Isotopic signatures (14C, 13C, 15N) as tracers of sources and cycling of soluble and particulate organic matter in the Santa Monica Basin, California

Measurements of Δ14C, δ13C and δ15N are reported for dissolved (plus colloidal), suspended and sinking particulate, and total sedimentary organic matter in the Santa Monica Basin (mid-basin and shelf sites) on CaBS cruises 1, 3, 4, 7, 8 and 10. These isotopic signatures were indicative of the following processes occurring within the basin regime: (1) terrestrial inputs of organic matter to the sinking and suspended particulate organic matter were of the order of 10% or less, and as high as 25% for the sedimentary organic matter; (2) Δ14C values of the UV-oxidizable dissolved organic matter below 5m were similar to those measured in open ocean waters, while the Δ14C values in the suspended, and, to a lesser degree, in the sinking organic matter decreased markedly with depth. This latter decrease was primarily attributed to episodic resuspension of shelf and slope sedimentary organic matter, and secondarily to natural and/or anthropogenic petroleum inputs; and (3) the isotopic signatures of the UV-oxidizable dissolved organic matter, coupled with total dissolved carbon and amino acid and carbohydrate concentrations were strikingly similar in the deep basin and at an open-ocean site, suggesting a common history for the subsurface (>300m) and deep water at both sites. In addition, total mass and organic carbon and nitrogen fluxes from five particle trap deployments are described in detail. Mass fluxes increased with depth, especially on the shelf, suggesting that particle input from the basin slopes may reach the mid-basin site. We conclude that there is minimal perturbation of all organic phases in the basin from terrestrial sources, and that the properties of the UV-oxidizable dissolved organic matter are not greatly influenced by particles of local origin. © 1992.

Origin of A~4C gradients in mid-basin and shelf organic phases (water column) 280 5.3 Origin of the sedimentary organic matter 282 5.4 Phase inter-relationships deduced from isotopic signatures 283 5.5 C/N ratios in the sinking organic matter 283 5.6 Comparison of isotopic signatures and other properties between the Santa Monica Basin and the North Central Pacific gyre 284 6. Summary 286 7.

INTRODUCTION
The principal goal of the California Basin Study (CABS) is to delineate the sources, transformation, dispersal and ultimate fate of particulate organic particles derived from marine and terrestrial sources. An understanding of particle dynamics is essential in predicting the impact of energy-derived, possibly deleterious, substances on the southern California borderland regime. Inherent in understanding particle dynamics are concurrent studies on all organic phases in the basin system. This includes sinking and suspended particulate, dissolved (plus colloidal), and sedimentary organic matter. Especially important are transformations between these phases. This report focuses on the use of multiple isotopic signatures (14C, 13C, ~SN) to trace the origin and cycling of organic phases in the Santa Monica-San Pedro basin system. Radiocarbon (14C) is derived from two sources: natural cosmogenic production in the stratosphere, and atmospheric bomb production during the late 1950s and early 1960s. The high levels of ~ac in surface dissolved inorganic carbon (DIC) subsequent to 1960 (NYDAL, LOVESETH and SKOGSETH, 1980) result in organic matter produced in surface waters during photosynthesis also containing high levels of ~4C (A~4Ca = +50 to +150%0). Thus, surface-derived organic matter photosynthetically-fixed since 1957, and found in sub-surface waters and in the sediments, contains bomb ~ac. Lower levels of '4C (A~4C<-75% o) indicate pre-bomb levels (BERGER, TAYLOR and LIBBY, 1966;ROBINSON and THOMPSON, 1981), which in turn, suggests a minimal turnover of carbon in the past 35 years. Thus, 14C is useful as both a long and short-term tracer because of the natural and bomb-produced components, respectively. The high '4C activity of recently synthesized (<35y) terrestrial organic matter (N4C --+200%0) can also be reflected in marine organic phases impacted from terrestrial (and/or atmospheric) sources.
Carbon-13 (~3C)a in conjunction with nitrogen-15 (8~5N) ' signatures reflect both the origin and fractionation of organic matter as it flows through ecosystems and undergoes phase transformations. Measurements of both 5~3C and 8~5N on the same organic components are obligatory in sorting out organic matter cycles because of the wide spread in the ~13C and 5~5N signatures for the various terrestrial and marine end-members and within specific trophic levels. Generally speaking, 13C isotopic compositions have proven most valuable for delineating dietary sources and less so in following isotopic fractionation along food-chains, while 15N reflects trophic levels and specific processes such as nitrogen fixation and denitrification and particle transformations (see, for example, FRY and SHERR, 1984;FRY, 1988;ALTABET, 1988). Table 1 summarizes previously-published A14C, ~13C and 8~5N values measured in the Southern California Bight regime for various organic pools and end-members which are relevant for interpreting the measurements presented here for the Santa Monica Basin. Some ~13C and 8~SN data are omitted for the sedimentary column below the upper 0-1cm horizon.

2.BACKGROUND
It is evident that A14C values reported to date are sparse for all the organic phases. Few DIC 14C activities were measured in California coastal waters from 1959 to 1987, and no values are available for sewage or river (storm water) soluble or particulate organic components. Thus it is difficult to rigorously assign A14C values to these important end-members. The A14C measurements of EMERY and BRAY (1962) for sedimentary organic carbon (SOC) are primarily from subsurface samples where surficial A14C values represent extrapolations. Nonetheless, these numbers are valuable in that they represent pre-bomb activities.
There are no reported measurements of 8~5N in dissolved organic nitrogen (DON) -off southern California or anywhere else owing to the difficulty of performing such measurements -and only a few results for suspended particulate organic matter (POM) at a location seaward of the Southern California Bight (SAINO and HATTORI, 1987). There are, however, some 8~aC and 5~5N measurements which can be used to assign limits to the relevant end-member values necessary for estimating marine versus terrestrial contributions to the Santa Monica mid-basin organic phases. a A14 C (per mil deviation from the 14C activity of 19th century wood) = 814C°/oo-216~3C (%0) + 25.0] [1 + ~514C (%0) 10-3]; where the Age (BP) = 8033 In [1 + A14C 10-3] "1, and 813C (%) = [(13C/12C)Rample . (t 3C/12C)poB] / ( ~ 3C/12C)po B X 10 a. All A~4C V alues are corrected for: 81 ~C of the sample, age to 1950AD (STUIVER and POLACH, 1977), and relevant blanks for quartz filters and/or combustion tubes.   Table 2 lists the cruises, dates and samples taken. Note that all mid-basin and shelf samples are from the same two locations for all cruises. The box core (BC-89, CABS-l) was taken ~ 10km southwest of the main mid-basin station.

