I have been contacted by a whistleblower with a remarkable story of corruption of the academic peer-review process involving a paper published in 2022. The whistleblower has provided me with relevant emails, reviews and internal deliberations from which I recount this disturbing episode — which ends with an unwarranted and politically-motivated retraction of a paper that some climate scientists happened to disagree with.
The paper at the center of this story is not particularly significant, as it mainly reviews the conclusions of the Intergovernmental Panel on Climate Change (IPCC) on trends in weather extremes. The paper does venture a bit too far (in my view) into commentary, but that is neither unique nor a basis for retracting a paper – if it were we’d have a lot of retractions!
To be clear, there is absolutely no allegation of research fraud or misconduct here, just simple disagreement. Instead of countering arguments and evidence via the peer reviewed literature, activist scientists teamed up with activist journalists to pressure a publisher – Springer Nature, perhaps the world’s most important scientific publisher – to retract a paper. Sadly, the pressure campaign worked.
The abuse of the peer review process documented here is remarkable and stands as a warning that climate science is as deeply politicized as ever with scientists willing to exert influence on the publication process both out in the open and behind the scenes.
I have contacted the publisher and the co-chief-editor of the journal with several questions (which you can find at the bottom), and a request for a reply by close-of-business today. It is now after 7PM in Europe, where both are based, and I have not received a response. My invitation for comment remains open and I will update this article should they respond.
In January 2022, the European Physical Journal Plus (EPJP), a peer-reviewed academic journal under the Springer Nature umbrella of journals, published a paper titled, “A critical assessment of extreme events trends in times of global warming,” by Gianluca Alimonti and colleagues. The paper reviews trends in various extreme events and disasters, drawing heavily on IPCC reports.
Following its publication, the paper was discussed a bit on several blogs but did not get much attention (I Tweeted on it at the time). Then, eight months later following some discussion of the paper in the Australian media, The Guardian wrote an article severely criticizing the paper. The Guardian quoted four scientists critical of the paper: Greg Holland, Lisa Alexander, Steve Sherwood, and Michael Mann.
Michael Mann was scathing and personal in his comments:
“another example of scientists from totally unrelated fields coming in and naively applying inappropriate methods to data they don’t understand. Either the consensus of the world’s climate experts that climate change is causing a very clear increase in many types of weather extremes is wrong, or a couple of nuclear physics dudes in Italy are wrong.”
Less than a week later, the AFP followed up with an article also critical of the paper, with the headline, “Scientists urge top publisher to withdraw faulty climate study.”
The AFP quoted two scientists calling for the paper to be retracted by Springer Nature. One was Friederike Otto, of the Grantham Institute for Climate Change and the Environment, who stated:
“They are writing this article in bad faith. If the journal cares about science they should withdraw it loudly and publicly, saying that it should not have been published.”
The other scientist calling for the paper’s retraction was Stefan Rahmstorf, Head of Earth Systems at the Potsdam Institute for Climate Impact Research who stated,
“I do not know this journal, but if it is a self-respecting one it should withdraw the article”
Two other scientists quested by the AFP, Peter Cox and Richard Betts, both of the University of Exeter, explicitly opposed the idea of retraction. Cox noted that retraction could “lead to further publicity and could be presented as censorship.”
Two days later, on 29 September 2022, Christian Caron of Springer Nature and the editorial manager of the Italian Physical Society, Barbara Ancarani (and why she was involved is unclear), contacted Alimonti et al. to let them know that based on the two media stories, an investigation had been opened of their paper, cc’ing EPJP co-editor-in-chief, Beatrice Fraboni:
Dear Prof. Alimoti,
We are contacting you today regarding your article
Included in these reports are numerous concerns of scientists who are considered highly expert in this subject.
As a result of these circumstances it is now necessary that the journal carry out an investigation to assess the validity of these concerns, in line with good practice when concerns of this type are brought to a journal. An editorial note on the homepage of the above mentioned article will be added stating:
Readers are alerted that the conclusions reported in this manuscript are currently under dispute.
The journal is investigating the issue.
We will be contacting you shortly to provide further details of next steps.
Thank you for your cooperation on this matter.
Barbara Ancarani (SIF) & Christian Caron (Springer Nature)
cc Editor-in-Chief of EPJ Plus (Prof. Beatrice Fraboni)
The next day, 30 September 2022, Fraboni, the EPJP co-chief-editor, contacted the associate editor responsible for handling the review process of Alimonti et al., Jozef Ongena.
Fraboni explained that the journal’s publisher, Springer Nature, wanted action:
“. . . we are facing some issues with a paper in your area. The publishers have asked the Editors to take action.”
Ongena responded immediately:
“The article has undergone the usual peer review. There should be no blame and shame… Peer reviewing is the common practice. That there is a discussion seems not abnormal and seems a very healthy thing.”
