From Energy Dependence to Spatial Intelligence: A Spatial Data-Based Carbon Emission Estimation Model for Urban Built-Up Area

As urban built-up areas are the main generators of carbon emissions, scientific and accurate estimation of carbon emission levels in urban built-up areas is an important method to help implement the carbon neutrality target. Nowadays, developing a spatial data–based carbon emission estimation model that reduces dependence on energy consumption data, shortens the estimation cycle, and enhances its applicability to urban spatial development remains an urgent challenge. In this study, we developed a spatial data-based carbon emission estimation model for urban built-up areas using data from five winter cities in China over a 15-year period as an example. The estimation model not only strengthens the connection between carbon emission results and urban spatial elements, but also gets rid of the over-reliance on energy data, which in turn greatly shortens the estimation cycle of urban carbon emissions. We also used the model to investigate the distribution of carbon emissions in urban built-up areas. Compared with the traditional carbon emission estimation model based on energy consumption, the correlation coefficient between the two models is greater than 0.95, and the error between the two models is extremely small, indicating that this model has important practical value. On this basis, we propose applications for this model. We apply the model to Harbin, China, to estimate built-up area carbon emissions without using energy consumption data, thereby improving estimation efficiency. We also assess how the current urban planning strategy influences low-carbon construction. Additionally, we use the SHAP method to rank each spatial element’s contribution to carbon emissions. Based on these results, we propose low-carbon optimization strategies for winter cities in China.