Rdistance is back on CRAN

Distance analysis package Rdistance is back on CRAN after an extended hiatus

Distance Analysis

Trent McDonald


June 23, 2023

Rdistance, a package for line and point transect distance analysis, is back on CRAN after a 2-year hiatus. Rdistance has always been available on Github. Now, it passes all CRAN checks and is easily installed using the standard install.packages routine.

A Major Update

Version 3.0.0 of Rdistance represents a major update from prior releases. Perhaps the largest change is that measurement units are now required on all distances. This means that users must now specify units like ‘m’ (meters) or ‘ft’ (feet) for off-transect distances, transect length, and other physical measurements. Rdistance’s help pages have been significantly updated and clarified. One-sided transects are now allowed, and bootstrap intervals for effective sampling distances (ESW and EDR) are now computed.

The Biggest Changes

Following are the most substantial updates in version 3.0.0 of Rdistance. A list of all updates can be found in the Rdistance NEWS file.

  • Measurement Units: Rdistance now requires units be defined on all distances (e.g., off-transect distances, strip widths, study area size, etc.). All computations and outputs are converted internally to common measurement units, and those common units can be specified by the user. Internal unit conversion is controlled by the awesome units package.
  • One-sided Transects: Rdistance version 3.0.0 adds the ability to analyze single-sided transects. Uses can now set parameter singleSided = TRUE in calls to abundEstim. One-sided transects occur most commonly during aerial surveys when observers sit on only one side of the aircraft.
  • Confidence Intervals for Effective Sampling Distances: Version 3.0.0 automatically computes bootstrap confidence intervals for effective sampling distances (ESW and EDR).
  • NA Length Transects: Version 3.0.0 allows NA length transects. Distance observations associated with NA length transects are used to estimate distance functions, but are dropped when estimating density. This allows, for example, off-transect distance observations from one study area or year to be included in distance function estimation but not included in density estimates. In situations where sightings are rare, it is common to pool detection distances across study areas or years even though density is estimated for individual study areas or years.