Abstract/Summary

A consequence of the rapid urbanization process around the world has been the deterioration of the urban environment caused by climate change, air pollution, the increasing demand for energy, etc. Numerical models can assist sustainable urban environmental planning and strategic solutions. This thesis documents a series of developments aimed at improving the practicality and ability of climate models to assess and optimize urban planning using outdoor thermal climate simulations based on a Geographic Information System (GIS). The research focuses on efficiently representing the physical and behavioural processes of heat exchange at the neighbourhood scale. Major contributions and key findings include: 1) A new urban surface feature description scheme based on raster data was developed. By integrating the representation of urban surface properties into a multi-band raster dataset, the new scheme can more briefly and comprehensively reflect the morphological features of urban surfaces. It can also use map algebra algorithms integrated into climate simulation models to reduce computational expense. Also, the input data acquisition method to describe the urban surface features has been improved by combining it with the multi-data source fusion method. Improvements have been made to the pre-processing of input data for urban climate simulations. 2) A new scheme for simulating surface thermal radiation interactions based on a 2.5D urban structure model was developed. The scheme provides additional façade information for the 2.5D building model by the gradient analysis of raster data. This makes up for the lack of vertical surface results when analysing solar radiation distribution in traditional 2.5D models. The algorithm for calculating the solar radiation distribution within districts based on the 2.5D model has been improved. 3) A new numerical simulation framework for neighbourhood thermal climates using the GIS spatial analysis method was developed. The new framework includes important urban surface energy transfer processes, as well as new algorithms based on geographic information system spatial analysis. The framework integrates the raster data map algebra algorithm as the main body to simulate the energy exchange process at the urban neighbourhood scale and realizes a novel wind direction judgment in a short time with the help of the raster data flow direction algorithm. The new framework has been evaluated and compared with experimental data to confirm that it can accurately and quickly represent the outdoor thermal environment of different urban morphologies. These developments are compiled into a free open-source model using the Python language. The integrated model tool has the ability to efficiently quantify neighbourhood thermal climate values under different urban morphologies during the planning stage. The outputs should help improve urban environmental design solutions and promote sustainable urban development.