Bottom Water Velocity Research Articles

Seiche- and storm-driven benthic oxygen uptake in a eutrophic freshwater bay determined with aquatic eddy covariance

Oxygen depletion in bottom waters of lakes and coastal regions is expanding worldwide. To examine the causes of hypoxia, we quantified the drivers of benthic oxygen uptake in Green Bay, Lake Michigan, USA, using 2 techniques, aquatic eddy covariance and sediment core incubation. We investigated benthic oxygen uptake along a gradient in C deposition, including shallow water near the riverine source of eutrophication and deeper waters of lower Green Bay where high net sediment deposition occurs. Time-averaged eddy covariance oxygen uptake was high near the source of eutrophication (11.5 mmol m−2 d−1) and at the shallower of the high deposition sites (9.8 mmol m−2 d−1). The eddy covariance technique revealed a decrease in benthic oxygen uptake with depth at the high deposition sites. These patterns were consistent with benthic uptake being driven by the deposition of autochthonous production. Additionally, eddy covariance revealed a nearly proportional relationship between benthic oxygen uptake and current velocity at all sites. Specifically, because of the lake seiche, water velocity typically varied 3× at a site and caused a 3× variation in benthic oxygen uptake. A summer storm also doubled bottom-water velocities and caused a further doubling of uptake to 28 mmol m−2 d−1. This high sensitivity of benthic oxygen uptake to seiche-driven water velocities indicates that redox conditions in surficial cohesive sediments are highly dynamic.

Flume Experiments to Assess the Effect of Bottom Roughness, Temperature, Water Velocity and Oil Condition on Critical Shear Stress Estimates of Heavy Fuel Oil

Non-floating oil is challenging to detect, track, and recover due to limited visibility inhibiting verification of the oil's location and subsurface movement. Oil that sinks to the bottom (i.e., sunken oil) can form large mats or small agglomerates on the bottom, mix into sediments, or remobilize into the water column and move with currents potentially impacting shorelines, benthic and pelagic organisms, intakes for drinking water, and power plants. Trajectory models exist that predict movement of floating and submerged oil; however, many models cannot accurately address sunken oil movement because the bed shear stress (BSS) necessary to mobilize oil (i.e., critical shear stress (CSS)), neglects the effects of bottom roughness and assumes an immobile bed. The goal of this research is to provide responders and modelers with more precise CSS estimates that include the effect of bottom roughness and incorporate results into a response tool to predict sunken oil movement. The transport of oil depends upon in-situ environmental conditions and oil properties. This research used the Coastal Response Research Center's (CRRC) 2180-liter straight flume to test the effects of water velocity, water temperature, oil mass, and bottom friction on fresh and weathered No. 6 Heavy Fuel Oil (HFO) on an immobile boundary. The flume's test section provided a uniform, one-dimensional flow field measured in 3D by an acoustic Doppler velocimeter (ADV), a Nortek AS (Norway) Vectrino II Profiling Velocimeter. The fresh or weathered (%Ev=5) HFO was mixed with kaolinite clay as a sinking agent, and 100 grams of the mixture was injected into static water via subsurface injection. The water velocity was incrementally increased in a stepwise manner by 0.07 m/s intervals and held for 15 minutes at each velocity. This occurred until: (1) oil had stopped eroding or was completely eroded from the substrate, or (2) the maximum velocity of 1.04 m/s was reached. Bottom roughness was evaluated using the velocity profile and bed shear stress (BSS) was calculated using multiple methods applicable to lab and field conditions. The oil's behavior was documented by downward- and side-facing GoPro cameras and reviewed to estimate mass loss per velocity interval, the distance the oil migrated along the bottom, and the corresponding CSS. In the case of an oil spill, responders can compare CSS estimates, determined through this research, with in-situ BSS estimates predicting under what conditions the sunken oil will become mobile.

Dispersal and transport of microplastics in river sediments.

Rivers are viewed as major pathways of microplastic transport from terrestrial areas to marine ecosystems. However, there is paucity of knowledge on the dispersal pattern and transport of microplastics in river sediments. In this study, a three dimensional hydrodynamic and particle transport modelling framework was created to investigate the dispersal and transport processes of microplastic particles commonly present in the environment, namely, polyethylene (PE), polypropylene (PP), polyamide (PA), and polyethylene terephthalate (PET) in river sediments. The study outcomes confirmed that sedimental microplastics with lower density would have higher mobility. PE and PP are likely to be transported for a relatively longer distance, while PA and PET would likely accumulate close to source points. High water flow would transport more microplastics from source points, and high flow velocity in bottom water layer are suggested to facilitate the transport of sedimental microplastics. Considering the limited dispersal and transport, the study outcomes indicated that river sediments would act as a sink for microplastic pollutants instead of being a transport pathway. The patchiness associated with the hotspots of different plastic types is expected to provide valuable information for microplastic source tracking.

