spOccupancy: An R package for single-species, multi-species, and integrated spatial occupancy models

Methods in Ecology and Evolution DOI:10.1111/2041-210X.13897

Jeffrey W. Doser, Andrew O. Finley, Marc Kéry, Elise F. Zipkin

Package Website and Repository

Code/Data DOI:

Please contact the first author for questions about the code or data used in the empirical case studies: Jeffrey W. Doser (doserjef@msu.edu)

---

Abstract

1. Occupancy modeling is a common approach to assess species distribution patterns, while explicitly accounting for false absences in detection-nondetection data. Numerous extensions of the basic single-species occupancy model exist to model multiple species, spatial autocorrelation, and to integrate multiple data types. However, development of specialized and computationally efficient software to incorporate such extensions, especially for large data sets, is scarce or absent.
2. We introduce the spOccupancy R package designed to fit single-species and multi-species spatially-explicit occupancy models. We fit all models within a Bayesian framework using Pólya-Gamma data augmentation, which results in fast and efficient inference. spOccupancy provides functionality for data integration of multiple single-species detection-nondetection data sets via a joint likelihood framework. The package leverages Nearest Neighbor Gaussian Processes to account for spatial autocorrelation, which enables spatially-explicit occupancy modeling for potentially massive data sets (e.g., 1000s-100,000s of sites).
3. spOccupancy provides user-friendly functions for data simulation, model fitting, model validation (by posterior predictive checks), model comparison (using information criteria and k-fold cross-validation), and out-of-sample prediction. We illustrate the package's functionality via a vignette, simulated data analysis, and two bird case studies.
4. The spOccupancy package provides a user-friendly platform to fit a variety of single and multi-species occupancy models, making it straightforward to address detection biases and spatial autocorrelation in species distribution models even for large data sets.

Repository Directory

All code and resulting model objects were created and saved using spOccupancy v0.3.0.

BBS

Contains all code and data for case study of the Black-throated Green Warbler distribution across the eastern USA.

• data: directory containing the raw BBS data used in the analysis. These data were downloaded directly from the USGS.
• bbs-PGOcc-cross-val.R: script to run nonspatial single species occupancy model with cross-validation.
• bbs-PGOcc.R: script to run nonspatial single species occupancy model.
• bbs-data-prep.R: script to prepare the raw data in the data subdirectory for analysis in spOccupancy.
• bbs-pred-data-prep.R: script to prepare the raw data in the data subdirectory for prediction across the eastern US in spOccupancy.
• bbs-predict.R: code to predict occurrence across the eastern US.
• bbs-spPGOcc-GP-cross-val.R: script to run spatial single species occupancy model using a full Gaussian process with cross-validation.
• bbs-spPGOcc-cross-val.R: script to run spatial single species occupancy model using an NNGP with cross-validation.
• bbs-spPGOcc.R: script to run spatial single species occupancy model using an NNGP.
• bbs-spPGOccGP.R: script to run spatial single species occupancy model using a full Gaussian process.
• pfile-1, pfile-2, pfile-3, pfile-sp-1, pfile-sp-2, pfile-sp-3: files used to specify initial values when running files across multiple cores from the command line.
• summary-bbs.R: script to perform summary analyses of all model results. Code to produce Figure 1 and Table 2 is in this script.

HBEF

Contains all code and data for case study of the foliage-gleaning bird community in the Hubbard Brook Experimental Forest.

• hbef-spatial: directory containing shapefiles of the Hubbard Brook Experimental Forest boundaries for creation of Figure 2 in the manuscript.
• hbef-msPGOcc-cross-val.R: script to run nonspatial multispecies occupancy model with cross-validation.
• hbef-msPGOcc-int.R: script to run nonspatial multispecies occupancy model with an intercept only model for occurrence.
• hbef-msPGOcc.R: script to run nonspatial multispecies occupancy model.
• hbef-predict.R: code to predict species-specific occurrence and species richness from a nonspatial multispecies occupancy model.
• hbef-spMsPGOcc-cross-val.R: script to run spatial multispecies occupancy model with cross-validation.
• hbef-spMsPGOcc-int.R: script to run spatial multispecies occupancy model with an intercept only model for occurrence.
• hbef-spMsPGOcc-nn.R: script to compare spatial multispecies occupancy models fit with different numbers of nearest neighbors.
• hbef-spMsPGOcc.R: script to run spatial multispecies occupancy model.
• pfile-1, pfile-2, pfile-3, pfile-sp-1, pfile-sp-2, pfile-sp-3: files used to specify initial values when running files across multiple cores from the command line.
• summary-hbef.R: script to perform summary analyses of all model results. Code to produce Figure 2, Table 3, Figure S1, and Figure S2 is in this script.

Simulations

Contains code and data for analysis of a simulated data set.

• simulation-data.rda: simulated data set obtained using simIntOcc().
• spIntPGOcc-sim.R: script to fit spatial integrated occupancy model for the simulated data set and analyze the results. Code to produce Figure S3 and Table S1 is in this script.