Search

Vertical water mass structure in the Polar North Atlantic Ocean plays a critical role in planetary climate by influencing the formation rate of North Atlantic deepwater, which in turn affects surface heat transfer in the northern hemisphere, ventilation of the deep sea, and ocean circulation on a global scale. However, the response of upper stratification in the Nordic seas to near-future hydrologic forcing, as surface water warms and freshens due to global temperature rise and Greenland ice demise, remains poorly known. While past major interglacials are viewed as potential analogues of the present, recent findings suggest that very different surface ocean conditions prevailed in the Polar North Atlantic during Marine Isotope Stage (MIS) 5e and 11 compared to the Holocene. It is thus crucial to identify the causes of those differences in order to understand their role in climatic and oceanographic variability. To resolve this, we pair here bulk sediment δ15N isotopic signatures with planktonic foraminiferal assemblages and their isotopic composition across major past interglacials. The comparison defines for the first time stratification-induced variations in nitrate utilization up to 25% between and within all of these warm periods that highlight changes in the thickness of the mixed-layer throughout the previous interglacials. That thickness directly controls the depth-level of Atlantic water inflow. The major changes of nitrate utilization recorded here thus suggest that a thicker mixed-layer prevailed during past interglacials, probably related to longer freshwater input associated with the preceding glacial termination. This would have caused the Atlantic water to flow at greater depth during MIS 5e and 11. These results call for caution when using older interglacials as modern or near-future climate analogues and contribute to the improvement of our general comprehension of the impact of freshwater input near a globally important deep-water formation site like the Nordic Seas. This is crucial when assessing the negative impacts on the Greenland Ice Sheet of climate change and global warming.

Continental margin sediments are commonly studied using trace metal enrichments as proxies for characterizing their modern redox conditions and tracking past changes in bottom water ventilation and marine primary productivity. Currently little is known about the sedimentary redox history of the continental shelf west of the Antarctic Peninsula. Yet, environmental conditions in the Antarctic Peninsula region have changed rapidly in recent decades in response to global warming. Paleoclimate archives including bulk sediment trace metal enrichments can provide insight into both past and future environmental changes in this climatically sensitive region, and characterization of redox conditions in surface (modern) sediments is essential for establishing a framework for paleoredox interpretations. In this study we measured concentrations of trace metals (Ag, Cd, Re, and Mo) and productivity proxies (total Ba, organic carbon, and biogenic silica) in surface sediments from the Marguerite Bay, Gerlache Strait, and Bransfield Strait areas. Proxy concentrations suggest that sediments are generally suboxic due to seasonally high export of organic carbon from surface waters. Comparison of West Antarctic Peninsula trace metal/Al ratios to trace metal data from continental margins outside the Antarctic region demonstrates that although the West Antarctic Peninsula water column differs from other locations (e.g., it lacks a strong oxygen minimum zone), sedimentary redox chemistry is similarly controlled by organic matter decomposition, through slow recycling of highly seasonal export production falling through a cold water column.

A high-resolution sedimentary record of dinoflagellate cysts from Ocean Drilling Program Hole 1017E (off Point Conception, California margin) reflects how marine primary productivity has changed in response to major shifts in climate and ocean circulation along the California margin over the past 42 kyr. Throughout the studied sequence, dinoflagellate cyst assemblages are dominated by upwelling-related taxa, signifying the continued presence of coastal upwelling on the margin during the late Quaternary. The cyst record suggests that marine primary productivity was enhanced during the Holocene and Bolling, and to a lesser extent, during the late glacial and most Dansgaard-Oeschger events, while an apparent reduction in primary productivity can be seen during the Younger Dryas. The best analogue technique, based on a modern dinoflagellate cyst assemblage database from the northeast Pacific, was used for quantitative reconstruction of past sea surface conditions. It points to dynamic changes in annual marine primary productivity (~235–331 g C m−2 yr−1) and sea surface temperature (~10.1–12.6°C in winter; ~13.1–14.3°C in summer), while sea surface salinity appears to be confined to a narrower range (~32.9–33.4 in summer). Our results also indicate noticeable climate variability during the Holocene in this region.

A high-resolution, 55 kyr long record of chalcophile and redox-sensitive trace element accumulation (Ag, Cd, Re, and Mo) from MD02-2515, western Guaymas Basin, is investigated in conjunction with patterns in stratigraphy and productivity. High opal concentrations (~59 wt. %), representing increased diatom production, coincide with laminated sediments and dilute the concentrations of organic carbon (Corg) and metals. A similarity between opal and normalized Corg, Ag, and Cd concentrations suggests delivery to the sediments by diatom export production, while patterns in normalized Re and Mo accumulation suggest a different emplacement mechanism. Although Mo enrichment in organic-rich, laminated sediments typically represents anoxic conditions at other locations, Mo (and Re) in Guaymas Basin is enriched in nonlaminated and bioturbated sediments that are representative of oxygenated conditions. Adsorption onto Fe- and/or Mn-oxyhydroxide surfaces during oxygenation inadequately explains both the Re and Mo enrichments. Thus, recently published mechanisms invoking direct Re and Mo removal from the water column and bioturbation-assisted irrigation of Re into the sediments are used to explain the counterintuitive observations in Guaymas Basin. The MD02-2515 stratigraphic and proxy records are also different from other records in the northeast Pacific in that there is little correspondence with Greenland Dansgaard-Oeschger interstadials. There is some correlation with Heinrich events, suggesting that ventilation of intermediate waters and/or reduced productivity may be important in controlling stratigraphy and trace element accumulation. The results question whether MD02-2515 records can be compared to northeast Pacific open-margin records, especially before 17 kyr B.P.

Deglacial sediments in piston core JT96-02 from the continental slope off western Canada contain numerous centimeter-thick, iron sulfide-rich layers. Major and minor element chemistry, as well as C and N isotopic data suggest that these layers represent background hemipelagic sedimentation. Intervening gray silty clay layers, each of which exhibits a fining-upward trend, are distal turbidites. The iron sulfides precipitated in response to the emplacement of each turbidite, which slowed oxygen diffusion into the sediment allowing anoxic conditions to develop in the underlying relatively organic-rich deposits. Resulting sulfate reduction and precipitation of Fe-sulfide fostered Mo enrichment (> 2 μg/g). It may also explain Ni, Cr, V, and Co enrichments, although differences in detrital mineralogy cannot be ruled out, particularly for Ni and Cr. Rhenium is also enriched, albeit slightly, within the sulfide layers, but low Re/Mo ratios (~ 0.78 ng/g/μg/g) suggest that anoxic conditions developed very rapidly. Metals with high lithogenic concentrations (Cd and U) are not obviously enriched in the sulfide layers because there was insufficient time for noticeable authigenic accumulation before deposition of the next turbidite which severed diffusive communication with overlying seawater, the primary source of metals. Silver is also slightly higher in the sulfide layers, possibly linked to higher marine organic matter content. In summary, the pulsed character of sedimentation during the deglaciation led to episodic enrichment of Mo and occurred independently of changes in bottom water oxygen or organic carbon content. This observation potentially complicates the use of Mo as a paleo-proxy in some continental margin environments.