Sampling
Dissolved organic carbon (DOC) and DIC were sampled with a new, 12 1 Go-Flo Niskin bottle on CABS-7 (trace ~4C contaminations of DOC samples from CABS-4 may have resulted from a "used" 30 1 Niskin). Aliquots for DOC were passed directly from the Niskin through ignited GF/ C glass fiber filters (nominal pore diameter -1.0~m) into 1 gallon brown glass bottles and frozen at -20°C. Unfiltered, 500ml samples for DIC were taken initially from the same Niskin bottle by overflowing three successive volumes in a 500ml brown bottle. These samples were preserved with 0.5ml of saturated HgCl 2 solution and stored at 3°C.
Suspended particulate organic matter was collected using an in situ filtration system (WILLIAMS, CARLUCCI and OLSON, 1980) based on an earlier version of LAIRD, JONES and JENTSCH (1967). The current system employs two Pelagic Electronics, sealed in situ pumps powered by two marine high-amperage, 12-volt lead-acid batteries to pull seawater through four 142mm diameter Whatman, quartz fiber filters (nominal pore diameter 0.8~tm) pre-ignited at 800°C) held in PVC holders. Flow rates are initially 10-15 1 m -~ and total filtration volumes are measured using individual ARAD ~ flowmeters. The pumps are turned on and off at depth with a mechanical timer, and at deployment are filled with distilled water which is displaced by ambient seawater after dissolution of a soluble wafer (Lifesaver). The intake water is prefiltered through 505~m pore diameter nitex netting. Contamination is minimized by coating the supporting framework with heat-cured epoxy resin and using stainless steel hardware. The amounts of seawater filtered per single 142ram filter unit varied from 6501 at 100m to 1440 1 at depth for 3 to 6h of pumping, respectively. After deployment, each filter pad was folded and frozen at -20°C in cleaned glass jars.
Sinking particles were collected using paired, fiberglass"Soutar" cones, each with an effective mouth opening of 0.062m 2 (see BRULAND, FRANKS, LANDING and SOUTAR, 1981, for basic trap design). The mouth of each trap was baffled with a cellular grid (lcm x I cm x 4cm deep) to restrict internal water flow and constrain particle capture to the trap mouth. The lucite cod-end collectors (2 l) were charged with 30g of solid HgC12 in a perforated plastic bottle prior to deployment, and filled with surface seawater at launching. On retrieval, the cod-ends were capped and stored at 3-5°C prior to splitting and processing the sample (usually 1 to 14 days later, depending on cruise duration).
All of the above shipboard sampling procedures were done on the R/V New Horizon, where 14C productivity studies were routinely carried out. Thus, it was necessary to rigorously clean the samplers with dilute HC1, methanol and distilled water and to use plastic gloves, etc., when appropriate. No ~4C contamination was evident in the A~4C results documented in this manuscript.

Analyses
DOC was oxidized to CO 2 by high-energy UV irradiation following procedures given in WILLIAMS and GORDON (1970) as modified in DRUFFEL, WILLIAMS, ROBERTSON, GRIFFIN, JULL and DONAHUE (1989). Five litre, thawed seawater samples are acidified to pH = 2.2-2.5 with H3PO 4, sparged free of DIC with clean N 2, saturated with 02, and oxidized for 6h with a 1200W Conrad-Hanovia UV lamp at 70°C. The resultant CO 2 is purified and its 14C content determined at the NSF-TAMS facility at University of Arizona after conversion to graphite. 813C (+0.1°/oo) values for the same DOC were measured on a V.G. Micromass 602E isotope ratio mass spectrometer at the Woods Hole Oceanographic Institution (WHOI).
DIC was collected as CO 2 by acidifying 150ml of the preserved samples with 3 ml of 5% H3PO 4 (final pH = 2.3-2.4) in a CO2-free closed system. The CO 2 was collected and purified as with the DOC-CO 2./~14C and 813C measurements were done at the University of Arizona and WHOI, respectively.

Suspended, filterable POM [POC (susp) and PON (susp)]
At WHOI the thawed quartz fiber filters were acidified with 5ml of 1% HzPO 4 and dried in vacuo overnight to remove inorganic carbon and water. The filters were then oxidized for 1 h with CuO (wire form) and Ag foil in double quartz tubes at 850°C following the procedure of MINAGAWA, WINTER and KAPLAN (1984). The resultant CO 2 derived from the POC (susp) was converted to graphite (JULL, DONAHUE, HATHEWAY, LINICK and TOOLIN, 1986) at the University of Arizona and the N4C and 813C values measured at the University of Arizona and WHOI, respectively. N4C values for CO 2 obtained from all closed tube combustions were corrected for blank CO 2 from the quartz combustion tubes and quartz filters. Blanks composed 0.3 to 5.7% of the sample CO 2 volumes. A~4C values of blank CO 2 samples averaged-480%. 8~3C and 8~5N values for replicate filters were also measured by D. Winter at the Institute of Geophysics and Planetary Physics, University of California, Los Angeles (UCLA). These latter values are those reported in Tables 3, 4 and 5, where the precision is +0.1%o for both the 8~3C and 8~5N measurements.

