Speaker
Description
The rapid urbanization process has led to greater thermal stress on pedestrians, and greenery arrangement can provide an effective mitigation strategy. However, the impact of vegetation on microclimate and building heat load (BHL) has primarily been explored in street canyon scenarios, with limited research at the residential scale. Furthermore, traditional methods for assessing the environmental impact of different urban forms and greenery arrangements often require extensive time for simulation.
Therefore, this study proposes an integrated cascade modelling chain with idealised model batch simulation based on ENVI-met and EnergyPlus, and analytical prediction based on XGBoost, SHapley Additive Projection (SHAP) and Partial Dependency Plot (PDP), for the comprehensive assessment of outdoor environmental indicators affected by greening and urban form during both summer and winter seasons. By creating a simulated dataset containing common thermal climate indices(UTCI), wind speed, temperature, mean radiant temperature(Tmrt) and relative humidity for seven major climate zones in China (East, Central, South, North, Northeast, Northwest and Southwest), a strong interaction between building shadow exposure (BSE) and frontal area index (FAI) as well as between BSE and greenery porosity (Po) was found.
The result shows that XGBoost is reliable and robust in the assessment of synergistic effects of built form and greening in urban settlements. Specifically, frontal area index (FAI), greenery porosity (Po), average building height (AH), building shadow exposure (BSE), and distance between tree and building (ADB) were the main influences on wind-thermal environmental comfort. The proportion of podium-type buildings significantly affects the ventilation of residential areas, while the courtyard-type buildings have a greater impact on the thermal environmental comfort of residential areas than the other building forms. The greenery porosity (Po), on the other hand, is highly efficient in improving the relative humidity and the mean radiant temperature (Tmrt). The methodology in this study provides quantitative insights into sustainable urban planning and strategies for the placement of building forms and greenery in settlements, which will result in more targeted optimal placement strategies for different climate types. It is also demonstrated that the XGboost-based machine learning model can accelerate the simulation process and enable a more comprehensive study to support sustainable urban planning in the future.
References
Ge, J. et al. (2024). Effects of urban vegetation on microclimate and building energy demand in winter: An evaluation using coupled simulations. Sustainable Cities and Society, 102, 105199.
Li, H. et al. (2021) ‘Urban morphology in China: Dataset development and spatial pattern characterization’, Sustainable Cities and Society, 71, p. 102981.
Li, J. et al. (2022). A novel geometric parameter to evaluate the effects of block form on solar radiation towards sustainable urban design. Sustainable Cities and Society, 84, 104001.
Morakinyo, T.E. et al. (2020) ‘Right tree, right place (urban canyon): Tree species selection approach for optimum urban heat mitigation - development and evaluation’, Science of The Total Environment, 719, p. 137461.
Ouyang, W. et al. (2022) ‘Evaluating the thermalradiative performance of ENVI-met model for green infrastructure typologies: Experience from a subtropical climate’, Building and Environment, 207, p. 108427.
Wong, N.H. et al. (2021) ‘Greenery as a mitigation and adaptation strategy to urban heat’, Nature Reviews Earth and Environment. Springer Nature, pp. 166–181.
Zhou, S. et al. (2024). Synergistic assessment of multi-scenario urban waterlogging through data-driven decoupling analysis in high-density urban areas: A case study in Shenzhen, China. Journal of Environmental Management, 369, 122330.
| Keywords | Sustainable Urban Planning; greenery arrangement; thermal environmental comfort; XGBoost; Microclimate |
|---|---|
| Best Congress Paper Award | Yes |
