Urban Climate at Street Scale : Analysis and Adaptation

Abstract

It is well known that the urban environment changes local climate inside the city. This change of the local climate manifests itself mainly through di_erences in air temperature, where cities remain warmer than the rural environment during the night. This phenomenon is called the Urban Heat Island (UHI) e_ect, and is de_ned as di_erence in air temperature between the urban and rural environment. The UHI e_ect is found in many cities of di_erent sizes around the world, and ranges between 1 and 10oC during the night. The combination of the increasing urbanisation, global warming and the impact of increasing temperature on human health makes the urban heat island a topic that is gaining more and more attention. This thesis focusses on the urban micro-climate, which treats indivicual buildings and their direct surroundings. A numerical modelling approach is used in this thesis, such that the local urban climate can be investigated and perturbed in a systematic way. The developed 2D model, called URBSIM, combines computation of radiative transfer by a Monte-Carlo model, conduction of energy into the urban material and a Computational Fluid Dynamics (CFD) model to compute air ow and air temperature. With this model, it is shown that the main source of energy to the urban heat budget is due to radiative transfer. During the night, the long wave trapping e_ect (de_ned in this theses as radiation emitted by one surface and absorbed by an other) and absorbed long wave radiation emitted from the sky are of the same order of magnitude for a building height (H) over street width (W) ratio of H=W=0.5. With increasing building height, longwave trapping becomes the main source of energy to the urban energy budget. During the day time, absorbed shortwave radiation is the main source of energy, followed by the long wave trapping e_ect. The relative contribution of these radiative components is decreasing with increasing building height, vi Summary and the conductive heat ux becomes more important. The large impact of radiation sparked the question which high albedo adaptation measure (white surfaces) is best suited to reduce the Urban Heat Island e_ect. This thesis shows that there is a clear distinction between the atmospheric UHI (air temperature) and pedestrian heat stress. Lower air temperatures can be achieved by using high albedo materials, whereas thermal comfort at street level can be improved by using low albedo materials. By using a low albedo material, less radiation is reected back inside the canyon, thereby reducing the mean radiant temperature. The lowest pedestrian heat stress is found by using a vertical albedo gradient from high albedo at the bottom part to a low albedo at the top part of the wall for H=W=1.0. This study indicated that using a high albedo material can decrease the UHI e_ect, but increases pedestrian heat stress, which might not be the desired e_ect. The developed micro-scale model is also compared to a large-scale urban parametrisation scheme that is used in meso-scale models. In this parametrisation, a 2D geometry is used to compute the uxes of the 3D environment. Results indicate that radiative transfer is well captured in the parametrisation. Canyon wind speeds and the sensible heat ux showed much larger di_erences between the two models, which is most likely due to the 2D geometry that is used as a basis for the parametrisation. It is very likely that these parametrisations are adapted to better represent the 3D urban environment. The result of this thesis is an advanced numerical model that includes most processes relevant to the urban environment. Despite the fact that the model is limited to 2D cases, the studies presented in this thesis have aided the understanding of the elementary processes that control urban air temperature, the feedback processes and interactions between the di_erent mechanisms in the urban surface energy