Sediment Trap Measurements Research Articles

Using JGOFS<I> in situ</I> and ocean color data to compare biogeochemical models and estimate their parameters in the subtropical North Atlantic Ocean

How well do biogeochemical data sets serve to decide among models and model parameter values? Data at 21N, 31W from the French JGOFS EUMELI cruises and the SeaWiFS ocean color sensor were used to estimate parameters for three very different models of biological nitrogen flux in a water column. The three models are (1) an NPZD (Nutrients, Phytoplankton, Zooplankton and Detritus) model (Oschlies et al., 2000), (2) a seven-component model with two pools of dissolved organic matter and detritus with different remineralization and sinking rates (Dadou et al., 2001) and (3) a model of nutrients and phytoplankton including aggregates (Kriest and Evans, 1999). Parameters of the three models are estimated using the same sets of data within the same one-dimensional physical framework. A combination of local and nonlocal optimization methods is used. It is not easy to decide among candidate models based on their fit to the data. Parameters that mean the same thing in the three models, like the half-saturation concentration for nitrate uptake, were estimated at not very different values in different models. The model with dissolved organic matter, based on its primary production and sediment flux data time evolutions, seems to exhibit the more reasonable annual behavior. Large seasonal changes in deep nitrate data suggest an unexpected role of lateral advection and may vitiate the 1-D approach even at the EUMELI oligotrophic site. The small number of sediment trap measurements are very powerful for constraining the biological nitrogen. Ocean color data did not add extra constraining power.

Comment on “determination of areal sedimentation rates in rivers by using plate sediment trap measurements and flow velocity—settling flux relationship” by Hans-Peter Kozerski

Comment on “determination of areal sedimentation rates in rivers by using plate sediment trap measurements and flow velocity—settling flux relationship” by Hans-Peter Kozerski

More about plate sediment trap measurements—Reply to the comment of Sukhodolov et al.

More about plate sediment trap measurements—Reply to the comment of Sukhodolov et al.

Determination of areal sedimentation rates in rivers by using plate sediment trap measurements and flow velocity—settling flux relationship

Not much is known about sedimentation of organic particles in rivers. Plate sediment traps deployed in fast flowing River Spree and in a slowly flowing side arm yielded trapping rates between 0.7 and 52 gDW m −2 d −1. A linear regression between flow velocity and trapping rates was used to extrapolate from point measurements to areal average trapping rates, which were 11 gDW m −2 d −1 in the side arm and 1.4 gDW m −2 d −1 in the free flowing river section. These rates are typical for the River Spree, but different from other rivers. The plate trap measurements provide the basic variables for the sedimentation rate sed= C· W s *(1− V/ V crit,sed), i.e. concentration of particles ( C), sinking velocity ( W s), flow velocity ( V), and a threshold velocity ( V crit,sed).

Sediment-Trap Measurements of Suspended Mine Tailings in Shallow Water Cover

Disposal of pyrites mine tailings under shallow water covers is an effective and commonly used method for reducing tailings oxidation, thus minimizing the formation of acid rock drainage. However, wind-wave-induced resuspension of tailings under the water cover could compromise suspended solids discharge criteria and increase the overall sulfide oxidation, since dissolved oxygen concentration are higher in the water cover than in the pore water. This article reports field measurements of suspended tailings collected approximately 2-10 cm above the tailings surface by means of 16 sediment traps partly buried in the tailings bed. The traps were distributed over ab area of about 600x300 m, where the water cover depth varied from 0 to 2 m. The traps were sampled on three occasions during a total study period of 93 days distributed over two field seasons. Wind speed and direction were monitored at the site during the study. Hourly average wind speeds ranged from 0 to 10 m/s. Suspended tailings were recovered in every sediment trap on each sampling event. Comparison of particle size distributions for suspended solids and bed tailings suggests that at this site and for the duration of this study, tailings erosion and resuspension occurred primarily in areas where the water cover depth was 1.0 m or less, and that suspended tailings were transported by currents to other parts of the water cover. The amount of tailings resuspension was related to the frequency of strong winds blowing in the direction of maximum fetch. Comparison of suspended solids and bed tailings compositions indicates that suspended tailings oxidized at a larger rate than bed tailings.

