CaCO3, foraminifer fragmentation, benthic carbon isotopes, coarse lithic counts and radiogenic isotopes (Sr, Nd, Pb) of the Pliocene and earliest Pleistocene terrigenous component (3.3-2.4 Ma) of IODP Site 306-U1313

We present Plio-Pleistocene records of sediment color, %CaCO3, foraminifer fragmentation, benthic carbon isotopes (d13C) and radiogenic isotopes (Sr, Nd, Pb) of the terrigenous component from IODP Site U1313, a reoccupation of benchmark subtropical North Atlantic Ocean DSDP Site 607. We show that (inter)glacial cycles in sediment color and %CaCO3 pre-date major northern hemisphere glaciation and are unambiguously and consistently correlated to benthic oxygen isotopes back to 3.3 million years ago (Ma) and intermittently so probably back to the Miocene/Pliocene boundary. We show these lithological cycles to be driven by enhanced glacial fluxes of terrigenous material (eolian dust), not carbonate dissolution (the classic interpretation). Our radiogenic isotope data indicate a North American source for this dust (~3.3-2.4 Ma) in keeping with the interpreted source of terrestrial plant wax-derived biomarkers deposited at Site U1313. Yet our data indicate a mid latitude provenance regardless of (inter)glacial state, a finding that is inconsistent with the biomarker-inferred importance of glaciogenic mechanisms of dust production and transport. Moreover, we find that the relation between the biomarker and lithogenic components of dust accumulation is distinctly non-linear. Both records show a jump in glacial rates of accumulation from Marine Isotope Stage, MIS, G6 (2.72 Ma) onwards but the amplitude of this signal is about 3-8 times greater for biomarkers than for dust and particularly extreme during MIS 100 (2.52 Ma). We conclude that North America shifted abruptly to a distinctly more arid glacial regime from MIS G6, but major shifts in glacial North American vegetation biomes and regional wind fields (exacerbated by the growth of a large Laurentide Ice Sheet during MIS 100) likely explain amplification of this signal in the biomarker records. Our findings are consistent with wetter-than-modern reconstructions of North American continental climate under the warm high CO2 conditions of the Early Pliocene but contrast with most model predictions for the response of the hydrological cycle to anthropogenic warming over the coming 50 years (poleward expansion of the subtropical dry zones).

Quantitative component analysis of the coarse fraction of surface sediments from the Persian Gulf

