Indirect determination of the heat budget of tidal flats

A method is presented to calculate indirectly the heat budget of a tidal flat area from downstream observations of temperature and horizontal velocity in a tidal channel. It is only necessary to establish a relationship between the velocity and the volume flux. Then the heat budget of the upstream region is determined by integrating the heat flux over one tide. The proposed method is applied to long-term measurements obtained in 2004 at two sites in a tidal channel in the Hörnum Basin, German Wadden Sea. At the site located farther downstream in the channel, the upstream catchment area is diagnosed to export heat (heat gain in the interior) from March to August, while import is diagnosed for the same period of time at the other upstream site. From September to November the situation is reversed. An analytical estimate suggests that the sign of the budget is controlled by the tidal prism and the length of the dry-falling period of the flats in the respective upstream region. In addition, a simple model is developed which can be used to determine the integral bottom heat flux of the tidal flats.     

Interannual (1999-2005) morphodynamic evolution of macro-tidal salt marshes in Mont-Saint-Michel Bay (France)

This paper provides a detailed study on the sedimentation patterns and the recent morphodynamic evolution affecting the macro-tidal salt marshes located west of the Mont-Saint-Michel (France). Twenty-two stations along three transects on the marshes were seasonally monitored for marsh surface level variations from 1999 to 2005, using a sediment erosion bar. The corresponding erosion/accretion rates were obtained together with data on topography, vegetation cover, and grain size of surface sediment. To examine the mechanisms contributing to the salt marsh sedimentation, the data and their evolution were treated with respect to tides, relative mean regional sea level, and wind speed/frequency variations.From 1999 to 2005, the marsh was globally accreting (from 3.45 to 38.11 mm yr⁻¹ in the low marsh, up to 4.91 mm yr⁻¹ in the middle marsh, and up to 1.35 mm yr⁻¹ in the high marsh), while the study was conducted during a window of decreasing trend in mean regional sea level (−2.45 mm yr⁻¹ according to regional-averaged time series). These sedimentation rates are one of the highest recorded worldwide; however, the sedimentation was not found to be continuous over the period in question. This pattern is illustrated by the strong extension of the marshes from 1999 to 2002, and the relative stability observed from 2003 to 2005. The imported and reworked sediments are trapped and fixed by the dense vegetation (Puccinellia maritima, Halimione portulacoides), inducing the general seaward extension of the marshes. The processes governing sediment budget (accretion/erosion) show annual, seasonal, and spatial variability on the marsh. Spatial variations display contrasted patterns of erosion/sedimentation between the low, middle, and high marsh, and between the different transects. These patterns are a result of distance from sediment sources, strong heterogeneity in vegetation cover (human induced or not), and contrasting topographic and micro-topographic characteristics. The higher accretion rates are observed in distal settings in the low marsh, and strongly decrease toward the middle and high marsh. This evolution results from a decrease in accommodation space/water column thickness, and frequency of inundation coupled with an increase in station elevation, but also from the cumulated effects of vegetation cover and micro-topography. The vegetation cover of the low and middle marsh enhance the settling and fixing of fine sediments imported through tides or dispersed by flood and ebb currents.The seasonal evolution of the marshes is marked by contrasting effects of water storage in the sediment. The overall seasonal sediment budget is controlled by the variation of the frequency of inundation relative to tidal range and marshes topography. Autumns are influenced by the tide (equinoxes), relative mean regional sea level, and variations in wind speed/frequency. Winter wind speed and frequency in relation with tidal variations appear to be the main parameters regulating winter marsh evolution. Summers are predominantly under the influence of local variations in water storage (desiccation) while external parameters generally display a low influence. Although it is not governed by any one parameter, springtime sediment budget seems to result from strong interaction between the above-cited parameters, despite the significant frequency of inundation (equinoxes).     

Currents in a small channel on a sandy tidal flat

Channels affect drainage and bed stresses on tidal flats. Here, near-bottom currents observed on a sandy tidal flat are compared with those observed 35 m away inside a shallow (≈0.3 m deep) channel. For water depths between 0.5 and 2.5 m (when both current meters are submerged), current speeds 0.13 m above the bed on the flat are about 30% greater than those observed 0.13 m above the bed in the channel, and are approximately equal to those observed 0.58 m above the channel bed (0.26 m above the flat elevation). Flow directions on the flat are similar to those in the channel. For flows directed across the channel axis, the ratio of speeds increases from about 1.3 to about 2.2 with increasing water depth. The corresponding ratio of the vertical velocity variances (a proxy for turbulence) decreases from about 1.5 to about 0.2, suggesting that the turbulence near the bed of the channel is greater than that near the bed of the flat for water depths greater than about 1.0 m. Drag coefficients estimated with the vertical velocity variance are approximately 70% larger in the channel than over the visually smoother flat, consistent with prior studies suggesting that channels may increase tidal-flat roughness. For flows directed along the channel axis (in the cross-flat direction), the ratio of speeds (1.2) is similar to the ratio predicted by a cross-flat momentum (along-channel) balance.     