Seasonal nutrient depletion and carbon export in the Subantarctic and Polar Frontal zones of the Southern Ocean south of Australia

Seasonal surface water nitrate and phosphate depletions were calculated for the Subantarctic Zone (SAZ) and Polar Frontal Zone (PFZ) of the Southern Ocean along the World Ocean Circulation Experiment SR3 repeat section south of Tasmania, Australia, and three other transects slightly to the west. The depletions were calculated between a single winter (July) section and eight other sections from October to March in the years 1991‐1998 and corrected for frontal movements and changes in water masses between the cruises using observed quasi‐linear nutrient‐salinity relationships. Arranged in a seasonal progression, depletion began in approximately October and continued through March, in good agreement with satellite biomass‐based estimates of production. Nitrate depletion in the SAZ (510 mmol N m−2) was approximately twice that of the PFZ (250 mmol N m−2), but phosphate depletions were similar (29 and 24 mmol P m−2 for the SAZ and PFZ, respectively). The SAZ exhibited a Redfield N/P depletion ratio (15.1±2.3), but the PFZ ratio was much lower (8.3±5.4). While the SAZ exhibits greater seasonal nutrient depletion, the relative magnitudes of export in the two zones remain uncertain because supply of nutrients from south of the Polar Front may lower the observed nutrient depletions in the PFZ. However, comparisons with sediment trap and production measurements for the region south of Australia suggest the SAZ and PFZ differences derive from differences in production. Assuming a Redfield C/N ratio (6.6), carbon export for the longest period examined (July‐March) averaged 3400±430 mmol C m−2 in the SAZ, with a range between 2700 and 3900 mmol C m−2 depending on latitude. In the PFZ, carbon export averaged 1600±350 mmol C m−2, with a range of 1100 to 2000 mmol C m−2. These averages represent minimum annual export production estimates because neither resupply nor export outside the July‐March observational period is included.

Moored sediment trap measurements of carbon export in the Subantarctic and Polar Frontal zones of the Southern Ocean, south of Australia

Sediment trap moorings were deployed from September 21, 1997 through February 21, 1998 at three locations south of Australia along 140°E: at ∼47°S in the central Subantarctic Zone (SAZ) with traps at 1060, 2050, and 3850 m depth, at ∼51°S in the Subantarctic Front with one trap at 3080 m, and at ∼54°S in the Polar Frontal Zone (PFZ) with traps at 830 and 1580 m. Particle fluxes were high at all the sites (18–32 g m−2 yr−1 total mass and 0.5–1.4 g organic carbon m−2 yr−1 at −1000 m, assuming minimal flux outside the sampled summer period). These values are similar to other Southern Ocean results and to the median estimated for the global ocean by Lampitt and Antia [1997], and emphasize that the Southern Ocean exports considerable carbon to the deep sea despite its “high‐nutrient, low chlorophyll” characteristics. The SAZ site was dominated by carbonate (>50% of total mass) and the PFZ site by biogenic silica (>50% of total mass). Both sites exhibited high export in spring and late summer, with an intervening low flux period in December. For the 153 day collection period, particulate organic carbon export was somewhat higher in all the traps in the SAZ (range 0.57–0.84 gC m−2) than in the PFZ (range 0.31–0.53), with an intermediate value observed at the SAF (0.60). The fraction of surface organic carbon export (estimated from seasonal nutrient depletion, Lourey and Trull [2001]) reaching 1000 m was indistinguishable in the SAZ and PFZ, despite different algal communities.

Seasonal dynamics of coastal ecosystems and export production at high latitudes: A modeling study

Export of organic matter from the surface to deeper waters often shows much smaller seasonal variations than primary production or nitrate‐based new production in mid‐ to high‐latitude marine systems. The mechanisms underlying this pattern remain poorly understood, but seasonal shifts in food web structure and dynamics have been implicated. We report here on an ecosystem modeling analysis of a high‐resolution (biweekly) time series of biomass, production, and export flux (sediment trap) measurements conducted in 1991 in Bonne Bay (Newfoundland). This time series shows the classical pattern of a spring bloom followed by a summer low biomass period, yet export is bimodal, with maxima during spring and late summer. The ecosystem model was forced by diagnostic vertical mixing calculations based on temperature and salinity records taken every 3 d and hourly wind data. The physical analysis indicated that the nitrate flux into the euphotic zone during summer was equivalent to that during the spring and fall seasons and accounted for half of the summer export. Statistical adjustments of the parameters of the ecosystem model indicated that strong production of dissolved organic carbon during the spring bloom, high temperature dependence of microbial activity, and physico‐chemical particle aggregation played key roles in explaining the remainder of the summer export. Seasonal changes in trophic pathways between spring and summer, such as a shift from a herbivorous to a microbial food web, played a comparatively smaller role. Our modeling analysis suggests that physical mixing processes and physico‐chemical aggregation processes are at least as important as shifts in food web trophic pathways in explaining the postbloom export flux in mid‐ to high‐latitude marine systems.