Sinking POM [POC (sink)] and PON [PON(sink)]
Approximately 1200-1500ml of the supernatant liquid in the lucite cod-ends was syphoned off, and the remaining 500-800ml mixed and subsampled for isotope analyses, mass and C and N flux measurements (SIO), and trace organic compounds (M.I. VENKATESAN, UCLA). Aliquots for A~4C, 8~3C, and 8~5N analyses were acidified to a stable pH = 1.5-3 to remove carbonates and filtered through 42mm diameter GF/C glass fibre filters using a ~15psi vacuum. The collected solids were briefly rinsed with double-distilled water, dried, and pulverized in a mortar. Portions of these homogenized solids were: (1) oxidized to CO 2 at WHOI as for the POC (susp), and the CO 2 -~4C content measured at the University of Arizona; and (2) oxidized at UCLA and their 813C and 815N signatures measured. Additional aliquots of the original cod-end samples were analyzed in duplicate for their total mass (dry weight) by drying at 60°C to constant weight (NELSON, BEERS, EPPLEY, JACKSON, McCARTHY and SOUTAR, 1987) and in duplicate for their CHN content (SHARP, 1974) using either a Hewlett-Packard Model 185B or a Perkin-Elmer Model 2400 CHN analyzer.
The sediment core was extruded from its lucite liner after slightly thawing on the outside, rinsed briefly with double-distilled water, and cut into 1.25cm sections with a hacksaw. The outer edges were trimmed, and the centre portions acidified to pH = 2.6-3.0 by titration with successive portions of 50% H3PO 4. These slurries were dried to constant weight at 60°C, pulverized, and aliquots analyzed for A~4C, 813C, and 8~5N [as for POC(sink)].
All of the above operations concerned with isotopic measurements were carried out at either the Mt Soledad Radiocarbon Laboratory (SIO) or the WHOI Radiocarbon Laboratory. Both of these facilities are free of t4C contamination. TABLE 3./il3C, 615N, A~'C and totalorganic carbon data: POC(susp) and PON(susp), IX)C, and DIC.
A~4C v flues are corrected for 6t3C and age to 1950 AD, according to STUIVER and POLACH (1977).  Tables 3, 4 and 5 contain the original A14C, ~tsC and 815N measurements for the dissolved, sinking, suspended, and sedimentary organic phases in the Santa Monica Basin. These same isotopic signatures are plotted in Figs 1,2 and 3. Mass and organic C and N fluxes are contained in Tables 6, 7   ~Calculated from cumulative mass (rag cm-2)/mass flux (rag cm-2y -~) using mass flux = 14. Img cm "~ y-I for box core no. BC-89 (see HUH et al, 1990). Cumulative mass values for 1.25cm sections interpolated graphically from 1.0cm sections measured by HUH, ZAHNLE and SMALL (1986). Ages are mean values for 1.25cm sections + 3 years.
The ~4C values in DIC (DP4C) at 3-5m in surface waters south of San Diego and in the Santa Monica Basin (Tables I and 3) were 109 and 94°/ooin 1980 and 1987, respectively, compared to 140-Z-_10% o at 3m in the North Central Pacific in (WILLIAMS and DRUFFEL, 1987DRUFFEL and WILLIAMS, 1990). These lower DII4C values in the coastal areas probably result from both sporadic upwelling and the shoaling of 100m isopycnal surfaces, resulting in low ~4C activity, subsurface DIC reading the surface. This low t4C activity is then reflected in the carbon fixed by photosynthesis which is the source of suspended and sinking particulate organic matter and, eventually, DOC. The bomb:4C signal is apparenteven in the limited measurements of DI14C from surface waters off Southem California (Tables 1 and 3), where A~4C increased from ~ -90°/ooin 1959 to ~+140°/ooin 1969 and then decreased to ~90% oin 1987. The corresponding average A~4C values for atmospheric CO 2 during these same years were approximately 230, 550 and 200% o, respectively, reflecting the bomb inputs in the late 1950s and early 1960s (LEVIN, KROMER, SCHOCH-FISCHER, BRUNS, MUNNICH, BERDAN, VOGEL and MUNNICH, 1985;MAN-NING, LOWE, MELHUISH, SPARKS, WALLACE, BRENNINKMELIER and McGILL, 1990). If we assume that photosynthetically-derived terrestrial inputs to the Southern California borderland have the same ~4C activity as atmospheric CO v then the A14C values for marine and terrestrial end-members are ~90 and ~2000/o0, respectively. At the other extreme, petroleum hydrocarbons (seeps, spills, chemicals) will probably contain little or no ~4C, resulting in A~4C values approaching or equal to -1000%o.   The differences in ml4C values (Table 3) between DOl4C and DI14C (DOC-DIC) are -291%o at 5m and -293°/0o at 850m, differences which are of the same order as those found in the NCP at these depths (WILLIAMS and DRUFFEL, 1987). This suggests that this fraction of the DOC (oxidizable by UV radiation) is recycling throughout the oceanic water column including the contiguous, semi-isolated coastal basins. However, the ~4C activity of the DOC (or some fraction thereof) oxidizable only by high temperature combustion (HTC) (SUGIMURA and SUZUKI, 1988) must be significantly higher than that of the UV-oxidizable fraction to account for the observed covariance between DOCrrrc concentration with apparent oxygen utilization (AOU) in some oceanic regimes. There is some indirect evidence for this "extra" DOC having an elevated 14C content , but there are no hard data. The significance of this paradox of the dissolved/colloidal organic matter (see WILLIAMS and DRUFFEL, 1988) to organic carbon cycling in coastal basins in general, and the Santa Monica Basin in particular, is in assessing the role of DOC in balancing particle fluxes, in situ biochemical oxidation, and sedimentary organic matter burial against carbon fixed by photosynthesis and/or entering the system from non-marine sources (JACKSON, AZAM, CARLUCCI, EPPLEY, FINNEY, GORSLINE, HICKEY, HUH, JAHNKE, KAPLAN, LANDRY, SMALL, VENKATESAN, WILLIAMS and WONG, 1989).
Perhaps the most important impact of the "old" recycled DOC is via its possible association with both suspended and sinking POC which may result in low A~4C values for these particulate phases at depth (Fig. 1) and hence, ambiguity in reading the bomb-~4C signal or radiocarbon clock of these phases vis-3-vis their origin and eventual burial in the sediments.
6t3C values for UV-oxidizable DOC (DOC-UV) in the basins (Table 3) vary little with depth or location and only slightly between cruises -the mean values for all depths for CABS-4 and CABS-7 samples were -21.4 + .5°oo (n=6) and -21.8 + 2°0o (n=5) respectively. The average 8 ~3 value for the North Central Pacific (3-900m, n=17) in 1985 and 1987 was -21.1 + .3% o(n=22) (WILLIAMS and DRUFFEL, 1987, unpublished data); and for the SIO pier and San Diego Trough area (Table 1) in 1967-1968was -22.5°/o0 (WILLIAMS and GORDON, 1970; results which could be 1%o more negative because of earlier procedural problems). The DOC-UV concentrations for CABS-4 and CABS-7 (Table 3) were nearly identical at all mid-basin and shelf depths, and were ~4~tmol higher at depth than the North Central Pacific DOC-UV values. Thus, the concept of a common origin for the UV-oxidizable fraction of DOC present in both the open ocean and coastal basin is strengthened by 613C values and total DOC-UV content-in addition to the A~4C signatures.

Suspended particulate organic matter: POC(susp) and PON(susp)
Suspended POM, whose carbon concentration is of the order of 1% of the DOC-UV (Table  3), has estimated residence times in the upper mixed layer (photic zone) of 20-39 days ) and 3 to >100 days (EPPLEY, RENGER and BETZER, 1983) based on new production and sinking fluxes, alternatively 4 to 7 days based on thorium isotope measurements (HUH and BEASLEY, 1987). In the deeper basin waters, these POC (susp) residence times may be of the order of 5 to 10 years, as has been reported for thorium isotope measurements in the open ocean (BACON and ANDERSON, 1982). Particles collected in the SMB by in situ filtration were nominally > 1 ~tm in size but would include finer particles scavenged on larger ones (McCAVE, 1984) and trapped on clogged filters. Thus, the origin of suspended POM is not necessarily localized to the basin system, but is a combination of: (1) POM advected into the basin by surface currents and deep water exchange across the sills; (2) POM derived from the basin slopes through resuspension and subsequent incorporation into various levels within the water column; and (3) POM resulting from fractionation of sinking POM into finer, suspended particles. Transmissometer measurements within the SMB water column (NELSON et al, 1987;HUH, SMALL, NIEMNIL, FINNEY, HICKEY, KACHEL, GORSLINE and WILLIAMS, 1990;B. laCKEY, personal communication) indicate that there are no permanent nephaloid layers beneath the euphoric zone but do not preclude sporadic injections of suspended particles at depth from the basin shelf or slopes. Surface and deep current measurements and tracer experiments give residence times for water parcels (with their associated suspended POM) ranging from 3-20d in the upper 200m, up to several years for sporadic renewal of deep basin waters (JACKSON et al, 1989). These renewal times are short compared to estimated suspended particle residence times in the surface mixed layer, suggesting caution in relating chemical and isotopic properties of POC (susp) and PON (susp) to locally-sinking POM or other indigenous sources of suspended organic particles. Clearly, advected particles may dominate those of local origin, especially in deeper waters.
A14C values for POC (susp) decrease significantly with depth (Table 3, Fig. 1) with a minimum value of-46%o at 700m. Neither the 813C signature nor the POC (susp) concentration is strikingly different at 700m from those same parameters measured in the 510 and 850m samples, suggesting that this particular sample was not significantly "contaminated" with oil or other hydrocarbons. Without replicate samples, this low A14C value cannot be attributable to the filtration of a spurious cloud of"old" suspended particles; nor can it arise from resuspended surficial sedimentary organic particles whose A14C values were -10 and -2 l°/oo in the upper 2.5cm of bulk sediment (Table 5). Two facts are evident from the POC (susp) ~4C activities: (1) Reduced carbon, recently fixed photosynthetically at the surface, whose DI4C value should be similar to the surface water DI~4C value (94°/oo), does not constitute all of the suspended POC. Low 14C-containing organic particles and/or DOC must contribute to the POC (susp) A~4C values below 100m assumaing that we are not dealing with spurious oil contamination; and (2) the pOI4C (susp) activities are lower than those for the sinking POC collected on the same cruise at the same depths (Fig. 1). This might seem to support longer turnover times for the POC (susp) than for the POC (sink), but this is only true if inclusion of "old" sedimentary particles and/or DOC are excluded, and if the POC (susp) has a common origin with the POC (sink).
813C and 815N values for POC (susp) do not reflect the isotopic signatures in either POC (sink) or DOC (Figs 2 and 3). ~13C values for POC (susp) are more negative by ~1% othan for DOC at comparable depths (Table 3), as was earlier observed by WILLIAMS and GORDON (1970) (Table  1). JEFFREY, PFLAUM, BROOKS and SACKEVr (1983), who measured 8t3C values for POC (susp) in numerous shallow profiles (<400m) from various oceans, suggested that biological reworking ofPOC (susp) in the pycnocline caused fractionation of POC-St3C towards lighter isotopic values so that 813C values increased with depth. The opposite was found in the SMB where the two shallow samples (mid-basin and shelf) were lighter in ~3C than samples at 500m and below by 3.0%o and 1.39%o, respectively. This depletion in ~13C suggests a terrestrial contribution is possible; but, if so, none of the shallow POC (sink) ~13C values (Table 4) is isotopically lighter than those measured in the deeper samples. Flocculents from the ouffalls and suspended particles from Los Angeles River storm waters (Table l) have 8'aC values which vary from -23.7 to -24.9%o, and 815N values from 2.4 to 3.0%o, but there is no compelling evidence from 8~5N values for PON (susp) supporting an appreciable contribution of ouffall-derived suspensoids (this general question of terrestrial contributions to all organic phases in the Santa Monica Basin system is amplified in the Discussion section). The approximately 2%o increase in 8~5N at 700 and 850m relative to the 80, 100 and 510m samples (Table 3, Fig.,2) follows previous measurements of lighter 8~5N values in PON (susp) in the euphoric zone than at depth (ALTABET, 1988).

