Peer-to-Peer Review (Part II): How 'Climategate' Marks the Maturing of a New Science Movement

Please click for Part I and Part III.

PART II – The “hockey stick” graph inadvertently incites a new camp of “lukewarmer” skeptics. Climategate files make first appearance on the internet, but were in the hands of one person days before they were made public.

If history tells us anything, it shows us that inciting an audience is an extremely precarious undertaking. Inspiring one group of followers with a call-to-action can just as easily unleash the furor of another. Arousing a community to attack an indisputable villain can surprisingly lead to a circling of the wagons by the scoundrel’s close associates. Inciting is an unpredictable endeavor, best left for those with an innate ability to read a situation or an army on-the-ready to quickly take advantage of an opening.

So is the story of the hockey stick – just as quickly as it was used as a rallying call for warmists, it also inadvertently gave birth to a camp of skeptics.


Enter Steve McIntyre, stage right. A retired mineral explorer and math scholar from Toronto, McIntyre became interested in climate science and the hockey stick due to its seeming inconsistencies. A Medieval Warming Period (seen above) had been a well-documented event in which the earth’s temperature increased considerably sometime between 1000 and 1300 AD, followed by a cooling trend known as the Little Ice Age. A graph of this cycle was even included in the IPCC First Assessment Report in 1990. These events were, however, absent from the Mann et al hockey stick graph. McIntyre was curious about how the graph was made and as chance would have it, the discipline of temperature reconstructions, largely an exercise in statistics, fit right within his mathematics wheelhouse.

Steve_McIntyreSteve McIntyre

On investigating the hockey stick, McIntyre happened upon what he viewed as some errors in the application of his field of expertise along with some misuses of data. He contacted Ross McKitrick, an environmental economist with a PhD in economics, and the two worked on a paper that would highlight the errors in the original hockey stick article.

The article, entitled Corrections to the Mann et al. (1998) Proxy Data Base and Northern Hemisphere Average Temperature Series, was published in the journal Energy and Environment in 2003.

In addition to their MM paper, as it came to be known, they performed other investigations into the work of climate scientists. To McIntyre and McKitrick’s surprise, it didn’t appear that there was much review of the work on which so much government policy was being derived. They claimed that, in addition to Mann et al adding in temperatures where there were holes in the data through a guessing method called extrapolation, they also found improper duplication of data sets and a strong cherry-picking effect that would result in a hockey stick formation no matter what type of temperature data was entered into the climate models. They input what is called “red noise”, or random data, into the model and out popped a hockey stick graph.

Eventually, the MM paper led to Mann et al publishing a modification to provide more detail to their original article. However, the scientists did not change any of their underlying results. The MM Paper also led to a congressional hearing and multiple counter studies were published, with one known as Wahl and Ammann (2006) supporting Mann et al’s position and another, known as The Wegman Report, supporting the MM paper.

To journalize his investigations into climate science, McIntyre created a blog called ClimateAuditand this is when the peer-to-peer review network first took solid form.

Others followed suit. Anthony Watts, a weather specialist, created WattsUpWithThat, Lucia Liljegren created The Blackboard, and Jeff “Id” Condon created the Air Vent. Both the contributors and commenters at these sites meticulously picked apart the work of the scientists featured in the Climategate emails – that’s how this community works.

When I asked Gavin Schmidt about whether this group has had any impact on the science, he responded, “On the science? No.” He continued, “There is an old joke about a professor reading over one of his not-so-promising student’s work: ‘This paper contains much that is novel and correct.’ [the professor] continues, ‘Unfortunately, that which is novel is not correct, and that which is correct is not novel.’ It fits the bill here.”

In response to Schmidt’s claim that the group had no impact on the science, Steve McIntyre told me that the then-Chairman of the National Academy of Science Committee on Statistics “endorsed our findings on [Mann et al].”

One of the main contention points of the peer-to-peer reviewers has been a concept termed divergence.

Some climate scientists noticed a problem with tree ring temperature reconstructions beginning around 1960. What they found was that the tree rings showed a downward trend in temperatures starting at this time while the actual thermometer readings showed a temperature increase. This is termed a divergence because the reconstructions didn’t match the actual temperature. So what the climate scientists did in some of the hockey stick graphs, according to the skeptics, was delete the tree ring data starting in 1960, replacing them with the actual temperatures. The climate scientist’s rationale was that tree-rings had a change in response to the environment over recent decades. The peer-to-peer group claimed that the climate scientists were participating in bad science. Why? Because the divergence raises the question that if the tree-ring reconstructions could not read the higher temperatures of today, how could the scientists be sure that there weren’t higher temperatures throughout the last thousand years that have also gone undetected by the tree-rings.

The spike in temperatures that are there today could also have been there one thousand years ago – inferring that the earth is just going through a natural cycle. This divergence problem appeared to show a serious problem with the science that was the underpinning of the hockey stick graphs.

It was this concern, continually highlighted by the peer-to-peer reviewers, that has brought so much perspective to the now infamous Climategate email of “Mike’s Nature trick” to “hide the decline” sent by Phil Jones. Several of the hockey stick graphs appeared to be underpinned by data that mixed apples with oranges. Even in graphs where the actual temperatures were not added, the divergence appears to be artfully hidden as seen in a hockey stick graph included in the IPCC Third Assessment Report.



Detail of above

As can be seen, the Briffa et al reconstructions (in green) come to a halt at 1960 and the end of the graph, which is trending downwards, is hidden behind others that trend up – thus providing a strong rhetorical impression of a scientific consensus of unprecedented temperature increase. The pressure applied to climate scientists to reach this consensus would be later revealed in the Climategate emails.

