UK Administrative Shapefiles clipped to buildings (simplified at 100m)

This dataset includes a series of modified UK administrative boundary shapefiles based on the 2011 census which are intended for use in more accurate visualisation of UK geospatial data analysis. There are two key features of these shapefiles: (1) administrative shapes have been clipped to the Ordnance Survey buildings shapefile, so that in choropleth visualisations relating to demographic data filled spaces represent populated areas of the UK rather than large undifferentiated blocks. (2) Shapefiles have been simplified to reduce loading and processing time, in the case of this repository at 100m. After testing, we have settled on a procedure to render buildings layer visually comprehensible at high zoom levels, by adding a small buffer, dissolving (so that individual overlapping shapes combine into a single more easily visualised shape) and then simplifying at 150m. It is important to emphasise that because of the use of simplification (using a Ramer–Douglas–Peucker algorithm), these shapefiles are not suitable for analysis as boundaries may not be suitably precise or accurate. For users interested in the process used to generate these files you can consult the codebase deposited on github.

Many thanks to colleagues including Alasdair Rae for recommendations on technique used here. Computations were performed using the University of Birmingham's BEAR Cloud service, which provides flexible resource for intensive computational work to the University's research community. See http://www.birmingham.ac.uk/bear  for more details. Given the massive size of datasets involved (including the district buildings vector shapefile which is 1.4gb and consists of hundreds of thousands of individual shapes), this work would have been impossible without this invaluable resource. I hope that these files will be of use to colleagues who may not have access to similar large computational arrays and make the process of visualising UK boundary and census data more accurate and efficient.

Original files are under OGLv3 licenses. Derived data files, where possible are licensed for use under CC BY 4.0.

Files include the following:

Original unmodified data:

• infuse_ctry_2011.zip - original country level shapes, based on 2011 census, downloaded from https://borders.ukdataservice.ac.uk/ukborders/easy_download
• infuse_dist_lyr_2011.zip - original local authority shapes, based on 2011 census, downloaded from https://borders.ukdataservice.ac.uk/ukborders/easy_download
• TermsAndConditions.html - UK Data Service license details (OGLv3), applies to all the above
•  GB_Postcodes.zip - UK postcode district shapes, prepared by Addy Pope, https://datashare.ed.ac.uk/handle/10283/2597

Derived data files:

• OS_Open_Zoomstack_district_buildings.zip - buildings layer extracted from Ordnance Survey Zoomstack package, licensed under OGLv3 and exported to gpkg format.
• *_simplified_100m.gpkg - Administrative shapes from above, simplified in R at a resolution of 100 metres.
• *_simplified_100m_buildings_overlay_simplified.gpkg - Administrative shapes from above, simplified in R at a resolution of 100 metres, and then clipped to the buildings layer.
• *_simplified_100m_buildings_overlay_simplified.gpkg - Administrative shapes from above, simplified in R at a resolution of 100 metres, and then run against the buildings layer as a difference layer. Suitable for using as an overlay as the shapes are inverse.

Users who wish to use these shapefiles in a reproducible research context may want to download individual files directly from this repository. To do so, you could use the following R code:

# load packages
require(sf) # load simplefeature data class, supercedes sp() and used for st_read
# given the size and complexity even of simplified files here, ragg is highly recommended 
# for users on macos given inefficiencies in default R graphics device
require(ragg)

# create paths as needed
if (dir.exists("data") == FALSE) {
  dir.create("data")
}

# download data files only if they aren't already present
if (file.exists("data/infuse_dist_lyr_2011.shp") == FALSE) {
  download.file("https://borders.ukdataservice.ac.uk/ukborders/easy_download/prebuilt/shape/infuse_dist_lyr_2011.zip", destfile = "data/infuse_dist_lyr_2011.zip")
  unzip("infuse_dist_lyr_2011.zip", exdir = "data")}
local_authorities <- st_read("data/infuse_dist_lyr_2011.shp")