Premium
Using Distance Sampling‐Based Integrated Population Models to Identify Key Demographic Parameters
Author(s) -
Schmidt Joshua H.,
Robison Hillary L.
Publication year - 2020
Publication title -
the journal of wildlife management
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.94
H-Index - 111
eISSN - 1937-2817
pISSN - 0022-541X
DOI - 10.1002/jwmg.21805
Subject(s) - population , population model , vital rates , wildlife , sampling (signal processing) , inference , statistics , population growth , distance sampling , abundance (ecology) , demography , mark and recapture , population size , geography , ecology , biology , computer science , mathematics , sociology , filter (signal processing) , artificial intelligence , computer vision
Effective wildlife management relies on rigorous estimates of population parameters, although data for small populations are often sparse, limiting inference. Integrated population models (IPMs) offer a potential solution by formally combining data sets in a unified analysis, thereby improving precision and allowing the estimation of latent parameters. We expected that incorporating open‐population distance sampling models into an IPM framework would provide further advantages for assessing population dynamics, particularly for rare species. We present an open‐distance IPM combining separate sources of abundance, composition, survival, and harvest data to better understand the dynamics of a small (~200 individuals) muskox ( Ovibos moschatus ) population in northwestern Alaska, USA. There was a 75% chance the muskox population in our study area was declining (λ < 1.0), primarily because of a −4.3%/year decline in adult females, and estimated survival probabilities were 0.70, 0.87, and 0.89 for yearlings, adult females, and adult males (harvest excluded), respectively. Insufficient numbers of recruits drove the decline in adult females, and harvest likely limited the adult male component of the population, accounting for up to 50% of mortalities. Together, these results suggest more conservative harvest management might be appropriate moving forward. In contrast, the results from a more conventional analysis were largely ambiguous, which would inevitably lead to delays in the application of appropriate management actions. Our work furthers the development of open‐population distance sampling models and IPMs and demonstrates an efficient approach for managing small populations when extensive marking of individuals is not possible. Published 2019. This article is a U.S. Government work and is in the public domain in the USA.