I remember an interview with the producthead of Google Earth (the desktop client), she said when photographing all the streets, the cars also checked for air pollution:
She mentioned a capital in Europe, where the amount of particles under certain sizes differed by 10x from one crossing to the next.
The measurement may very well be accurate but statements like that should set off massive red flags and not be taken at face value. A factor of ten difference for something that just kinda diffuses through the air doesn't "just exist". You don't get gradients like that "naturally" for the most part. It's the result of something. Maybe there's a source something is upwind of and something else is downwind of. Maybe there's conditions causing it to concentrate. Or it varies 10x day to day, but on an average basis it equals out. Etc. Etc.
Doesn't it really depend? Like I recall Oxford Street in London at one point was notoriously bad because it was a bottleneck for a lot of slow moving traffic so that one length of road was especially bad. 10x bad I don't know... But it's not hard to imagine some of the quieter roads filtering off like Berwick St or Dean St would be considerably better
Slow moving traffic at one intersection and free flowing traffic at another could easily account for a 10x ratio of particulate pollution, especially in European capitals where diesels are prevalent.
But a 10x ratio on the same road is also plausible if the Google car is following a large truck on one pass and then driving by itself on the second.
Especially the small particles in large cities are mostly caused by Biodiesel fuels - at least from my anecdotal experience.
I have a Phillips air purifier which includes a sensor for particles - whenever Diesel cars drove by the particles spiked and it went full throttle for a while if I had my window open.