Assessing changes in extreme sea levels: Application to the English Channel, 1900–2006

A recently extended and spatially rich English Channel sea level dataset has been used to evaluate changes in extreme still water levels throughout the 20th century. Sea level records from 18 tide gauges have been rigorously checked for errors and split into mean sea level, tidal and non-tidal components. These components and the interaction between surge and tide have been analysed separately for significant trends before determining changes in extreme sea level. Mean sea level is rising at 0.8–2.3 mm/year, depending on location. There is a small increase (0.1–0.3 mm/year) in the annual mean high water of astronomical tidal origin, relative to mean sea level, and an increase (0.2–0.6 mm/year) in annual mean tidal range. There is considerable intra- and inter-decadal variability in surge intensity with the strongest intensity in the late 1950s. Storm surges show a statistically significant weak negative correlation to the winter North Atlantic Oscillation index throughout the Channel and a stronger significant positive correlation at the boundary with the southern North Sea. Tide–surge interactions increase eastward along the English Channel, but no significant long-term changes in the distribution of tide–surge interaction are evident. In conclusion, extreme sea levels increased at all of the 18 sites, but at rates not statistically different from that observed in mean sea level.     

A synthesis of morphological evolutions and Holocene stratigraphy of a wave-dominated estuary: The Arcachon lagoon,SW France

This work presents the first synthesis of secular to millenary morphological evolutions and stratigraphy of a wave-dominated estuary, the Arcachon lagoon, from a combination of unpublished bathymetric maps (1865 and 2001), core results and high-resolution seismic profiles recorded for the first time in this lagoon. The Arcachon lagoon is located on the Atlantic coast of France, facing the wave-dominated shelf of the Bay of Biscay. It is a mesotidal semi-enclosed environment of about 160 km².The sediment budget of the Arcachon lagoon was computed by subtracting the 1865 bathymetric map from that of 2001. The computed volume difference is low (?9.9±35×10⁶ m³ in 136 yrs) and is the result of the balance between erosion and accretion that occurs within tidal channels and tidal flats, respectively. This morphological evolution pattern is explained by low sediment supply and also by the tidal distortion resulting from the morphology of the lagoon. Deep channels connected to the inlet are dominated by ebb currents inducing erosion. Tidal flats and transverse channels display weak or flood-dominated tidal currents leading to the deposition of silts. The areas of tidal flat siltation locally correlate with the presence of oyster farms, suggesting the influence of Man on the lagoon sediment-fill. Transverse channel-infill is related to weak tidal currents resulting from the hydraulically inefficient orientation of these channels which served as an ancient drainage network.Evidence for tidal channel-infill and channel abandonment are also provided by seismic profiling and cores. The upper stratigraphic succession of the lagoon (about 10 m thick) includes four main stratigraphic units dominated by channel-fills. The two lower units (around 7500–2800 yrs BP) display tabular-shape sandy channels interpreted to be records of the open estuarine phase of the Arcachon lagoon. The two upper units (around 2800 yrs BP to present-day) display U-shaped mixed sand-and-mud channel-fills interpreted to be records of the closure of the lagoon. Given that the basal estuarine units are transgressive and the upper lagoonal units are regressive, the main stratigraphic change at around 2800 yrs BP is interpreted as being the maximum flooding surface (MFS). This late MFS is explained by the low sediment supply. It is proposed that the transition from the estuarine to the lagoonal phase is related to the development of the Cap-Ferret spit in response to an increase in the ratio between wave power to tide power. This change in wave-to-tide ratio may be triggered by wave power increase following the Subboreal/Subatlantic climate instability or a decrease in tide power following a decrease in tidal prism related to the lagoon sediment-fill.Thus, the evolution of the Arcachon lagoon over the last millenaries was mainly controlled by its spit development, leading to a wave-dominated estuary in terms of its geomorphology. Once it was partially closed, extensive mud flats developed in the lagoon which became ebb-dominated.     

