Impact of roof and building geometry on urban flows: from an idealised canopy to a real case study

Nowadays, cities occupy 2% of the world soil, collecting 50% of global population. In a urban environment so complex and fragile, 75% of pollutant and 80% of CO2 are emitted. As a consequence, finding healthy way to live cities is pressing, especially in relation to the continuous urbanisation process. The ventilation inside the urban pattern plays a key role in dispersion and comfort, so that many research dealt with it by facing several specific subjects and by employing different models, scales and techniques. This thesis investigated the influence of two specific geometrical parameters: the building aspect ratio (i.e. the building width to height ratio) and the gable roof slope variability instead of the more common flat roofs. This was made at first considering some series of 2D street canyons and then by modelling a district of a real city. More specifically, numerical (LES) simulations were carried out performing street canyons with two different canyon aspect ratios (ARC = 0.5, 1.0) and varying the building aspect ratio (ARB = 0.1, 0.2, 0.3, 0.4, 0.5, 1.0, 1.5 and 2.0), in order to study wind flow and ventilation mechanism in narrow canyons, that are typical of the Mediterranean cities. Then by fixing the canyon width as 1.0, 2.0, and by changing the roof slope from 0° to 45° (0°, 10°, 20°, 30°, 45°), the variability of gable roof inclination was investigated. In addition to this, the influence of gable roofs 45° sloped was experimentally investigated (water channel), exploring the entire range of flow regimes (Oke, 1988), by means of the canyon width from 1 to 6. Finally, a quasi-real 1:200 scaled model was employed, by means of the EnFlo wind tunnel and the study site on the South Bank of the Thames River in London, as part of the MAGIC-air-UK project. Here the influence of roofs with different shapes was evaluated in terms of ventilation mechanism and pollutant dispersion. All the outcomes demonstrate the positive influence of roofs on ventilation and air exchange mechanism. Roofs increase turbulence, particularly in the upper part of the canyon, even if the behaviour at pedestrian level is not obvious. Roof employment appears to considerably affect the flow dynamics also in more complex models, suggesting that not employing roofs entails significant errors in results.