Norwegian Sea Overflow Water Research Articles

Holocene evolution of deep circulation in the northern North Atlantic traced by Sm, Nd and Pb isotopes and bulk sediment mineralogy

Bulk mineralogy, Sm, Nd and Pb elemental and isotopic compositions of the clay‐size fraction of Holocene sediments were analyzed in three deep North Atlantic cores to trace the particle provenance. The aims of the present paper are to identify the origin of the particles driven by deep currents and to reconstruct deep circulation changes over the Holocene in the North Atlantic. The three cores are retrieved in fracture zones; two of them are located in the Island Basin along the gyre of North Atlantic Deep Water, and the third core is located off the present deep circulation gyre in the Labrador Sea. Whereas sedimentary supplies in the Labrador Sea were constantly derived from proximal sources, the geochemical mixing trends in the Iceland Basin samples indicate pronounced changes in the relative contribution of continental margin inputs over the past 6 kyr. Supplies from western European margin that sharply increased at 6 kyr were progressively diluted by a larger contribution of Scandinavian margins over the last 3 kyr. Changes in composition of the particles imply significant reorganization of paleocirculation of the deep North Atlantic components in the eastern basins: mainly reorganizations for both Iceland‐Scotland Overflow Water and Norwegian Sea Overflow Water. Moreover the unusual Pb isotopic composition of the oldest sediments from the southern Iceland Basin indicates that distal supplies from Greenland margin were driven into the Iceland Basin, supporting a deep connection between Labrador Sea and Iceland Basin through the Charlie Gibbs Fracture Zone prior the Holocene Transition period.

Holocene sea-surface conditions in the North Atlantic—contrasted trends and regimes in the western and eastern sectors (Labrador Sea vs. Iceland Basin)

Two sediment cores from the Labrador Sea and one from the Iceland Basin were analysed in order to compare Holocene sea-surface conditions across the northern North Atlantic. δ 18O measurements on meso- and epi-planktonic foraminifera ( Neogloboquadrina pachyderma left-coiling and Globigerina bulloides), along with sea-surface condition reconstructions using transfer functions based on dinoflagellate cyst assemblages, were used to document thermohaline properties of the upper water column. At the Labrador Sea sites, a decreasing trend in sea-surface temperature following an early Holocene maximum, is observed. At the Iceland Basin site, a hiatus between 13 and 8 ka BP prevents us from documenting the entire Holocene, but a stable to slightly decreasing trend is observed from 8 ka BP onwards. Sea-surface salinity and potential density ( σ θ ) show little variation at the Orphan Knoll site in the Labrador Sea since 7 ka BP, whereas they depict a decreasing trend at the two other sites since ca 8 ka BP. This shift in σ θ values suggests a progressive enhancement of Labrador Sea Water formation and a relative decrease in the Nordic seas components of the North Atlantic Deep Water (i.e., Denmark Strait Overflow Water and Norwegian Sea Overflow Water). Discrepancies are also observed with regard to millennial-frequency oscillations in sea-surface conditions. At the easternmost site, they likely match previously documented cold water pulses in the north-eastern Atlantic, whereas they do not show any clear pattern at the westernmost sites. Hence, the western and eastern areas show different behaviours with respect to both long-term and high-frequency fluctuations.

Norwegian Sea overflow through the Faroe–Shetland gateway as documented by its bedforms

