Abstract: With the advancement of dual-carbon goals, the low-carbon operation of power systems has become a research focus. However, the coordinated dispatch between virtual power plants (VPPs) and distribution networks (DNs) faces challenges due to insufficient multi-time scale carbon emission quantification. To address this issue, this paper proposes a coordinated dispatch model for VPPs and DNs that incorporates multi-time scale carbon emission factors. First, the phenomenon of renewable energy curtailment is integrated into the carbon emission factor calculation, and a multi-time scale correction mechanism is introduced to enhance the spatiotemporal accuracy of the carbon emission factors. Second, a coarse-fine tuning multi-time scale coordinated dispatch framework is established: the day-ahead scheduling prioritizes economic efficiency and security, while the intraday scheduling adjusts carbon emission factors and optimizes operational strategies based on real-time data. Finally, the model is solved using the analytical target cascading method. Case study shows that the improved carbon emission factor can effectively differentiate the accommodation differences of wind and solar power during zero-carbon periods. Compared with traditional carbon emission factor calculation methods, it reduces the DNs' carbon emissions by 4.7 tons and cuts carbon emission costs by 17.5%. The multi-time scale coordination mechanism significantly enhances the renewable energy accommodation capacity and economic efficiency, providing strong support for low-carbon dispatch of the power system.