In the Persian Gulf and the Gulf of Oman marl forms the primary sediment cover, particularly on the Iranian side. A detailed quantitative description of the sediment components > 63 µ has been attempted in order to establish the regional distribution of the most important constituents as well as the criteria governing marl sedimentation in general. During the course of the analysis, the sand fraction from about 160 bottom-surface samples was split into 5 phi° fractions and 500 to 800 grains were counted in each individual fraction. The grains were cataloged in up to 40 grain type catagories. The gravel fraction was counted separately and the values calculated as weight percent. Basic for understanding the mode of formation of the marl sediment is the "rule" of independent availability of component groups. It states that the sedimentation of different component groups takes place independently, and that variation in the quantity of one component is independent of the presence or absence of other components. This means, for example, that different grain size spectrums are not necessarily developed through transport sorting. In the Persian Gulf they are more likely the result of differences in the amount of clay-rich fine sediment brought in to the restricted mouth areas of the Iranian rivers. These local increases in clayey sediment dilute the autochthonous, for the most part carbonate, coarse fraction. This also explains the frequent facies changes from carbonate to clayey marl. The main constituent groups of the coarse fraction are faecal pellets and lumps, the non carbonate mineral components, the Pleistocene relict sediment, the benthonic biogene components and the plankton. Faecal pellets and lumps are formed through grain size transformation of fine sediment. Higher percentages of these components can be correlated to large amounts of fine sediment and organic C. No discernable change takes place in carbonate minerals as a result of digestion and faecal pellet formation. The non-carbonate sand components originate from several unrelated sources and can be distinguished by their different grain size spectrum; as well as by other characteristics. The Iranian rivers supply the greatest amounts (well sorted fine sand). Their quantitative variations can be used to trace fine sediment transport directions. Similar mineral maxima in the sediment of the Gulf of Oman mark the path of the Persian Gulf outflow water. Far out from the coast, the basin bottoms in places contain abundant relict minerals (poorly sorted medium sand) and localized areas of reworked salt dome material (medium sand to gravel). Wind transport produces only a minimal "background value" of mineral components (very fine sand). Biogenic and non-biogenic relict sediments can be placed in separate component groups with the help of several petrographic criteria. Part of the relict sediment (well sorted fine sand) is allochthonous and was derived from the terrigenous sediment of river mouths. The main part (coarse, poorly sorted sediment), however, was derived from the late Pleistocene and forms a quasi-autochthonous cover over wide areas which receive little recent sedimentation. Bioturbation results in a mixing of the relict sediment with the overlying younger sediment. Resulting vertical sediment displacement of more than 2.5 m has been observed. This vertical mixing of relict sediment is also partially responsible for the present day grain size anomalies (coarse sediment in deep water) found in the Persian Gulf. The mainly aragonitic components forming the relict sediment show a finely subdivided facies pattern reflecting the paleogeography of carbonate tidal flats dating from the post Pleistocene transgression. Standstill periods are reflected at 110 -125m (shelf break), 64-61 m and 53-41 m (e.g. coare grained quartz and oolite concentrations), and at 25-30m. Comparing these depths to similar occurrences on other shelf regions (e. g. Timor Sea) leads to the conclusion that at this time minimal tectonic activity was taking place in the Persian Gulf. The Pleistocene climate, as evidenced by the absence of Iranian river sediment, was probably drier than the present day Persian Gulf climate. Foremost among the benthonic biogene components are the foraminifera and mollusks. When a ratio is set up between the two, it can be seen that each group is very sensitive to bottom type, i.e., the production of benthonic mollusca increases when a stable (hard) bottom is present whereas the foraminifera favour a soft bottom. In this way, regardless of the grain size, areas with high and low rates of recent sedimentation can be sharply defined. The almost complete absence of mollusks in water deeper than 200 to 300 m gives a rough sedimentologic water depth indicator. The sum of the benthonic foraminifera and mollusca was used as a relative constant reference value for the investigation of many other sediment components. The ratio between arenaceous foraminifera and those with carbonate shells shows a direct relationship to the amount of coarse grained material in the sediment as the frequence of arenaceous foraminifera depends heavily on the availability of sand grains. The nearness of "open" coasts (Iranian river mouths) is directly reflected in the high percentage of plant remains, and indirectly by the increased numbers of ostracods and vertebrates. Plant fragments do not reach their ultimate point of deposition in a free swimming state, but are transported along with the remainder of the terrigenous fine sediment. The echinoderms (mainly echinoids in the West Basin and ophiuroids in the Central Basin) attain their maximum development at the greatest depth reached by the action of the largest waves. This depth varies, depending on the exposure of the slope to the waves, between 12 to 14 and 30 to 35 m. Corals and bryozoans have proved to be good indicators of stable unchanging bottom conditions. Although bryozoans and alcyonarian spiculae are independent of water depth, scleractinians thrive only above 25 to 30 m. The beginning of recent reef growth (restricted by low winter temperatures) was seen only in one single area - on a shoal under 16 m of water. The coarse plankton fraction was studied primarily through the use of a plankton-benthos ratio. The increase in planktonic foraminifera with increasing water depth is here heavily masked by the "Adjacent sea effect" of the Persian Gulf: for the most part the foraminifera have drifted in from the Gulf of Oman. In contrast, the planktonic mollusks are able to colonize the entire Persian Gulf water body. Their amount in the plankton-benthos ratio always increases with water depth and thereby gives a reliable picture of local water depth variations. This holds true to a depth of around 400 m (corresponding to 80-90 % plankton). This water depth effect can be removed by graphical analysis, allowing the percentage of planktonic mollusks per total sample to be used as a reference base for relative sedimentation rate (sedimentation index). These values vary between 1 and > 1000 and thereby agree well with all the other lines of evidence. The "pteropod ooze" facies is then markedly dependent on the sedimentation rate and can theoretically develop at any depth greater than 65 m (proven at 80 m). It should certainly no longer be thought of as "deep sea" sediment. Based on the component distribution diagrams, grain size and carbonate content, the sediments of the Persian Gulf and the Gulf of Oman can be grouped into 5 provisional facies divisions (Chapt.19). Particularly noteworthy among these are first, the fine grained clayey marl facies occupying the 9 narrow outflow areas of rivers, and second, the coarse grained, high-carbonate marl facies rich in relict sediment which covers wide sediment-poor areas of the basin bottoms. Sediment transport is for the most part restricted to grain sizes < 150 µ and in shallow water is largely coast-parallel due to wave action at times supplemented by tidal currents. Below the wave base gravity transport prevails. The only current capable of moving sediment is the Persian Gulf outflow water in the Gulf of Oman.

