Search

Close-up stereo photographs of the seafloor were taken at five stations on the Nova Scotian rise. Stereo viewing reveals abundant relief with heights in the range 1–50 mm which are not apparent in oblique monoscopic pictures. Objects with dimensions of 0.1 mm can be resolved, and areas up to 1.0 cm 2 can be viewed. Photogrammetric contour mapping at one millimeter intervals of longitudinal ripples gives in-situ dimensions of 12 cm height, 90 cm full width, and 410 cm 2 cross-sectional area. A field of ripples spaced at 5 m with a sediment density of 0.7 g cm −3 contains 5.7 kg dry sediment per m 2 . Seafloor roughness on the scale of centimeters in abyssal, high-energy boundary layers reflects both recent sedimentary and biological processes. The spatial density of eight classes of bed relief varies on scales of tens of kilometers downslope and is consistent with downslope trends in the occurrence of larger bedforms, texture of surface sediments, and bottom boundary layer conditions. Most relief is biogenic, directly or indirectly, and most is streamlined.

A longitudinal ripple was obtained in a box core taken from the Nova Scotian continental rise in July 1982. Soft brown mud comprising 1.5—10% sand, ∼60% silt and ∼35% of μ m clay forms the 5 cm high and ∼40 cm wide ripple. A maximum thickness of 8 cm of this mud under the crest, thinning to 2 cm on the ripple's flank, overlies stiff muddy foram ooze. The vane shear strength of the brown mud is ∼0.4 kPa whereas that of the ooze is ∼4 kPa. X-radiographs show the mud to contain many fine burrows and a few larger ones as well as winnowed horizons rich in foraminifera of both benthonic and planktonic origin. Gentle wet sieving of the sand fraction on board ship showed the sand fraction to contain very few faecal pellets. Thus it appears unlikely that bedload movement played a large part in formation of the ripple. Rapid initial deposition from a concentrated suspension is suggested to have formed the structure, but this was followed by periods of erosion to yield winnowed horizons and further rebuilding with material deposited from suspension. Radiochemical data ( 234 Th and 210 Pb) and X-radiographs suggest intense particle mixing ( D B ∼90 cm 2 yr −1 ) and rapid sediment deposition (∼1.5 cm month −1 ). Thus it is possible that the bedform (and associated structures) are destroyed and recreated on a time-scale of a few months.

Bottom photographs across the Nova Scotian continental rise show effects of an intense bottom current at 4800–5000 m that decreases in strength upslope to tranquil near-bottom conditions at 3200–4000 m. Apparently weak currents are present above 3200 m. Long-term, geologically significant flow parallels the bathymetric contours toward the southwest, but significant short-term variability in direction is indicated by orientations of small-scale bedforms. The Western Boundary Undercurrent, documented elsewhere along the continental margin at depths less than 4000 m, has relatively little effect in modifying the sea floor on the Nova Scotian rise, but the deep currents (> 4000 m) have a significant effect. Repeated photography of small seafloor areas suggests that many bedforms, especially those of smaller scale, are constructed and destroyed by strong, variably directed, current events (“benthic storms”) on time scales of 3–6 months or less. Thus, a significant part of the geologic response of the seabed to currents may be “event-controlled” on time scales not very different from those in coastal environments.

New multi-beam bathymetric data from the Philippine Sea and northwest Pacific Basin reveal linear chains of small (less than 40 km 3 ) domed-shaped volcanoes (Philippine) and coned-shaped volcanoes (Pacific) rising 100–1000 m above the 6 km deep ocean floor. Some appear to have well-developed collapsed calderas and spines. Their morphology suggest recent formation in supposedly stable mid-plate regions and their occurrence in linear chains approximately parallel to plate motion may suggest an origin by extrusion from “mini-hot spot” plumes.

Fast currents, high concentrations of suspended sediment and grooved mud beds are associated with erosion in frequent abyssal storms where a fast deep mean flow is augmented (or reversed) by intense intermittent currents. This occurs about 5 km below the Gulf Stream or its rings. The waning phase of a storm results in development of bedforms and rapid deposition of a mud blanket. Several other regions of the world ocean display evidence of abyssal storm activity.

