On 4 February 2017, the British tabloid Mail on Sunday (and the Daily Mail‘s online site) published an article by David Rose — a longtime proponent of climate change conspiracy theories whose analyses the scientific community widely regards as flawed and deceptive — alleging that scientists at the National Oceanic and Atmospheric Administration (NOAA) used misleading data in order to rush publication of a groundbreaking climate study and thereby “dupe” world leaders:
The Mail on Sunday today reveals astonishing evidence that the organisation that is the world’s leading source of climate data rushed to publish a landmark paper that exaggerated global warming and was timed to influence the historic Paris Agreement on climate change.
A high-level whistleblower has told this newspaper that America’s National Oceanic and Atmospheric Administration (NOAA) breached its own rules on scientific integrity when it published the sensational but flawed report, aimed at making the maximum possible impact on world leaders including Barack Obama and David Cameron at the UN climate conference in Paris in 2015.
At issue is a June 2015 paper published in Science titled “‘Possible artifacts of data biases in the recent global surface warming hiatus,” authored by (among others) Thomas R. Karl, director of the NOAA’s National Centers for Environmental Information.
As described in an analysis accompanying the piece in Science, the paper dealt with an issue that deniers of anthropogenic climate change have rallied behind for years — that there has been a slowdown or “hiatus” in global warming since 1998 despite continually rising CO2 levels:
Previous analyses of global temperature trends during the first decade of the 21st century seemed to indicate that warming had stalled. This allowed critics of the idea of global warming to claim that concern about climate change was misplaced. Karl et al. now show that temperatures did not plateau as thought and that the supposed warming “hiatus” is just an artifact of earlier analyses. Warming has continued at a pace similar to that of the last half of the 20th century, and the slowdown was just an illusion.
Because of its direct attack on a talking point popular with climate change skeptics, the study has been a common target of those seeking to undermine climate science. In this latest round of controversy, the Daily Mail article cited a “whistleblower” (retired NOAA scientist John Bates) who came forward to reveal what he described as ethical lapses in the way that study’s data were selected and archived by its authors.
The allegations made against Karl and his co-authors fall into three general groups:
1) That the team unethically selected flawed data that was supportive of their cause.
2) That they failed to archive their data as required by both NOAA and the journal Science.
3) That they rushed through the required internal review process to get the paper published before the 2015 Paris Climate Summit.
First, and most “astonishing,” were claims that the data used in Karl’s paper were manipulated to derive a specific result, according to Rose’s Daily Mail report:
The sea dataset used by Thomas Karl and his colleagues — known as Extended Reconstructed Sea Surface Temperatures version 4, or ERSSTv4, tripled the warming trend over the sea during the years 2000 to 2014 from just 0.036C per decade — as stated in version 3 — to 0.099C per decade …
But Dr Bates said this increase in temperatures was achieved by dubious means. Its key error was an upwards ‘adjustment’ of readings from fixed and floating buoys, which are generally reliable, to bring them into line with readings from a much more doubtful source — water taken in by ships. This, Dr Bates explained, has long been known to be questionable: ships are themselves sources of heat, readings will vary from ship to ship, and the depth of water intake will vary according to how heavily a ship is laden — so affecting temperature readings.
Here, Rose offered two implications. The first was that Karl’s methodology was intentionally opaque in order to hide the fact that it went against a well-known ship measurement bias. In reality, the entire methodology was spelled out in the paper, and the ship data correction Karl et al selected had previously been published. This correction, far from ignoring the differences between boat and buoy data, actually took into account the generally superior buoy data in its calculation:
Several studies have examined the differences between buoy- and ship-based data, noting that the ship data are systematically warmer than the buoy data. This is particularly important because much of the sea surface is now sampled by both observing systems, and surface-drifting and moored buoys have increased the overall global coverage by up to 15%. These changes have resulted in a time-dependent bias in the global SST record, and various corrections have been developed to account for the bias.
Recently, a new correction [Huang et al 2015] was developed and applied in the Extended Reconstructed Sea Surface Temperature (ERSST) data set version 4, which we used in our analysis. In essence, the bias correction involved calculating the average difference between collocated buoy and ship SSTs. The average difference globally was −0.12°C, a correction that is applied to the buoy SSTs at every grid cell in ERSST version 4. […] More generally, buoy data have been proven to be more accurate and reliable than ship data, with better-known instrument characteristics and automated sampling. Therefore, ERSST version 4 also considers this smaller buoy uncertainty in the reconstruction.
