Learn R Programming

s2

The s2 R package provides bindings to Google’s S2Geometry library. The package exposes an API similar to Google’s BigQuery Geography API, whose functions also operate on spherical geometries. Package sf uses this package by default for nearly all its geometrical operations on objects with ellipsoidal (unprojected) coordinates; in cases where it doesn’t, such as st_relate(), it emits a warning.

This package is a complete rewrite of an earlier CRAN package s2 with versions up to 0.4-2, for which the sources are found here.

Installation

You can install the released version of s2 from CRAN with:

install.packages("s2")

And the development version from GitHub with:

# install.packages("remotes")
remotes::install_github("r-spatial/s2")

Example

The s2 package provides geometry transformers and predicates similar to those found in GEOS, except instead of assuming a planar geometry, s2’s functions work in latitude and longitude and assume a spherical geometry:

library(s2)

s2_contains(
  # polygon containing much  of the northern hemisphere
  "POLYGON ((-63.5 44.6, -149.75 61.20, 116.4 40.2, 13.5 52.51, -63.5 44.6))",
  # ...should contain the north pole
  "POINT (0 90)"
)
#> [1] TRUE

The sf package uses s2 for geographic coordinates with sf::sf_use_s2(TRUE), and will become the default after sf version 1.0.0. The sf package also supports creating s2 vectors using as_s2_geography():

library(dplyr)
library(sf)

nc_s2 <- read_sf(system.file("shape/nc.shp", package = "sf")) %>% 
  mutate(geometry = as_s2_geography(geometry)) %>% 
  as_tibble() %>% 
  select(NAME, geometry)

nc_s2
#> # A tibble: 100 × 2
#>    NAME        geometry                                                         
#>    <chr>       <s2_geography>                                                   
#>  1 Ashe        POLYGON ((-81.4528885 36.2395859, -81.4310379 36.2607193, -81.41…
#>  2 Alleghany   POLYGON ((-81.1766739 36.4154434, -81.1533661 36.4247398, -81.13…
#>  3 Surry       POLYGON ((-80.4530106 36.2570877, -80.4353104 36.5510445, -80.61…
#>  4 Currituck   MULTIPOLYGON (((-75.9419327 36.2943382, -75.9575119 36.2594528, …
#>  5 Northampton POLYGON ((-77.1419601 36.4170647, -77.1393204 36.4564781, -77.12…
#>  6 Hertford    POLYGON ((-76.7074966 36.2661324, -76.7413483 36.3151665, -76.92…
#>  7 Camden      POLYGON ((-76.0173492 36.3377304, -76.0328751 36.3359756, -76.04…
#>  8 Gates       POLYGON ((-76.46035 36.3738976, -76.5024643 36.4522858, -76.4983…
#>  9 Warren      POLYGON ((-78.1347198 36.2365837, -78.1096268 36.2135086, -78.05…
#> 10 Stokes      POLYGON ((-80.0240555 36.5450249, -80.0480957 36.5471344, -80.43…
#> # … with 90 more rows

Use accessors to extract information about geometries:

nc_s2 %>% 
  mutate(
    area = s2_area(geometry),
    perimeter = s2_perimeter(geometry)
  )
#> # A tibble: 100 × 4
#>    NAME        geometry                                           area perimeter
#>    <chr>       <s2_geography>                                    <dbl>     <dbl>
#>  1 Ashe        POLYGON ((-81.4528885 36.2395859, -81.4310379 3… 1.14e9   141627.
#>  2 Alleghany   POLYGON ((-81.1766739 36.4154434, -81.1533661 3… 6.11e8   119876.
#>  3 Surry       POLYGON ((-80.4530106 36.2570877, -80.4353104 3… 1.42e9   160458.
#>  4 Currituck   MULTIPOLYGON (((-75.9419327 36.2943382, -75.957… 6.94e8   301644.
#>  5 Northampton POLYGON ((-77.1419601 36.4170647, -77.1393204 3… 1.52e9   211794.
#>  6 Hertford    POLYGON ((-76.7074966 36.2661324, -76.7413483 3… 9.68e8   160780.
#>  7 Camden      POLYGON ((-76.0173492 36.3377304, -76.0328751 3… 6.16e8   150430.
#>  8 Gates       POLYGON ((-76.46035 36.3738976, -76.5024643 36.… 9.03e8   123170.
#>  9 Warren      POLYGON ((-78.1347198 36.2365837, -78.1096268 3… 1.18e9   141073.
#> 10 Stokes      POLYGON ((-80.0240555 36.5450249, -80.0480957 3… 1.23e9   140583.
#> # … with 90 more rows

Use predicates to subset vectors:

nc_s2 %>% 
  filter(s2_contains(geometry, "POINT (-80.9313 35.6196)"))
#> # A tibble: 1 × 2
#>   NAME    geometry                                                              
#>   <chr>   <s2_geography>                                                        
#> 1 Catawba POLYGON ((-80.9312744 35.6195908, -81.0035782 35.6970558, -81.0547791…

