Towards typology-based management of urban commuting carbon emission characteristics: Identification of commuting behavior and carbon emission accounting based on individual spatiotemporal big data

• Identification method based on time matching, frequency ranking, and speed thresholds. • Revealing the most consistent commuting behavior patterns of individual residents. • Achieving large-scale, multi-category, and fine-grained commuting CO 2 accounting. • Multi-indicator joint description method for commuting carbon emission characteristics. • Typology-based precise carbon emission reduction management framework. Quantifying commuting carbon dioxide (CCO 2 ) emissions is a crucial method for evaluating the rationality of urban spatial layouts and the operational efficiency of transportation systems, becoming a key focus for precise carbon reduction in urban low-carbon development. Current research predominantly relies on limited-sample questionnaire surveys, lacking new methods for exploring large-sample, multi-category, and fine-grained CCO 2 emission accounting under spatiotemporal big data sources. Addressing this, this study focuses on Qiaoxi District, Shijiazhuang City, China, where 5 G communication infrastructure is well-developed. Utilizing one month's high-precision mobile phone signaling data (HMPSD) to provide residents' high-precision home and workplace locations and actual commuting durations for each trip, combined with estimated commuting times for different transportation modes and routes between home and workplace from navigation data, we constructed a method for identifying commuting behavior patterns using time matching, frequency ranking, and speed threshold assessment. This method identified the most consistent commuting modes (i.e., driving, public transportation, walking, bicycling, and electric biking) and corresponding commuting routes for residents' daily commutes, facilitating the accounting of CCO 2 emissions. The spatial distribution of residents' CCO 2 emission characteristics (i.e., proportion of residents commuting by car, proportion of residents commuting by public transportation, per resident commuting distance, per resident commuting duration, and per resident CCO 2 emission) was revealed at the land parcel scale. Finally, through constructing a multi-indicator joint description method and hierarchical cluster analysis, we categorized the CCO 2 emission characteristics of 115 parcels into five types and formulated low-carbon planning strategies matching these types. This study promotes the realization of supply-side planning responses from the demand-side perspective, benefiting the comprehensive consideration and precise implementation of planning layout adjustments and facility configuration optimization, thereby achieving efficient resource allocation and optimal utilization in the process of low-carbon urban construction.