Skip to contents

Reads the Li Nighttime data

Source: R/data_linight.R
read_linight.Rd

Downloads, preprocesses, and harmonizes the Li et al. global nighttime light dataset (v8). This dataset provides global, annual composites of nighttime light intensity, harmonized across multiple satellite sensors to produce a consistent multi-decadal time series.

Usage

read_linight(overwrite_files = FALSE)

Arguments

overwrite_files

Logical. If TRUE, previously fixed rasters are recalculated and overwritten. Defaults to FALSE.

Value

A SpatRaster object

Details

The function:

  • Downloads the zipped Li Nighttime Lights raster files from the PRIO-GRID data repository

  • Extracts TIF files, caching results to avoid repeated unzipping

  • Identifies rasters with extent mismatches (common in the dataset)

  • Resamples problematic rasters to a standardized global template (EPSG:4326, extent -180/180, -90/90) using nearest neighbor resampling

  • Stores corrected rasters with a "fixed_" prefix for reuse

  • Combines corrected rasters into a multi-layer SpatRaster

  • Assigns layer names as dates, aligned to PRIO-GRID temporal conventions (January 1 of each year by default)

Note

  • Initial preprocessing (extent harmonization) may take time, but is cached for faster subsequent runs

  • Large raster files may require substantial disk space and memory

  • Nighttime lights are influenced by sensor calibration, atmospheric conditions, and moonlight; Li et al. provide harmonization but residual inconsistencies may remain

References

Li X, Zhou Y, Zhao M, Zhao X (2020). “A Harmonized Global Nighttime Light Dataset 1992–2018.” Scientific Data, 7(1), 168. ISSN 2052-4463, doi:10.1038/s41597-020-0510-y , https://www.nature.com/articles/s41597-020-0510-y.

Examples

if (FALSE) { # \dontrun{
# Read harmonized Li Nighttime Lights data
linight <- read_linight()

# Inspect structure
print(linight)

# Plot nighttime lights for year 2000
terra::plot(linight[["2000-01-01"]],
            main = "Global Nighttime Lights 2000")

# Compare change between 2000 and 2020
lights_2000 <- linight[["2000-01-01"]]
lights_2020 <- linight[["2020-01-01"]]
change <- lights_2020 - lights_2000
terra::plot(change, main = "Nighttime Lights Change 2000–2020")

# Extract regional time series
# example_extent <- terra::ext(100, 120, 20, 40) # East Asia
# region_lights <- terra::crop(linight, example_extent)
# terra::plot(region_lights[[1]], main = "Regional Nighttime Lights")
} # }