A spatiotemporal model for multivariate occupancy data

Abstract

We present a multivariate occupancy model to simultaneously model the presence/absence of multiple species, and demonstrate its use with a goal of estimating parameters related to occupancy. The proposed model accounts for both spatial and temporal dependence within each species, as well as dependence across all species. These dependencies are addressed through random effects, defined so there is no confounding with estimating occupancy covariate effects. Data augmentation and specific choices for the random effects permit all Gibbs updates in the Markov chain Monte Carlo algorithm, making the model computationally efficient and scalable with the number of species and size of spatial domain. A simulation study shows that the model outperforms single‐species spatiotemporal occupancy models with regard to estimating occupancy parameters. We demonstrate the model with a three species camera trap study on Thomson's gazelle, wildebeest, and zebra in the Serengeti National Park of Tanzania, Africa.

Integrating Public Health Surveillance and Environmental Data to Model Presence of Histoplasma in the United States

“…We assume a probit link for each probability, ψi and pit, and relate these quantities to environmental variables and spatial or temporal random effects. 23,24 The probit link was chosen for computational considerations (see eAppendix Section 1.2; http://links.lww.com/EDE/B927 for more details). More specifically, we assume the probability of the presence of H. capsulatum relates to environmental covariates and a spatial random effect that accounts for our belief that H. capsulatum is more likely to be present in a county if it is present in neighboring counties.…”

Outstanding challenges and future directions for biodiversity monitoring using citizen science data

“…For example, the Polya‐Gamma (PG) data‐augmentation scheme for logistic regression models (Polson et al, 2013) is orders of magnitude more efficient than standard Metropolis–Hastings algorithms for occupancy models with a logit‐link function (Clark & Altwegg, 2019). Similarly, several advancements have been made for probit‐link occupancy models accounting for spatial and spatio‐temporal autocorrelation using efficient algorithms (Hepler & Erhardt, 2021; Johnson et al, 2013; Mohankumar & Hefley, 2022). Finally, implementing the PG scheme within a variational Bayes framework (Diana et al, 2021) leads to additional and substantial savings in computation time.…”

“…Similarly, several advancements have been made for probit-link occupancy models accounting for spatial and spatio-temporal autocorrelation using efficient algorithms (Hepler & Erhardt, 2021;Johnson et al, 2013;Mohankumar & Hefley, 2022). Finally, implementing the PG scheme within a variational Bayes framework (Diana et al, 2021) leads to additional and substantial savings in computation time.…”

Next-Generation Camera Trapping: Systematic Review of Historic Trends Suggests Keys to Expanded Research Applications in Ecology and Conservation

“…Increased experimental focus of anthropogenic and environmental factors on behavior generally and interspecific interactions specifically would benefit understanding even more if coordinated across studies and sites (Frey et al, 2017). Recent advances in the use of CT data to model dynamics of perceived risk (Palmer et al, 2017), use intensity from spatially recurrent events (Keim et al, 2019), and joint overlap along spatiotemporal dimensions (Cusack et al, 2017;Hepler and Erhardt, 2020) promise to expand study of interspecific interactions still further.…”