Benoit Guenard figures out the easiest places to record new ant genera

If you follow myrmecology on the internet, you probably know about Benoit Guenard’s Global Ants database. Benoit has spent years combing disparate biological literature and natural history collections to compile a comprehensive map of where all the 300-some ant genera are known to live. This information is useful in its own right (want to know which ants live in that tropical vacation destination?) but the database is more powerful that that. It can be used to make predictions about where in the world we are most and least likely to make new genus & species records.

Top: the number of ant genera recorded from various political divisions (darker=more). Bottom: model predictions of undercollected regions (yellow & blue are different models; black is where both models agree). Modified from Figures 1 & 3 in Guenard et al 2012.

In a clever paper out this week in the Proceedings of the National Academy of Sciences, Benoit and his colleagues Michael Weiser and Rob Dunn apply a pair of mathematical models to the database to locate spots on the map with far fewer known ant genera than their location might predict. Because ant researchers have tended to work more in particular countries and less in others, what this project has effectively done is pinpoint the under-studied corners of the globe. Places where even common ants have gone uncollected.

Off to Cambodia it is, then.


source: Guenard, B. et al 2012. Global models of ant diversity suggest regions where new discoveries are most likely are under disproportionate deforestation threat. PNAS published online before print doi:10.1073/pnas.1113867109.

9 thoughts on “Benoit Guenard figures out the easiest places to record new ant genera”

    1. There is also the consideration of the probability of making it back alive …

      Just sayin…not terribly smart to collect in a perpetual warzone, heh.

  1. Strange paper… The genera map is indeed helpful and I guess I get the vaguely suggested conservation implications but I’m struggling with specific value of using genera for conservation purposes. So you find Pheidole in a new region, no big deal. But if that Pheidole is a new group of trimorphic species – then it is a big deal. But these models can’t inform such a scenario. Both models cannot provide any specific predictive value that applies to one of the grossest level for conservation decision making (e.g. total species diversity per unit area) that is typically used for assigning “hotspot” categories either. I would also argue that most of the people that are actually likely to seek out areas that are already under-collected have already identified many of these areas (all of the ant taxonomists I have had conversations about their wishlists included several spots in southeast Asia, including Cambodia and Vietnam, southern Mexico, and a number of others on the maps).

    1. I wouldn’t be so critical. The study is obviously crude, but:

      1. We go to war with the data we have, not the data we wish we have. Species-level data is superior, but mostly unavailable.

      2. Having a quantitative assessment of what we qualitatively “know” already does have some value, especially in areas relevant to policy-making.

      1. I’m not sure I’m being critical, I’m just not sure what to think of the paper, and my stream of consciousness pooped that out in the comment above. My limited intelligence and complete lack of imagination probably are the root cause for this.

        But playing devil’s advocate to your comment on “… with the data we have…” why not just publish this as a genera map? That is the most useful part of their data gathering exercise.

        I teach Conservation Biology (undergrad.) and Conservation Biology Theory (grad) and I have a hard time coming up with uses for the models presented that aren’t already redundant with much of the (very limited) current insect conservation practice that is out there and focused on hotspots. Again, I’m not being critical just trying to imagine how to use this in a way other than “factoid.”

        1. If we can use taxonomic rank as a (very) relative indicator of magnitude of diversity, this highlights not only areas which are known to have greater diversity, but areas which need exploration. The “pin-pointed” areas are proverbial X’s on treasure maps. I myself am pleased to see that some of the pin-pointed areas are those which will be explored by Jack Longino’s prospective LLAMA (Leaf Litter Arthropods of Meso-America) II expedition. Should funding be granted for that expedition, I look forward very highly to the discoveries which await in the winklers (and baiting and beating samples). Jack’s rigorously quantitative sampling methods will provide an excellent test of Benoit’s predictions.

Leave a Reply