Previously, the World Bank had used off-the-shelf luminosity data night time satellite imagery. However, this data could often be ‘noisy’ due to contaminated satellite images (as a result of cloud cover, reflections of light on water, etc.).
Nighttime satellite imagery is a useful proxy for economic activity in the absence of official statistics. Luminosity is particularly useful as it is frequent (it is available monthly), timely (it is made available within a few weeks following the end of the month), and is granular (it can be used to measure economic activity in small regions).
In order to address these sources of noise, we conducted pixel-level processing of satellite imagery to address cloud cover, pixels falling on water bodies, and ephemeral light (e.g. lightning). This was done using a high performance computing environment.
We found that our cleaned luminosity estimates were significantly less noisy and volatile than unprocessed/off-the-shelf satellite imagery: