evidence for resuspension of rebound particles from near-bottom sediment traps

Deep-Sea Research, Vol. 35, No. 1, pp. 59-70, 1988. 0198-0149/88 $3.00 + 0.00 Printed in Great Britain. © 1988 Pergamon Journals Ltd.

Evidence for resuspension of rebound particles from near-bottom sediment traps

IAN WALSH,*t KATHY FISCHER,* DAVID MURRAY*t a n d JACK DYMOND*

(Received 30 April 1986; in revised form 2 April 1987; accepted 30 June 1987)

Abstract--Near-bottom sediment trap moorings were recovered at three sites in the North Equatorial Pacific. Total particulate fluxes recorded within 30-350 m of the ocean floor were greater than those recorded in the mid-water column. A simple two-component mixing model using the most organic-rich surface sediments sampled and an extrapolated mid-water column flux does not account for increases in the major biogenic components (organic carbon, calcium carbonate and opal). The resuspension of rebound particles (those particles that have settled through the water column but have not become incorporated into the sediments) may account for the observed flux near the ocean floor.

I N T R O D U C T I O N

FECAL pellets and "marine snow" composed of biogenic detritus derived from euphotic zone productivity form the largest component of the total flux of particles to the sediment surface (SCHRADER, 1971; McCAvE, 1975; HONJO, 1978; SHANKS and TRENT, 1980; DEUSER et al., 1981; URRERE and KNAUER, 1981). With an average settling rate of 100- 150 m day -i, these particles take 1-2 months to reach the bottom of the deep ocean basins (FOWLER and SMALL, 1972; SMALL et al., 1979; SHANKS and TRENT, 1980; SUESS, 1980; KOMAR et al., 1981; DEUSER, 1986). The major components of the particle flux; organic carbon, CaCO3 and biogenic opal, are partially recycled during settling (WALSH et al., 1988). Consequently, the primary particle flux (the material settling through the water column derived from the overlying photic zone) decreases with depth if there are no inputs from other sources.

Sediment traps moored within the nepheloid layer collect both the primary particle flux and resuspended sediment (SPENCER et al., 1978; ROWE and GARDNER, 1979; FISCHER, 1984; GARDNER et al., 1985). Since sediments are enriched in clay and strongly depleted in biogenic material relative to the primary particle flux measured in mid-water traps (FISCHER, 1984), and nepheloid layer particles reflect the composition of the underlying surface sediments (GARDNER et al., 1985), near-bottom sediment traps record strong increases in the flux of aluminosilicate-bound elements such as Al, Fe and Sc, as compared to sediment traps moored above the nepheloid layer.

Settling rates of clay-sized particles are too slow to contribute significantly to the downward flux of particles (McCAVE, 1975); thus suspended nepheloid layer particles

* College of Oceanography, Oregon State University, Corvallis, OR 97331, U.S.A. t Present address: Department of Oceanography, Texas A&M University, College Station, TX 77843,

U.S.A. $ Present address: Department of Geological Sciences, Brown University, Providence, RI 02906, U.S.A.

59

60 I. WALSH et al.

should not contribute significantly to the observed near-bottom flux increases without some process of aggregation. Aluminosilicate-rich fecal pellets produced by filter feeding benthypelagic zooplankton in the nepheloid layer may be a biologically mediated method of aggregation contributing to the increased fluxes observed in the near-bottom (SPENCER et al. , 1978).

Within the nepheloid layer, production of fecal pellets probably increases towards the bottom as the biomass of benthypelagic zooplankton is greater 10 m above the bottom (m.a.b.) than at 100 m.a.b. (WISHNER, 1980a,b). Additionally, the ratio of the eddy diffusivity coefficient to settling rate decreases with increasing particle size and density, thus vertical mass transport by currents is also inversely related to distance above the bottom. Therefore, the settling flux of resuspended particles is inversely related to the distance above the bottom.

The composition of nepheloid layer particles, however, may reflect seasonal changes in productivity. The primary particle flux reflects seasonal productivity changes (DEUSER et al. , 1981; BILLET et al. , 1983; FISCHER, 1984; DEUSER, 1986), resulting in variability in the near-bottom environment (BILLET et al. , 1983; LAMPn-r, 1985). At a site 4000 m deep in the northeast Atlantic the seafloor became covered with a carpet of particles derived from the spring bloom, which had settled at a rate of 100-150 m day -1 (LAMPI~, 1985). The particles were not mixed into the sediment, were resuspended when current speeds 1 m.a.b, exceeded 7 cm s -1, and were no longer visible within a month after the peak of deposition.

