This dissertation covers the development of two numerical models to investigate the impact of blue-green solutions on the urban microclimate; in particular, how vegetation and green infrastructure in cities influences the surface energy balance, how much water and heat is exchanged by the surface, and how the resulting sensible and latent heat fluxes feed back on the atmospheric flow. The first model is a relatively simple energy balance model and is envisioned to be of use in the master planning stage. The second model is a sophisticated three-dimensional unsteady large-eddy simulation (LES) code which provides information with spatial resolutions of one metre and sub-second time-intervals. This model will be of use primarily in the detailed design phase and for scientific investigations.
This dissertation covers the development of two numerical models to investigate the impact of blue-green solutions on the urban microclimate; in particular, how vegetation and green infrastructure in cities influences the surface energy balance, how much water and heat is exchanged by the surface, and how the resulting sensible and latent heat fluxes feed back on the atmospheric flow. The first model is a relatively simple energy balance model and is envisioned to be of use in the master planning stage. The second model is a sophisticated three-dimensional unsteady large-eddy simulation (LES) code which provides information with spatial resolutions of one metre and sub-second time-intervals. This model will be of use primarily in the detailed design phase and for scientific investigations.\\
The energy balance model MTEB is based on the Town Energy Balance \citep{Masson2000}. It was extended to incorporate the physics of green roofs; is capable of predicting the urban climate on a neighbourhood scale and to assess the scalability of climate mitigation strategies. The MTEB model solves a set of ordinary differential equations describing the evolution of heat and water on idealised urban surfaces such as roads, walls and roofs. Air movement is not modelled explicitly; instead aerodynamic resistances are used to describe the turbulent heat and moisture transport in the atmosphere. This model was used to investigate the effect of evaporative cooling from green roofs on outdoor air temperatures in an urban environment. Simulations were performed for different climates and urban geometries, with varying soil moisture content, green roof fraction and urban surface layer thickness. All simulations show a linear relationship between surface layer temperature reduction ($\Delta T_\mathrm{sl}$) and domain averaged evaporation rates from vegetation. The slope of this relationship is called the evaporative cooling potential with a value of $\approx -0.35$\UC{K}{}{day}{}{mm}{-1}. This value is independent of the method by which water is supplied. Further, a simple algebraic expression to predict the evaporative cooling potential was derived using a Taylor series expansion. For a London summer climate the results imply a maximum achievable mean temperature reduction $\le 0.4$K. Thus, without additional watering the outdoor cooling effect of green roofs on a neighbourhood scale is rather limited.\\
The detailed LES model, DALES-Urban \citep{Heus2010,Tomas2015}, was extended in order to study the complex exchanges between the urban morphology and the atmospheric air. Indeed, turbulence is very heterogeneous and depends strongly on the local morphology and temperature distribution. Turbulence in turn is a major contributor to momentum, energy and moisture fluxes. Wall functions were introduced to model scalar and momentum fluxes at horizontal and vertical structures such as roads, roofs and walls. Energy balances for all urban surfaces, including wall heat flux, heat storage and radiation were fully integrated, where the radiosity approach was being pursued to simulate short- and longwave radiative exchanges. A detailed verification was carried out to ensure that the physics were implemented correctly. These model capabilities allow to simulate urban processes in unprecedented detail allowing to gain new insight into the complex urban microclimate and help urban planning and development.\\
DALES-Urban was used to perform a study of the effect of a green roof on the ``Eastside'' building in South Kensington. The 3D building morphology was created from LIDAR data. The results indicate that the presence of a single green roof does neither affect the average wind velocities nor the average air temperature in any substantial way. The green roof surface energy balance shows a clear shift from a sensible to a latent heat flux, leading to cooler vegetated surfaces. The corresponding humidity increase in the air is found to be marginal.
The detailed LES model, DALES-Urban \citep{Heus2010,Tomas2015}, was extended in order to study the complex exchanges between the urban morphology and the atmospheric air. Indeed, turbulence is very heterogeneous and depends strongly on the local morphology and temperature distribution. Turbulence in turn is a major contributor to momentum, energy and moisture fluxes. Wall functions were introduced to model scalar and momentum fluxes at horizontal and vertical structures such as roads, roofs and walls. Energy balances for all urban surfaces, including wall heat flux, heat storage and radiation were fully integrated, where the radiosity approach was being pursued to simulate short- and longwave radiative exchanges. A detailed verification was carried out to ensure that the physics were implemented correctly. These model capabilities allow to simulate urban processes in unprecedented detail allowing to gain new insight into the complex urban microclimate and help urban planning and development.\\
DALES-Urban was used to perform a study of the effect of a green roof on the ``Eastside'' building in South Kensington. The 3D building morphology was created from LIDAR data. The results indicate that the presence of a single green roof does neither affect the average wind velocities nor the average air temperature in any substantial way. The green roof surface energy balance shows a clear shift from a sensible to a latent heat flux, leading to cooler vegetated surfaces. The corresponding humidity increase in the air is found to be marginal.