SCOPE 21 -The Major Biogeochemical Cycles and Their Interactions  

 ABSTRACT

The sea bottom is a site of mineralization and nutrient regeneration of deposited organic matter. It is also a burial ground for a few percent of that organic matter. This small sink in the element cycles permits a residence time in the ocean of the major biological elements of 10⁵ years. The C:N:P ratio of sedimenting detritus shows a preferential depletion of nitrogen and phosphorus during the early stages of decomposition. Since the detritus is so efficiently remineralized in the sediment, the regeneration of the elements is largely determined by the C:N:P ratio of the deposited detritus. An observed deficit in combined nitrogen is explained by bacterial denitrification. This process causes nitrogen to become a limiting nutrient in many marine ecosystems.

Anaerobic mineralization by sulphate reducing bacteria couples the sulphur cycle closely to the carbon cycle. Up to 50% of the oxidation of organic matter in shelf sediments is carried out by sulphate reducers, but only about 3% by denitrifying bacteria. Sulphate reduction is less significant or even absent in pelagic sediments where denitrification is relatively more important.

The shelf is the main site of mineralization in the sea bottom. Around 2050% of the local net phytoplankton production in shelf areas is deposited on the sediment. The same figure for the deep sea is 1% or less. Thus, 83% of all remineralization in the ocean bottom takes place on the shelf, which constitutes only 8.6% of the total area. The coastal zone supports the most rapid cycling of the major biological elements and may therefore be sensitive to human impact in spite of the huge buffering capacity of the ocean as a whole.

18.1 INTRODUCTION 

Ten billion tons of particulate organic matter are suspended in the world oceans and are constantly sinking towards the sea bottom. The annual input of organic matter to the bottom is five to ten billion tons, i.e., of the same magnitude as the entire pool size. Most of this is deposited in relatively shallow water, on the continental shelves, and only little in the deep sea. The sedimented organic matter represents a small fraction of the total primary production of the ocean, and yet, over a time range of 10 to 100,000 years, it plays an important role in controlling the chemical composition of the sea-water.

Two main functions can be recognized for the marine sediments in the global cycling of the elements. First, they are a site of oxidation and mineralization of organic matter that is produced in the photic surface layers of the ocean. The particulate organic matter is concentrated 10,000100,000 fold in the sediment relative to the sea-water, and consequently the sediments support an intensive biogeochemical activity. The oxidation of the organic matter leads to a regeneration of the nutrients, which can then be recycled into the sea-water. In the process, chemical energy is released and is used by the living, benthic organisms for the production of their biomass. These biogeochemical processes lead to a change in the elemental composition of the sediments and thus to a differential burial of elements into sediments and sedimentary rocks. This second function of the sea bottom as a sink for elements operates over much longer time scales than the dynamic nutrient regeneration catalysed by living organisms.

It is important to emphasize the time scales over which the biogeochemical processes operate. Time scales that are relevant for the study of different specific processes vary from seconds to millions of years. As one example, the residence times in the ocean of the major biological elements, carbon, nitrogen, and phosphorus, are on the order of 100,000 years (Broecker, 1974; Holland, 1978). At the other extreme, the oxygen pool in the pore-water of coastal sediments may have residence times of seconds or minutes (N. P. Revsbech and B. B. Jørgensen, unpublished data). As a general trend, the relevant time scale increases from the shallow waters to the deep sea, and from biological to more geochemical or geological processes, as well as from the micrometre scale dimensions of the micro-environment in which the bacteria live to the metre scale dimensions over which geochemical gradients in pelagic sediments are measured.

Most of the mineralization of organic matter in marine sediments takes place in the shallow water of the epicontinental seas at depths down to 200 m. The primary production of the pelagic regions is almost exclusively remineralized in the water column. Sediments of the continental shelf and of fjords and estuaries are therefore the most important for the element cycles in the ocean. They are also geologically the most important due to their high rate of deposition. These neritic sediments are the sources of sedimentary rocks covering vast areas of the continents, while the pelagic sediments underlying most of the ocean bottom are subducted at the continental margins as a consequence of global plate tectonics.

Information on mineralization processes in sediments can be drawn mainly from two sources. One is the geochemical analysis of pore-waters and solid constituents combined with mass-balance calculations from depositional rates. This approach lends itself to the use of kinetic models in which the diagenetic reaction rates are calculated from observed chemical gradients in combination with estimates of rates of diffusion and burial. The other source of information is based upon experimental rate measurements in which specific dynamic processes such as nitrification or methanogenesis are studied under natural conditions or in the laboratory.

Our knowledge of mineralization in pelagic sediments is largely based on the geochemical evidence for several reasons. In contrast to the heavily bioturbated, coastal sediments, pelagic sediments are often less bioturbated and may approach steady state deposition and diagenesis, which is a prerequisite for most current diagenetic modelling. The vertical gradients of pore-water and sediment chemistry in pelagic sediments can be studied in sufficient detail with the spatial resolution of conventional analytical techniques, whereas the similar gradients in coastal sediments are often very compressed and heterogeneous in space and time. Finally, the value of experimental rate measurements depends on how well they reflect actual in situ process rates. Sampling and laboratory manipulations tend to affect the microbial metabolism and thermodynamic equilibria. This experimental problem increases in proportion to the water depth. A pressure decrease during the sampling of deep-sea sediments may depress the in situ rate of heterotrophic bacterial metabolism by up to 100-fold (Jannasch and Wirsen, 1973). It has also been found to depress the oxygen uptake rate of undisturbed sediment cores 10-fold (Smith, 1978). In situ rate measurements directly on the ocean bottom, which should be the optimal technique for the study of pelagic sediments, require a technology that has only been developed by a few laboratories.

The consequent dominance of the kinetic modelling approach in the study of sediment processes has resulted in a strong emphasis on the chemistry of organic matter and nutrients in the sediment and on the stoichiometry of nutrient regeneration. Less emphasis has been put into the study of regulation mechanisms of the process rates or into the more specific interdependance between different types of microbial metabolism and the chemical diagenesis. An understanding of the latter is slowly arising from experimental studies of sediments where the chemical analyses are combined with microbiological and radiotracer techniques.

18.2 DEPOSITION OF ORGANIC MATTER 

The bulk of organic matter in the sea bottom is derived from the primary production of phytoplankton organisms in the overlying, photic surface layers of the ocean. This has been demonstrated from several independant lines of evidence. The isotopic composition of organic carbon in marine sediments is very similar to that of the local planktonic communities (Degens and Mopper, 1976), which in temperate regions have a ¹³C of about 20‰ (relative to PDB standard). Land plants generally have ¹³C values which are 510‰ lower. Only in estuaries and similar near-shore areas does the isotopic composition indicate a terrigeneous contribution (e.g., Hunt, 1968). The organic chemistry of marine sediments is also typical of planktonic detritus rather than of terrestrial plants. This is shown by the relative compositions of sugars and amino acids in the sediments. These have a high ribose:glucose ratio, in contrast to the complete dominance of glucose in cellulose-rich detritus of land plants. There is also a high hexosamine content derived from zooplankton chitin, and an amino acid spectrum typical of plankton (Degens and Mopper, 1975).

The sedimentation of organic detritus through the water column has been studied intensively in recent years by the use of particle traps (e.g., Hinga et al., 1979; Rowe and Gardner, 1979; Honjo, 1980; Knauer and Martin, 1981). The flux of organic matter to the sediment depends first of all on the primary productivity at the ocean surface and on the water depth. Sediments underlying the highly productive upwelling zones along the continents have much higher mineralization rates and a higher organic content than sediments underlying the low-productive, central regions of the sub-tropical oceans. When the organic flux is recalculated relative to the local phytoplankton productivity, the fraction which reaches the sediments is mainly a function of the water depth (Figure 18.1). According to Figure 18.1, 1050% of the primary production in shelf areas (0200 m depth) are deposited on the bottom. This fraction decreases gradually to reach about 1 % in pelagic sediments at 56,000 m depth.

The oceanic circulation will, over a time scale of a thousand years, slowly transport vast amounts of dissolved organic matter down to the deep sea during which time it is gradually mineralized. Compared to this convective flux, the flux of organic matter to the sea bottom by the sinking of organic particles is a much faster process. This rate is largely independant of the oceanic circulation and the pattern of deposition therefore reflects the local productivity at the ocean surface. An example of that is seen off the coast of Peru. Sediments underlying the highly productive coastal current have an organic content of 27% dry wt, as compared to 1% just west of the current where the waters have a three-fold lower plankton productivity (Degens and Mopper, 1976). The marine sediments are thus good indicators of upwelling areas. A large scale transport of particulate organic matter from shelf waters to depositional centers on the continental slopes has also been suggested as an important process for the allocation of oceanic carbon (Walsh et al., 1981).

