## Swiss Federal Institute of Technology, Department of Geography (ETHZ), Switzerland

### Main topics:

- Identification of length and velocity scales in urban street canyons.
- Dispersion modelling within a street canyon using a Lagrangian particle dispersion model.

The boundary layer research group is part of the Climatology Section within the Geography Department at ETHZ, Switzerland. More information concerning the activities of other groups within the department can be obtained at http://www.geo.umnw.ethz.ch.
The research topics of the group are:

#### Dispersion modelling over inhomogeneous terrain

Investigation and development of applied and research dispersion models (Lagrangian stochastic particle dispersion model, Puff-Particle approach). Special emphasis on urban air pollution problems. The research is focused on the influence of the roughness sublayer turbulence structure on near-surface pollutant characteristics. Specific projects:
- Development of the puff-particle approach (PPM) for short-range dispersion modelling. Introduction of roughness sublayer characteristics and test on tracer data sets. Hierarchy of models involved: Lagrangian stochastic particle model, PPM, Gaussian (operational) multi-source/multi-receptor model (OML).
- Development of an urban turbulence parameterization for an atmospheric mesoscale model. Collaboration with L.P.A.S., EPFL.
- TRAPOS, see below

#### Turbulent exchange over inhomogeneous surfaces

Investigation of the effect of inhomogeneity (patchiness) of a given surface upon turbulent exchange of momentum and/or sensible and latent heat. In particular, the problem of spatial averaging is considered. In the framework of BAT (Biosphere Atmosphere Transfer, with partners from the University of Bern and IUL/FAC, Liebefeld) two specific projects are running:
- Large Eddy Simulation (LES) of patchy surfaces typical for Swiss agricultural sites. Introduction of drag effects at the surface.
- Identification of source areas (area of influence) for observations at a given height over inhomogeneous patchy surfaces. Extension of existing source area/ footprint models using an appropriate Lagrangian particle dispersion model (project within EUROTRAC-2, BIATEX-2).

#### Turbulence in stable boundary layers

Analysis of turbulence and other data from the ETH Greenland expedition (1990/91). Exchange processes of momentum and sensible heat, comparison with theory and derivation of revised parameterizations. Derivation of budget estimates for, e.g. the conservation of sensible heat. Investigation and modelling of katabatic wind.
#### Scintillation

Investigation of a new measurement technique to measure areally averaged turbulence fluxes. In particular, the adequacy of the scintillometric technique for applications over non-homogeneous surfaces is examined since the approach is presently based on the intrinsic assumption of horizontal homogeneity (collaboration with the Institute of Geodesy and Photogrammetry, ETHZ, which is the grant holder for this project).
### ETHZ-Contribution to TRAPOS:

Street-level dispersion models require meteorological input such as wind speed or dispersion parameters (e.g. velocity variances) at a particular location within the street canyon to be modelled. Often, meteorological information is only available from another position (for example: roof level wind) or even from a remote street canyon or another site (e.g., airport). Simple parameterizations are therefore required to relate the information from where it is available to where it is required. This, in turn, makes it necessary to investigate the characteristic flow and turbulence structure within and above a street canyon, i.e. within the urban roughness sub-layer. With this available, the meteorological information from any location within the roughness sub-layer can be used to derive the input into dispersion model, wherever it is required. A specific objective of the ETH group is therefore the identification of length and velocity scales in urban street canyons to be used in these parameterizations. These will be derived from theoretical and heuristic considerations as well as from the exploitation of as much as available full scale and wind tunnel data.
A second area of involvement will be dispersion modelling within a street canyon using a Lagrangian particle dispersion model in combination with simulated flow and turbulence fields from the CFD approach of some other groups within TRAPOS. The two objectives are linked in that the dispersion modelling exercise can be repeated using the parameterized turbulence characteristics in order to investigate to what extent the simplified fields can reproduce the complex structure of dispersion processes within a street canyon.

Positions offered