We have used seabed information from three deep-tow side-scan sonar surveys in order to trace the high-energy current core of Norwegian Sea Overflow Water (NSOW) from the Faroe–Shetland Channel (FSC) through the Faroe Bank Channel (FBC) towards the southern flank of the Iceland–Faroe Ridge. A hierarchy of bedform types was used for estimating maximum mean near-bottom current speed, which has been compared with results from current meter measurements. We conclude that in the FSC maximum mean NSOW near-bottom flow speed at some sites may occasionally approach or even exceed 1.0 m/s. Both seabed data and oceanographic information indicate that the current core is concentrated along the Faroes slope at the 500–600-m depth stratum. With a change of the large-scale channel topography towards the southern extremity of the Faroe Plateau, where the FSC turns west towards the FBC, the slope current core gradually descends towards basin depth. Along the Wyville–Thomson Ridge (1100–1200 m water depth) maximum mean NSOW near-bottom current speed decreases slightly towards the west as the high-speed current core runs upslope the ridge. At the entrance to the narrow channel between Faroe Bank and the Faroe Plateau funnelling results in a renewed current acceleration at basin depth. Seabed evidence shows that further west in the FBC the NSOW high-speed current core is detached from the channel floor over larger areas. From the FBC outlet most of the overflow waters continue as a high-energy contour current concentrated along the southern flank of the Iceland–Faroe Ridge between 600 and 1000 m water depth.

The Faroe–Shetland Gateway: Late Quaternary water mass exchange between the Nordic seas and the northeastern Atlantic

Thirteen piston and gravity cores from the Faroe–Shetland area were investigated for their planktic and benthic foraminiferal and oxygen isotopic distributions. Eight time-slices between 18 ka BP and the present were reconstructed to study variations in surface and deep water exchange between the SE Norwegian Sea and the northeast Atlantic Ocean. Today, a relatively strong northward flow of warm North Atlantic surface water is counterbalanced by a southward outflow of newly convected cold bottom water, the Norwegian Sea Overflow Water. During the last glacial maximum at 18 ka BP both the surface and bottom flows were slow and the climate conditions were Arctic. The convection north of the Faroe area was weak and unstable. The first indication of the deglaciation is a decrease in the planktic oxygen isotope values discernible southwest of the Faroe Islands at 15.5 ka BP. The deglaciation proceeded northeast and eastward synchronous with a gradual intensification of northward flowing warmer Atlantic Intermediate Water along the sea bottom. Meltwater fluxes increased between 14 and 13 ka BP producing cold surface waters, and the climatic cooling was extreme. There was no southward overflow of cold bottom water during this time period and the exchange of water masses between the Nordic seas and the North Atlantic Ocean was essentially reversed, i.e. estuarine. During the Bølling Interstadial at 12.5 ka BP northward flowing warm surface water was present to the east of the Faroe–Shetland Channel, wedged below a tongue of polar water spreading from the northwest and reaching into the Faroe–Shetland Channel. Convection in the Nordic seas and overflow of cold deep water started during the Bølling Interstadial. The polar water spread more eastward and southward during the following cold spell, the Younger Dryas, around 10.3 ka BP. The polar water was overlying the warmer, but more saline Atlantic water, which flowed northward below the cold surface water. The overflow of cold bottom water was supposedly only slightly weaker than during the Bølling Interstadial. Strong inflow of warm surface water took place during the Early Holocene at 9.5 ka BP and relatively dense cold water flowed southward along the bottom. The rate of water mass exchange reached a maximum at 6.5 ka BP, when both the inflow of warm Atlantic surface water and the outflow of cold dense bottom water appear to have been stronger than today.

Late Quaternary glacigenic contourite, debris flow and turbidite process interaction in the Faroe-Shetland Channel, NW European continental margin

Abstract The Faroe-Shetland Channel is an important conduit or gateway for the southward flow of cold bottom waters formed in the Norwegian Sea. This Norwegian Sea Overflow Water (NSOW) finds several spillover channels across the Wyville-Thomson Ridge, eventually descending into the northern Rockall Trough and Iceland Basin. The Neogene channel floor succession predominantly displays a broad sheeted drift geometry. Bottom current scours and channels were apparently inherited from an episode of enhanced bottom current activity in late Oligocene/early Miocene. The late Quaternary channel-floor succession is dominated by distal glaciomarine sediments, derived from the shelf and slope during glacial stages and mostly transported by ice-rafting. Glacigenic debris flows and minor turbidity currents were also active across the slope region. Consequently, the principal channel-floor facies are glacigenic contourites that show extensive bioturbation, rare primary structures, mixed composition and marked grain size variation. These features indicate the important influence of cyclical fluctuations in bottom current velocity throughout both stadial and interstadial or interglacial periods. However, the concentration of sandy contourites, erosive surfaces and top-only contourites during interstadials/interglacials and during phases of marked cooling or warming testify to the significance of climate-control on contourite deposition.