Abyssal furrows and ripples, formed in cohesive sediments, have been observed and sampled by submersible on the Bahama Outer Ridge. These features have developed as bed forms of steady bottom currents presently flowing at 5–20 cm/sec. Small furrows developed high on the secondary outer ridge are asymmetrical, and have existid for at least the last 11,000 years. Cross-furrow variations in microtopography, sediment type, and accumulation rate suggest that the furrows are maintained by present-day currents. Transverse ripples on furrow wails are developed in cohesive sediments and have small corniches similar to those observed in snow downwind from mountain ridges. Regional changes in furrow type and morphology may reflect changes in sedimentary conditions and furrow evolution (current velocity, sediment type, sedimentation rate), or they may reflect different furrow origins.

Consolidation tests were conducted on sediment samples from four cores (17–31 m lengths) obtained with the Giant Piston Corer on the Blake—Bahama Outer Ridge (water depths 4758–4962 m). Three of the cores (GPC-7, 8, 9) had similar composition (calcareous lutites) and texture (60% clay) and similar vertical variability in water content, shear strength and carbonate profiles with very large decreases in water content (average decrease of 52%) occurring at 20 m depth. The fourth core (GPC-11) has considerable sand- and silt-sized material (calcareous tests) and much higher shear strength. The consolidation results of core GPC-11 showed a significant amount of overconsolidation suggesting that more than 10 m of sediment has been eroded and that the core was taken in the bottom of a large erosional furrow. Correlations between consolidation parameters and composition are proposed through use of a mineralogy factor ( E ) such that the mineralogy of the clay fraction is combined with effects of stress history and clay fraction influence as follows: E=[(M ∗ + I) − (%CaCo 3 )] × [(w 0 − w p ) % where: • M∗ = percentage of montmorillonite plus mixed-layer montmorillonite—illite, • I = percentage of illite, • w 0 = in-situ water content, and • w P = plastic limit. The following regression-line equations were obtained based on consolidation test results from Blake—Bahama Outer Ridge: Compression Index = C c = 0.0162 E − 0.0333 Coef. of Secondary Compression = C ∝ = (0.461 E − 0.126) · 10 −3 Although clay mineralogy does have an effect on consolidation behavior, the effects of consistency coupled with percentage of calcium carbonate exert the greatest influence on the consolidation characteristics of these sediments.

The Giant Piston Corer has been used to obtain good quality sediment samples from several sedimentological/acoustic regimes in the northwestern Atlantic. A total of 185 m of sediment were recovered in nine cores with lengths ranging from 6.5 to 30.5 m. Sediment types include calcareous ooze, pelagic brown clay, organic‐rich gray‐green silty clay, and graded sand. Data are presented from three of these cores, taken from three different acoustic provinces. Seismic profiles and 3.5 kHz records from our study area on the eastern Bermuda Rise (4–5 km water depth) reveal marked differences in acoustic characteristics that are consistent with physical property variations at three of the core sites. This area contains three major features: (l)a broad, gently rolling plateau covered with about 300 m of sediment containing numerous conformable reflectors; (2) a slope region with maximum slope angle of 0.275 rad (16°), which is underlaid by acoustically transparent sediments; and (3) a region at the toe of ...

The papers in this special volume represent a small segment of the published work undertaken at the HEBBLE site. Because the program is interdisciplinary in nature, specialized results of import in biology and physical oceanography for example have been published elsewhere. This volume represents the early, site-descriptive results from the program. While many of the results are descriptive, considerable effort has been made by the authors to show the implications for understanding of marine sedimentological processes. The objective of this final paper is to summarize these results and implications and on occasion add our own personal interpretation in an attempt to obtain for the reader a greater synthesis of the project. The organization of the HEBBLE project is largely reflected in the resultant organization of the major sections in this volume. Most of the results here focus on the sedimentology and stratigraphy of the HEBBLE site, with some attention being paid to the characteristics of the biological community at the site as needed to refine a truly marine sediment transport model. The success of estimating the deposition rate and of relating the bed modification and deposition dynamics to fluid forcing by the boundary layer are the key goals for the next two years. The results on the characteristics of the suspended sediment show that much information can be obtained by careful and repeated sampling of material, if the results are placed in their appropriate fluid mechanical setting (see McCave, this volume). The key implication of the field observations and laboratory and theoretical studies relates to the generation of longitudinal ripples. Of greatest import is the conclusion that such features are depositional; thus we are looking at bedforms that may well be preserved into the stratigraphic record. But if longitudinal ripples are depositional, then one of the cornerstones in discriminating between turbidites and contourites is now removed. Turbidites are preserved because of the rapid and high rate of local downslope deposition. Longitudinal ripples may likely be produced from the waning period of a storm when extreme rates of deposition occur; without knowing the local slope it may well be impossible to discriminate between turbidite and contourite deposits.

