sample_recordings {ARUtools}

Sample recordings

Description

Sample recordings based on selection weights from calc_selection_weights() using spsurvey::grts().

Usage

sample_recordings(
  meta_weights,
  n,
  os = NULL,
  col_site_id = site_id,
  col_sel_weights = psel_std,
  seed = NULL,
  ...
)

Arguments

meta_weights

(Spatial) Data frame. Recording meta data selection weights. Output of calc_selection_weights(). Must have at least the columns identified by col_site_id and col_sel_weights, as well as the probability of selection columns (those starting with psel) and doy.

n

Numeric, Data frame, Vector, or List. Number of base samples to choose. For stratification by site, a named vector/list of samples per site, or a data frame with columns n for samples, n_os for oversamples and the column matching that identified by col_site_id.

os

Numeric, Vector, or List. Over sample size (proportional) or named vector/list of number of samples per site Ignored if n is a data frame.

col_site_id

Column. Unquoted column containing site strata IDs (defaults to site_id).

col_sel_weights

Column. Unquoted name of column identifying selection weights (defaults to psel_std)

seed

Numeric. Random seed to use for random sampling. Seed only applies to specific sampling events (does not change seed in the environment). NULL does not set a seed.

...

Extra named arguments passed on to spsurvey::grts().

Value

A sampling run from grts. Note that the included dataset is spatial, but is a dummy spatial dataset created by using dates and times to create the spatial landscape.

Examples

s <- clean_site_index(example_sites_clean,
  name_date_time = c("date_time_start", "date_time_end")
)
m <- clean_metadata(project_files = example_files) |>
  add_sites(s) |>
  calc_sun()

params <- sim_selection_weights()
w <- calc_selection_weights(m, params = params)

# No stratification by site
samples <- sample_recordings(w, n = 10, os = 0.1, col_site_id = NULL)

# Stratification by site defined by...

# lists
samples <- sample_recordings(w, n = list(P01_1 = 2, P02_1 = 5, P03_1 = 2), os = 0.2)

# vectors
samples <- sample_recordings(w, n = c(P01_1 = 2, P02_1 = 5, P03_1 = 2), os = 0.2)

# data frame
samples <- sample_recordings(
  w,
  n = data.frame(
    site_id = c("P01_1", "P02_1", "P03_1"),
    n = c(2, 5, 2),
    n_os = c(0, 0, 1)
  )
)