Ongena followed up with a second email with a proposal:
“I would invite the colleagues that have objections to send in their objections and to pass them on to the authors. To start a discussion in the press as they already did is certainly worse than publishing a critical paper. They could later also be invited to publish a comment.
We should as a journal not refrain or be afraid from a scientific discussion, but it should be in a correct way.”
Alimonti responded to the publishers email (Ancarani/Caron above) on 4 Oct 2022 with a similar proposal to that made by Ongena — specifically to engage the different views in the literature, as is normally done in science:
“Dear Dr. Caron,
after confronting with the other authors, we believe a possible correct way to criticize a scientific paper would be to write a detailed summary about what is supposed to be not correct and complete it with references; in other words a paper with precise counter arguments or at least a
detailed report that should be sent to the journal where the original paper has been published; at this point the authors of the criticized paper may give detailed answers and the journal may decide further steps.
Have Springer or EPJP been somehow formally contacted with a detailed counter analysis? If so,
please forward us any comment so that we can properly answer; if not, we believe that considering “under discussion” a scientific paper that underwent a peer review process just on the basis of interviews appeared on online newspapers or blogs, even if authoritative, is not what a scientific method requires.”
Alimonti continued and took particular issue with the comment of Michael Mann about one of his co-authors:
“Prof. Prodi, a distinguished climatologist, not just “a nuclear physics dude”, reminds me that he also served as Editor of Springer for many years: criticizing him as author would be a critic to Springer in selecting reviewers and editors. The Publisher should defend its scientific integrity in a resolute way, in order not to lose prestige itself, by moving at the request of newspapers or by denying its role.”
The co-chief-editor of EPJP, Fabroni, initially appears to have accepted this proposed course of action on 9 Oct 2022:
“After having received various feedbacks we have decided to contact the colleagues who expressed concern on the paper to provide a scientific comment that we will then send out to independent reviewers. If and when the Comment will be approved by them, we will share it with the authors so that they will be able to address the issues raised. Also their reply will be peer-reviewed.”
The eight “colleagues who expressed concern” via the media (and listed above) all apparently chose not to provide a scientific comment on Alimonti et al. and no further discussion of the comment was made in subsequent correspondence that I have seen.
However, the investigation proceeded.
On 17 Nov 2022, after Alimonti emailed Fabroni to ask for an update on the investigation, Fabroni responded to Alimonti et al. that she had consulted with the scientists who had criticized the paper in the two media stories. She noted that her reply to Alimonti was crafted with assistance from Springer:
“The reply has been drafted with the assistance of the Springer Research Integrity Department, after carefully taking into consideration the feedbacks received from the colleagues who criticised the paper in the media”
Fabroni’s reply to Alimonti stated:
“Thank you very much for your patience – we have analyzed the case now in-depth. While we acknowledge that the media coverage has certainly made the case temporarily bigger than necessary, it has also uncovered a clear weakness of your paper that we believe must eventually be addressed.”
The “clear weakness” was described by Fabroni as a “main criticism” and it was that the paper did not reference the IPCC AR6, which had not been published at the time that Alimonti et al. was written, reviewed or published.
“Indeed, a main criticism is that your paper refers essentially only to the Intergovernmental Panel on Climate Change AR5 of 2013. . . “
Alimonti et al. were given a choice to prepare an “erratum” or not. If they chose to prepare an “erratum” then Fraboni asked that it discuss floods, drought and heavy precipitation, drawing on relevant sections if the IPCC AR6 Working Group 1. Here are how the choices were presented to Alimonti:
1) You will submit an Erratum taking the final, published version of AR6 into account, where the above criticism is explicitly addressed and any conclusion that needs to be revised will be detailed. This Erratum paper, where we expect ample references to the published AR6, will be thoroughly assessed by also involving scientists from the cited parts of AR6.
The Erratum has to be submitted before Dec 31st, 2022.
2) If you decide not to submit such an Erratum or the Erratum is not submitted by the above deadline, the journal will publish an Editorial where we summarize our findings, very much as outlined above and the present Editorial Note on your article will be changed to a permanent Editorial Expression of Concern that will refer to this Editorial.
Alimonti responded six days later, on 23 November 2022, noting that it should be an “Addenda” to their paper and not an “Erratum”:
we thank you for your message where you acknowledge that “the media coverage has certainly made the case temporarily bigger than necessary”.
Since the main request arising from your message is to write a paper with ample references to a document that was not referenceable at the time our article was submitted, we believe an Addenda from our side is the most appropriate paper that would answer your request, as specified in
Alimonti et al. then prepared an Addendum to their paper, which the journal sent out for review.
There were four total reviews, two were solicited by Fabroni (who I call Reviewers 1 and 2 below) and two from Ongena (Reviewers 3 and 4). A fifth individual was enlisted by Fraboni as an “Adjudicator.” The Adjudicator apparently had access to only the reviews of Reviewer 1 and 2.
