Short description
(English)
|
Weather forecasting models have been for some time now on a steady path of improvement. Today, the ECMWF provides every 6 hours at about 50 km (0.5°) resolution wind speed and direction, temperature and water content. Based on such data, models like aLMo (Alpine Model) are able to provide the corresponding values at 7 km resolution, and a 2 km resolu-tion is looming on the horizon. The same applies to the AROME model for example, used in France.Regarding dispersion models, many types have appeared: from the oldest and simplest (gaussian, box model) to the most recent (second generation gaussian model, multibox eulerian model, lagrangian and eulerian-lagrangian models), they all share the need of high resolution, on the order of 100m., in order to deal with primary pollutants which exhibit high concen-tration levels close to the sources.SEDE SA has developed an eulerian nonstationnary multibox model of pollution dispersion, POLYTOX. Its main strengths are to include an horizontal parametrisation of the diffusion processes, an despite being mostly a diagnostic model, to include a vertical turbulence representation. It is geared towards maximizing the use of available ground measurements. This is also its weakness, because although it is not stationnary, it does not take into account the external conditions.With state of the art meteorological models providing data at a 2 km resolution, and dispersion boxmodels with 100 meters resolution, it is now possible to integrate both in a single step, which will enhance significantly the dispersion calculation outputs, especially for long time period simulations of primary pollutants concentrations. Peak concentration modelling, decontamination scenarios will also greatly benefit from this integration.This will involve wind interpolation, using the continuity equation (e.g. mass conservation), dispersion parametrisation, and in a following step, urban dispersion parametrisation.
|
Partners and International Organizations
(English)
|
AT, BE, BG, CH, CY, DE, DK, EE, ES, FI, FR, GR, HU, IT, LT, NL, NO, PL, PT, RO, SE, TR, UK
|
Abstract
(English)
|
The present work deals with air quality modelling, focusing on meteorological data assimilation. The local dispersion model Polytox is modified in order to use meteorological fields from the meso-scale meteorological model COSMO instead of local measurements. As a result, the impact of input meteorological fields on the air quality modelling can be analyzed. As a practical example, we use a real case study located in 'Entre-deux-Lacs', a region in Switzerland where it is planned to build a gas power plant. Three critical meteorological situations, with high pollutant concentrations records, have been identified. Polytox has been ran for those 3 situations with different configurations: i) Input meteorological fields are provided by measurements ii) Input meteorological fields (wind fields, mixing height, diffusion coefficients) are provided by the meso-scale model. The results show the Polytox ability to use both local meteorological measurements and meso-scale meteorological fields. The COSMO data, especially when used to determine the mixing height, improves the Polytox air quality simulation results. With the meso-scale meteorological data, Polytox is able to compute air quality anywhere in Switzerland if the emission inventory is available. Furthermore, Polytox is no more dependent on the measurements quality and quantity. However, the base Polytox configuration with meteorological fields provided by measurements gives satisfactory results and remains valid for air quality simulations when the measurements are highly representative of the meteorological situation.
|
