Artificial light at night in the Arctic

Data

The consistent and corrected nighttime light (CCNL) dataset is based on DMSP and available as GeoTIFF format at https://zenodo.org/record/6644980

Global Human Settlement Layer (GHSL) Population grid for the years 1995, 2000, 2005, and 2010 is available as Mollweide projection with 1000 m resolution at https://ghsl.jrc.ec.europa.eu/download.php?ds=pop

Database of Global Administrative Areas (GADM 4.1) is available at https://gadm.org/download_country.html

Code

Required packages: terra, raster, rgdal, tictoc, reshape, ggplot2

Data preparation

ALAN. We downloaded CCNL rasters for each year from Zenodo repository, stacked and cropped them above 45°N and filtered out the auroras via following scripts

downloading_ALAN_layers.R

stacking_and_cropping_ALAN_layers.R

filtering_of_auroras.R

Human settlement. We downladed Global Human Settlement Layer (GHSL) via the following script. We then reprojected it to the standard WGS 84 coordinate system using QGIS 3.28.0.

downloading_human_set_data.R

Stats for ALAN area and development during 1992-2013 (Figure 1, Table 1, Suppl. Table 2)

We calculated total lit area for each region and subregion for each year. We then calculated ARIMA slope and p-value and the annual growth in ALAN extent.

calculating_lit_area_per_year_and_growth.R

We plotted Figure 1 with plotting_Fig1.R

We calculated total area, newly lit area and ALAN intensity-based annual growth rate in human activity for regions and subregions. Additionally, we calculate the percent of significantly increasing/decreasing area to total area based on the ALAN intensity trend map (see the next section).

calculating_areas_stats.R

We calculated proportion of lit areas containing human settlement to the total lit area for each region and subregion.

calculating_proportion_of_inhabited_lit_areas.R

We then created Table 1 and Supplementary table 2 with creating_tables.R

Creating ALAN trend maps (Figure 2,3)

We calculated and saved ARIMA slope and p-value for each pixel of CCNL data across 1992-2013. The code is parallelized to 32 cores for computational efficiency and takes about 5h to run on 32 cores, 32GB RAM.

calculating_arima_slope_pval.R