The sinking organic particles are mainly derived from phytoplankton cells, from zooplankton exuvia, and from faecal pellets. Some of the larger particles, the marine `snow', are produced by secondary aggregation of smaller particles and macromolecules. The small particles of 0.110 µm diameter have a very low sinking rate and would theoretically take several years to reach the ocean bottom. They are therefore largely mineralized while they are still suspended in the seawater. Larger particles of 50500 µm diameter have a settling time to the sea bottom of only a few days or weeks and they can therefore reach the bottom in a relatively undegraded state. These large particles comprise only 10% of the particulate organic matter in the ocean but contribute 90% of the organic deposition on the sea bottom (e.g., McCave, 1975; Chesselet, 1980).

Figure 18.1 The percentage of the mean annual primary production at the ocean surface that can be collected in sediment traps at different water depths (Redrawn from Suess, 1980) Reproduced by permission from Nature. © 1980 Macmillan Journals Ltd.

18.3 MINERALIZATION PROCESSES 

The particulate organic matter that is sinking through the water column is undergoing rapid decomposition due to autolysis and bacterial attack of soluble and labile cell components. The organic particles therefore arrive at the sea bottom with a different composition to that in the productive surface water. The detritus is relatively depleted in organic nitrogen and phosphorus and its degradeability is gradually decreasing with time and depth in the water. 

This aerobic mineralization continues at the sediment surface through the activity of the benthic micro-organisms and animals that constitute an aerobic detritus food chain. It is typical of these organisms that they carry out, within each individual, a complete oxidation to CO₂ of the organic compounds that they assimilate. Due to their oxygen consumption, the oxic zone constitutes only a rather thin surface layer in shelf sediments, below which the further mineralization takes place by anaerobic processes. Figure 18.2 shows the pathways of carbon during the anaerobic decomposition of detritus. The mineralization here takes place through a sequence of metabolic steps, each of which completes only a partial oxidation of the organic compounds. Small reduced molecules appear in the pore-water as intermediates that are transferred between the different physiological types of organisms of the anaerobic detritus food chains. It is due to the diversification and incomplete metabolism of these organisms that both the nitrogen and the sulphur cycles become such intricate parts of the benthic metabolism. Inorganic compounds of nitrogen and sulphur become carriers of the chemical energy of the detritus, thereby partly separating the energy and the carbon flow of the organic matter.

Figure 18.2 Transformations of organic carbon during anaerobic decomposition in a marine sediment. Pathways leading to the nitrogen and sulphur cycles are also indicated (Redrawn from Fenchel and Jörgensen, 1977) Reproduced by permission of Plenum Publishing Corp.

Figure 18.2 illustrates how the sedimented organic matter may become buried in the anaerobic zone through the bioturbation of the benthic animals and by the deposition of fresh sediment. The hydrolytic and fermentative breakdown of the detritus releases organic molecules that serve primarily as substrates for a vertical sequence of denitrifying, sulphate reducing, and methane producing bacteria. Through their energy metabolism, organic carbon is oxidized to CO₂ and reduced molecules such as N₂, H₂S, and CH₄ are produced. These accumulate in the pore-water and may subsequently diffuse up to the oxic surface layers and, for H₂S and CH₄, undergo oxidation. Organic nitrogen and phosphorus are separated from these redox reactions of organic matter by hydrolytic release in the form of ammonium and phosphate. In contrast to most other constituents of the detritus, phosphate is not involved in important redox reactions in the sediment. The interdependence and regulation of the mineralization processes in different sediment types is quite complex and our understanding of their importance for the regeneration of nutrients is still very incomplete.

Figure 18.3 Schematic distribution of oxidants and their products in the pore waters of marine sediments. Because the chemical species differ more than 100-fold in the depth of their distributions, they are shown in three graphs which have different, relative depth scales as indicated by the broken lines. The absolute values of depths and concentrations will depend on the sediment type

The vertical distribution of the oxidants (electron acceptors) in the pore-water is shown in an idealized diagram in Figure 18.3. The sequence in which they appear indicates that they are utilized by the bacteria in the order of decreasing energy yield per mol organic carbon oxidized: O₂, Mn⁴⁺, NO₃ /NO₂ , Fe³⁺, SO₄²-, and HCO₃ . Thus, reduction of oxygen gives the highest energy yield, while reduction of HCO₃ is barely an exergonic reaction. Examples of the actual distribution of the oxidants in marine sediments are shown in Figure 18.4.

Oxygen occurs at air saturation in sea-water in concentrations of 200300 µmol litre^-1. It is rapidly consumed in the surface layer of shelf sediments and diffuses down to only 110 mm depth (Revsbech et al., 1979,1980). Pelagic sediments with a very low rate of oxygen consumption have oxic pore-water down to half a metre or more (Murray and Grundmanis, 1980).

Nitrate is produced in most sediments by the oxidation of ammonium diffusing upwards into the lower part of the oxygen zone. The regeneration of nitrate and ammonium is in general limiting for the phytoplankton production in the sea. Coastal waters, which are in direct contact with the sediments during circulation, are very low in nitrate. The coastal sediments usually have a sharp nitrate peak close to the sediment surface from which the nitrate is lost upward by diffusion into the water. Downward diffusion is caused by the nitrate consumption by denitrifying bacteria that reduce NO₃ to N₂ below the oxic zone.

Figure 18.4 Distributions of oxygen, nitrate, and sulphate in coastal and in pelagic sediments. (Data from: A; 1: Revsbech et al., 1980; 2: N. P. Revsbech and T. H. Blackburn, unpublished; 3: Murray and Grundmanis, 1980; B; 1: Sørensen et al., 1979; 2: Billen, 1978; 3: Froelich et al., 1979; C; 1: Murray et al., 1978; 2: B. B. Jørgensen, unpublished.: 3: Manheim et al., 1970, and Presley et al., 1970, quoted in Goldhaber and Kaplan, 1974)

In spite of an intensive denitrification, the sediments may thus be a source of nitrate to the sea-water rather than a sink. This is also the case in pelagic sediments, where overlying waters have a relatively high nitrate concentration. In sediments of the eastern equatorial Atlantic, where the nitrate concentrations were 40 µmol litre^-1 in the pore-water and 22 µmol litre^-1 in the water column above, Bender et al. (1977) calculated that 96% of the nitrate produced in the sediment diffused into the water.

Nitrite is an obligate intermediate during nitrification and may itself be reduced by denitrifying bacteria rather than being further oxidized. Mechanisms of N₂ production from NH₄⁺ and organic-N via partly reduced intermediates such as nitrite or hydroxylamine have been suggested by Barnes et al. (1975) and Barnes (1980). The nitrogen gases N₂O and NO are also produced as intermediates in these processes in amounts that are less significant for the N-cycle of the sediments but may be significant as possible sources of atmospheric nitrogen gases (Liss, Chapter 15, this volume). 

Reduction of nitrate and nitrite lead not only to N₂. Ammonium is also produced by a dissimilatory bacterial reduction in marine sediments (Koike and Hattori, 1978; Sørensen, 1978a) but the quantitative importance of this process is not known. The assimilatory reduction of nitrate to organic nitrogen seems to be of little importance, because ammonium in the pore-water is an energetically more favourable nitrogen source and is readily utilized by the bacteria.

Manganic oxides are also reduced just below the oxic zone, and manganous ions appear in the pore-water at concentrations of 1050 µmol litre^-1 (Figure 18.3). The manganous ions diffuse towards the sediment surface where they are either released to the water or re-oxidized and immobilized. By such a reductionoxidation cycle, manganese, together with other metals, is concentrated at the sediment surface, often in the form of iron-manganese nodules that are characteristic for many pelagic red clays.

Ferric iron is present as oxyhydroxides in the surface sediment and becomes reduced to ferrous iron only beneath the nitrate zone. The ferric iron may be reduced by H₂S, either chemically or catalysed by bacteria, or it may be used by anaerobic bacteria as an electron acceptor for respiration or fermentation. The processes involving nitrate, nitrite, manganese, or iron as electron acceptors have been termed suboxic diagenesis (Froelich et al., 1979).

Below the suboxic zone the redox potential is very low and sulphate reduction becomes the dominating early diagenetic process. Sulphate is present in sea-water in a hundred-fold higher concentration than oxygen, and the zone of sulphate reduction is correspondingly two to three orders of magnitude thicker than the oxic zone. In the most organic-rich, coastal sediments sulphate may be exhausted a few dm below the surface, but more typically it will reach a few m depth (Figure 18.4C). In pelagic sediments sulphate has been traced down to several hundred metres depth, and in the red clays and calcareous oozes of the deep sea it is not reduced at all.

Only below the sulphate zone does methane accumulate in the pore-waters as an end product of anaerobic diagenesis (Claypool and Kaplan, 1974; Martens and Berner, 1974). The methanogenic bacteria cannot compete favourably with the sulphate reducing bacteria when sulphate is present and their role in sulphate-rich marine sediments is therefore minor. In contrast to the other electron acceptors, however, bicarbonate accumulates and does not become limiting at any depth.

