Starting on 2021-08-30, the driving speed limit within the Paris municipality was reduced from 50 km/h to 30 km/h. There is a debate on the impact that this will have on air pollution.
The Paris municipality justifies lists the benefits as better safety, noise reduction, and a fairer sharing of public space, all of which are out of scope of this question.
Le Point published an inflammatory article titled “30 km/h in town, a promise of worsened pollution”. Le Figaro has a more factual title “Cars pollute more at an average speed of 30 km/h than 50 km/h, according to a study”. Le Journal du Dimanche, on the other hand, quotes an expert stating that “from one study to the next, and within the same study, the results are highly variable”. A Libération article merely cites a 2014 study with “‘contrasted’ short term results”.
The claims that the speed reduction will increase air pollution are mostly based on 2021 report by Cerema (a French public agency) (in French). This report compared fuel consumption and NOx, PM and CO2 emissions depending on the average speed, ranging from 10 km/h to 130 km/h. This report shows that for cars and light utility vehicles, pollution per km traveled has a U shape: consumption is higher at low speeds and at high speeds than at medium speeds, and the optimum speed is over 50 km/h. (Other studies show somewhat different effects; see Does lowering speed limits by 10 km/h have any impact on air pollution? for a discussion of pollution at higher speeds.)
However, it is not clear to me that this report is applicable to urban driving, in a city that wasn't designed for cars and doesn't have long stretches where a car can drive at a sustained speed. Urban traffic is stop-and-go, with frequent intersections and pedestrian crossings. Streets don't follow much of a grid shape and synchronized traffic lights only concern a small number of streets (“The longest ‘green wave’ is (…) 3.2 km long. (…) In practice, traffic has to be fluid to drive the whole 3.2 km without stopping, which is rare (…) except at night or on Sunday morning.” — Le Parisien, 2018). The average vehicle speed in Paris during daytime was 11,6 km/h in 2019Q4, down from 13,6 km/h two years earlier (source). The relevant comparison is not between a smooth 30 km/h and a smooth 50 km/h, but between frequently speeding up from 0 to the limit (or less) and braking back to 0. A lower speed limit reduces the amount of energy spent accelerating and of braking (which emits particles).
The speeds considered in the report are average speeds based on COPERT version 5 methodology. It is not clear to me what this means: COPERT includes some speed variation, but does it correspond to typical urban traffic or to typical road traffic? And if it does model urban traffic, how does the average speed figure in the Cerema report relate to the speed limit?
Reducing the speed limit is likely to slightly reduce traffic as some drivers decide to reroute or use a different means of transportation. My primary question assumes the same amount of traffic, but if there's a relevant empirical data in places where the speed limit was reduced in a similar way and there was a reduction in traffic, I'm interested as well.
TL,DR: in dense urban driving conditions, for cars, does reducing the speed limit from 50 km/h to 30 km/h increase air pollution?