An overview of advanced urbanization of mesoscale models

The presence of buildings of variable height, construction materials with variable thermal and reflecting properties, and surfaces oriented in various direction make a city a very challenging area to simulate in atmospheric mesoscale models. The flow features (both turbulent and not) induced by the buildings, in fact, are significant in size, but still smaller than the grid size usually employed in mesoscale models. This means that their effect must be parameterized. However, the strong heterogeneity of the flow in the urban canopy layer makes impossible the use of the approximations adopted over homogenous surfaces. On the other hand, urban areas are extremely important for many applications, because people live in cities. A high degree of precision is expected from mesoscale models in urban areas. The history of mesoscale model urbanization is characterized by the tension between the amount of details and complexity used to represent the urban area in the models, and the computational time used. The compromise between these two (often conflicting) aspects is determined by the specific application of the model. Since the computational power available and (in a certain sense) also the applications are dynamically evolving, also the methodology used to urbanize mesoscale models is evolving. In this contribution the evolution of the techniques used to urbanize mesoscale models is briefly presented and put in relations with the most common applications, like air quality, emergency response, urban climatology, urban planning and weather forecast. Moreover, the future directions and challenges (in the view of the author) of model urbanization will be presented.