Working With Shape Files in R
Loading the Shape file in R
To load the previously downloaded Shape file containing the polygons for the NSW LGAs we will use the function readOGR() that is part of the package rgdal.
# Reads the Shape file and loads it into R
NSWLGA <- readOGR(dsn = "NSW_LGA_POLYGON_shp", layer = "NSW_LGA_POLYGON_shp")
#> OGR data source with driver: ESRI Shapefile
#> Source: "/Users/juanpablo/UTS/STDS/AT1/NSW_LGA_POLYGON_shp", layer: "NSW_LGA_POLYGON_shp"
#> with 197 features
#> It has 10 fields
NSWLGA_Df <- as.data.frame(NSWLGA@data)
head(NSWLGA_Df)
#> LG_PLY_PID DT_CREATE DT_RETIRE LGA_PID NSW_LGA_sh NSW_LGA__1
#> 0 649 2011/11/29 <NA> NSW153 2011/11/29 <NA>
#> 1 650 2011/11/29 <NA> NSW153 2011/11/29 <NA>
#> 2 651 2011/11/29 <NA> NSW153 2011/11/29 <NA>
#> 3 652 2011/11/29 <NA> NSW153 2011/11/29 <NA>
#> 4 653 2011/11/29 <NA> NSW153 2011/11/29 <NA>
#> 5 654 2011/11/29 <NA> NSW153 2011/11/29 <NA>
#> NSW_LGA__2 NSW_LGA__3 NSW_LGA__4 NSW_LGA__5
#> 0 UNINCORPORATED UNINCORPORATED <NA> 1
#> 1 UNINCORPORATED UNINCORPORATED <NA> 1
#> 2 UNINCORPORATED UNINCORPORATED <NA> 1
#> 3 UNINCORPORATED UNINCORPORATED <NA> 1
#> 4 UNINCORPORATED UNINCORPORATED <NA> 1
#> 5 UNINCORPORATED UNINCORPORATED <NA> 1Simple Shape file visualisation
Now that we have our Shape file loaded in R we can start working with it. First, I would like to plot a very simple map of the NSW LGAs using the function plot().
# Plots a simple outline of the map
## Plot already recognises the object as a map after having installed the rgdal package
plot(NSWLGA) 
Notice that the map plotted is very simple and doesn’t provide any insight or helps us to tell any data story for NSW.
Visualising especific polygons
The next step is to visualise specific LGAs of NSW. This may be useful in the case you want to put special emphasis in one location or area of NSW, for this particular case I am interested in plotting the polygons that correspond to the LGAs in the Sydney metropolitan area.
# Create a logical object with TRUE values for the LGAs in the Sydney metropolitan area
sydmetro <- NSWLGA$LGA_PID %in% c("NSW335", "NSW283", "NSW268", "NSW264", "NSW272", "NSW282", "NSW332",
"NSW315", "NSW295", "NSW325", "NSW270", "NSW234", "NSW229", "NSW331",
"NSW258", "NSW117", "NSW275", "NSW232", "NSW329", "NSW213", "NSW314",
"NSW217", "NSW216", "NSW181", "NSW196", "NSW298", "NSW200", "NSW175",
"NSW294", "NSW180")
# Plots a map of the LGAs that are part of the Sydney Metropolitan Area
plot(NSWLGA[sydmetro, ], col = "deepskyblue2", main = "Sydney Metro", sub = "GRS 80")
Lets add some color
Now that we know how to plot shape files and know how to focus on spcific polygons, we can then fill the polygons with different colors to represent for example the quantiles of a specific distribution. In this particular case we will be using the function qtm() that is included in the package tmap to plot a map filled with the quantiles of a normal distribution.
# First we create a new column in our NSWLGA data frame containing values from a normal distribution
set.seed(574)
NSWLGA_Df$Rnorm <- rnorm(nrow(NSWLGA_Df))
# Then we add the new column to the Spatial Polygon Data Frame using the function left_join() from the package dplyr.
NSWLGA@data <- left_join(NSWLGA@data, NSWLGA_Df, by = "LG_PLY_PID")
# Finally we plot the map
qtm(NSWLGA[sydmetro,], "Rnorm", title = "Sydney Metro", sub ="GRS 80")
#> Linking to GEOS 3.6.1, GDAL 2.1.3, proj.4 4.9.3
#> Variable "Rnorm" contains positive and negative values, so midpoint is set to 0. Set midpoint = NA to show the full spectrum of the color palette.