NON-ERODING WATER VELOCITIES FOR INCOHERENT BOTTOM SEDIMENTS

There are the values of non-eroding water velocities for various types of bottom sediments (incoherent and cohesive) given in the normative documentation in the form of tables and graphs. Also there are a number of regulatory documents containing methods for calculation such velocities. These methods are based on empirical dependencies adapted to specific conditions. The calculated mean non-eroding water velocities are proportional to the depth of flow and bottom particle size in the case of incoherent bottom sediments erosion. The authors made an attempt to estimate non-eroding water velocity by a physical approach to the problem depending on the internal friction angle, the calculated clutch of incoherent bottom sediments and the depth of the water over the bottom. This approach should be universal. An analysis of the results indicated that the proposed formula for calculating bottom non-eroding water velocities in all considered cases gives results significantly higher than the values given in the regulatory documents for the corresponding size of incoherent bottom sediments. As a result authors obtained non-eroding water velocities, which were overestimated at times on the basis of expert evaluation. When the depth changes from 0.5 to 10 m, the spread of estimated bottom velocities varies from 14 to 22%, depending on the size of the incoherent soil. It was concluded that for smaller particles of incoherent soil, the less deviation of the calculated values of bottom non-eroding water velocities from the normative ones (for massive gravel sands at a 10 m flow depth, the deviation from the normative values reaches 375-510%). In addition, the dependence of the value bottom non-eroding water velocity on the depth is traced, which is not provided in regulatory documents. The authors offer the scientific community to join to discussion of the reasons for these discrepancies.

On the reconstruction of ocean circulation and climate based on the “Gardar Drift”

AbstractSediment‐based reconstructions of bottom water velocity at the Gardar Drift are commonly interpreted to reflect changes in the eastern Nordic Seas overflows. Here we investigate the relationship between changes in the water that overflows through the Faroe Shetland Channel and downstream bottom velocity at the location of the Gardar Drift as represented in a 500 year long simulation with the Bergen Climate Model. We identify a region in our simulation proximal to the geographical location of the northern Gardar Drift where 76% of the variance in bottom velocity can be explained by changes in the volume transport and density of the Faroe Shetland Channel overflow. By contrast, Labrador Sea Water changes do not appear to play a significant role in the bottom flow over the Gardar Drift. Our findings support the assumption in the paleo literature that reconstructions of bottom water velocity at the Gardar Drift reflect past changes in the eastern Nordic Seas overflows. However, our results suggest that velocity changes downstream are not a simple metric for the strength of the overflow, rather overflow density plays the largest role.

The impact of bedform migration on benthic oxygen fluxes

AbstractPermeable sediments are found wide spread in river beds and on continental shelves. The transport of these sediments is forced by bottom water currents and leads to the formation of bedforms such as ripples and dunes. The bottom water flow across the bedforms results in pressure gradients that drive pore water flow within the permeable sediment and enhance the supply of reactive substrates for biogeochemical processes. This transport‐reaction system has been extensively studied for the case of stationary bedforms, whereas bedform migration—the most ubiquitous form of sediment transport—has been often ignored. To study the impact of sediment transport on pore water flow, we incorporated an empirical model of bedform migration into a numerical transport‐reaction model for porous media, using oxygen as reactive solute. The modeled oxygen flux changes significantly as soon as the sediment divides into an upper mobile layer (migrating bedform) and a stationary layer underneath. The bedform is increasingly flushed with oxic bottom water, whereas pressure gradients and pore water flow reverse at increasing rate underneath the bedform. This suppresses net pore water displacement and reduces the oxygen penetration depth up to 90%. In effect, the overall oxygen uptake decreases significantly with bedform migration although bottom water velocities increase. This counterintuitive effect is systematically described for a range of different sediment types, current velocities, and respiration rates and should be considered in future studies.