Hydrographic sections of the northern valley of the Charlie-Gibbs fracture zone show that below 2000 m this passage through the Mid-Atlantic Ridge is filled mainly with Iceland-Scotland Overflow Water (ISOW). Two current meters moored within this water mass for 9 months, at sites just downstream from the 3675-m sill of the passage, recorded mean westerly currents of 3.8 and 4.4 cm s−1 These averages include mean westerly velocities of 7 to 8 cm s−1 during fall and spring and almost zero during the winter. Another long-termmeter and an array of meters deployed within 100 m of the sea floor showed steering of the Overflow Water current by the complex topography near the intersection of the transform fault and the spreading center. West of 35°W a different, denser water mass fills the deepest parts of the valley below about 3400 m; it also flows westward, as shown by meters moored 10 and 20 m above the valley floor for 1.5 to 4 days. However, its properties (potential temperature

Slumping on the eastern margin of Rockall Bank (eastern North Atlantic) 15,000–16,000 years before present, produced a series of debris flows and turbidity currents which have disrupted the Feni Drift around 56°N 13°W. Near-bottom investigations with a deeply towed instrument package show that a large area of slightly irregular hyperbolic echoes (11,000 km2) on the Feni Drift reflects the presence of a large debris flow. Sediment cores indicate that associated turbidity-current deposits are present over a larger area. The irregular topography of the debris-flow deposit results from large blocks of relatively undisturbed sediments protruding from the surface of the flow on the shallower, erosional portion of the flow. The topography on the deeper, depositional region of the flow is due to folded sediments. The debris flow obliterated part of a field of sediment waves, the northern portion of which now forms its northern boundary. Near-bottom investigations indicate that these sediment-wave troughs have been rapidly filling in during most of the last 73,000 years, suggesting that a large sediment input to the trough preceded slumping. Possibly an influx of sediment during the last glacial period, combined with increased wave activity during lower sea-level stands, overloaded these sediment slopes and made them more susceptible to failure.

HEBBLE's precise aim is to develop and to test explicit predictions about the response of adhesive/cohesive marine sediments to imposed and controlled stresses. Pursuit of this goal has necessitated a coordinated, interdisciplinary effort, to date including physical oceanographers, sedimentologists, radiochemists and biochemists, and biological oceanographers.

The Gulf Stream System contains a clockwise rotating set of bottom currents which influence the sea bed in the northern North American Basin. It is possible to interpret the present-day and historical record of current activity in this basin in terms of the deep flow of the Gulf Stream. This interpretation provides a much more satisfactory and consistent explanation of the geologic record than previous interpretations based strictly on the influence of the classic abyssal circulation. The present-day circulation of the Gulf Stream actively resuspends large quantities of sediment on and near the northern Bermuda Rise, and it is suggested that this resuspension may be due to the eddy field which is embedded in the westward flowing return flow. During the late Cenozoic the Gulf Stream System was responsible for depositing and shaping the major acoustic/sedimentological units on the relatively smooth surface of the Horizon A complex. Fine-grained, mont-morillionite-rich sediments derived from the chemically weathered saprolites of the Hudson and St. Lawrence drainage basins, were injected into the deep ocean basins by turbidity currents during the late Paleogene and the Neogene. Fine-grained turbidites from the St. Lawrence spread southward down the southeastern Sohm Abyssal Plain across the eastward-flowing Gulf Stream and its westward flowing return flows. Portions of these fine-grained sediments were entrained by the main and return flows and deposited downstream as acoustically non-laminated accumulations on the Corner and Bermuda Rises, respectively. The Hudson River input was injected into the deep basin through the Hudson Canyon System. The interaction of these fine-grained turbidites with the northward flowing deep flow of the Gulf Stream System led to the formation of the Gulf Stream Outer Ridge, here defined for the first time. During the late Neogene or early Pleistocene the Gulf Stream Outer Ridge was partially eroded by the Gulf Stream System and portions subsequently covered by coarse-grained Pleistocene turbidites. The coarse-grained illite-rich turbidites that flowed across the southeastern Sohm Abyssal Plain also crossed the main and return flows of the Gulf Stream System. The finer portions of these sediments were entrained and deposited as Pleistocene acoustically laminated sediments on the plateaus of the Bermuda and Corner Rises.

