Ants in the past: Journal pulls insect-global warming paper after questions arise over results

insectscicoverA group of ecologists in Germany who published a paper on the potential impact of global warming on ants in the Harz Mountains — northern Germany’s highest range — have retracted the paper after becoming, well, a bit antsy about the validity of their findings.

The article, “Diversity of ants across an altitudinal gradient in and outside a spruce forest in the Harz Mountains, Germany,” appeared in August 2012 in the journal Insect Science, a publication of the Chinese Academy of Sciences. The last author of the paper was Christoph Scherber, of the University of Göttingen.

We found the abstract of the paper on this site:

Altitudinal gradients provide an excellent opportunity to examine the effects of climate change. Ants (Hymenoptera) are ideally suited for studying ß-diversity along altitudinal gradients because not only are they one of the most ecologically significant components of an ecosystem, many species are restricted to their preferred (micro-) habitats. Here we show the possible effects of global warming on the diversity of ants in the Wurmberg Mountain (Harz Mountains), Germany. We identified three subfamilies, ten genera, and 41 species, collecting 971 ants on Wurmberg Mountain. Higher diversity indices, numbers of individuals and species of ants were observed outside the forest than inside the pine forest, as well as with the decrease of the altitudinal gradient. We observed 18 species of Myrmicinae and 12 of Formicinae (total 3 species) at lower altitudes against seven and five (total 12 species), respectively, at higher altitudes. We found that the ants of the Harz Mountains react as expected to changes in altitude. That said, some species, specifically low-altitude thermophilic ones, show signs of expanding into higher altitudes, a possible reaction to climate warming. The distributions of host-specific herbivorous insects along altitudinal gradients, particularly within montane environments, provide useful analogs for predicted future changes that are likely to occur over time at any one location, given a gradually changing thermal environment.

But as the retraction notice indicates, those results are fraught:

The following article from Insect Science, ‘Diversity of ants across an altitudinal gradient in and outside a spruce forest in the Harz Mountains, Germany’ by Marc Srour, Germano Leão Demolin Leite, Torsten Wappler, Teja Tscharntke and Christoph Scherber, published online on 2 August 2012 in Wiley Online Library (, has been retracted by agreement between the authors, the journal Editor in Chief, Le Kang, and Wiley Publishing Asia Pty Ltd. The retraction has been agreed due to concerns having been raised about the validity of the species richness values derived by the authors, and regarding the validity of the species determinations.

We’ve tried to reach Scherber and the journal for more information and will update this post if we learn anything.

Update, 8 a.m. Eastern, 7/11/13: Added “northern” to description of Harz Mountains as Germany’s highest range. Thanks to commenters below for pointing out the error.

23 thoughts on “Ants in the past: Journal pulls insect-global warming paper after questions arise over results”

    1. No, you are right, the Harz is not Germany’s highest mountain range. Brocken, with 1141 metres above sea level the highest mountain in the Harz region, is Germany’s highest mountain outside the Alps.

    2. And it quite accurately states in the linked wikipedia entry that it is the highest mountain range in northern Germany, but I suppose the retractionwatch bloggers are too busy to bother with these details…

  1. Just some interesting details on the case:
    1) One of the authors is a Brazilian scientist with a remarkable CV record. He publishes in a quite diverse array of different specialities — maybe this makes it more difficult to keep a solid record of every published output result.
    2) Species identification in ants is absolute hell. One common practice is to ID them as “morphospecies” — by comparison with deposited specimens in official collections. Since the method is slack on its own nature, and many collections are not reliable, this approach is questionable. This is a common flaw in many published studies with ant biology.

      1. I did a pot doc once. Don’t remember much of it. I think it must have been at the same time as my bachelors. Don’t remember much of it.

  2. Concerning retractions in Entomology — which are bound to become and more numerous with time–, the following interesting Opinion Letter was just published:

    “Evidence of Photo Manipulation in a Delusional Parasitosis Paper” by Matan Shelomi, Department of Entomology, University of California–Davis.

