Skip to contents

Process Year-Indexed Rasters Against Time-Varying Admin Boundaries

Source: R/process_rasters.R
process_rasters_by_year.Rd

Wrapper around process_raster_with_boundaries() for the case where the admin boundaries change over time (one row per admin x year in the shapefile, identified by a year column). For each raster, detects the year from the filename, filters the shapefile to just that year's boundaries, extracts zonal statistics against those polygons, and row-binds the results.

Usage

process_rasters_by_year(
  raster_files,
  shapefile,
  id_cols = c("adm0", "adm1", "adm2"),
  year_col = "year",
  aggregations = c("mean"),
  layer_to_process = 1,
  raster_is_density = FALSE,
  year_extractor = NULL
)

Arguments

raster_files

Character vector of raster file paths.

shapefile

An sf object whose rows are admin units valid for a specific year. Each (admin x year) combination is a row; the year is stored in the column named by year_col.

id_cols

Character vector of shapefile columns to carry into the output (e.g., admin names or codes). Default: c("adm0", "adm1", "adm2").

year_col

Character. Name of the year column in shapefile. Default: "year".

aggregations

Vector of aggregation methods. Same options as for process_raster_with_boundaries(). Default: c("mean"). For population count rasters, use "sum".

layer_to_process

Integer or character. Layer index for multi-layer rasters. Default: 1.

raster_is_density

Logical. Convert density to counts using cell area before extraction. Default: FALSE.

year_extractor

Optional function (filename) -> year, used when the default "last plausible 4-digit year in filename" heuristic picks the wrong year. See extract_time_components().

Value

A tibble with one row per (admin, year) combination, holding id_cols, year, and the requested aggregation columns. Rasters whose year is not present in the shapefile are skipped with a warning.

Examples

if (FALSE) { # \dontrun{
# population rasters with year-varying districts
shp_by_year <- sf::read_sf("districts_by_year.gpkg")
raster_paths <- list.files(
  "data/worldpop/raw", pattern = "_pop_.*\\.tif$", full.names = TRUE
)
pop_by_admin <- process_rasters_by_year(
  raster_files = raster_paths,
  shapefile = shp_by_year,
  id_cols = c("adm0", "adm1", "adm2"),
  aggregations = "sum"
)
} # }