From the continent to the coast: the bedload transport across the lower sector of the Guadiana River Mouth (Spain-Portugal)

The estuary of the Guadiana River is located in the southwestern Iberian Peninsula and serves as border between Spain and Portugal for more than 40 km. This estuary consists of a narrow rock bounded channel with meandering morphology imposed by the geologic materials. Its mouth is affected by a mesotidal regime, with tidal ranges comprised between 3 and 1 m (spring vs neap tides). This work analyzes the sediment transport patterns across a section located in the low marine sector of the estuary. The transversal profile is composed by a deep channel and a lateral bar. The bed of the deeper area is constituted by well sorted medium-to-coarse sand, displaying large ebb oriented sand waves (mean wavelength of 12 m), in consequence can be deduced that serves as a bypassing channel. The bed of the lateral bar is composed of cohesive bad-sorted sediment with an important population of clayey silts including a variable amount of fine-to-medium sand. The most frequent bottom feature is the plane bed, but sand patches with bedforms of a mean wavelength of 2.5 m can be observed. The fieldwork was developed in July 2010 during two opposite (neap and spring) mean tides. During the surveys, near bottom water velocity was measured using an Acoustic Doppler Current Profiler (ADCP). The speed data were used to calculate values of potential bedload transport (Qb) according with the equations suggested by Bagnold (1963). The values of bed density and rugosity were determined by using of a Laser Difractometer in sediment samples taken with a Van Veen grab sampler. The final values of Qb were compared to the real bedload transport (Sb) obtained from the amount of sand gathered in a Poliakoff sediment trap. A preliminary analysis of the results of Qb shows as the transport in the contrary senses (ebb and flood) is nearly balanced. The values of Qb oscillate between the 69.8 gr/cm during a neap tidal cycle and 329.1 gr/cm during a spring tidal cycle, whereas Sb varies between 14.6 and 72.1 gr/cm during the same tides. We can conclude that the volumes of real transport are always minor than those of potential transport, being in an approximate relation of 1/5.

Brown trout redd superimposition in relation to spawning habitat availability

Abstract –The relationship between redd superimposition and spawning habitat availability was investigated in the brown trout (Salmo truttaL.) population inhabiting the river Castril (Granada, Spain). Redd surveys were conducted in 24 river sections to estimate the rate of redd superimposition. Used and available microhabitat was evaluated to compute the suitable spawning habitat (SSH) for brown trout. After analysing the microhabitat characteristics positively selected by females, SSH was defined as an area that met all the following five requirements: water depth between 10 and 50 cm, mean water velocity between 30 and 60 cm s−1, bottom water velocity between 15 and 60 cm s−1, substrate size between 4 and 30 mm and no embeddedness. Simple regression analyses showed that redd superimposition was not correlated with redd numbers, SSH or redd density. A simulation‐based analysis was performed to estimate the superimposition rate if redds were randomly placed inside the SSH. This analysis revealed that the observed superimposition rate was higher than expected in 23 of 24 instances, this difference being significant (P < 0.05) in eight instances and right at the limit of statistical significance (P = 0.05) in another eight instances. Redd superimposition was high in sections with high redd density. High superimposition however was not exclusive to sections with high redd density and was found in moderate‐ and low‐redd‐density sections. This suggests that factors other than habitat availability are also responsible for redd superimposition. We argue that female preference for spawning over previously excavated redds may be the most likely explanation for high superimposition at lower densities.

Benthic foraminiferal distribution on the southeastern Brazilian shelf and upper slope

Foraminiferal data were obtained from 66 samples of box cores on the southeastern Brazilian upper margin (between 23.8°–25.9°S and 42.8°–46.13°W) to evaluate the benthic foraminiferal fauna distribution and its relation to some selected abiotic parameters. We focused on areas with different primary production regimes on the southern Brazilian margin, which is generally considered as an oligotrophic region. The total density (D), richness (R), mean diversity ( $$ \overline{{H^{\prime } }} $$ ), average living depth (ALD X ) and percentages of specimens of different microhabitats (epifauna, shallow infauna, intermediate infauna and deep infauna) were analyzed. The dominant species identified were Uvigerina spp., Globocassidulina subglobosa, Bulimina marginata, Adercotryma wrighti, Islandiella norcrossi, Rhizammina spp. and Brizalina sp.. We also established a set of mathematical functions for analyzing the vertical foraminiferal distribution patterns, providing a quantitative tool that allows correlating the microfaunal density distributions with abiotic factors. In general, the cores that fit with pure exponential decaying functions were related to the oligotrophic conditions prevalent on the Brazilian margin and to the flow of the Brazilian Current (BC). Different foraminiferal responses were identified in cores located in higher productivity zones, such as the northern and the southern region of the study area, where high percentages of infauna were encountered in these cores, and the functions used to fit these profiles differ appreciably from a pure exponential function, as a response of the significant living fauna in deeper layers of the sediment. One of the main factors supporting the different foraminiferal assemblage responses may be related to the differences in primary productivity of the water column and, consequently, in the estimated carbon flux to the sea floor. Nevertheless, also bottom water velocities, substrate type and water depth need to be considered.