Accurate interpretation of paleoproductivity proxies in the Southern Ocean requires “ground truthing” these proxies. Here the concentrations of total Ba, organic carbon (C org ), and biogenic silica (Si bio ), along with other redox-sensitive trace metals (Cd, Re, and Mo), are compared with the distribution of silver (Ag) in core top sediments from the Pacific sector of the Southern Ocean. Conventional thinking holds that enrichment of trace metals should be muted or absent in Southern Ocean sediments despite high export of C org from surface waters because deposition under well-oxygenated bottom waters inhibits accumulation of reduced metal phases. While our data support this prediction, detectable preserved vertical fluxes of Ag, Cd, and Re exist beneath the opal belt and faithfully record surface productivity patterns. Therefore we suggest Ag fluxes into Southern Ocean sediments occur via biodetritus associated with high export production rather than accumulation by redox-driven diagenetic processes. Sedimentary Ag concentrations are likely altered at the seafloor by biogenic particle remineralization due to high bottom water O 2 . Nevertheless, zonal Ag fluxes are similar to surface productivity patterns, highlighting the potential of Ag to serve as a proxy for diatom export production.

Moorings with sequential sediment traps to study downward sediment flux in the Canada Basin of the Arctic Ocean were maintained in 3350 m of water year-round between September 1990 and July 1994. Sediment was collected nominally at 600-m depth and instruments measuring current, temperature, salinity and pressure were placed at several levels between 45 and 1515 m. Total dry weight particle fluxes were low (4.2, 2.1, and 8.2 g m −2 a −1 ) compared to those found in other world ocean basins. Particle flux varied greatly intra-annually. There were peaks during each summer, with differing seasonal timing and particle composition suggesting a correlation with inter-annual differences in summertime ice clearance. In winter, the particle flux was higher if the ice of the preceding summer was light. Enhanced primary production in summers with wider ice-free seas is a possible explanation, but inconsistent with the high lithogenic (LITH) content of most samples: A significant fraction of the particulate organic carbon (POC) content is refractory carbon. Another possibility is that the south-easterly wind pushing ice to the north-west in summers of reduced ice drives the Mackenzie River plume, laden with lithogenic sediment, behind it and out to the mooring site. There is also another factor in play: Peaks in particle flux frequently coincided with eddies, mesoscale circulation features with strongest current well below the surface. The highest measured flux occurred in the winter of 1994, synchronous with a baroclinic eddy which enveloped the mooring for 60 days while rotating cyclonically and moving slowly north-westward. Particles trapped at this time had high LITH and low POC contents, a relatively high molar ratio of biogenic silica to POC and appreciable chlorophyll a and phaeophytin pigments. Subsequent trap intervals coincided with the passage of a deeper anti-cyclonic eddy, also on a north-westward trajectory, which deposited material of different composition. These observations demonstrate that an understanding of the lateral transport of sediment by energetic physical phenomena is critical to the insightful interpretation of the particle flux measured with sediment traps at sites remote from the coast in the southern Canada Basin.

Five vertical profiles of silver (Ag) in the subarctic northeast Pacific are presented. Dissolved (

Determining climate variations over the Holocene requires high-resolution records with well-developed age models. A 40 m long marine sediment core raised from Effingham Inlet, an anoxic fjord on the west coast of Vancouver Island, British Columbia, Canada, yields such a record. Forty six14C accelerator mass spectrometry (AMS) dates determined from terrestrial plant material form the age model. Downcore sampling at both 5 cm (20 year) and 1.5 cm (7 year) resolution indicates that high-frequency oceanographic variability has prevailed at this site over the last 10 000 years. Spectral analysis of wt.% opal, a proxy for diatom productivity in the basin, reveals the bidecadal and pentadecadal periods of the Pacific decadal oscillation (PDO) – North Pacific index (NPI) that are related to changes in the strength of the Aleutian Low. Coherence analysis between the Effingham Inlet data and δ18O records from Jellybean Lake (a high elevation site in southwest Yukon) indicates regional coherence at periods of 45, 70, and 510 years between productivity in Effingham Inlet and changes in the Aleutian Low strength. Over the entire Holocene, the strength of decadal variability has changed. Both 20- and 50-year periods are present to some degree in the early Holocene, and only the 50 year period is evident in the late Holocene. These data imply that regime shifts would have been more frequent in the early Holocene relative to the last several thousand years.

Multiproxy analysis of a 38 m long sediment core recovered from the Vancouver Island margin (48.97°N, 127.04°W, water depth 1243 m) has yielded a millennial-scale history of upwelling and export production over the last ∼50 ka. Higher concentrations of marine organic carbon, opal, and trace Mo suggest that production was higher, and sedimentary pore waters more anoxic, during the warm Holocene, Bølling–Allerød, and interstadial events between 31 and 44 ka BP. Relatively lower production and higher inputs of terrigenous organic matter occurred during the last glacial (14.7–31 ka BP; Cordilleran ice sheet proximal to coring site at ∼19.5 ka BP) and from 44–50.4 ka BP. Enrichments in sedimentary δ15N during interstadial events are interpreted to reflect episodic delivery and upwelling of isotopically heavy nitrate to the surface waters and subsequent vectoring to the seafloor via settling planktonic detritus. Similar patterns are seen in southern California and other areas along the western margin of North America, implying that heavier nitrate generated by denitrification in the Eastern Tropical North Pacific has in the past been carried northward in the California Undercurrent at least as far as central Vancouver Island. This inference is consistent with modern hydrographic observations in the region. Comparison of the coherent Vancouver Island, Oregon, California, and northwest Mexico margin records with late Pleistocene climate history in Greenland reinforces the conclusion that a tight physical and biogeochemical coupling has existed for at least 50 ka between the North Atlantic and North American margin waters, including those off Vancouver Island.

