Abstract
Understanding how built environment features are associated with travel-related carbon-dioxide (CO2) emissions is essential for planners to encourage environmentally sustainable travel through transportation and land use policies. Applying gradient boosting decision trees to the data from the Minneapolis-St. Paul metropolitan area, this study addresses two gaps in the literature by identifying critical built environment determinants of CO2 emissions, and more importantly, illustrating threshold effects of built environment elements. The results show that three neighborhood-level built environment factors have the strongest influences on CO2 emissions: distance to the nearest transit stop, job density, and land use diversity. The distance to downtowns also has a substantial impact. This study further confirms that built environment variables are effective only within a certain range. These threshold effects offer valuable implications for planners to achieve desirable environmental benefits efficiently.
Original language | English (US) |
---|---|
Pages (from-to) | 1-12 |
Number of pages | 12 |
Journal | Transportation Research Part D: Transport and Environment |
Volume | 75 |
DOIs | |
State | Published - Oct 2019 |
Bibliographical note
Publisher Copyright:© 2019 Elsevier Ltd
Keywords
- Climate change
- Greenhouse gas emissions
- Machine learning
- Travel behavior
- Urban form