Commuting　carbon　emissions　(CCE)　constitute　a　significant　portion　of　urban　carbon　emissions　and　exert　a　notable　influence　on　urban　land　use　policies　and　spatial　planning　strategies.　Traditional　methods　for　estimating　CCEs　have　many　disadvantages　and　are　less　accurate　in　terms　of　space　and　time.　In　this　study,　a　typical　weekday　in　the　main　urban　area　of　Beijing　is　selected　as　the　research　object.　First,　taxi　CCEs　are　estimated　by　an　improved　VSP　model　based　on　GPS　trajectory　data.　Second,　taxi　CCEs　are　obtained　by　filtering　the　commuting　trajectories　based　on　POI　data　and　taxi　trajectories.　Finally,　the　spatial　distribution　characteristics　of　taxi　CCEs　are　analysed　by　spatial　autocorrelation.　The　results　show　that　(1)　the　morning　peak　commuting　carbon　emission　in　Beijing　reaches　5338.04　kg/h,　accounting　for　26.13　%　of　the　total　carbon　emission　of　cab　trips;　the　evening　peak　commuting　carbon　emission　reaches　3858.73　kg/h,　accounting　for　18.89　%　of　the　total　carbon　emission　of　cab　trips;　(2)　the　spatial　distribution　of　CCE　in　Beijing　shows　that　the　western　part　of　the　core　area　is　higher,　while　the　northeast　and　southwest　of　the　central　city　are　higher;　(3)　High-value　commuting　carbon　emission　areas　are　mainly　concentrated　in　regions　with　a　more　developed　financial　industry;　the　carbon　emissions　from　the　city　ring　road　and　airport　highway　significantly　surpass　those　of　other　areas.　This　research　introduces　an　innovative　approach　to　examine　CCE　and　measure　it　within　various　functional　zones　of　urban　areas.　The　findings　of　this　study　can　serve　as　valuable　guidance　for　informing　future　urban　spatial　configurations　and　land　use　policies,　leading　to　more　effective　and　sustainable　urban　development.