Of the four reviews of the Addendum, three recommended publication and offered minor suggestions. Reviewer 1 opposed publication. The Adjudicator agreed with Reviewer 1. Here are excerpts from all four reviews and the Adjudicator’s summary judgment.
Reviewer 1 wrote:
“I totally agree that the origin of the large increase in the number of weather and climate related catastrophic events is largely due to increased exposure and vulnerability connected to demographic and economic growth rather than to climate change”
Reviewer 1 also notes:
“In this manuscript, as well as in [Alimonti et al.], it is stressed that despite the existence of detectable trends in mean variables, in most cases no trends in extreme events exist. I understand what the authors mean, but caution has to be paid to the exact wording of the sentences. . . Detecting trends in extreme events is much more difficult than detecting trends in mean variables. Clearly, the limited amount of data for extreme events makes much more difficult to detect changes in a statistically significant way. The large interannual variability of extreme event statistics means that even if changes are present, the limited amount of data at our disposal makes them undetectable for long times.”
Reviewer 1 concludes:
“in my opinion the manuscript cannot be published”
Reviewer 2 recommended “Accept as it” and noted:
“The statements made by the authors are generally in agreement with the assessment produced by the working group 1 of the IPCC in their Sixth Assessment Report (AR6).”
The journal sent the comments of Reviewer 1 to Alimonti et al. who then revised their Addendum and responded to EPJP with the revision and response to reviews. Reviewer 1 (and apparently inly Reviewer 1) was sent the revised Addendum. Reviewer 1 re-reviewed the Addendum and responded to the journal with the following conclusion (my emphasis added):
“In summary, the claims in the addendum are correct (and in line with cherry-picked statements in IPCC AR6 and in selected publications), but they are presented in a way that does not give the full picture. Especially considering that typical readers of EPJP journal are not climate experts, I think editors should seriously consider the implications of the possible publication of this addendum.”
What implications?
Meantime, the original paper’s original editor, Ongena, also solicited two further reviews of the Addendum, which he shared with Fabroni.
Reviewer 3 wrote:
“The original article is a straightforward recitation of credible, key data about several types of extreme weather events. I find nothing selective, biased, or misleading in what they present. While there’s hardly anything written that isn’t well-known to experts, it’s useful for non-experts to see the underlying data, which are most often obscure in the IPCC reports. . .
The addendum is an on-point discussion of the extent to which the original paper agrees with the IPCC on three types of extremes. The document is up to professional standards -specific, detailed, and with citations.”
Reviewer 4 wrote:
“The most important contribution of the authors is to look further back into the climate record (including early 20th century), when many types of extreme events were comparable to today. The paper doesn’t specifically focus on the attribution (cause) of any trend (or lack thereof).
I don’t see any grounds for criticizing this work. Further, most of their conclusions are supported by the IPCC AR6 WG1.”
Fabroni invited a fifth person to serve as an “Adjudicator” of the reviews of the Addendum and revised Addendum. Apparently the Adjudicator was provided only the reviews of Reviewer 1 and 2.
The Adjudicator began their report by discussing his/her views of the original paper, noting that such commentary was not within their remit:
“This is a challenging task, as I do not think that the original manuscript meets the standards required by a scientific publication. . . While I have not be asked to comment on the original paper, I would strongly recommend to re-assess the review process of this paper.”
The Adjudicator focused primarily on the original paper, despite not having been invited to discuss that paper, and concluded with a recommendation to retract the original paper.:
“I agree with referee 1 that the addendum does not meet the scientific standards that would allow for publication. Furthermore, I recommend retraction of the original manuscript.”
The editor, Fabroni, than emailed the handling editor, Onega, the following on 13 July 2023 notifying him of the journal’s and Springer Nature’s decision to retract the original paper:
The adjudicator report – from a leading expert in the field – leaves no other choice but to reject the Addendum altogether under these circumstances.
The failure of the Addendum to mend the problems with the original article as shown by this in-depth post publication review, necessarily re-opens the question of the fate of the original article. After an in-depth consultation with the publishers (copied here) we came to the conclusion that a retraction is inevitable, a decision fully backed by the publishers.
A retraction based not on any claims of scientific misconduct, but simple disagreement.
I emailed Fraboni and Springer Nature with the following questions:
I understand that you have decided to retract Alimonti et al. Is this correct?
What is your motivation for the investigation of Alimonti et al.?
What is your rationale for the post-publication actions that you have taken, including now the decision to retract?
Under what specific journal procedures have you conducted your post-publication investigation and decision making?
Can you provide me a timeline of your actions under the formal procedures?
Should I get a response, I will update.
What to take from this saga? Shenanigans continue in climate science, with influential scientists teaming up with journalists to corrupt peer review.