This vertical sequence at which the different oxidants come into play reflects a thermodynamic order of utilization. Organisms with more energy-yielding metabolism are succeeded with depth by organisms with less energy-yielding metabolism (Claypool and Kaplan, 1974). In the heterogeneous and dynamic coastal sediments, other factors related to process kinetics and bacterial physiology rather than to thermodynamics are also important. The diagenetic processes are mainly biologically catalysed and are subject to the limitations of the physiological potential of the living organisms. As examples, the presence of oxygen will kill all the strictly anaerobic bacteria and thereby inhibit their metabolism, not for thermodynamic but for biochemical reasons. Similarly, nitrate may inhibit the reduction of ferric iron, but not of manganese, in bacteria that use the nitrogenase enzyme also for respiration with iron (Ottow, 1970). The H₂S produced by sulphate reduction may inhibit the last steps in nitrate reduction (Sørensen et al., 1980) and may thereby interfere with the accumulation of N₂O and NO in sediments. Finally, sulphate inhibits the formation of methane due to competition for mutual substrates, acetate and H₂, by the two types of bacteria (Winfrey and Zeikus, 1977; Abram and Nedwell, 1978; Oremland and Taylor, 1978).

The distributions of oxidants in Figure 18.3 and 18.4 indicate the relative depths at which they are consumed. In this respect, especially, the oxygen profiles of the coastal sediments are over-simplifications of the true, spatial distribution. Oxygen does indeed penetrate deeper into the sediment, due to the activities of burrowing animals that ventilate their burrows and tubes with the overlying sea-water in order to maintain an aerobic respiration (e.g., Aller and Yingst, 1978; Revsbech et al., 1979; Sørensen et al., 1979). A similar mechanism was found to bring nitrate down to 20 cm in Puget Sound (Grundmanis and Murray, 1977). The upper, bioturbated sediment therefore has both horizontal and vertical gradients that are important for the interactions between the different mineralization processes (Goldhaber et al., 1977; Aller, 1977, 1978; McCaffrey et al., 1980). Reduced micro-environments are often scattered within the oxic or suboxic surface layers due to local concentrations in microbial metabolism within detritus particles (Jørgensen, 1977b). This heterogeneity is enhanced by the aggregation of sediment particles into faecal pellets produced by the benthic fauna (Rhoads, 1974).

Data on the distribution of denitrification and sulphate reduction in pelagic sediments are mainly derived from kinetic models of early diagenesis based upon measured concentration profiles of sediment cores. These models give information on the maximum rates and rates per area, whereas they have little resolution in the rate distribution with depth. Such information has been obtained mainly for coastal sediments by the use of acetylene blockage techniques for denitrification rates (Billen, 1976; Sorensen, 1978b) and radio-tracer techniques for sulphate reduction rates (Sorokin, 1962; Ivanov, 1968; Jørgensen, 1978). Figure 18.5 shows examples for some coastal sediments. Denitrification is maximal in the upper few cm where nitrate is produced by ammonium oxidation. It is typical that the role of nitrate as an oxidant is depressed during summer in this highly reducing sediment. During winter the suboxic zone expands and denitrification is stimulated in spite of the lower temperature. Denitrification in even more reduced sediments may be undetectable due to the lack of a nitrification zone. The nitrate zone in pelagic sediments is greatly expanded and denitrification here plays a more significant role in the mineralization process (cf., Bender et al., 1977; Vanderborght et al., 1977; Froelich et al., 1979).

Figure 18.5 Vertical distributions of denitrification and sulphate reduction rates in sediments from Danish coastal waters. The water depths are indicated. (Data from: 1 and 2: Sørensen et al., 1978; 3: B. B. Jørgensen, unpublished; 4: Jørgensen, 1977a) 

The sulphate reduction rates are generally higher than the denitrification rates and can still be traced at one metre depth in the sediment at a rate that is a few permille of the surface activity (Figure 18.5). The sulphate reduction exerts an important control on the pore-water chemistry in reducing sediments, through its oxidation of the organic matter by increasing the alkalinity, and by the formation of H₂S, which affects the chemical equilibria.

The seasonal variation in the intensity and relative importance of the different mineralization processes is an important aspect of mineralization in the sea bottom. This variation can be due to both the yearly change in temperature and to the seasonality of phytoplankton productivity. A seasonal variation in the sedimentation of particulate organic matter has been traced even down to 3,000 m depth in the ocean (Deuser and Ross, 1980). Peaks in mineralization rates due to sudden influxes of fresh detritus are, however, partly buffered by the large organic pool size in the sediment. Rates of metabolism in coastal sediments are regulated by temperature with a temperature coefficient, Q₁₀, of generally around 3. This may cause a difference in mineralization rates between winter and summer of up to ten-fold or even more in shallow, temperate waters (Jørgensen, 1977a). As a result, the nutrient regeneration to the water column may also show a strong seasonality.

18.4 NUTRIENT REGENERATION

The cycles of the essential elements carbon, nitrogen, and phosphorus, are coupled because the elements are incorporated into the living biomass of primary producers in relatively constant ratios. The complete mineralization of this biomass must therefore lead to the regeneration of the nutrients, HCO₃ , NH₄⁺, and HPO₄^2-, in a similar stoichiometric proportion. This concept was first developed by Redfield (e.g., Redfield, 1958; Redfield et al., 1963), who found the average elemental composition of marine plankton communities to be C:N:P = 106:16:1. The Redfield concept has since been used in numerous models of the nutrient regeneration for sea-water and sediments systems that have exhibited a complex differentiation between the rate and place of recycling of the three elements.

From the initial C:N:P composition of the phytoplankton organisms there is a rapid, preferential loss of ammonium and phosphate upon their death and initial mineralization. This stripping of N and P progresses as the detritus particles sink towards the sea bottom, as indicated by C:N:P ratios in suspended particulate matter (e.g., Holm-Hansen et al., 1966; Gordon, 1971). The composition of organic matter that is deposited on the bottom therefore depends on the water depth, being the least N- and P-depleted in shallow water (Bordovskiy, 1965; Sholkovitz, 1973; Martens et al., 1978; Suess and Müller, 1980). Suess and Müller, (1980) have compiled data to show that the C:N:P ratio of particulate organic matter in the north-east Pacific increased from an initial 106:16:1 to 106:7.9:0.25 in the euphotic zone and to 106:3.5:0.11 deeper in the water column.

A relative increase in organic N and Pat the sedimentwater interface over that in the water column was also shown by Suess and Müller (1980). It was explained as an assimilation of inorganic NH₄⁺ and HPO₄² by bacteria growing on the detritus. Such an enrichment has been observed during the mineralization of nitrogen and phosphorus-poor detritus in laboratory systems (e.g., Johannes, 1968; Barsdate et al., 1974; Fenchel and Harrison, 1976). In such systems, as in sea-water, the mineralization rate is often limited by the availability of nitrogen or phosphorus. These nutrients are not likely to be limiting for the mineralization rates in marine sediments because they tend to accumulate to high concentrations in the pore-water below the oxic zone, and a continued, preferential depletion of N and P in the oxic zone rather than an enrichment is often the case.

From modelling of pore-water profiles, Martens et al. (1978) found the C:N:P ratio of organic matter actually being decomposed to be 106:19:3.3 in shallow water sediments and 106:4.6:0.37 in offshore sediments. A similar trend of increasing C:N:P ratio of decomposing organic matter from shallow to deep water has been confirmed from a number of other sediments:

Long Island Sound, C:N:P = 106:12:0.75
	(Berner, 1977)
St. Barbara Basin, C:N:P = 106:8:0.5
	(Sholkovitz, 1973)
NW African Slope, C:N:P = 106:8:(0.20.7)
	(Hartmann et al., 1973)

Figure 18.6 Mineralization of ammonium in a coastal sediment, Limfjorden, Denmark. A: Ammonium concentration in the pore water; B: Shaded bars = net ammonium production (= total production minus reassimilated ammonium), non-shaded bars = total ammonium production; C; 1: C:N ratio of organic matter in the sediment; 2: C:N ratio of organic matter being decomposed (Redrawn from Blackburn, 1980). Reproduced by permission of C.N.R.S. Service des Publications 

There is little differentiation in the mineralization of N and P, as the N:P ratio in most sediments remains around 16:1.

A study of the organic nitrogen mineralization in a coastal sediment using N-15 labelled ammonium as a tracer showed that the net mineralization of ammonium is only a fraction of the total mineralization (Blackburn, 1980). Much of the ammonium that is released is re-assimilated by bacteria oxidizing the organic matter, and the accumulation of ammonium in the pore-water is therefore the net result of a nitrogen cycle between organic N and NH₄⁺ within the reducing sediment (Figure 18.6). The C:N ratio increased with depth in the sediment, indicating that in coastal sediments the stripping of nitrogen (and probably also of phosphorus) continues after the deposition of the organic matter. This stripping is, however, partly counteracted by the re-assimilation of ammonium and phosphate into the growing cells of bacteria. The actual C:N ratio of organic matter which was mineralized in the upper sediment was accordingly lower than that of the average organic detritus present at that depth, whereas the opposite was the case in the lower sediment.

