Spencer & Braswell 2011: Proof that global warming is exaggerated? Or just bad science?

This guest post by Michael Ashley (which includes opinions from the Climate Science Rapid Response Team matchmakers) examines the peer-reviewed article by Spencer & Braswell, published on 25 July 2011, and reported in the press as “blow[ing] a gaping hole in global warming alarmism”.

If true, this would make the article one of the most important published in the last decade.

Background

There are very few articles published in the peer-reviewed scientific literature that challenge the consensus that human-induced climate change is real and significant.

None have stood the test of time.

A new article by Spencer & Braswell is following the common trajectory of many such papers:

1. The article is published in a non-mainstream journal, following inadequate peer-review.
2. Press releases from the authors exaggerate/distort the contents of the article to inflate its significance and increase the attention given to it.
3. News of the article spreads like wild-fire around the blogosphere.
4. Some media outlets take the press release and exaggerate it further still, so that the information that finally reaches the public has almost no relation to the original article.
5. Within days, experts in the field show that the original article is fatally flawed; but by now the damage is done.
6. For years into the future, the article is quoted by deniers of human-induced climate change as evidence that the science is uncertain.

The publication of Spencer & Braswell 2011

On 25 July 2011 the journal Remote Sensing published an article entitled “On the Misdiagnosis of Surface Temperature Feedbacks from Variations in Earth’s Radiant Energy Balance” by Roy Spencer and William Braswell, both from the University of Alabama in Huntsville. The article has since attracted considerable attention in some media outlets, with various claims being made, including that it “blow[s] a gaping hole in global warming alarmism”.

What does Spencer & Braswell 2011 actually show?

The article compares measurements of air temperature with the energy that the earth is radiating into space.

It does this using data for the ten year period 2000–2010.

The article goes on to examine IPCC models over the 100 year period 1900–1999, and finds a different relation between air temperature and energy radiated. The article draws various conclusions from these differences.

Why is this interesting?

Any difference between observations and models could point to problems with either the observations or the models, so it is worth investigating.

Have Spencer & Braswell found a significant difference between observations and the IPCC models?

No. Their article contains a number of errors that have since been identified by climate scientists. These errors range from the trivial (using the wrong units for the radiative flux anomaly), to the serious (treating clouds as the cause of climate change, rather than resulting from day-to-day weather; comparing a 10 year observational period with a 100 year model period and not allowing for the spread in model outputs).

Within three days of the publication of Spencer & Braswell 2011, two climate scientists (Kevin Trenberth & John Fasullo) repeated the analysis and showed that the IPCC models are in agreement with the observations, thus refuting Spencer & Braswell’s claims. An independent analysis by Andrew Dessler also confirms the Trenberth & Fasullo result.

Furthermore, Trenberth and Fasullo showed that the better-performing IPCC models were distinguished by their ability to track the El Niño-Southern Oscillation, not by their climate sensitivity as claimed by Spencer & Braswell.

In other words, there is no evidence from the 10 years of satellite data that forecasts of global warming are too high. There are additional problems with the article, but these new analyses are sufficient to invalidate the conclusions made by Spencer & Braswell.

So, the outcome is?

Spencer & Braswell 2011 is a flawed article that unfortunately adds nothing to our understanding of climate science. In fact, the article is misleading and detracts from the real state of knowledge.

What about headlines such as “New NASA Data Blow Gaping Hole In Global Warming Alarmism”?

These headlines bear no relation to the actual contents of Spencer & Braswell’s article.

The University of Alabama press release

Coincident with the publication of the article, the University of Alabama issued a press release in which Spencer made erroneous claims as to what the article shows. For example, Spencer says “The result is climate forecasts that are warming substantially faster than the atmosphere”; this is not mentioned anywhere in the abstract, discussion, or conclusion of the article.

What is motivating Spencer & Braswell? Why are they publishing flawed science?

A charitable interpretation of the Spencer & Braswell article is that it is simply an unfortunate mistake, and that the authors will learn from this experience and write better articles in the future. However, there are a number of factors that make this interpretation unlikely to be correct.

Foremost is that Spencer has disagreed that flaws have been found in the article. This fits an unfortunate pattern with his research where flaws have been found and not acknowledged. It is common knowledge in the climate science community that Spencer’s work is often wrong. Some of these errors show a fundamental lack of understanding of some of the most elementary concepts used when comparing models and data. See for example http://bbickmore.wordpress.com/2011/07/26/just-put-the-model-down-roy/.

Spencer is also well known for his erroneous articles with John Christy in the 1990’s that purported to show that the troposphere was warming more slowly than other groups had found, once again supporting Spencer’s belief that global warming is exaggerated—it took nearly 13 years for scientists to identify the faults in Spencer and Christy’s calibration of the satellite data. For further information on Spencer’s long history of mistakes, see http://www.skepticalscience.com/search.php?Search=spencer.

Another clear indication that there is something amiss with the Spencer & Braswell article is that the authors didn’t publish it in a recognized journal in the field. They selected a journal that was not one of the more than 8300 science journals covered by the Web of Science, the usual reference that scientists use for following the literature. Moreover, the journal’s field is only tangentially-related to climate science, and has an editorial board without the expertise to properly evaluate the article. The editors would likely not have had access to a list of appropriate expert referees to comment on the article prior to publication. The errors in the article are so clear that any expert in the field would have found them. The fact that the errors were not corrected prior to publication argues that the article was not adequately refereed, which in turn reflects poorly on the journal.

A final disturbing aspect of the Spencer & Braswell article is that, as we have seen, the associated press release makes claims that are unsupported by the contents of the article.

Spencer & Braswell is political propaganda, not good science

Most academic scientists employed at universities would regard their jobs as a combination of trying to do the best science research that they can, and teaching/mentoring/inspiring their students.

Spencer, on the other hand, has said “I view my job a little like a legislator, supported by the taxpayer, to protect the interests of the taxpayer and to minimize the role of government.”

In the light of this statement, the Spencer & Braswell article can be understood as being motivated by a political agenda, and not for the advancement of science.

For expert commentary on the science, please consult Kevin Trenberth, John Fasullo, and/or Andrew Dessler.

1 August 2011