Sinking particulate organic matter [POC(sink) and PON(sink)], and total mass and organic carbon and nitrogen fluxes
There are two caveats concerning the Soutar particle trap deployments. The f'trst is that the shallower traps (100 and 200m) collected significantly fewer organic particles than smallerdiameter, multicylinder "Vertex" traps (WELSCHMEYER and LORENZEN, 1985;SMALL et al, 1989) deployed at these same shallower depths (see LANDRY, PETERSON and ANDREWS, 1992). This apparent "under-collection" by the shallow Soutar cones is probably related to motioninduced orientation problems in the upper 200m. We consider that the deeper traps (500-700 and 850m) did collect representative flux samples based on the following observations. JAHNKE (1990) has shown that organic carbon burial rates correspond to sinking carbon fluxes at 850m adjusted for sediment oxidant consumption (see also JACKSON et al, 1988). Also ratios of organic carbon fluxes to total mass fluxes (from Tables 6 and 8) decrease from 0.13-0.20 at 100 and 200m to 0.055-0.084 at 850m, prim adly because of a decrease in organic matter trapped at depth coupled with an increase in total mass flux with depth (Fig.4). These ratios do not preclude under-trapping of sinking organic matter in the upper 200m with Soutar traps, but do show an approximately twofold decrease in the amount of organic C trapped relative to total mass at 850m compared to 100 or 200m.
The second caveat is that the 100, 200 and 350m traps contained variable but often significant numbers of swimmers, primarily copepod nauplii, all of which could not be separated from the sinking particles. Such swimmers will have A14C values close to that of surface DIC, and ~13C and filSN values representative of marine end-members, so the isotopic signatures for sinking organic matter will be biased in these shallow samples. The swimmers will also lead to the total organic C and N fluxes for passively sinking particles being over-estimated. Conclusions based solely on fluxes and total organic matter collected in the shallow traps are most subject to these errors.
There are no N4C values reported previously for POC (sink) in coastal basins. In the North Central Pacific, pO14C (sink) at 4120 and 5120m were 85 _+ 7 and 114 _+ 7%o, respectively (DRUFFEL and WILLIAMS, 1990). These values are higher than those measured at depth in the SMB, possibly reflecting the higher surface DI~4C content in the North Central Pacific.
N4C values for POC (sink) from five CaBS cruises (Table 4 and Fig. l) generally decrease to 700m at the mid-basin station with no further reduction in ~4C at 850m. All of the N4C values are greater than -75°oo, the maximum pre-bomb value for surface DIC prior to 1950 (AD); or, conversely, all of the POC (sink) samples contain bomb-14C -as would be expected. The interesting point is that this diminuition of ~4C activity with depth requires that some fraction of the POC (sink) contains organic carbon which is not surface-derived and/or derived from surface waters >35 years ago. At 850m (Fig. l) the PO14C (susp) and PO~4C (sink) values converge to a mean N4C = 27 _+ 21 o/oo (range = 2-52%o, n=4) from a mean value of 96_+ 14%o (range = 78-110°/ oo, n=4) at 80-200m, the same value as DI~4C at 5m. The general correspondence between the ~4C contents of sinking and suspended POC suggests a common source and similar histories for these phases, although it does not rule out their having disparate sources with closely related isotopic signatures.
One clue to the source of"older" sinking POC is the lower A~4C values for all of the particle trap POM samples from the shelf. A14C decreased from a mean of 67 + 7°oo at 100m to 39 _+ 6°0o at 200m and -13 + 14%o at 350m, ~50m above the seafloor. Perhaps, fortuitously, the differences between the surface and deep samples at mid-basin (69%0) and on the shelf (80°oo) are not significantly different. This might be interpreted as a "common source", e.g. resuspended, low-~4C activity, shelf sedimentary organic matter (SOM), mixing with recently-fixed organic carbon, where the amount of common-source POM decreases with distance from its origin on the shelf. Since persistent nephaloid layers have not been observed at depth in mid-basin waters, these low A14C values for POC (sink) on the shelf are more likely from resuspended 14C-poor SOM, while another process (e.g. adsorption of"old" DOC) may also occur with the mid-basin samples.
~3C and 815N values for sinking POM (Table 4, Figs 2 and 3) vary significantly between cruises, between depths, and between the mid-basin and shelf stations. The 515N values are particularly variable at mid-basin where 815N at 850m (cruise CABS-3) was 5.50%0, and that at 500m (cruise CABS-7) was 10.20%0. Only the CABS-4 deployment showed smooth 8~3C and 8~5N profiles (Figs 2 and 3). ff all measurements at each depth are averaged, vertical changes in 5~5N are much less pronounced (see Fig.3 and JACKSON et al, 1989), and the average 815N values at 100 and 500m (8.87 and 8.42%0) are similar to those measured at the same two depths (8.0 and 8.0%0, respectively) for samples from two, 2.5 day, free-floating conical trap deployments in 1983 (NELSON et al, 1987).
It is apparent from Fig.3 that the vertical profiles of 513C (with the exception of 700 and 850m on CABS-3) are about 1%o heavier in 13C on cruises CABS-3, 7 and 10 than on CABS-4, both at mid-basin and on the shelf. No similar differences occur with the fi~SN values. 815N on cruise CABS-4 is not systematically lower as would be expected if the lighter 5~3C values on CABS-4 originated from terrestrial inputs. It is difficult to attribute variations in 813C values, be they dissolved, suspended or sinking organic phases to terrestrial sources, since 8~3C values for the terrestrial end-members (-23.7 to -26.4%0, Table 1) are only about 3%0 or less depleted in ~3C relative to the dissolved and particulate organic phases in question. Variations in the fi~3C signatures for the sinking organic matter are most likely a function of differing 813C values for the various sources compromising the sinking POM, and do not arise from trophic level fractionation processes. These sources include (see Table 1): phyto-and zooplankters (fi~3C ----19 to -22°/0o); kelp (fit3C = -14 to -21%o); sole and prawns (~13C ~ -16 to -18°/oo), resuspended sedimentary organic matter (813C = -20 to -23% o), and sewage or storm water particles (~3C = -23.7 to -26.4%0). The uniform, CaB S-4 8t3C signatures for sinking POM may have been a result of predominantly one source of particles (e.g. copepod-derived organic matter) being trapped during this deployment.
Some variation in the 813C signatures could have arisen from changes in CO s (aq) concentrations (RAU, TAKAHASHI and DES MARAIS, 1989) which would alter the 513C values of the organic phases. However, the small interannual variations in CO 2 (aq) or temperature in this region would result only in small changes in the 8~3C signatures.
How then, are the corresponding ~tSN signatures interpreted? The large, and seemingly unrelated excursions in 8~SN between deep and shallow trap samples and between deployments must be a function of isotopic fractionation during phase changes, and reutilization and recycling in the euphotic zone coupled with a sporadic input of sewage particulates. Such processes as: (1) ~SN enrichment in higher trophic levels of the order of 3.5%0 per trophic level (FRY, 1988 and references therein); (2) ~SN depletion in fecal pellets (~ 1-2%o) relative to the parent zooplankton body tissue coupled with an approximate 2%0 fecal pellet enrichment over the apparent zooplankton food source (ALTABET and SMALL, 1990); (3) variable 8~SN values in photosynthetically fixed NO 3-and NH4 + reflecting variations in NO 3-and NH4 + during denitrification and recycling of nitrogen in the euphotic zone (LIU and KAPLAN, 1989;CHECKLEY and ENTZEROTH, 1985;CHECKLEY and MILLER, 1989); and (4) ~SN fractionation between sinking and suspended particles (ALTABET, 1988). Any or all of these processes could be operating within the complex nitrogen cycle in the SMB system, making it difficult, if not impossible, to relate any specific ~15N profile of PON (susp) to any one (or more) direct sources or processes. What is evident is that on cruise CABS-3, ~SN-depleted particles (presumably sewage-derived) dominated the trapped particles at 350m on the shelf (8~5N = 3.55) and even appeared to have entered the 850m trap at the mid-basin station. Such "sewage" contributions were not apparent in the other mid-basin PON (susp) profiles (Fig.3), but could have enhanced the regular depletion of 13N in the 350m traps relative to the 100 and 200m samples on the shelf. This depletion could also arise from resuspension of sedimentary organic particles whose 613C values (e.g. 6.4 at 0-1.25cm, Table 5 for core BC-89) were more similar to the PON (sink) 615N values at 350m on the shelf (and at 850m at mid-basin) than to, for example, zooplankton tissue (Table 1). The significant increases in mass and C and N fluxes on the shelf, and in mass fluxes at mid-basin (Fig.4), support the hypothesis for such a sedimentary contribution from the shelf/slope regions; so do the lower 14C activities in the 350 and 700 and 850m trapped particles and in the surface sedimentary organic matter.