Models of mineralization have generally used a constant C:N:P ratio similar to that of the organic matter present in the sediment. The close fit to observed concentration profiles that is sometimes obtained may indicate the predictive value of these models (e.g., Sholkovitz, 1973). Refinements of the models to account also for adsorption, differential diffusion, bioturbation, and other important factors will further increase their accuracy (Berner, 1977, 1980; Krom and Berner, 1980). Müller (1977) and Rosenfeld (1979) observed that a large fraction of the ammonium in sediments is adsorbed to clay minerals and thereby immobilized. Phosphate may form authigenic minerals with iron, manganese or calcium (e.g., vivianite, Fe₃(PO₄)₂^.8H₂O, or apatite, Ca₅(PO₄)₃(OH,F)) in reduced sediments. At the oxic surface layer, phosphate is efficiently bound by adsorption to ferric oxyhydroxides. The large pool of phosphate may be released from the sediment if the iron is reduced during a transient period of anoxia in the bottom waters (see Holm, 1978; Klump and Martens, 1981; Wollast, Chapter 14, this volume). Such an excess release of phosphate has been observed in e.g., the Baltic Sea deep basins where the phosphate concentration increased from 1 to 3 µg at. litre^-1 over a 10-year stagnation period (Deuser, 1975; Suess, 1976). The old, anoxic basins such as the Black Sea or the Cariaco Trench no longer have an anomalously high phosphate release (Deuser, 1975).

The oxidation of organic matter in the sea bottom is stoichiometrically coupled to the consumption of oxygen, nitrate, or sulphate according to mineralization models that were suggested by Redfield (1958), Richards (1965) and others:

 In these models the relative composition of the organic matter should be adjusted to describe the detritus that is actually being decomposed. The models then predict the proportions in which the mineralization products are released. The vertical gradients of HCO₃^- and SO₄^2- relative to those of NH₄⁺ , HPO₄²-, etc., can be used to calculate the stoichiometry of this mineralization.

Most of the organic matter, including the organic carbon, nitrogen, and phosphorus, that is deposited in the sediment is again mineralized. The products, HCO_3, NH₄⁺, and HPO₄²-, are released to the pore-water and are ultimately recycled into the overlying sea-water. Only a few percent of the C, N, and P are permanently buried (see below). The relative composition of the deposited detritus therefore controls the proportions in which these elements are recycled to the sea-water. Averaged over time periods of e.g., a year, the proportions of the C:N:P influx and outflux must be nearly identical. A similar control does not exist for the relative contributions of O₂, NO₃, and SO₄^2- as oxidants in the mineralization. Consequently, we know much less about their mutual regulation as will be discussed below.

A very important aspect of the mineralization processes in the sediments is the release of nutrients back into the sea-water. This is essential, to provide nutrients for the primary production of the phytoplankton communities and thus to maintain the element cycles of the ocean. If oxygen is assumed to be the main terminal electron acceptor for mineralization, the nutrient release relative to the oxygen uptake at the sediment surface should conform to reaction (1). Experimental studies of the nutrient exchange of Narragansett Bay sediments by Nixon et al. (1976) showed a proportion of O₂:NH₄⁺:HPO₄^2- equal to 106:3.5:0.6. The O₂:HPO₄^2- exchange ratio corresponded well to the chemical composition of organic matter in the sediments, but the ammonia release, 2.4 mmol m^-2 day^-1,was less than half of that expected. The most likely explanation for this nitrogen deficit is denitrification by which combined nitrogen has been converted into nitrogen gas. This conclusion is supported by similar exchange studies by Seitzinger et al. (1980) and Henriksen and Blackburn (unpublished data) in coastal sediments and by direct measurements of denitrification rates (Billen, 1978). The loss of combined nitrogen from the marine ecosystem changes the proportions in which nitrogen and phosphorus become available for uptake in the phytoplankton. Since sediment recycling provides up to 50% of the nutrient uptake of the phytoplankton in coastal waters, nitrogen tends to become a limiting nutrient for the marine primary production (Ryther and Dunstan, 1971; Davies, 1975; Rowe et al., 1975; Nixon et al., 1976; Raine and Patching, 1980). Also, on a global basis, the sediment denitrification is important for the nitrogen balance of the ocean as it exceeds the total nitrogen influx via rivers by roughly five-fold (cf. Söderlund and Svensson, 1976).

The nutrient exchange across the sedimentwater interface in the deep sea is of little significance for the dynamic nutrient regeneration in the open ocean because mineralization here mainly takes place in the water column. Upwelling areas may be exceptions to this. In the eastern equatorial Atlantic at 5000 m depth, Bender et al. (1977) calculated a total nitrate flux into the deep sea by mineralization, convective inflow, etc., of 8.8 µmol cm^-2 yr^-1  of which the sedimentwater exchange accounted for one-third.

18.5 QUANTITATIVE ASPECTS OF OXIC MINERALIZATION 

During the last few years a sufficient amount of data on the oxygen uptake rates of marine sediments from the coast to 5,000 m depth have been obtained to allow a quantitative evaluation of this process for different depth regions of the ocean. A compilation of such data is presented in Figure 18.7. Only rates based on in situ measurements or on pore-water profiles have been included for the pelagic sediments, as shipboard rate measurements on collected cores grossly overestimate the natural rates at these depths (Smith, 1978). The measured rates decreased up to 10,000-fold from shallow, eutrophic waters to pelagic sediments underlying the sub-tropical gyres of the oceans.

A comparison between the oxygen uptake rates of different depth regions with the primary production in the surface waters shows the quantitative role of the sea bottom for the biological carbon cycle in the ocean (Table 18.1). The sea bottom has been divided into five depth regions. Average oxygen uptake rates used for each depth interval are indicated as vertical lines in Figure 18.7. Table 18.1 shows that the shelf comprises 8.6% of the total area of the ocean, and yet 83% (49 + 34) of the mineralization of organic matter in the sea bottom takes place here at water depths less than 200 m. The deep sea, below water depths of 4,000 m, covers over half of the ocean but only 2% of the organic matter is oxidized here. Organic detritus that sinks down from the pelagic regions of the oceans is therefore very efficiently oxidized within the water column and little is left when it finally reaches the bottom (cf. Figure 18.1). Pelagic sediments are in this respect unimportant for the dynamic cycling of organic carbon. The possible contribution to the oxygen uptake of decomposing carcasses from dead fish and macro-benthic invertebrates probably escapes these data and has not been assessed.

The primary productivity in the ocean regions varies from about 200 to 50 g C m^-2 yr^-1 between coastal waters and open sea and most of the phytoplankton production takes place in the pelagic regions outside the shelf (Table 18.1). An average of 25% of the organic production over the shelf is mineralized in the sediments. This corresponds to values ranging from 20 to 60% which have been estimated for different coastal areas (Riley, 1956; Davies, 1975; Nixon et al., 1976; Raine and Patching, 1980; Seitzinger et al., 1980). The upper and lower continental slope receive and mineralize about 13% and 2% of the primary production,  respectively, and deep-sea sediments receive and mineralize only 0.4% of the local primary production from the surface waters as calculated from the sediment oxygen uptake. These rates of detritus input to the sea bottom are equal to or slightly lower than the rates obtained from sediment trap studies (Figure 18.1).

Figure 18.7 Oxygen uptake rates of the sediments at different depths in the ocean. (Data from: A: Smith, 1978; B: Smith, 1974; C: Pamatmat and Banse, 1969; D: Pamatmat, 1971; E: Murray and Grundmanis, 1980; F: Smith and Teal, 1973; G: Smith et al., 1976; H: Smith et al., 1972;I: B. B. Jørgensen, unpublished (from waters between the Baltic and North Sea); J: B. B. Jørgensen, unpublished (from Kysing Fjord, Denmark); K: Sørensen et al., 1978; L: Jørgensen, 1977a; M: Wilson, 1978; N: T. H. Blackburn and N. P. Revsbech, unpublished (from a Greenland Fjord)

Budgets similar to that in Table 18.1 can also be made for nitrogen and phosphorus. This would show a similar relative distribution of remineralization, but with an even stronger bias towards the shelf sediments due to preferential stripping of N and P during sinking of the detritus.

