Accurate　traffic　prediction　plays　a　pivotal　role　in　the　development　of　intelligent　transportation　systems　(ITSs).　However,　traffic　data　often　exhibits　multiscale　temporal　dependencies,　spatiotemporal　heterogeneity,　and　highly　nonlinear　characteristics,　which　pose　significant　challenges　to　existing　prediction　models.　To　address　these　challenges,　we　propose　a　general　decoupled　graph　convolutional　recurrent　network　(GDGCRN).　At　the　core　of　the　GDGCRN　is　its　ability　to　handle　multiscale　temporal　dynamics.　By　learning　temporal　embeddings　for　daily　and　weekly　cycles,　the　model　adaptively　captures　short-term　fluctuations　as　well　as　long-term　patterns.　On　the　spatial　side,　the　GDGCRN　incorporates　a　sensor-specific　mechanism　that　adapts　to　the　unique　behavior　of　individual　traffic　sensors,　allowing　the　model　to　capture　complex　spatial　dependencies　across　diverse　and　heterogeneous　traffic　environments.　Moreover,　the　GDGCRN　introduces　a　signal　decoupling　mechanism　that　separates　steady-state　and　nonsteady-state　signals,　enabling　the　model　to　robustly　handle　both　predictable　traffic　trends　and　sudden　disruptions.　Through　comprehensive　experiments　on　multiple　benchmark　datasets,　including　traffic　flow　prediction　on　PeMS04　and　PeMS08,　demand　prediction　on　the　NYCBike　dataset,　and　speed　prediction　on　PeMSD7(M),　we　demonstrate　that　the　GDGCRN　achieves　highly　competitive　performance　across　these　datasets.　Notably,　on　the　PeMS08　dataset,　the　GDGCRN　reduces　the　mean　absolute　error　(MAE)　metric　by　9.92%　compared　to　the　MVSTT　model.　©　2001-2012　IEEE.