Abstract In this paper, we explore the secondary flows in the Danish Straits using observations and numerical simulations performed with the unstructured-grid hydrodynamic model SCHISM covering the North Sea and Baltic Sea. The straits are resolved on scales of up to ∼ 100 m. Given that large-scale atmospheric variability dominates the transport in these straits, we focus on the processes with subtidal time scales. Similarities and differences between the in- and outflows in the straits and flood and ebb flows in estuaries are analyzed. Contrary to the tidal straining in estuaries, the Danish Straits feature substantial differences in the stratification stability during the outflow and inflow phases. With a resolution of ∼ 100 m, new transport and mixing pathways that were previously unresolved appear fundamental to the strait dynamics. The variety of the strait morphology leads to high variability in the appearance of secondary circulation. Helical cells, often with a horizontal extension of ∼ 1 km, develop in the deep parts of the channels. A comparison between the high-resolution simulation and a simulation with a coarse grid of ∼ 500 m in the straits suggests that the coarser resolution overestimates the stratification and misrepresents the transport balance; the axial velocities and transport through the Sound are underestimated by ∼ 12%. These differences are explained by the missing secondary circulation when the coarse resolution is used (approximately two grid-points per cell instead of ten grid-points per cell in the fine resolution model), along with the resulting changes in mixing along the straits. In conclusion, the use of ultrafine resolution grids is essential to adequately resolve secondary flow patterns and two-layer exchange. Thus, the problems caused by the failure to resolve the secondary circulation in straits appear similar to the problems caused by the failure to resolve mesoscale eddies in ocean models.

中文翻译：

---

浅海海峡的二次环流：丹麦海峡的观测和数值模拟

摘要 在本文中，我们使用覆盖北海和波罗的海的非结构化网格水动力模型 SCHISM 进行的观测和数值模拟，探索了丹麦海峡的二次流。海峡的分辨率可达 ∼ 100 m。鉴于大尺度大气变率主导了这些海峡的运输，我们关注潮下时间尺度的过程。分析了海峡入流和出流与河口涨落流的异同。与河口的潮汐紧张相反，丹麦海峡在流出和流入阶段的分层稳定性存在显着差异。分辨率约为 100 m，以前未解决的新传输和混合路径似乎是海峡动力学的基础。海峡形态的多样性导致二次环流出现的高度可变性。螺旋细胞通常水平延伸约 1 公里，在通道的深部发育。高分辨率模拟与海峡中约 500 m 粗网格的模拟之间的比较表明，较粗的分辨率高估了分层并歪曲了运输平衡；通过声音的轴向速度和传输被低估了约 12%。这些差异是由于使用粗分辨率时缺少二次环流（每个像元大约两个网格点，而不是精细分辨率模型中的每个像元十个网格点）以及由此产生的沿海峡混合的变化来解释的。综上所述，使用超细分辨率网格对于充分解析二次流动模式和两层交换至关重要。因此，海峡二次环流无法解析导致的问题，与海洋模式中尺度涡旋解析失败所导致的问题类似。