A 40.32 m piston core recovered from Effingham Inlet, on the west coast of Vancouver Island, provides the basis for a high-resolution geochemical study of the last deglaciation and the Holocene. Glacial retreat, basin isolation, sea-level rise, and productivity variations are determined using proxies for sediment composition (K/Al, Fe/Al, Mg/Al), grain size (Ti/Al, Zr/Al), sedimentary redox conditions (Mo/Al, U/Al), and productivity (wt.% organic carbon, wt.% opal). As local ice retreated and marine waters inundated the basin, coarse-grained glacimarine sediments were replaced by finer grained, laminated, opal-rich sediments. During meltwater pulse-1a, the dominance of local crustal rise over eustatic sea-level rise resulted in the progressive restriction of ocean circulation in Effingham Inlet and the formation of a temporary freshwater lake. The transition into stable Holocene conditions was initiated at ∼12 700 BP, which corresponds to the onset of the Younger Dryas, as identified by the Greenland Ice core Project (GRIP) ice core δ18O record and was completed by 10 700 BP, ∼800 years after the GRIP ice core record stabilized. Holocene Mo/Al and U/Al ratios range between 12–35 (×104) and 1–3.4 (×104), respectively, indicating that although large-amplitude, high-frequency fluctuations occur, the sediments of Effingham Inlet inner basin have remained organic rich and oxygen depleted for the entire Holocene period. The combination of anoxic bottom waters and a Holocene sedimentation rate of 217 cm/ka have preserved a high-resolution record of environmental change in the northeast Pacific over the last 11 000 years.

Fifty-six surface sediment samples from the northeastern Pacific (43–25°N) were investigated in order to examine the spatial distribution of modern organic-walled dinoflagellate cysts in relation to hydrological conditions and marine productivity. The analyzed dinoflagellate cyst assemblages are diverse, and include over 60 taxa. Multivariate statistical analysis (CCA) of dinoflagellate cysts and environmental variables identifies annual productivity and sea-surface temperature as main factors affecting dinoflagellate cyst distribution in the region. In the studied area, marine productivity is greatly influenced by the strength of the coastal upwelling. Cyst assemblages from the coastal sites associated with active upwelling are characterized by the dominance of heterotrophic taxa, particularly Brigantedinium spp., Echinidinium spp. and cysts of Protoperidinium americanum. Taxa associated with low productivity offshore sites are Impagidinium spp., Nematosphaeropsis labyrinthus, Pyxidinopsis reticulata and Operculodinium centrocarpum. Dinoflagellate cyst species associated with warmer waters are Lingulodinium machaerophorum, Spiniferites mirabilis, S. ramosus, S. delicatus/bulloideus, Bitectatodinium spongium and Polykrikos cf. kofoidii, while those from cooler environments include Selenopemphix nephroides, Trinovantedinium variable and cysts of Pentapharsodinium dalei. Combining the present cyst dataset with other published cyst data from the Northeastern Pacific, we have compiled a “NE Pacific 188” database. This database can be used as a basis for the quantitative reconstructions of (paleo)temperature and productivity in the Northeastern Pacific. © 2008 Elsevier B.V. All rights reserved.

Increasing quantities of atmospheric anthropogenic fixed nitrogen entering the open ocean could account for up to about a third of the ocean's external (nonrecycled) nitrogen supply and up to ∼3% of the annual new marine biological production, ∼0.3 petagram of carbon per year. This input could account for the production of up to ∼1.6 teragrams of nitrous oxide (N 2 O) per year. Although ∼10% of the ocean's drawdown of atmospheric anthropogenic carbon dioxide may result from this atmospheric nitrogen fertilization, leading to a decrease in radiative forcing, up to about two-thirds of this amount may be offset by the increase in N 2 O emissions. The effects of increasing atmospheric nitrogen deposition are expected to continue to grow in the future.

Redox conditions in near-surface sediments from the continental margin off western Canada were characterized to determine how they influence the accumulation of redox-sensitive trace metals. Despite the development of suboxic conditions within millimeters of the sediment–water interface there is little enrichment of Re, U, Cd or Mo in the upper 10 to 15 cm of the sediment column. This observation is consistent with the low extractible solid Mn concentrations but high I/Corg ratios that imply that near-surface sediment redox conditions are poised between Mn and I reduction. In contrast, below ∼15 cm Re, and to a lesser extent Cd and U, are enriched (e.g., up to 65 ng/g Re) but only in those cores collected from within and below the oxygen minimum zone. Molybdenum concentrations remain low (b1.5 μg/g) in all cores indicating that these near-surface sediments never became fully anoxic.

High-precision isotope composition determinations using multicollector, magnetic-sector inductively coupled plasma mass spectrometry (MC-ICPMS) have recently revealed that some transition metal isotopes such as those of Mo, Fe, Cu, Zn etc. can be used as biogeochemical tracers in global change research. The Mo isotope system may be useful in paleoredox investigations indicating that δ97/95Mo in seawater may co-vary with changes in the relative proportions of anoxic and oxic sedimentation in the ocean, and that this variation may be recorded in δ97/95Mo of anoxic sediments. The Mo continental flux into the oceans and the global Mo isotope budget can be estimated from δ97/95Mo values. The Fe isotope composition in seawater is an important issue because Fe plays a controlling role in biological productivity in the oceans and its abundance in seawater may have substantial effect on climate changes. Iron isotope fractionations could result from bio- and abio-processes and have about 0.1% variation (δ56/54Fe), so Fe isotopes considered alone cannot be used to distinguish the products of abiotic and biotic Fe processing in geological records. Cu and Zn isotopes are also used as biogeochemical tracers, but the researches are relatively less. This review mainly focuses on the methods for preparation, purification and determination, of new isotope tracer samples, and on isotope applications in marine environmental changes.

The organic carbon content of marine sediments is commonly used as a proxy for export production. However, in continental margin sediments a large fraction of the organic matter may be of terrestrial origin, and it is necessary to correct the total organic carbon data accordingly. Radiocarbon dating of bulk organic carbon, organic geochemistry and isotope data (δ13C and δ15N) are used here to characterize the type of organic matter present in Core JT96-09 collected at a water depth of 920 m on the slope off Western Canada. The quantities of marine and terrestrial organic carbon are then estimated using the δ13C data. The 16 kyr record obtained from Core JT96-09 suggests that accumulation of total organic carbon was highest during the late glacial and deglacial, but geochemical data indicate that as much as 70% of this carbon is terrestrial in origin. When the palaeo-record is corrected for this terrigenous input it is observed that accumulation of marine organic matter, and presumably marine export production, increased at the end of the last glacial contemporaneous with the Bolling, and that it peaked during the Allerod. Data from other palaeoproductivity proxies (i.e., bio-barium, opal and alkenones) also indicate relatively high productivity during the deglacial. These results indicate a return to modern upwelling conditions and high marine export production at ∼14.3 calendar kyr BP and a period of enhanced upwelling, relative to the present, during the Allerod.

