You can track trade by watching the density of radio traffic from the shipping lanes. This is in some ways a better indicator than the Baltic Dry Index.

I was trying to do what these guys did 10 years ago - but the night images data was expensive and came off a defense meteorological satellite. Found a guy at NOAA who had done the work before me and was distributing it as a lights at night study. He thought this technique of estimating GDP had issues because different regions have rules and special conditions that cause a broad difference in light levels (e.g. Denmark has particularly strict light pollutions regulations) that do not always compose at the national level.

I'll be interested in these guys results to see how they addressed the problems.

I'd assume that developing nations do not have laws dealing with light pollution or anything of that sort just yet.

Developing nations have no shortage of quirks. For example, they can impose tariffs on the import of the equipment needed for domestic lighting while still having a lot of GDP generated by oil. That does funny things to the analysis.

Then there was the problem of how you uncover the sub-national GDP to ground-truth your data showing sub-national variations in light levels.

The lights from China were not too bright or widespread as compared to those in India. IIRC, China's GDP is about thrice that of India's.

It wasn't clear (to me; correct me if I'm wrong) from the article whether this looks at the spread of lighting across the country. Seems to reflect that fairly accurately on a continent-level, though.

The classic, absolutely terrifying example from the Korean peninsula:

http://www.globalsecurity.org/military/world/dprk/dprk-dark....

I have never seen an image which is so effective at showing the disparity between countries.

Paper:

http://www.econ.brown.edu/students/adam_storeygard/HSWLights...

I was fascinated to see the bright streak of light tracking the Nile.

To paraphrase, The lights are turning on all over the world. :D