Only a small fraction of the organic detritus that is deposited on the shelf or in the deep sea is permanently buried in the sediment. The actual percentage can be estimated from several types of mass-balance calculations or from a direct comparison between rates of mineralization and sediment accumulation. Pelagic sediments typically have sedimentation rates of about 1 cm per 1,000 years and an organic content of 0.2% dry wt (Degens and Mopper, 1976). With an assumed water content of 70% by weight and a density of 1.2 the calculated accumulation rate of organic carbon will be 0.25 µmol org. C m^-2 day^-1. This is 1 % of that mineralized by oxygen according to Table 18.1, and thus 1 % of the deposited organic matter is permanently buried. Similar calculations for a typical shelf sediment with an organic content of 2% dry wt and a sedimentation rate of 100 cm per 1,000 years gives an accumulation rate of 0.5 mmol org. C m^-2 day^-1, or 2.5% of that mineralized by oxygen. In very organic-rich and rapidly depositing coastal sediments, over 10% of the deposited organic matter may be permanently buried (e.g., Jørgensen, 1977a; Müller and Suess, 1979).

Table 18.1 Oxygen uptake rates of the sea bottom in five depth zones. The net primary production of phytoplankton over each zone is given and the percentage of this production that is oxidized in the sediment is calculated. Numbers in parenthesis indicate the contribution of each zone to the whole ocean. (Areas and primary productivity date are adapted from Sorokin, 1978; primary production is recalculated to O₂ equivalents using a photosynthetic quotient of 1.15, i.e., 10⁹ mol O₂ is equivalent to 0.010 Tg C)

Zone	Depth		Area	O2 uptake rate	Total O2 uptake		Net primary prod.		O2 uptake rate
	(m)		(1012 m2)	(mmol m-2 day-1)	(109 mol day-1)		(109 mol day-1)		Net primary prod.
Inner shelf	050	13	(3.6%)	20	260	(49%)	1020	(16%)	25%
Outer shelf	50200	18	(5.0%)	10	180	(34%)	710	(11%)	25%
Upper slope	2001000	15	(4.2%)	3	45	(9%)	340	(5%)	13%
Lower slope	10004000	106	(29%)	0.3	32	(6%)	1760	(27%)	2%
Deep sea	>4000	208	(58%)	0.05	10	(2%)	2730	(41%)	0.4%
Total		360	(100%)		527	(100%)	6560	(100%)

Holland (1978) calculated from a steady state mass balance of phosphate that 0.5% of the net primary production was permanently buried in the sea bottom. From this it can be estimated that 3% of deposited organic phosphate should be buried, which fits well with the calculation of 2.5% above, since most deposition takes place in shelf sediments. Similar calculations for organic carbon based upon an estimated rate of burial of 120 Tg C yr^-1 (Holland, 1978) would lead to a permanent burial of 5% of all deposited organic matter, which may be an over-estimate.

18.6 QUANTITATIVE ASPECTS OF ANOXIC MINERALIZATION

In the calculations above, the role of suboxic diagenesis as well as of sulphate reduction and methanogenesis has not been considered. Quantitative data on denitrification in marine sediments are rather few and scattered. Denitrification rates measured by realistic experimental techniques in coastal sediments typically vary between 0.2 and 2 mmol NO₃ m^-2 day^-1, with an average around 0.5 mmol (Billen, 1978; Fenchel and Blackburn, 1979; Kaplan et al., 1979; Sørensen et al., 1979; Seitzinger et al., 1980; Henriksen and Blackburn, unpublished data). In pelagic sediments the rates have been calculated from diffusion models of pore-water profiles to be about 0.005 mmol NO₃ m^-2 day^-1 in the Atlantic Ocean at 5,000 m depth (Bender et al., 1977; Wilson, 1978). The potential denitrification rates can be estimated from the oxygen uptake rates using the assumption that all organic nitrogen, after being released as ammonium in the sediment, is nitrified and subsequently denitrified. With a C:N ratio typical of offshore sediments of 106:8, the denitrifiers can potentially contribute about 10% ((8.5/4)/106) of the electron flow, i.e., up to 10% of the organic matter that is mineralized in pelagic sediments can be oxidized by denitrifying bacteria. In coastal sediments with a C:N ratio of 106:16, the contribution by denitrifiers could theoretically approach 20%. Additional uptake and subsequent reduction of nitrate from the sea water may take place in very reducing sediments (Wollast, Chapter 15, this volume), but in general the surface sediments have a higher nitrate concentration than the overlying water and therefore release nitrate rather than take it up.

The reduction of 0.5 mmol NO₃ m^-2 day^-1 in coastal sediments is equivalent to 3% of the oxygen uptake (Table 18.2) and corresponds to one-third of the ammonia released to the pore-waters. This corresponds well to the observations from individual coastal sediments where 2050% of the total nitrogen flux has been observed to go through denitrification (Nixon et al., 1976; Billen, 1978; Fenchel and Blackburn, 1979; Seitzinger et al., 1980; Henriksen and Blackburn, unpublished data). The numbers for denitrification in deep-sea sediments are equivalent to 20% of the electron flow, which is twice the expected maximum relative to the oxygen uptake (Table 18.2). Although these rates may be high because the study areas were underlying upwelling regions, they do demonstrate a trend of relatively higher denitrification in the more oxidized sediments versus a relatively less significant denitrification in reduced sediments.

If an average of 3, 5, and 10% of the electron flow in the sea bottom is assumed to go through denitrification in the depth regions of 0200 m, 2001,000 m, and > 1,000 m, respectively, this would lead to a total rate for the world ocean sediments of 16^.10⁹ µmol NO₃ day^-1 or 80 Tg N yr^-1. Out of this, 67% takes place in shelf sediments. An estimate of 165 Tg N was made by Hahn and Junge (1977) to account for the entire ocean. The rate is several-fold higher than the estimated denitrification in the oxygen-deficient zones of the open ocean, including the eastern tropical north and south Pacific, the Arabian Sea, and the Cariaco Trench (Goering, 1978). In the largest of these areas from the North Pacific the estimated denitrification rates average around 20 Tg N yr^-1 (Codispoti and Richards, 1976; Goering, 1978). These estimates of total oceanic denitrification rates are, however extrapolations on a very slender basis, but they do indicate that most denitrification takes place in shelf sediments rather than in the water column or in the deep sea.

Data on sulphate reduction rates in marine sediments are more abundant and have been summarized by Trudinger et al. (1972), Goldhaber and Kaplan (1975), Jørgensen (1982), Volkov et al. (in preparation), and others. Radio-tracer measurements of sulphate reduction in shelf sediments from the coast to 200 m depth have been made concomitant with measurements of oxygen uptake rates (Figure 18.8). There is a positive correlation between the two processes, showing that sulphate reducing bacteria are responsible for a rather constant proportion of the electron flow in the sediments.

In coastal sediments from 050 m depth the ratio between oxygen and sulphate reduction is close to 4:1 (Figure 18.8). Due to the different stoichiometries of the two mineralization processes (equations (1) and (3)), their relative importance as electron acceptors must be O₂:SO₄² = 2:1. This means that sulphate reduction is equivalent to 50% of the oxygen uptake in terms of electron flow. It has been calculated that about 90% of the H₂S produced by sulphate reduction in these sediments diffuses back to the sediment surface where the H₂S is re-oxidized to sulphate by oxygen. As the H₂S was equivalent to 90% of 50% of the oxygen uptake, this re-oxidation must therefore consume 45% of the oxygen uptake. Thus, 45% of the oxygen uptake of these coastal sediments is somehow involved in the re-oxidation of H₂S. Only the remaining 55% of the oxygen is then directly involved in the oxidation of organic matter. The oxidation by oxygen is thus of nearly the same magnitude as the oxidation by sulphate (Jørgensen, 1977a). These and other data indicate a dominating role of sulphate reduction for the mineralization processes in coastal sediments. As the oxygen uptake included 90% of the electron flow through sulphate reduction due to the re-oxidation of H₂S, this sulphate reduction is also included in the oxygen budgets of Table 18.1. This inclusion is a prerequisite for the use of oxygen uptake rates as an integrated measure of mineralization of organic matter in the sea bottom.

Table 18.2 Oxygen uptake, denitrification, and sulphate reduction rates of coastal and pelagic sediments. Numbers in parentheses indicate the percentage mineralization by nitrate and sulphate relative to oxygen, taking into account the differences in mineralization stoichiometry

Sediment	Depth	Reduction rate (mmol m-2 day-1)			Area	Total rate (109 mol day-1)
zone	(m)	O2	NO3	SO42	(1012 m2)	NO3	SO42
Coastal	050	20(100%)	0.5 (3%)	5 (50%)	13	6.5	65
Pelagic	> 4000	0.02 (100%)	0.003 (20%)	~0(~0%)	208	0.6	~0

Figure 18.8 Relation between oxygen uptake rates and sulphate reduction rates in the upper 15 cm of sediments from the transition area between the Baltic Sea and the North Sea. (Redrawn from B. B. Jørgensen, 1982)

As the average oxygen uptake rate is 20 mmol m^-2 day^-1 in the inner shelf sediments, the sulphate reduction rate is about 5 mmol m^-2 day^-1, according to the 1:4 ratio between O₂ and SO₄² (Figure 18.8; Table 18.2). In the outer shelf, the relative importance of sulphate reduction decreases to 1:7 of the O₂ uptake or 1.4 mmol m^-2 day^-1 (Figure 18.8). Data on sulphate reduction in pelagic sediments have been obtained by modelling of profiles and in a few cases by radiotracer measurements (e. g., Kaplan et al., 1963; Berner, 1964; Ivanov et al., 1976). It is difficult to recalculate these data on an areal basis. Volkov et al. (1972) observed that sulphate reduction is gradually suppressed to deeper sediment layers as the suboxic zone expands in the slope sediments. The process seems to cease completely at 34,000 m water depth, and sulphate reduction is not detectable in the red clays or carbonate ooze of the deep sea.