Grain size distribution and percentiles of the sand component from sediments obtained in the southern Great Belt, Baltic Sea

Im Sedimentationsraum der südwestlichen Ostsee verdient der nordöstliche Teil der Kieler Bucht besonderes Interesse. Dort öffnet sich die wichtigste Verbindung zwischen Ostsee und Nordsee. Von den Austauschvorgängen, durch welche diese Meeresräume aufeinander Einfluß nehmen, ist gerade jenes Gebiet entscheidend betroffen. Die Beobachtung der Dynamik des Austausches, die Beobachtung der Transportlast, welche von den Wassermassen bewegt wird, und schließlich auch die Beobachtung der Beziehungen, welche sich zwischen dem Zusammentreffen von Wassermassen unterschiedlicher physikalischer Eigenschaften und der Sedimentbildung ergeben, läßt deshalb vor allem dort wesentliche Hinweise zum Verständnis der Sedimentationsvorgänge in der südlichen Ostsee erhoffen. In der vorliegenden Arbeit wurden an 49 Durchschnittsproben die Korngrößenverteilungen und Schwermineralgehalte von Sedimenten aus dem Südausgang des Großen Beltes untersucht. 1. Es wurden sechs in sich morphologisch etwa gleichwertige Gebiete ausgegliedert, die jeweils durch Sedimente mit ähnlichen Korngrößenverteilungen ausgezeichnet sind. Nach Lage, Typ und genetischer Ausdeutbarkeit fügen sich diese Gebiete dem von O. PRATJE (1939, 1948) gegebenen Modell der Sedimentationszonen gut ein. 2. Innerhalb dieser Gebiete ergibt sich für Sande in mehr als 20 m Wassertiefe südwärts gerichteter Transport. Oberhalb dieser Tiefe läßt sich stellenweise nordwärts gerichteter Transport nachweisen. 3. Der Schwermineralgehalt der Sedimente bleibt immer unter zwei Prozent. Die höchsten Anteile (1,7 bzw. 1,9%) werden in den Sedimenten der Tiefen Rinne und der ufernahen Bereiche des Großen Beltes angetroffen. 4. Die Korngrößenverteilungen der Sedimente werden nach der Lage der Modi in bis zu drei (Kies-, Sand-, Silt-) Komponenten zerlegt. Die Beteiligung der Silt-Komponente wird entscheidend von der Salzgehaltssprungschicht beeinflußt. 5. Es bestehen offensichtlich Zusammenhänge zwischen der Schlicksedimentation und der Salzgehaltsschichtung auch in der weiteren südlichen Ostsee.