    I thought it would interest RW readers. It prevents evidence of fraudulent image manipulation on a paper and suggests its retraction, discussing over practical issues. Quite remarkable.

  3. Hi all. I am the one who did the ant identifications for this paper, and the sole cause of the retraction. I had done these identifications as an undergrad when I was too inexperienced to do so, and too young to know to ask for help. The paper got flagged by several leading myrmecologists right after it was posted online early, and we moved to retract it the very next day, with the retraction notice coming out a week or so later, so that the damage would be very limited.

    I took full responsibility for it, since the others were merely building on my own mistakes and had done nothing wrong. I also pulled out all other papers I had in review that involved any sort of taxonomic identification by me, pending independent reinvestigation.

    I hope that was enough.

    1. well done Marc! You must have collected huge data from your undergrduate studies. What is your advice to undergraduates who do research and plan to write research reports?

      1. The biggest thing I learned from this self-caused debacle is to always get everything checked out by someone you know. I remember submitting my sections and thinking that if there was any mistake, the anon peer reviewers would catch them. Never depend on that, or else this will happen. And you do *not* want to go through a retraction experience, it’s horrible, especially as a young scientist (I had done the IDs when I was writing my unrelated palaeontological BSc. thesis, 21 years old; the paper and retraction came out 2 years later when I wasn’t in the uni anymore, a research associate at an NGO).

        More general things:

        – Keep a research notebook. Always write down anything you notice, no matter how small. You never know when a little factoid might actually be the key to solving the riddle. And as long as you have derived all your data with a proper methodology, you can include that stuff in your papers. It’s tough to get into the habit unless you grew up keeping notes, but it’s worth to learn it. I’m moving to continue my research elsewhere, and half my baggage allocation is just scribbled-on papers and notebooks. That’s how important these things can be.

        – Read papers and get a feel for what each section is meant to do and its target audience. Not many people will read the entire paper. A fellow researcher will read the methods, results, and discussion. Someone outside the field will read the abstract, intro, and probably discussion. Most people will stay at the abstract. So the language and level needs to be adjusted for each section, and you need to make sure that pertinent facts get repeated in case someone skips through the paper.

        – Get a citation manager. Zotero or Mendeley are free and adequate. My initial draft of my part of the introduction had 60 references, since I wrote a mini-review of ants at altitudinal gradients. It was a huge and time-consuming pain to edit all the references and get them right when we changed the journal destination or when I rewrote stuff. With a citation manager, it’s all done in 2 minutes rather than 2 hours.

        1. This is something grand you are doing here, Marc. Congrats. You cleary mean this was honest move all along, and your attitude is commendable. I would suggest adding any useful note on the retraction on its respective PubPeer entry, as retraction notices are usually uninformative and many readers might get the wrong idea.
          Indeed keeping useful notes is very hard.

        2. …but still I really think other co-authors (and lab colleagues) could have questioned the nature of the core data used in this publication. I personally think an undergrad is free to err so he can learn. Getting unsupervised data published does not look right for any professional scientist. Probably the greatest lesson to be drawn here.

  4. While seraching for something else, I stumbled across this retraction, labelled as “withdrawn”:
    Withdrawn: A new record of longicorn beetle, Acanthophorus rugiceps, from India as a root borer on physic nut, Jatropha curcas, with a description of life stages, biology, and seasonal dynamics. Mathyam Prabhakar, Y.G. Prasad, G.R. Rao, B. Venkateswarlu
    First published online: 2 November 2014
    The reason: “This paper (originally published as 12.141) described the root borer, Acanthophorus rugiceps Gahan (Coleoptera: Cerambycidae), as a new record on physic nut, Jatropha curcas L. (Malpighiales: Euphorbiaceae), in India. The species involved has since been identified as Acanthophorus serraticornis (Olivier) by Dr. Alain Drumont (Entomology Department, Institut Royal des Sciences naturelles de Belgique, Bruxelles Belgium), which is known to be present in India. The original paper has therefore been withdrawn.
    Received December 10, 2011. Accepted February 23, 2012.”

Leave a Reply

Your email address will not be published. Required fields are marked *