covid19_r_map_ew

R code for creating a map of R(t) for England and Wales using data from LocalCovidTracker.

Data requirements

Two datasets are required to run this script: shapefiles of the UK and estimates of R(t).

• Estimates of R(t) by Lower Tier Local Authority (LTLA) are available from LocalCovidTracker. Within the "Daily tracker" tab, adjust the date range to include the date of interest and select "Lower Tier Local Authority" from the "Area type of interest" drop down menu, and click "Download CSV" on the "Instantaneous reproduction number" panel. The downloaded filename should be of the form estimated.R.ltlas.*.csv when downloaded from the Shiny App. This file should be placed within the folder data/raw and it's full name should be updated within the Makefile (the localcovidtracker_filename variable).
• Shapefiles of the UK with 2019 Local Authority District codes are available from the ONS. Download the "Shapefiles" from this website and put the uncompressed folder within data/raw of this repository. The uncompressed folder name should be called Local_Authority_Districts__December_2019__Boundaries_UK_BUC-shp and should include five different files.

Software requirements

• The R scripts within this repository require the following packages: rgdal, broom, tidyverse, ggplot2.

Usage

• Once downloaded, the following command wil clean both the R(t) and Shapefile data and produce a map for the date specified.

make r_date="2020-04-23" map

• Simply calling make map will produce a map for for 21st July 2020 (2020-07-21) by default.
• Dates can also be specified by day after 31st Dec 2019 (such that 1 is 1st day of the year of 2020): make r_date=4 map.

Additional commands and notes

• make data_covid19 will clean the R(t) COVID19 data (automatically called by make map).
• make data_shapefile will clean the Shapefile (automatically called by make map). Within this project, this script simply moves the shapefiles to the data/processed folder but can include more elaborate scripts to simplify the shapefiles if necessary.
• Generation of these plots on Oxford's Rescomp Cluster requires adjustments to the plotting script. See the commented command at the bottom of src/visualisation/create_map_r_ew.R for plotting on Rescomp. The script create_maps_rescomp.sh will run these plotting commands in parallel (across multiple dates) on Rescomp.

Output

Example output from calling make map is the following file that is produced in output/figures and shown at the top of this README.md file.