Another complaint by these skeptics was the lack of transparency and access to data used by the climate establishment. In repeated attempts, Steve McIntyre attempted to acquire the computational code from Michael Mann and data from Keith Briffa and Phil Jones that were used to build the hockey stick graphs. Like the US, the UK has Freedom of Information legislation that allows for individuals to request information related to government-funded projects. The peer-to-peer network requested the data directly from the University of East Anglia’s Climate Research Unit and McIntyre even approached the peer review journals that the global warming establishment used to legitimize their research, in hopes for access to the underlying data and code that the scientists were using – but all were continually met with some form of resistance.

It wasn’t until the Climategate files were leaked that we would understand the significance of these efforts.


Jeff “Id” Condon moderates a small blog called the Air Vent which is frequented by lukewarmers – a group of global warming enthusiasts that believe that CO2 does trap energy and could be warming the planet, but questions the magnitude of the problem and the certainty of climate science. Moderates is actually a loose descriptor for what Jeff does to posted comments because it was actually his lack of moderation that put his blog at ground zero of Climategate.

On Tuesday November 17th, Jeff was hunting deer, as he puts it, “unsuccessfully” in the Upper Peninsula of Michigan. On his return he had one hundred or so emails and comments that needed to be moderated before they would go live. Jeff was at his parent’s house at the time and didn’t pay much attention to the all-important comment. “What I did was read in a huge hurry, [because] I was at my folkshouse and kept being interrupted, and I read the email and posted the link. There was no cussing, it seemed to be about climate so I investigated nothing.”

Jeff then jumped on the road for the five-hour drive home.

Unknown to Jeff though, was that one of the comments was the morsel of information that would validate the peer-to-peer review network and provide clear perspective to their efforts.

The comment was from a user named FOIA, posted to the blog from a computer with a Saudi Arabian IP address. “We hereby release a random selection of correspondence, code, and documents. Hopefully it will give some insight into the science and the people behind it.” The comment linked to a file named on a Russian server.

By the time Jeff got home he noticed an email from Steven Mosher bringing attention to the content of the link. “Steve Mosher is as smart a guy as any of us will ever meet so I quickly opened and looked at a couple of [the] emails.” Understanding the implications of the files, Jeff immediately pulled down the link.

Not knowing exactly what to do next, Jeff returned to the trusty peer-to-peer review network. He emailed Steven Mosher, Lucia, Anthony Watts, and Steve McIntyre questioning what they should do about the files. Little did he know that most of this group had already downloaded them and begun investigating the content.

An hour had passed with no response, but he didn’t let the hour go to waste. With his jaw on the floor, Jeff read “one damning email after another,” getting more acclimated with the content of the files.

Now Jeff is not a lawyer, he’s an aeronautical engineer by trade, but he was concerned about the legal implications of posting personal emails. What he did notice, however, was that both Anthony Watts’ blog and Lucia had posted several of the emails. With two from the network opening the door, Jeff thought to himself, “the others have already posted on this, what is wrong with you, put it up.” He posted the article Leaked FOIA files 62 mb of gold.

The Air Vent was the little blog that could – it was not, however, the first place that the files turned up on the internet. The first place they turned up was on the very site moderated by those implicated in the Climategate files.


It was Tuesday November 17th at 6:20am EST when the Climategate files made their first known poke onto the internet. An unknown user logged into, the climate science blog of co-founders Michael Mann and Gavin Schmidt, disabled all legitimate users, then uploaded from an IP address in Turkey a zip file with the title containing the Climategate documents. Michael Mann is, of course, the father of the hockey stick graph and Gavin Schmidt is one of the top climatologists at the NASA Goddard Institute for Space Studies (GISS). This unknown user then attempted to create a draft post identical to that posted on the Air Vent, but the RealClimate admin stopped the foreign user before the post could go live.

A short while later, another user, likely the same RealClimate foreign user, posted a comment to an article on the blog of Steve McIntyre (ClimateAudit) at 7:24am EST from a Russian IP address. The comment was much more subtle than the one that would later be posted on the Air Vent. It read only, “A miracle just happened.” The user was listed as RC, and linked to the file on RealClimate. Four downloads occurred from this post before the link was made inactive by the admin. This comment had apparently gone unnoticed by Steve McIntyre, who wasn’t made aware of it until Gavin Schmidt highlighted it’s existence in a post on November 23rd.

The post on ClimateAudit was not exactly the grand announcement that was seen on the Air Vent and didn’t lead initially to an explosion of the story. Why did the promoter of the Climategate files choose such a subtle post? Did the leaker know anything about the ClimateAudit server that made him or her avoid making a stronger proclamation and drawing more traffic to the site?

The closest that you can currently get to the leaker is, intriguingly, someone from within the peer-to-peer review network.


As anyone who really researches Climategate will learn, the name Steven Mosher continues to pop up. When the story broke, it was Mosher who drew attention to the comment at the Air Vent. He was also the man that alerted Lucia of The Blackboard blog to the files. And he was the first to alert followers of ClimateAudit with a series of posts that included some of the emails.

Steven_Mosher_secretSteven Mosher

Why was Steven Mosher so ubiquitous when it came to the breaking of the Climategate story? Because Steven Mosher had the files several days before they reached the internet.


Watch for Part Three, which concludes this series, coming soon on Big Journalism.


Please let us know if you're having issues with commenting.