Search

The problem of flow separation around islands is investigated using a dynamically adaptive finite element model to allow for resolution of the shear layers that form in the advent of separation. The changes in secondary circulation and vertical motion that occur in both attached and separated flows are documented, as is the degree of closure of the wake eddies. In the numerical experiments presented, the strongest motion always takes place at the sides of the idealised island, where flow curvature and shear act together to induce ascent. In contrast, it is the slower motion within the wake eddies that allow streamlines to extend from the bottom to the surface. We find no evidence for closure of the wake eddies. Rather, all of our separated experiments show that streamlines that pass through the eddies originate outside of the shear layers and frictional boundary layers on the upstream side of the idealised island. The numerical experiments demonstrate the potential for dynamically adaptive, unstructured meshes to resolve the separated shear layers that occur downstream of the idealised island, as well as the narrow boundary layers that form on the island itself. [ABSTRACT FROM AUTHOR]

Unstructured grid ocean models are advantageous for simulating the coastal ocean and river–estuary–plume systems. However, unstructured grid models tend to be diffusive and/or computationally expensive, which limits their applicability to real-life problems. In this paper, we describe a novel discontinuous Galerkin (DG) finite element discretization for the hydrostatic equations. The formulation is fully conservative and second-order accurate in space and time. Monotonicity of the advection scheme is ensured by using a strong stability-preserving time integration method and slope limiters. Compared to previous DG models, advantages include a more accurate mode splitting method, revised viscosity formulation, and new second-order time integration scheme. We demonstrate that the model is capable of simulating baroclinic flows in the eddying regime with a suite of test cases. Numerical dissipation is well-controlled, being comparable or lower than in existing state-of-the-art structured grid models. [ABSTRACT FROM AUTHOR]

Highlights: [•] Ocean circulation beneath ice shelves is simulated using an unstructured-mesh. [•] We represent a smoothly varying ice base in the presence of a vertical ice front. [•] We implemented ice shelf/ocean interaction in the context of finite-element method. [Copyright &y& Elsevier]

Lower to Middle Turonian deposits within the Bohemian Cretaceous Basin (Central Europe) consist of coarse-grained deltaic sandstones passing distally into fine-grained offshore sediments. Dune-scale cross-beds superimposed on delta-front clinoforms indicate a vigorous basinal palaeocirculation capable of transporting coarse-grained sand across the entire depth range of the clinoforms ( ca 35 m). Bi-directional, alongshore-oriented, trough cross-set axes, silt drapes and reactivation surfaces indicate tidal activity. However, the Bohemian Cretaceous Basin at this time was over a thousand kilometres from the shelf break and separated from the open ocean by a series of small islands. The presence of tidally-influenced deposits in a setting where co-oscillating tides are likely to have been damped down by seabed friction and blocked by emergent land masses is problematic. The Imperial College Ocean Model, a fully hydrodynamic, unstructured mesh finite element model, is used to test the hypothesis that tidal circulation in this isolated region was capable of generating the observed grain-size distributions, bedform types and palaeocurrent orientations. The model is first validated for the prediction of bed shear stress magnitudes and sediment transport pathways against the present-day North European shelf seas that surround the British Isles. The model predicts a microtidal to mesotidal regime for the Bohemian Cretaceous Basin across a range of sensitivity tests with elevated tidal ranges in local embayments. Funnelling associated with straits increases tidal current velocities, generating bed shear stresses that were capable of forming the sedimentary structures observed in the field. The model also predicts instantaneous bi-directional currents with orientations comparable with those measured in the field. Overall, the Imperial College Ocean Model predicts a vigorous tide-driven palaeocirculation within the Bohemian Cretaceous Basin that would indisputably have influenced sediment dispersal and facies distributions. Palaeocurrent vectors and sediment transport pathways however vary markedly in the different sensitivity tests. Accurate modelling of these parameters, in this instance, requires greater palaeogeographic certainty than can be extracted from the available rock record. [ABSTRACT FROM AUTHOR]

Due to a lack of modern analogues, debate surrounds the importance of tides in ancient epi-continental seas. However, numerical modelling can provide a quantitative means of investigating palaeo-tidality without recourse to analogues. Finite element modelling of the European Upper Carboniferous epi-continental seaway predicts an exceedingly low Lunar tidal range ( ca 5 cm in the open water regions of the UK and Southern North Sea). The Imperial College Ocean Model (ICOM) uses finite element methods and an unstructured tetrahedral mesh that is computationally very efficient. The accuracy and sensitivity of ICOM tidal range predictions were tested using bathymetric data from the present-day Mediterranean Sea. The Mediterranean Sea is micro-tidal and varies in depth up to 5·4 km with an average depth of 1–2 km. ICOM accurately predicts the tidal range given both a realistic, but smoothed, bathymetry and a straight sided basin with a uniform depth of 1 km. Variation in uniform depth from 100 to 3000 m with and without islands consistently predicts micro-tidality, demonstrating that the model is robust and the effect of bathymetric uncertainty on model output is relatively small. The extremely low tidal range predicted for the European Upper Carboniferous is thus deemed robust. Putative Upper Carboniferous tidal deposits have been described in the UK and southern North Sea, but are represented by cyclic rhythmites and are limited to palaeo-estuaries. Calculations based on an embayed coast model show that the tidal range could have been amplified to ca 1 m in estuaries and that this is sufficient to form cyclic rhythmites. Without tidal mixing, the tropical equatorial heat and salinity enhancement would promote stratification in the open water body. The introduction of organic matter probably caused anoxia, biotic mortality and carbon accumulation, as evidenced by numerous black ‘marine-band’ shales. [ABSTRACT FROM AUTHOR]