A correlation between warm interstadials implied by δ18Oice in the GISP (Greenland) ice core and anoxic events recorded by the sediments drilled at ODP Site 893 in Santa Barbara Basin (SBB), has been used previously to relate variations in bottom water oxygenation to high-frequency changes in the ventilation of the SBB (Nature, 379 (1996) 243). However, an additional control on the oxygen content at depth is the local settling flux of metabolizable organic matter. To assess the impact of oxygen consumption related to this source, Mo, Cd, Re and U concentration data are used here in conjunction with % COrg and δ15N to deduce past redox conditions, variations in the sedimentary redox boundary depth and the impact of local organic flux on the oxygenation of SBB deep waters. Comparisons between ODP Sites 893, 1017 and 1019 allow the differentiation of regional signals, dominated by ventilation changes, from local signals, dominated by vertical organic flux. It is concluded that prior to the Holocene, local productivity changes in the Santa Barbara Basin are not solely the cause of sedimentary anoxia. Episodic injections of low oxygen intermediate water, identified on the regional scale, strongly influenced the oxygen content of the SBB during this time. However, over the last 10,000 years the influence of local productivity has increased and the regional similarity among the three sites is complicated by local signals.

Detailed examination of the water column, sediments, and interstitial waters was conducted in Balmer Lake, Ontario, Canada, in 1993-1994 and 1999 in order to assess the seasonal and interannual controls governing the behavior of As. High-resolution profiles of dissolved (

Concentration and mass accumulation rate profiles from Southeastern Atlantic sediment cores located off Namibia show that an exceptional episode in benthic carbonate dissolution occurred during early glacial isotope stage 6 (substages 6.6 and 6.5) between about 186 000 and 170 000 yr BP. Although this episode is restricted to or is more pronounced in this region than in other areas of the Atlantic Ocean, its exceptional character with respect to older and younger climatic episodes at the same site cannot be fully explained by local factors alone, but requires a combination of local and global influences. The onset of the carbonate dissolution episode is related to a more efficient transfer of organic matter from surface eutrophic areas to the lower and is due to low sea level, while its termination relates to a change in either global ocean alkalinity or bottom water circulation. An evaluation of the magnitude of this local carbonate dissolution episode suggests that its contribution to a global alkalinity change may have been significant. Carbonate dissolution was probably amplified by stronger upwelling activity of the Benguela System linked to an exceptional northern excursion of the boreal summer ITCZ during early glacial isotope stage 6. This low latitude global linkage may explain how this carbonate dissolution event as well as other ‘anomalies’ observed for early stage 6, like an important Dole effect minimum or a ‘cold’ Mediterranean sapropel, are related.

Changes in palaeoproduction in Saanich Inlet, British Columbia, were examined from sediments using total organic carbon contents and stable carbon isotope ratios. Sediments were obtained from Holes 1033B and 1034B of Ocean Drilling Program Leg 169S. These cores contain two distinct zones: (1) carbon poor (%Corg

Coring at site ODP 1033B in Saanich Inlet recovered 59.4 m of mainly laminated olive-grey diatom ooze and an underlying 55.15 m of massive grey to olive-grey silty clay. Based on AMS radiocarbon dating, the boundary between the two units is between 11,000 and 13,800 calibrated years BP, and represents the Holocene-Pleistocene boundary. The lower unit represents glaciomarine deposition, whereas deposition of the upper unit began when the modern semi-restricted physiography of the fjord was established following glacial rebound and highly productive marine conditions were established. The glaciomarine clay is almost entirely terrigenous, whereas the diatom ooze contains 2–3 wt.% organic C and 20–40 wt.% biogenous silica; CaCO 3 contributions are minor, but there are several peaks in carbonate abundance in the upper unit. The isotopic composition of organic C and total N suggests that organic matter in the glaciomarine clay is dominantly terrestrial ( δ 13 C organic δ 15 N total =ca. 3‰) and in the diatom oozes it is mainly marine ( δ 13 C organic >−22‰ and δ 15 N total =ca. 10‰). The heavy δ 15 N total values probably record a contribution of isotopically heavy nitrate to the surface waters of the inlet that is transported to British Columbia (BC) coastal waters from the eastern tropical Pacific by the California Undercurrent. Major and minor elemental data suggest that the composition of the terrigenous material and its grain-size has changed over the last 15 kyr, and there are marked enrichments in several redox-sensitive elements in the diatom oozes. Thus, Cu, Mn, Mo, Ni, Pb, V and Zn have higher concentrations in the upper unit; Br and I are also enriched because of their association with organic matter. Mn is enriched in the anoxic diatom oozes due to the presence of manganoan carbonate (Mn peaks generally corresponding with carbonate peaks) formed in the sediment when deep water renewal caused precipitation of Mn oxyhydroxides, which dissolved in the anoxic sediment and was precipitated as a diagenetic phase. The remaining metals are enriched because of their removal to the sediment as sulphides (Cu, Mo, Ni, Pb and Zn) or as particle-reactive reduced species (V). Cr enrichment is obscured by the presence of Fe-rich chlorite. The lag in the enrichment of Mo with respect to organic C in the sediments indicates that anoxia developed some time after marine production increased following the semi-isolation of the fjord.

A palynological investigation was undertaken on the upper 29 m of sediment at Ocean Drilling Program (ODP) Site 887, spanning the last 430 000 years (i.e., isotopic stages 12 to 1). Pollen and dinocyst assemblages reveal a major ecostratigraphical boundary at the Middle–Late Pleistocene transition. The Middle Pleistocene pollen data document the occurrence of a spruce forest vegetation in the source area, likely located on the adjacent Alaskan coast, whereas the Late Pleistocene is marked by higher inputs of pine, shrub, and herb taxa, suggesting predominant inputs from a more open landscape. The Middle Pleistocene is characterized by a low diversity in dinocyst assemblages, which are dominated by Operculodinium centrocarpum, whereas the Late Pleistocene is marked by the significant occurrence of Pentapharsodinium dalei, Pyxidinopsis reticulata, and by high percentages of Brigantedinium spp. Such assemblages suggest open oceanic and cool temperate conditions during the Middle Pleistocene, changing toward generally colder and less saline conditions during the Late Pleistocene. In addition, large fluctuations in the dinocyst assemblages during the Late Pleistocene are recorded in phase with the main shifts in the isotopic stratigraphy. A new dinocyst taxon, Spiniferites alaskensis sp. nov., exclusively recorded in sediments of the isotopic substage 5e, is described herein.

