March 28, 2013

Speeding-up the description of species

Yesterday I was reading two different but related papers dealing with accelerating the pace of species description by taxonomists. Both were published as open access, thus the interested reader can freely download them here and here. The paper from Zookeys dealt with the description of 101 new species of weevils (Coleptera, Curculionidae) from Papua New Guinea, while the one from Frontiers in Zoology is more about how to embark oneself in this thrilling fast-pace description of species.

Those two papers, published this week, are not the first ones to propose such a fast pace. Six months ago a paper in Zootaxa described 179 new species from Thailand belonging to the braconid genus Aleiodes and proposed the term "turbo-taxonomy" for such fast and quick descriptions. I am also aware of other papers in Hymenoptera following similar approaches (for example, a series of papers from the Platygastroidea Planetary Biodiversity Inventory, and two soon-to-be-submitted papers on the braconid genera Heterospilus and Apanteles). And without doubts, there are more papers published (or being written) which describe lots of species in a relatively fast way.

This topic, very fascinating and also controversial, has interested me for the past few years. Actually, any taxonomist dealing with hyperdiverse groups should be interested in considering and evaluating any help to speed-up the pace at which species descriptions are made! 
What characterizes such efforts? Are they done by "super-humans" taxonomists? Or "super-smart" ones? Is there a magic wand that allows to generate such impressive (and massive) papers?

There are a few common things shared between those papers. The three most conspicuous are:

1) Using DNA barcodes as one of the main tools to separate species. In some cases, barcoding is THE ONLY mean used to separate species, and is even added to taxonomic keys (for example, see some couplets of the Aleiodes paper mentioned above). Barcoding is not only used to tell species apart, but also to reconstruct phylogenies and to characterize individual species (in many cases the actual barcode is included as part of the description process).

2) Relying on high resolution pictures to depict characters and illustrate species.

3) Providing only short descriptions of the species (at least relatively short compared to more "traditional" taxonomic papers). Obviously this is a result of the previous point (and a confirmation of the old adage that "One picture is worth a thousand words").

There are other technology-driven tools, such as the use of software to generate keys and species descriptions (e.g. Delta or Lucid). Or the use of Wiki pages (see the weevil papers cited at the beginning of this post). Some papers (soon to come) also rely heavily on biological information i.e. host records, or detailed geographical distribution patterns.

Personally, I believe that fast taxonomy is the way many things will have to be dealt with. At least for hyperdiverse and morphologically cryptic species -such as those braconid wasps that are covered in this blog! 

In fact, during the past two years I have been involved in a series of taxonomic papers produced at a fast pace. One of them -about the microgastrine genus Iconella in the New World- is at present under review in Zookeys. And by this summer we (a group of authors) should have ready a revision of Mesoamerican Apanteles, including the description and keying of 206 species, 186 of them new to Science. Both papers were done in a relatively short period of time. And we have more papers coming, all to be done in a similar fashion.

Of course, there are many challenges when adopting this fast approach. For the sake of being fair and provide the two sides of the story, let me briefly mention some of the risks I can think of as having a negative influence.

One of the most important is the temptation to go TOO FAST. Sounds silly, but it might be likely to happen in some cases. A few examples: errors in coding characters (when using software to generate species descriptions), examining very few specimens per species (which risks overlooking the variation within and between species), or over-reliance on DNA barcodes (in the sense of uncritically accepting species boundaries as delimited by barcoding, without further analysis of other characters). It might be argued that one can be equally careful while describing hundreds of species, or a total mess while describing just one; it would all come down to the work ethics of the person(s) involved. While that is true, dealing with many species at once, and describing them in a quick way, increases the chances for human mistakes.

Another negative side is that "turbo-taxonomy" requires resources which are not equally accessible to researchers. If we look at the three main points mentioned above that characterize fast descriptions of species, it becomes evident that this is a rather expensive endeavour. While DNA barcoding has become relatively cheap -and no doubts it will continue that trend in the future- it still costs 10-15 dollars per specimen to be sequenced. If one is going to rely heavily on barcodes to deal with and describe species, then ideally one would need to barcode hundreds or thousands of specimens per taxonomic review. Likewise, higher resolution pictures are only good if using photographic equipment that presently costs several thousand dollars. Indeed, fast taxonomy is not suited for the amateur, or the scientist living in a not-so-wealthy country, at least not at present.

Another complication comes from the need to compare the putative new species against the species already described and deposited in museums. This actually represents two challenges. First, not many holotypes will be allowed to be barcoded -although museums policies might change in the future to allow for more DNA sampling of primary types, the current situation is not too favorable in that regard; and many old types will simply render no or very few DNA. The second challenge arises from the difficulties of finding old types, especially from species described in Europe and North America, with so many collections and over 250 years of taxonomic history. As many taxonomists know well, it is quite common that describing new species from the tropics is actually easier than from temperate areas (even though tropical diversity is much higher, usually there are much less described species to compare against). It is not a coincidence that the known examples of fast description of species mentioned above come, in all cases, from tropical and relatively poorly-studied areas (Costa Rica, Papua New Guinea, Thailand).

But the potential pitfalls should not prevent a taxonomist to consider with an open mind the speeding-up of species descriptions, and the potential use of several tools already available for that. Hopefully some improvements will follow, and down the road more researchers would embrace those approaches. It might be the only sensible way to advance the description of the millions of undescribed species...

[There are more considerations to make when discussing fast description of species. I hope to prepare a small article in the near future with more details... unless someone else decides to do it before. In any case, I welcome any comments and suggestions the readers might have. Thanks!]

No comments:

Post a Comment