The carbon cycle at the ocean margin in the northern Gulf of Biscay

Abstract The aim of this paper is to provide a synthesis of several multidisciplinary studies performed during the OMEX I project mostly presented in this volume. Our final goal is to construct a carbon cycle for the shelf break of the northern Gulf of Biscay. This margin is characterized by a steep slope adjacent to a broad continental shelf, under limited influence of terrestrial input. Remote sensing images indicate the occurrence of cold water masses at the shelf break induced by complex physical processes related to the presence of the bathymetric discontinuity and the existence of frequent heavy storms. The primary production at the ocean margin remains however moderately high, around 200 g C m−2 a−1, with a f-ratio of 0.5. Most of the resulting new production is respired either in the water column below the euphotic zone (∼50 g C m−2 a−1) or at the sediment-water interface (∼20 g C m−2 a−1). According to sediment trap measurements, the difference between new production and respiration (∼30 g C m−2 a−1) may be exported to the slope and the open ocean. OMEX I benthic studies have shown that organic carbon (∼0.1 g C m−2 a−1) is not accumulated significantly in sediments and that no depocenter can be identified in this margin area. As expected, the primary production and thus the fluxes of organic carbon decrease with distance away from the shelf break and do not exceed ∼140 g C m−2 a−1 at a distance of 150 km from the shelf break. Despite some lateral inputs, there is again no significant accumulation of organic carbon in the sediments of the adjacent slope or abyssal plain. A nitrogen cycle is also described, based on the carbon cycle, demonstrating the importance of vertical mixing at the margin.

Food web structure and nutrient enrichment: effects on sediment phosphorus retention in whole-lake experiments

A series of whole-lake manipulations of both food web structure and nutrient loading rate revealed that the downward vertical P flux was significantly affected by food web structure. Food webs and nutrient input rates of two lakes were manipulated while a third lake served as a reference system. Phosphorus transport to the sediments was examined by three independent methods: mass-balance budgets, sediment traps, and sediment cores. After 2 years of pretreatment study, manipulated lakes were fertilized for 5 years at rates that were similar within each year but varied among years from 0.97 to 6.0 mg P·m-2·day-1. Increased vertical P flux was associated with the increased abundance of large-bodied zooplankton grazers such as Daphnia pulex. Vertical P flux increased with P input rates, but the increase was often insufficient to prevent P accumulation in the water column. Sediment trap measurements of the vertical P flux were significantly higher than the long-term sediment P retention measured by sediment cores and the seasonal sediment P retention calculated by mass balance. Using 210Pb data from the sediment cores, we corrected the sediment trap measurements of vertical P flux for focusing and brought them into better agreement with the seasonal sediment P retention calculated by mass balance.

Remineralization of organic carbon in eastern Canadian continental margin sediments

Undisturbed sediment samples were collected for chemical analyses at six sites during winter and summer cruises to the eastern Canadian continental margin. Micro-electrode oxygen profiles were obtained in freshly collected multicorer samples, and replicate cores were incubated at in situ temperature for 48 h to monitor changes in the concentrations of dissolved oxygen and nitrate. In addition, box cores were subsampled vertically for porewater chemistry, porosity, and particulate carbon. The data obtained are combined with estimates of sedimentation rate based on sediment trap measurements, 210 Pb dating and historical data to evaluate the role of benthic processes in the carbon cycle on the eastern Canadian continental margin. With one exception, oxygen uptake rates determined from incubations and calculated from micro-profiles were very similar, indicating that exchange of oxygen across the sediment–water interface was dominated by molecular diffusion. On the basis of this observation, transport by diffusion is assumed for the calculation of the flux rates for other solutes from their respective porewater gradients. The fluxes of oxygen into the sediments were low, but generally comparable to other continental margins at comparable depths. They varied from 1.4 to 1.8 mmol/m 2/d in December 1993 and from 2.8 to 4.5 mmol/m 2/d in June 1994. Uptake of nitrate by the sediment occurred at all sites except for the continental slope off Nova Scotia. Both oxygen and nitrate uptake were higher in summer than in winter, indicative of a lingering response to the input of organic matter associated with the early spring bloom. At one of the sampling sites, Miscou Channel, the measured oxygen uptake rate far exceeded the flux calculated from the oxygen gradient. The difference suggests biologically enhanced exchange with the overlying waters at this site, consistent with the greater abundance of benthic organisms. The rate of organic carbon mineralization at the seafloor (1.6 – 4.2 mmol/m 2/d) was estimated as the sum of the sediment oxygen and nitrate uptake rates. The contribution of other electron-acceptors to the mineralization of organic carbon is assumed to be accounted for by the oxidation of most of their reduced by-products by O 2 and NO 3 −. The return fluxes of dissolved inorganic carbon (∑CO 2 efflux corrected for carbonate dissolution) were generally comparable to the carbon oxidation rates. A detailed carbon budget was established at one of the sampled stations for which a complete set of measurements is available. The vertical flux of total particulate carbon through the water column (measured with a sediment trap at 150 m depth) accounts for approximately 9% of the annual primary production (PP). In the sediment, about 6% of the PP is remineralized and total carbon equivalent to 4.5% PP is buried. The self-consistency of the flux data at this station is used to justify similar budget calculations at the other stations where a limited data set was gathered. Sequestration of organic carbon via burial with the accumulating sediments (on average 0.5 mol/m 2/yr) is partially offset by the release of CO 2 associated with carbonate precipitation and burial (about 0.2 mol/m 2/yr).