Studies of the interaction of deep ocean current systems with the sea bed are briefly traced from their beginnings in the decade of the mid-1950's to mid-1960's. Subsequent lines of study in physical oceanography, suspended sediment measurement, echocharacter mapping, microphysiography, sedimentology and developments of instrumentation are noted up to 1978 when the High Energy Benthic Boundary Layer Experiment started.

The Nuclear Research Submarine NR-1 was used to study morphology, sediment, and sediment—water interactions off the northwest coast of Puerto Rico, New detailed bathymetry from the surface-support ship, USS “Portland”, shows several submarine canyons in the area, some of them unreported previously. The north coast canyons, Arecibo, Tiberones and Quebradillas, are primarily erosional features although no recent turbidity-current evidence is seen. The canyons are presently filling with river-transported sediments. High-resolution seismic reflection profiles show the tectonic control of Mona Canyon. Fault zones can be outlined. The effects of sediment input, turbidity currents and bottom currents act to smooth and shape the canyon topography. River alluvium and biogenic debris are draped around the canyon and on the narrow benches between fault scarps on the north wall. Turbidity currents have carried large amounts of sediment from shallow water to the lower reaches of the canyon where the axis is being filled and leveled. Current measurements from NR-1 show oscillating flow with tidal frequency in the head of Mona Canyon.

Pelagic carbonate oozes interlayered with mixed carbonate-volcanic silts and sands have been the principal sedimentary facies deposited on the northwest African continental rise during the late Quaternary. Terrigenous sediments are absent, having been trapped along the slope and upper continental rise in basins formed by islands or seamounts. Most quartz silts and kaolinite clays have probably been supplied by eolian sources, and montmorillonite clays by the diagenesis of volcanic rocks. 3.5 kHz profiles show reflectors which are generally parallel to the seafloor. Individual reflectors are traceable for hundreds of kilometers. Faunal analysis and radio-carbon dating indicate sedimentation rates of $2-3 cm/10^{3} yr$ during the late Quaternary. Sedimentary parameters (mineralogy, carbonate abundance, lithology) apparently have remained unchanged during late Quaternary climatic fluctuations. Core data and bottom photographs indicate that reworking of bottom sediments has principally been by benthic organis...

A near-bottom survey has been made on the Eastward Scarp (32°50′N, 57°30′W) of the Bermuda Rise, which rises 1150 m above the 5500-m deep Sohm Abyssal Plain in the western North Atlantic. The survey reveals evidence of erosion and deposition at present and in the late Quaternary by the deeper levels of the westward flowing Gulf Stream Return Flow. Four distinct regions of increasing bed gradient show increasing sediment smoothing and scour in the transition from plateau to abyssal plain. Bedforms observed are current crescents, crag and tail, triangular ripples, elongate mounds, transverse mud ripples, lineations, and furrows ranging from 10 to 1 m or less in depth, decreasing generally with bed gradient. Measured near-bottom current speeds are up to 20 cm s −1 . Temperature structure on the lower, steep, slopes suggests that detachment of bottom mixed layers may occur there. Extensive net erosion appears to be confined to the lower steep slopes of the scarp. Reflection profiles (4 kHz) show that there has been erosion in areas thinly draped with recent sediments and in areas that show development of small scarps. The distribution of subsurface acoustic characteristics of the region corresponds broadly to the areas characterized by bed gradient and distinct sedimentation conditions. Subsurface hyperbolae, possibly caused by buried furrows, show furrow persistence through several tens of metres of deposition. Erosion occurs up to the top of the scarp during episodes of presumed stronger currents, which may correspond with intensified circulation during glacials.

