Skip to contents

Flag records that are close to each other in the geographic space

Source: R/thin_geo.R
thin_geo.Rd

Marks occurrence records for thinning by keeping only one record per species within a radius of 'd' kilometers.

Usage

thin_geo(
  occ,
  species = "species",
  long = "decimalLongitude",
  lat = "decimalLatitude",
  d,
  prioritary_column = NULL,
  decreasing = TRUE,
  remove_invalid = TRUE,
  optimize_memory = FALSE,
  verbose = TRUE
)

Arguments

occ

(data.frame or data.table) a data frame containing the occurrence records to be flagged. Must contain columns for species, longitude, and latitude.

species

(character) the name of the column in occ that contains the species scientific names. Default is "species".

long

(character) the name of the column in occ that contains the longitude values. Default is "decimalLongitude".

lat

(character) the name of the column in occ that contains the latitude values. Default is "decimalLatitude".

d

(numeric) thinning distance in kilometers (e.g., 10 for 10km).

prioritary_column

(character) name of a numeric columns in occto define retention priority (e.g., quality score, year). See details.

decreasing

(logical) whether to sort records in decreasing order using the prioritary_column (e.g., from most recent to oldest when the variable is "year"). Only applicable when prioritary_column is not NULL. Default is TRUE.

remove_invalid

(logical) whether to remove invalid coordinates. Default is TRUE.

optimize_memory

(logical) whether to compute the distance matrix using a C++ implementation that reduces memory usage at the cost of increased computation time. Recommended for large datasets (> 10,000 records). Default is FALSE.

verbose

(logical) whether to display messages during function execution. Set to TRUE to enable display, or FALSE to run silently. Default is TRUE.

Value

The original occ data frame augmented with a new logical column named thin_geo_flag. Records that are retained after thinning receive TRUE, while records identified as too close to a higher-priority record receive FALSE.

Details

This function is similar to the thin() function from the spThin package, but with an important difference: it allows specifying a priority order for retaining records.

When a thinning distance is provided (e.g., 10 km), the function identifies clusters of records within this distance. Within each cluster, it keeps the record with the highest priority according to the column defined in prioritary_column (for example, keeping the most recent record if prioritary_column = "year"), and flags the remaining nearby records for removal.

If prioritary_column is NULL, the priority follows the original order of rows in the input occ data.frame.

Examples

# Load example data
data("occurrences", package = "RuHere")
# Subset occurrences for Araucaria angustifolia
occ <- occurrences[occurrences$species == "Araucaria angustifolia", ]
# Thin records using a 10 km distance threshold
occ_thin <- thin_geo(occ = occ, d = 10)
#> Removing 1 invalid records
sum(!occ_thin$thin_geo_flag)  # Number of records flagged for removal
#> [1] 235
# Prioritizing more recent records within each cluster
occ_thin_recent <- thin_geo(occ = occ, d = 10, prioritary_column = "year")
#> Removing 1 invalid records
sum(!occ_thin_recent$thin_geo_flag)  # Number of records flagged for removal
#> [1] 235