Effects of spilled oil on microbial communities in a tidal flat

Effects of spilled oil on microbial communities in tidal flats were examined by use of a simulator for a tidal flat ecosystem. The simulator is composed of a wave generator, a tide control device, and a tidal flat. Sediment for the tidal flat was obtained at a natural tidal flat in Hiroshima Bay, Japan. After stabilizing the benthic organisms, fuel oil C was added to the surface of the flat at 1 lm(-2). Although the total number of micro-organisms remained at 1.5-3.5 x 10(9) cells g(-1) dry sediment irrespective of the addition of oil, bacterial communities which were analyzed based on the 16S rDNA showed clear changes after the addition of fuel oil C and after a subsequent recovery period. Bacterial colonies were randomly isolated from the oil-supplemented sediment during the experiments, and the isolates were examined for susceptibility to hydrocarbons in order to screen the oil-susceptible bacteria. The proportion of oil-susceptible bacteria in the isolates decreased with the addition of the oil. Oil-susceptible bacteria showed an inability to assimilate petroleum compounds as well as an inhibition of growth. The possibility of using oil-susceptible bacteria as an indicator of bioremediation in tidal flats was discussed.     

Interactions of river discharge and tidal modulation in a tropical estuary, NE Brazil

Observations of thermohaline properties and currents were undertaken in the Curimataú River estuary (6°18′S), Rio Grande do Norte state (RN), Brazil, during consecutive neap–spring tidal cycles in the austral autumn rainy season. Highly asymmetric neap tide along channel velocities (−0.4 to 0.9 m s⁻¹) and highly stratified conditions were generated by an increase of the buoyancy energy from the freshwater input (R _iE≈5.6). During the spring-tidal cycle the river discharge decreased and the longitudinal velocity components were higher, less asymmetrical (−0.8 to 1.1 m s⁻¹) and semidiurnal, associated with moderately stratified conditions (R _iE≈0.1) due to the increase of the kinetic tidal energy forcing mechanism. The overall salinity variation from surface to bottom during two tidal cycles was from 20.5 to 36.3 and 29 to 36.7 in the neap and spring tide experiments, respectively; in the last experiment, the tropical water (TW) mass intrusion was enhanced. The net salt transport reversed from down to up estuary during the neap and spring tide experiments, respectively, varied from 6.0 to –2.0 kg m⁻¹ s⁻¹, an indication of changes in the main forcing of the estuary dynamics. Evaluation of a classical steady analytical model, in comparison with nearly steady experimental vertical profiles of velocity, shows an agreement classifiable as reasonably fair.     

Analysis of 19-year tidal data

19-year tidal data of the 3 stations, Huludao, Qinhuangdao and Kanmen, have totally been analysed, and the amplitudes and phases of 472 tidal constituents have been calculated with a resolution of Δσ⩾ 0.002 2°/h. Based on the draconitic tide, the anornalistic tide and pole tide obtained, the ultra-long-period variations of the mean sea level have been predicted. The annual tidal analysis of 19-year data at the above-mentioned stations and at Tanggu, Longkou has been carried out. The stability of the annual tidal analysis has been investigated with regard to the astronomical factors, the nonlinear effects and the variations of sea-bottom topography.     

Surface water–groundwater exchange dynamics in a tidal freshwater zone

In coastal rivers, tides can propagate for tens to hundreds of kilometres inland beyond the saltwater line. Yet the influence of tides on river–aquifer connectivity and solute transport in tidal freshwater zones (TFZs) is largely unknown. We estimate that along the TFZ of White Clay Creek (Delaware, USA), 11% of river water exchanges through tidal bank storage zones. Additional hyporheic processes such as flow through bedforms likely contribute even more exchange. The turnover length associated with tidal bank storage is 150 km, on the order of turnover lengths for all hyporheic exchange processes in non‐tidal rivers of similar size. Based on measurements at a transect of piezometers located 17 km from the coast, tides exchange 0.36 m³ of water across the banks and 0.86 m³ across the bed per unit river length. Exchange fluxes range from ?1.66 to 2.26 m day^?1 across the bank and ?0.84 to 1.88 m day^?1 across the bed. During rising tide, river water infiltrates into the riparian aquifer, and the downstream transport rate in the channel is low. During falling tide, stored groundwater is released to the river, and the downstream transport rate in the channel increases. Tidal bank storage zones may remove nutrients or other contaminants from river water and attenuate nutrient loads to coasts. Alternating expansion and contraction of aerobic zones in the riparian aquifer likely influence contaminant removal along flow paths. A clear need exists to understand contaminant removal and other ecosystem services in TFZs and adopt best management practices to promote these ecosystem services. Copyright © 2015 John Wiley & Sons, Ltd.     