Sediment traps were deployed to sample vertical fluxes of particulate matter both immediately above and within the nepheloid layer on the continental slope and rise of the western North Atlantic. The data from the traps are used to estimate the settling flux of particles and to attempt to distinguish between the primary flux (settling material from the upper water column) and the resuspended flux (settling material that has been resuspended from some region of the sea floor). From these data and from measurements of the net nepheloid standing crop of particles, estimates of the mean residence time for particles resuspended in the nepheloid layer are on the order of weeks to months from 15 to 500 m above bottom. Horizontal advection and diffusion are more important than vertical mixing in maintaining the upper portion of the nepheloid layer. The estimated flux of Al to the North Atlantic as atmospheric dust is one to two orders of magnitude less than the primary flux of Al measured just above the nepheloid layer, yet the range of accumulation rates of Al measured on the sea floor with radioisotopes is close to the values we define as the primary flux. Horizontal transport of Al-rich particles (mostly aluminosilicates) must therefore be significant even above the nepheloid layer. The size of 80—90% of the particles collected in the traps both above and within the nepheloid layer was less than 63 μm. The mean size of particles trapped in the nepheloid layer was about 22 μm, and above the nepheloid layer, 11 μm. The particles were probably in large aggregates when they encountered the traps, but identifiable fecal pellets are only a small fraction of the material collected in the traps. Between 6 and 15% of the primary production of particulate organic carbon in the euphotic zone was collected in floating sediment traps at the 100 m depth, indicating that most particulate organic carbon is recycled in the surface water. Only 2—3% of primary production of organic carbon was collected as primary flux 500 m above the sea floor. From an intercomparison of the composition of particles falling rapidly (collected in traps), falling slowly or not at all (collected in water bottles), and resting on the sea floor (from core tops), it was found that elements associated with biogenic matter, such as Ca, Sr, Cu and I, were carried preferentially by the rapidly falling particles. Once these particles reached the sea floor the concentration of those elements was decreased through decomposition, respiration, or dissolution. Dissolution appears to be rapid near or on the sea floor, because radiolarians, diatoms, and juvenile foraminifera were rare in core samples even though abundant in all traps. The dynamic nature of the nepheloid layer makes it possible for particles to be resuspended many times before they are finally buried, enabling sediment to be carried long distances from its origin. The recycling of particles near the sea floor may also increase dissolution of siliceous and carbonate matter.

A new large piston-coring device utilizing well-casing pipe 14 cm (o.d.), 11 cm (i.d.) has obtained cores over 20 m in length in water depths of 100 to 5000 m. Recovery ratios range from 0·76 to 0·87. Penetration varies between 24 and 29 m with a total core weight of 3200–4000 kg and 2 to 3 m of free fall.

The Northwest Atlantic Mid-Ocean Canyon is a depositional–erosional linear feature which extends southward from the northern Labrador Sea to the Sohm Abyssal Plain. Analysis of the data accumulated over the past 20 yr allows us to conclude: (1) The canyon does not have any major tributaries; (2) it is Quaternary in age; and (3) it is not a tectonic feature. However, neither the source nor the sedimentary process which constructed the canyon can as yet be established.

A large-diameter (14 cm) long piston corer (20–40 meters) has been used successfully to obtain relatively undisturbed sediment samples in water depths from 80 to 5500 meters with recovery ratios ranging between 0.76 and 0.87. The first recovered core (KN-10-1) was taken in the Stellwagen basin, western Gulf of Maine, at 81-meter water depth using a pipe length of 33.5 meters and total core weight of 3500 kg. The 21.74 meters of recovered sediment is black, gray to olive green silty illitic clay. Visual observations and results of consolidation studies indicate that structural disturbance to the sediments was not severe, that it is normally consolidated, and that no postdepositional compaction occurred. Another core (KN-27-1) was taken about 3 km east of the first core location, and 19.65 meters of sediment were recovered. An anomalous zone in core KN-10-1 reveals high water content (96%) and high shear strength (260 g/cm2). A marked abundance of diatoms may explain the high water content, and the highly flocculated nature of the abundant clay minerals observed with a scanning electron microscope accounts for the high shear strength. A similar anomalous zone in core KN-27-1 occurs where the water content is 53% and shear strength is about 250 g/cm2. Preliminary correlations with the 3.5 kHz subbottom profiles suggest that changes in both texture and water content may produce reflecting horizons.

