Fortunately, the methodology is terribly sloppy anyway, so there's nothing serious to discuss. The researchers directly measured 30 trucks. Then they measured the total cloud of particles downwind of the traffic. There was more carbon than they'd expect given the measured value for trucks and the estimated value for cars. Therefore the cars must be emitting much more on average. Oddly, they never directly measured any cars. The idea that the additional black carbon might be due to some other source besides the cars was apparently not considered.

Knowing one of the authors (I wrote the software that runs one of the instruments), I asked him for a comment on this comment (above). Here is his response:

Actually, the black carbon was measured upwind and downwind of the highway, along with many, many other parameters (including wind speed and direction). We were able to correct the results for back ground black carbon levels (other sources). The measurements were conducted very close to the highway, so nothing else other than traffic could have generated the increase over the background. The traffic flow along the highway was monitored (with licence plate identification to determine the number of cars vs. trucks). The part about chasing 30 trucks on the highway was just a corollary study, and was not essential to the conclusions.

Other papers have addressed the use of LII (Laser Induced Incandescence) on both gasoline and diesel cars and trucks (both on-road and on test stands).