Persistent Wintertime Thermal Inversion in the City. The case of Madrid. Part II: Dispersion

A correct simulation of the dispersion during wintertime thermal inversion is important not only to improve the physical understanding of the phenomena but also to design and evaluate strategies to improve air quality. In Part II of this study the simulations performed in Part I using WRF with the multilayer urban canopy scheme BEP-BEM, are used to simulate the dispersion during one of those episodes over Madrid, Spain, using a passive tracer representing NOx. The results are compared against NOx measurements obtained at 24 ground stations, grouped in 5 zones. Coherently with Part I, the analysis shows that the best results for dispersion are obtained when BEP-BEM is used and the soil moisture is reduced to 25% of the value provided by the global model. The results also show a strong spatial variability of dispersive conditions, with a tendency of the pollutants to accumulate in the topographical depressions. Furthermore, a simple methodology is proposed to derive NO2 from NOx concentrations based on an empirical fitting of two months of observations. The statistical indicators for NO2 computed in this way are similar to those obtained with photochemical models, making this approach a valid, and computationally efficient, alternative, in particular for forecasting purposes. In the last part of this contribution, the model is used to design and evaluate simple air pollution reduction strategies. In particular, it has been found that: 1) there is no pollutant accumulation from one day to the following one, so the air quality of one day is entirely determined by the emissions and meteorological conditions of that day, 2) displacing part of the emissions of the rush hours to the central hours of the day, when the dispersive conditions are more favorable, can significantly reduce the peaks, and to a lesser extent the daily mean, without changing the total amount of pollutants emitted.