Speaker
Description
Paper Title: Investigating the impact of Urban Form Elements on carbon emissions at Different Development Stages: Based on GWR and Random Forest Models
*Presenter:* Yihuan Wang (Southeast University)
Author(s): Yihuan Wang (Southeast University)
Topic Category: Carbon emissions, Urban development Levels, Urban forms, Geographically Weighted Regression Model, Random forest
Abstract Text:
With rapid urban expansion, cities, the main vehicles for human activity, consume 78 percent of the world's energy. Massive carbon emissions have led to a rise in global average temperatures and frequently triggered extreme weather events. Due to the uneven urbanization process and socioeconomic differences, Chinese cities are at various stages of development, Consequently, the relationship between urban form and carbon emissions varies significantly among cities. Therefore, understanding the characteristics of carbon emissions and their influencing factors at different stages of development is crucial for formulating differentiated carbon emission reduction policies, promoting the development of low-carbon cities, and achieving the goal of carbon neutrality.
To more accurately reveal the impacts of urban form elements on urban carbon emissions at different development stages and to explore targeted sustainable development models, this study aims to use Geographically Weighted Regression (GWR) and Random Forest Model to analyze the impact of urban form elements on carbon emissions in Chinese cities at different development stages for 10 consecutive years, from 2012 to 2022, such as the total area of the city, road network density, average building height, etc. (Wang et al., 2019)
The main steps of this study include the following five aspects:
1. Classification of Development Stages: Relevant literature was consulted to classify Chinese cities into distinct development stages, enabling a comparative analysis of urban carbon emissions across different stages.
2. Carbon Emission Estimation: National emission inventories and Night Time Light (NTL) data were used to estimate city-level carbon emissions and validate the accuracy of the estimates with city-level carbon emission statistics.
3. Spatial Autocorrelation Analysis: The spatial autocorrelation of carbon emissions was analyzed, and regression analysis and the GWR model were used to examine the spatial heterogeneity in the relationship between urban form elements and carbon emissions.
4. Random Forest Model Construction: The random forest was constructed to learn the nonlinear relationship between urban form elements and urban carbon emissions and evaluate the impact of various urban form elements on carbon emissions.
5. Results and Recommendations: The results from the GWR and Random Forest models were analyzed to provide evidence-based recommendations for carbon reduction strategies and sustainable urban development policies.
The findings reveal that the factors influencing urban carbon emissions vary significantly across different stages of urban development. Notably, the average building height and the number of buildings emerge as key contributors to carbon emissions. These results underscore the importance of tailored approaches to urban planning and policy-making that consider the unique characteristics of cities at different developmental stages.
By combining the GWR and Random Forest Models, this study provides a more nuanced and accurate assessment of the relationship between urban form and carbon emissions. The insights gained from the analyses provide practical recommendations for reducing carbon emissions and promoting sustainable urban development. Ultimately, this study aims to support the broader goal of achieving carbon neutrality by identifying the most effective urban planning strategies for different development contexts.
References
Wang, S. et al. (2019) Examining the spatial variations of determinants of energy-related CO2 emissions in China at the city level using Geographically Weighted Regression Model, Applied Energy, 235, pp. 95–105.
References
Wang, S. et al. (2019) Examining the spatial variations of determinants of energy-related CO2 emissions in China at the city level using Geographically Weighted Regression Model, Applied Energy, 235, pp. 95–105.
| Keywords | Carbon emissions; Urban development Levels; Urban forms |
|---|---|
| Best Congress Paper Award | Yes |
