Population ecology for wildlife, and for the people who manage it.
Most of the animals that matter to conservation and management are hard to see. They are nocturnal, wide-ranging, rare or simply small, and the decisions made about them are only as good as the estimates behind them.
We are a research group in the Department of Fisheries, Wildlife, and Conservation Sciences at Oregon State University, and we work on both halves of that problem. In the field, we run studies of carnivores, small mammals and birds across Oregon and beyond, using remote cameras, scent-detection dogs, GPS collars and biologging tags, acoustic recorders, live-trapping and DNA collected without ever handling an animal.
At the desk, we develop and test the statistical models that turn those data into defensible answers: how many animals there are, where they are, how they interact, and whether a monitoring programme could even detect a change if one happened.
Nearly everything we do is designed with the agencies, growers and forest managers who will use the results. The best questions come from real problems, and the best answers are the ones that get used.
Wildlife management runs on estimates. When they are wrong, or missing, the costs fall on species and on people.
Coastal martens persist in a handful of small populations. Western gray squirrels are declining. Recovery plans for species like these depend on knowing how many animals remain and whether numbers are rising or falling, and on finding out in time to act.
In outbreak years, gray-tailed voles strip crops across the Willamette Valley. After wildfire, forest managers must choose what to do with burned stands. We work with the people facing those choices so that the science answers the questions they actually have.
Every estimate is a model's reading of imperfect data. We test the tools ecologists rely on against simulated truth, and build better ones, so that agencies can trust the numbers they act on and afford to collect them.
Seven graduate-led field projects and several lab-wide programmes, organised around a small set of recurring questions about populations and how we measure them.
Photo: Jonny ArmstrongMartens, fishers, bobcats and cougars: how many there are, how they share landscapes, and what that means for conservation.
Voles that boom and crash, squirrels in decline, and the monitoring problems both create.
Wildfire, invading owls and expanding agriculture: how whole communities respond to change.
Hierarchical models and survey designs that are honest about uncertainty and practical to run.
Our remote-camera grids and acoustic recorders run day and night across Oregon. These are a few of the animals that walked past.
Oregon packs an unusual range of ecosystems into one state. Our field sites follow that gradient, and a few reach well beyond it.
We are part of the Cascadia Wildlife Lab, a collaborative research group based at Oregon State University that brings genomics, artificial intelligence, statistical modelling and field ecology together for evidence-based conservation.
Through Cascadia we work alongside colleagues at OSU, the USDA Forest Service and NCASI, and share genetics, AI and analysis capacity, from non-invasive DNA surveys to bioacoustics and camera-trap image recognition. Josh is part of the Cascadia Analysis Lab, which specialises in estimating density, distribution and movement from monitoring data.
Camera-trap surveys and hierarchical models show how much wildlife the forest set aside on working farms can hold.
Simulations show when co-occurrence models recover real interactions, and when they confidently report ones that are not there.
A way to ask, before the first camera goes out, whether a monitoring programme can detect the changes it exists to detect.
Our best projects start with a question an agency, grower group or forest manager needs answered. We also hire field technicians most years, and prospective graduate students are welcome to get in touch.
