|
Activity Number:
|
192
- Contributed Poster Presentations: Section on Statistics and the Environment
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #330181
|
|
|
Title:
|
Improving Spatial Occupancy Model Parameter Estimation Using Citizen Science Data
|
|
Author(s):
|
David Huberman* and Brian Reich and Krishna Pacifici
|
|
Companies:
|
North Carolina State University and North Carolina State University and North Carolina State University
|
|
Keywords:
|
spatial;
citizen science;
occupancy modeling;
ecology;
bayes
|
|
Abstract:
|
In the past decade, ecological statistics has explored the utility of volunteer collected (citizen science) data in improving our understanding of species distributions. While professionally collected (standardized) data is preferable in terms of data quality, the abundance of citizen science data can provide otherwise difficult to obtain spatial coverage. This paper explores a method of incorporating citizen science data with standardized data to improve estimates of the effect of various habitat covariates on bird distributions. The method borrows ideas from causal mediation analysis for linear systems to decompose the total habitat effect using the citizen science data as a mediator variable to obtain indirect and direct effects. To obtain these effects, a spatial occupancy model is used with the mediator and habitat variables as covariates. Additionally, the mediator is regressed linearly on habitat. We hypothesize this method could be useful in contexts where the standardized data has poor site coverage relative to the citizen science data. The project and associated simulation study are in progress at the time of the abstract submission.
|
Authors who are presenting talks have a * after their name.