Food web structure and nutrient enrichment: effects on sediment phosphorus retention in whole-lake experiments

A series of whole-lake manipulations of both food web structure and nutrient loading rate revealed that the downward vertical P flux was significantly affected by food web structure. Food webs and nutrient input rates of two lakes were manipulated while a third lake served as a reference system. Phosphorus transport to the sediments was examined by three independent methods: mass-balance budgets, sediment traps, and sediment cores. After 2 years of pretreatment study, manipulated lakes were fertilized for 5 years at rates that were similar within each year but varied among years from 0.97 to 6.0 mg P·m-2·day-1. Increased vertical P flux was associated with the increased abundance of large-bodied zooplankton grazers such as Daphnia pulex. Vertical P flux increased with P input rates, but the increase was often insufficient to prevent P accumulation in the water column. Sediment trap measurements of the vertical P flux were significantly higher than the long-term sediment P retention measured by sediment cores and the seasonal sediment P retention calculated by mass balance. Using 210Pb data from the sediment cores, we corrected the sediment trap measurements of vertical P flux for focusing and brought them into better agreement with the seasonal sediment P retention calculated by mass balance.

Pyrophaeophorbide- a as a tracer of suspended particulate organic matter from the NE Pacific continental margin

Pyrophaeophorbide- a, a degradation product of chlorophyll- a, is predominantly formed by grazing processes in sediments as well as in the water column. Water column profiles of pyrophaeophorbide- a/suspended particulate organic carbon (SPOC) concentrations, at an abyssal site in the northeast (NE) Pacific (Sta M, 34°50′N, 123°00′W; 4100 m water depth), show low concentrations (0.01–0.1 ng/μg SPOC) at surface and mesopelagic depths, and increasing concentrations with closer proximity to the sea floor (0.05–0.6 ng/μg SPOC). However, in June 1992, the deep maximum of pyrophaeophorbide- a/SPOC in the water column of Sta M extended higher into the water column, as much as 1600 m above the bottom (mab) (2500 m water depth); in other seasons they only extended up to 650 mab (3450 m water depth). Previous studies have demonstrated lateral transport of particulate matter from the continental shelf to the deep ocean off the coast of northern California. Recent work suggests that the benthic boundary layer (BBL) extends to 50 mab, based on sediment trap and transmissometry measurements (Smith, K.L., Kaukmann, R.S., Baldwin, R.J., 1994. Coupling of near-bottom pelagic and benthic processes at abyssal depths. Limnology and Oceanography 39, 1101–1118.), and that lateral transport is significant only during summer, which is consistent with our observations. A partial vertical profile of pyrophaeophorbide- a/SPOC from the north central (NC) Pacific provides some evidence that the deep maximum may be absent due to the distance of this site from the continental margin. Thus, the observed deep maximum of pyrophaeophorbide- a/SPOC at Sta M is likely due mainly to lateral transport from the continental slope rather than to local vertical resuspension in the BBL exclusively. Pyrophaeophorbide- a concentrations in SPOC at Sta M were negatively correlated with Δ 14C values of SPOC (SPOC samples from Druffel, E.R.M., Bauer, J.E., Williams, P.M., Griffin, S.A. and Wolgast, D., 1996, Seasonal variability of particulate organic radiocarbon in the northeast Pacific Ocean. Journal of Geophysical Research 101, 20543–20552), further supporting our contention that pyrophaeophorbide- a peaks in the deep water column are derived from “older” resuspended sediments that were laterally transported from continental margin sediments. Our molecular biomarker (pyrophaeophorbide- a) data are reflective of a specific fraction of SPOC that likely remains in the water column for long periods of time or is derived from resuspended sediments.

