Hsu, Fucent Hsuan WeiFucent Hsuan WeiHsuChen, Shih-NanShih-NanChen2025-12-312025-12-312025-08https://www.scopus.com/pages/publications/105013847568https://scholars.lib.ntu.edu.tw/handle/123456789/734880This study revisits the front-trapping problem previously studied by Chapman and Lentz (hereafter CL94), which describes a coastal bottom-attached front being arrested at a certain depth due to convergence of bottom Ekman transport. The focus is on the robustness of the trapping mechanism when the fronts supported by a buoyancy discharge are unstable. Sensitivity experiments suggest that CL94’s solution, which was based on reduced physics and coarse resolution, is not physically realistic. Repeating CL94’s experiment with a more complete model physics and higher resolution leads to an outflow and coastal front that are inherently unstable to baroclinic instability. Yet, the instability does not negate the trapping process. In the time-mean flow, key trapping signatures characterized by the collocation of near-bottom reversed alongshore flow, onshore-directed Ekman transport, and elevated density gradients can be identified in all unstable outflows considered. Comparisons of the evolution of a density front in an alongshore periodic domain reveal further that front trapping is a 3D process: Energy budget analyses suggest that the along-shore periodic setup has a bottom-attached front and instabilities, but it lacks an upstream energy supply to maintain the time-mean front to allow trapping to persist. Additionally, in all unstable outflows, a secondary bottom convergence zone seaward of the trapped front is found. It is shown that this secondary convergence is associated with periodic generation and subsequent propagation of coherent vortices. Potential relevance of this secondary feature to field conditions is discussed.Coastal flowsEkman pumping/transportInstabilityOceanjournal article10.1175/jpo-d-24-0231.1