These are the two words you never want to see on Wikipedia: [citation needed]

Why? Because Wikipedia isn't infallible. The world's fifth most popular website – according to um… Wikipedia – is an awesome resource of almost endless knowledge, but that knowledge is only as sturdy as the bedrock of primary sources it's built on top of.

That's why you never want to see [citation needed].

Every time you do, it becomes a bit harder to rely on whatever it was you just read – and because no source is listed, it's harder to track down further information if that's what you need.

With those sorts of things in mind, and to publicise open accessibility of scholarly citation data, researchers at the Wikimedia Foundation recently compiled a dataset of all academic citations across Wikipedia, combing each of its 297 editions published in different languages.

For the purposes of this analysis, each citation had to have a research-oriented identifier: either an ISBN (books), a DOI (research papers), a PMID or PMC (biomedical research), or an arXiv ID (pre-print research).

In other words, Wikipedia citations referencing regular web pages (like the story you're reading right now), newspaper or magazine articles, or TV shows wouldn't be included – but basically all circulating research in scientific and non-scientific literature would be.

So, with that condition in mind, what are the most cited sources across all of Wikipedia's almost 15.7 million citation records? Here they are:

  1. Updated world map of the Köppen-Geiger climate classification, 2,830,341 citations
  2. Prediction of Hydrophobic (Lipophilic) Properties of Small Organic Molecules Using Fragment Methods: An Analysis of AlogP and CLogP Methods, 21,350 citations
  3. The status, quality, and expansion of the NIH full-length cDNA project: the Mammalian Gene Collection (MGC), 20,247 citations
  4. The de Vaucouleurs Atlas of Galaxies, 19,068 citations
  5. The Complete New General Catalogue and Index Catalogues of Nebulae and Star Clusters by J. L. E. Dryer, 19,060 citations
  6. Galaxies and How to Observe Them, 19,058 citations
  7. A Concise History of Romania, 15,597 citations
  8. Catalog of Fishes California Academy of Sciences, 11,980 citations
  9. Dictionary of Minor Planet Names, 10,651 citations
  10. National and religious composition of the population of Croatia, 1880–1991: By settlements, 8,230 citations

Okay, so don't be surprised if you haven't heard of most (or probably any) of those.

While these are the 10 most frequently cited academic sources on Wikipedia, they're culled from some 4.5 million unique references, so even though they're definitely popular in terms of academic citations, they're a drop in the ocean of what is technically citable.

That said, they're all notable for the high number of citations they receive, and none more so than the top result, which totally eclipses everything else on the list – let alone the rest of Wikipedian academia – with its stunning 2.8 million citations.

What's even more surprising is the Australian authors of the paper – an update of a widely used climate classification scheme originally developed by Russian-German scientist Wladimir Köppen – had no idea their study was so widely referenced on the world's largest encyclopaedia.

"Those numbers blew me away," retired geography professor, Brian Finlayson, formerly of the University of Melbourne, told Wired.

"None of us had any idea about this. We didn't know Wikipedia collected this information or anything about it."

The team suspects the reason their paper might be so widely cited is because it constitutes an important update of one of the most important climate classification schemes known to science, which was originally devised well over a century before their own amendments in 2007.

"There's nothing scientifically new in it, we simply used Köppen's classification and added new data to it and then drew a world map," Finlayson told Wired.

"The reason it's so widely cited is because it's useful, and I think that's the important point about it, it's not that we suddenly dropped into the system this brand-new thing that had never been done before."

Another factor is that they consciously elected to publish it in a free, open access journal – meaning their research, and the contribution it makes to scientific endeavour, wasn't locked behind commercial paywalls.

That decision not only made their climate classification update available to a broader audience – it also helped make Wikipedia history, garnering almost 3 million citations in a movement and medium Köppen could never have imagined.

"The journal we originally published the paper in is free and open access, and we chose the journal for that reason," Finlayson told The Guardian.

"Research is no good to anyone locked in a cupboard, or published in a journal you have to pay a lot of money to access."

You can read about the Wikimedia research here, and find the citation datasets they compiled on Figshare.