The East China Sea Shelf has an unusually wide and low gradient shelf, supplied from sediment-charged rivers and large river delta systems, with bottom currents sweeping the sea floor and located in the path of strong typhoons. Sediment gravity flow deposits, including four hybrid event beds and a high density turbidite, are identified in a core from the mid-shelf of the East China Sea. The hybrid event beds typically comprise three or two internal divisions from the base to the top: (i) H1, H3 and H5; or (ii) H3 and H5. Radiocarbon ages of the hybrid event beds were in the range of 3821 to 8526 yr bp. Based on correlation with surrounding cores, the hybrid events may have happened at any time between 1930 yr bp and 3890 yr bp. The delta C-13 values in hybrid event beds together with bathymetry data suggest local erosion on the shelf. The average delta C-13 value for the H1 division is similar to the H3 division in the hybrid event beds, implying that the organic matter in the H1 and H3 divisions may come from the same source area. Cross-plots of upper continental crust normalized rare earth elements in the five units reveal that the sediment source of the four hybrid event beds and the turbidite was ultimately primarily from Korean rivers. Partial transformation from a moderate-strength debris flow with the additional role of erosional bulking can explain occurrences of hybrid event beds on the East China Sea Shelf. The data indicate that hybrid sediment gravity flow deposits were sourced from intra-shelf failures and subsequently transformed and deposited as hybrid event beds. The study shows that hybrid sediment gravity flows and turbidity currents may not necessarily indicate proximity to a major fluvial or deltaic system and that intra-shelf sedimentation can be a sediment source. It is unlikely that the debris flows and turbidity currents were triggered by a hyperpycnal flow or tsunami, because both can carry continental and/or coastal signals which have not been recognized in the core. Typhoons are the probable triggering mechanism.