Contourites of the Nova Scotian continental rise and the HEBBLE area

Abstract The Nova Scotian continental rise is swept by a Deep Western Boundary Current system comprising layers of Labrador Sea Water overlying a core of Norwegian Sea Overflow Water at depths of 3100-3900 m, and below about 4600 m a cold stream of southern-source water. Seismic-reflection data show that the rise contains sediments transported downslope in channels and debris lobes, but there is also evidence of current-controlled deposition and erosion in the post-Eocene sequence. The rise is now mantled by Holocene contourites that have accumulated at a rate of c . 6 cm ka -1 and are <1 m thick. Bottom photographs show a zonation in current effects and bedform types, with longitudinal ripples and strong currents prevalent at 4800-5000 m, smaller bedforms and progressively weaker currents up to c . 4000 m, and mostly tranquil seafloor above 4000 m. Bedform scales and orientations also suggest significant short-term (hours to weeks) variability in current velocity but a mean contour-following flow to the southwest at longer time scales (months to years). These structures are not preserved in the sediment because of pervasive bioturbation and the uppermost layers have negligible preservation potential. The sediments display clear current controlled effects in their grain-size structure involving both percentage of (foraminiferal) sand, and size and percentage in the 10-63 µm range, the 'sortable silt'. There is a sand-rich zone at 4800-4900 m and below 5000 m, and a decreasing silt/clay ratio from 5100 m up to 4000 m. Although much of the sedimentary sequence probably has been emplaced by downslope processes, it has been significantly modified by the Deep Western Boundary Current. Particularly strong and variable currents which rework sediments below c . 4800 m probably are engendered by interaction of Gulf Stream eddies with the DWBC. Although strong currents and upstream input from tur-bidites and debris flows might be thought to favour a coarse-grained deposit, the facies at the HEBBLE (High Energy Benthic Boundary Layer Experiment) site is muddy contourite with ≤ 12% sand.

Late Quaternary palaeo-oceanography of the Denmark Strait overflow pathway, South-East Greenland margin

Direct interaction between the atmosphere and the deep ocean basins takes place today only in the Southern Ocean near the Antarctic continent and in the northern extremity of the North Atlantic Ocean, notably in the Norwegian–Greenland Sea and Labrador Sea. Cooling and evaporation cause surface waters in the latter region to become dense and sink. At depth, further mixing occurs with Arctic water masses from adjacent polar shelves. Export of these water masses from the Norwegian–Greenland Sea (Norwegian Sea Overflow Water) to the North Atlantic basin occurs via two major gateways, the Denmark Strait system and the Faeroe– Shetland Channel and Faeroe Bank Channel system (e.g. Dickson et al. 1990; Fig.1). Deep convection in the Labrador Sea produces intermediate waters (Labrador Sea Water), which spreads across the North Atlantic. Deep waters thus formed in the North Atlantic (North Atlantic Deep Water) constitute an essential component of a global ‘conveyor’ belt extending from the North Atlantic via the Southern and Indian Oceans to the Pacific. Water masses return as a (warm) surface water flow. In the North Atlantic this is the Gulf Stream and the relatively warm and saline North Atlantic Current. Numerous palaeo-oceanographic studies have indicated that climatic changes in the North Atlantic region are closely related to changes in surface circulation and in the production of North Atlantic Deep Water. Abrupt shut-down of the ocean-overturning and subsequently of the conveyor belt is believed to represent a potential explanation for rapid climate deterioration at high latitudes, such as those that caused the Quaternary ice ages. Here it should be noted, that significant changes in deep convection in Greenland waters have also recently occurred. While in the Greenland Sea deep water formation over the last decade has drastically decreased, a strong increase of deep convection has simultaneously been observed in the Labrador Sea (Sy et al. 1997).

