Keywords: atmospheric dispersion models, surface roughness, urban air pollution, urban turbulence, meteorological pre-, processors, environmental pollution, modelling
Towards a meteorological pre-processor for dispersion models in an urban environment
The main results from an extensive observational study in an urban roughness sub-layer are presented and discussed in view of possible applications in the context of meteorological pre-processing for dispersion modelling. These findings mainly concern the turbulence structure, the height variation of Reynolds stress, and the implications for scaling within the upper part of the roughness sub-layer (i.e. above roof level). In order to estimate average flow and dispersion parameters closer to the surface (i.e. within the canopy) from information above roof level, characteristic scaled profiles extending down to about a mid canyon level may be used. Using the above information, a method to estimate meteorological parameters in an urban environment crucially depends on the knowledge of a friction velocity in the inertial sub-layer, i.e. at the upper end of the roughness sublayer. This can be inferred from on-site measurements within the roughness sublayer (if available) by using the height dependence of Reynolds stress, or by relating remote (e.g. airport) measurements to the urban surface using information on the different surface roughness. Once this friction velocity is established, the local scaling approach within the upper part of the roughness sublayer allows – in conjunction with the stress profile – the determination of dispersion parameters, such as the velocity variances at any height within the roughness sublayer above roof level. Finally, the characteristic scaled profiles may be used if information on dispersion parameters is required closer to the surface.