Quantifying Stressors and Natural Fish Habitat Quality: Linking Data to Fish Habitat Conservation

Our research team has been working with the Midwest and Great Plains Fish Habitat Partnerships to quantify fish habitat conditions and identify habitat conservation priorities throughout the central U.S.  Our goals include: 1- a compilation of current data on landscape characteristics and fish communities; 2- statistical analyses of these data to construct independent measures of natural habitat quality and cumulative anthropogenic stress for each Fish Habitat Partnership separately and for the Region as a whole; 3- assess differential vulnerabilities of fish habitats to climate change; and 4- combine information on fish habitat conditions with socio-economic factors to identify habitat conservation priorities.  Here, we will present an overview of the Data-Model-Application process as well as details of our modeling approach.  Data is compiled at the 1:100K segment-level watershed scale and is sequentially up-scaled to the 12-digit HUC and 8-digit HUC watersheds.  Modeling uses boosted regression trees to quantify relative contributions of landscape features to fish condition metrics and to predict current fish habitat conditions continuously throughout the region.  Applications include stressor identification, conservation prioritization, as well as threats and alternative futures assessment.  Finally, we will describe how the entire Data-Model-Apply process can be embedded within a spatially-explicit interactive GIS-based modeling system.