Relatively strong bottom currents in the deep-sea can be inferred from sedimentary structures observed in bottom photographs and cores. The majority of photographs of high and steep topographic prominences, such as seamounts, escarpments and the crests of major ridges, show ripples, scour and rock outcrops. Although photographic current evidence is uncommon in the deeper waters of the ocean basin floor, striking and significant examples do occur. Current scour and ripples are observed beneath the Antarctic bottom current in the western South Atlantic, below the Gulf Stream in the Florida Straits and on the Blake Plateau, beneath the outflowing Mediterranean water west of Gibraltar and beneath the deep currents in the Drake Passage. The three types of deep-sea sands and silts are turbidite, accretionary, and residual. The latter two types are always associated with currents and in some areas turbidites are reworked by bottom currents. Traction velocities necessary for the transport of deep-sea sediment particles probably range from 4–60 cm/sec, velocities similar to those found through dynamic computations of geostrophic currents and observed by recent deep-sea direct current measurements.

Data from hydrographic stations, current measurements of four to six months' duration, and bottom photographs are used to infer the pattern of abyssal circulation in the vicinity of the Greater Antilles Outer Ridge north of Puerto Rico. A deep contour-following current of Antarctic Bottom Water with a small admixture of North Atlantic Deep Water flows southeast along the Bahama Banks, counterclockwise around the Silver Abyssal Plain, clockwise around the northwest end of the Greater Antilles Outer Ridge and then eastward along the north flank of the ridge. The Antarctic Bottom Water enters the basin of the Hatteras Abyssal Plain through Vema Gap, flows counterclockwise around the abyssal plain and then south along the lower continental rise into the region of the Greater Antilles Outer Ridge as a deep extension of the Western Boundary Undercurrent. This deep circulation pattern may provide a mechanism by which terrigenous detritus from the North American continental margin has been transported to the southeast and deposited to form the Greater Antilles Outer Ridge.

A study of compass-oriented sea-floor photographs, echograms and sediment cores on the Atlantic continental margin of North America has been made in order to evaluate the role of deep-sea contour currents in the shaping of the continental rise. The sediments on the upper continental rise consist of lutites which are being deposited on the sea floor in an environment devoid of strong bottom currents. Below an abrupt change in regional slope, that marks the boundary between the upper and lower continental rise, a swift bottom current is observed which flows to the southwest parallel to the contours. Beneath this current the surface sediments are distinctly coarser grained and long cores show many quartz silt laminations in the sedimentary column. Further downslope on the lower continental rise the currents are variable in direction and weaker. Measurements of northerly directions indicate that at certain locations the Gulf Stream may intermittently scour the sea floor. In the area of the Lower Continental Rise Hill a swift southwesterly current is again observed. Hyperbolic echo traces on echograms, prolonged multiple echo sequences, and wedging of sub-bottom reflecting interfaces can be mapped as distinct zones. These zones of sea floor micromorphology parallel the regional contours of the continental rise, and are produced by erosional and depositional processes of bottom currents. We conclude that the continental rise is a large sediment wedge which owes its shape to deep geostrophic contour currents.

The central North Pacific, one of the largest sedimentary provinces in the world ocean, is covered by red clays and abundant manganese nodules1,2. Few studies have focused on the sedimentation history in this area, presumably because of the difficulty of dating these sediments; biostratigraphically useful fossils are rare or absent, and palaeomagnetic stratigraphy is limited in this area to the past 2–3 Myr (refs 3–5 and R. A. Prince et al., in preparation). We present here a history of Cenozoic sedimentation for the central North Pacific based on lithologie, stratigraphie and sedimentological data from a single 25-m long giant piston core (B.H.C. and C.D.H. in preparation). A sedimentation model incorporating the present-day sedimentation patterns in the central North Pacific and the north-northwest movement of the Pacific plate during the Cenozoic seems to explain the observed sedimentological variations seen in this unique core.

Funding was provided by the International Cable Engineering Division of A T & T Communications under contract number 47-5697.

