, McKitrick and Michaels found, has led the Intergovernmental Panel on Climate Change to overstate the overall average of warming rates in its surface data. The amount of warming resulting from those signals varies from between 26% to more than 90% of the observed warming. The range of possible bias is large because it is impossible to determine exactly the interaction between some national indicators and climate itself.
Similar to earlier work by Michaels et al. (2000), the new study showed that the most significant determinant of warming in cold, dry regions was the amount of cold, dry air, which is consistent with a greenhouse warming. But for all other seasons and moisture combinations, non-climatic factors were more important than others in accounting for observed trends.
One interpretation is that the IPCC’s claims about detecting a “greenhouse fingerprint” are largely related to winter warming in dry regions such as Siberia, and little else. Further, it is clear that the IPCC’s claim that non-climatic signals in weather station data have been identified and accounted for is highly questionable.
One of the arguments people use to support the claim that humans are causing climate change is that, regardless of how much the global average of temperatures goes up, the places where the warming is happening fits the predictions of climate models.
Some writers like to refer to the “fingerprint” of greenhouse warming, implying that if it’s caused by carbon dioxide, there will be a telltale spatial pattern, with more warming in some places than others, based on how the global atmospheric circulation responds to the greenhouse gases.
It has long been known that the vertical distribution of observed temperature change is not consistent with model projections. So it should not be surprising that the distribution through much of the year and through much of the surface moisture spectrum may also be non-climatic.
Much previous work compared the spatial pattern of warming in temperature data to the pattern generated in model predictions, where a key assumption is that the only things that can affect the pattern in temperature data are changes in solar intensity, atmospheric aerosols, greenhouse gases and natural cycles like El Nino. On that assumption, if the model says solar changes and other natural causes can’t explain the spatial pattern, but greenhouse warming can explain some of it, then the “fingerprint” of greenhouse warming has been detected.
In identifying a “net warming bias,” McKitrick and Michaels have shown is that there is another explanation, namely that economic, social and political conditions in the countries where the data are collected. Indeed, they appear to explain a large proportion of the spatial pattern of measured warming, though, as the authors note, “a more precise estimation of [the bias’s] magnitude will require further research.”
So, before anyone can claim to have “detected” a greenhouse fingerprint in the data, they must filter out the pattern of non-climatic influences on the global data set.
Many researchers claim that this has already been done, but the results in this paper show that it hasn’t. On the contrary, McKitrick and Michaels’ study confirms the recent finding of deLaat and Maurellis (2004) that the spatial pattern of surface temperature change is driven by patterns of industrialization and land-use change, not the greenhouse effect (for more information see here).
As shown in landmark work by Kalnay and Cai (2003), land-based weather station temperature data have been affected by local factors related to economic growth and land-use changes. The McKitrick and Michaels study provides further evidence that this contamination has not been removed, earlier claims not withstanding. Furthermore, it adds up to a net warming bias at the globally averaged level. Consequently IPCC surface temperature data should not be interpreted as if it measures only “climate.” Instead the temperature data reflect the influence of many things, including a complex blend of local economic and social factors.
Some of those socioeconomic factors exert an indirect influence on local temperatures but have nothing to do with the global climate, while others have nothing to do with temperature at all but instead affect data quality control. The McKitrick-Michaels study provides evidence that after controlling for these, the observed rate of temperature change is noticeably lower in a global sample, depending on how economic influences are removed, as noted above. Hence, attempts to identify the magnitude of a global greenhouse climate signal on surface data without properly removing the extraneous biases risks exaggerating the perceived influence anthropogenic greenhouse emissions.
de Laat, A.T.J., and A.N. Maurellis, 2004. Industrial CO2 emissions as a proxy for anthropogenic influence on lower tropospheric temperature trends. Geophysical Research Letters, 31, L05204, doi:10.1029/2003GL019024.
Michaels, P.J., et al., 2000. Observed warming in cold anticyclones. Climate Research, 14, 1-6.
Kalnay, E., and M. Cai, 2003. Impact of urbanization and land use change on climate. Nature, 423, 528–531.
McKitrick, R., and P.J. Michaels, 2004. A test of corrections for extraneous signals in gridded surface temperature data. Climate Research, 26, 159-173.