Sedimentary organic matter (SOC)
AI4C, ~13C and 515N values for the upper 20cm of box core No. BC-89 are listed in Table 5 and depicted in Figs 1 and 2.
The AI4C values measured in the surficial sediments of the SMB (-10 + 6 and -21 + 6%0, 0 to 2.5cm) clearly show the presence of bomb 14C. As expected, a AlnC result obtained for a grab sample collected during the pre-bomb period of 1956 in the SMB (-147 + 19%0, 1280 + 150y BP;EMERY and BRAY, 1962) was much lower than our 1985 values. There is, however, reasonable agreement between the 1956 A14C values at lcm and 15cm (-147 + 19%o and -258 + 19%o, respectively, EMERY and BRAY, 1962) and our 1985 Nac values at 7cm and 13cm (-154 + 5%0 and -307 + 5%0). Our A14C value at 20cm appears to be anomalously low (-416 + 4%0) when compared with any of EMERY and BRAY'S results for the upper 90cm in three cores from the Santa Monica Basin and the adjoining San pedro Basin. This may be evidence of "old", redeposited carbon that EMERY and BRAY (1962) postulated to be in surficial SOC from the basins off Southern California. For the Santa Monica and San Pedro basins, EMERY and BRAY (1962) calculated six sedimentation rates ranging from 38 to 145cm 10-3y ~ (mean = 66cm 10-3y 1) based on the difference in 14C activities between a surface and deep (>300cm) N4C measurement. These sedimentation rates are similar to the 31cm 103y -I calculated for core BC-89 from 2~°Pb measurements in the top 20cm (Table 5, 20cm x 103/642y). The ~4C sedimentation rate for Core BC-89 is much less (~4cm 10-3y -~) based on near-surface (6.2 to 20cm) N4C results. This latter rate may be unique to core BC-89, as large variability in AI4C results were also observed by EMERY and BRAY (1962) in near-surface SOC.
813C and fi~SN values for SOC (Table 5, Fig.2) do not show extreme excursions within the sedimentary column. ~3C (-21.79 to -22.62°/oo, mean = -22.11%o) and 5~5N (6.09 to 8.74%0, mean = 6.98%0) values fall within the envelope of data points on a regression plot of 8~3C vs. 615N for sedimentary organic matter from continental shelf sediments of the northeast Pacific (PETERS, SWEENEY and KAPLAN, 1978), but are not themselves significantly correlated (Fig.5F). There are no striking correlations of 8~3C or 8~5N with sedimentation events which have occurred during the past 500 years; e.g. the decline in sedimentation rates in core BC-89 which occurred 200 years ago (HUH, ZAHNLE and SMALL, 1990) at the time of an extensive turbidite intrusion into the basin (MALOUTA, GORSLINE and THORNTON, 1981), or with episodic, high, storm-derived run-off at 10-30y intervals (MALOUTA, GORSLINE and THORNTON, 1981). Nor are there any tight correlations of 8~3C or 8~5N with past productivity maxima (SMITH and EPPLEY, 1982) or total SOC content for the 1922-1979 period. Note that total SOC concentration profiles at mid-basin and shelf sites show a maxima in SOC content in 1965SOC content in -1970, when inputs of wastewater particulates from the White's Point outfall reached a maximum EGANHOUSE and KAPLAN, 1988). There is also no reflection in the isotopic ratios for core BC-89 of the SOC and SON concentration maxima and 5~3C and filSN minima observed at 6-10cm in a core taken at 60m bathymetric depth on the San Pedro shelf in 1981 (EGANHOUSE and KAPLAN, 1988). These SOC and SON maxima and isotopic minima on the shelf are a direct result of the maximum discharge of terrestrially-derived sewage particles, which occurred in the late 1960s. Thus, the terrestrial C and N isotopic signatures have been poorly preserved in the mid-basin, sedimentary column (at least in 1.25cm sections), implying that significant biochemical oxidation-solubilization of sewage particulates occurs en route to their eventual sequestration in the sediments.