Fine grain sediment transport and deposition in the Patos Lagoon–Cassino beach sedimentary system

Extensive mud deposits superimposed on the predominantly sandy inner continental shelf adjacent to the Patos Lagoon estuary, indicates that the Lagoon is a potential source of fine sediments to the coastal sedimentary system. The lagoon is large and shallow, and the water movement is mainly controlled by wind-driven set-up and set-down. The mean river inflow is around 2000 m³ s⁻¹, although peak flow rates exceeding 20,000 m³ s⁻¹ have been observed during El Niño periods. Though the tidal elevations are small, tidal velocities in the lagoon's inlet can be significant due to the large extension of the backwaters. Moreover, significant exchange flows can be generated between the estuary and coastal area due to barotropic pressure gradients established as a function of wind and freshwater discharge. The predominant net flow is seawards, but opposite near-bed flows due to pronounced vertical salinity stratification can also be observed. The coastal area is characterized by small tidal effects, large scale ocean circulation, wind-induced residual flows and wave-driven currents, where the waves originate from swell or are locally generated.     

Decadal evolution of tidal flats and channels in the Outer Weser estuary,Germany

This study focuses on the medium scale morphodynamics of the tidal flat and channel system Fedderwarder Priel, located in the Outer Weser estuary (Wadden Sea, Germany). Tidal channels and adjacent flats are highly dynamic systems whose morphologic evolution are driven by tidal, wind, and wave forcings. These coastal environments are an important ecosystem and react to changes in hydrodynamic conditions in various spatial and temporal scales. Based on annual medium-resolution digital elevation models from 1998 to 2016, we describe changes in the surface area over depth with hypsometries and use vertical dynamic trends in order to analyze and visualize the morphologic evolution of the Fedderwarder Priel and adjacent tidal channels. It is shown that several intertidal flats rise in the order of 1.3 to 5.6 cm/year. The findings indicate that the Outer Weser estuary was not in an equilibrium state for the investigated period, and tidal flats accreted with a rate exceeding mean sea level rise.     

Tides and mixing in the northwestern East China Sea Part I: Rotating and reversing tidal flows

Bottom-mounted ADV and ADCP instruments in combination with CTD profiling measurements taken along the Chinese coast of the East China Sea were used to study the vertical structure of temperature, salinity, and velocity in reversing tidal currents on a shallow inner shelf and in rotating tidal flows over a deeper sloping bottom of the outer shelf. These two regimes of barotropic tide affect small-scale dynamics in the lower part of the water column differently. The reversing flow was superimposed by seiches of ∼2.3 h period generated in semienclosed Jiaozhou Bay located nearby. As the tidal vector rotates over the sloping bottom, the height of the near-bottom logarithmic layer is subjected to tidal-induced variations. A maximum of horizontal velocity U_max appears at the upper boundary of the log layer during the first half of the current vector rotation from the minor to the major axis of tidal ellipse. In rotating tidal flow, vertical shear generated at the seafloor, propagated slowly to the water interior up to the height of U_max, with a phase speed of ∼5 m/h. The time-shifted shear inside the water column, relative to the shear at the bottom, was associated with periodically changing increases and decreases of the tidal velocity above the log layer toward the sea surface. In reversing flows, the shear generated near the bottom and the shear at the upper levels were almost in phase.     

Diel variation of dissolved organic carbon during large flow events in a lowland river

Diel variation in dissolved organic carbon (DOC) within lotic systems has been reported on numerous occasions. However, to our knowledge there has been no published work on diel DOC variation within lowland rivers during high flow events. We sampled DOC at 4 h intervals from two sites across two distinct flow regimes in the regulated lower Namoi River, Australia. This included a large flood (mean flow 224 m³ s⁻¹ and a peak flow of 376 m³ s⁻¹) sampled every 4 h for 10 consecutive days. DOC concentrations were significantly greater at night than during the day (P < 0.05) and the mean DOC concentration was 23.4 mg L⁻¹ at night compared to 18.9 mg L⁻¹ during daylight hours. The magnitude and duration of flow within this lowland river system and the mobilisation of large quantities of allochthonous carbon appeared to play a role in increasing DOC concentration and the diel difference.     

