This project estimates the global rural population by combining data from the World Bank and WorldPop. The methodology involves data acquisition, preparation, and processing to estimate rural populations for each country.
-
World Bank Data:
- Source: World Bank API
- Content: Rural population as a percentage of the total population.
-
WorldPop Data:
- Source: WorldPop
- Content: High-resolution population density data.
- Download and unzip World Bank rural population data.
- The World Bank data is unzipped and read into a DataFrame.
- A mean of the last 5 years (2015-2019) of rural population percentage data is calculated. If data is missing, a mean of the last 10 years (2010-2019) is used instead.
- Clean and filter DataFrame to include
Country.Name,ISO_A3,Indicator.Name,rural_pop_pct_mean. - Ensure dataset contains complete cases with positive rural population percentage.
- Load WorldPop population density raster data.
- For each country (by ISO code):
- Crop and mask WorldPop raster to country boundaries.
- Crop and masks the WorldPop raster data to the country's boundaries.
- Calculate the total population estimate by summing up the values in the raster.
- A frequency table for each population density class is created to determine the threshold population density that defines rural areas.
- The rural population is estimated by masking out areas with population density higher than the threshold.
- Raster files for estimated rural population at 1km resolution for each country.
- Files saved with LZW compression in specified directory.
- R packages for spatial data (
raster,sf,sp). - R packages for data downloading and processing (
httr,dplyr).
- Rural areas defined based on population density thresholds.
- Estimates rely on WorldPop and World Bank data accuracy and resolution.
- Assumes spatial uniformity within raster cells.
- Integrate additional or more recent data sources.
- Refine rural area definition using country-specific criteria.