We present here the results of 3 year long deployments of sediment traps at MANOP (Manganese Nodule Project) sites C, H and M in the north equatorial Pacific (Fig. 1). The near-bottom traps at these sites recorded increased fluxes of biogenic and refractory components compared to mid-water traps (Figs 2 and 3). We present simple multiple- component mixing models, including resuspension of surface sediments and resuspension of particles, with compositions extrapolated from the primary particle flux to account for the near-bottom flux.

The resuspension of primary flux particles observed by BILLET et al. (1983) and LAMPITr (1985), and modeled here, differs from the resuspension of a fluff layer in that

36 N

z6

16

0 c

I I I

t C

I I I 140 ° W

~... f -

120 ° I 0 0 ° 8 0 °

Fig. 1. Map of the North Pacific showing locations of MANOP sites C, H, and M. The traps were deployed at site C from December 1982 to February 1983, and at site H and M from

September 1980 to October 1981.

650

1300

1950

DEPTH

2600

(m)

3250

3900

4500

BULK Corg CaCO3

0 950 1450

OPAL

7?7777" I I I I

200 400 600 800

Fig. 2. Bulk and biogenic fluxes measured at MANOP sites C (open circles), M (filled circles) and H (open squares). Bulk flux is in mg cm -2 y-1 biogenic fluxes are in lag cm 2 y 1. Note differences in scales and near-bottom flux increases. Primary fluxes to the near-bottom traps were extrapolated using a first order decay model from the measured flux in the 3075 m trap at H, the

1565mtrap at M, and the 3495mt rap at C.

AI Fe Ba Sc

°- I 650 --

1 3 0 0 - ~

1950"-~T ~ DEPTHIII /

oo-ii i 3250---~

] 2 24 36 48

75--

r r - - - ~ ~ i

5 15 25 1.5 4.5 7.5 ] 4 8 12 16 Fig. 3. Refractory element fluxes measured at MANOP sites C (open circles), M (filled circles) and H (open squares). Fluxes are in p_g cm 2 y t, Sc x 10 3. Note differences in scales and near- bottom flux increases. Primary fluxes to the near-bottom traps were assumed equal to the

measured flux in the 3075 m trap at H, the 1565 m trap at M, and the 3495 m trap at C.

62 I. WALSH et al.

the particles are not mixed into the sediment and the sediment surface is not necessarily disturbed. Additionally, the concentration of organic carbon in the fluff layer is much closer to the concentration in underlying sediment than primary flux material (SOUTAR et al., 1981; FISCHER, 1984), and therefore the addition of sufficient fluff to account for the organic carbon flux increase would be apparent in the dilution of the sediment trap material by aluminosilicates. In particular, at MANOP site H the highest organic carbon concentration in the surface sediment is 1.19% whereas the organic carbon concentration in the material recovered in the sediment trap 500 m.a.b. (assumed to have little or no resuspension input) was 5.4%. Similar relationships between organic carbon concentrations in surface sediments and sediment traps were found at MANOP sites M and C. To distinguish resuspension of surface sediment from resuspension of particles that have reached the sediment surface but have not become incorporated into the sediments we adopt the term "rebound" for such particles.

METHODOLOGY

Data were collected from long-term deep ocean mooring arrays at three MANOP (Manganese Nodule Project) sites in the Pacific (Fig. 1). Site H is located in the Guatemala Basin about 900 km east of the crest of the East Pacific Rise (EPR) and 900 km west of the continental margin. Local relief is approximately 25 m. The sediments are hemipelagic and lie below the CaCO 3 compensation depth. During the deployment current velocities 4.7 m.a.b, averaged 3 cm s -1 with a maximum of 10 cm s -~ (GARDNER e t al., 1984).

Site M is located in a sediment pond approximately 25 km east of the crest of the EPR. Local relief is dominated by fault scarps and is on the order of 150 m. The sediments are comprised of metaliferous particles from an EPR source in addition to particles settling from surface waters. Current velocities 200 m.a.b, during the deployment averaged 4 cm s -~ with a maximum of 13 cm s -~.

