Large disparities in spatiotemporal distributions of building carbon emissions across China

Carbon dioxide (CO2) emissions from China's building sector constitute a significant portion of the total emissions, offering considerable potential for mitigation. However, the absence of fine-scale data on building carbon emissions has limited studies on their spatiotemporal characteristics, especially in urban and rural China. To address this issue, this study proposed a downscaling methodology to estimate high-resolution CO2 emissions from China's building operation stage from 2000 to 2020 at five-year intervals. In the methodology, we first constructed provincial-level emission inventories. Then, an extended STIRPAT model and other machine learning algorithms were employed to establish relationships between provincial-level CO2 emissions and multiple variables such as building volume. Based on these relationships, provincial-level emissions were downscaled to a 1 km grid resolution. Through a multi-model comparison, the extended STIRPAT model achieved higher fitting accuracy and captured more accurate and reasonable spatial characteristics of building CO2 emissions. The downscaled results derived from this model were further substantiated by very high correlations with existing data at both municipal and grid scales, confirming their validity and reliability. The study found that high levels of building CO2 emissions are predominantly concentrated in eastern China, while core cities experience slower growth rates. Notably, there are significant and widening urban-rural differences in building CO2 emissions, especially in northern regions and certain developed cities. Spatial heterogeneity analysis indicated that urban areas have a more uniform distribution of emissions compared to rural areas. In rural regions, the spatial heterogeneity of building CO2 emissions is particularly high and increasing in the west and northeast due to unbalanced development. To mitigate emissions while maintaining economic growth, it is crucial to strategically optimize the industrial structure, alongside harnessing rural potential to invigorate local economies and promote integrated urban-rural development.