marjoleinbruijning/covid19-burden-madagascar
Domaine:
geospatialType de record:
project
Créateur:
mar
Hôte:
Mapping the expected burden of Covid-19 in Africa.
Spatial estimates of burden by age cfrs for Africa
================
4/1/2020
---
This project is aimed at projecting and mapping the expected Covid-19 burden in Africa, based on population age distributions. For a similar project focusing on the USA, see this page . For an interactive tool of our project, see here .
---
## Table of Contents
* Analysis
* Data sources
* Calculations
* Output Files
* Code
* Preliminary results
* Next Steps
* Session Info
---
## Analysis
### Data
We use two sources of data from WorldPop:
- Africa Continental (@ 1x1 km
distribution)
- 0 - 65 in 5 year intervals, 65+ is one age group
- Country level (\~ 100m
resolution)\]
- 0 - 1, 1 - 80 in 5 year intervals, 80+ is one age group
- If you want to download data for another country, you can use
the `wp_age.sh` bash script included here. From your working
directory:
bash wp_age.sh {output directory} {isocode} {yr}
For example: `bash wp_age.sh mada_age MDG 2020` - You will need to
have GNU `parallel` and
`curl` installed, to use this script.
- Then edit the script R/mada\_1x1.R with the new
directories & iso code
- A table of country iso codes is included
here
- Shapefiles from malariaAtlas using the R
package
- I used rmapshaper to simplify polygons for easier plotting
(except for admin3 which was too big\!)
- The continental shapefiles are a bit patchy and you end up with
some invalid geoms
**Spatial files and other large files are stored on dropbox\!** -
Download
here
---
### Calculations
For both Madagascar @ \~ 1x1 km scale and AFR at \~ 10 km scale:
1. Aggregate rasters up to make them easier to work with
2. Add male and female populations in each age group
3. Match to admin codes (country iso code, admin 1 - 3) from
MalariaAtlas shapefiles
4. Apply cfrs across age groups
---
### Output files
These data (zipped csv files) with cell\_id corresponding to raster &
admin codes corresponding to shapefiles:
- Africa gridded @ 10x10 km: output/afr\_dt.gz
- Mada gridded @ 1 …