mshmandal.github.io - Map with ggplot2

First load required packages

library(ggplot2)
library(geodata)
library(dplyr)
library(sf)

Get country vector file from GADM.org. We use gadm() function from geodata package. Here, to download a vector file we need to specify the country, level and path arguments. Read the function help document for more details

# help("gadm")
bd_level1 <-geodata::gadm(
    country = "Bangladesh", # the country name
    level = 1,              # The level of administrative subdivision requested
    path = tempdir()        # path to download
    )

The bd_level1 variable now contains our file. This is a vect class object native for the “terra” package. However, I prefer using the “sf” object for vector file types.

So, first using “st_as_sf()” lets convert it from terra “class” to “sf”

bd_sf <- bd_level1 %>% st_as_sf() # to simple feature (sf) types
# here using pipe ( %>% ) operator from dplyr package
# without pipe operator, we also could use it simply like below.

bd_sf <- st_as_sf(bd_level1) # this is equivalent to the previous code.

# but later we will use the pipe approach to write shorter 
# and easier to understand codes. But first let's check our data

head(bd_sf,2)  # check first few rows

Simple feature collection with 2 features and 11 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: 89.85194 ymin: 20.74111 xmax: 92.67366 ymax: 24.26797
Geodetic CRS:  WGS 84
    GID_1 GID_0    COUNTRY     NAME_1                        VARNAME_1
1 BGD.1_1   BGD Bangladesh    Barisal                        Bakerganj
2 BGD.2_1   BGD Bangladesh Chittagong Chattagram|Parbattya Chattagram|
  NL_NAME_1 TYPE_1 ENGTYPE_1 CC_1 HASC_1 ISO_1                       geometry
1      <NA> Bibhag  Division   10  BD.BA  <NA> MULTIPOLYGON (((90.77085 21...
2      <NA> Bibhag  Division   20  BD.CG  <NA> MULTIPOLYGON (((92.33417 20...

Next we use ggplot2 to make a map

ggplot()+
    # we specify the data inside geom_sf()
    # within the geom_sf, we specify "fill" variables to fill the polygons
    geom_sf(data = bd_sf,aes(fill=NAME_1))+
    # next line we using viridis color palettes for discrete types
    scale_fill_viridis_d(name="Divisions")+
    # specify the x,y axis namea as well as title
    labs(x="Longitude",y="Latitude",title="Bangladesh")+
    # next line not strictly necessary, as ggplot by deafult will use
    # "EPSG:4326" coordinate system for geom_sf() types geometry. 
    # But I like to specify it explicitely
    coord_sf(crs = "EPSG:4326")

To save this plot use “ggsave”

# Uncomment the following line to run the code
# ggsave("./bd_map.jpg",dpi = 300,width = 6,height = 6)