Publicação
Urban Factors' impact in CO₂ emissions: A Spatial Analysis in Oeste, Portugal
| Resumo: | Understanding the spatial determinants of CO₂ emissions is essential for effective environmental policy, especially in regions with diverse economic and social dynamics. While global models can identify broad trends, they often overlook local variations in emission drivers such as traffic, population, economic activity, and tourism. To address this gap, this study aimed to explore and understand the spatial heterogeneity of the factors impacting CO₂ emissions across the Oeste region of Portugal using spatial modelling techniques. Two methods were applied, Geographically Weighted Random Forest (GWRF) and Geographically Weighted Regression (GWR), to model emissions and examine the varying impact of predictors at different spatial scales. This study presents the spatial patterns of predictor importance and the explanatory power of both models across the region. The GWR model demonstrated substantial improvement over the global regression model, increasing the explained variance from approximately 2% 62% and significantly reducing the Residual Sum of Squares. Energy consumption and economic transactions exhibited an overall small but positive effect across most areas. The GWRF model showed moderate predictive performance (cross-validated R² = 0.39), highlighting the spatially varying contributions of economic activity, population, traffic congestion, and tourism. The spatial distribution of Local R² values from both models also reflected the differences in model performance across the region. In conclusion, spatial heterogeneity is a critical factor in understanding the drivers of CO₂ emission patterns. Geographically adaptive models offer meaningful improvements over global approaches by capturing localized dynamics. These methods can inform targeted environmental strategies and may also be extended to explore other factors or applied to different regions and sustainability challenges in spatial planning. |
|---|---|
| Autores principais: | Antunes, Mafalda de Oliveira e Sousa |
| Assunto: | Carbon Neutrality Oeste Region GWR GWRF Urban factors SDG 11 - Sustainable cities and communities SDG 13 - Climate action SDG 15 - Life on land SDG 17 - Partnerships for the goals |
| Ano: | 2025 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | Understanding the spatial determinants of CO₂ emissions is essential for effective environmental policy, especially in regions with diverse economic and social dynamics. While global models can identify broad trends, they often overlook local variations in emission drivers such as traffic, population, economic activity, and tourism. To address this gap, this study aimed to explore and understand the spatial heterogeneity of the factors impacting CO₂ emissions across the Oeste region of Portugal using spatial modelling techniques. Two methods were applied, Geographically Weighted Random Forest (GWRF) and Geographically Weighted Regression (GWR), to model emissions and examine the varying impact of predictors at different spatial scales. This study presents the spatial patterns of predictor importance and the explanatory power of both models across the region. The GWR model demonstrated substantial improvement over the global regression model, increasing the explained variance from approximately 2% 62% and significantly reducing the Residual Sum of Squares. Energy consumption and economic transactions exhibited an overall small but positive effect across most areas. The GWRF model showed moderate predictive performance (cross-validated R² = 0.39), highlighting the spatially varying contributions of economic activity, population, traffic congestion, and tourism. The spatial distribution of Local R² values from both models also reflected the differences in model performance across the region. In conclusion, spatial heterogeneity is a critical factor in understanding the drivers of CO₂ emission patterns. Geographically adaptive models offer meaningful improvements over global approaches by capturing localized dynamics. These methods can inform targeted environmental strategies and may also be extended to explore other factors or applied to different regions and sustainability challenges in spatial planning. |
|---|