York Road Air

Four steps, each of which you can check:

emissions  = AADT x class split x diurnal share x g/vehicle-mile
plume      = Gaussian line source, Briggs urban coefficients
dilution   = f(wind speed, stability class, distance)
breathed   = regional background + traffic increment

Traffic is every MDOT SHA segment within about 6 km of Towson — — segments carrying — vehicle-kilometres a day, with nine years of counts each. The class split — cars, light trucks, buses, single-unit and combination trucks — is published for 75 of them; for the rest it is imputed from the median split of segments that do report, grouped by road class. Interstates carry about 2.5× the heavy-vehicle share and roughly 19× the tractor-trailer share of a local road, so road class is the variable that matters.

Why heavy vehicles dominate. Diesel trucks and buses are a couple of percent of the traffic and, on a typical arterial hour, around a fifth of the nitrogen oxides. Overnight their share of the fleet roughly triples while total traffic collapses — which is why the model puts the worst air before dawn rather than at rush hour.

Dispersion discretises each road into point sources every 80 m and sums a Gaussian plume from each, using Briggs urban coefficients plus the initial mixing that traffic's own turbulence creates. Stability class comes from hour and wind speed, blended across dawn and dusk rather than switched on the hour. It is the same family of model as EPA's R-LINE, simplified: no street-canyon effects, no terrain, flat ground.

Background matters. The model predicts the increment the roads add. What you breathe is that plus the regional background the whole city sits in, so both are shown separately — the increment is the part that is about your street rather than your region.

Does it produce sane numbers? Taking the busiest York Road segment alone — 36,780 vehicles a day — with the wind blowing straight across it on a calm, stably layered morning, the model gives roughly 52 µg/m³ of nitrogen oxides at 20 m from the kerb, 18 at 50 m and 5 at 100 m. Published near-road measurements put arterial NOₓ increments in the tens of µg/m³ at that distance, decaying 60–80% within the first 100 m, so the magnitude and the shape both land where the literature says they should. That is a consistency check, not a validation — only a sensor is that.