Holocene sediment transport from the glaciated margin of East/Northeast Greenland to the N Iceland shelves

We examine variations in the ice-rafted sources for sediments in the Iceland/East Greenland offshore marine archives by utilizing a sediment unmixing model and link the results to a coupled iceberg-ocean model. Surface samples from around Iceland and along the E/NE Greenland shelf are used to define potential sediment sources, and these are examined within the context of the down-core variations in mineralogy in the <2 mm sediment fraction from a transect of cores across Denmark Strait. A sediment unmixing model is used to estimate the fraction of sediment <2 mm off NW and N Iceland exported across Denmark Strait; this averaged between 10 and 20%. Both the sediment unmixing model and the coupled iceberg-ocean model are consistent in finding that the fraction of "far-travelled" sediments in the Denmark Strait environs is overwhelmingly of local, mid-East Greenland, provenance, and therefore with a significant cross-channel component to their travel. The Holocene record of ice-rafted sediments denotes a three-part division of the Holocene in terms of iceberg sediment transport with a notable increase in the process starting ca 4000 cal yr BP. This latter increase may represent the re-advance during the Neoglacial period of land-terminating glaciers on the Geikie Plateau to become marine-terminating. The contrast in spectral signals between these cores and the 1500-yr cycle at VM28-14, just south of the Denmark Strait, combined with the coupled iceberg-model results, leads us to speculate that the signal at VM28-14 reflects pulses in overflow waters, rather than an ice-rafted signal.

Area-averaged mean seasonal cycles of chlorophyll-a concentration, sea surface temperature, alongshore wind stress, westward wind speed and dust component of the aerosol optical depth at 550 nm

Nutrient supply in the area off Northwest Africa is mainly regulated by two processes, coastal upwelling and deposition of Saharan dust. In the present study, both processes were analyzed and evaluated by different methods, including cross-correlation, multiple correlation, and event statistics, using remotely sensed proxies of the period from 2000 to 2008 to investigate their influence on the marine environment. The remotely sensed chlorophyll-a concentration was used as a proxy for the phytoplankton biomass stimulated by nutrient supply into the euphotic zone from deeper water layers and from the atmosphere. Satellite-derived alongshore wind stress and sea-surface temperature were applied as proxies for the strength and reflection of coastal upwelling processes. The westward wind and the dust component of the aerosol optical depth describe the transport direction of atmospheric dust and the atmospheric dust column load. Alongshore wind stress and induced upwelling processes were most significantly responsible for the surface chlorophyll-a variability, accounting for about 24% of the total variance, mainly in the winter and spring due to the strong north-easterly trade winds. The remotely sensed proxies allowed determination of time lags between biological response and its forcing processes. A delay of up to 16 days in the surface chlorophyll-a concentration due to the alongshore wind stress was determined in the northern winter and spring. Although input of atmospheric iron by dust storms can stimulate new phytoplankton production in the study area, only 5% of the surface chlorophyll-a variability could be ascribed to the dust component in the aerosol optical depth. All strong desert storms were identified by an event statistics in the time period from 2000 to 2008. The 57 strong storms were studied in relation to their biological response. Six events were clearly detected in which an increase of chlorophyll-a was caused by Saharan dust input and not by coastal upwelling processes. Time lags of <8 days, 8 days, and 16 days were determined. An increase in surface chlorophyll-a concentration of up to 2.4 mg m**3 after dust storms in which the dust component of the aerosol optical depth was up to 0.9 was observed.

Sediment facies of the Indian-Pakistan continental margin, Indian Ocean Expedition

