21st Annual Student Research Conference
Descriptive Statistics of Echolocation Calls From Northeast Missouri Bats
Phillip D. Vance
Dr. Scott Burt and Dr. Jason E. Miller, Faculty Mentors
Most bats emit ultrasonic sounds known as echolocation calls, which many investigators assume can be used to differentiate species. However, the practice of determining species apart by their calls has proven difficult. Our goal has been to develop a statistical model that will accurately and reliably predict a bat's identification based on quantitative acoustic data. However, a high degree of variability exists for call characteristics in a species, variation that has impeded our ability to develop accurate statistical models. To further understand the models' predictive ability we have compiled descriptive statistics for the calls that were recorded. From these data we have determined the temporal and frequency characteristics and ranges for seven bat species. Such basic data are surprisingly lacking in the primary literature. From these results, we have also discovered that some call features are superior to others in discrimination among the species. Such knowledge will be of great use to anyone interested in distinguishing bats from their echolocation calls and will serve to enhance our models.
Keywords: bats, echolocation, calls, statistical model
Topic(s):Mathematical Biology
Presentation Type: Oral Paper
Session: 52-5
Location: VH 1428
Time: 3:45