Regularly increasing radiocarbon age–depth profiles and near-constant sediment composition with depth have demonstrated that sediment accumulation has been relatively constant during the late Holocene at two north-east Atlantic sites on Rockall Bank and Feni Drift (UK Benthic Boundary Layer Experiment sites). In such a quasi-steady state situation, the geochemical responses to early diagenesis of the redox-sensitive elements Cd, Mn, Mo, Re, Se and U are readily discerned through changes in element concentrations with depth. Collectively, a colour change in the sediments, surficial MnOx-enriched layers and 210Pbexcess profiles allow an estimate of the mean position of the oxic–post-oxic boundary in the sediments at both sites. This boundary is situated deeper than the 210Pbexcess surface mixed layer but shallower than the 14C surface mixed layer which represent mixing on 102 and 103 year time scales, respectively. This implies that the well-defined surface MnOx enrichment and its associated 226Ra and Mo fractions must have been efficiently recycled first by downwards mixing into anoxic conditions, and then by subsequent reduction and diffusive migration back up into oxic conditions. This occurs in

A biogeochemical study of recent (multicores) sediments of the northwest African slope was undertaken to understand how the sediment composition varies with respect to the location of core sites relative to the centers of coastal upwelling, and how this has affected the palaeoceanographic record. Sedimentary organic carbon contents are inversely correlated with the nitrogen isotopic composition (δ 15 N), high C organic concentrations and low δ 15 N occurring at proximal (shallow) sites and the opposite at distal (deep) ones. These spatial differences are interpreted to result from higher relative nutrient utilization and a decrease in production as waters are advected offshore from the zone of upwelling. Highest C organic contents also correlate positively with highest concentrations of redox-sensitive elements (U, Mo and S) that are fixed diagenetically in the sediments. These results suggest that the sedimentary regime at a fixed position depends on the spatial location of the productive areas relatively to a given core site. Downcore records of Zr/Al, Ti/Al, mean grain size of the terrigenous fraction, δ 15 N, C organic , biogenic Ba, U, Mo and sulfur at a single site on the slope are interpreted to reflect glacial-interglacial changes in the core location relative to the coastline (sea-level effect), and hence changes in production as the area of coastal upwelling moved on- and offshore as sea-level changed, as well as undoubtedly changes in upwelling intensity through wind forcing. Further studies are needed to fully understand the interrelationships of all these processes, which are required for building more reliable paleoceanographic-paleoclimatic records.

The late Quaternary history of water-column denitrification of the eastern Pacific margins and the Arabian Sea is reconstructed using sedimentary delta(15)N measurements. The delta(15)N values in six piston cores raised from these regions show remarkably similar cyclic variations, being heavy (9-10.5 parts per thousand) during the interglacials and 2-3 parts per thousand lighter during the glacials. This implies that denitrification in these regions decreased substantially during the glacial periods. The glacial decline in denitrification is attributed to reduced upwelling and flux of organic material through the oxygen minimum zone. Since water-column denitrification in these areas accounts for about half of the fixed-nitrogen loss in the modern ocean, the inferred decrease in denitrification should have increased the oceanic nitrate inventory during glacial periods. Because nitrate is a limiting nutrient, oceanic productivity and attendant changes in CO(2) may therefore have been modulated on glacial-interglacial timescales by variations in the oceanic NO(3)(-) content.

A common feature in δ15N profiles downcore in continental margin sediments is that the heaviest δ15N values are frequently observed during the deglaciation (i.e., between 12,000 calendar years BP and the climatic optimum at 6000 yrs BP), not at the warmest stage. Using a conceptual model across the northwestern Africa margin, a region of pronounced modern upwelling, as well as data from a core in the area, we show that this feature can be explained as a consequence of postglacial sea level rise. The model is based on a simplified twodimensional physical circulation scheme orthogonal to the margin and uses the topographic profile at the latitude of the core as well as a simplified biological model for nitrate utilization and nitrogen isotope fractionation. Shore-parallel influences are ignored. The most recently published age model of sea level rise for the last deglaciation is used [Bard et al., 1996]. The trangression causes a progressive increase in the area of shallow regions where large amounts of nutrients are recycled relative to deep regions, to which a significant portion of the nutrients is exported. This causes first an increase and then a decrease in the δ15N of the organic matter accumulating at a fixed point on the upper slope. Although the deglacial δ15N maximum is more pronounced in areas where there is not a marked oxygen minimum layer [Holmes et al., 1997; this paper], it does exist in areas where an oxygen minimum layer is present in the water column [Altabet et al., 1995; Ganeshram et al., 1995]. In such areas, the major δ15N contrast between glacial and interglacial episodes is explained by higher denitrification during interglacial stages, but it is probable that transgressing sea level contributes to this effect. The model has implications for the changes of vertical oceanic nutrient fractionation [Boyle, 1988] and the respiratory dissolution of deep carbonates [Archer and Maier-Reimer, 1994] and hence could have important potential implications for the timing of global CO2 exchanges between ocean and atmosphere and their feedback to climate.

Although a clear explanation has yet to emerge for the observed decline in atmospheric CO2 content during glacial episodes, numerous hypotheses have been offered. Selected examples are reviewed or evaluated here. A strengthened biological pump resulting from intensified upwelling at low latitudes was suggested early on as a likely direct cause of the glacial CO2 drawdown. Two problems have since arisen with this hypothesis. First, on some upwelling continental margins, for example off California, Oregon, northwestern Mexico, Peru, and parts of NW Africa, organic matter accumulation decreased, in some areas markedly, during the Last Glacial Maximum (LGM). Second, upwelled cool CO2-rich water degasses carbon dioxide in warm regions, and for a key area such as the modern equatorial Pacific to become a net sink for atmospheric CO2, rather than a source as it is today, the rate of downward organic carbon export must exceed the rate of degassing. Recent sedimentary carbon and nitrogen isotopic data suggest however that the equatorial Pacific has remained a strong CO2 source for at least the last 40,000 yr and probably much longer. Although the direct extraction of carbon from surface waters and burial in the sediments in productive regions now appears unlikely to provide the sought-for explanation, indirect effects related to export productivity may hold the key. These can modulate the chemical character of the ocean so as to increase its uptake of CO2 (a rheostat effect) or they could increase indirectly the nutrient inventory of the sea (an amplifier effect). For example, it is now recognized that denitrification was greatly reduced during glacial maxima in all three principal oxygen minimum zones in the oceans, i.e. the subtropical north and south Pacific and the Arabian Sea. The implication is that nitrate, the limiting nutrient in the modern ocean, must have been more abundant during glacial periods, and that this surfeit would have supported increased export production in the meso or oligotrophic areas that are today nitrate-limited. Dust may represent another rheostat. Increases in atmospheric turbidity during glacials are clearly recorded by ice and marine sediment cores in both hemispheres. In addition to fostering enhanced export production by adding needed iron to nitrate-rich areas, it has been suggested recently that increased dust inputs to nitrate-depleted regions during glacials might have encouraged growth of iron-hungry N2 fixing cyanobacteria, thus alleviating nitrate limitation. No sedimentary δ 15 N data yet exist to support this hypothesis, but it remains viable. Finally, regional changes in physical hydrography may have played a major and hitherto underestimated role. For example, in the glacial Southern Ocean south of the Polar Front, recent nitrogen isotope and other data imply that the upper water column was well stratified during the LGM. By limiting upwelling, this would have reduced the ocean-atmosphere CO2 `leak’ in the area. This could have made a significant contribution to the pCO2 drawdown.

