Simplifying polygons in rgeos and maintaining data in SpatialPolygonsDataFrame

Question

Background

I'm interested in simplifying polygons with use of the gSimplify function available through the rgeos package.

Reproducible example

A reproducible example can be generated with use of the code below:

# Data sourcing -----------------------------------------------------------

# Download an read US state shapefiles
tmp_shps <- tempfile()
tmp_dir <- tempdir()
download.file(
    "http://www2.census.gov/geo/tiger/GENZ2014/shp/cb_2014_us_state_20m.zip",
    tmp_shps
)
unzip(tmp_shps, exdir = tmp_dir)

# Libs
require(rgdal)
require(rgeos)

# Read
us_shps <- readOGR(dsn = tmp_dir, layer = "cb_2014_us_state_20m")

# Simplified --------------------------------------------------------------

# Simplifiy
us_shps_smpl <- gSimplify(spgeom = us_shps,
                          tol = 200,
                          topologyPreserve = TRUE)

Preview

par(mfrow = c(2,1))
plot(us_shps_smpl, main = "Simplified")
plot(us_shps, main = "Original")

Problem

In addittion to simplifying polygons the gSimplify function changed classes of the resulting object:

>> class(us_shps)
[1] "SpatialPolygonsDataFrame"
attr(,"package")
[1] "sp"
>> class(us_shps_smpl)
[1] "SpatialPolygons"
attr(,"package")
[1] "sp"

>> names(us_shps)
[1] "STATEFP"  "STATENS"  "AFFGEOID" "GEOID"    "STUSPS"   "NAME"     "LSAD"     "ALAND"    "AWATER"  
>> names(us_shps_smpl)
 [1] "0"  "1"  "2"  "3"  "4"  "5"  "6"  "7"  "8"  "9"  "10" "11" "12" "13" "14" "15" "16" "17" "18" "19"
[21] "20" "21" "22" "23" "24" "25" "26" "27" "28" "29" "30" "31" "32" "33" "34" "35" "36" "37" "38" "39"
[41] "40" "41" "42" "43" "44" "45" "46" "47" "48" "49" "50" "51"

Questions

• How can I safely reattached the data that was initially available in the original object and transform the resulting SpatialPolygons object to a SpatialPolygonsDataFrame

• I reckon that one approach would simply involve attaching data frame;but this depends on the order of elements not changing. Are there any other better approaches (ideally preserving initial object class)?

Answer 1

The sf package is based entirely on data frames, so its geometry manipulations always preserve the data attached to each feature. The package hasn't caught up with all the standard spatial packages in R yet, but it's fairly easy to go back and forth between sf and sp objects when you need more functionality.

Here, st_simplify() does the work, but you'll need to project your polygons first:

library(sf)

# Download and read example data
tmp_shps <- tempfile()
tmp_dir <- tempdir()
download.file(
  "http://www2.census.gov/geo/tiger/GENZ2014/shp/cb_2014_us_state_20m.zip",
  tmp_shps
)
unzip(tmp_shps, exdir = tmp_dir)

us_shps <- st_read(paste(tmp_dir, "cb_2014_us_state_20m.shp", sep = "/"))

# st_simplify needs a projected CRS
us_shps_merc <- st_transform(us_shps, 3857)
simple_us_merc <- st_simplify(us_shps_merc)

# Change back to original CRS
simple_us <- st_transform(simple_us_merc, st_crs(us_shps))

# Change to sp object, if you like
simple_us_sp <- as(st_zm(simple_us), "Spatial")