Conspicuously absent from Rose’s article was any mention of a 4 January 2017 study that critically investigated the choices referenced above, demonstrating that the record made by Karl et al tracked the buoy data and other modern sources of data more accurately than any other model. The lead author of that more recent study discussed those findings in response to the Daily Mail article:
I recently led a team of researchers that evaluated NOAA’s updates to their ocean temperature record. In a paper published last month in the journal Science Advances, we compared the old NOAA record and the new NOAA record to independent instrumentally homogenous records created from buoys, satellite radiometers, and Argo floats. Our results, as you can see in the chart below, show that the new NOAA record agrees quite well with all of these, while the old NOAA record shows much less warming.
This was due to two factors: the old NOAA record spliced together warmer ship data with colder buoy data without accounting for the offset between the two; and the new NOAA record puts more weight on higher-quality buoy records and less weight on ship records (versus the old NOAA record which treated ships and buoys equally) …
The fact that the new NOAA record [Karl et al 2015] is effectively identical with records constructed only from higher quality instruments (buoys, satellite radiometers, and Argo floats) strongly suggests that NOAA got it right and that we have been underestimating ocean warming in recent years.
The second implication, that Karl et al intentionally selected data and corrections with the intent of exaggerating global warming, was a serious charge based only on Bates’ testimony. But in a 7 February 2016 interview with EE News, Bates explicitly stated that he did not mean to suggest that Karl et al had manipulated data:
Bates accused former colleagues of rushing their research to publication, in defiance of agency protocol. He specified that he did not believe that they manipulated the data upon which the research relied in any way.
“The issue here is not an issue of tampering with data, but rather really of timing of a release of a paper that had not properly disclosed everything it was,” he said.
With respect to the team’s use of this new correction, Rose reported that “Dr Bates said he gave the paper’s co-authors ‘a hard time’ about this, ‘and they never really justified what they were doing.”
However, Karl told us via e-mail that Bates’ only role in the paper was to organize the internal review process, during which time he says no such conversations regarding their use of ERSSTv4 ever took place:
[There were] no discussions, nor emails to me. Dr. Bates was asked to coordinate the internal review of the paper (I am not sure who made the request) since this was normally handled by Tom Peterson, but he was one of the authors. The responsibility of the coordinator for the internal review is to identify an individual within the [National Centers for Environmental Information] who could review the paper and pass those comments back to the authors for their response. In our case, there were no comments by the reviewer.
(Neither John Bates nor David Rose returned our requests for comments.)
Rose further alleged that Karl et al used a particular land-temperature dataset because it demonstrated findings more amenable to their cause, even though (according to Rose) that dataset was both too preliminary to be approved and was plagued by “devastating” computer bugs:
The second dataset used by the [Karl et al.] paper was a new version of NOAA’s land records, known as the Global Historical Climatology Network (GHCN) […]. This new version found past temperatures had been cooler than previously thought, and recent ones higher — so that the warming trend looked steeper. For the period 2000 to 2014, the paper increased the rate of warming on land from 0.15C to 0.164C per decade.
In the weeks after the [Karl et al.] paper was published, Dr Bates conducted a one-man investigation into this. His findings were extraordinary. Not only had Mr Karl [sic] and his colleagues failed to follow any of the formal procedures required to approve and archive their data, they had used a ‘highly experimental early run’ of a programme that tried to combine two previously separate sets of records …
As discussed above, Bates has asserted that he did not mean to suggest Karl and his colleagues manipulated the data by selecting (as Rose implied) a new buggy model in order to exaggerate the records. Instead, Bates said that he simply felt it improper to use that data without a disclaimer clearly indicating that it was derived from an experimental run, as discussed in a blog post authored by Bates:
Clearly labeling the dataset would have indicated this was a highly experimental early GHCN-M version 4 run rather than a routine, operational update. As such, according to NOAA scientific integrity guidelines, it would have required a disclaimer not to use the dataset for routine monitoring.
Bates, it should be noted, was the lead on a massive NOAA effort that aimed to overhaul procedures for the standardization of which datasets to use, as described in that same blog post:
As a climate scientist formerly responsible for NOAA’s climate archive, the most critical issue in archival of climate data is actually scientists who are unwilling to formally archive and document their data. I spent the last decade cajoling climate scientists to archive their data and fully document the datasets. I established a climate data records program that was awarded a U.S. Department of Commerce Gold Medal in 2014 for visionary work in the acquisition, production, and preservation of climate data records (CDRs), which accurately describe the Earth’s changing environment.
