I’m very pleased to announce that our multi-year investigation into functional habitat connectivity priorities for the South Atlantic LCC is finished at last and ready for public consumption/use!
I’ll be giving a Third Thursday Web Forum presentation on the project on Thursday, March 17 2016 from 10-11 am (see link for details).
In a nutshell–we used supercomputers to model habitat connectivity for seven species: box turtles, pine snakes, timber rattlesnakes, eastern diamondback rattlesnakes, black bears, red wolves, and Florida panther/eastern cougar. We ran two different types of models, Circuitscape and Connectivity Analysis Toolkit, yielding two different metrics.
Circuitscape shows us the expected density of flow of these animals across the landscape, using a set of defined nodes as the starting points. In some areas, where there are pinchpoints caused by urban development, roads, etc., you see high density flow. In other areas with generally suitable habitat, the predicted flow spreads out. Circuitscape gives much more complex results than least cost path analysis, as it shows the flow of the species across all of the possible pathways between a pair of nodes all at once (some pathways end up used more, as you would expect).
Connectivity Analysis Toolkit produced a metric of “shortest path betweenness centrality”, which is basically what you’d get if you drew least cost paths between every pair of grid cells on the landscape, and then stacked those paths up and counted the number of overlaps. Places where lots of paths stack up have high centrality, so are important to the overall architecture of connectivity across the landscape.
The other cool thing we did was to project future landscape conditions in the year 2100 (using SERAP‘s urbanization projections and a coarse prediction of high-end sea-level rise) and then re-run our connectivity models for those conditions. As you might expect, the high levels of projected urbanization in the region reduced the predicted flow of animals a great deal, almost to the point where it is difficult to show current and future connectivity conditions on the same scale. Sea-level rise also pushed important connectivity pathways inland and away from low-lying areas such as NC’s Albemarle Peninsula and SC’s ACE Basin.
The final report is available here (with images of all of the layers we produced).
If you would like to receive the other data layers and a more detailed description of our GIS methods, please email me at ron at wildlandsnetwork.org
Looking ahead: now that the South Atlantic LCC has made great progress refining the Conservation Blueprint for the region, it makes sense to step back and consider how our model results might be used to complement and refine the Blueprint’s identified corridor regions for wildlife. There is also an urgent need to calibrate our connectivity model predictions with actual field data–so we’re better able to say whether two places on the landscape are indeed connected by actual flow of a given species. We’re also already thinking about how to use our results and the Blueprint to help develop a set of priority locations for wildlife road crossings across the region.
Lastly, I want to take a moment to thank all of the people who helped with the project: Paul Leonard and Rob Baldwin of Clemson University, Derek Fedak, Rachael Carnes, and Alison Montgomery (all interns for Wildlands Network), and Maggie Ernest (our new hire at Wildlands Network), who has been actively working to distribute the results to land trusts and other conservation professionals around the region. And then there are the dozens of expert biologists who helped us develop the resistance layers that underlie each of our connectivity models. These experts gave generously of their time and we cannot thank them enough (read their names in our report linked above).
For the wild,
Ron Sutherland, Ph.D.
Conservation Scientist, Wildlands Network