Hole 887B of the Ocean Drilling Program (ODP) comprises a 44 m (750 kyr) long continuous section recovered from the Patton–Murray Rise, an elevated plateau that is largely isolated from turbidite deposition. The Patton–Murray area is centered under the Alaska Gyre, a region characterized by the domal upwelling of nutrient-rich waters. Marked increases in productivity and rapid settling of biogenic matter are suggested throughout the section by the episodic accumulation of diatomaceous oozes up to ∼1 m thick that are accompanied by barium enrichments. Significant δ13Corg maxima in the major diatomaceous bands suggest that mixed-layer [CO2(aq)] may have been drawn down significantly during some of the productivity events. The episodes of enhanced productivity at Site 887 occur synchronously with short-lived minima in planktonic foram δ18O, suggesting a direct link to low salinity, or less likely, warming, events in the Gulf of Alaska. There is no obvious explanation for the events, but they may be related to seasonal incursions of meltwater from Alaska. We speculate that episodic input of meltwater- or dust-borne iron from Asian or Alaskan sources may have promoted the extraordinary diatom production events recorded in the sedimentary record.

The surface waters of the modern subarctic Pacific Ocean are isolated from the nutrient-rich waters below by a steep vertical gradient in salinity (halocline), a feature which is a dominant control on upper-ocean stratification in polar environments1,2,3. The physical processes which maintain the halocline and, in turn, its physical, biological, and geochemical effects have long been subjects of intense inquiry3,4. The stratification of polar surface waters influences the exchange of CO2 between ocean and atmosphere5,6,7,8, so the history of the subarctic Pacific halocline may have played a role in past changes in atmospheric CO2 concentration. Here we report opal accumulation rates and nitrogen-isotope data from sediments in this region which indicate that the subarctic Pacific halocline developed abruptly 2.73 million years ago, coincident with the onset of extensive Northern Hemisphere glaciation. The halocline would have reduced the transport of nutrient-rich deep water into the euphotic zone, leading to a decrease in biological production but an increase in the fraction of nutrient stocks utilized. This increase in the efficiency of the ‘biological pump’ would have lowered the rate of CO2 evasion from ocean to atmosphere, potentially reducing atmospheric CO2 concentrations from the suggested higher level of the preceding mid-Pliocene warm interval9,10.

Factors controlling the burial of organic carbon (OC) in Late Quaternary sediments on the NW Mexican continental margin are assessed using a suite of box and piston cores strategically located on the shelf-slope rise with respect to the intense oxygen minimum in this region. An OC concentration maximum occurs on the mid-slope, below the core of an intense water-column O2 minimum, due to current winnowing on the outer shelf, the preferential accumulation of organic matter in fine-grained deposits, and the offshore decrease in the settling flux of organic detritus. The organic matter at all water depths is overwhelmingly marine. Hydrogen indices (HI) are higher on the slope (>300 mg HC/g TOC) than on the shelf (

Intervals of organic C- and carbonate-rich laminated sediments occur in the Sea of Japan with roughly the same frequency as temperature changes observed in Greenland ice cores, providing clear evidence of rapid oceanographic change during the past 36 kyr. Planktonic foraminiferal δ18O data suggest that only the laminated sediments deposited during the Last Glacial Maximum (LGM), and perhaps one other interval formed during a period of increased water column stratification. Sedimentary Re and Mo data are consistent with bottom waters that were sulfidic during the LGM and suboxic during other laminated intervals. Results of a numerical model of Corg and Re burial are consistent with a mechanism whereby an increased Corg flux to the seafloor drove oxygen concentrations toward depletion during times of deposition of the suboxic laminated intervals. Such a process could have resulted from increased upwelling driven either by increased deep water formation due to colder and/or more saline surface waters or by stronger northeasterly monsoonal winds.

Laminated sediments deposited under anoxic bottom waters in the Japan Sea during the last glacial maximum (LGM) contain extremely well preserved calcareous microfossils and eolian carbonates. The radiocarbon age-difference between bulk sediment and monospecific planktonic foraminifera in discrete laminae from a core in the southern Japan Sea implies that ∼40% of the total carbonates in the sediments at the LGM are of eolian origin. Extrapolation of this result yields a rate of supply of eolian carbonates of ∼2800 tons d−1 to the entire Japan Sea during the LGM. The climatic significance of this flux potentially lies in its broader geographic extension, particularly in the interaction of the carbonate-bearing dust with shallow, corrosive North Pacific waters and with rain in the atmosphere. By increasing the alkalinity of such waters and by enhancing the biological pump the dust flux could have increased CO2 absorption by both the ocean and rain during the LGM.

Sedimentary accumulation of biogenic components (organic carbon, opal, and biogenic barium) on the northwestern Mexican margin declined during every glacial interval of the past 140 kyr, indicating decreases in upwelling-induced productivity during cold periods. The glacial-interglacial contrasts in upwelling on this margin are attributed to reversals in land-ocean thermal contrast, the waxing and waning of the Laurentide Ice Sheet, and consequent responses of the western hemisphere wind fields. This scenario is consistent with three independent lines of evidence: terrestrial paleoclimatic data, general circulation model results, and our marine records. This pattern of glacial-interglacial variability in upwelling off NW Mexico is opposite to that observed in other low-latitude and midlatitude upwelling areas, such as the eastern equatorial Pacific. These results add to a growing pool of observations that the response of oceanic upwelling to glacial climatic forcing has been regionally variable.

The effect of sea level change on nutrient supply to the anoxic Cariaco demonstrates the fundamental importance of nitrogen (N2) fixation and phosphate to oceanic production. As N2 fixation produces biomass of low δ15N and has been reported to be an important component of the nitrogen cycle in the modern Cariaco Basin, we propose that it contributes to the light interglacial δ15N (∼2‰–3‰) values observed in the Ocean Drilling Program (ODP) site 1002 sediment record. During the glacials the sediments are bioturbated (oxic conditions) with low total organic carbon (TOC) contents and sedimentary δ15N values of ∼5‰, suggesting that nitrogen (N2) fixation contributed little to the N nutrition of Cariaco surface waters. The most plausible explanation for the inferred glacial/interglacial changes in N2 fixation in the Cariaco is that they have occurred in response to variations in the N/P ratio of the nutrient supply, driven by changes in denitrification.

Micropaleontological data of core PAR87A-10 reveal that the last glacial interval, prior to 13 ka, was marked by low biogenic fluxes and poor CaCO3 preservation. Quantitative estimates of sea-surface conditions based on dinocyst assemblages suggest that cold temperatures and freezing winter conditions existed during this period. The glacial to interglacial transition, i.e., the 13–8 ka interval, was characterized by an increase in fluxes of microfossils indicating enhanced productivity in surface waters. A higher biogenic carbonate production probably resulted in better preservation of CaCO3. This interval was marked by relatively low salinity and by sea-surface temperatures increasing toward modern values. Relatively high pollen flux during the transition suggests nutrient inputs through atmospheric and/or fluvial transport from the adjacent North American continent. After 8 ka, diminished fluxes of plankton, concomitant with a decline in pollen input, are associated with decreasing nutrient supply as predominantly eastward winds became established over the North Pacific.

