Remember when I said that the satellite pixels cover half a square kilometer? That really limits our ability to understand exactly what it is that is changing.

I'm part of a group that will propose a nighttime light satellite with unprecedented sensitivity to #ESA for consideration as their #EarthExplorer 13 mission. It would allow us to understand the nature of these changes far better (e.g. who exactly is responsible for the changes? Are whole areas changing gradually, or are we seeing the impact of specific buildings or parking lots?)

So, please cross your fingers for us, and if you would be interested in using such data, please reach out

I'm very grateful to Tian Li and Zhe Zhu from #UConn and Zhuosen Wang from NASA for involving me and my (former @GFZ) colleague Theres Kuester in the work. And also very grateful to @GFZ, #ESA, and especially @ruhr-uni-bochum.de for making it possible for me to work in this field!

That's what I made the images in the thread with. #RemoteSensing #LightPollution #GIS #GEE

Almost 30% of the increase in total light emissions for the entire world took place in China.

Out of wealthy countries that aren't in crisis, France stands out for really dramatic reductions in total light emissions. Partly this is due to their light pollution law, and partly it's because so many communities in France now turn off their streetlights late at night when there's no one on the street to see them.

So here's the main finding: yes, Earth is getting brighter on average. But it's certainly not getting brighter everywhere - there are lots of places where light emissions are decreasing!

Here's the "gradually changing" areas in Berlin, for example. Within the city, there are places and neighborhoods that are brightening, and others that are darkening!

Nighttime lights data are really noisy. Part of the noise comes about because the satellite can view a scene from different directions. When that happens, it's viewing LITERALLY DIFFERENT LIGHTS, as in the aerial photos here.

When something changes, the observations stop agreeing with the prediction (middle). If the change persists for at least 14 consecutive observations, then a break point in the trend is assigned. From that point on, a new model predicts future changes. And so on and so forth (right) for the following years.

There were 1.16 million daily (NASA Black Marble) *images*. Now think about how many pixels make up each image... 🤯

Those studies looked at (very) large areas, and used monthly or annual composite images. The new study analyzes change at the spatial scale of individual pixels (~0.5 square km) and at the temporal scale of individual nights.