Machine Learning-Based Carbon Emission Estimation at the Urban Functional Zone Scale

Strategic Insights -

Research from Zhejiang Agriculture & Forestry University has introduced a new approach to estimating carbon emissions at the urban functional zone scale using machine learning techniques. The study aimed to improve the accuracy and spatial resolution of carbon emissions estimation by integrating multi-source data. The research found that analyzing carbon emissions based on urban functional zones is effective and supports low-carbon city construction and management.

Key Takeaways:

  • The study utilized Open Street Map (OSM) and point-of-interest (POI) data to identify urban functional zones, which were combined with multi-source remote sensing data to construct a machine learning-based carbon emission model.
  • The identified functional zones were divided into 3,861 zones with total emissions of 51,124,900 tons, where residential, commercial, and industrial zones were the main sources of emissions.
  • Multifunctional urban green space plazas also contributed significantly to emissions, a factor often overlooked in previous studies.
  • The study found that conventional methods relying on nighttime light data had limited accuracy and spatial resolution, especially in industrial and green space zones.
  • The machine learning-based model improved the accuracy and spatial resolution of carbon emissions estimation, especially in industrial and green space zones.
  • The research confirmed that analyzing carbon emissions based on urban functional zones is effective and supports low-carbon city construction and management.

Statistics:

  • Total emissions estimated: 51,124,900 tons
  • Number of identified functional zones: 3,861
  • Main sources of emissions: residential (23.5%), commercial (20.6%), and industrial (17.1%) zones
  • Contribution of multifunctional urban green space plazas to emissions: 10.2%
  • Improvement in accuracy and spatial resolution of carbon emissions estimation using machine learning-based model: 15.6% (in industrial zones) and 21.7% (in green space zones)

Sources:

  • Carbon Emission Estimation At the Urban Functional Zone Scale: Integrating Multi-source Data and Machine Learning Approach. Energy and Buildings, 2025;341.
  • Elsevier Science Sa, PO Box 564, 1001 Lausanne, Switzerland (www.elsevier.com)
  • Zhejiang Agriculture & Forestry University, College of Mathematics and Computer Science, Hangzhou 311300, Zhejiang, People's Republic of China