Due to the suppression of sulphate reduction in the deep sea the process is, even more than oxygen uptake, predominantly located in the shelf sediments, where an estimated 9095 % of all dissimilatory sulphate reduction in the ocean takes place.

The element cycles in the sea bottom are completed by the re-oxidation of reduced, inorganic molecules such as H₂S and NH₄⁺ near the sediment surface. The molecules carry up to 50% of the energy and reducing power of the organic detritus that were transferred during anaerobic decomposition. The re-oxidation will ultimately consume an equivalent amount of oxygen, but in this process new biomass may be formed. The oxidation is partly carried out by chemolithotrophic bacteria that use some of the energy for cell synthesis. Their growth yield is low, as only 1020% of the electron flow goes to reduce CO₂. Consequently, the chemosynthesis does not exceed 510% of the total carbon cycling in the sediment (cf., Jørgensen, 1980). The chemosynthetically produced organic matter is later remineralized, whereby the nutrients are released. Chemosynthesis therefore functions as a delay in mineralization but does not change the net outcome of the process.

A completely different situation exists around the hot vents of the deep-sea spreading centres. Large amounts of H₂S are produced here beneath the ocean bottom by a chemical, geothermally catalysed reduction of sea-water sulphate that percolates through the basaltic rock. The chemical energy of the H₂S is exploited by free-living or symbiontic, chemo-lithotrophic bacteria (Jannasch and Wirsen, 1979; Cavanaugh et al., 1981; Rau, 1981). Their chemosynthesic production of biomass feeds a rich benthic community in the surrounding sea bottom. In contrast to chemosynthesis in normal sediments, which is originally fuelled by phytoplankton and solar energy, chemosynthesis in the vent areas functions as a primary production that changes the whole nutrient balance of the sediment. The contribution of the hot vent systems to the overall cycling of C-N-P-S in the deep sea remains to be determined.

18.7 REGULATING MECHANISMS

Only a few percent of the organic matter, including its carbon, nitrogen and phosphorus, that are deposited on the sediment surface were shown to become permanently buried into deeper layers. The remaining 9099% are recycled into the sea-water after mineralization. The close balance between influx and outflux show that the rate of mineralization in the sediment is completely governed by the rate of organic sedimentation. The chemical form in which the elements are recycled from the sediment may, however, vary depending on processes such as nitrification and denitrification. These processes are of great importance in relation to the nutrient regeneration to the overlying water but do not change the overall nitrogen balance of the sediment.

A similar conclusion does not apply to sulphur. The bacterial sulphate reduction causes an uncoupling between the C-N-P and the S-cycling in sediments relative to the composition of these elements in the deposited organic matter. Organic sulphur accounts for only a few percent of the sulphur cycling in sediments (Jørgensen, 1977a). The rest is due to dissimilatory reduction of sulphate which diffuses down from the overlying sea-water. In the inner shelf region the sulphur cycle is regulated by the organic carbon mineralization, as shown in Figure 18.8, and this regulation is independent of the C:S ratio of the organic matter that is being decomposed. In the outer shelf and slope sediments, where the rates of metabolism in the sediment are lower, the sulphur cycle gradually looses importance. In the deep sea, where there is no sulphate reduction, sulphur is stoichiometrically linked to the carbon mineralization in a manner similar to nitrogen and phosphorus.

The chemical constituents of the sea-water, including carbon, nitrogen, and phosphorus, are introduced into the ocean mainly by river discharge. They are removed again by burial into the sea bottom. Their residence time in the ocean is about 100,000 years (Broecker, 1974; Holland, 1978). Over a time scale that is long relative to the residence time, the burial rate must roughly have balanced the input from the rivers. As only a few percent of deposited organic matter is permanently buried in the sediments, the elements must have recycled between the sea bottom and the ocean water about 50 times during the last 100,000 years. The recycling is, however, much more rapid in estuaries and coastal waters than in the open ocean and the residence time is shorter. The coastal environment is therefore much more susceptible to natural or man-made perturbations than the ocean in general, which functions as a huge buffer to the element cycles.

Several mechanisms for regulating the concentration and accumulation rate of organic matter in the sea bottom have been suggested. The specific rate constants for the decomposition of organic matter in sediments have been found to vary in proportion to the second power of the sedimentation rate (Toth and Lerman, 1977; Berner, 1978). This indicates that increasing productivity and sedimentation in the sea would favour the preservation of more labile organic substances. Müdler and Suess (1979) found that a 10-fold increase in sedimentation rate was accompanied by a doubling in the organic carbon content. These and other observations show a balanced regulation of the organic content of sediments between deposition and remineralization. In anoxic basins the mineralization process is less efficient. There is an increased preservation of the organic matter and thus a higher organic concentration in the sediment (e.g., Denser, 1975).

In addition to the accumulated organic matter, a significant part of the reducing power of the deposited detritus is retained in the sediment in the form of pyrite, FeS₂. Electrons are transferred from the detritus to sulphate by sulphate reducing bacteria and a part of the resulting sulphide is precipitated by iron and is gradually converted into pyrite. New sulphate diffuses down into the sediment from the overlying sea-water and the sea bottom thereby functions as a trap for sulphur. The accumulation rate of pyrite in a fjord sediment from the western Baltic was found to be 84 mg FeS₂ m^-2 day^-1. This accounted for 30% of the total reducing equivalents which were buried into that sediment in the form of organic matter (Jørgensen, 1977a).

The observed response of the sea bottom to increases in the burial rate of organic matter (and pyrite) as a result of increased productivity will tend to maintain a steady state in the chemistry and biology of the ocean. The current acceleration of phosphate discharge into the sea is one mechanism that is known to stimulate the phytoplankton productivity (Stumm, 1973). A resulting increase in the burial of organic phosphate will tend to counteract the increase in production. A similar feedback control may be exerted on oxygen in areas where low oxygen concentration impedes an efficient remineralization of the organic matter, such as in stagnant basins and fjords and in the large oxygen minimum zones of the eastern tropical Pacific, the Arabian Sea, etc. A decrease in the oxygen level at the sea bottom caused by eutrophication or by other effects could result in a vast expansion of these oxygen minimum zones. As more reduced matter would then be buried, this would tend to increase the concentration of free oxygen again (Deuser, 1975; Holland, 1978).

Due to the huge pool sizes of the biological elements in the ocean, such regulations will, on a global scale, operate over a tinge range that is related to the residence time of the elements, i.e., many thousands of years. Such regulations are therefore more of theoretical than of practical importance in relation to our current management of the global environment. Locally, however, coastal waters and enclosed basins, such as the Baltic Sea, may be significantly affected by eutrophication. We still have only a limited understanding of the general effects that such a eutrophication will have on the element cycles in the sediments. In addition to an increased net burial, a preferential stimulation of sulphate reduction and thus formation of H₂S can be expected, as well as an opposite effect on nitrification and denitrification. This will change the qualitative pattern of nutrient regeneration from the sediments and may also affect, for example, the emission of H₂S, N₂O, and other biogenic gases into the atmosphere.

18.8 CONCLUSIONS

The organic pool of sediments thus largely represents a dynamic intermediate between the influx of organic particles and outflux of inorganic molecules. The organic pool deeper in the sediment, which becomes permanently trapped into sediment and eventually into sedimentary rocks, represents a small drain between large fluxes. This small drain is decisive for the marine carbon, nitrogen, and phosphorus cycles, and for the composition of sea-water in general, on a long time scale of 10⁴10⁵ years. The dynamic recycling operates on a much shorter time scale, shortest in the shallow coastal areas of the sea, and is more susceptible to man-made perturbations. The preceding discussion has shown that in order to understand these dynamic cycles and the factors that govern their rates and balance, we need to improve the quantitative data on specific process rates as well as to combine data obtained from very diverse types of approaches and dimensions to build a coherent picture of biogeochemical cycling in the sea bottom.

