Skip to contents

Download and process Global Road Density layers

Source: R/roads.R
roads.Rd

This function downloads, processes, and extracts road density variables. Each variable corresponds to a global raster (~1 km resolution) reporting the total length of roads (in metres) within each grid cell, for a single road class or for a group of classes.

Usage

roads(x, vars = "all", ...)

Arguments

x

The output from `par_set()` defining the area or locations for extraction, the reference system, and the buffer. Leave this empty and use `par_set()` to define parameters for download.

vars

Character vector of one or more variables to download and process. Defaults to "all" (all road classes combined).

...

Additional arguments (currently unused).

Value

If `par_set()` contained a raster/polygon/points with buffer: a `SpatRaster` stack of processed variables. If `par_set()` contained spatial points or data.frame of points without buffer: a `data.frame` of x, y, and extracted values.

Details

The dataset
Road density at 1 km grid resolution over the globe, derived from the OpenStreetMap database (https://www.openstreetmap.org/) accessed through the GeoFabrik (https://www.geofabrik.de/) functionalities, as of 30 January 2026. Out of the original OpenStreetMap categories, roads were classified into five classes:

  • class1 - highway (sum of the original categories "motorway" and "motorway_link")

  • class2 - primary (sum of "primary", "primary_link", "trunk" and "trunk_link")

  • class3 - secondary (sum of "secondary" and "secondary_link")

  • class4 - tertiary (sum of "tertiary" and "tertiary_link")

  • class5 - quaternary and other (sum of "residential", "living_street", "unknown" and "unclassified")

These five classes are further grouped into three aggregated layers:

  • primary - sum of classes 4 and 5

  • other - sum of classes 1, 2 and 3

  • all - sum of all five classes

Note that the aggregated layer "primary" is a group of the minor-road classes (4 and 5) and is not the same as the single class "class2" (the OpenStreetMap "primary" category), while the aggregated layer "other" groups the major-road classes (1, 2 and 3). In the three aggregated layers, cells with no road cover are already stored as NA (rather than 0) to streamline download and analyses; the five single-class layers keep their original values.

All layers report the length in metres of roads within each ~1 km cell (i.e. road density per grid cell).

Available variables (working synonyms in parentheses):

  • "class1" ("class 1", "roads 1", "road class 1", "highways", "highway", "motorway")

  • "class2" ("class 2", "roads 2", "road class 2", "primary class", "trunk")

  • "class3" ("class 3", "roads 3", "road class 3", "secondary roads", "secondary")

  • "class4" ("class 4", "roads 4", "road class 4", "tertiary roads", "tertiary")

  • "class5" ("class 5", "roads 5", "road class 5", "quaternary", "local roads", "local", "residential")

  • "primary" ("primary roads", "primary group", "classes 4 and 5")

  • "other" ("other roads", "other group", "classes 1 2 and 3")

  • "all" ("all roads", "total", "total roads", "all classes", "combined")

If `vars` is not specified, only "all" is downloaded.

Data source:
Data are hosted in an embargoed Figshare repository and are retrieved through private links. Access is provided for the use of this package while the repository is under embargo; please check with the data authors before redistributing the layers.

Note: Data extent is [-180, 180, -60, 84].

Examples

# \donttest{
# Example 1: Download total road density for Italy
processed <- par_set(country = "Italy", crs = 3035) %>%
roads()

# Example 2: Download single road classes
processed <- par_set(country = "Italy", crs = 3035) %>%
roads(vars = c("highways", "class5"))

# Example 3: Download the two aggregated groups of classes
processed <- par_set(country = "Italy", crs = 3035) %>%
roads(vars = c("primary", "other"))
  # }