Site C lies one degree north of the equator under the high productivity region associated with the equatorial divergence, Local relief is slight, though the site lies in a basin between a number of seamounts (MURRAY, 1987). The calcite-rich sediments at this site are typical of the carbonate ooze belt of the central equatorial Pacific. Current velocities 45 m.a.b, from December 1982 to February 1984 averaged 5.6 cm s -~ with a maximum of 24.6 cm s -1.

The moorings were equipped with OSU single-cone sediment traps (1 m 2 baffled collection area) that collect sequential samples, and double-cone Soutar traps (0.5 m 2 baffled collection area, each cone) that collect a single sample for the deployment period. Fluxes from the OSU traps are yearly averages to allow comparison with the average flux of the Soutar cones. Current velocities were measured with Aanderaa current meters.

The sample cups on the OSU traps were poisoned with sodium azide to prevent bacteriological decay. The Soutar traps were poisoned with buffered formalin. Recent comparisons between formalin and azide poisoned cups on a Soutar trap show agreement within 5% for concentrations of the biogenic components (DYMOND, unpublished data).

Trap samples were sieved through a 1 mm nylon mesh, freeze dried, ground, and split for analysis. Salt corrections were made by difference in weight after freeze drying assuming a salinity of 35 ppt. A1, Ba, Si, Fe and Sc concentrations were determined by Instrumental Neutron Activation Analysis (INAA). Organic carbon and CaCO3

Resuspension of rebound particles 63

measurements were made with a LECO carbon analyser using a phosphoric acid/wet oxidation procedure (WELIKY et al . , 1983). Biogenic opal was calculated by assuming all particulate A1 is detrital aluminosolicate with a Si/A1 value of 3.0, and all remaining Si is in opal (FISCHER, 1984). Since non-biogenic Si in our samples never exceeds 10% of total Si, this normative determination of opal is insensitive to errors in the assumed Si/AI ratio of aluminosilicates. Details of the analytical techniques have been reported previously (FIscHER, 1984). The fluxes for site H differ slightly from those previously reported (F~sCHER, 1984) due to recalculation of the trap collection areas.

Microfossil and size fraction data were obtained by wet sieving a 1/32 split of the <1 mm trap samples, and surface sediment samples, into four size fractions (<38 ~tm, 38--63 pm, 63-150 ~tm and >150 tam). Each size fraction was examined under a binocular microscope, then dried and weighed. CaCO3 and opal in each sample were extracted using leach solutions of 1.0 M NaOAc, 0.516 M HOAc and 2.0 M Na2CO3, respectively (MURRAY, 1987). Ca and Si were measured in the supernatants by nitrous oxide- acetylene Atomic Adsorption analysis. Weight percent Si was converted to weight percent opal on the assumption that all Si was from biogenic opal with a formula of SiO2 • 0.7H20.

Sediment samples were taken from the upper 5 mm of sediments and from material suspended in the overlying water from Soutar box cores, which recover a relatively undisturbed sediment surface (SouTAR et al . , 1981). These samples contain the highest concentration of the biogenic components of sediment samples at each site, representing a transition zone between the settling particles and accumulating sediment (SoUTAR et

al. , 1981). The primary particle flux at each site was estimated by extrapolating the flux measured

in a trap judged free of resuspended material. This judgement was made on the basis of the stability of the refractory element fluxes from the mid-water maximum flux (Fig. 3). Extrapolations were made using a first order decay equation and reaction constants estimated for the mid-water column:

p i = pioe-kZ ' (1)

where P~ is the measured flux of component i in the primary flux trap (at site H the 3075 m trap, at site C the 3495 m trap, at site M the 1565 m trap), pi is the extrapolated flux, k is the reaction rate constant for i, and z is the difference in depth between the primary flux trap and the depth of extrapolation (Tables 1 and 2). Reaction rate

Table 1. Primary fluxes and their extrapolations to the sediment surface at MANOP sites C, H and M

Organic CaCO 3 Opal carbon

Depth Site (m) (lag cm 2 y a)

AI

H

M

Primary trap 3495 119 1230 700 1.83 Sediment surface 4470 108 1160 635 1.83



64 I. WALSH et al.

Table 2. First order dissolution~degradation constants used to extrapolate primary fluxes at M A N O P sites C, H, and M

Site Organic carbon CaCO3 Opal

C 1.0 x 10 ~ 6.3 x 10 5 1.0 x 10 ~ H 3.5 x 10 4 1.2 x 10 ~ 2.2 x 10 4 M 2.2 x 10 ~ 8.5 x 105 1.2 x 10 ~

All values are in units of m -~. Site M values are the average of mid-water column values at MANOP sites C, H, and S (WALSH et al., 1987).

constants were estimated at sites C and H by solving equation (1) for k using biogenic flux data normalized to the AI flux for two traps in the mid-water column (WALsH et al., 1988). For the refractory elements (A1, Fe, Ba and Sc) primary particle fluxes were assumed constant for all depths below the primary flux trap. The bulk flux was extrapolated by subtracting the loss of opal, CaCO3 and organic matter assuming organic carbon is present as CH20.