Use transformers to create new geometries:

nc_s2 %>% 
  mutate(geometry = s2_boundary(geometry))
#> # A tibble: 100 × 2
#>    NAME        geometry                                                         
#>    <chr>       <s2_geography>                                                   
#>  1 Ashe        LINESTRING (-81.4528885 36.2395859, -81.4310379 36.2607193, -81.…
#>  2 Alleghany   LINESTRING (-81.1766739 36.4154434, -81.1533661 36.4247398, -81.…
#>  3 Surry       LINESTRING (-80.4530106 36.2570877, -80.4353104 36.5510445, -80.…
#>  4 Currituck   MULTILINESTRING ((-75.9419327 36.2943382, -75.9575119 36.2594528…
#>  5 Northampton LINESTRING (-77.1419601 36.4170647, -77.1393204 36.4564781, -77.…
#>  6 Hertford    LINESTRING (-76.7074966 36.2661324, -76.7413483 36.3151665, -76.…
#>  7 Camden      LINESTRING (-76.0173492 36.3377304, -76.0328751 36.3359756, -76.…
#>  8 Gates       LINESTRING (-76.46035 36.3738976, -76.5024643 36.4522858, -76.49…
#>  9 Warren      LINESTRING (-78.1347198 36.2365837, -78.1096268 36.2135086, -78.…
#> 10 Stokes      LINESTRING (-80.0240555 36.5450249, -80.0480957 36.5471344, -80.…
#> # … with 90 more rows

Finally, use the WKB or WKT exporters to export to sf or some other package:

nc_s2 %>% 
  mutate(geometry = st_as_sfc(s2_as_binary(geometry))) %>% 
  st_as_sf()
#> Simple feature collection with 100 features and 1 field
#> Geometry type: GEOMETRY
#> Dimension:     XY
#> Bounding box:  xmin: -84.32385 ymin: 33.88199 xmax: -75.45698 ymax: 36.58965
#> CRS:           NA
#> # A tibble: 100 × 2
#>    NAME                                                                 geometry
#>    <chr>                                                              <GEOMETRY>
#>  1 Ashe        POLYGON ((-81.45289 36.23959, -81.43104 36.26072, -81.41233 36.2…
#>  2 Alleghany   POLYGON ((-81.17667 36.41544, -81.15337 36.42474, -81.1384 36.41…
#>  3 Surry       POLYGON ((-80.45301 36.25709, -80.43531 36.55104, -80.61105 36.5…
#>  4 Currituck   MULTIPOLYGON (((-75.94193 36.29434, -75.95751 36.25945, -75.9137…
#>  5 Northampton POLYGON ((-77.14196 36.41706, -77.13932 36.45648, -77.12733 36.4…
#>  6 Hertford    POLYGON ((-76.7075 36.26613, -76.74135 36.31517, -76.92408 36.39…
#>  7 Camden      POLYGON ((-76.01735 36.33773, -76.03288 36.33598, -76.04395 36.3…
#>  8 Gates       POLYGON ((-76.46035 36.3739, -76.50246 36.45229, -76.49834 36.50…
#>  9 Warren      POLYGON ((-78.13472 36.23658, -78.10963 36.21351, -78.05835 36.2…
#> 10 Stokes      POLYGON ((-80.02406 36.54502, -80.0481 36.54713, -80.43531 36.55…
#> # … with 90 more rows

Acknowledgment

This project gratefully acknowledges financial support from the

Copy Link

Version

Install

install.packages('s2')

Monthly Downloads

236,052

Version

1.1.7

License

Apache License (== 2.0)

Issues

Pull Requests

Stars

Forks

Maintainer

Last Published

July 17th, 2024

Functions in s2 (1.1.7)

s2_data_tbl_countries

Low-resolution world boundaries, timezones, and cities
s2_data_example_wkt

Example Geometries
s2_plot

Plot S2 Geographies
s2_point

Create an S2 Point Vector
wk_handle.s2_geography

Low-level wk filters and handlers
s2-package

s2: Spherical Geometry Operators Using the S2 Geometry Library
s2_options

Geography Operation Options
s2_lnglat

Create an S2 LngLat Vector
s2_cell_union

Create S2 Cell Union vectors
s2_cell_is_valid

S2 cell operators
s2_cell_union_normalize

S2 cell union operators
s2_bounds_cap

Compute feature-wise and aggregate bounds
s2_closest_feature

Matrix Functions
s2_cell

Create S2 Cell vectors
s2_boundary

S2 Geography Transformations
s2_contains

S2 Geography Predicates
as_s2_geography

Create an S2 Geography Vector
s2_earth_radius_meters

Earth Constants
s2_geog_point

Create and Format Geography Vectors
s2_project

Linear referencing
s2_is_collection

S2 Geography Accessors