Pollutant transport between a street canyon and a 3D urban array as a function of wind direction and roof height non-uniformity

Measurement data/results

Results

Figs 9 and 11 show the mean dimensionless total pollution fluxes through the top (at z/H = 0.6) and side openings of the studied street canyons for the vertical (90°) and oblique (45°) wind directions, respectively. The mean dimensionless total vertical pollution fluxes were calculated for the top opening as

${\frac {\overline {c^{*}w}}{U_{ref}}}={\frac {\overline {\frac {cU_{ref}HLw}{Q}}}{U_{ref}}}={\frac {HL}{Q}}{\overline {cw}}$ , (1)

where c* is the instantaneous dimensionless concentration, c is the instantaneous concentration in ppm, w is the instantaneous vertical velocity component in m s^-1, U_ref is the reference velocity in m s^-1(here the freestream velocity), H is the reference height in m (here the height of the building of uniform roofs), L is the length of the line source in m, Q is the volumetric flow of ethane from the line source in ml s^-1, and the overbar denotes the time averaging. Similarly, the mean dimensionless total latera pollution fluxes were calculated for the lateral opening as

${\frac {\overline {c^{*}v}}{U_{ref}}}={\frac {\overline {\frac {cU_{ref}HLv}{Q}}}{U_{ref}}}={\frac {HL}{Q}}{\overline {cv}}$ , (2)

where v is the lateral velocity component.

Fig 9a shows that in the case of the uniform roof, the pollutant is transported up the leeward wall and down the windward wall in the middle of the street canyon due to vertical recirculation (see Fig 10). Near the lateral ends of the street canyon, the pollutant is transported by horizontal recirculation, also known as corner vortex. This type of transport can also be observed at the lateral openings of the street canyon. The pollutant is removed from the street canyon through the all openings (top and laterals) mainly by the turbulence, as shown by the relative contribution of turbulent pollutant fluxes to the total fluxes, which is about 70 %. However, in the case of the uneven roof heights, these eddies are either absent altogether, as in the case of the right uneven street canyon (Fig 9b), or they are enhanced, as in the case of the left uneven canyon (Fig 9c). In the first case, pollutant transport from the canyon through the openings is mainly driven by the advection, in the second case by the turbulence. Especially near the right lateral end (seen from downstream) of the left non-uniform canyon, there is a very strong horizontal recirculation that also retains the pollutant on the windward wall (se y = H in Fig 9c). It can also be seen from Fig 9 that both the uniform and the left non-uniform street canyon have a higher re-emission (negative total pollutant flux) of the pollutant than the right non-uniform street canyon and are therefore worse ventilated.

Figure 9: Mean dimensionless total pollution flux fields (coloured contours) through the top (c*w/U_ref in the middle) and lateral (c*v/U_ref, at the sides) of the a) uniform (A1-R) and non-uniform b) right (A2-R) and c) left (A2-L) street canyons for the perpendicular wind direction. The positive and negative values of the fluxes represent the outgoing and incoming flux from/into the canyon, respectively. The grey contour represents the dimensionless height (z/H) of the buildings, where H is the mean urban height.

Figure 10: Typical mean vortices (blue streamlines) developed in the street canyon with the uniform roof height. The streamlines were obtained from later particle image velocimetry (PIV) measurements.

When the wind blows at an angle of 45°, a spiral vortex is created in the street canyons (Fig 12), which carries the pollutant up the leeward wall and down the windward wall along the street canyon (Fig 11). But the non-uniformity of the roof height also influences the development of this pollutant transport here. The best ventilated street canyon is now the left non-uniform one, since less pollutant is transported into the canyon through the lateral opening (cf. the negative pollutant flows for the left openings of the street canyons). In general, all studied street canyons are better ventilated and the pollutant transport is mainly driven by the advection through the downstream lateral openings.

Figure 11: The same as in Fig. 9 but for the oblique wind direction.

Figure 12: Sketch of the mean spiral vortex mean vortices (blue streamline) developed in the street canyon with the uniform roof height.

Experimental data for download

The following experimental data are presented in Tecplot block format. Thus, all of the values for the first variable are given in a block, then all of the values for the second variable, then all of the values for the third, and so forth. The list of the all variables presented in the data can be found in Measurement Quantities and Techniques. The normalised time-averaged momentum fluxes ( ${\overline {u'w'}}$ or ${\overline {u'v'}}$ ) and the turbulent ( ${\overline {c'w'}}$ or ${\overline {c'v'}}$ ) and total pollution ( ${\overline {cw}}$ or ${\overline {cv}}$ ) fluxes are given in angle brackets (< >). The name of the data (zones in Tecplot) reflects the configuration studied, e.g. A2-L_45_z=0.6H stands for the city model A2, the left street canyon (L), the oblique wind direction (45°) and the height of the upper opening (0.6H). The data are packed according to the measured street canyon opening (TOP or LAT) and wind direction (90 degrees or 45 degrees).