Habitat Associations of Blotchside Logpearch (Percina burtoni) in the Little River, Tennessee, and the Availability of Preferred Habitats in Abrams Creek, Great Smoky Mountains National Park

Cary, Joshua

ABSTRACT: Blotchside Logperch (Percina burtoni) are a large and long-lived darter native to the creeks and rivers of the Tennessee and Cumberland river drainages. Although the species once occupied large regions in these drainages, it is currently limited to small, disjunct populations in the Tennessee River watershed. One of the areas where Blotchside Logperch no longer occur is Abrams Creek in Great Smoky Mountains National Park. Great Smoky Mountains National Park is seeking to reintroduce Blotchside Logperch to Abrams Creek but the availability of suitable habitat within the system is unknown and the creek’s lower reaches are now inundated by the Chilhowee Reservoir, greatly reducing the amount of mid- to large-sized river habitat available. The goal of this thesis is to guide the potential reintroduction of Blotchside Logperch in Abrams Creek by determining the habitat associations of the species and assessing to what extent comparable habitats are present in the receiving system. To most accurately assess the habitat availability of Abrams Creek, the habitat associations of Blotchside Logperch were assessed at the landscape (105 m – 106 m), reach (10 m – 103 m), and microhabitat (1 m) scales. Maxent modeling was used to analyze range-wide collection records of Blotchside Logperch and correspond occurrences with landscape scale variables. The habitat associations of Blotchside Logperch at the reach and microhabitat scales were determined by sampling the Little River population, which is the closest remaining population to Abrams Creek. Occupancy modeling was used to assess reach scale presence while also accounting for imperfect detection across 31 sites in the Little River. Within occupied Little River sites, random forest models were used to distinguish occupied microhabitats from available microhabitats. Stream size was the strongest predictor of Blotchside Logperch occurrence at the landscape (variable: cumulative drainage area) and reach scales (variable: mean wetted width). At the microhabitat scale, habitat unit (riffle, run, or pool), dominant substrate class, and percent of substrate embeddedness were the variables with the most importance. The most supported model at each scale was used to predict the availability of preferred habitats in Abrams Creek and, while the models predict preferred habitats are present, availability and location of preferred habitat is dependent on scale of observation. Even so, successfully reestablishing a population of Blotchside Logperch would contribute to the recovery of Abrams Creek and would create a new, protected population for this vulnerable species.