Influences of Urban Geometry on Local Climate A case study of the city of Stockholm

Abstract

Climate change of anthropogenic origin is having a pivotal impact on urban climate. Urbanisation will likely continue excelling in the future in conjunction with a growing population, making the understanding of urban climate change and the modelling of accurate urban climate conditions of interest. The aim of this study is to investigate the effect of urban geometry on the local climate, using the Weather Research and Forecasting (WRF) model with a focus on the summer of 2018 in the city of Stockholm. To well capture the impact of urban land use on the local climate, two multi-layer urban canopy schemes are used in the WRF experiments: the Building Effect Parameterisation (BEP) and the Building Energy Model (BEM). Five simulations have been conducted, i.e., the default BEM and BEP, as well as BEM and BEP with building height which has been adjusted according to Stockholm (BEMLowBH and BEPLowBH), and one with the urban areas being replaced by the nearest no urban land use type (NoUrban). The ability of the model to capture local climate conditions was evaluated by comparison of modelled and observed data. Furthermore, the urban heat island (UHI) is examined based on observations between Stockholm city stations and surrounding rural areas. Surface air temperature maps were produced, and an UHI index between an urban station (Stockholm-Observatoriekullen) and a rural station (Tullinge) was calculated. The results indicate that the WRF model simulations, except the NoUrban simulation, are uniform in their simulation of temperature, demonstrating a tendency to overestimate the temperatures during daytime. There is a high correlation between modelled and observed temperature, however, standard deviation is considerably higher for the observed data compared to the modelled. The BEM, BEP, BEMLowBH and BEPLowBH simulations all manage to recreate a daytime UHI, while they are all unable to fully capture the nocturnal UHI, and the urban geometry influences on local climate were not sensitive to the changes in building heights in the current simulations. Model improvements could consist of better simulations in heat storage in the daytime and redistribution during the night-time than the current version.