Accounting for sampling error when inferring population synchrony from time-series data: a Bayesian state-space...

Area of Science:

• Ecology
• Population Dynamics
• Statistical Modeling

Background:

• Population size data often contain sampling errors, which can bias ecological models.
• Ignoring sampling error in population dynamics can lead to inaccurate parameter estimation, such as density-dependence.
• This bias can distort the understanding of factors driving population synchrony, like the Moran effect versus dispersal.

Purpose of the Study:

• To demonstrate the bias introduced by neglecting sampling error in population synchrony studies.
• To introduce a novel space-state modeling approach for quantifying population synchrony that explicitly incorporates sampling error.
• To provide practical examples and tools for researchers to apply this improved methodology.

Main Methods:

• Developed a space-state modeling framework to explicitly account for sampling error in population size data.
• Applied the model to empirical datasets with pre-estimated and jointly estimated sampling variances.
• Compared the results from the new approach with a standard method that ignores sampling variance.

Main Results:

• Neglecting sampling variance can obscure true population synchrony patterns.
• The classical estimator for synchrony strength is biased downward when sampling errors are independent.
• Averaging population size replicates is an insufficient method to correct for this bias.

Conclusions:

• The proposed state-space model accurately quantifies population synchrony, even with over-dispersed count data.
• This flexible approach accounts for uncertainty in population size estimates, crucial for field studies.
• A user-friendly R-program and tutorial are provided to facilitate the adoption of this method in ecological research.