The Sediments and Benthic‐Pelagic Exchange in the Shallow Lake Müggelsee (Berlin, Germany)

AbstractThe sediments and exchange processes of the shallow (mean depth 4.9 m), polymictic, eutrophic Müggelsee have been studied over more than 30 years. Here, previous results are summarised and completed by new data. Sediment analyses were used to identify changes in loadings and migration processes within the sediments of the 12000 year old lake. The surficial oligohumic sediments are rich in Fe (8.8 ± 1.9% DW) and in P (2.7 ± 0.5 mg P g−1 DW). Mass balances of P and Fe as well as sediment trap measurements and sinking velocity analyses showed that the retention of matter strongly varies within seasons and years. The lake amplifies the seasonal fluctuations of P in the River Spree system by a strong redox‐controlled, benthic P‐release in summer. Since 1988, P‐release has exceeded retention so that the lake changed from a P‐sink to a source. The retention rate dropped in summer to ‐7.3 mg P m−2 d−1 and in winter to 1.4 mg P m−2 d−1.

Under which conditions can we expect realistic sediment trap measurements?

Under which conditions can we expect realistic sediment trap measurements?

Vertical fluxes of particulate matter in Aarhus Bight, Denmark

Sediment traps, transmissometer readings and sediment core-dating have been used to describe sediment dynamics within a shallow micro-tidal coastal environment, characterized by intense resuspension. A method to separate the downward fluxes of primary settling matter and resuspended sediments is proposed. It is shown that these fluxes can be separated during periods when a marked pycnocline exists in the water column, preventing resuspended sediment from reaching the sediment traps above the pycnocline. At such times it is possible to make quantitative calculations of the flux of primary settling matter at the sediment-water interface. Good agreement between particulate fluxes calculated on the basis of sediment trap measurements and accumulation rates determined from Pb-210 core dating have been obtained in this study, supporting the idea of using sediment trap measurements to calculate sediment accumulation at the bottom.

Unbalanced particle flux budgets in Crater Lake, Oregon: Implications for edge effects and sedimentfocusing in lakes

Flux estimates show that upward mixing of the deep‐water nitrate pool accounts for more than 85% of the total new nitrogen input to the euphotic zone of Crater Lake. Because measured primary productivity (360 mg C m−2 d−1) is 10–30 times higher than a level supported solely by the input of new nitrogen into the euphotic zone, nitrogen must be recycled in the euphotic zone many times before it is transferred by particles to the deep lake. Nitrogen recycling in the deep lake is also very efficient. Sediment trap measurements of particulate nitrogen fluxes reveal a major imbalance in our estimates of the lake’s internal nitrogen budget. We propose the imbalance reflects an “edge effect,” whereby enhanced biological production occurs near the lake margin and the shallower portions of the lake. Our measurements also reveal that Al, an element carried by refractory phases, is accumulating in basin sediments at a rate 15–30 times higher than the flux we measure with sediment traps. This difference is maintained by the near‐bottom transport of lithogenic particles from the lake margins to the deep basins of the lake. These comparisons of nutrient and refractory element fluxes reflect two important lake processes— enhanced productivity at the lake margin that may be due to greater availability of macro‐ and micronutrients at the lake edges and focusing of particulate material into the deep lake basins.

Fecal pellet flux in Dabob Bay during a diatom bloom: Contribution of microzooplankton

During a spring bloom primarily composed of the centric diatom Thalassiosira pacifica in Dabob Bay, Washington, an unarmored dinoflagellate, Gyrodinium sp., phagocytized chains of Thalassiosira. The diatoms were egested in peritrophic membrane‐bound fecal pellets that were collected in sediment traps. The dinoflagellate pellets accounted for an average of 29% (SD = 16%) of the volume of fecal pellet flux at 50 m. The diatoms in the pellets were intact but void of plasma content, thus reducing the flux of both POC and pigments per volume of fecal pellets, as was apparent in the sediment trap measurements. Grazing by these dinoflagellates on the large diatoms and the export of relatively carbon‐ and pigment‐deplete pellets have much different implications for organic material cycling than if the diatoms were consumed by mesozooplankton or if they sank.