The studies described here base mainly on sedimentary material collected during the "Indian Ocean Expedition" of the German research vessel "Meteor" in the region of the Indian-Pakistan continental margin in February and March 1965. Moreover,samples from the mouth of the Indus-River were available, which were collected by the Pakistan fishing vessel "Machhera" in March 1965. Altogether, the following quantities of sedimentary material were collected: 59.73 m piston cores. 54.52 m gravity cores. 33 box grab samples. 68 bottom grab samples Component analyses of the coarse fraction were made of these samples and the sedimentary fabric was examined. Moreover, the CaCO3 and Corg contents were discussed. From these investigations the following history of sedimentation can be derived: Recent sedimentation on the shelf is mainly characterized by hydrodynamic processes and terrigenous supply of material. In the shallow water wave action and currents running parallel to the coast, imply a repeated reworking which induces a sorting of the grains and layering of the sediments as well as a lack of bioturbation. The sedimentation rate is very high here. From the coast-line down to appr. 50 m the sediment becomes progressively finer, the conditions of deposition become less turbulent. On the outer shelf the sediment is again considerably coarser. It contains many relicts of planktonic organisms and it shows traces of burrowing. Indications for redeposition are nearly missing, a considerable part of the fine fraction of the sediments is, however, whirled up and carried away. In wide areas of the outer shelf this stirring has gained such a degree that recent deposits are nearly completely missing. Here, coarse relict sands rich in ooids are exposed, which were formed in very shallow stirred water during the time when the sea reached its lowest level, i.e. at the turn of the Pleistocene to the Holocene. Below the relict sand white, very fine-grained aragonite mud was found at one location (core 228). This aragonite mud was obviously deposited in very calm water of some greater depth, possibly behind a reef barrier. Biochemic carbonate precipitation played an important part in the formation of relict sands and aragonite muds. In postglacial times the relict sands were exposed for long periods to violent wave action and to areal erosion. In the present days they are gradually covered by recent sediments proceeding from the sides. On the continental margin beyond the shelf edge the distribution of the sediments is to a considerable extent determined by the morphology of the sea bottom. The material originating from the continent and/or the shelf, is less transported by action of the water than by the force of gravity. Within the range of the uppermost part of the continental slope recent sedimentation reaches its maximum. Here the fine material is deposited which has been whirled up in the zone of the relict sands. A laminated fine-grained sediment is formed here due to the very high sedimentation rate as well as to the extremely low O2-content in the bottom water, which prevents life on the bottom of the sea and impedes thus also bioturbation. The lamination probaly reflects annual variation in deposition and can be attributed to the rhythm of the monsoon with its effects on the water and the weather conditions. In the lower part of the upper continental slope sediments are to be found which show in varying intensity, intercalations of fine material (silt) from the shelf, in large sections of the core. These fine intercalations of allochthonous material are closely related to the autochthonous normal sediment, so that a great number of small individual depositional processes can be inferred. In general the intercalations are missing in the uppermost part of the cores; in the lower part they can be met in different quantities, and they reach their maximum frequency in the upper part of the lower core section. The depositions described here were designated as turbid layer sediments, since they get their material from turbid layers, which transport components to the continental slope which have been whirled up from the shelf. Turbidites are missing in this zone. Since the whole upper continental slope shows a low oxygen-content of the bottom water the structure of the turbid layer sediments is more or less preserved. The lenticular-phacoidal fine structure does, however, not reflect annual rhythms, but sporadic individual events, as e.g. tsunamis. At the lower part of the continental slope and on the continental rise the majority of turbidites was deposited, which, during glacial times and particularly at the beginning of the post-glacial period, transported material from the zone of relict sands. The Laccadive Ridge represented a natural obstacle for the transport of suspended sediments into the deep sea. Core SIC-181 from the Arabian Basin shows some intercalations of turbidites; their material, however, does not originate from the Indian Shelf, but from the Laccadive Ridge. Within the range of the Indus Cone it is surprising that distinct turbidites are nearly completely missing; on the other hand, turbid layer sediments are to be found. The bottom of the sea is showing still a slight slope here, so that the turbidites funneled through the Canyon of the Swatch probably rush down to greater water depths. Due to the particularly large supply of suspended material by theIndus River the turbid layer sediments show farther extension than in other regions. In general the terrigenous components are concentrated on the Indus Cone. It is within the range of the lower continental slope that the only discovery of a sliding mass (core 186) has been located. It can be assumed that this was set in motion during the Holocene. During the period of time discussed here the following development of kind and intensity of the deposition of allochthonous material can be observed on the Indian-Pakistan continental margin: At the time of the lowest sea level the shelf was only very narrow, and the zone in which bottom currents were able to stir up material by oscillating motion, was considerably confined. The rivers flowed into the sea near to the edge of the shelf. For this reason the percentage of terrigenous material, quartz and mica is higher in the lower part of many cores (e.g. cores 210 and 219) than in the upper part. The transition from glacial to postglacial times caused a series of environmental changes. Among them the rise of the sea level (in the area of investigation appr. 150 m) had the most important influence on the sedimentation process. In connection with this event many river valleys became canyons, which sucked sedimentary material away from the shelf and transported it in form of turbidites into the deep sea. During the rise of the sea level a situation can be expected with a maximum area of the comparatively plane shelf being exposed to wave action. During this time the process of stirring up of sediments and formation of turbid layers will reach a maximum. Accordingly, the formation of turbidites and turbid layer sediments are most frequent at the same time. This happened in general in the older polstglacial period. The present day high water level results in a reduced supply of sediments into the canyons. The stirring up of sediments from the shelf by wave action is restricted to the finest material. The missing of shelf material in the uppermost core sections can thus be explained. The laminated muds reflect these calm sedimentation conditions as well. In the southwestern part of the area of investigation fine volcanic glass was blown in during the Pleistocene, probably from the southeast. It has thus become possible to correlate the cores 181, 182, 202. Eolian dust from the Indian subcontinent represents probably an important component of the deep sea sediments. The chemism of the bottom as well as of the pore water has a considerable influence on the development of the sediments. Of particular importance in this connection is a layer with a minimum content of oxygen in the sea water (200-1500 m), which today touches the upper part of the continental slope. Above and beyond this oxygen minimum layer somewhat higher O2-values are to be observed at the sea bottom. During the Pleistocene the oxygen minimum layer has obviously been locatedin greater depth as is indicated by the facies of laminated mud occuring in the lower part of core 219. The type of bioturbation is mainly determined by the chemism. Moreover, the chemism is responsible for a considerable selective dissolution, either complete or partial, of the sedimentary components. Within the range of the oxygen minimum layer an alkaline milieu is developed at the bottom. This causes a complete or partial dissolution of the siliceous organisms. Here, bioturbation is in general completely missing; sometimes small pyrite-filled burrowing racks are found. In the areas rich in O2 high pH-values result in a partial dissolution of the calcareous shells. Large, non-pyritized burrowing tracks characterize the type of bioturbation in this environment. A study of the "lebensspuren" in the cores supports the assumption that, particularly within the region of the Laccadive Basin, the oxygen content in the bottom sediments was lower than during the Holocene. This may be attributed to a high sedimentation rate and to a lower O2-content of the bottom water. The composition of the allochthonous sedimentary components, detritus and/or volcanic glass may locally change the chemism to a considerable extent for a certain time; under such special circumstances the type of bioturbation and the state of preservation of the components may be different from those of the normal sediment.

