Want to see the code? Click on the black boxes on the right to show/hide the code.
Lines. Lines can be boundaries - they can show us where one thing stops and another starts.
If you travel the length of the UK, from North to South, you might notice something about placenames - the word Kirk is replaced by the word Church.
For today’s map we’re going to use the power of data visualisations to find out where Kirk become Church!
For this map we’ll use OpenNames - a dataset containing named places provided by the Ordnance Survey. You can get the data as a GeoPackage from OS Data Hub
Let’s read in the places from the geopackage, pulling out all the towns, villages etc where the name starts or ends with ‘church’ or ‘kirk’ and display them on a map…
#First, we'll need to load a bunch of libraries so we can handle and view geospatial data
library('sf')
library('raster')
library('mapview')
library('terra')
# We can use the st_read function from the sf library to read in the data from the geopackage
# and limit it to populated places where the name starts or ends with 'church' or 'kirk'
# using a simple SQL query-like syntax
churchAndKirk <- st_read('./data/02/opname_gb.gpkg', quiet = TRUE, query='SELECT geometry, name1 as placeName from named_place where type = "populatedPlace" and (name1 like "kirk%" or name1 like "%kirk" or name1 like "church%" or name1 like "%church")')
# Next we'll create a new field called 'nameType' and fill this with either 'Church' or 'Kirk'
# depending on the placename
churchAndKirk$nameType <- ifelse((startsWith(churchAndKirk$placeName,"Church") | endsWith(churchAndKirk$placeName,"hurch")), "Church", "Kirk")
mapviewOptions(basemaps = c("Esri.WorldShadedRelief"),
layers.control.pos = "topright",
leafletWidth = "100%",
legend = TRUE
)
mapview(churchAndKirk, zcol = c("nameType"), layer.name="Placename type", label = "placeName") So we can see some patterning there, but point maps like this can make it challenging to discern patterns or identify areas of high or low density.
Density maps mitigate this issue by aggregating data into continuous regions, reducing visual clutter. We can create a density map showing which areas are are mostly ‘Kirk’ and which are mostly ‘Church’
# First we'll create a raster 'template' that extends to the bounds of our churchAndKirk
# data frames. This splits the area into 40km x 40km squares and initialises it with zeros.
rasterTemplate = rast(ext(churchAndKirk), resolution = 40000,
crs = st_crs(churchAndKirk)$wkt, vals=0)
# Next we'll create two rasters - one where the value of each cell is the number of 'Kirk' places,
# one where it's the number of 'Church' places in each square.
kirkRaster = rasterize(subset(churchAndKirk, nameType == "Kirk"), rasterTemplate, fun="length")
kirkRaster <- classify(kirkRaster, cbind(NA, 0))
churchRaster = rasterize(subset(churchAndKirk, nameType == "Church"), rasterTemplate, fun="length")
churchRaster <- classify(churchRaster, cbind(NA, 0))
# Finally we'll combine these two rasters into a single raster - the value of each cell will be
# based on whichever contains the most Churches or Kirks. If there are more Churches than Kirks
# we'll assign it a value of 1, if there are more Kirks than Churches we'll assign it a value of 2.
nameType <- ifel((churchRaster > kirkRaster), 1,
ifel((churchRaster < kirkRaster), 2, NA))
# We'll also set the names of the levels to 'Mostly Churches' and 'Mostly Kirks'
levels(nameType) = data.frame(value = c(1,2), name = c("Mostly Churches", "Mostly Kirks"))
densityMap = mapview(nameType,na.color="transparent",layer.name="Placenames")
densityMapSo we can see a little more clearly where the ‘Church’ and ‘Kirk’ placenames are. And if we look at the map we can see that there’s a bit of a divide between the two.
If you know a bit about the history of the UK, you might know that there was a time when the country was divided into two parts - the Danelaw and the Saxon Kingdoms.
The Danelaw represented areas where Danish (ie “Viking”) legal systems were enforced, often in contrast to Anglo-Saxon laws of the time.
I’ve created a geopackage that contains the boundary of the Danelaw - let’s overlay this on our map to see if there’s any correlation
And here we have it - a map showing the Danelaw and how, despite being 1000 years old, it left a lasting impact on the UK particularly in terms of language, culture, and place names. Many English towns and cities in the Danelaw area still bear names of Scandinavian origin, especially those ’Kirk’s!
Data: Ordnance Survey Open Places OGL https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/