Late Quaternary palaeo-oceanography of the Denmark Strait overflow pathway, South-East Greenland margin

Direct interaction between the atmosphere and the deep ocean basins takes place today only in the Southern Ocean near the Antarctic continent and in the northern extremity of the North Atlantic Ocean, notably in the Norwegian–Greenland Sea and Labrador Sea. Cooling and evaporation cause surface waters in the latter region to become dense and sink. At depth, further mixing occurs with Arctic water masses from adjacent polar shelves. Export of these water masses from the Norwegian–Greenland Sea (Norwegian Sea Overflow Water) to the North Atlantic basin occurs via two major gateways, the Denmark Strait system and the Faeroe– Shetland Channel and Faeroe Bank Channel system (e.g. Dickson et al. 1990; Fig.1). Deep convection in the Labrador Sea produces intermediate waters (Labrador Sea Water), which spreads across the North Atlantic. Deep waters thus formed in the North Atlantic (North Atlantic Deep Water) constitute an essential component of a global ‘conveyor’ belt extending from the North Atlantic via the Southern and Indian Oceans to the Pacific. Water masses return as a (warm) surface water flow. In the North Atlantic this is the Gulf Stream and the relatively warm and saline North Atlantic Current. Numerous palaeo-oceanographic studies have indicated that climatic changes in the North Atlantic region are closely related to changes in surface circulation and in the production of North Atlantic Deep Water. Abrupt shut-down of the ocean-overturning and subsequently of the conveyor belt is believed to represent a potential explanation for rapid climate deterioration at high latitudes, such as those that caused the Quaternary ice ages. Here it should be noted, that significant changes in deep convection in Greenland waters have also recently occurred. While in the Greenland Sea deep water formation over the last decade has drastically decreased, a strong increase of deep convection has simultaneously been observed in the Labrador Sea (Sy et al. 1997).

Norwegian Sea overflow variability and NE Atlantic surface hydrography during the past 150,000 years

The long-term variability of Norwegian Sea Overflow through the Faeroe Bank Channel and its relation to NE Atlantic surface hydrography is reconstructed based on sedimentological, micropalaeontological and stable isotope analysis, as well as AMS 14C dating, of a sediment core (ENAM-33) retrieved near the outlet of the channel. Additional palaeoceanographic information has been obtained from three nearby sediment cores (ENAM-30, -32, 94-13). Our results demonstrate enhanced circulation of Norwegian Sea Overflow Water (NSOW) at times of increased Sea Surface Temperature (SST), and a reduction or cessation of NSOW flow at times of low SST. During the Last Glacial Maximum (LGM), NSOW activity was at a minimum from about 22 ka to 16 ka. After the LGM, NSOW circulation was weak until a major flow pulse of NSOW occurred, dated at between 13.4 ka and 12.0 ka. Thereafter, NSOW flow was once again reduced. Renewed intensification of NSOW flow did not occur until after 9.8 ka, whereas a further increase in NSOW activity is observed after the early Holocene. The timing of both glacial–interglacial and deglacial NSOW variability is similar to the timing of changes in deep-water circulation reconstructed from cores from the open North Atlantic in previous studies. Whereas most of these studies involve the reconstruction of changes in the proportion of North Atlantic Deep Water (NADW) relative to southern-source deep waters, the lithology of the cores from near the Faeroe Bank Channel outlet provides a qualitative but direct indication of the strength of NSOW currents. This study confirms that the NADW changes observed in the North Atlantic are due to a reduction in NSOW, and cannot be attributed only to an increase in southern-source deep waters. Furthermore, our results indicate that NSOW flow through the Faeroe Bank Channel was not a source of Glacial North Atlantic Intermediate Water (GNAIW), and that a subpolar source for this water mass is likely.