Contamination of sediment trap fluxes by vertically migrating phototrophic micro-organisms in the coastal Baltic Sea

MEPS Marine Ecology Progress Series Contact the journal Facebook Twitter RSS Mailing List Subscribe to our mailing list via Mailchimp HomeLatest VolumeAbout the JournalEditorsTheme Sections MEPS 122:45-58 (1995) - doi:10.3354/meps122045 Contamination of sediment trap fluxes by vertically migrating phototrophic micro-organisms in the coastal Baltic Sea Heiskanen AS Sedimentation rates of phytoplankton and particulate organic carbon (POC) were studied in the northern Baltic Sea, SW coast of Finland, during the spring bloom and early summer periods in May and June 1988. Sediment traps were moored at 15 and 30 m depths, and the cylinders were either preserved with formaldehyde (F) or remained as unpreserved control cylinders (C). The phytoplankton spring bloom in May was dominated by motile dinoflagellates. Due to the high abundance and vertical migration of the chain-forming dinoflagellate Peridiniellacatenata, the POC fluxes measured by the F-cylinders were, on average overestimated by 12%, and at most by 32% during the spring period. Between 3 and 12% of the P. catenata population was estimated to migrate daily below 15 m depth; part of the population also reached 30 m depth. Another migrating phototrophic organism present in the water column was the ciliate Mesodiniumrubrum. Generally, a larger fraction of the M. rubrum population (2 to 26%) migrated, but they did not contaminate the measurements to the same degree as P. catenata (&LT3% of total POC fluxes) due to their lesser abundance in the water column. In June, Eutreptiella sp. (Euglenophyceae) had significantly higher cell numbers in the F- than in the C-cylinders. They were estimated to contaminate the POC fluxes up to 13% at 15 m depth (F-cylinder). The difference in the Eutreptiella cell numbers between the 2 treatments could have been partly due to increased copepod grazing rates inside the C-cylinders (up to 90% of the total difference if they fed on phytoflagellates only). Contamination caused by the total apparent vertical migration of P. catenata, M. rubrum and Eutreptiella sp. accounted for 17 and 12%, on average, of the total POC sedimentation (F-cylinders) at 15 and 30 m depths, respectively. This study indicates that vertical migration of the phototrophic micro-organisms can significantly bias sediment trap measurements in shallow coastal areas where sediment traps are often deployed within the vertical range of the migrating microplankton (i.e. above 30 to 40 m). Sediment traps · Sedimentation · Preservation · Swimmer contamination · Vertical migration · Dinoflagellates · Mesodiniumrubrum · Eutreptiella sp. Full text in pdf format PreviousNextExport citation RSS - Facebook - Tweet - linkedIn Cited by Published in MEPS Vol. 122. Publication date: June 15, 1995 Print ISSN:0171-8630; Online ISSN:1616-1599 Copyright © 1995 Inter-Research.

Carbon-cycle imbalances in the Sargasso Sea

THE net exchange of carbon dioxide between the atmosphere and the ocean, and thus the nature of the oceanic carbon sink, is dominated by the seasonal dynamics of carbon cycling in the upper ocean. This cycle represents a balance between abiotic and biotic carbon transport into, and export out of, the ocean's upper layer. Here we report measurements of these processes made over five years in the Sargasso Sea off Bermuda, as part of the US Joint Global Ocean Flux Study (JGOFS). We find that the decrease in carbon stocks from the spring to the autumn in the upper 150 m of the ocean is three times larger than the measured sum of biotic and abiotic fluxes out of this layer. This discrepancy can be explained either by failure to account for horizontal advection of carbon or by inaccuracies in the fluxes of sinking particles as measured using sediment traps. Either the traps miss 80% of the sinking particles, or 70% of the carbon cycling is due to advection (or a combination of both processes is responsible). Sediment-trap measurements of the 234Th flux during this period suggest that most of the discrepancy may be due to inaccuracies in the trap methods, which would require a very general reassessment of existing ideas about particle export and remineralization of carbon in the oceans. If, on the other hand, advection is the main source of the discrepancy, the traditional one-dimensional (vertical) modelling of the oceanic carbon cycle cannot give a full account of carbon dynamics.