Improved Bayesian Analysis of Metapopulation Data With an Application to a Tree Frog Metapopulation

Abstract

Metapopulation models are important tools to predict whether a species can persist in a landscape consisting of habitat patches. Here a Bayesian method is presented for estimating parameters of such models from data on patch occupancy in one or more years. Earlier methods were either ad hoc, produced only point estimates, or could only use turnover information. The new method is based on the assumption of quasi‐stationarity, which enables it to use not only turnover data, but also snapshot data. Being Bayesian, the method produces reliable information about the uncertainty of the parameters and model‐based predictions in the form of posterior distributions. It is computationally demanding, but considerably faster than a recently developed Bayesian method extended beyond turnover data. The method is compared with existing methods (placed in a Bayesian framework) by fitting an extended incidence function model to a data set on a tree frog metapopulation with many missing values and by predicting its viability, mean occupancy, and turnover rate after 100 yr. Corresponding Editor: O. N. Bjornstad.

Limits of large metapopulations with patch-dependent extinction probabilities

“…other extreme, a SPOM might explicitly include a number of features of the environment, such as connectivity of patches, distances between patches, and the size and quality of patches (see [21] for an application to a tree frog metapopulation). The complexity of these models typically means that their properties need to be studied using simulation [16].…”

From Individual Behavior to Metapopulation Dynamics: Unifying the Patchy Population and Classic Metapopulation Models

“…The habitat patches are simply classified as occupied or empty, hence the name stochastic patch occupancy models (SPOMs) by Moilanen (1999). The SMT has turned out to provide an effective framework both for parameter estimation (Moilanen et al 1998;Hanski 1999;Moilanen 1999Moilanen , 2000Moilanen , 2002O'Hara et al 2002;ter Braak and Etienne 2003) and mathematical analysis (Day and Possingham 1995;Hanski and Ovaskainen 2000;Ovaskainen and Hanski 2001, 2003Frank and Wissel 2002;Ovaskainen 2002), and it has become a standard tool in quantitative metapopulation studies (Sjögren-Gulve and Hanski 2000;.…”

A model for a spatially structured metapopulation accounting for within patch dynamics

“…Let x(t, x 0 ) be the solution to (10). Suppose there exists bounded Lipschitz (7) and assume also that the functions b i satisfy (8). If X N (0) → x 0 ∈ E\∂E as N → ∞ and x(s, x 0 ) ∈ E\∂E for 0 ≤ s ≤ t. Then, for every t > 0 and δ > 0,…”

“…More generally, the presence-absence assumption has simplified modelling, data collection and analysis for a number of metapopulations [7,8,9,10,11,12,13,14]. However, this assumption is not always adequate, for example in stock dynamics where more detail is required [15].…”