Date of Award

January 2020

Degree Type

Open Access Thesis

Document Type

Master Thesis

Degree Name

Master of Science (MS)


Biological Sciences

First Advisor

Luke E. Dodd

Department Affiliation

Biological Sciences

Second Advisor

David R. Brown

Department Affiliation

Biological Sciences

Third Advisor

Cy L. Mott

Department Affiliation

Biological Sciences


Recovery of bat species impacted by white-nose syndrome (WNS) will necessarily require population growth. I assessed reproductive capability of the endangered Indiana bat (Myotis sodalis) in Kentucky, where WNS was first detected in hibernacula in April 2011. Due to loss of fat reserves associated with WNS infection, coupled with the energetic expenditures associated with pregnancy, I hypothesized mass of reproductively-active females captured during the maternity season would decrease across my data collection period (2017-2019). Further, I predicted that reproductive rates in the study population would be lower than historic rates for Kentucky. Mist net surveys around artificial maternity roosts at three focal management areas resulted in the capture of 866 Indiana bats across 22 netting events during the 2017-2019 maternity seasons. I examined changes in female mass (a measure of reproductive health) across years in order to determine how bats responded to WNS, and also assessed population trends in female pregnancy and lactation timing over the course of a season. A linear mixed model (LMM) was used to examine the effects of year (an analog for WNS impact) and wing score on the mass of female bats during the maternity season within my study term. To assess and account for the influence of other variables (reproductive condition, right-forearm length, capture date, site) on female mass, I conducted Akaike information criterion (AIC) model selection and restricted analysis to relevant models. Generalized linear mixed models (GLMM) with binomial distribution were implemented to determine if reproductive proportion differed between my field-collected data versus historic records for Kentucky. Counter to expectations, data at the population-level indicate 92% of adult female Indiana bat captures exhibited signs of reproduction (pregnancy, lactation, or post-lactation). Further, juvenile bats comprised 55% of captures in surveys during the post-parturition phase of the maternity season, and multiple juveniles were recaptured as reproductive adults during the study. Although reproductive proportion trended lower in post-WNS captures (92% compared to 99% pre-WNS) no significant difference was found between pre and post-WNS reproductive proportions. Across the three years of my study, post-WNS within-colony reproductive proportions did not differ. These data suggest reproductive potential has been sustained in the studied colonies despite bats’ exposure to WNS during hibernation; promising evidence for persistence of the species. Even so, optimized management of maternity habitat remains critical to support population recoveries from WNS.

Additionally, I reported band recovery data of bats at the study colonies. While banding provides valuable information on migratory behavior, travel distances, survival rates and reproductive rates, among other behavioral insights, band recovery rates are persistently low. My study was designed to apply bands in large numbers to determine recapture rates at focal management locations of the Indiana bat during the maternity season, and to investigate bat movement within and from these locations. Relying on the same mist net surveys as my first chapter, a total of 119 recaptures were observed at an overall recapture rate of 14%. Within-year recapture rates in individual years of the study increased from 1% in 2017 to 18% in 2018, leveling at 17% in 2019. Most recaptures at each site occurred within June of each year. Recaptures at the more intensively sampled Veterans Memorial Wildlife Management Area (VWMA) demonstrated occurrences of roost switching and movement between three discrete clusters of roosts. Recoveries were recorded in 4 Kentucky hibernacula, two of which (Bat and Saltpeter Caves) ranked within the 10 most populous Indiana bat hibernacula in the 2019 range-wide population survey. Links were also revealed between a hibernaculum in Indiana and two capture sites, Bernheim Arboretum and Research Forest (BARF) and Yellowbank Wildlife Management Area (YWMA). Five of these recoveries were recorded in Jug Hole Cave, reported as the second largest Indiana bat hibernaculum in 2019 range-wide surveys. Most female hibernacula recaptures with known reproductive condition (80%) showed signs of reproduction upon first capture. Additionally, two females exhibited reproductive capability prior to and following their recovery in caves. These observations demonstrate the ability of female Indiana bats to continue the migratory cycle despite WNS-related energy depletions coupled with expenditures of pregnancy, and provide evidence that female Indiana bats can engage in reproduction following potential WNS exposure during the hibernation period.

Finally, I investigated temperature conditions and use of artificial roosts by Indiana bats. Deployment of artificial roost structures is an increasingly common strategy in bat research and management, and understanding potential benefits and impacts of these structures on roosting bats is imperative to management of imperiled species. Although all roosts fluctuate in temperature, roosts that remain cool (<15°C) for extended periods may cause bouts of torpor, potentially impacting females during maternal seasons. Overheating roosts (>40°C) can cause heat stress, while extreme overheating events cause mortality. I hypothesized that temperatures would differ across three roost clusters based on differing levels of solar exposure at our focal study site, VWMA, a location with well-documented presence of Indiana bats in Kentucky and discrete sections of BrandenBark™ artificial roosts spread across the landscape. I also predicted that live potential roost trees at the site would be cooler than artificial roosts, and when comparing artificial habitat only, that bats would be found in greater densities at warmer artificial roosts due to preference for solar exposed snags. To record roost temperatures, data loggers were placed on the exterior of false bark on roost structures at northeast and southwest aspects (n = 38) and set to capture hourly temperature through the maternity season. Data loggers were placed in the same orientation on one live tree in each cluster (n = 6). Bat use of structures was indexed via standardized guano screens, with guano collected every 2-3 days from different clusters. Generalized linear models (GLM) approximating a gamma distribution and AIC model selection processes were used to determine the most influential variables on daily mean, minimum, and maximum roost temperature. Mann-Whitney U tests were conducted to determine if artificial roosts differed in temperature from live trees. Graphing approaches were implemented to explore relationships between temperature and roost use by bats. North, Central and South roost clusters did not vary significantly from one another in temperature profile, and aspect placement of roost data loggers did not cause deviations in mean, minimum, or maximum daily temperature of roosts. Comparisons between live trees and artificial roosts resulted in no significant difference between daily mean and minimum temperature, while daily maximum temperature was slightly higher at artificial roosts. No temperature measurements were collected that exceeded the maximum temperature threshold of 40°C. Daily minimum temperature for individual loggers fell below the minimum threshold on 648 instances across 29 days, representing temperatures across all roosts, leading to the conclusion that bats at VWMA are under greater threat from roosts falling under 15°C than overheating. However, results may have differed had loggers been placed on the interior of false bark on roosts. Bats exhibited the highest densities at the South cluster of roosts, followed by the Central and North clusters. However, the colony appeared to spread across VWMA throughout the season, and cumulative collection from the greatest-use roost in each cluster accounted for 36% of all guano collected in the study. The lack of significant temperature difference between clusters, and persistent use of all roost clusters at VWMA indicates each provides suitable habitat for Indiana bat maternity colonies. All artificial roosts were used at VWMA in 2019 and roost switching was confirmed, indicating suitability of roosts placed in a variety of habitat conditions, and the ability of bats to successfully relocate. Reproduction and pup rearing were observed at all three clusters, indicating the importance of variation in roost placement on the landscape for opportunistic selection. Selection of roosts by Indiana bats is likely the result of additional unknown factors.