Simple Features

Simple feature standard

  • Simple features is an open standard (ISO 19125-1:2004) developed and endorsed by the Open Geospatial Consortium (OGC)
  • The standard is widely implemented in spatial databases (such as PostGIS), desktop GIS (such as ArcGIS, QGIS), scripting languages (such as R, Python) and forms the vector data basis for libraries such as GDAL.

What is a feature?

The standard says:

A simple feature is defined […] to have both spatial and non-spatial attributes. Spatial attributes are geometry valued, and simple features are based on 2D geometry with linear interpolation between vertices.

  • A feature is thought of as a thing / an object in the real world, such as a building or a tree.
  • Features have:
    1. a geometry describing where on Earth the feature is located
    2. attributes, which describe other properties.
  • For example:
    1. The geometry of a tree can be the delineation of its crown, of its stem, or the point indicating its center
    2. attributes (properties) may include its height, color, diameter at breast height at a particular date, and so on

Simple Feature Model

  • Simple features is a hierarchical data model that represents a wide range of geometry types.
  • All geometries are composed of points in a 2-, 3- or 4-dimensional space
  • Of 18 geometry types supported by the specification, only the following seven (see Figure 1.1 and Table 1.1) are used in the vast majority of geographic research:
    • Three basic types: points, linestrings, polygons
    • Three composite types: mutlipoints, multilinestrings, multipolygons
    • One special case: geometrycollection (which can be a conglomarate of all the afore mentioned)
  • These seven core geometry types are fully supported by the R package sf (Pebesma 2018)
Figure 1.1: Image source: Dorman (2023)
Table 1.1: Source: Pebesma (2018)
Type Description # of Dimension
POINT zero-dimensional geometry containing a single point 0
LINESTRING sequence of points connected by straight1 line pieces 1
POLYGON sequence of points form a closed2 ring3 2
MULTIPOINT set of points 0
MULTILINESTRING set of linestrings 1
MULTIPOLYGON set of polygons 2
GEOMETRYCOLLECTION set of geometries of any of the above types NA

Simple features in R

Simple Features in R is modelled in three levels:

  1. Simple feature geometries (sfg): Individual Simple Feature objects
  2. Simple Feature geometry columns (sfc): A list column of sfgs
  3. Simple Features (with attributes): A sfc with attributes, i.e. additional columns

sfg: simple feature geometry

  • Simple feature geometries are implemented as R native data, using the following rules:
    1. a single POINT is a numeric vector
    2. a set of points, e.g. in a LINESTRING or ring of a POLYGON is a matrix, each row containing a point
    3. any other set is a list

(however, creator functions are rarely used in practice, since we typically bulk read and write spatial data. They are useful for illustration)

Points

library(sf)
fracht <- st_point(c(2685374, 1256519))

fracht
plot(fracht)

Linestrings

coords <- c(
  2684336, 1255553, 
  2682705, 1258929
  ) |> 
  matrix(ncol = 2, byrow = TRUE)

coords
        [,1]    [,2]
[1,] 2684336 1255553
[2,] 2682705 1258929
piste <- st_linestring(coords)

piste
plot(piste)

Polygons

coords_2 <- c(
  2684142, 1255702, # ↰ 
  2685600, 1256958, # start and end must
  2682534, 1259699, # be identical (closed)
  2684142, 1255702  # ↲ 
) |> 
  matrix(ncol = 2, byrow = TRUE) |> 
  list()

coords_2
[[1]]
        [,1]    [,2]
[1,] 2684142 1255702
[2,] 2685600 1256958
[3,] 2682534 1259699
[4,] 2684142 1255702
flughafen <- st_polygon(coords_2)

flughafen
plot(flughafen)

sfc: Simple feature geometry columns

  • Usually, a sfc contains more than one sfg.
  • This is not mandatory, and for convenience we will just use a single sfg to showcase sfc
  • As you might have noticed, we didn’t specify a CRS when creating the sfg objects. This is because we weren’t able to. With sfc, we can (and should) specify a crs.
fracht_sfc <- st_sfc(fracht, crs = 2056)

fracht_sfc
Geometry set for 1 feature 
Geometry type: POINT
Dimension:     XY
Bounding box:  xmin: 2685374 ymin: 1256519 xmax: 2685374 ymax: 1256519
Projected CRS: CH1903+ / LV95
piste_sfc <- st_sfc(piste, crs = 2056)

