Spatial analysis with R

spatial_example.R

# An example of spatial analysis with R that involves grabbing data
# from a PostGIS database, joining to tabular data, projecting
# simplifying and creating an interactive map.

# This data is publicly available but requires some processing
# if you want the county boundaries you can find them here:

# cftp://ftp2.census.gov/geo/tiger/TIGER2016/COUNTY/

# The tabular data requires some processing but the raw data
# is here:

# https://www.census.gov/data/datasets/2016/demo/saipe/2016-state-and-county.html

# If you want to learn more about processing data with the sf
# package you can go to this DataCamp course:

# https://www.datacamp.com/courses/spatial-analysis-in-r-with-sf-and-raster


library(dplyr)
library(sf)
library(RColorBrewer)
library(mapview)

# Read the poverty data
pov <- readr::read_csv("data.csv")

# Make a connection to a local PostGIS database
con<-RPostgreSQL::dbConnect(RPostgreSQL::PostgreSQL(), 
               host="localhost",
               user= "postgres", 
               dbname="data_census")

# Read data from PostGIS to R with sf pacakge
cnty <- st_read_db(con, table = c("geo", "tl_2016_us_county"))
cnty <- mutate(cnty, geoid = paste0(statefp, countyfp))
cnty <- st_transform(cnty, crs = 2163) # project

# Original size of file
object.size(cnty) # 127 megabytes

# Simplify
cnty_simp <- st_simplify(cnty,dTolerance = 500, preserveTopology = TRUE)

# New size of file
object.size(cnty_simp) # 5 megabytes

# Link to tabular data
cnty_dat <- filter(pov, year == 2016) %>% 
  inner_join(cnty_simp, ., by = "geoid")

cnty_dat <- filter(cnty_dat, !substring(geoid,1,2)%in%c("02", "15"))

# Create an interactive map
mapviewOptions(legend.pos = "bottomright")
mapview::mapview(cnty_dat, zcol="pov_per", 
                 col.regions = colorRampPalette(brewer.pal(7, "YlOrBr")),
                 legend = TRUE,
                 layer.name = "Percent poverty",
                 lwd = 0.2)