Overview

The distanceto package is designed to quickly sample distances from points features to other vector layers. Normally the approach for calculating distance to (something) involves generating distance surfaces using raster based approaches eg. raster::distance or gdal_proximity and subsequently point sampling these surfaces. Since raster based approaches are a costly method that frequently leads to memory issues or long and slow run times with high resolution data or large study sites, we have opted to compute these distances using vector based approaches. As a helper, there’s a decidedly low-res raster based approach for visually inspecting your region’s distance surface. But the workhorse is distance_to.

The distanceto package provides two functions:

distance_to
distance_raster

Install

# Enable the robitalec universe
options(repos = c(
    robitalec = 'https://robitalec.r-universe.dev',
    CRAN = 'https://cloud.r-project.org'))

# Install distanceto
install.packages('distanceto')

`distance_to`

library(distanceto)
library(sf)
#> Linking to GEOS 3.12.1, GDAL 3.8.4, PROJ 9.4.0; sf_use_s2() is TRUE

Long-lat / unprojected coordinates

# Load nc data
nc <- st_read(system.file("shape/nc.shp", package="sf"))
#> Reading layer `nc' from data source `/github/workspace/pkglib/sf/shape/nc.shp' using driver `ESRI Shapefile'
#> Simple feature collection with 100 features and 14 fields
#> Geometry type: MULTIPOLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -84.32385 ymin: 33.88199 xmax: -75.45698 ymax: 36.58965
#> Geodetic CRS:  NAD27

# Set number of sampling points
npts <- 1e3

# Sample points in nc
ncpts <- st_sample(nc, npts)

# Select first 5 of nc
ncsub <- nc[1:5,]

# Measure distance from ncpts to first 5 of nc
dist <- distance_to(ncpts, ncsub, measure = 'geodesic')

# or add to ncpts
ncpts$dist <- dist

head(dist, 30)
#>  [1]  76158.89 254412.95 107602.44  71285.77 196834.15 246284.98 114131.38
#>  [8] 127865.03 105894.85 131787.28 226724.24  96505.14  87023.11  11718.99
#> [15] 116266.01  80358.29 174489.64 212503.50      0.00  18795.10  59375.03
#> [22]  84271.03  35701.13 121144.63 114161.99  54626.76   5190.77 133178.95
#> [29]  92771.40 122502.83
hist(dist)

Projected coordinates

# Transform nc data to local projected coordinates (UTM 18N)
nc_utm <- st_transform(nc, 32618)

# Set number of sampling points
npts <- 1e2

# Sample points within nc data
nc_utm_pts <- st_sample(nc_utm, npts)

# Select one polygon within nc data
nc_utm_select <- nc_utm[1, ]

# Measure distance from seine points to seine
dist <- distance_to(nc_utm_pts, nc_utm_select)

# or add to seine points
nc_utm_pts$dist <- dist

head(dist, 30)
#>  [1]   8180.195 104933.379 295498.963 120748.089 388455.066 337473.837
#>  [7]  86273.899  28105.740 127427.339 289570.017  66051.274 112201.142
#> [13] 206097.699 321682.102 261371.640 482740.166 121025.955 239569.611
#> [19] 104886.889 261447.352 164956.330 266604.771  79150.919 414226.024
#> [25] 388698.535  92125.719 424134.391 382234.571 125786.804 246685.506
hist(dist)

`distance_raster`

library(raster)
#> Loading required package: sp

rdist <- distance_raster(nc_utm_select, 1e4, extent = st_bbox(nc_utm))

plot(rdist)

Introduction to distance-to