The　multi-community　integrated　energy　system　(MCIES)　has　become　a　practical　approach　to　improve　energy　efficiency　and　reduce　carbon　emissions　with　the　promotion　of　urbanization.　However,　the　current　optimal　scheduling　of　MCIES　neither　balances　the　total　operation　cost　of　the　MCIES　and　the　individual　benefits　of　community　microgrids　effectively　nor　considers　the　carbon　emission　accounting　and　carbon　trading　of　terminal　integrated　energy　systems　adequately.　In　this　study,　a　bi-level　optimal　scheduling　framework　dominated　by　the　regional　dispatching　center　(RDC)　and　fully　participated　by　the　community　integrated　energy　systems　(CIESs)　is　developed　for　the　scheduling　of　MCIES,　which　considers　the　energy　interaction　and　carbon　trading　between　CIESs.　Firstly,　a　refined　carbon　emission　accounting　model　considering　direct　and　indirect　carbon　emissions　is　proposed,　which　can　calculate　the　carbon　emissions　of　each　CIES　in　detail　from　the　energy　flow　perspective　based　on　the　definition　of　virtual　and　actual　carbon　flow.　Then,　a　one-master-multi-slave　game　model　is　established,　where　the　RDC　acts　as　the　leader　pursuing　the　minimum　operation　cost　of　the　region　by　formulating　energy　interaction　and　carbon　trading　schemes　between　CIESs,　while　the　CIESs　are　regarded　as　the　followers　that　adjust　the　energy　consumptions　to　minimize　their　operation　costs.　Finally,　the　effectiveness　of　the　proposed　optimization　model　is　validated　through　the　case　studies,　which　are　tested　in　a　specific　urban　area.　The　simulation　results　demonstrate　that　the　proposed　optimal　scheduling　model　can　not　only　effectively　coordinate　the　benefits　between　RDC　and　CIESs,　but　also　reduce　the　overall　operational　cost　and　carbon　emission　of　the　MCIES.　©　2023　Elsevier　Ltd