The trace elements Re and Mo both behave conservatively in seawater yet are strongly enriched in reducing sediments. Their potential for authigenic enrichment above crustal concentrations is greater than for many other elements, due to the high ratio of [metal] sw /[metal] crust . We present sedimentary Re and Mo data from box- and multi-cores spanning a range of redox conditions, from well-oxygenated sites to locations with substantial sulfide concentrations. At the oxic sites, Re and Mo, as expected, accumulate at concentrations close to their crustal abundances. Re shows substantial enrichment in suboxic (absence of O 2 and H 2 S) sediments of the Sea of Japan, in sediments within the oxygen minimum of the Pakistan margin, as well as in sediments underlying the sulfide-bearing waters of the Black Sea and Saanich Inlet. Re enrichment occurs in these cores just below the depths of Fe and U reduction. Only in the sediments underlying the sulfidic waters of the Black Sea and Saanich Inlet is there substantial authigenic Mo accumulation. Absence of Re and Mo enrichments in sediment trap samples from the sulfide-bearing basins suggests that the addition of both metals occurs at or below the sediment-water interface. The Re and Mo concentration profiles are modeled to hindcast the removal depth of Re and Mo in the sediments which, along with the Re/Mo ratio, determine whether oxic, suboxic or sulfidic conditions were present in the water column or in the sediments in the past. Using this approach, historical redox conditions can be inferred even in environments such as continental margins where a substantial lithogenic component can obscure authigenic enrichments of other metals.

The sedimentary geochemistry of manganese is dominated by the redox control of its speciation, higher oxidation states (Mn (super 3+) and (super 4+) ) occurring as insoluble oxyhydroxides in well-oxygenated environments and the lower oxidation state (Mn (super 2+) ) being much more soluble in oxygen-deficient settings. Its geochemical behavior is therefore quite different in oxic and anoxic environments, and where oxic and anoxic conditions are juxtaposed, Mn is recycled between the two environments. In modern marine sediments, Mn is present above its crustal abundance as an oxyhydroxide in all slowly accumulating (pelagic) sediments of the deep ocean and in surficial deposits of continental margin environments. Diagenetic recycling of Mn in the latter causes surficial deposits to have larger Mn enrichments than in many pelagic sediments. Bottom sediments of permanently anoxic basins show no enrichments and have Mn concentrations that are controlled solely by the aluminosilicate fraction. Manganese carbonates (kutnohorite and calcic rhodochrosite) are found only in anoxic sediments accumulating beneath surface oxic horizons (and therefore under oxygenated bottom waters) in many nearshore environments. Such enrichments are due to delivery of Mn by burial of surface oxyhydroxides into the subsurface anoxic environment where they are dissolved. Pore-water Mn levels can reach saturation with respect to a mixed Mn-Ca carbonate phase in such sediments. The diagenetic origin for these phases is shown by their carbon isotope compositions, which typically indicate a carbon source from decomposing organic matter. The presence of Mn carbonates therefore signifies that the host sediment must have accumulated under oxygenated bottom waters.On the basis of this information it is proposed that, in contrast to several current explanations for the formation of Mn carbonates (kutnohorite and rhodochrosite) in ancient organic-rich shales, limestone, and marl sequences and in many Mn ore deposits, the occurrence of these mineral phases indicates that the sediments originally accumulated beneath oxygenated bottom waters. By analogy with the present, Mn carbonates could not have formed in the bottom waters of anoxic basins. These diagenetic phases, however, did form where Mn was supplied at a high rate, namely, by the burial of oxyhydroxide-enriched surface sediments, to a subsurface anoxic environment. This situation could only have occurred under oxygenated bottom waters. The presence of Mn carbonates in ancient black shales (and in some carbonate-rich rocks) lends strong support to the notion that these rocks did not necessarily form in anoxic basins but owe their carbon richness to a high supply of organic matter to sediments deposited under oxygenated bottom waters, probably in continental margin settings.

CHANGES in ocean chemistry and circulation have been invoked to explain the lower atmospheric CO2 concentrations of glacial periods observed in ice-core records1. The processes that modulate these concentrations are not well understood, but an increase in the nutrient inventory of the ocean is one mechanism that could lower atmospheric CO2 levels by enhancing oceanic biological productivity and CO2 storage1-3. The oceanic concentrations of one such nutrient, nitrate, may be regulated by changes in the rate at which it is degraded by bacteria (denitrification) in oxygen-deficient subsurface waters. Denitrification constitutes a significant global sink for oceanic nitrate4, and the eastern tropical North Pacific Ocean is particularly important in this respect as it accounts for at least a third of global oceanic fixed-nitrogen removal by water-column denitrification4,5. Here we present 15N/14N records from marine sediment cores, which show that water-column denitrification in the eastern tropical North Pacific Ocean was greatly diminished during glacial periods. We suggest that, because nitrate limits biological productivity in much of the modern ocean, a consequent increase in the oceanic nitrate inventory during glacial periods could have contributed to the observed decrease in atmospheric CO2 concentration.

reflects transient iron limitation. Our new d 30 Si record from the Guaymas Basin shows dramatic variations at millennial timescales. Low d 30 Si values synchronous with Heinrich events are interpreted as resulting from the decline in Si(OH)4 utilization at times of decreased upwelling strength, while nearly complete Si(OH)4 utilization was observed at times of invigorated upwelling and increased opal burial during the Holocene, the Bolling-Allerod and the last glacial period. We attribute the complete utilization of Si(OH)4 to the occurrence of transient Fe limitation at these times. Our study highlights the importance of Fe limitation on Si and C cycling in coastal upwelling regions and suggests that upwelling dynamics, in combination with Fe availability, have the potential to modulate marine Si distribution and opal burial even at short timescales. Citation: Pichevin, L., R. S. Ganeshram, B. C. Reynolds, F. Prahl, T. F. Pedersen, R. Thunell, and E. L. McClymont (2012), Silicic acid biogeochemistry in the Gulf of California: Insights from sedimentary Si isotopes, Paleoceanography, 27, PA2201

[1] Particle fluxes over the Beaufort continental slope (70.548°N/140.045°W; 710 m) were measured at two depths (290 and 490 m) during the fall 1992 to spring 1993 period. Sequential sediment traps recorded a peak in particle flux under heavy ice cover in midwinter. The fine-grained, highly terrigenous material was consistent in composition with bottom sediments from the shelf and upper slope. The peak occurred coincident with high current speed and isopycnal displacements characteristic of cyclonic eddies, and upwelling-downwelling (UW-DW) cycles driven by winds and drifting ice. Such phenomena may be the operative mechanisms for transporting sediments from shelf to basin in the Arctic. Available evidence favors the origin of the midwinter peak to be rainout of particles from the cyclonic eddies; we suggest that the particles were earlier entrained into the eddy when it impinged upon the upper slope. Estimates show that sediment-laden water in an eddy could sustain rainout at 250 mg m−2 d−1 for ∼120 days, so that sediment could have been transported 100–500 km with the eddy before reaching our site. Current speeds generated by wind-driven UW-DW events were insufficient to resuspend sediment on the outer shelf, and the timing of strong events was unconvincing as a link to the enhanced sediment flux over the slope 50 km to the north. Bottom current must exceed 19–36 cm s−1 depending on the sediment type to resuspend material from the seabed of the Mackenzie shelf.