Terrestrial (sewage, storm run-off industrial organics) contribution to the SMB organic phases
The average 6~3C and 5~5N signatures for White's Point sewage flocculents are -24 and 2.5%0, respectively (Table 1). fi~3C in Los Angeles river storm water suspended solids is -25°/oo (Table 1). No values for fi~SN in river solids, or for A~4C in sewage or runoff organics are available. A14C values in natural oil seeps, anthropogenic petroleum products and spills, and petroleumderived industrial chemicals will be -1000°/oo, and in modern organic matter (grass, leaves, paper, etc.) will be taken as 2000/00, equivalent to that of contemporary atmospheric CO 2 (Table 1).
A rough estimate of the contribution of sewage-derived organic flocculents to the SMB organic phases can be calculated using the two-source mixing equation: where X s and X M are the fractions of sewage and marine end-members in the SMB organic phases having the appropriate measured isotopic signatures (8~SCsMB and 5~SNsMB). Table 9 lists the percentages of sewage-derived organics in the SMB organic phases and includes end-members and organic phase isotopic signatures. The latter values are means of both mid-basin and shelf measurements.  I  I  I  I  I  I  I  I  I  I   -21.6 (n---1 I) -- Table 3  The percentage contribution of sewage flocculents, X s, to the SMB organic phases (Table 9) is 3-8% for the sinking and suspended POM (excluding the 72% value for POC (susp) based on ~13C) and 25-28% for the surficial SOM. The wider spread in the ~13C end-members makes the percentages calculated using 615N values more credible. The 25% calculated for surficial SOM corresponds to the value found by EGANHOUSE and KAPLAN (1988) for SOM at 23-34cm depth in a core from the San Pedro shelf at 60m, where the corresponding percentages for the 0 to 23cm sections varied from -40% at 0-2cm to 95% at 2-4cm. Thus, the contribution of sewage flocculents to the water column organic phases in the mid-basin and shelf waters is probably less than 10%, while the surficial, mid-basin SOM could contain -25% terrestrial components, assuming the terrestrial end-member assignments are valid.
There are also indications from 8~5N values for the POC (sink) of episodic inputs of terrestrial POM at 350m on the shelf (815N = 3.55°/oo, CABS-3), 850m mid-basin (815N = 5.50°/oo, CABS-3), and 850m, mid-basin (815N = 5.97°/oo, CABS-10). Using the 815N s and ~15N M end-member values from Table 9, the estimated contribution of sewage to the POM particle trap contents at 350 and at 850m for CABS-3 and 850m for CABS-10 are 83%, 50% and 42%, respectively. However such high sewage inputs are not reflected in the 813C signatures of the POC (sink), which suggests that there is reworking and fractionation of the sewage-derived POC without corresponding changes in the ~15N values (see MYERS, 1974, andSWEENEY, KALIL and. Independent of the isotopic evidence, the doubling in the mean mass flux between depths of 100 and 200m to 350m on the shelf, coupled with a 40% increase in the corresponding CN flux (Fig.4), implies that there is export of POM, including terrestrial components, across the shelf. As well as the increased mass flux on the shelf, there is a 40% increase in total mass flux between depths of 100 to 850m at mid-basin (Fig.4), although its origins could be from the seaward slopes of the basin as well as the coastal shelf and slope. These inputs augment the POM derived directly from marine organisms.
With respect to the A14C signatures, sewage inputs cannot be assessed without knowing the A14C value for sewage flocculents and solubles. Sewage is composed mainly of faeces, cellulose (toilet paper) and variable quantities of chloroform-soluble hydrocarbons (R. EGANHOUSE, personal communication). Its A~4C signature will be a mix of recent (A14C = 200%0) and ancient (AlnC =-1000°/oo) components. The hydrocarbon fraction, however, could contribute to lowering A~4C values in the water column and sedimentary organics (see below).