piste_sfc
Geometry set for 1 feature 
Geometry type: LINESTRING
Dimension:     XY
Bounding box:  xmin: 2682705 ymin: 1255553 xmax: 2684336 ymax: 1258929
Projected CRS: CH1903+ / LV95
flughafen_sfc <- st_sfc(flughafen, crs = 2056)

flughafen_sfc
Geometry set for 1 feature 
Geometry type: POLYGON
Dimension:     XY
Bounding box:  xmin: 2682534 ymin: 1255702 xmax: 2685600 ymax: 1259699
Projected CRS: CH1903+ / LV95

sf: Simple features

  • Simple features have attributes
  • In order to add attributes to a simple feature column (sfc), we need to convert it to a simple feature (sf)
  • A sf object is a subset of a data.frame. Many (all?) things which can be done with a data.frame, can be done with an sf object
  • Many tidyverse functions are implemented to nicely work with sf
  • Note: Geometries are sticky. This means they aren’t dropped implicitly
fracht_sf <- st_as_sf(fracht_sfc)

fracht_sf$name <- "Ost"
# sf objects are also dataframes
is.data.frame(fracht_sf)
[1] TRUE
# Geometries are sticky
fracht_sf[, "name"] 
Simple feature collection with 1 feature and 1 field
Geometry type: POINT
Dimension:     XY
Bounding box:  xmin: 2685374 ymin: 1256519 xmax: 2685374 ymax: 1256519
Projected CRS: CH1903+ / LV95
  name                       x
1  Ost POINT (2685374 1256519)
piste_sf <- st_as_sf(piste_sfc)

piste_sf$nr <- 34

flughafen_sf <- st_as_sf(flughafen_sfc)

flughafen_sf$name <- "Flughafen Zürich"

flughafen_sf
Simple feature collection with 1 feature and 1 field
Geometry type: POLYGON
Dimension:     XY
Bounding box:  xmin: 2682534 ymin: 1255702 xmax: 2685600 ymax: 1259699
Projected CRS: CH1903+ / LV95
                               x             name
1 POLYGON ((2684142 1255702, ... Flughafen Zürich

Plotting simple features

  • Various libraries support plotting sf objects:
    • base R (sf provides a plot-method)
    • General purpose libraries such as ggplot2
    • Dedicated geospatial plotting libraries such as tmap
  • Use:
    • base R: If you want to take a quick look at your data. Base R has the most compact syntax and is extremely fast in plotting
    • ggplot2: If you only have (small-ish) vector data (no raster) and/or want to leverage the power of ggplot2
    • tmap: If you want to use all features a dedicated library for geospatial data has to offer: North arrow, scale bar, interactive (web) maps
#
# Using base R
plot(st_geometry(flughafen_sf))
plot(piste_sf, add = TRUE)
plot(fracht_sf, add = TRUE)
#
# Using ggplot2
library(ggplot2)
ggplot() +
  geom_sf(data = flughafen_sf) +
  geom_sf(data = piste_sf) +
  geom_sf(data = fracht_sf)
#
# Using tmap
library(tmap)
tm_shape(flughafen_sf) + tm_polygons() +
  tm_shape(piste_sf) + tm_lines() +
  tm_shape(fracht_sf) + tm_dots()

base R

library ggplot2

library tmap
  • The tmap library can render the map either in a static plot as above (the default) or as an interactive web map (see below)
# set tmap_mode to "view" for an interactive web map
tmap_mode("view")

tm_shape(flughafen_sf) + tm_polygons() +
  tm_shape(piste_sf) + tm_lines() +
  tm_shape(fracht_sf) + tm_dots()

🪕 Tasks

  1. Create some simple feature geometries (sfg) of objects you know “by hand”
  • Create at least one POINT, one LINESTRING and one POLYGON geometry
  • You can capture the coordinates of the nodes from map.geo.admin if these are is Switzerland and openstreetmap (or similar) if they aren’t
  1. Create simple feature columns from you sfgs. Make sure that you assign the correct CRS
  2. Create simple features (sf) from your sfcs and add some attributes

  1. non-selfintersecting↩︎

  2. non-selfintersecting↩︎

  3. the first ring denotes the exterior ring, zero or more subsequent rings denote holes in this exterior ring↩︎