In light on a couple recent posts on parks and conservation lands (on which I hope to comment soon), some of you might be interested in participating or monitoring the discussion being help this week here.  The Designing the Parks initiative is sponsored by the National Park Service, UVA, and a mess of NGOs.  In a conference earlier this year they attempted to “to assess lessons learned through an examination of park planning and design history” and, come December, they will reconvene in San Francisco to hash out a picture of the future look of our parklands.  The latter is the topic of the current online forum.  White it’s aimed primarily at landscape architects, we certainly can’t leave them to their own devices.

The basic approach in species distribution modeling / habitat suitability modeling is you collect occurence data for the target species, collect spatially-explicit environmental variables that might determine the limits of the species’ range, and then model where you would expect to find the species in your study area. In general, it works okay (no models are real, some are good, &c &c). It can be useful, especially in conservation, when rigorous surveys aren’t feasible. But people have been using these methods to model current distribution, and then predict range shifts based on climate change. As with yesterday’s biophysics approach to modeling flyways, such a study ignores things like behavior and adaptation. What limits the current generation might not do so in other conditions.

It surely says something about me that (with reservations) I think this is very cool! It’s always neat to have this sort of cross-disciplinary stuff: modeling the energetics of an individual is such a physicist way of thinking about movement ecology. It also doesn’t hurt to have the computing power of NASA. There’s been plenty said about the coming convergence of computing power, neural networks, and the human brain. As computers become more powerful, and our ability to model evolutionary and ecological processes improves, I think we’ll be provided with incredibly useful tools to study conservation.

Of course, field biologists will be, rightfully, crying foul: nature is incredibly complex and while it may follow certain patterns, the details are nearly impossible to define. Smith’s model doesn’t include multi-species interactions (e.g. competition and predation); it also didn’t appear to predict the fact that his study species, the Pectoral Sandpiper, is a frequent migrant to the UK. No doubt that tug of war between folks in the field and modelers will continue, but there’s a balance to be struck. Models like this, by people removed from the traditions of the field, can help to expose previously-ignored pressure points in ecology. I think the word I’m looking for is “refreshing.”

NatureServe, as I understand it, grew out of a TNC initiative to develop spatial datasets and provide them to researchers on-line. It seems that, throughout their existence, they’ve struggled with presenting conservation content on the internet: who is their audience? How do they get the data they need, and how is it updated? Much like the rest of the web, a lot of these projects try to do too much (“Let’s get researchers to upload all of their spatial data to one place!”), and the effort dies out. The really useful sites are ones that present novel data-sets for specific purposes. If you need a land-cover dataset, it’s much easier to find it at the host institution rather than going through some aggregated site. “One-stop shops” tend not to carry the stock that they promise. Regardless, it’s National Geographic, so the site is definitely beautiful. I hope they push through that nebulous barrier and keep this thing going.

[Edit: Oops! I forgot that NatureServe developed its NatureServe Explorer and InfoNatura, presenting datasets of species ranges in North and South America, which I've used in the past. They are useful -- but, again, not comprehensive.]