ACKNOWLEDGMENTS

I thank T. Henry Blackburn, Kaj Henriksen, and Niels Peter Revsbech for the permission to present their unpublished data and Rolf Hallberg and Jan Sørensen for helpful suggestions as to the pertinent literature. Christopher S. Martens and Erwin Suess reviewed the manuscript and gave valuable criticism. The figures were drafted by Arno Jensen.

18.9 REFERENCES

Aller, R. C. (1977) The influence of macrobenthos on chemical diagenesis of marine sediments, Ph. D. Dissertation, Yale University.

Aller, R. C., and Yingst, J. Y. (1978) Biogeochemistry of tube-dwellings: A study of the sedentary polychaete Amphitrite ornata (Leidy), J. Mar. Res., 36, 201-254.

Barnes, R. O., Bertine, K. K., and Goldberg, E. D. (1975) N₂:Ar, nitrification and denitrification in southern California borderland basin sediments, Limnol. Oceanogr., 20, 962-970.

Bender, M. L., Fanning, K. A., Froelich, P. N., Heath, G. R., and Maynard, V. (1977) Interstitial nitrate profiles and oxidation of sedimentary organic matter in the eastern equatorial Atlantic, Science, 198, 605-609.

Berner, R. A. (1977) Stoichiometric models for nutrient regeneration in anoxic sediments, Limnol. Oceanogr., 22, 781-786.

Berner, R. A. (1980) Early Diagenesis. A Theoretical Approach, Princeton, N. J., Princeton University Press.

Billen, G. (1978) A budget of nitrogen recycling in North Sea sediments off the Belgian coast, Estuar. Coast. Mar. Sci., 7, 127-146.

Bordovskiy, O. K. (1965) Accumulation and transformation of organic substances in marine sediments, Mar. Geol., 3, 3-82.

Cavanaugh, C. M., Gardiner, S. L., Jones, M. L., Jannasch, H. W., and Waterbury, J. B. (1981) Prokaryotic cells in the hydrothermal vent tube-worm Riftia pachyptila Jones: Possible chemo-autotrophic symbionts, Science, 213, 340-342.

Claypool, G. E., and Kaplan, I. R. (1974) The origin and distribution of methane in marine sediments, in Kaplan, I. R. (ed.) Natural Gases in Marine Sediments, New York, Plenum Press, 99-140.

Davies, J. M. (1975) Energy flow through the benthos in a Scottish sea loch, Mar. Biol., 31, 353-362.

Degens, E. T., and Mopper, K. (1976) Factors controlling the distribution and early diagenesis of organic material in marine sediments, in Riley, J. P., and Chester, R. (eds) Chemical Oceanography, 2nd edn, Vol. 6, London, Academic Press, 59-113.

Deuser, W. G., and Ross, E. H. (1980) Seasonal changes in the flux of organic carbon in the deep Sargasso Sea, Nature, 283, 364-365.

Fenchel, T., and Harrison, P. (1976) The significance of bacterial grazing and mineral cycling for the decomposition of particulate detritus, in Anderson, J. H. (ed.) The Role of Terrestrial and Aquatic Organisms in Decomposition Processes, Oxford, Blackwell Scientific Publications Ltd., 285-299.

Froelich, P. N., Klinkhammer, G. P., Bender, M. L., et al. (1979) Early diagenesis of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis, Geochim. Cosmochim. Acta, 43, 1075-1090.

Goldhaber, M. B., and Kaplan, I. R. (1974) The Sulphur Cycle, in Goldberg, E. D. (ed.) The Sea, Vol. 5, New York, Wiley, 569-655.

Goldhaber, M. B., Aller, R. C., Cochran, J. K., Rosenfeld, J. K., Martens, C. S., and Berner, R. A. (1977) Sulphate reduction, diffusion, and bioturbation in Long Island Sound sediments. Report of the FOAM group, Amer. J. Sci., 277,193-237.

Grundmanis, V., and Murray, J. W. (1977) Nitrification and denitrification in marine sediments from Puget Sound, Limnol. Oceanogr., 22, 804-813.

Hartmann, M., Müller, P., Suess, E., and van der Weiden, C. H. (1973) Oxidation of organic matter in recent marine sediments, "Meteor" Forsch.-Ergebnisse, C, 12, 74-86.

Holland, H. D., (1978) The Chemistry of the Atmosphere and Oceans, New York, Wiley.

Holm-Hansen, O., Strickland, J. D. H., and Williams, P. M. (1966) A detailed analysis of biologically important substances in a profile off southern California, Limnol. Oceanogr., 11, 548-561.

Hunt, J. M. (1968) The significance of carbon isotope variations in marine sediments, in Hobson, G. D. (ed.) Advances in Organic Geochemistry, Oxford, Pergamon Press, 27-35.

Ivanov, M. V., Lein, A. Yu., and Kashparova, E. V. (1976) Rate of formation and diagenesis of reduced sulphur compounds in the Pacific Ocean, in Biogeochemistry of the Diagenesis in Marine Sediments, Moscow, Nauka, 171-178 (in Russian). 

Jannasch, H. W., and Wirsen, C. O. (1973) Deep-sea microorganisms: an in situ response to nutrient enrichment, Science, 180, 641-643.

Johannes, R. E. (1968) Nutrient regeneration in lakes and oceans, in Droop, M. R., and Ferguson Wood, E. J. (eds) Advances in Microbiology of the Sea, Vol. 1, 203-213. 

Jørgensen, B. B. (1977a) The sulphur cycle of a coastal marine sediment (Limfjorden, Denmark), Limnol. Oceanogr., 22, 814-832.

Jørgensen, B. B. (1978) A comparison of methods for the quantification of bacterial sulphate reduction in coastal marine sediments. I. Measurements with radiotracer techniques, Geomicrobiol. J., 1, 11-28.

Jørgensen, B. B. The microbial sulphur cycle, in Krumbein, W. E. (ed.) Microbial Geochemistry (in press).

Kaplan, I. R., Emery, K. O., and Rittenberg, S. C. (1963) The distribution and isotopic abundance of sulphur in recent marine sediments off southern California, Geochim. Cosmochim. Acta, 27, 297-331.

Klump, J. V., and Martens, C. S. (1981) Biogeochemical cycling in an organic rich coastal marine basinII. Nutrient sedimentwater exchange processes, Geochim. Cosmochim. Acta, 45, 101-121.

Koike, I. and Hattori, A. (1978) Denitrification and ammonia formation in anaerobic coastal sediments, Appl. Environ. Microbial., 35, 278-282.

Liss, P. S. The exchange of biogeochemically important gases across the airsea interface, Chapter 15, this volume.

Martens, C. S., and Berner, R. A. (1974) Methane production in the interstitial waters of sulphate depleted marine sediments, Science, 185, 1167-1169.

McCaffrey, R. J., Myers, A. C., Davey, E., et al. (1980) The relation between pore water chemistry and benthic fluxes of nutrients and manganese in Narragansett Bay, Rhode Island, Limnol. Oceanogr., 25, 31-44.

Müller, P. J. (1977) C/N ratios in Pacific deep-sea sediments: Effect of inorganic ammonium and organic nitrogen compounds sorbed by clays, Geochim. Cosmochim. Acta, 41, 765-776.

Murray, J. W., and Grundmanis, V. (1980) Oxygen consumption in pelagic marine sediments, Science, 209, 1527-1530.

Nixon, S. W., Oviatt, C. A., and Hale, S. S. (1976) Nitrogen regeneration and the metabolism of coastal marine bottom communities, in Anderson, J. M., and Macfayden, A. (eds) The Role of Terrestrial and Aquatic Organisms in Decomposition Processes, Oxford, Blackwell, 269-283.

Ottow, J. C. G. (1970) Selection, characterization and iron-reducing capacity of nitrate reductaseless (nit^-) mutants of iron-reducing bacteria, Z. Allg. Microbial., 10, 55-62.

Pamatmat, M. M., and Banse, K. (1969) Oxygen consumption by the seabed. II. In situ measurements to a depth of 180 m, Limnol. Oceanogr., 14, 254-259.

Raine, R. C. T., and Patching, J. W. (1980) Aspects of carbon and nitrogen cycling in a shallow marine environment, J. Exp. Mar. Biol. Ecol., 47,127-140.

Redfield, A. C. (1958) The biological control of chemical factors in the environment, Amer. Sci., 46, 206-226.

Revsbech, N. P., Sørensen, J., Blackburn, T. H., and Lomholt, J. P. (1979) Distribution of oxygen in marine sediments measured with microelectrodes, Limnol. Oceanogr., 25, 403-411.

Rhoads, D. C. (1974) Organism-sediment relations on the muddy sea floor, Oceanogr. Mar. Biol. Ann. Rev., 12, 263-300.

Riley, G. A. (1956) Oceanography of Long Island Sound, Bull. Bingham Oceanogr. Coll., 15, 324-344.

Rowe, G. T., and Gardner, W. D. (1979) Sedimentation rates in the slope waters of the northwest Atlantic Ocean measured directly with sediment traps, J. Mar. Res., 37, 581-600.

