Measurement-Dependent Asymmetry in Chinese Urban Carbon-Emission Inventories: Reference States, Scale Sensitivity, and Cross-Inventory Concordance

Urban carbon-emission inventories can yield different conclusions depending on accounting construction and analytical scale, yet these two sources of variation are rarely evaluated within a common framework. We define asymmetry as a measurement-dependent departure from an explicit reference state, rather than as a synonym for inequality, temporal variability, or spatial clustering. Using CEADs annual city inventories for 2001–2019, Carbon Monitor Cities daily estimates for 2019–2021, and harmonized city geometries, we examine distributional, temporal, and local spatial dimensions and assess cross-inventory concordance. The results show that long-term inequality trends are sensitive to sample composition; seasonal contrasts are substantially larger than weekday–weekend differences; and weak global spatial dependence can coexist with persistent local tail organization. Cross-inventory city rankings are comparatively robust despite material differences in absolute emission magnitudes, whereas fine-grained local spatial classifications are more sensitive to inventory construction. These findings indicate that the apparent urban carbon-emission asymmetry is a measurement property jointly conditioned by the selected reference state, sample, temporal resolution, spatial definition, and accounting boundary. For inventory compilers and comparative researchers, multiscale reporting and the explicit disclosure of these choices are therefore necessary for credible city-level comparison.