農業用排水路と溜池における生物を媒介とした物質循環

Living things play unique and important roles in material circulations in a rural area. Their abilities seem to get weak after land improvement.This study consists of four examinations, using stable isotope ratio. The first one examined nitrogen transfer into living world by periphyton. The second one attempted to analysis the foodweb in Habahiro suiro canal. Detritus is principal producer like algae. The third one showed an area of distribution of stoney (Sterwpsyche sp.) and nitrogen restoration to terristrial area with the invertebrate. The last one made clear that nitrogen derived from farm house drainage was absorbed by Reed Canary grass (Phragmites australis).These results suggested that not only preservation of continuity in aquatic area, but also various bottom materials and water velocity were essential measures to promote material circulations by living things.

Effects of hanging ice dams on winter movements and swimming activity of fish

Formation of a hanging dam, a large, thick sub-surface accumulation of frazil ice that blocked river flow, altered physical environmental conditions dramatically and common carp Cyprinus carpio, and brown trout Salmo trutta, evacuated the pool in which the dam developed. A mean of 80·1% of the pool volume was filled with frazil ice. Mean and bottom water velocities in the pool increased from 6 to 27 cm s−1 and from 4 to 21 cm s−1, respectively, when the hanging dam formed, and water depth decreased from a mean of 2·25 to 0·45 m. Activity levels of fish leaving the pool varied. Two carp had higher swimming activities when in the over wintering pool, compared to the period when they were out of it, and a third showed opposite results.

Effects of hanging ice dams on winter movements and swimming activity of fish

Formation of a hanging dam, a large, thick sub‐surface accumulation of frazil ice that blocked river flow, altered physical environmental conditions dramatically and common carp Cyprinus carpio, and brown trout Salmo trutta, evacuated the pool in which the dam developed. A mean of 80·1% of the pool volume was filled with frazil ice. Mean and bottom water velocities in the pool increased from 6 to 27 cm s−1 and from 4 to 21 cm s−1, respectively, when the hanging dam formed, and water depth decreased from a mean of 2·25 to 0·45 m. Activity levels of fish leaving the pool varied. Two carp had higher swimming activities when in the over wintering pool, compared to the period when they were out of it, and a third showed opposite results.

Magnetic characterization of Holocene sedimentation in the South Atlantic

Surface sediment samples representative for the tropical and subtropical South Atlantic (15°N to 40°S) were investigated by isothermal magnetic methods to delineate magnetic mineral distribution patterns and to identify their predominant Holocene climatic and oceanographic controls. Individual parameters reveal distinct, yet frequently overlapping, regional sedimentation characteristics. A probabilistic (“fuzzy c‐means”) cluster analysis was applied to five concentration independent magnetic properties assessing magnetite to hematite ratios and diagnostic of bulk and fine‐particle magnetite grain size and coercivity spectra. The resultant 10 cluster structures establish an oceanwide magnetic sediment classification scheme tracing the major terrigenous eolian and fluvial fluxes, authigenic biogenic magnetite accumulation in high‐productivity areas, transport by ocean current systems, and effects of bottom water velocity on depositional regimes. Distinct dissimilarities in magnetic mineral inventories between the eastern and western basins of the South Atlantic reflect prominent contrasts of both oceanic and continental influences.

