A fast light pollution map generation software package has been developed,
written entirely in Rust, to compute and deliver up-to-date light pollution
maps as and when required. The project aims to be open source and avail-
able to the public as a fast and accurate light pollution model. Although
the project has not yet reached the intended level of accuracy at the time
of writing, it has established a solid foundation.
The project makes use of NASA’s black marble VNP46A2 BRDF Corrected
nighttime lights product [1]. To facilitate this, VNP46A2 tiles are fetched
concurrently through the LAADS DAAC API, converted to the GeoTIFF
format, and merged to form a more extensive GeoTIFF that spans the
entire globe. While the project currently operates with tiles, transitioning
to use the larger GeoTIFF isn’t anticipated to pose a significant challenge.
After the download and conversion to GeoTIFF, R.H Garstang’s light pol-
lution model is applied on each radiant pixel in the tile. Upon completion,
a gradient is applied, resulting in the creation of a PNG image. This image
effectively demonstrates the application of the light pollution model.