Skip to contents

Classify landcover data for simple features object

Source: R/classify_landcover.R
classify_landcover.sf.Rd

Prepare a land cover data set for use with the other functions in this package by aligning the land cover classifications with those required by the metrics data as provided in the key and corresponding predictors required for fitting species distribution models.

Usage

# S3 method for class 'sf'
classify_landcover(x, source = "LSPT", ...)

Arguments

x

object of class sf, sfc or sfg

source

character string indicating the source of the land cover polygons; currently only "LSPT" is supported

...

Unused

Details

This function provides support for cross-walking polygon-based vegetation data, such as the "Habitat_types_modern_forLSPT" layer published in the Landscape Scenario Planning Tool v3 (LSPT), to the land cover classes required for use with this package. Ultimately, this function may be updated to extend support to other land cover data sources, but it currently relies on the fields provided in that layer, especially: Habitat_Type, Source_classification, Crop2016, and, Tidal.

The function returns an sf object with additional fields:

  • CODE_NAME: most specific land cover class designation, matching those provided in the key

  • CODE_NUM: corresponding code number, matching those in the key, for creating rasters

  • PREDICTOR_RIPARIAN: corresponding predictor name used in the riparian landbird SDMs

  • PREDICTOR_WATERBIRD_FALL: corresponding predictor name used in the waterbird SDMs for the fall season

  • PREDICTOR_WATERBIRD_WIN: corresponding predictor name used in the waterbird SDMs for the winter season

  • PREDICTOR_TIMA: corresponding predictor name used in the tidal marsh bird SDMs

The CODE_NAME field can be rasterized for further analysis with the metrics data and converted to predictors for fitting the SDMs by calling this function again on the raster (see classify_landcover.SpatRaster(). Alternatively, each predictor field can be rasterized directly from this output. The results should be mostly identical, except for a few special cases for the riparian predictors; see predictors_riparian for more information.

See also

key, predictors_riparian, predictors_waterbird_win, predictors_waterbird_fall, predictors_tima