Effect of wave-induced oscillatory flow on grazing by a subtidal sea urchin Strongylocentrotus nudus (A. Agassiz)

Oscillating-flow-tank experiments were conducted to evaluate the effect of wave-induced oscillatory flow on feeding by subtidal sea urchins Strongylocentrotus nudus (A. Agassiz). Feeding rates by two size groups (mean test diameter=53 and 80 mm) were measured for rehydrated dried blades and fresh live thalli of kelp Laminaria spp. Feeding rate of the larger sea urchins was markedly reduced at the peak velocity of 0.3 m/s and virtually ceased beyond ≈0.40 m/s. The peak velocities at which feeding rate began to decrease and reached almost nil were somewhat lower for the smaller sea urchins than for the larger ones. The velocity limit for feeding does not appear to be a function of temperature. Movement by sea urchins was reduced to half the rate in still water at the peak velocity of 0.30–0.40 m/s. Sea urchins could hardly move beyond 0.70 m/s. The depth variation in the mean of the wave-induced significant peak bottom water velocity, predicted from 6-yr offshore wave data available on the northeastern Pacific coast of Honshu, Japan, was consistent with depth distributions of S. nudus and kelp. The predicted frequency of the significant peak velocity <0.40 m/s, in which sea urchin grazing might be possible, was as high as 70% at 12 m depth but was only 3% at 2 m depth. Mirroring this, sea urchins were almost absent at 2 m depth and increased in density with increasing depth while the kelp decreased in density with increasing depth and were almost absent at 12 m depth.

Diel Microhabitat Preferences of Juvenile Bull Trout in an Artificial Stream Channel

Abstract We measured day and night microhabitat preferences of age-1 juvenile bull trout Salvelinus confluentus in an artificial stream channel in early spring, and observed a significant diel shift in both microhabitat use and preference. Microhabitat variables examined were water depth, stream bottom water velocity, overhead cover, and substrate. Fish showed a stronger preference for cover during the day than at night, and generally preferred the deeper and faster areas in the channel at night compared to the daytime. The results are of importance in determining how juvenile bull trout use cover, and they suggest that maintaining or increasing rearing capacity and survival of juveniles is related to ensuring that adequate amounts of large, coarse rock substrate remain in streams. Our observations also suggest that daytime snorkeling counts of juvenile bull trout may provide inaccurate estimates of abundance because of the juvenile's extensive use of cover during the day.

Larval habitat preference of the endemic Hawaiian midge, Telmatogeton torrenticola Terry (telmatogetoninae)

Telmatogeton torrenticola Terry is a large endemic chironomid (lastinstar >20 mm) commonly found in high gradient Hawaiian streams on smoothrock surfaces with torrential, shallow flow and in the splash zones ofwaterfalls. We have quantified benthic water flow in larval habitat in a 50m segment of Kinihapai Stream, Maui using a thermistor-based microcurrentmeter. Under base flow conditions at sites suitable for larval attachment,depth was measured and bottom water velocity measurements were made ≈2 mmabove populations. Larval densities ranged from 386.9–1178m−2, habitat bottom water velocities from 13.4–64.2 cms−1, and water depths from 1.5–50 cm. Bottom velocitiesof sites with zero larvae ranged from 20.8–21.8 cm s−1with depths from ≈50 to >160 cm. Larval densities were greatest inareas with high bottom water velocities and shallow depths. Stepwisemultiple regression analyses showed that density could be confidentlypredicted best by Froude number (r=0.81; p=0.008). In the absence of Froudenumber as a regression term, the best variable to predict larval density wasbottom velocity ratio: relative depth ratio (r=0.75; p=0.019). In addition,the torrential habitat of the larvae was always characterized by aperiphyton community that appeared to be the primary food resource for thelarvae. These data suggest that torrential flows over appropriate substratesare important factors regulating habitat availability for T. torrenticolaand that reduced discharge (e.g. affected by water diversions) couldsignificantly reduce the amount of available habitat for this organism andother flow sensitive stream fauna.

Multi‐level Effects of Severe Storms on an Offshore Temperate Reef System: Benthic Sediments, Macroalgae, and Implications for Fisheries

Abstract. Episodic events have been shown to strongly affect structure and function of marine benthic ecosystems. Severe storms can have profound effects on the distribution of marine sediments which could, in turn, influence the development of benthic communities. The rich and diverse epibenthic communities on the United States mid‐Atlantic continental shelf owe their existence to the presence of a complex sequence of rocky outcrops. An unusually strong storm struck this shelf system in March 1993. Two of these carbonate platforms had been characterized by moderate sediment cover for at least the previous two years, but bottom water velocities generated by this storm removed considerable amounts of sediment from these upper flat hardbottom habitats. Macroalgal cover on these platforms dramatically increased between 1992 and 1993 with the increased exposure of hard substrate for attachment. The edges of the outcrops (scarps and rubble ramps), which are usually free of sediment, maintained their dense algal cover. Settlement blocks placed in various sub‐habitats showed little variation in algal cover among flat hardbottom and scarp areas during both years, indicating that available hard substrate habitat may be the primary limiting factor for algal growth on the North Carolina continental shelf. Since macroalgal meadows provide food and shelter for juvenile fish, the increase in critical habitat following these storms may have implications for recruitment of economically important fish species. Thus, indirect effects of episodic storms, i.e., redistribution of sand bodies leading to algal meadow development over large spatial scales, may have important consequences for benthic community development and persistence in temperate reef systems.