Origin of Al4C gradients in mid-basin and shelf organic phases (water column)
The most obvious carbon source resulting in the diminution of 14C activities in the DOC, POC (susp) and POC (sink) is petroleum derived from natural seeps, sewage discharges, oil spills and atmospheric inputs.
As stated earlier, the DOC A14C profile can be primarily attributed to the influx of open-ocean water into the basin, rather than to soluble hydrocarbons, whose concentrations (in the boiling point range of n-tetradecane to n-dotriacontane) in 1-3m surface waters and in samples collected 10m off the bottom have been reported to be 0.03 to 80~tg1-1 (~0.001 to 3gtmol C 1-1) (RISEBROUGH and BURLINGAME, 1978). Assuming a mean total DOC concentration of44gtmol 1 1 at 500 to 850m (Table 3), and a surface (5m) A~4C value of-197%0 for the source DOC, then it would require an additional 10gtmol (23% of the total UV oxidizable DOC!) of hydrocarbon-C per liter (A14C = -1000% o) to attain the measured value of ~ -470°/oo (Table 3) for the deep water DOC. However, the resultant 813C value of this DOC would then be -23.3°/oorather than the mean measured value of-21.75 (Table 3), assuming a 813C = -28°/oo for petroleum products.
With respect to the sinking particles, CRISP, BRENNER, VENKATESAN, RUTH and KAPLAN (1979) reported a total hydrocarbon flux of 0.07 lg (~0.006mol C) m2y -1 for the Santa Monica Basin at ~850m. The mean C flux at 850m (Table 7)( is ~ 15g m-2y -~ and 45g m-2y= -I at 350m on the shelf, which gives hydrocarbon contents of -0.24% (mid-basin) and 0.08% (shelf) assuming the same flux of hydrocarbons at both sites. If the AI4C value for the POC sinking from the surface was equal to surface DIC (94°/oo), then the A~4C for POC (sink) at 850m (or at 350m on the shelf) would be lowered only to 91.3°/oo (and to 93.1 °/oo on the shelf) with admixture of 0.24% (or 0.08% on the shelf) petroleum hydrocarbons (A~4C = -1000% o). The measured means of A14C at 850m and 350m, however, are 41 and -13%o, respectively (Table 4).
A similar calculation for POC (susp) can be made assuming (1) the A14C source for POC (susp) has a value of 94%0 with a mean C concentration of 0.27ktmol 1 1 at 500-850m (Table 3), and (2) the mean concentration of hydrocarbon -C in suspended particles is 0.005ktmol 1 1, based on data from RISEBROUGH and BURLINGAME (1978) and DeLAPPE, RISEBROUGH and WALKER (1983). These assumptions give a POC (susp) 14C activity of 74°/oo versus the mean measured AI4C value of --3%0 at 510 to 850m (mid-basin).
These crude estimates of the contribution of petroleum to the organic phases in mid-basin and shelf waters while not precluding it's presence in these phases, do suggest that the contribution is minimal (< 1%) based on the data available for hydrocarbon contents of DOC, POC (susp) and POC (sink) in the basin.
If natural oil seeps and/or anthropogenic petroleum inputs are not responsible for the diminuation of ~4C activities with depth, then the most plausible explanation (as mentioned earlier) is that there are contributions of low tac activity, surficial sedimentary organic particles which either enter the mid-basin water column sporadically from the basin slopes, or, in the case of the shelf waters, are directly injected into the overlying wters via resuspension by slumping and internal waves. Downslope resuspension of subsurface sediment (5-10cm) injected into the nearbottom water by the activities of benthic animal activity may also contribute significant amounts of low 14C activity SOC. The observed increases in mass and CN fluxes with depth (Tables 6 and 7, Fig.4) are evidence for these processes, most notably at the shelf stations; although the frequency with which these inputs occur, especially the introduction of slope particulates into the deep basin waters, is unknown. The sporadic presence of nephaloid layers in the deep basin waters (B. HICKEY, personal communicaion) lends credence to the interdigitation mechanism.
Assuming the A14C values of the POC (susp) derived from the slope surficial sediments to be -150/00 (the same as the mean value measured for the mid-basin 0-2.5cm SOC, Table 5), and the original source A~4C of POC (susp) to be 940/00 (same as 5m DIC), then the resultant percentage contribution of SOC to the POC (susp) at 510 and 850m is estimated to be 61 and 78%, respectively.
The corresponding percentage contributions of resuspended sedimentary POC to the POC (sink) at 500, 700 and 850m in the basin will be 27, 58 and 50% respectively, where the mean measured A14C values at these depths are 65, 31 and 390/00 respectively (Table 4), and the two source A~4C values are again taken as 95%0 [surface POC (sink)] and -15%0 (sedimentary POC). For the 350m POC (sink) at the shelf stations, the SOC contribution would be 98% of the total POC (sink) to give its resultant mean A14C value of -13% o.
Obviously, the A~4C value of the SOC derived from the basin slopes and/or shelf may be higher or lower than the assumed value of -15%o, depending upon how thick a surficial layer of sedimentary POC was injected into the deep basin or resuspended on the shelf. The A~4C values may also be a function of sedimentary grain size. Coarse particles are rich in fresh plant debris while fine-grained particles contain more fossil organic matter (THOMPSON and EGLINTON, 1978). ThUS one might expect that At4C values would vary at any given sediment depth depending on grain size. This mechanism does not account for the ~4C activity (A14C = -46%0) of the suspended POC at 700m (Table 3). The only explanation for this anomalously low result is contamination with "dead" (petroleum) carbon. An addition of 6.3% "dead" carbon (A14C = -1000) is needed to lower the A~4C of POC (susp) at 700m (assumed to be 200/00) to -46%0. This invokes a spurious contamination event, probably sampling related, as opposed to the unlikely ubiquitous contribution of petroleum to the POC (susp) at 510m and 850m discussed above.
We feel that the estimated 27, 58 and 50% contributions of SOC to the POC (sink) are high, but not unreasonably so, considering the probability of sporadic SOC inputs occurring during the 98-233d particle trap deployments. However, the estimates of 61 and 78% SOC contributions to the POC (susp) are based solely on a single profile with 6h in situ particle filtrations at 510,700 and 850m. This collection may have occurred fortuitously during a period of significant SOC input, so may be unrepresentative of the normal background POC (susp). If a significant fraction of the POC (susp) was derived from near-bottom waters from the slope being advected into the basin, then the resultant isotopic signatures would be strongly influenced by the isotopic characteristics of this advected POC (susp).
There is also the intriguing possibility of adsorption of low 14C activity DOC-UV onto suspended (or sinking) particles which, if it occurred during the 5 to 10 year residence time of the POC (susp) in the water column, could contribute to the gradient in AI4C. It would require 12, 25 and 15 % of the POC (susp) to be adsorbed DOC at 510, 700 and 850m respectively, to give the measured POC (susp) AInc values (27, -46 and 9%0) at these depths (Table 3), assuming the A14C of the surface-derived POC (susp) to be 940o (same as DIC at 5m). This calculation also assumes a A14C of -450°/oo for the DOC-UV at 500m based on the North Central Pacific value at 700m. Again, the SMB 700m POC (susp) A~4C value of -46%0 is suspect (it's A14C value would be 12°/00 assuming a 15 % DOC contribution). Although a 12 to 15 % contribution of DOC to the POC (susp) intuitively seems too high, it only amounts to 0.025 to 0.054txmol (0.3-0.65~tg) C 1 I of the total 0.21 to 0.36~tmol 1-1 of POC (susp), or 0.06 to 0.13% of the total DOC-UV at those depths.

