Melded Integrated Population Models

Justin J. Van Ee (), Christian A. Hagen, David C. Pavlacky, David A. Haukos, Andrew J. Lawrence, Ashley M. Tanner, Blake A. Grisham, Kent A. Fricke, Liza G. Rossi, Grant M. Beauprez, Kurt E. Kuklinski, Russell L. Martin, Matthew D. Koslovsky, Troy B. Rintz and Mevin B. Hooten
Additional contact information
Justin J. Van Ee: Colorado State University
Christian A. Hagen: Oregon State University
David C. Pavlacky: Bird Conservancy of the Rockies
David A. Haukos: Kansas State University
Andrew J. Lawrence: New Mexico State University
Ashley M. Tanner: Caesar Kleberg Wildlife Research Institute, Texas A &M University-Kingsville
Blake A. Grisham: Texas Tech University
Kent A. Fricke: Kansas Department of Wildlife and Parks
Liza G. Rossi: Colorado Parks and Wildlife
Grant M. Beauprez: New Mexico Department of Game and Fish
Kurt E. Kuklinski: Oklahoma Department of Wildlife Conservation
Russell L. Martin: Texas Parks and Wildlife Department
Matthew D. Koslovsky: Colorado State University
Troy B. Rintz: Western EcoSystems Technology, Inc.
Mevin B. Hooten: The University of Texas at Austin

Journal of Agricultural, Biological and Environmental Statistics, 2025, vol. 30, issue 3, No 9, 769-799

Abstract: Abstract Integrated population models provide a framework for assimilating multiple datasets to understand population dynamics. Understanding drivers of demography is key to improving wildlife management, and integrated population models have informed conservation practices for many species of conservation concern. Motivated by multiple surveys of lesser prairie-chicken (Tympanuchus pallidicinctus), we developed a flexible integrated population modeling framework for assimilating demographic data with multiple surveys of abundance. Measurements of abundance are derived from aerial and ground surveys that vary in their observational uncertainty, sampling design, temporal coverage, and survey effort. Our proposed integrated population model draws from the strengths of each survey and prevents their sampling biases from compromising inference. We facilitate posterior inference for our integrated population model using chained Markov melding, which induces the joint distribution for all data sources by linking inference across several submodels. Using Markov melding, we extend the modeling framework previously proposed for analyzing the individual data sources while still obtaining joint Bayesian inference. We fit the melded model with a multistage Markov chain Monte Carlo algorithm that decreases run time and improves mixing. We assimilate data from several state and federal wildlife agencies and over a dozen independent researchers to infer lesser prairie-chicken abundance and vital rates across its entire range over the last 18 years. Supplementary materials accompanying this paper appear online.

Keywords: Data assimilation; Integrated modeling; Lesser prairie-chicken; Chained Markov melding; Multistage MCMC; State-space modeling (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s13253-024-00620-2 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:jagbes:v:30:y:2025:i:3:d:10.1007_s13253-024-00620-2

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/13253

DOI: 10.1007/s13253-024-00620-2

Access Statistics for this article

Journal of Agricultural, Biological and Environmental Statistics is currently edited by Stephen Buckland

More articles in Journal of Agricultural, Biological and Environmental Statistics from Springer, The International Biometric Society, American Statistical Association
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().