At site M, because all of the traps near the bottom showed elevated fluxes of refractory elements, the extrapolations were made from the 1565 m trap using the average rate constant from the mid-water data at MANOP sites H, C and S (WALSrt et al., 1988).

Resuspended sediment model At all three sites the near-bottom traps measured higher bulk fluxes than the

extrapolated primary flux (Fig. 4). Flux increases were measured in the biogenic components as well as in the refractory elements. The percentage of the total particulate flux contributed by the primary flux decreased towards the bottom at H and M, indicating greater resuspension input closer to the bottom. At site C the fraction of the total particulate flux contributed by the primary flux is similar in both traps. The difference between the sites in this respect may be because the deepest traps at M and H were only 30 m.a.b, compared to 175 m.a.b, at C.

Assuming that the observed flux increases were the result of resuspension of surface sediment, the flux of AI in the near-bottom traps can be partitioned according to the relationship:

A1 m = Alp + {[A1]~ x S}, (2)

where A1 m is the measured flux of AI in gg cm -2 y-~. Alp is the extrapolated primary flux of A1 in gg cm -2 y-1 and is assumed equal to the flux of A1 measured in the primary trap (Table 1). [Al]s is the concentration of A1 in the surface sediments in gg mg -1, and S is the bulk flux of resuspended sediment in mg cm -2 y-l.

As all but S are known, this equation can be solved for S for any trap. We used A1 for the partitioning because it is an effective tracer of refractory aluminosilicates (SPENCER et al., 1978; DYMOND, 1981; HONJO, 1982). The sediment flux of any other component (i) is found by multiplying S by the concentration of that element in the sediment. Thus:

M i = pi + {[i]s x S} + R i, (3)

where M i is the measured flux of component i in gg cm -2 y-l, U is the extrapolated primary flux of i in gg cm -2 y-I given by equation (1), [i]s is the concentration of i in the surface sediment in lag mg -I, S is the bulk sediment flux in mg cm -2 y-a determined from


1.0

0.2

1.0

0.2

1.0

0.I

Site C " ~ ( , ~

~295m

' \ \ x 27°°" \+ ~ /x

\

0.01 I I I I I I I I Bu lk Corg CaCO 3 Opal Ba AI Fe Sc

Fig. 4. Flux ratio plot of extrapolated primary fluxes divided by the measured fluxes for near- bottom traps at MANOP sites C, H, and M. Included is the total particulate flux and major and minor components. Ratios of <1 indicate that the primary flux is insufficient to account for the

measured flux.

equation (2), and R i is the residual flux of i in lag cm -2 y-1. The contribution of i from resuspended sediment is given by the term ([i]s × S).

This approach shows that most of the refractory material could be supplied by resuspended sediments and the primary flux (Fig. 5). However, substantial (>20%) portions of the measured biogenic fluxes cannot be accounted for by this model. Additionally, the Fe and Sc fluxes are overestimated ([P + SlIM > 1) in the 3225 m trap (350 m.a.b.) at H, and the 2700 m trap (380 m.a.b.) at M, indicating that the amount of resuspended sediment is overestimated. This is likely due to attributing all of the Al increase to resuspended sediment.

At site C, Fe and CaCO3 fluxes are overestimated by the model, indicating that the resuspended sediment input to the traps is also overestimated at site C.

