Skip to contents

Thin observations

thin_observations.Rd

Removes redundant detections by grouping consecutive records within thin_minutes minutes of each other into clusters (per site and species), retaining only the record with the highest count from each cluster.

Usage

thin_observations(
  observations,
  deploymentID = deploymentID,
  scientificName = scientificName,
  eventStart = eventStart,
  count = count,
  thin_minutes = 30
)

Arguments

observations

A dataframe of observation records. Must contain columns deploymentID, eventStart, scientificName, and count (or equivalents specified via the corresponding arguments).

deploymentID

<data-masking> Column name for sites (ARUs). Default: deploymentID.

scientificName

<data-masking> Column name for species names. Default: scientificName.

eventStart

<data-masking> POSIXt. Column name for observation timestamps. Default: eventStart.

count

<data-masking> integerish.Column name for number of individuals per observation record. Default: count.

thin_minutes

Non-negative numeric. Observations within thin_minutes minutes of each other (per site and species) are thinned to a single observation, retaining the record with the highest count. Default: 30.

Value

A dataframe of thinned observation records, sorted by deploymentID, scientificName, and eventStart, or the original dataframe if thin_minutes = 0, with thin_minutes stored as attribute.

See also

make_data()