Sensitivity of tidal characteristics in double inlet systems to momentum dissipation on tidal flats: a perturbation analysis

In a tidal channel with adjacent tidal flats, along–channel momentum is dissipated on the flats during rising tides. This leads to a sink of along–channel momentum. Using a perturbative method, it is shown that the momentum sink slightly reduces the M₂ amplitude of both the sea surface elevation and current velocity and favours flood dominant tides. These changes in tidal characteristics (phase and amplitude of sea surface elevations and currents) are noticeable if widths of tidal flats are at least of the same order as the channel width, and amplitudes and gradients of along–channel velocity are large. The M₂ amplitudes are reduced because stagnant water flows from the flats into the channel, thereby slowing down the current. The M₄ amplitudes and phases change because the momentum sink acts as an advective term during the fall of the tide, such a term generates flood dominant currents. For a prototype embayment that resembles the Marsdiep–Vlie double–inlet system of the Western Wadden Sea, it is found that for both the sea surface elevation and current velocity, including the momentum sink, lead to a decrease of approximately 2% in M₂ amplitudes and an increase of approximately 25% in M₄ amplitudes. As a result, the net import of coarse sediment is increased by approximately 35%, while the transport of fine sediment is hardly influenced by the momentum sink. For the Marsdiep–Vlie system, the M₂ sea surface amplitude obtained from the idealised model is similar to that computed with a realistic three–dimensional numerical model whilst the comparison with regard to M₄ improves if momentum sink is accounted for.     

A test of the influence of tidal stream polarity on the structure of turbulent dissipation

Vertical mixing by the tides plays a key role in controlling water column structure over the seasonal cycle in shelf seas. The influence of tidal stirring is generally well represented as a competition between surface buoyancy input and the production of turbulent kinetic energy (TKE) by frictional stresses, a competition which is encapsulated in the Qh/u³ criterion. An alternative control mechanism arises from the limitation of the thickness of the bottom boundary layer due to the effects of rotation and the oscillation of the flow. Model studies indicate that, for conditions typical of the European shelf seas, the energy constraint exerts the dominant control but that for tidal streams with large positive polarisation (i.e. anti-clockwise rotation of velocity vector), some influence of rotation in limiting mixing should be detectable. We report here measurements of flow structure (with ADCPs) and turbulent dissipation (FLY Profiler) made at two similar locations in the Celtic Sea which differ principally in that the tidal currents rotate in opposite senses with approximately equal magnitude (polarity P=±0.6). A clear contrast was observed between the two sites in the vertical structure of the currents, the density profile and the rate of dissipation of TKE. At the positive polarity (PP) site (P≈+0.6), the bottom boundary layer in the tidal flow was limited to ∼20 mab (metre above the bed) and significant dissipation from bottom boundary friction was constrained within this layer. At the negative polarity (NP) site (P≈−0.6), the dominant clockwise rotary current component exhibited a velocity defect (i.e. reduction relative to the free stream) extending into the upper half of the water column while significant dissipation was observed to penetrate much further up the water column with dissipation levels ∼10^−4.5 W m⁻³ reaching to the base of the pycnocline at 70–80 mab. These contrasting features of the vertical distribution of dissipation are well reproduced by a 1-D model when run with windstress and tidal forcing and using the observed density profile. Model runs with reversed polarity at the two sites, support the conclusion that the observed contrast in the structure of tidal velocity, dissipation and stratification is due to the influence of tidal stream polarity. Increased positive polarity reduces the upward penetration of mixing which allows the development of stronger seasonal stratification, which, in turn, further inhibits vertical mixing.     

Variability of baroclinic tidal currents on the Mackenzie Shelf,the Southeastern Beaufort Sea

Semidiurnal tidal currents on the outer shelf of the Mackenzie Shelf in the Beaufort Sea were found to be strongly influenced by the locally generated baroclinic tide. Two primary factors are involved in this process: (1) the sharp shelf break along the northeastern Mackenzie Shelf, promoting the generation of vigorous internal tidal waves; and (2) the proximity to critical latitudes for M₂ and N₂ motions locking these waves and preventing them from leaving the source region. As a result, internal tides are resonantly trapped between the shelf and critical latitudes. The physical properties and temporal variations of tidal motions were examined using current meter measurements obtained from 1987–1988 at four sites (SS1, SS2, SS3, and SS4) offshore of the shelf break at depths of ∼200 m. Each mooring had Aanderaa RCM4s positioned at ∼35 m below the surface and ∼50 m above the bottom. Complex demodulation was used to compute the envelopes (amplitude modulation) of these components. A striking difference in the variability of clockwise (CW) and counterclockwise (CCW) tidal currents was found. The CW tides are highly variable, have greater amplitude, exhibit a burst-like character associated with wind events and contain about 80% of the total energy of the semidiurnal tidal currents. In contrast, the CCW components have a more regular temporal regime with distinct monthly, fortnightly and 10-day modulation at astronomical periodicities associated with frequency differences M₂–N₂ (0.03629 cpd), S₂–M₂ (0.06773 cpd), and S₂–N₂ (0.10402 cpd). Significant horizontal correlation of the CW current envelopes was found only between stations near the northeast Mackenzie Shelf, indicating this to be the main area of baroclinic internal wave generation.     