Neogene seismic facies and deep-water gateways in the Faeroe Bank area, NE Atlantic

Interpretation of sedimentary structures on high-resolution multichannel sleeve-gun seismic records has been used to illustrate changes in the Neogene deep-water flow regime around the Bill Bailey Bank. At the southern end of the Faeroe–Shetland Channel, the main part of the Norwegian Sea Overflow Water (NSOW) turns into the Faeroe Bank Channel. Leaving the Faeroe Bank Channel, the NSOW forms an erosive flow along the southern part of the Faeroe Platform where the basalt is outcropping. It is indicated that part of the overflow waters continues to the northwest entering the N Atlantic Basin to the west of the Faeroe–Rockall Plateau. Intermittent overflow of NSOW occurs via the Wyville–Thomson Ridge Complex. Between the Bill Bailey Bank and Faeroe Bank, sedimentary structures seen on the seismic profiles indicate the existence of a former NSOW pathway which was active until presumably the Late Pliocene. Along the southern rise of the Bill Bailey Bank the seismic profiles indicate that deep-water flow was still active in the late Quaternary, which presumably can be related to intermittent overflow of the Wyville–Thomson Ridge Complex.

Quaternary sedimentation and Norwegian Sea overflow pathways around Bill Bailey Bank, northeastern Atlantic

Using high-resolution multi-channel sleeve gun seismics and TOBI side-scan sonar and subbottom profiler acoustic data, as well as sediment cores and published hydrographic measurements, a study has been made of Quaternary sedimentation patterns and depositional processes around Bill Bailey Bank (BBB) in the northeastern Atlantic. From the western outlet of the Faeroe Bank (FB) Channel, which is the main conduit of Norwegian Sea Overflow Water (NSOW), there is a complex pattern of NSOW-controlled sediment transport pathways extending to areas north of Lousy Bank (LB). The acoustic data and sediment cores reveal the existence of a contour current at water depths exceeding 1000 m flowing from the Wyville–Thomson Ridge via the southern rise of FB and BBB into the valley between BBB and LB and further northward towards the area north of BBB. We conclude that this contour current is related to intermittent NSOW overflow of the Wyville–Thomson Ridge. The maximum bottom flow speed along this secondary NSOW flow pathway is inferred to reach about 0.50 m/s. Bedforms observed in the northeastern part of the study area, north of BBB, suggest a maximum bottom flow speed of locally at least 1.0 m/s. The maximum thickness (0.5 s TWT) of the Pliocene–Quaternary seismic sequence, at water depths in excess of 1000 m, south of BBB is nearly twice that found north of BBB. On average, higher accumulation rates in the former area can be ascribed to a hydrographic setting favouring sediment accumulation which, especially under glacial conditions, markedly differed from conditions north of BBB. This applies to both bottom and surface circulation. Linear bulk sedimentation rates north of BBB were largest during deglacial warming when the rates were about three times higher than the average Holocene rate (4.5 cm/ka). Increased fluxes during deglaciation reflect transport and depositional conditions associated with reactivation of the NSOW current system. South of BBB, average linear bulk accumulation rates were persistently higher, i.e. in the range 15–23 cm/ka, during various Late Pleistocene stages between 30 and 10 ka.

Sedimentation and climate-induced sediments on Feni Ridge, Northeast Atlantic Ocean

