The rapid expansion of dockless shared bicycle systems creates persistent challenges for parking management and the efficient use of urban public space, particularly around transport hubs and major commuting corridors. Conventional hotspot-identification methods mainly indicate where parking demand is concentrated but provide limited information on the spatial form and potential public-space impact of bicycle accumulation. This study develops a morphology-aware framework using shared bicycle order and trajectory data from Guangzhou. After removing operational repositioning events, dense parking cores were extracted through grid-based connected components. Principal component analysis and robust spread measures were then used to quantify cluster orientation, length–width ratio, and bandwidth. Road proximity and orientation consistency were incorporated to distinguish roadside-band (RB), intersection-corner (IC), designated-parking-area-like (DPA), and scattered points (SP) morphologies. Short-duration demand surges during morning and evening peaks were identified separately and combined with high demand and band-shaped morphology to screen priority governance locations. Analysis of 6.902 million orders identified 27,687 dense parking cores, with the top 1% accounting for 43.7% of within-cluster orders. Band-shaped clusters represented 13.4% of all cores but accounted for 32.1% of orders. Among 26,142 strictly classified cores, RB clusters accounted for 3.23%, and only 21.8% were aligned within 30° of adjacent roads, indicating that oblique and multi-row occupation was more common than simple linear roadside parking. Among 1,842 high-demand cores, 320 showed morning-peak surges and 80 evening-peak surges. High-demand locations within 200 m of metro entrances recorded 1.7 times the mean order volume of those outside the metro buffer. The combined surge–demand–morphology criterion identified 240 priority governance points, 64.6% near arterial or sub-arterial roads and 50.0% near designated parking facilities. The findings show that parking pressure is shaped by both demand intensity and accumulation morphology, providing a practical basis for differentiated parking management, pre-peak dispatch, and improved use of urban public space.