Mineralogy and isotope chemistry of GISP2 and potential source areas

Samples of dust from the Greenland Ice Sheet Project 2 (GISP2) ice core, Summit, Greenland, dated within marine isotope stage 2 (between 23,340 and 26,180 calendar years B.P.) around the time of the coldest, local, last glacial temperatures, have been analyzed to determine their provenance. To accomplish this, we have compared them with approximately Coeval aeolian sediments (mostly loesses) sampled in possible source areas (PSAs) from around the northern hemisphere. The <5-µm grain-size fraction of these samples was analyzed on the basis that it corresponds to the atmospheric dust component of that time and locale, which was sufficiently fine grained to be transported over long distances. On the basis of comparison of the clay mineralogy and Sr, Nd and Pb isotope composition with ice dust and PSAs and assuming that we have sampled the most important PSAs, we have determined that the probable source area of these GISP2 dusts was in eastern Asia. The dust was not derived from either the midcontinental United States or the Sahara, two more proximal areas that have been suggested as potential sources based on atmospheric circulation modeling. Except for a brief period during an interstadial, when dust transport was exceptionally low (for glacial times) and had a mineralogical composition indicative of a slightly more southern provenance, the source area of the dust did not change significantly during times of variably higher fluxes of dust with larger mean grain size or lower fluxes of dust with smaller mean grain size. This includes the high-dust period that correlates with the Heinrich 2 period of major iceberg discharge into the North Atlantic. Variable wind strengths must therefore be invoked to account for these abrupt and significant changes in dust flux and grain size.

