Importance of isotope measurements in marine organic geochemistry

Abstract We present here some specific areas of research, prefaced by brief and selected reviews, that address the cycling of organic matter in the marine environment. These are areas in which isotopic signatures will provide vital answers or clues.


INTRODUCTION
The largest exchangeable organic carbon reservoir on Earth is in the form of dissolved organic matter (DOM) in seawater. Less than 10% of this DOM has been characterized either with regard to organic compound classes or specific organic molecules. Moreover, little is known about the sources of marine DOM. For instance, it remains a mystery whether or not an appreciable part of this DOM is of terrestrial origin, and if so, whether it is significantly altered with time. Isotopic signatures (A'4C, 6 ~3C and 6'5N) have proven powerful tracers in helping establish the sources and sinks of DOM and living and detrital particulate organic matter (POM) in the oceans, and in assisting in determining the extent and direction of transformation processes between the soluble and particulate phases. In conjunction with chemical data (i.e. total DOM, oxygen, nutrients, total CO2, alkalinity and salinity), isotopic measurements are essential for assessment of the processes controlling organic matter distributions in the oceans. STABLE CARBON ISOTOPES (6~3C) Recent carbon and nitrogen isotopic measurements have revealed information about the sources of organic matter and the transformation processes that occur in the oceanic water column. Studies of ~3C have been used to distinguish between terrestrial (-30 to -25%o) and marine (-23 to -15%0) sources of POM. However, various factors complicate this method: (1) the dependence of ~3C on CO2~aq) concentrations in the water (Rau et al., 1989b); (2) kinetic effects on ~3C of organisms that grow slowly (i.e. at the base of the euphotic zone) (Bishop et al., 1977); (3) fractionation that occurs in some regimes during diagenesis/remineralization (McNichol et al., 1991); (4) biomagnification of ~3C in the food web, which causes the enrichment of 13C in higher trophic level organisms, and possibly in their excretion products (McConnaughey and McRoy, 1979); (5) the wide range of 613C values in source biocarbon, ranging from -10%o in macroscopic algae to -28%0 in some phytoplanktons. The reader is referred to a comprehensive review by Fry and Sherr (1984) of ~3C measurements as indicators of carbon flow in aquatic environments.
Studies of organic 613C in nearshore sediments have been useful for determining relative quantities of marine-derived vs. terrestrially derived organic carbon (see, for instance, Gearing et al. (1977)). The 613C signature of UVoxidizable DOC is suggestive of a marine origin for oceanic DOM (Williams and Gordon, 1970;. This agrees with the results of Meyers-Schulte and , who used lignin-derived phenols as molecular-level probes of DOM to demonstrate that only about 10% of DOM in open ocean water had its origin from terrestrial sources. Mantoura and Woodward (1983), however, projected a mainly terrestrial origin for DOM in the ocean based on the conservative behavior of DOM in an estuary in the UK. Unequivocal attribution of so much terrestrial or marine components to marine DOM is qualitative at best using 6~3C signatures, as there are large overlaps between the 613C values of terrestrial and marine end-members.
Thus, it is clear that 613C alone cannot unambiguously answer the questions that confront us. Instead, tracers with different input functions must be used in conjunction with ~ ~3C to determine sources and sinks of organic matter and the transformation processes that they undergo in the water column and in the sediment. Two such tracers are radiocarbon (14C) and 15N. RADIOCARBON (At4C) Radiocarbon is derived from two sources: natural cosmogenic production in the stratosphere, and bomb production during the late 1950s and early 1960s. In the dissolved inorganic carbon (DIC) pool, bomb radiocarbon is restricted mainly to the upper 1000m of the water column in non-polar oceans, the exception being North Atlantic Deep Water laden with bomb Iac that has penetrated as far south as 30°N. Organic matter produced in seawater during photosynthesis contains high levels of 14C (A14C ----F 50 to + 150%0) that are retained because of the long half-life (5730 years) of 14C. Thus, surface-derived material found in the deep sea and in the sediments contains bomb 14C. Lower levels of 14C (At4C ~< -50%0) indicate pre-bomb levels, suggesting that there has been minimal turnover of carbon in the past 35 years. Thus, ~4C is useful as both a long-and short-term tracer because of the natural and bomb-produced components, respectively. Hedges et al. (1986) measured radiocarbon levels in suspended particulate organic matter and dissolved humic and fulvic acids transported by the Amazon River and found bomb radiocarbon in all fractions analyzed. This indicates rapid turnover for riverine organic matter, where different values for the coexisting fractions reflect varying degrees of retention by soils in the drainage basin.
