Research

I am interested in the interface between the ecosystem sciences and public policy. In particular, I use remote-sensing, field data, and modeling tools to prioritize landscapes for the conservation of threatened species and the provision of ecosystem services.

Projects

Combining Presence and Fitness Data to Model Habitat Quality

In collaboration with Doug Bruggeman, Matt Simon, Austin Milt, and Jen Costanza, I developed a method to fuse presence and fitness data to model habitat quality for the Red-cockaded Woodpecker (Picoides borealis, RCW) across the Onslow Bight region of North Carolina. The model uses habitat structural information that I derived from a regional (1 million ha) LiDAR dataset. We used LiDAR, forest inventory data, and the maxent algorithm to locate forest patches that meet existing habitat standards for the species.

Our results indicate that less than 3% of the study region can be considered RCW habitat, and that a substantial portion (45%) is on lands without formal legal protection. We then refine our habitat model by regressing family group size, an indicator of fitness in RCWs, against LiDAR-derived forest structural characteristics and landscape attributes at multiple spatial scales. Our results indicate that group size is strongly related to forest structure near RCW nesting sites as well as their position in the larger landscape. Our analysis generates a high resolution map of habitat quality for this species and reveals novel relationships between RCW presence, fitness and habitat structure that could improve habitat management and recovery efforts for RCWs. The data used to drive this type of analysis is rapidly becoming available over large extents, and we anticipate that our methodolgy could be useful for modeling habitat quality for a number of species that are sensitive to vegetation structure worldwide. A paper based on this work is currently in review in the journal Biological Conservation.

Guiding Connectivity Conservation in the NC Sandhills

I have worked closely with the Sandhills Conservation Partnership to develop a GIS toolbox designed to help conservation practitioners easily incorporate habitat connectivity considerations into land management and land-use planning. The toolbox, called “Connect”, provides three GIS tools that build off of existing free and open-source software. The first tool creates simple animal dispersal models using circuit theory. The second tool uses the zonation algorithm to combine dispersal models from multiple species to prioritize habitat patches for conservation or restoration of landscape connectivity. Optionally, this tool can also incorporate social data into the habitat prioritization, including land acquisition costs and development threats. The third tool uses network analysis algorithms from graph theory to measure the impact of alternative land-use scenarios on landscape connectivity.

I developed two stakeholder-driven case studies that demonstrate the usefulness of our work. First, I used the Connect tools to prioritize private land parcels for connectivity conservation in fragmented habitats around Fort Bragg, NC, for a suite of species that are currently of conservation concern, including the eastern tiger salamander, ornate chorus frog, and the St. Francis saytr, a federally endangered butterfly. Second, I identify important dispersal corridors for the federally endangered red-cockaded woodpecker that are threatened by suburban development near the city of Fayetteville. Our results show that legal protection of lands currently targeted for conservation of red-cockaded woodpeckers would not protect important dispersal corridors for the other species in the study. Further, our analysis identifies "robust corridors" for RCW dispersal that are not likeley to be interupted by near-future land-use change. Based on our analysis and feedback from our project partners, we conclude that the Connect tools provide a useful platform for planning the adaptive management of multi-species habitat connectivity in complex landscapes. You can learn more about Connect and download a public beta version of the toolbox here .