Morphological controls in sandy estuaries: the influence of tidal flats and bathymetry on sediment transport

The morphodynamics of shallow, vertically well-mixed estuaries, characterised by tidal flats and deeper channels, have been investigated. This paper examines what contributes to flood/ebb-dominant sediment transport in localised regions through a 2D model study (using the TELEMAC modelling system). The Dyfi Estuary in Wales, UK has been used as a case study and, together with idealised estuary shapes, shows that shallow water depths lead to flood dominance in the inner estuary whilst tidal flats and deep channels cause ebb dominance in the outer estuary. For medium sands and with an artificially ‘flattened’ bathymetry (i.e. no tidal flats), the net sediment transport switches from ebb-dominant to flood-dominant where the parameter a/h (local tidal amplitude ÷ local tidally averaged water depth) exceeds 1.2. Sea level rise will reduce this critical value of a/h and also reduce the ebb-directed sediment transport significantly, leading to a flood-dominated estuarine system. A similar pattern, albeit with greater transport, was simulated with tidal flats included and also with a reduced grain size. This suggests that analogous classifications for flood/ebb asymmetry of the tide in estuaries as a whole may not represent the local sediment transport in sufficient detail. Through the Dyfi simulations, the above criterion involving a/h is shown to be complicated further by augmented flow past a spit at the estuary mouth which gives rise to a self-maintaining scour hole. Simulations of one year of bed evolution in an idealised flat-bottomed estuary, including tidal flow past a spit, recreate the flood/ebb dominance on either side of the spit and the formation of a scour hole in between. The erosion rate at the centre of the hole is reduced as the hole deepens, suggesting the establishment of a self-maintaining equilibrium state.     

Characteristics of river discharge and its indirect effect on the tidal bore in the Qiantang River,China

Based on field data of river discharge, tide, tidal bore, and riverbed topography, the characteristics of river discharge, the effect of river discharge on riverbed erosion and sedimentation, and the feedback effect of riverbed erosion and sedimentation on the tide and tidal bore in the Qiantang River are analyzed. The results show that the inter-annual and seasonal variation of river discharge in the Qiantang River is noticeable, while the seasonal distribution of river discharge tends to be un...     

Modelling the tidal mixing fronts and seasonal stratification of the Northwest European Continental shelf

We investigate mixing processes under stratified conditions on the Northwest European Continental shelf using a numerical model (POLCOMS). Our results indicate that convection induced by vertical shearing of horizontal density gradients (‘shear-induced convection’) is a regularly occurring feature in the bottom and surface boundary layers in this open shelf-sea situation. Two types of turbulence models are investigated to study their capability for reproducing the observed location of tidal mixing fronts, and the physical processes occurring in seasonally stratified waters. The first model is a one-equation variant of the Mellor–Yamada model, whereas the second model combines a more recent second-momentum closure with a two-equation model. It is found that generally mean frontal positions (as estimated from ICES data) are predicted more accurately by the two-equation model. The one-equation model reproduces the mean frontal locations to 18.1 km (<3 grid spacings) and the two-equation model to 17.1 km; although in the Celtic Sea the accuracy is ∼33 and ∼12 km, respectively. Comparison with historical tide gauges, current metres, CTD stations, and thermistor chain data from the North Sea Project all show an improvement in accuracy when the two-equation model is used. This is particularly apparent in the model's ability to reproduce the spring–neap variability during stratification. We find that in the presence of shear-induced convection the routinely applied clipping of the turbulent length-scale, previously thought to be a minor ingredient in a turbulence model, has a dramatic effect on the results: if the length-scale clipping is not applied, substantial over-mixing is observed to occur. The causes and possible remedies of this effect are investigated. Overall our results demonstrate a sensitivity to the details of the turbulence model that is significantly greater than previously thought.