Long-term (i.e. 100 year) climate predictions for high-latitude stations indicate these places may experience significant annual and seasonal changes in air temperature, precipitation and type and amount of evaporation. Such changes would have a pronounced effect on the water balances of many mine sites and strategies to manage water, effluents and environmental impacts post-closure. This paper describes methods for the development of climate-scaled water balances for mine closure planning. The methods focus on: 1) the compilation of site-specific and regional climate data; 2) usage of a weather generator to produce long-term synthetic datasets representative of the location of interest; 3) assessment of climate change scenario data (e.g. general circulation model (GCM) and regional climate model (RCM)) and the selection of data scalars for model parameters of interest; and 4) the steps to combine elements of climate variability (weather generated) and climate change to produce a climate-scaled daily dataset representative of the location. Data are presented for a case study in northern Europe. The paper concludes with a discussion of the uncertainties inherent in the approach and opportunities and limitations in applying it to other locations.

The chemical and mineralogical composition of the surface sediments of the Sulu Sea, a deep, hydrographically-restricted basin in the Indonesian archipelago, and the neighbouring area of the South China Sea have been examined. The aim of the study was to characterize the geochemistry of the sediments of the isolated, oxygen-deficient basin and to compare them with sediments at equivalent water depths outside the basin, as a prelude to a further study of the response of the composition of the sediments to oceanographic changes, including possible anoxia, during the last glacial maximum. The concentrations of most of the chemical constituents in the surface sediments along four transects change markedly with water depth in the Sulu Sea (SS) and the South China Sea (SCS). Thus, CaCO 3 decreases from around 60% by weight in the shallowest samples (800–1000 m) to around a few percent in the deepest sample at 4203 m in the SCS, where the CCD lies at around 4000 m. In the SS, however, the CCD lies much deeper, so that the CaCO 3 contents are still approximately 40% at 4500 m depth. Organic C also decreases with depth in both basins, from values of around 1–2% in the shallowest samples to less than 1% in the deepest samples. The ratios of Si, Mg, Na, K, Co, Cr, Cu, Ni, Pb, Rb, Y, Zr and Zn to Al all decrease with increasing water depth because of the change in mineralogy and sediment texture with water depth. The P Al is higher in shallower water in both areas because of the higher abundance of detrital phosphorite or phosphatized limestone in these sediments. The Sr Al ratio is also higher in shallower water, reflecting the supply of aragonitic shell debris from the shallow banks and shelves around the basin. The Ba Al ratio has a maximum at 2000 m depth in the SCS, due to higher primary production in overlying surface waters, whereas it is uniformly low in the SS. Manganese is also enriched at intermediate water depths in both the SS and the SCS because of the presence of oxyhydroxides, and this also produces a slight enrichment in Mo in the same samples because of their association in this authigenic phase. The concentration of I and the I C organic ratio decreases with water depth in both areas due to the higher abundance of fresh, marine organic matter at the shallow water sites, which scavenges proportionally more I from sea water. Finally, the δ 13 C organic values tend to decrease irregularly with increasing water depth probably reflecting a higher contribution of terrestrial organic matter in deeper water. A discriminant analysis of the full data set reveals that, compared with the SCS samples, the oxygen-deficient SS sediments have higher CaCO 3 contents, lower N and S contents and lower I C organic , Ba Al , Fe Al , K Al , Pb Al , Rb Al , Ti Al , V Al , Y Al and Zn Al ratios. These contrasts are brought about by the different depths of the CCD, differences in the nature of the organic matter and different types of aluminosilicate detritus in the two areas. Although the I C organic ratios are slightly but significantly lower in the SS, which may be due to the low dissolved oxygen contents of the waters of this basin, the Mn and Mo contents and other minor elements that respond significantly to ambient bottom water oxygen levels are not significantly different in the SS and the SCS. Moreover, the organic C contents are also indistinguishable in the two areas; hence, the low bottom water oxygen levels in the SS do not lead to higher organic carbon contents in the surface sediments. The composition of sediments in contact with waters having even lower oxygen concentrations should be evaluated in order to conduct a more definitive test of the effect of low oxygen conditions on carbon preservation. Comparison of glacial-interglacial sediment sections from this area must take particular care to separate the effects of source variations on those potentially caused by low oxygen levels.

Powell and Sakinaw Lakes are stably stratified ex-fjords, which became isolated from the Strait of Georgia approximately 11000 years ago by emerged sills due to post-glacial isostatic rebound. Both lakes contain highly sulphidic relict seawater (Powell 3.0 mM; Sakinaw 5.5 mM) at depth chemically analogous to porewaters of reducing marine sediments. Sakinaw has a very sharp chemocline located just below the oxic/anoxic interface, whereas in Powell, the interface is spread out over 200 m of the water column. Sulphate is present in the oxic part of the water column at much higher concentrations in Sakinaw Lake (up to 65 μM) than in Powell Lake (10 μM), and increases in concentration with depth until reaching the interface, below which it is rapidly depleted. Sulphur oxyanions of intermediate oxidation state (thiosulphate, sulphite, and tetrathionate) are not detectable in either lake. Particulate elemental sulphur was also not found in either lake. High concentrations of reduced iron and hydrogen sulphide result in formation of iron monosulphides and pyrite in the anoxic water columns of both lakes. The presence of these two minerals correlates well with their calculated saturation states. Pyrite precipitates with no apparent monosulphide precursor at depths where dissolved sulphide concentrations are low and monosulphide phases are undersaturated. As sulphide levels increase at greater depths, the waters become saturated with respect to iron monosulphides and particulate FeS species are detected in the water column. The large separation of the oxic/anoxic interface and the chemocline in Sakinaw (~10 m) and especially in Powell Lake (~100 m) relative to that of sediment porewaters allows excellent resolution of these processes.