Feni Ridge has been studied in detail in the area between 54 and 56°N and 12 and 16°W by using high-resolution profiling, water column sampling with transmissometer and CTD, and by piston and box core sampling. The crest of Feni Ridge is absent in the north of the study area, but on passing towards the south it begins to take shape and to increase in height; its distance from the Rockall slope also increases towards the south. On the lower southeastern flank a secondary ridge crest has locally developed, and sediment waves along the flanks have amplitudes of up to 45 m. The highest and steepest waves are found in the south, the waves in the north being somewhat lower. Bottom nepheloid layers of 175 m near the foot of Rockall Plateau and 125 m near the crest are attributed to a Norwegian Sea Overflow Water current with a core at 2100 m. Bottom nepheloid layers of 75–100 m on the lower flanks of the ridge are thought to have been caused by Northeast Atlantic Deep Water. Piston cores reflect alternations of glacial and interglacial periods. Interglacials have a high carbonate and C org content, a positive Eu anomaly and negative Ce anomaly, and a low Rb La ratio. They are dominated by a subpolar and transitional foraminiferal assemblage. Glacials show a low carbonate and C org content, a reduced Eu anomaly and a high Rb La ratio, suggesting increasingly continental sediment sources. These deposits are dominated by a polar planktonic foraminiferal assemblage characterized by N. pachyderma (s) and G. quinqueloba. Sedimentation rates reflect differential deposition. At crestal station NA 87-25 rates are 17.1 cm/k.y. for stage 1, 14.6 cm/k.y. for stage 2, 3.4 cm/k.y. for stage 3 and 5.4 cm/k.y. for stage 4. Stage 5a–e has a mean sedimentation rate of 6.1 cm/k.y. Over the last 125 k.y. the development of Feni Ridge has taken place under a relatively low energy current regime.

Glacial-Holocene stratigraphy, chronology, and paleoceanographic observations on some North Atlantic sediment drifts

Several sediment drifts in the deep North Atlantic have been studied using oxygen isotopes, percent CaCO 3, clay mineralogy, tephra content and Accelerator Mass Spectrometer (AMS) radiocarbon dating. Because enhanced deposition of fine-grained terrigenous sediment increases rates of sedimentation by as much as 100 times the regional mean, sediment drifts contain high-resolution records of climate and oceanographic change. The major results of this study are that: (1) patterns of sedimentation rate are regionally variable. In the Western North Atlantic (Bermuda Rise, Blake-Bahama Outer Ridges), rates of sedimentation have been relatively low during the Holocene, but were as high as 200 cm/1000 years during the latest glacial episode, reflecting a greater supply of sediment due to glaciation and sea level lowering. Deep circulation patterns were probably similar to today. In the northern North Atlantic (Gardar Drift), which is dominated today By Norwegian Sea Overflow Water (NSOW), sedimentation rates were lowest during the glacial, despite the availability of sediment, because NSOW production was stopped. Rates peaked during the early Holocene as NSOW resumed. Glacial and Holocene rates of sedimentation were roughly comparable in the northeast Atlantic (Feni Drift) because sedimentation at that location is influenced less by NSOW than by Northeast Atlantic Deep Water (NEADW), which has a large southern-source component. (2) Stable isotopic events of deglacial age are preserved with unprecedented clarity in the sediment of North Atlantic drifts. Where the Younger Dryas cooling and the Vedde Ash occurred together, at 56°N on Feni Drift, direct AMS dating suggests that the radiocarbon age difference between the surface ocean and atmosphere reservoirs in the northeast Atlantic was the same 10,500 years ago as it is today (about 400 years). A δ 18O maximum of Younger Dryas age indicates cooler or more saline surface waters above the Bermuda Rise at 33°N and as far south as 28°N on the Bahama Outer Ridge. Two δ 18O minima at 12,000 and 13,500 radiocarbon years BP on the Bahama Outer Ridge and the Bermuda Rise probably resulted from lowered surface water salinity. These low salinity events most likely originated as meltwater discharge down the Mississippi and into the Gulf of Mexico, followed by advection into the open North Atlantic via the Gulf Stream.

