Developing　a　high-resolution　CO2　emissions　inventory　for　China　is　challenging　because　of　limited　detailed　parameter　information　in　bottom-up　approaches.　This　study　integrated　socioeconomic　attributes,　point　emission　data,　industrial　heat　sources,　and　improved　night-time　light　data　to　develop　an　advanced　top-down　dynamic　CO2　emissions　estimation　model.　Using　this　model,　a　0.01　degrees　resolution　CO2　emissions　inventory　for　China　from　2012　to　2022　was　created.　The　results　demonstrated　that　the　model　enhances　spatial　precision,　distribution　accuracy,　and　timeliness.　Spatiotemporal　dynamics　help　identify　high　emission　periods　and　regions,　and　reflect　the　impact　of　geographical　and　social　activities.　The　driver　factor　analysis　indicated　that　GDP　per　capita,　energy　intensity,　and　carbon　emissions　intensity　were　the　main　drivers　of　changes　in　emissions.　Each　region　should　develop　emissionreduction　strategies　based　on　the　dynamic　variations　of　these　drivers.　This　study　offers　a　reliable　tool　for　carbon　emissions　inventory　research,　supporting　accurate　carbon　emissions　estimation　and　policy　formulation.