Oklahoma Project: Assessment of co-occurrence modeling when closure is not met

Oklahoma Project

Assessment of co-occurrence modeling when closure is not met

January 2021 - June 2022

Personnel

Robert Lonsinger, Principal Investigator

Participating Agencies

Species interactions and patterns of co-occurrence are commonly assessed with two-species occupancy models. Occupancy models assume that the occupancy state is closed to changes within a sampling season. When evaluating multiple species simultaneously to assess patterns of co-occurrence, the assumption of closure is likely violated due to variation in life history characteristics (e.g., movement capacity, home range size, social behaviors). This study is employing a simulation-based approach to evaluate the influence of violating the closure assumption on inferences from two-species occupancy models. The results of this study will help improve our understanding of occupancy modeling methods and provide insights on how best to interpret two-species occupancy modeling results when one or both species violate the closure assumption.

Research Publications

Research Publications	Publication Date
Lonsinger, R. C. 2022. Co-occurrence models fail to infer underlying patterns of avoidance and aggregation when closure is violated. Ecology and Evolution 7: e9104. https://doi.org/10.1002/ece3.9104 \| Abstract	June 2022	Advances in multi-species monitoring approaches have led to an increase in the use of multi-species occupancy analyses to assess patterns of co-occurrence among species, even when data were collected at spatial or temporal scales that likely violated the assumption that sites were closed to changes in occupancy state for the target species. Violating the closure assumption may lead to erroneous conclusions related to patterns of co-occurrence among species. Patterns of occurrence for two hypothetical species were simulated under patterns of avoidance, aggregation, or independence, when the closure assumption either was or was not met. Simulated populations were sampled at two levels (<i>N = </i>250 or 100 sites) and two scales of temporal resolution for surveys. Sample data were analysed with conditional two-species occupancy models and performance was assessed by comparing the proportion of simulations recovering the true pattern of co-occurrence. Estimates of occupancy were unbiased when closure was met, but biased when closure violations occurred; bias increased when sample size was small and encounter histories were collapsed to a large-scale temporal resolution. When patterns of avoidance and aggregation were simulated, conditional two-species models tended to correctly find support for non-independence when closure was met and estimated species interaction factors aligned with predicted values. Models tended to incorrectly infer a pattern of independence when closure was violated and power to detect simulated patterns of avoidance or aggregation decreased with smaller sample size. Simulation results suggest that when the closure assumption is violated, occupancy-based models of co-occurrence often fail to detect underlying patterns of avoidance or aggregation and incorrectly identify a pattern of independence among species, which could have negative consequences for our understanding of species interactions and conservation efforts. When closure is not met, inferred patterns of independence from multi-species occupancy models should be interpreted cautiously. In contrast, when violations of the closure assumption likely occur, evidence of avoidance or aggregation is likely a conservative estimate of true pattern or interaction.

Cooperative Fish and Wildlife Research Units Program: Oklahoma Education, Research and Technical Assistance for Managing Our Natural Resources