In that post, as well as in the Daily Mail piece, Bates made the claim that the use of the more experimental dataset by Karl et al contradicted NOAA policy because the new dataset had not undergone an “operational readiness review” (ORR). He also alleged that the use of this data set, and a computer failure, resulted in no record being created of what the paper’s authors did, putting that paper in conflict with both Science’s editorial standards and NOAA’s internal standards — a point Rose brought up multiple times:
The lack of archival of the GHCN-M V3.X [their experimental version of the land temperature data] and the global merged product is also in violation of Science policy on making data available. This policy states: “Climate data. Data should be archived in the NOAA climate repository or other public databases”. Did Karl et al. disclose to Science Magazine that they would not be following the NOAA archive policy, would not archive the data, and would only provide access to a non-machine readable version only on an FTP server?
Zeke Hausfather, the lead author on a 2017 study that looked into Karl et al’s use of this data, disputed Rose’s characterization:
In his [Daily Mail] article, David Rose relies on reports from a researcher at NOAA who was unhappy about the data archiving associated with the Karl et al paper. While I cannot speak to how well the authors followed internal protocols, they did release their temperature anomalies, spatially gridded data land and ocean data, and the land station data associated with their analysis. They put all of this up on NOAA’s FTP site in early June 2015, at the time that the Karl et al paper was published.
As someone who works on and develops surface temperature records, the data they provided would be sufficient for me to examine their analysis in detail and see how it compared to other groups. In fact, I used the data they provided shortly after the paper was published to do just that. While it would have been nice for them to publish their full analysis code online as well as the data, I’m sure they would have provided it to any researchers who asked.
For what it’s worth, Karl told us that he has no knowledge of a computer failure that wiped out critical information, saying that “This allegedly happened after I retired, but I have been informed that is simply not true.”
Indeed, information about the crash appears to have been secondhand information which Bates passed on to Rose, according to his blog post:
I later learned that the computer used to process the software had suffered a complete failure, leading to a tongue-in-cheek joke by some who had worked on it that the failure was deliberate to ensure the result could never be replicated.
The final class of allegations, that these missteps were accepted specifically to influence the Paris Climate Summit, relied (as did everything else) on the testimony of Bates, who is quoted in the Daily Mail piece as claiming that Karl “rushed so that he could time publication to influence national and international deliberations on climate policy.”
As reported by EENews, this issue is likely moot, as the State Department indicated that the NOAA study had little to no bearing on the outcome of the Paris meeting:
Whether the research was published to influence the Paris climate talks is a moot point, said Andrew Light, a senior member of the State Department’s climate talks negotiating team in 2015. He said the talks had already been underway for about four years when the paper was published and that 188 nations were relying on a tremendous amount of research to support their goal of reducing humans’ carbon emissions to slow the warming of the planet. They had also already crafted proposed reductions by the time the research was published, he said.
“I never heard it discussed once, let alone this one NOAA report, discussed in Paris, the run-up to Paris or anything after Paris, so this is really just an incredibly bizarre claim,” Light said.
All these claims of malfeasance used the testimony of a single scientist, upset with how more senior scientists dealt with data acquisition and archiving, to paint an excruciatingly technical matter as a worldwide conspiracy. Rose made no effort to contact Karl or other members of the team (according to Karl), and outside of Bates’ conversation with him, Rose provided no corroborating evidence or relevant background for his assertions.
Despite these discrepancies, the story gained additional traction when Texas representative Lamar Smith, chair of the U.S. House Committee on Science, Space, and Technology, tweeted the flawed narrative and issued a press release about it:
NOAA sr officials played fast & loose w/data in order 2 meet politically predetermined conclusion on climate change
While Karl et al might reasonably be criticized for having been less than rigorous in their data documentation, their findings have been independently verified, contrary to allegations that the authors manipulated data to reach a desired conclusion:
What David Rose fails to mention is that the new NOAA results have been validated by independent data from satellites, buoys and Argo floats and that many other independent groups, including Berkeley Earth and the UK’s Met Office Hadley Centre, get effectively the same results.
Rose’s claim that NOAA’s results “can never be verified” is patently incorrect, as we just published a paper independently verifying the most important part of NOAA’s results.