|
Activity Number:
|
177
- Observations Through a Foggy Lens: Modeling Complex Measurement Error and Non-Random Missingness in Ecological and Environmental Health Data
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, August 8, 2022 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Consulting
|
|
Abstract #320402
|
|
|
Title:
|
To Model or Design Your Way to Defensible Conservation Information on At-Risk Bat Populations?
|
|
Author(s):
|
Kathryn Irvine* and Christian Stratton
|
|
Companies:
|
US Geological Survey and Montana State University
|
|
Keywords:
|
sensor network;
ecology;
conservation;
trend estimation;
sampling design;
species assemblages
|
|
Abstract:
|
Bats are probably the most at-risk group of terrestrial vertebrates in North America. Bats are under-studied relative to other taxonomic groups because of their cryptic and nocturnal behavior. The easiest way to collect bat data is using automatic recording units that record echolocation calls as bats navigate through the environment. Then proprietary software algorithms are used to classify a bat pass to a species. Bat acoustic surveys present an intoxicating blend of statistical challenges: false positives, false negatives, lack of independence, purposive sampling, and uncertain and nonoverlapping species ranges. This laundry list of concerns results in potential bias in statistical inferences. An additional complication is that our proposed solutions must be practical and feasible for bat biologists to adopt and in turn operationalize. Our research explores the tension between providing guidance to mitigate sources of error through the design and data processing workflow versus advancing statistical model development.
|
Authors who are presenting talks have a * after their name.