Hydraulic ffactionation of resuspended sediment based on particle size and density, either through differential resuspension or sorting by settling of larger/denser particles during transport could complicate this model. We have examined the effects of hydraulic fractionation by measuring the concentration of CaCO3 and opal in different size

66 I. WALSH et al.

2.0

1.0

0.5

2.0

I.O

0.5

2.(

1.0

0.5

429.5 m~

4170m

Site C

J 2 2 5 m,~

Site H

2 7 0 0 m

Site M

I I I I I I I I Bulk Corg CoCO 3 Opal Ba AI Fe Sc

Fig. 5. Flux ratio plot of the sum of the extrapolated primary and resuspended surface sediment fluxes divided by the measured fluxes in the near-bottom traps at MANOP sites C, H, and M. Ratios <1 indicate that the primary and the resuspended sediment flux are insufficient to account for the measured flux. Ratios >1 result from an overestimation of the flux supplied by the resuspended sediment and primary material. Note that A1 is fixed at one by the modeL's

assumptions.

fractions of the surface sediments at sites C and H. Generally, the weight percentages of opal and CaCO3 within a sediment size fraction increase with increasing size. Tests of foraminifera and radiolarians make up >98% by weight of the >150 gm fraction in the site C sediment samples, and radiolarian tests dominate the course fraction in the calcite- free surface sediments at site H. Since currents preferentially transport small particles, resuspended sediment would be depleted in opal and CaCO3 relative to the clay fraction. Thus, hydraulic fractionation cannot explain an underestimated flux of CaCO3 or opal.

On the other hand, processes of hydraulic fractionation may explain why the model overestimates the CaCO3 flux at site C. The >150 p.m size fraction at site C comprises 15% of the surface sediment and is >90% CaCO3 . The >150 gm fraction requires a larger bed shear stress to be resuspended than the clay fraction in unconsolidated sediment (MILLER, 1977). Therefore the CaCO3 to A1 ratio in the resuspended material may be less than that for the sediment as a whole, resulting in the model overestimating the resuspended sediment flux of CaCO3.

Effects of other assumptions in the model have been considered. Dissolution and oxidation reactions during resuspension will result only in decreasing biogenic/Al ratios as compared to the sediment ratios, and therefore these reactions cannot explain the model 's underestimation of the biogenic fluxes. CaCO3 produced by benthic foraminifera


is included in the surface sediment analysis, and comprises <5% of the surface sediment foraminifera assemblage. Chemolithotrophy by bacteria may be a source of organic carbon (KARL et al., 1984), but has not been investigated in the lower water column and cannot explain enhanced opal or CaCO3 fluxes.

Resuspension of sediment and rebound particles

To account for the flux components not explained by the sediment resuspension mode, we consider resuspension of rebound particles (Rv). We assume that A1 and organic carbon are fully accounted for by the primary flux, resuspended sediment, and resuspended rebound particles. Organic carbon was chosen because of its low concentration in the surface sediments at all three sites. For AI:

Alto = Alp + {[AI]~ × S} + AI . . . . (4)

where Alm, Alp, [Al]s, and S are defined as in equation (2). The contribution of A1 from resuspended sediment is given by the term {[A1]s x S}, and Alrv is the flux of A1 contributed by the resuspended rebound particles in lag cm -~- y-1. For organic carbon:

Cm = Cp + {[C]s × S} + {(C/A1)r,, × Air,,}, (5)

where Cm is the measured flux of organic carbon in I~g c m-2 y-l, Cp is the primary flux of organic carbon in lag cm -z y-a, [C]s is the concentration of organic carbon in the surface sediments in lag mg -1, S is the bulk flux of the sediment component in mg cm -2 y-l, (C/Al)rv is the weight ratio of organic carbon to A1 in the rebound particles, and Alr,, is the flux of AI contributed by the resuspended rebound particles in p.g cm -2 y l . We assume that (C/A1)rv is the same as (C/AI) of the primary flux extrapolated to the sediment surface (Table 1). The contribution of C from resuspended sediment is given by the term ([C]s × S]. The contribution of C from resuspended rebound particles is given by the term {(C/A1)rv x Alto.}. As only S and Alr, are unknown in equations (3) and (4), the equations can be solved simultaneously.

For all other flux components (i) the partitioning equation is

M i = U + {[i]~ × S} + {(i/Al)rv × Air,,} + R i, (6)

where M', P/, [i]~, S, and R i are defined as in equation (3), (i/A1),.v is the ratio of i to AI in the resuspended rebound particles (in this model, the primary flux extrapolated to the sediment surface), and AI~,, is the flux of A1 in ~tg cm -2 y-1 contributed by the resuspended rebound particles determined using equations (4) and (5). The contribution of i from resuspended sediment is given by the term {[i]s × S}. The contribution of i from resuspended rebound particles is given by the term {(//Al)rv × Ally}.

