How much is traffic responsible for upwind dispersion of a tracer in central London?

Tracer dispersion experiments carried out in central London between 2003 and 2007 have indicated that the dimensionless dosage reduces with the inverse square of the distance from source according to D* = 12(R/H)^-2, where D* is the dimensionless dosage, R the straight-line distance away from source and H the average building height.

A detailed examination of the field data suggests that there is similar upper bound also to UPWIND dispersion, given by D* = 1.4(R/H)^-2. Although local flows around buildings can create some limited upwind spread, the distances over which substantial upwind dispersion was observed suggests an additional mechanism. This is generally thought to be due to moving traffic. We adapt concepts from building-wake dispersion to estimate such processes for a single vehicle, demonstrating that large vehicles would be the dominant factor. Case studies drawn from DAPPLE (www.dapple.org.uk; Wood et al. 2009, BAMS 90: 955--969) field experiments are used to isolate cases of upwind dispersion and then contrast the measured levels on a case-by-case basis both with a simple model (based on building wake theory) and with the upper-bound limit.

The field data show substantial upwind transport over distances well in excess of 150 m along a busy two-way, multi-lane road. The decay rate of upwind dosage with increasing distance from the source was less than the simple model predicts: suggesting that further mechanisms were involved in the dense-traffic conditions encountered during the experiments. Since the wind direction was approximately the same on two days of the campaign, the field data also provide some insight into the level of variability in short-range urban dispersion, which is seen to be substantial. The level of variability probably masks any differences that might exist between the data on the two days due to different mean wind speeds between the two days.