On the other hand, marine dissolved organic carbon (DOC) contains very low radiocarbon activities. In the deep North Pacific, A~4C values are relatively constant at -520%0 , whereas the surface values are -150%o. The similarity between the surface-to-deep gradients in the DO~4C and the DII4C activities suggests that bomb radiocarbon has penetrated each of these pools to similar depths of about 1000 m in the North Pacific by 1987. The very old 14C activities of the deep DOC (6000 years B.P.) indicate that at least a portion of the UV-oxidizable organic matter has been cycling within the oceans on time-scales several times that of the oceanic mixing time. With the discovery by Sugimura and Suzuki (1988) that there is an 'extra' fraction of DOC in seawater that is undetected by conventional oxidative methods, the question arises: 'What is the L4C age of this "extra" DOC?' Problems in answering this question include obtaining sufficient oxidative-CO2 (50-100 #g C) for accelerator mass spectrometry (AMS) 14C measurements.
We also must use 14C to date individual compound classes (for instance, lipids, and total hydrolyzable amino acids and carbohydrates) within the dissolved and particulate organic carbon pools, to understand the origins of the various organic constituents within these reservoirs. If these components are surface derived they will contain post-bomb levels of ~4C; if their ~4C activities are low, however, production of these compounds at depths greater than 1000 m will be inferred.
Another intriguing observation is the gradual reduction with depth of ~4C activities in sinking and suspended POC (Druffel and Williams, 1990) and in the organisms that feed on POC (Williams and Linick, 1975;Pearcy and Stuiver, 1983;. As the turnover time of suspended POC (5-10 years; Bacon and Anderson, 1982) is much shorter than the lifetime of bomb ~4C in the oceans (35 years), it is difficult to envision a purely surfacederived source for deep-sea POC. Instead, there must be a deep-derived source of carbon to the POC pool, perhaps mediated by heterotrophic bacterial activity and/or the result of adsorption of DOC onto POC.
The potential to measure the ~4C content in micrograms of carbon is a most exciting development. Dissolved, colloidal and particulate organic phases in seawater, sediment pore-water, river water and precipitation, as well as biocarbon derived from micro-and macroorganisms can now be fractionated into individual compound classes and even into individual molecular compounds, which in turn may be age-dated by AMS. Such data will help resolve such questions as: (1) How rapidly has bomb ~4C been incorporated into deep-sea organic carbon pools?
(2) How important are coagulation and dissolution processes with respect to suspended, sinking and soluble organic phases in the water column?
(3) How much does the total amount of DOM in the deep oceans change with time, if at all, as reflected in the ~4C content of humic substances and/or the UV-oxidizable fraction of DOM?
(4) Is the colloidal fraction older or younger than the truly dissolved organic matter, and are there differences between the tac content of the organic compound classes in these two fractions vis-a-vis the suspended and sinking particles?
(5) Can dietary carbon sources for deep-sea organisms be differentiated via ~4C changes in muscle and visceral organs?
(6) What are the 14C contents of heterotrophic bacteria --do they reflect pathways of specific organic energy sources?
