Added new Trafford Creative Clusters custom geographies and amended the root README to better explain the contents of this folder.

Author: itsozz

council_defined/2024/trafford_creative_clusters.R

@@ -0,0 +1,85 @@
+# Council-defined geography: Trafford Creative Clusters
+# -----------------------------------------------------
+
+# 2024-07-29
+# These are a council-defined construct grouping the electoral wards into 10 groups.
+# They are aligned to the new ward boundaries which came into effect from May 2023.
+# They have been created as part of the Cultural Transformation work and link to the "Culture, sport and heritage for everyone" priority defined in the Trafford Corporate Plan "Our Trafford, Our Future 2024-27"
+# https://www.trafford.gov.uk/about-your-council/strategies-plans-and-policies/Corporate-Plan/Corporate-Plan-2024-27.pdf
+
+
+# load necessary packages
+library(sf) ; library(tidyverse) ; library(nngeo) ;
+
+
+# Full resolution
+# Source: https://geoportal.statistics.gov.uk/datasets/ons::wards-may-2023-boundaries-uk-bfc/about
+wards_full_res <- read_sf("https://services1.arcgis.com/ESMARspQHYMw9BZ9/arcgis/rest/services/WD_MAY_2023_UK_BFC/FeatureServer/0/query?where=LAD23NM%20%3D%20'TRAFFORD'&outFields=*&outSR=4326&f=json") %>%
+    select(area_code = WD23CD, area_name = WD23NM, lon = LONG, lat = LAT, area = Shape__Area)
+
+# Generalised
+# Source: https://geoportal.statistics.gov.uk/datasets/ons::wards-may-2023-boundaries-uk-bgc/about
+wards_generalised_res <- read_sf("https://services1.arcgis.com/ESMARspQHYMw9BZ9/arcgis/rest/services/WD_MAY_2023_UK_BGC/FeatureServer/0/query?where=LAD23NM%20%3D%20'TRAFFORD'&outFields=*&outSR=4326&f=json") %>%
+    select(area_code = WD23CD, area_name = WD23NM, lon = LONG, lat = LAT, area = Shape__Area)
+
+# Super Generalised
+# Source: https://geoportal.statistics.gov.uk/datasets/ons::wards-may-2023-boundaries-uk-bsc/about
+wards_super_generalised_res <- read_sf("https://services1.arcgis.com/ESMARspQHYMw9BZ9/arcgis/rest/services/WD_MAY_2023_UK_BSC/FeatureServer/0/query?where=LAD23NM%20%3D%20'TRAFFORD'&outFields=*&outSR=4326&f=json") %>%
+    select(area_code = WD23CD, area_name = WD23NM, lon = LONG, lat = LAT, area = Shape__Area)
+
+
+# =========
+# VERY IMPORTANT NOTE REGARDING SF PACKAGE AND COORDINATE WINDING ORDER 2023-12-21:
+# The IETF standard for GeoJSON has made certain changes over the original non-IETF specification.  The changes can be viewed here: https://datatracker.ietf.org/doc/html/rfc7946#appendix-B
+# One key change is that polygon rings MUST follow the right-hand rule for orientation (counter-clockwise external rings, clockwise internal rings).
+# This change has caused issues with certain libraries which have historically used the left-hand rule, i.e. clockwise for outer rings and counter-clockwise for interior rings.
+# D3-Geo, Vega-Lite and versions of sf below 1.0.0 (default behaviour) use the GEOS library for performing flat-space calculations, known as R^2 (R-squared) which produce polygons wound to the left-hand rule.
+# The sf package from version 1.0.0 onwards now uses the S2 library by default which performs S^2 (S-squared) spherical calculations and returns polygons wound according to the right-hand rule.
+# At the time of writing, if we want our geography files to work in D3 and Vega-Lite, they must use the left-hand rule and so we need sf to use the GEOS library not S2.
+# More information regarding this can be found at: https://r-spatial.github.io/sf/articles/sf7.html#switching-between-s2-and-geos
+# =========
+
+
+# Function to create the creative cluster boundaries from the wards ---------
+group_wards_into_creative_clusters <- function(ward_geometry, resolution_name) {
+    
+    sf_vers <- package_version(packageVersion('sf')) # packageVersion returns a character string, package_version converts that into numeric version numbers (major.minor.patch) e.g. 1.0.0
+    
+    # Only run the following if we are using sf version 1.0.0 or above
+    if (sf_vers$major >= 1) {
+        useS2 <- sf_use_s2()    # store boolean to indicating if sf is currently using the s2 spherical geometry package for geographical coordinate operations
+        sf_use_s2(FALSE)        # force sf to use R^2 flat space calculations using GEOS which returns polygons with left-hand windings
+    }
+    
+    ward_geometry %>%
+        # Assign the wards into their relevant creative cluster names and collapse the internal boundaries into a single area for each
+        mutate(area_name = case_when(
+            area_name %in% c("Ashton upon Mersey", "Brooklands", "Manor", "Sale Central", "Sale Moor") ~ "Ashton upon Mersey, Brooklands, Manor, Sale Central and Sale Moor cluster",
+            area_name %in% c("Longford", "Stretford & Humphrey Park") ~ "Longford and Stretford & Humphrey Park cluster",
+            area_name %in% c("Davyhulme", "Flixton", "Urmston") ~ "Davyhulme, Flixton and Urmston cluster",
+            area_name %in% c("Altrincham", "Bowdon", "Broadheath") ~ "Altrincham, Bowdon and Broadheath cluster",
+            area_name %in% c("Timperley Central", "Timperley North") ~ "Timperley Central and Timperley North cluster",
+            area_name %in% c("Hale", "Hale Barns & Timperley South") ~ "Hale, Hale Barns & Timperley South cluster",
+            TRUE ~ paste0(area_name, " cluster"))) %>%
+        group_by(area_name) %>% 
+        summarise() %>%
+        
+        # Calculate and store the coordinates of each cluster's centroid as a "lat" and "lon" property
+        mutate(lon = map_dbl(geometry, ~st_centroid(.x)[[1]]),
+               lat = map_dbl(geometry, ~st_centroid(.x)[[2]])) %>%
+        
+        # Some "artifacts" (small polygons) may be present - probably loose ends when creating the ward boundaries. This removes them    
+        st_remove_holes() %>%
+        
+        # Create the new spatial file
+        st_write(paste0("trafford_creative_clusters_", resolution_name, ".geojson"))
+    
+    # Ensure sf_use_s2() is reset to the state it was in before running the code above, i.e. whether to use the S2 library (for S^2 spherical coordinates) or GEOS (for R^2 flat space coordinates). Only if using v1 or above of the sf package
+    if (sf_vers$major >= 1) sf_use_s2(useS2)
+}
+
+
+# Create the output files ---------
+group_wards_into_creative_clusters(wards_full_res, "full_resolution")
+group_wards_into_creative_clusters(wards_generalised_res, "generalised_resolution")
+group_wards_into_creative_clusters(wards_super_generalised_res, "super_generalised_resolution")