“This paper went through as rigorous a review as it could have received,” said Ginger Pinholster, chief of communications for the American Association for the Advancement of Science, which publishes Science. “Any suggestion that the review was ‘rushed’ is baseless and without merit.”
She said the paper, submitted to the journal in December, went through two rounds of peer review by other scientists in the field before it was accepted in May. The number of outside reviewers was larger than usual, and the time from submission to online publication was about 50 percent longer than the journal’s average of 109 days, Pinholster said.
During the review, the research was sent back to NOAA for revision and clarification, she said. And because it was based on such an “intensive” examination of global temperature data, the reviewed was handled by one of the journal’s senior editors, she said, “so it could be more carefully assessed.”
The study is widely considered to be a bombshell in the climate change world because it contradicted the notion of a “pause” in global warming and thus undercut the arguments of global warming skeptics. Among them is Rep. Lamar Smith (R-Tex.), who chairs the House Committee on Science, Space and Technology. NOAA’s data sets are used by climate scientists to take temperature measurements worldwide.
Smith has subpoenaed four top NOAA officials seeking internal e-mails and documents relating to the study, which he alleged last week in a letter to Commerce Secretary Penny Pritzker was “rushed to publication” and may have violated the agency’s scientific integrity standards. The chairman also has threatened to subpoena Pritzker, whose department includes NOAA, if she does not turn over the internal deliberations.
What makes this feud so difficult to referee is not just the complexity of climate science. Smith and his committee have yet to offer details of the allegations that the research was rushed.
Smith told Pritzker in his letter that his claim is based on information from whistleblowers who have told the committee that some employees at NOAA had concerns about the research. But committee staff members have declined to provide details about these concerns, saying that disclosing the specifics could jeopardize the panel’s sources and their anonymity.
Pressed for more details, committee aides last week pointed as an example to new temperature data that was made publicly available earlier this month and questioned how the scientists used it. The data came from a larger number of measuring stations around the world than previous data sets.
“NOAA should not be publishing headline-grabbing results based on data sets that have not been adequately vetted and were not available to the public,” an aide to the House committee said.
Smith has said the scientists were in a hurry to have their findings published because they wanted to influence Obama’s Clean Power plan, which aims to cut carbon emissions from power plants, and upcoming international negotiations in Paris on climate change.
“This isn’t an easy high school science experiment where you do it and you get results and write them up,” another aide said. “There are huge data sets from all over the world. They need to be studied. Every time the sets are changed they have to be worked on to make sure the data set is now valid.”
The scientists, however, say their research was based on an earlier version of the data that had been made public and examined by other climate experts. The study published in Science was not based on the updated data released earlier this month, although the two versions are very similar, according to NOAA officials and one of the study’s authors.
That author, Thomas Peterson, described in an interview some of the internal tensions at NOAA between the scientists and computer engineers who were writing software code for the data and wanted more time to make sure it was reliable. The scientists felt confident using the data that had already been made public and were ultimately vindicated by the latest version.
The conclusions of the Science paper were based on corrections and adjustments to even earlier land and sea temperature measurements. These were intended to address what scientists described as measurement biases in readings taken of ocean temperatures and land temperatures that did not fully account for the rapidly warming Arctic.
NOAA published the first updates to the land temperature data set in October 2013 in the Geoscience Data Journal. The revised sea surface data were published in the Journal of Climate in October 2014. These updates were the basis of the study in Science, NOAA officials said.
That combined data set was available publicly in July 2014, officials said.
As NOAA scientists examined the data, they discovered that warming trends over the past few decades would be substantially larger than what the earlier data set indicated, recalled Peterson, who retired from NOAA as principal scientist in July.
“Was there a rush to get [the research] out? No,” he said. “Did we want to get this out to advance the science? Of course.”
Peterson acknowledged that tensions over timing developed between the scientists and a team of computer engineers — some contractors, some civil servants — who were rewriting the software code to process the data once it was collected from stations worldwide. The engineers wanted more time to test and retest the software to ensure its reliability, he recalled. The scientists argued that it was taking too long.
“We’re talking about a time lag of years between the science and when they thought the software testing would be ready because of this question of whether one piece of software might develop a glitch, ” said Peterson, now president of the World Meteorological Organization’s Commission for Climatology.
To accommodate the engineers, he said, the submission to GeoScience was delayed by six months.
Still, the engineers were frustrated that the scientists were pressing ahead with research based on the data. “They viewed it as putting the cart before the horse,” Peterson said. “Those of us, like myself, who knew the importance of getting this work out literally years ago were very frustrated by the process … that slowed the release down.”
Peterson stressed that the scientific analysis itself was not rushed. “Indeed just the opposite is true,” he said.
Peterson and NOAA officials said the scientists decided to submit their research to Science before the most recent data set was ready for release this month. They said the new version, which contains even more data than the set used for the Science study, confirms its findings.