by Judith Curry
Here is further explanation why I think Michael’s testimony is significant, and why I think the issue of the attribution since 1950 will be the battleground in U.S. CO2 policy. Michaell’s stated purpose for conducting this analysis was:
demonstration that the Finding of Endangerment from greenhouse gases by the Environmental Protection Agency is based upon a very dubious and critical assumption.
Michaels’ is seeking to establish reasonable doubt to the EPA’s CO2 endangerment finding, which is based on the statement (very similar to the IPCC’s statement):
Most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic GHG [greenhouse gas] concentrations.
There are two main elements to Michael’s argument regarding the trend in global surface temperature anomalies and their attribution: the global surface temperature anomaly data record 1950-2009, and forcing. The analysis focuses on new research since the IPCC AR4:
1a. Adjustments to the global surface temperature anomalies prior to 1965 (following Thompson et al. 2008) reduces the total temperature increase by 0.15C, and the trend by about 20%. Thompson et al. (2008) state:
The adjustments immediately after 1945 are expected to be as large as those made to the pre-war data (,0.3C; Fig. 4), and smaller adjustments are likely to be required in SSTs through at least the mid-1960s.
While it is my understanding that this temperature correction has not yet been applied to the CRU data set, Michaels’ application of this seems consistent with what Thompson et al. recommend. Thompson et al. state that:
Corrections for the discontinuity are expected to alter the character of mid-twentieth century temperature variability but not estimates of the century- long trend in global-mean temperatures.
The net effect of this correction is to make the mid century trend more continuous and reduce the appearance of the 1940’s “bump” that was followed by cooling. This also serves to increase the average global temperature ca. 1950 (the beginning point of the mid century in the attribution statements.)
1b. The second point considers “non climatic” trends over land associated with data quality and land use changes (McKitrick and Michaels, 2007), which they argue account for 0.08C of the global warming trend from 1980-2002 (which is completely independent of the ocean adjustment.)
If these two temperature corrections are correct, then the decadal rate of change in the period 1950-2010 is now probably slightly less than the decadal rate of change in the period 1910-1940 (which is unchanged in the Thompson et al. analysis.) Further, it is in principle easier to explain a smaller rate of temperature increase due to natural variability.
JC comments: I have no idea whether these adjustments to the temperature record are correct, but they certainly reflect the overall uncertainty in the data. This analysis indicates a 33% discrepancy in the size of the trend, which reflects uncertainty in data itself. The actual uncertainty, if a comprehensive error analysis was done, is possibly larger than this. Note, errors in surface temperature data will be subject of a future series.
2. The second part of the argument is forcing, and Michaels only includes two issues: stratospheric water vapor forcing, and forcing from black carbon (two factors that are almost certainly independent of each other.)
Here is where Michael’s argument becomes confusing. Michaels’ attempts to explain the 0.7C trend by saying the observations are wrong and the trend is less, and then finds a residual trend (reduced further by stratospheric water vapor and black carbon) to claim that he has explained more than half of the 0.7C trend without CO2.
Here is how I think he should proceed with his argument: If the observed temperature increase between 1950 and 2009 is 0.468C (trend 0.078C), then the challenge is to explain more than half of that trend with natural forcings. According to Michael’s analysis, black carbon and stratospheric H2O account for 34% of the 0.468C trend. So technically, Michael’s argument has not refuted the foundation of EPA’s endangerment finding (more than half of the observed warming, the magnitude of which has now been reduced). Adding uncertainty associated with solar variability may possibly make his argument work, but the argument as it stands, doesn’t hold up in my opinion (and not for the reason that Santer gave, in terms of including sulfate.) On the other hand, the original concern was raised over the magnitude of the original warming (0.708C), so Michaels’ broader argument does raise reasonable doubt (and would we be so worried if the observed trend was 0.47?)
IMO, the more significant thing that Michaels did was in adjusting the surface temperature time series, which may result in the rate of warming in the latter half of the 20th century being smaller than that between 1910-1940 (somebody needs to do the calculations, I don’t have time right now.)
In any event, I think this overall line of argument presented by Michaels is a very significant one in terms of the EPA CO2 endangerment issue. However, the logic of the argument needs refining and it needs extending before lawyers can use this as “reasonable doubt” in challenging the EPA endangerment ruling. And those defending the science the behind the EPA endangerment ruling (which is basically the IPCC) need to shore up their arguments. I think that this is the coming battleground issue in U.S. policy on this topic.