This model rather closely matches the observed fluxes of most components (compare Fig. 6 to Figs 4 and 5), but generally overestimates the bulk and biogenic fluxes. These subtle discrepancies could result from the use of a yearly average flux to model a seasonally dominated phenomena (BILLET et al., 1983; LAMPITr, 1985). This may be particularly important for CaCO3 and opal since the ratio of production by opal vs CaCO3 forming organisms changes through the year (PIS1AS et al., 1986; MURRAY, 1987). Additionally, our model assumes that the composition of the resuspended rebound particles does not change from the time they first reach the sediment to the time they are collected in the sediment trap. The validity of this assumption depends on the length of

68 I. WALSH et al.

2 . 0

1.0

0.5

2.C

I.O

\4/70m S i t e C

//5545m

~5225m o.~ Site H

2700m~ _ _ v ~ l 0 ~ ~ ~ X ~ ~

5050 m

0.5 S i t e M

I [ I I I I I I Bulk Corg CQCO 5 Opal Ba AI Fe Sc

Fig. 6. Flux ratio plot of the sum of the extrapolated primary, resuspended surface sediments, and resuspended rebound particle fluxes divided by the measured fluxes. Note that both organic

carbon (Corg) and AI are fixed at one by the model's assumptions.

time between the initial deposition and resuspension, and the distance between the area of resuspension and the traps. The assumption of no dissolution or oxidation is probably adequate for the traps 30 m.a.b. However, for the traps further above the bottom there may be compositional differences due to patchiness in the primary flux and oxidation and dissolution during the time of resuspension, as it is unlikely that rebound particles from the site itself were mixed directly upwards for 300 m. Rather, it is more likely that this material comes from some distance away, resuspended from topographic highs or dispersed along pycnoclinal surfaces via detached bottom mixed layers (ARMI and D'ASARO, 1980).

Since we have recorded increases in biogenic fluxes in near-bottom traps from three sites with different productivity, topographic, and sediment characteristics, and resuspension of rebound particles has been observed in the northeast Atlantic (BILLET et al., 1983; LAMPITT, 1985) and the eastern United States continental rise and slope (B1sCAYE e t

al., 1988), it appears likely that this is a common phenomenon in ocean basins. Resuspension of rebound particles probably results in dissolution and degradation

prior to incorporation in the sediment, as the resuspended rebound particles are fully exposed to the oxygenated and corrosive bottom water. Thus the primary flux of particles to the sediment surface is probably not equal to the flux across the sediment-water interface.


Acknowledgements--The authors thank Bruce Finney and Robert Collier for helpful discussions. We also thank Erwin Suess, Mitchell Lyle and Will Gardner for valuable criticisms and comments. Analytical and laboratory assistance was provided by Roberta Conard, Patricia Collier and Susan Miller. This work was funded by NSF Grant OCE-8315259.

REFERENCES

ARMI L. and E. D'ASARO (1980) Flow structures of the benthic ocean. Journal of Geophysical Reasearch, 85, 469-484.

BILLET D. S. M., R. S. LAMPITr, A. L. RICE and R. F. C. MANTOURA (1983) Seasonal sedimentation of phytoplankton to the deep-sea benthos. Nature, 302,520-522.

BISCAYE P. E., R. F. ANDERSON and B. L. DECK (1988) Fluxes of particles and constituents to the eastern United States continental slope and rise: SEEP--I. Continental Shelf Research, in press.

DEUSER W. G. (1986) Seasonal and interannual variations in deep-water particle fluxes in the Sargasso Sea and their relation to surface hydrography. Deep-Sea Research, 33,225-246.

DEUSER W. G., E. H. ROSS and R. F. ANDERSON (1981) Seasonality in the supply of sediment to the deep Sargasso Sea and implications for the rapid transfer of matter to the deep oceans. Deep-Sea Research, 28, 495-505.

DYMOND J. (1981) Geochemistry of Nazca plate surface sediments: An evaluation of hydrothermal, biogenic, detrital, and hydrogenous souces. In: NAZCA Plate: crustal formation and Andean convergence, L. D. KUt, M, J. DYMOND, D. J. DASCtf and D. M. HUSSONG, editors, Geological Society of America Memoir 154, pp. 133-173.

FISCHER K. (1984) Particle fluxes in the eastern tropical Pacific Ocean--Sources and processes. Ph.D. thesis, Oregon State Univeristy, 225 pp.