Spawning of the Shortnose Sturgeon in the Merrimack River, Massachusetts

We tracked 10 ultrasonically tagged shortnose sturgeons Acipenser brevirostrurn during spring in the Merrimack River to investigate spawning. Seven fish in 1989 and six fish in 1990 were tracked intensively to identify the timing and location of spawning and to characterize spawning habitat. In mid-April 1989 and 1990, fish moved upstream to just below head of tide, concentrating in a 2-km reach at river kilometers 30–32 (measured from the mouth) at Haverhill, Massachusetts. The estimated spawning time was a 5-d period (26–30 April) in 1989 and an 8-d period (22–29 April) in 1990. Spawning sites covered about 10.5 ha in 1989 and 13.5 ha in 1990. Fish spawned as river temperature increased from 9.6 to 14.0°C and river discharge decreased from 390 to 240 m3/s. Physical characteristics of spawning sites were boulder–rubble substrate, water depth of 1.8–5.5 m, and bottom water velocity of 0.3–0.7 m/s. We captured no ovulating females but verified successful spawning in 1990 by capturing two live embryos. Gill-net captures and telemetry during spring showed that some males moved to the spawning area annually. The low abundance estimates of spawning fish (9 in 1989 and 16 in 1990) indicate that the shortnose sturgeon population in the Merrimack River is the smallest yet identified and is likely vulnerable to extirpation.

Paleoflow history determined from mudwave migration: Argentine Basin

Abstract A mudwave site in the central Argentine Basin was investigated to determine its present and past mudwave activity. Site 7, on the south flank of the Ewing Drift in the northern Argentine Basin, has mudwaves that have migrated primarily throurgh preferential deposition, resulting in a relatively complete sediment on both wave flanks. Thus a history of mudwave migration can be determined for this site. The mudwave studied shows migration from before 420 ka until 32 ka, with paleocurrent speeds near 15 cm s−1 and with minor changes in current speeds around 110–120 ka. The wave migration rate became less from 32 ka to 26 ka with paleospeeds of 9 cm s−1. Wave migration ceased about 26 ka, suggesting a cessation in flow. This study indicates that bottom water velocities at this site were high during the last glacial period. but dropped abruptly near the end of the last glacial period. However, bottom current velocities appear to have been moderately high during the previous interglacial period (∼ 120 ka), suggesting that there is no simple relationship between flow speed and paleoclimate at this site.

Late quaternary changes in Antarctic bottom water velocity inferred from sediment grain size in the northern Weddell Sea

Newly formed bottom water ( θ ≤ −0.7°C) in the northern Weddell Sea flows E or NE at speeds up to 10–15 cm/s, with velocity decreasing towards the centre of the Weddell Gyre (preliminary results from long-term current meter moorings). Upper Quaternary sediments from this area contain a fine-grained terrigenous component (from the nepheloid layer) plus biogenic silica (mostly diatoms) with a small amount of ice-rafted debris. In cores from between 61° and 66°S and from 3300 m to 4700 m water-depth, the proportion of biogenic silica increases northwards (corresponding to increasing seasonal extent of open water vs sea-ice cover), and the proportion of silt and well-sorted fine sand in surface sediments increases with average current velocity. Downcore, diatom-rich and diatom-poor sediments alternate on a scale of 1–3 m, and intervals with more diatoms contain a higher proportion of silt to clay. Preliminary stratigraphy suggests the cyclicity in composition and texture is related to glacial-interglacial cyclicity. During warm periods (indirectly correlated with isotope stages 1, 3, 5 etc.) biogenic silica production takes place during several months of each year, and silt and fine sand are transported by bottom currents. During glacial periods with greater sea-ice cover than at present, biogenic productivity was suppressed and no silica was preserved in the sediments: in addition, a lower proportion of terrigenous silt implies that bottom currents were weaker. At sites with a present-day average velocity of 10 cm/s, a Last Glacial Maximum average velocity of 1 cm/s or less is inferred from grain-size measurements.