Late Quaternary deep-ocean circulation

A review of paleoceanographic studies dealing with late Quaternary deep-water circulation in the oceans is presented. These studies, which are based on the analysis of deep-sea benthic foraminiferal faunal and geochemical data and sedimentological data, are discussed in light of physical oceanographic conditions in the modern ocean. We include in this review a re-evaluation of benthic foraminiferal data from North Atlantic and Southern Ocean piston cores and suggest that the dominance of Uvigerina during glacial intervals reflects increased amounts of organic carbon at the sea floor compared to modern values. A number of geochemical studies have suggested that the production or characteristics of North Atlantic Deep Water (NADW) changed during glacial times. Although these changes have been thought to result from a cessation of overflow water from the Norwegian-Greenland Seas, it is suggested here that seasonal sea-ice cover was possible over southern portions of the Norwegian Sea during glacial intervals. The presence of seasonally open water would have allowed Norwegian Sea Overflow Water to have been produced, although perhaps at lower volumes and with different hydrographic properties than at present. The record of Antarctic Bottom Water (AABW) circulation does not show a simple relationship with paleoclimatic oscillations, indicating that changes in oceanographic conditions in the Southern Ocean had little effect on AABW formation. The AABW record contrasts with the glacial-interglacial cycles of NADW, suggesting no direct link between AABW and NADW circulation. A variety of data suggests that changes in Pacific Deep Water circulation occurred as a result of glacial production of North Pacific deep water or from an increased flux of Southern Ocean water.

Current Variability near the Southeast Newfoundland Ridge

Abstract A cooperative moored array experiment to measure currents and temperatures in the vicinity of the Southeast Newfoundland Ridge was carried out over a 14-manth period starting September 1979 and ending December 1980. Measurements were obtained from an incoherent exploratory army of nine moorings instrumented at 500, 800, 1500 and 4000 m levels. The objectives were to explore the amplitude, spatial variability and time scales of the eddy field in the decay region of the Gulf Stream. An additional experiment and mooring to examine the influx of Norwegian Sea Overflow Water into the western North Atlantic Basin was imbedded in the array. Supplementary CTD and XBT profiles and a single current profile were taken during deployment and recovery cruises to aid in interpretation of the moored measurements and identification of water masses. A description of the array experiment and preliminary results are reported. Mean velocities were weak compared with rms eddy velocities and were not determined with st...

Seismic evidence for Early Tertlary bottom-current controlled deposition in the Charlie Gibbs Fracture Zone

Buried, mounded and acoustically transparent sediment bodies banked up against a basement ridge at the eastern end of the Charlie Gibbs Fracture Zone are interpreted as a type of contourite drift. As they are probably of Eocene—?Palaeocene age, their origin has implications for the little-known deep circulation pattern in the NE Atlantic during this period. Possible sources for the bottom currents responsible for these drifts include Labrador Sea Water, Norwegian Sea Overflow Water, Antarctic Bottom Water or a Gulf Stream deep gyre system. The preferred source is an early development of Antarctic Bottom Water or an arm of a Gulf Stream Gyre flowing eastwards through the fracture zone.

The active part of Charlie‐Gibbs Fracture Zone: A study using sonar and other geophysical techniques

A short survey with Gloria side‐scan sonar and other geophysical instruments has revealed new information about Charlie‐Gibbs fracture zone between 29° and 36°W. The traces of two transform faults have been clearly delineated. They fit small circles about the pole of rotation with an rms error of only about 1 km, but they do not always follow the deepest parts of the transform valleys. The transforms are joined by a short spreading center at 31°45′W. The median transverse ridge appears to have been produced by normal seafloor spreading at this center and bears identifiable Vine‐Matthews magnetic anomalies. A transverse ridge along the eastern inactive part of the northern transform may be an intrusive feature. Considerable thicknesses of sediment appear to have been deposited in the northern transform valley from Norwegian Sea overflow water passing through the fracture zone, but transverse ridges have prevented the sediment reaching the southern valley.