(7) Can the flux of dissolved organic matter into seawater from sediments be evaluated by using A~4C in conjunction with 6~3C and ~15N isotopic signatures? NITROGEN ISOTOPES (615N) Nitrogen isotope ratios (61SN) are used to trace sources and transformation processes of marine organic matter. In a study of 6~SN in sedimentary nitrogen phases in the Santa Barbara Basin, Sweeney and Kaplan (1980a) concluded that most of the dissolved N in pore-waters was derived from planktonic organic matter degraded by bacterial activity and that more than 25% of the total N in the sediments was terrestrially derived. Peters et al. (1978) observed a linear correlation between fi13C and 61SN measured in organic matter separated from nearshore sediments of the Northeast Pacific, and attributed this correlation to the mixing of terrestrial and marine 'end-members' of refractory organic matter. Altabet (1988) observed a difference in the 615N of suspended and sinking POM, and concluded that the net transformation of suspended into sinking particles in the North Atlantic was the result of active processes such as macrozooplankton feeding, as opposed to passive physical effects. Rau et al. (1989a) used 615N and ~I3c of mesopelagic shrimps to illustrate inter-species feeding differences, despite considerable overlap in the reported depth distribution of each species.
In addition to assessing terrestrial inputs and food chain trophodynamics, the measurement of 6~5N in the total DOM and individual compound classes of the DOM should be an important tracer, with ~4C, in determining the origin of DOM. Do 6~5N values for DOM reflect those in the corresponding suspended and sinking POM or are they more closely related to 6~5N values of surface micro-and/or macroplanktonic organisms or to the surface DOM? There are no measurements of 6 lSN in DOM as yet, and these determinations should be a high priority.
There is great potential for understanding the fate of organic matter in the oceans using the multiple isotope tracer approach. Whereas 14C integrates the depths in the water column from where the organic matter was derived, the number of trophic levels and the degree of fractionation that has occurred are obtained from the fiLSN and 613C signatures, respectively.

OTHER ISOTOPES
Other isotopes are useful 'clocks' of processes in the upper ocean. For example, Lal et al. (1988) used 33p and 32p activities measured in DOM and POM to demonstrate that these tracers are suitable for determining phosphorus pathways through the pelagic food chain and for determining the flux of phosphorus through the thermocline, averaged over a period of about 1-2 months. Recently, 31p, the stable isotope of P, has been used to characterize organic P compound classes in marine sediments by 31p NMR (nuclear magnetic resonance) spectroscopy (Ingall and Ruttenberg, 1990). This measurement appears to be valuable for detecting trace amounts of anthropogenic organic phosphorus compounds (e.g. pesticides).
Stable sulfur isotope ratios (34S/32S and 634S) are an additional isotopic tracer for organic matter sources, especially in delineating terrestrial vs. marine origins. For example, Peterson and Howarth (1987) have used 634S values of upland plants, Spartina, and plankton to assess the energy sources for macroconsumers in the Sapelo Island salt-marsh estuaries. Sweeney and Kaplan (1980b) found that 634S values reflect both sewage and biogenic (marine bacterial-derived) components in nearshore sediments and suspended detritus from nearshore municipal outfalls on the San Pedro shelf. Organic ~345 values in suspended and sinking particles could be important in oxicanoxic transition zones for determining the extent of bacterial transformation of organic matter at these interfaces by reflecting the bacterial incorporation of sulfur derived from sulfate reduction. CONCLUSIONS We consider important areas of future research involving (~13C, 6~SN and A14C (and, to a lesser degree, ~34S, 31p, 32p and 33p) to include: (1) measurement of these isotopic signatures in particulate, colloidal and soluble organic phases, both the bulk fractions and in specific compound classes and discrete compounds within these fractions; (2) comparison of these same measurements with identical ones on terrestrial organic phases in rivers and within seawater-river-water mixing zones; and (3) measurement of 613C, 6~5N and At4C in bacteria, plankton and higher organisms whose identities, feeding depths and trophic levels are known.
Such measurements will improve resolution of, if not solve, the confusion now posed by scattered and incomplete data sets in the determination of the origins and transformations, and oxidation and preservation of organic matter in the oceans.