Origin of the sedimentary organic matter
The contribution of sinking POC reaching the ocean floor at mid-basin is 43mg m-2d-~ (Table  8, mean value, n=4) which may be augmented by unknown, but presumably small, quantities of POC (susp) and DOC. JAHNKE (1990) has calculated that 43% of the total organic carbon reaching the seafloor (18.5mg C m-2d ~) is permanently buried. He concludes that carbon burial rates for the Southern Califomia Bight overall are primarily controlled by input rates and not by variations in in situ redox potentials and other preservation variables. This sinking POC includes: (1) terrestrially-derived POC; (2) detritus and associated macro-and micro-organisms derived from the euphoric zone and added during transit to the seafloor; and (3) recycled SOC derived from the basin slopes via episodic injections into the deep basin water.
The contribution of terrestrial organic matter from all sources to the mid-basin sediments, calculated from 6~3C and ~SN signatures, is about 25% (Table 9), based solely on the ~3C and 515N end-member signatures of sewage organics and the surficial SOM. However, the mass emission of sewage carbon (-5 x 107kg y-l) into the Southern California Bight calculated from sedimentary coprostanol contents (VENKATESAN and KAPLAN, 1990) has been estimated to be ~ 103 times less than carbon fixed photosynthetically in the Bight (~ 1.4 x 10 ~°kg y~). This sewage carbon amounts to 1.6% of the mid-basin surficial SOC, which coupled with a 4% organic carbon contribution from terrestrial plants (based on lignin-derived phenol measurements), gives a total of -6% terrestrially-derived organic carbon in the mid-basin sediments, excluding humic substances (VENKATESAN and KAPLAN, 1992). If isotopic fractionation, during in situ oxic mineralization of SOM, discriminates against both 13C and ~SN, it could result in isotopic endmember values 0.5 to 1%o heavier than their initial signatures as given in Table 9, so the contribution of terrestrial organic matter would then be even lower than those values calculated in Table 9. For example, if the ~3C and ~lSN signatures in the 0-3.75cm core section were altered to -21.7 and 8.0°/oofrom -22.2 and 7.0% by mineralization, then the terrestrial contribution would be reduced from 28% to 8% and from 25% to 8%, respectively, based on these heavier 8~3C and ~15N values. If we assume isotopic fractionation occurs, the estimated upper limit to the amount of terrestrially-derived organic carbon in the surficial SOC of 10% is closer to the sewage and land plant inputs calculated by VENKATESAN and KAPLAN (1992).
Assuming that our previous estimate of recycled SOC in the POC (sink) at 850m (50% of the total) is valid, then 40% of the POC (sink) is unrecycled organic carbon of local origin. How much of each of these three fractions is mineralized and how much is buried is largely conjectural. However, if57% (24.5mg C m2d ~) of the organic matter reaching the ocean floor is mineralized (J~E, 1990), then there must be nearly complete oxidation of the sinking particles of local origin, plus oxidation of a significant fraction (-15%) of the "already-oxidized", slope-derived SOC and also of a small amount (-2%) of the terrestrial component of the POC (sink). Neither isotopic signatures nor carbon contents of the POC (sink) and SOC (0-2.5cm) can be used to quantify these fractions, largely because we know neither the true isotopic signatures nor the C and N concentrations of the terrestrial and slope inputs. If, for example, the slope-derived surficial SOC had a A~4C value less than that assumed (-15°/0o) because of contributions of "older" SOC from deeper sediment layers, then the calculated fraction of slope-derived SOC in the POC (sink) would be lower, and the proportion of recycled SOC in the basin sediment (0-2.5cm) reduced. It is clear, however, that the A14C value for the SOC (0-2.5cm) at mid-basin (-15%0) does not arise from bioturbation in these distinctly varved sediments, and hence there has to be admixture ofprebomb organic carbon with the sinking POC (A~4C at 850m = 32%0). In the simplest case, 57% of pre-bomb carbon (A~4C = -75%0) plus 43% modern POC (sink) (ml4C = 32%0) would give an SOC •t4C value of -29%0 for the 0-2.5cm accumulated since 1960. This is of the same order as the observed average AI4C value for this depth horizon (-15%o).
This still leaves the question: What is responsible for these "old" SO~4C ages in a rapidly accumulating sediment that is not appreciably mixed by bioturbation or by ventilation with nearbottom water from burrowing organisms? Radiocarbon decay at 20cm (2~°Pb age since AD 1950 =607 years) is equivalent to a decrease of about 73°/oo in z~4C, and an additional decrease of 75%0 (or 625 yrs) is accounted for in pre-bomb SOC owing to the reservoir age of the surface ocean at this site (BERGER, TAYLOR and LIBBY, 1966;ROBINSON and THOMPSON, 1981). Thus, we can account for a A~4C value for SOC at 20cm of -150%0; instead, our measured value is -416°oo. EMERY and BRAY (1962) suggested that unexpectedly "old" ages for surficial SOC (mean ~4C age of 2500 yrs BP, Atnc = -270°/0o), may result in part from redeposition and mixing of ancient with modern organic carbon. An alternate explanation is to invoke the incorporation of DOC into the POC that sediments onto the sediment/water interface and becomes buried. The large gradient with depth observed in POC Aa4C results suggests that "old" carbon in some form is being incorporated into POC in the water column. Figure 5 portrays regressions between ml4C, ~13C and 8~5N for the organic phases at the midbasin and shelf stations. These number-6-shot patterns, in general, show common, but not distinctive, sources for the organic matter in all phases. The narrow spread in ~3C values precludes any meaningful and distinctive regressions of ~3C vs. A14C for any of the organic phases, and while there is a larger spread in ~tSN values, no distinctive patterns in 8t3N vs A14C regressions emerge which would uniquely characterize the origin(s) of any particular organic phases. There is a tight coupling of 8~3C vs. ~2N only for the SOC. Thus, these isotopic signatures basically reflect the multiplicity of source organic matter to all phases (at least in the water column), and are not specific enough for interpreting phase inter-relationships.

C/N ratios in the sinking organic matter
C/N ratios (by atoms) at mid-basin vary from 6. I to 9.9 (excepting at 200m, CABS-7), and from 7.0 to 8.3 (excepting at 200 and 350m, CABS-7) at the shelf (Table 8, Fig.4). The three exceptionally low ratios (2.6, 2.8, 3.1) cannot be attributed to analytical error, and must have resulted from both reduced carbon flux and increased nitrogen flux; all are lower than C/N ratios in phyto-and zooplankters (C/N = 5-6) and in pure protein (C/N = 3.5; GNAIGER and BITTERLICH,

Z 4C(%o)
-500 -400 -300 -2.~0 -I00 0 +IQO +200 With respect to the more labile DFAA, THAA and TCHO values (Fig.7), however, the close correspondence of their vertical profiles between the SMB and the NCP is more difficult to explain. The DFAA concentrations are low at both sites, where the increase below the sill (740m) in the SMB is a result of denitrification and possible benthic contributions. It appears that higher productivity in the coastal waters is not reflected in elevated concentrations of either THAA or TCHO in the surface or deep basin waters; and, in fact, TCHO in the upper 100m of the SMB is lower than in the NCP.
The A14C values for POC (susp) in the SMB and NCP both decrease with depth, but have significantly lower 14C activities in the basin. Aside from contributions of resuspended POC discussed above, some fraction of the POC (susp) in the basin may have had a lower initial A[4C value because of the lower surface DICJ4C activity (94% oin the basin at 5m versus 140°/ooat 3m in the NCP). This assumes that the source of the POC (susp) was from photosynthetically-fixed DIC whose value was indeed 94%0. Total concentrations of POC (susp) in the basin deep-waters (500-850m) ( Table 3) were 0.21 to 0.36txmol 1-1 compared to values of 0.24 and 0.15~tmol 1-1 at 450 and 900m in the NCP. The corresponding 513C signatures were -22.22 to -23.08°/oo (SMB) and -19.35 and -21.70 at 450 and 900m in the NCP. Unlike the dissolved organic matter, these above comparisons for POC (susp) do not indicate a common origin ofPOC (susp) for the SMB and NCP regimes.

SUMMARY
Isotopic signatures (A14C, ~13C, ~15N) of the soluble and particulate organic phases measured in this study did not prove, nor were they anticipated to be, ideal proxies for unequivocally defining the sources and cycling mechanisms of organic matter within the Santa Monica basin system.
However, these isotopic signatures, coupled with organic matter concentrations and fluxes, did provide clues to the sources of certain organic components and, to some extent, set quantitative limits to these sources. For example: (1) terrestrial inputs of organic matter from both natural and anthropogenic sources were estimated to be <5% for dissolved organic matter, <10% for suspended and sinking particulate organic matter, and <25% for sedimentary organic matter; (2) the isotopic signatures and concentrations of UV-oxidizable DOC were strikingly similar in the deep basin and at an open-ocean site in the North Central Pacific, suggesting a common history for a major fraction of the DOC in the subsurface waters at both sites; and (3) episodic resuspension of"old" (14C-depleted) shelf and/or slope sedimentary organic matter appeared to be reflected in the low ~4C activities of the POC (susp) and POC (sink) in the basin subsurface waters.
There were no significant correlations for the A~4C, 8~3C and ~SN signatures between the suspended and sinking particulate organic matter, sedimentary organic matter, and soluble organic matter which would have been useful in delineating phase transformations among these organic pools. The multiplicity of fractionation processes within and between organic phases, plus the wide ranges in end-member isotopic signatures, makes it difficult, if not impossible, to rely upon isotopic measurements alone for future studies of this nature; studies which must couple isotopic measurements with chemical analysis of process-oriented tracers (e.g. organic biomarkers).