Wind tunnel measurements often under-predict the intensity of lateral dispersion,
because the large turbulent scales are difficult to reproduce in experiments. The
same problem applies to Large Eddy Simulation (LES) based dispersion models, which
use a steady uniform driving force in a periodic domain. This paper presents the Transient
Wind Forcing method (TWF), which is able to take into account the effect of turbulent
scales exceeding the domain size. The driving force components are controlled in order
to adjust the horizontal velocity to time-resolved field observations in a selected
point. Field experiments were performed in 1.2 m high street canyon series models,
simultaneously recording the three velocity components in five locations in each configuration.
The performance of the TWF model was assessed by using common validation metrics.
Furthermore, the impact of two important model parameters, the domain length and the
propulsion profile width were investigated. The dispersion of neutral particles emitted
from a point source was analyzed by Lagrangian particle tracking. The comparison of
the particle distributions obtained from the flow fields of the transient and static
wind forcing methods revealed that the TWF results exhibit a wider plume and stronger
lateral pollutant transport in the canyon containing the source.