Static Maps w/ ggplot

Schwab

Libraries

library(tidyverse)
library(sf)
library(tidygeocoder)
library(maps)

tidygeocoder

Let’s find the latitude and longitude for a couple of places.

addresses<- tibble(address=c("Smith College", "2 Tyler Ct, Northampton, MA 01060", "Holyoke Community College")) |>
geocode(address, method="osm") |>
  mutate(name = c("Smith College","McConnell Hall","Holyoke Community College"))

addresses

# A tibble: 3 × 4
  address                             lat  long name                     
  <chr>                             <dbl> <dbl> <chr>                    
1 Smith College                      42.3 -72.6 Smith College            
2 2 Tyler Ct, Northampton, MA 01060  42.3 -72.6 McConnell Hall           
3 Holyoke Community College          42.2 -72.7 Holyoke Community College

My first map

Let’s plot those with ggplot.

ggplot(addresses) +
  geom_point(aes(long,lat)) 

Not very exciting. We’ll add it it in a moment.

Mass Map data

Let’s build a map of Massachusetts with the map_data() function

# map_data comes from ggplot2

ma_counties <- map_data("county", "massachusetts") |>
  select( long, lat, group, id = subregion)

head(ma_counties)

       long      lat group         id
1 -70.67435 41.73997     1 barnstable
2 -70.53683 41.79727     1 barnstable
3 -70.53683 41.79727     1 barnstable
4 -70.51392 41.78008     1 barnstable
5 -70.47954 41.75716     1 barnstable
6 -70.41078 41.73425     1 barnstable

Mass Map

Notice there are a lot of points for barnstable. Those are the vertices of a polygon. Let’s draw the polygon with the points.

ma_counties |> ggplot(aes(long,lat,group=group))+
  geom_polygon(fill = "white", colour = "grey50")+
  geom_point()  

  #coord_sf()

Mapping projections

Coordinate Reference Systems (CRS) - Is the type of map projection employed To get a 3D Earth to a 2D surface.

Which one will coord_sf default to?

?coord_sf

This is the crs commonly used in gps.

West wing explanation

West Wing

Adding the 4326 crs

ma_counties |> ggplot(aes(long,lat,group=group))+
  geom_polygon(fill = "white", colour = "grey50")+
  geom_point()+
  coord_sf()

Add some cities

Let’s grab just Massachusetts cities from the data frame us.cities.

# us.cities comes from the maps package

ma_cities <- us.cities |>
  filter(country.etc =="MA") |>
  select(name, long, lat)

head(ma_cities)

                name   long   lat
1         Andover MA -71.14 42.65
2       Arlington MA -71.16 42.42
3       Attleboro MA -71.30 41.93
4 Barnstable Town MA -70.30 41.70
5         Beverly MA -70.84 42.56
6       Billerica MA -71.26 42.56

Exercise:

Plot MA cities onto our county map. Then clean up our map.

ma_counties |> ggplot(aes(long,lat))+
  geom_polygon(fill = "white", colour = "grey50",aes(group=group)) +
  #geom_point(data=ma_cities,.......)
  
  #This is the default mapping. Think GPS
 coord_sf(default_crs= st_crs(4326) )

          # This is Mercator
          # st_crs(3857))

Shape files

Often times map data is saved as a shape file.

The US government has shape files here.