make_spatial_objects: load the R objects required for fitting spatio-temporal...

In pbs-assess/hookCompetition: An R package for estimating relative abundance indices from longline data using a censored likelihood approach

load the R objects required for fitting spatio-temporal models. These include: a delauney triangulation mesh for computing the GMRF approximation to the SPDE (Lindgren et al 2011) within r-INLA (Rue et al 2005); an INLA spde object with user-specified pc priors specified on the range and standard deviation parameters.

Description

load the R objects required for fitting spatio-temporal models. These include: a delauney triangulation mesh for computing the GMRF approximation to the SPDE (Lindgren et al 2011) within r-INLA (Rue et al 2005); an INLA spde object with user-specified pc priors specified on the range and standard deviation parameters.

Usage

make_spatial_objects(
  data,
  survey_boundaries = NULL,
  mesh_coast = NULL,
  hires_coast = NULL,
  prior_range = 20,
  prior_range_prob = 0.01,
  prior_sigma = 2,
  prior_sigma_prob = 0.01,
  nx = 150,
  ny = 150
)

Arguments

data	a sf points object containing the IPHC data
survey_boundaries	a sf polygons object containing the survey boundary definitions
mesh_coast	a simplified (smoothed) sf polygons object over which the triangulation mesh will be created. If NULL, a simplified shapefile of BC Coastline will be returned.
hires_coast	a high-resolution sf polygons object used for plotting. If NULL, a default BC coastline object will be returned.
prior_range	Choose a value in kilometers such that the species density at two locations separated by distances greater than this will be ‘approximately’ independent (typically a lower bound is chosen).
prior_range_prob	What is you prior probability that the spatial range will be smaller than this value (typically a small probability)?
prior_sigma	Choose a value for the marginal standard deviation of the random field. Typically an upper bound is used. Remember our linear predictor is on the log scale, so think about constraining exp(+- 2*SD) in terms of multiplicative regional differences.
prior_sigma_prob	What is your prior probability that the true marginal standard deviation exceeds this?
nx	How many columns of pixels to place over the survey_boundaries for plotting? Default 150, higher means smoother plots, but longer run times.
ny	How many rows of pixels to place over the survey_boundaries for plotting? Default 150, higher means smoother plots, but longer run times.

pbs-assess/hookCompetition documentation built on April 27, 2023, 12:47 p.m.