FOWLER S. W. and L. F. SMALL (1972) Sinking rates of euphausiid fecal pellets. Limnology and Oceanography, 17,293-296.

GARDNER W. D., L. G. SULLIVAN and E. M. THORNDIKE (1984) Long-term photographic, current, and nephelometer observations of manganese nodule environments in the Pacific. Earth and Planetary Science Letters, 70, 95-109.

GARDNER W. D., J. B. SOUTHARD and C. D. HOLLISTER (1985) Sedimentation and resuspension in the northwest Atlantic. Marine Geology, 65, 199-242.

HONJO S. (1978) Sedimentation of materials in the Sargasso Sea at a 5,367 m station. Journal of Marine Research, 36,469-492.

HONJO S. (1982) Seasonality and interaction of biogenic and lithogenic particulate flux at the Panama Basin. Science, 218, 883-884.

KARL D. M., G. A. KNAUER, J. H. MARTIN and B. B. WARD (1984) Bacterial chemolithotrophy in the ocean is associated with sinking particles. Nature, 309, 54-56.

KOMAR P. D., A. P. MORSE, L. F. SMALL and S. W. FOWLER (1981) An analysis of sinking rates of natural copepod and euphausiid fecal pellets. Limnology and Oceanography, 26, 172-180.

LAMPITT R. S. (1985) Evidence for the seasonal deposition of detritus to the deep-sea floor and its subsequent resuspension. Deep-Sea Research, 22,885-897.

MCCAvt~ I. N. (1975) Vertical flux of particles in the ocean. Deep-Sea Research, 22,491-502. MILLER M. C. (1977) Threshold of sediment motion under unidirectional currents. Sedimentology, 24,

507-527. MURRAY D. W. (1987) Spatial and temporal variations in sediment accumulation in the central tropical Pacific.

Ph.D. thesis, Oregon State University, 345 pp. PISIAS N. G., D. W. MURRAY and A. K. ROELOFS (1986) Radiolarian and silicoflagellate response to

oceanographic changes associated with the 1983 El Nifio. Nature, 320, 259-262. ROWE G. T. and W. D. GARDNER (1979) Sedimentation rates in the slope water of the northwest Atlantic

Ocean measured directly with sediment traps. Journal of Marine Research, 37,581-600. SCHRADER H. J. (1981) Fecal pellets: role in sedimentation of pelagic diatoms. Science, 174, 55-57. SHANKS A. L. and J. D. TRENT (1980) Marine snow: sinking rates and potential role in vertical flux. Deep-Sea

Research, 27, 137-143. SMALL L. F., S. W. FOWLER and M. Y. UNLU (1979) Sinking rates of natural copepod fecal pellets. Marine

Biology, 51,233-241. SOUTAR A., S. JOHNSON, K. FISCHER and J. DYMOND (1981) Sampling the sediment-water interface---evidence

for an organic rich surface layer. EOS, 62, 905. SPENCER D. W., P. G. BREWER, A. FLEER, S. HONJO, S. KRISHNASWAMI and Y. NOZAKI (1978) Chemical

fluxes from a sediment trap experiment in the deep Sargasso Sea. Journal of Marine Research, 36, 493-523.

SUESS E. (1980) Particulate organic carbon flux in the oceans--surface productivity and oxygen utilization. Nature, 288,260-263.

70 I. WALSH et al.

URRERE M. A. and G. A. KNAUER (1981) Zooplankton fecal pellet fluxes and vertical transport of particulate organic material in the pelagic environment. Journal of Plankton Research, 3, 369-388.

WALSH I., J. DYMOND and R. COLLIER (1988) Rates of recycling of biogenic components of settling particles in the ocean derived from sediment trap experiments. Deep-Sea Research, 35, 43-58.

WELIKY K., E. SUESS, C. A. UNGERER, P. J. MULLER and K. FISCHER (1983) Problems with accurate carbon measurements in marine sediments and water column particulates: a new approach. Limnology and Oceanography, 28, 1252-1259.

WISHNER K. F. (1980a) The biomass of the deep-sea benthopelagic plankton. Deep-Sea Research, 27,203-216. WISHNER K. F. (1980b) Aspects of the community ecology of deep-sea benthopelagic plankton, with special

attention to Gymnopleid Copepods. Marine Biology, 60, 179-187.

evidence for resuspension of rebound particles from near-bottom sediment traps

Documents