simIDS: Simulate data for an integrated distance sampling, point...

In AHMbook: Functions and Data for the Book 'Applied Hierarchical Modeling in Ecology' Vols 1 and 2

Simulate data for an integrated distance sampling, point count and occupancy study

Description

Generates hierarchical distance sampling data, point count data, and occupancy (detection/nondetection) data for an integrated distance sampling (IDS) model with shared density and availability processes, but possibly different detection/perceptability processes (i.e., different detection functions).

The function calls simHDSpoint to generate hierarchical distance sampling (HDS) data under a point transect protocol with a half-normal detection function. Density is modeled as a log-linear regression on a "habitat" covariate, with coefficients mean.density and beta.density.

Point count and occupancy data are also generated by calls to simHDSpoint, assuming that the underlying detection process involves reduced probability of detection with distance from the observer. The distances from the observer are discarded, and only the counts (for PC data) or species detection information (for occupancy) retained.

Availability is modeled according to Sólymos et al. (2013), where the probability of availability depends on the duration of the observation.

Usage

simIDS(mean.density = 1, beta.density = 1, mean.phi = 0.14, beta.phi = 0,
    nsites_HDS = 1000, sigHDS = 100, maxDist_HDS = 200, nbins = 4,
    range.dur.HDS = c(5, 5),
    nsites_PC = 10000, sigPC = 70, maxDist_PC = 500, range.dur.PC = c(3, 30),
    nsites_OC = 5000, sigOC = sigPC, maxDist_OC = maxDist_PC,
    range.dur.OC = range.dur.PC,
    show.plots = TRUE)

Arguments

mean.density	the expected value of density (animals per hectare) when the habitat covariate = 0; the intercept of the log-linear regression for density is log(mean.density).
beta.density	the slope of the log-linear regression for density on a habitat covariate.
mean.phi	the expected value of the availability parameter.
beta.phi	the slope of the log-linear regression for availability on a covariate - not yet implemented.
nsites_HDS	the number of sites (point transects) for distance sampling.
sigHDS	the value of the scale parameter of the half-normal detection function for distance sampling.
maxDist_HDS	the truncation distance for distance sampling (in meters); any observations beyond this distance are discarded.
nbins	the number of distance bins for grouping distance sampling data.
range.dur.HDS	the range of durations for distance sampling; durations for each site are simulated from a uniform distribution with this range.
nsites_PC	the number of sites for point counts; set to 0 to suppress generation of point counts.
sigPC	the value of the scale parameter of the half-normal detection function for point counts.
maxDist_PC	the maximum distance from the observer for detection of animals when conducting point counts (m).
range.dur.PC	the range of durations for point counts; durations for each site are simulated from a uniform distribution with this range.
nsites_OC	the number of sites for occupancy surveys; set to 0 to suppress generation of occupancy data.
sigOC	the value of the scale parameter of the half-normal detection function for occupancy surveys.
maxDist_OC	the maximum distance from the observer for detection of animals when conducting occupancy surveys (m).
range.dur.OC	the range of durations for point counts; durations for each site are simulated from a uniform distribution with this range.
show.plots	choose whether to show plots or not. Set to FALSE when using function in simulations.

Value

A list with the values of the arguments entered and the following additional elements:

dds	simulated durations for distance sampling: a vector length nsites_HDS with a value for each site
umf_hds0	an unmarkedFrameDS [package "unmarked"] with the simulated distance sampling data BEFORE allowing for availability
umf_hds1	an unmarkedFrameDS [package "unmarked"] with the simulated distance sampling data AFTER allowing for availability
dpc	simulated durations for point counts: a vector length nsites_PC with a value for each site
umf_pc0	an unmarkedFramePCount [package "unmarked"] with the simulated point count data BEFORE allowing for availability
umf_pc1	an unmarkedFramePCount [package "unmarked"] with the simulated point count data AFTER allowing for availability
doc	simulated durations for occupancy surveys: a vector length nsites_OC with a value for each site
umf_oc0	an unmarkedFrameOccu [package "unmarked"] with the simulated occupancy data BEFORE allowing for availability
umf_oc1	an unmarkedFrameOccu [package "unmarked"] with the simulated occupancy data AFTER allowing for availability

Note

A function to analyze these data, IDS, will appear in a future version of unmarked. In the meantime, a devel version can be installed with remotes::install_github("kenkellner/unmarked", ref="IDS").

Author(s)

Ken Kellner, Marc Kéry, Mike Meredith

References

Kéry, M. & Royle, J.A. (2016) Applied Hierarchical Modeling in Ecology AHM1 - 8.5.1.

Sólymos, P. et al. (2013) Calibrating indices of avian density from non-standardized survey data: making the most of a messy situation. Methods in Ecology and Evolution 4, 1047-1058.

Kéry, M. et al. (2022) Integrated distance sampling models for simple point counts. (Submitted to Ecology)

Examples

# Simulate a data set with the default arguments and look at the structure of the output
tmp <- simIDS()
str(tmp)

AHMbook documentation built on Sept. 12, 2024, 6:37 a.m.