Component NRM directions for sediment core HLY0503-06JPC

The Arctic oceans have been fertile ground for the recording of apparent excursions of the geomagnetic field, implying that the high latitude field had unusual characteristics at least over the last 1-2 Myrs. Alternating field demagnetization of the natural remanent magnetization (NRM) of Core HLY0503-6JPC from the Mendeleev Ridge (Arctic Ocean) implies the presence of primary magnetizations with negative inclination apparently recording excursions in sediments deposited during the Brunhes Chron. Thermal demagnetization, on the other hand, indicates the presence of multiple (often anti-parallel) magnetization components with negative inclination components having blocking temperatures predominantly, but not entirely, below ~ 350 °C. Thermo-magnetic tests, X-ray diffraction (XRD) and scanning electron microscopy (SEM) indicate that the negative inclination components are carried by titanomaghemite, presumably formed by seafloor oxidation of titanomagnetite. The titanomaghemite apparently carries a chemical remanent magnetization (CRM) that is partially self-reversed relative to the detrital remanent magnetization (DRM) carried by the host titanomagnetite. The partial self-reversal could have been accomplished by ionic ordering during oxidation, thereby changing the balance of the magnetic moments in the ferrimagnetic sublattices.

Radionuclides of bulk and terrigenous surface sediments in the Southern Ocean

Sr and Nd isotopic compositions of Late Quaternary surface sediment and sediment cores from the south Atlantic and southeast Pacific sectors of the Southern Ocean are used to constrain the provenance and transport mechanisms of their terrigenous component. We report isotopic and mineralogical data for core samples from three localities, the Mid-Atlantic Ridge at 41°S and the northern and southern Scotia Sea. In addition, data for surface sediment samples from the south Atlantic and southeast Pacific sectors of the Southern Ocean are presented. The variations of Sr and Nd isotopic compositions of the bulk sediment samples in all cores were correlated with the magnetic susceptibility of the sediment and with the inferred glacial-interglacial stages. The isotopic data indicate that, during glacial periods, sediment was delivered from continental crust with a shorter residence time than that supplying material during interglacial periods. At the core site near the Mid-Atlantic Ridge, Nd isotopic, combined with mineralogical evidence indicates interglacial period deposition of a relatively high amount of kaolinite and silt with low epsilon-Nd values < -8. The material was probably supplied by North Atlantic Deep Water from low latitudes. For glacial periods, a high contribution of silt and clay with epsilon-Nd > -4.5, probably derived from southern South America, was indicated. The glacial-interglacial shift in sources may be due to either a decreasing influence of North Atlantic Deep Water during glacial times or by a larger contribution of glaciogenic detritus from southern South America. At the core site in the northern Scotia Sea, sediment of interglacial periods is dominated by smectite with epsilon-Nd < - 6 and silt with epsilon-Nd > -4. We suggest that smectite was derived from the Falkland shelf and silt was derived from the Argentinian shelf. During glacial periods, the Argentinian shelf was an important source for silt and chlorite with epsilon-Nd > -4. The contribution from the Falkland shelf seems to have remained similar during glacial and interglacial periods. Hydrographic transport by bottom currents and turbidites could account for the high glacial detrital flux. An evaluation of the significance of an aeolian contribution to deep sea sediment suggests that it plays only a minor role. In the southern Scotia Sea, the Antarctic Peninsula is considered an important source for young material with epsilon-Nd > -4, in particular during glacial periods. During interglacial periods, sediment supply from the Antarctic Peninsula was lower than during glacial times, resulting in a relatively high contribution of old material (epsilon-Nd < -8) from East Antarctica. Deep water currents and icebergs could account for the transport of the old component to the southern Scotia Sea. The accumulation rates of material from the various source regions for glacial times are in agreement with an increase in the strength of the Antarctic Circumpolar Current. The production rate and the circulation pattern of bottom water in the Weddell Sea appear to have remained similar over most of the last 150 kyr.