Bat use of upland ponds within a hardwood forest ecosystem, Southern Indiana.

Abstract

The extensive forested uplands of Morgan-Monroe and Yellowwood State Forests (MMYSF) in southern Indiana are largely composed of ephemeral streams. The Indiana Department of Natural Resources created ponds within the state forests to provide year-round fresh water for wildlife. The goal of this study was to document bat activity at these ponds, and determine what habitat attributes influence bat activity levels. In 2018 and 2019, we conducted acoustic surveys at 25 ponds, and temporally paired sites in the surrounding forest, during the summer months. Vegetation structure surveys were conducted at each pond to quantify habitat attributes that could affect bat activity, and aquatic faunal surveys were used as a surrogate for water quality. Results showed that bat activity was significantly higher at ponds than forest sites. A two-way ANOVA revealed that only two of 25 ponds surveyed had significantly more bat activity than the average, and results were consistent between years. According to a random forest model, the amount of vegetation surrounding ponds had the strongest relationship with bat activity compared to the other variables. Ponds with the highest bat activity had the following characteristics: < 35% vegetation structure surrounding the pond, surface area ≥ 1000 m2, and surface area coverage with emergent vegetation between 40-60%. In 2019, a small treatment experiment was conducted on a subsample of three ponds that were the least active in 2018, and two random selected ponds, not previously surveyed. Four 11m x 2m flyways were created around each of the five ponds by removing midstory trees and shrubs, to test if lowering the amount of vegetation surrounding a pond would increase overall bat activity levels. We used a two-way ANOVA with repeated measure and observed no difference bat activity levels between acoustic surveys, suggesting that bat activity is influence by other habitat attributes, in addition to vegetation structure. However, there were multiple outside influences within this single year experiment that may have influenced our results.