North Atlantic Deep Water formed by the later middle Miocene

There is evidence that in Mesozoic and Palaeogene times both north and south subpolar areas had a mild climate1,2. Unlike the present day climatic symmetry, with warm equatorial regions and ice-covered polar regions, a strong planetary asymmetry developed in the middle Miocene, when the Antarctic ice cap was established while the northern high latitudes remained unglaciated3. The symmetry was restored when a Northern Hemisphere continental ice cover was established ∼3 Myr ago in the middle Pliocene4. These changes influenced the formation of deep water in the ocean. At present the cold, dense bottom waters originate in only two high latitude areas—the Weddell Sea producing Antarctic Bottom Water, and the Norwegian Sea, producing Norwegian Sea Overflow Water which after mixing is a major component of the North Atlantic Deep Water. The production of cold deep water in the Southern Ocean started at the Eocene–Oligocene boundary3. We present here new evidence from isotope analyses of benthic Foraminifera from DSDP site 116 indicating that the production of oxygenated deep water in the North Atlantic Ocean started in the late middle-Miocene, ∼12 Myr ago.

A heat flow survey at anomaly 13 on the Reykjanes Ridge: A critical test of the relation between heat flow and age

We obtained 17 successful 12.5 m heat flow measurements on Anomaly 13 east of the Reykjanes Ridge. All temperature gradients were linear below 7.5 m but showed a large increase toward the sediment surface in the upper 5m. From the linear gradients and thermal conductivities measured on the sediment recovered in the cores we computed a mean flow for Anomaly 13 (35 m.y. B.P.) of 1.98±0.24 μcal/cm2 s (82.9±10.1 mW/m2). Thirteen of the stations were taken in a detailed survey of a well‐sedimented area at 56°N, 29°30′W. These measurements have a mean of 2.01±0.25 μcal/cm2 s (84.2±10.5 mW/m2). Almost all of the scatter in the data in this area can be accounted for by a combination of the refractive effect of the high thermal conductivity of the basement rocks and instrumental error. However, it is not possible from the present data set to rule out hydrothermal circulation in a ‘permeable’ basement as an explanation of the variation in the data. The mean value for the survey area is in excellent agreement with the relation between heat flow and the reciprocal of the square root of age predicted for oceanic crust less than 120 m.y. old. The sharp decrease in the temperature in the upper 5 m of sediment is most plausibly accounted for by a yearly periodic temperature change of an amplitude of 0.40°C in the Norwegian Sea overflow water. We suggest that the major temperature change occurs in late spring and reaches the area of our survey by early summer.

The deep waters of the western North Atlantic during the past 24,000 years, and the re-initiation of the Western Boundary Undercurrent

Changes in composition of benthic foraminiferal faunas in two cores from the North American continental rise (at ∼ 3000 m and ∼ 4000 m depth) and one core from the center of the North American Basin (at ∼ 4600 m depth) reflect changes in the nature of bottom water in the western North Atlantic during the past 24,000 years. During late glacial times the portion of the basin below 4500 m depth was filled with cold, well oxygenated water, probably of local North Atlantic origin. This water was overlain by more than 1500 m of “old”, oxygen-deficient water, of probably southern (Antarctic) origin. A basinwide change of deep water occurred first at about 12,500 yr. B.P., which resulted in a strong vertical differentiation of the water masses. As the deepest water became isolated and was losing oxygen, a western boundary undercurrent became established which carried well-oxygenated water at ∼ 3000 m depth along the continental rise. “Old” oxygen-deficient water reoccupied the levels between these two. A second basinwide change occurred shortly after 9000 yr. B.P., marking the beginning of modern deep-water conditions. The Western Boundary Undercurrent enlarged, moved downslope and flushed out the “old” glacial-age deep water. Norwegian Sea overflow reached its maximum about 4000 years ago. The center of the basin experienced the second change with a delay of about 1000 years, when it filled with Norwegian Sea Overflow Water, and later (after 5000 yr. B.P.) received large proportions of Antarctic Bottom Water. These deep-water events are correlated with the initial retreat of the polar front into the Norwegian Sea, its southward readvance, and the final opening of the Norwegian Sea to the inflow of North Atlantic Drift water.