A meeting was held March 19-24, 1979, to produce an engineering plan for the High Energy Benthic Boundary Layer Experiment. This document outlines the principal conclusions of that meeting with a synopsis of the engineering requirements and with the reports of subcommittees on current sensing, conditional sampling, bottom mixed layer measurement, imaging, suspended particulate matter and sediment property and laboratory experimental program. In each report, details of sensors and properties to be determined are given. (Author)

The discovery of erosional and depositional bed forms in the deep ocean provides a means for estimating some properties of the benthic boundary layer and also raises questions about secondary circulation near the bed. Previously unexplained hyperbolic echoes are now seen (using a deeply towed instrument package) to be caused by furrows on the sea bed. Small furrows are 1–4 m wide, ¾–2 m deep, and spaced 20–125 m apart; large furrows are 50–150 m wide, 20 m deep, and spaced 50–200 m apart. Analogy with shallow-water furrows and bed forms produced in deserts and in laboratory experiments suggest an origin for the furrows through interaction of large-scale helical vortices developed in the boundary layer with the sediments. The thickness of the bottom-mixed water layer should be about one-half the furrow spacing. Measurements show that mixed layer thickness (20–100 m) are of the correct order of magnitude.

Since the development of high-resolution precision echo sounders in the 1950s, marine geologists have been characterizing different types of sea floor on the character of the echo return. Short-ping (less than 5 msec) high-frequency (3.5–12 kHz) echograms have been used to provide much information on the nature of the microtopography and the sediment stratification and have been used as a basis for interpreting the erosional and depositonal processes acting in the deep sea (Heezen et al., 1959; Heezen et al., 1966; Hollister, 1967; Schneider et al., 1967; Heezen and Johnson, 1969; Hollister and Heezen, 1972).

Geostrophic contour-following bottom currents involved in the deep thermohaline circulation of the world ocean appear to be the principal agents which control the shape of the continental rise and other sediment bodies.

Distinct latitudinal control of the distribution of marine microfaunas in the North Atlantic Ocean began in the early part of the Tertiary (Paleogene) and is related to the opening of the North Atlantic to the Arctic region about 60 m.y. ago when the present deep-water circulation pattern was probably initiated. At that time a boreal zoogeographic province was established in the Atlantic Ocean for the first time. Caribbean and Mediterranean benthonic foraminiferal faunas exhibit a marked degree of similarity during the early Tertiary. The gradual displacement of west-east current migration routes into higher latitudes and the compression of Spain against North Africa brought an end to this amphiatlantic distribution pattern about 15 m.y. ago. Initiation of glaciation abou 3 m.y. ago had a marked effect on circulation in the North Atlantic and temperature may have been the primary factor responsible for the many extinctions in planktonic Foraminifera. Sediment cores and bottom photographs provide evidence of measurable bottom circulation within the Western Boundary undercurrent in the western North Atlantic. This contour current has played a major role in controlling fine-grained sediment deposition since the early Tertiary. Data from recent deep drilling suggest that during middle Cretaceous to early to middle Tertiary times extensive unconformities occurred, and during the early Tertiary a sudden onslaught of deep circulation coupled with wholesale erosion and redeposition of deep-sea sediments may have occurred. Currents gradually diminished in strength during the late Tertiary and are now flowing at moderate velocities apparently sufficient to transport and deposit sediment with only local erosion. End_of_Article - Last_Page 331

ANTARCTIC bottom water (ABW) flows into the Pacific basin from the region between Macquarie Island and Antarctica1–3. Dynamic considerations suggest that the bottom water flows northwards as an intense current along the western boundary of the Pacific basin4,5. Oceanographers have noted that the Pacific is divided into at least four principal basins which are separated by relatively shallow sills that constrain the flow of water below 4,000 m (refs. 3, 6, 7). The narrow Samoan Passage, first recognized by Reid8, is perhaps the most important channel with depth sufficient to allow a significant flow of bottom water from the southern basin into the central basin. Bottom current velocities in the passage of 5 to 15 cm s−1 towards the north have been measured9 and Nansen casts in the passage revealed a sharp decrease in temperature of 0.27° C in 150 m at a depth of about 4,400 m10. This benthic thermocline marks the abrupt boundary between the deep water and the ABW and supports the observation of cold bottom water flow through the passage. (In a personal communication, P. Lonsdale and J. Reid suggested that the bottom water may include entrained high salinity Atlantic deep water as well as ABW.)

A preliminary study of more than 100 deep-sea cores from abyssal plains has revealed two examples of recent muddy sands of the graywacke type which, together with the microcrystalline matrix, form a bimodal-size distribution sands have a well-sorted framework of quartz, feldspar, and rock fragments which, together with the microcrystalline matrix, form a bimodal-size distribution that is also typical of ancient graywackes. The matrix is considered to be primary.