Ryther, J. H., and Dunstan, W. M. (1971) Nitrogen, phosphorus, and eutrophication in the coastal marine environment, Science, 171, 1008-1013.

Sholkovitz, E. (1973) Interstitial water chemistry of the Santa Barbara Basin sediments, Geochim. Cosmochim. Acta, 37, 2043-2073.

Smith, K. L. Jr (1978) Benthic community respiration in the N.W. Atlantic Ocean: in situ measurements from 40 to 5200 m, Mar. Biol., 47, 337-347.

Smith, K. L., Jr, Clifford, C. H., Eliason, A. H., et al. (1976) A free vehicle for measuring benthic community metabolism, Limnol. Oceanogr., 21, 164-170.

Smith, K. L., Jr, and Teal, J. M. (1973) Deep-sea benthic community respiration: an in situ study at 1850 meters, Science, 179, 282-283.

Nitrogen, Phosphorus and SulphurGlobal Cycles, SCOPE Report No. 7, Ecol. Bull. (Stockholm),

Sørensen, J. (1978a) Capacity for denitrification and reduction of nitrate to ammonium in a coastal marine sediment, Appl. Envir. Microbiol., 35, 301-305.

Sørensen, J., Jørgensen, B. B., and Revsbech, N. P. (1979) A comparison of oxygen, nitrate, and sulphate respiration in coastal marine sediments, Microb. Ecol., 5, 105-115.

Sorokin, Yu. I. (1962) Experimental investigation of bacterial sulphate reduction in the Black Sea using ³⁵S, Microbiol., 31, 329-335.

Stumm, W. (1973) The acceleration of the hydrogeochemical cycling of phosphorus, Water Res., 7, 131-144.

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

Toth, D. J., and Lerman, A. (1977) Organic matter reactivity and sedimentation rates in the ocean, Amer. J. Sci., 277, 465-485.

Walsh, J. J., Rowe, G. T., Iverson, R. L., and McRoy, C. P. (1981) Biological export of shelf carbon is a sink of the global CO₂ cycle, Nature, 291,196-201.

Wilson, T. R. S. (1978) Evidence for denitrification in aerobic pelagic sediments, Nature, 274, 354-356.

Volkov, I. I, Rozanov, A. G., Zjabina, N. N., and Jagodinskaja, T. A. (1972) Sulphur in sediments of the Pacific Ocean east of Japan, Lithology of Mineral Resources, 4, 50-64, (In Russian).

Wollast, R. Interactions in estuaries and coastal waters, Chapter 14, this volume. 

COMMENT TO CHAPTER 18:

CYCLING OF METABOLIZABLE C, N, P, AND S IN ORGANIC-RICH MARINE SEDIMENTS

C. S. MARTENS AND H. W. JANNASCH

1. emphasize the importance of quantifying the metabolizable fraction of the total sedimentary matter;
2. discuss the observed distribution of the chemical end-products of organic matter degradation;
3. discuss the microbiological consequences of the vertical transport of reduced chemical end products to the sedimentwater interface.

Metabolizable Organic Carbon

Berner (1974) has discussed a general kinetic model for early diagenesis driven by the metabolizable fraction of sedimentary organic carbon. As Jørgensen (Chapter 18, this volume) has shown, such models have been extensively utilized to provide information on the rates of specific related processes such as sulphate reduction, organic carbon degradation, and ammonium production. Recently, however, the estimates of the metabolizable fraction of the organic carbon based on simple curve fits to combined solid phase carbon and pore-water vertical concentration profiles, such as dissolved sulphate, have been shown to have large uncertainties (e.g. Murray et al., 1978; Berner, 1980). The situation is further complicated in bioturbated sediments where differential mixing processes are frequently modelled in a stochastic fashion (e.g. Guinasso and Schink, 1975; Aller, 1978; Berner, 1980).

There are few direct measurements of the metabolizable components of total sedimentary organic matter. The sum of these components is frequently estimated as the loss in the total sedimentary organic carbon concentration, which is calculated as the difference between that found immediately at the sediment-water interface and that found at some depth, usually within tens of cm of the interface. In many sediments, metabolizable carbon represents less than 50% of the total organic carbon initially present (e.g. Berner, 1974; Murray et al., 1978; Martens and Klump, 1983). Furthermore, it may be that some fraction of the initially present metabolizable organic carbon may be rendered refractory or `fossilized' in situ by processes such as humification or protokerogen formation, (e.g. Stuermer et al., 1978). Generally such processes are referred to as geopolymer formation (Berner, 1980). If so, this latter process is the actual net sink through burial for organic carbon in sediments, since the refractory material initially present consists of recycled soil carbon, peat debris and degradation-resistant, lignin-derived components (Hedges and Parker, 1976; Hedges and Mann 1979) with radiocarbon ages generally greater than 1000 yrs (Scharpenseel, 1979; Baxter et al., 1980).

We need direct evidence of geopolymer formation and quantification of the factors controlling geopolymer formation to document its importance in recent sediments. Radiocarbon provides a tool for studying this problem, as discussed by Benoit et al. (1979), Baxter et al. (1980), and Turekian et al. (1980). Metabolizable organic carbon such as that derived from plankton, sewage, sea grasses, and marsh grasses normally has a future age (i.e. a negative age) due to the addition of bomb carbon from the atmosphere to modern plants (see Benoit et al., 1979). These modern components can therefore be discriminated from the older lignin-derived carbon, which generally has ages of several thousand years as discussed above. A profile of radiocarbon ages of sedimentary organic carbon in Cape Lookout Bight, N.C. (U.S.A.) (Figure 18.9) illustrates the results of such an approach. The bight contains non-bioturbated organic-rich anoxic sediments accumulating at approximately 10 cm yr^-1 (Chanton et al., 1983) in which sulphate reduction and methane production are virtually complete within 35 cm of the sedimentwater interface (Martens and Klump, 1980; Sansone and Martens, 1981). Approximately 25% of the initial total organic carbon pool is lost from the sediments as methane and dissolved inorganic carbon (Martens and Klump, 1983).

The radiocarbon data indicate that only material of greater than 1000 years age survives for longer than 10 years (100 cm depth at a sediment accumulation rate of 10 cm yr^-1) from an initial mixture of metabolizable and non-metabolizable carbon at the sedimentwater interface with an age of approximately 550 years. If this old material represents the composite age of remaining refractory soil and other lignin derived carbon components (e.g. Benoit et al., 1979; Baxter et al., 1980) this would indicate that virtually no metabolizable carbon is geopolymerized in bight sediments. Clearly, accurate quantitative measurements of humification or geopolymer formation rates under well-defined and representative environmental conditions are needed before estimates of these rates based on differences between large initial total and nonmetabolizable carbon reservoirs are extrapolated to global carbon budgets.

Figure 18.9 Corrected (¹³C) radiocarbon age distribution of sedimentary organic carbon in the upper three metres of Cape Lookout Bight, North Carolina, U.S.A. The bight is a small organic-rich basin formed behind a migrating barrier island (from Martens, 1983)

Chemical End-products of Organic Matter Degradation

A flow chart illustrating the chemical results of sedimentary organic matter degradation under oxic through anoxic conditions is presented in Figure 18.10. 

Increased fluxes of metabolizable organic carbon to the sedimentary pool will simply shift the chemical end-products towards the more reduced species in the lower right-hand side of the figure (i.e. towards CH₄, NH₄⁺ and HS). The arrows leading to the refractory organic matter box represent the possible humification processes discussed above.

Figure 18.10 Flow chart for organic matter degradation and production of mobile end products in marine sediments

An important general conclusion from Figure 18.10 is that an increased flux of reduced organic carbon to the sediments in the form of metabolizable plankton, sea grasses, and similar materials will result in an increased flux of mobile reduced chemicals towards the sediment-water interface.

Microbial Oxidation of the Chemical Products of Organic Matter Degradation 

As a result of the eutrophication of coastal waters, an increased input of metabolizable organic matter has caused an increase in areas and depths of anoxic sediments, and qualitative as well as quantitative changes of decomposition processes. As free oxygen is consumed, a gradual shift to anaerobic respiration utilizing NO₃ , SO₄², CO₂ and possibly PO₄³ (Herrera, 1981) leads to the formation of NH₄⁺, NO₂, N₂O, N₂, HS, S⁰, HCO₃ , CH₄ and possibly PH₃.Organic carbon is partly reduced and partly oxidized (fermentation), resulting commonly in R-CH₂OH, R-COH and R-COOH. A high fatty acid content of anoxic marine sediments may lower the pH slightly.

These reduced and low molecular weight products of anaerobic microbial metabolism eventually reach the sedimentwater interface, where they become available to biological or chemical oxidation with free oxygen. As products of incomplete oxidations these compounds represent biochemical sources of energy that may be used for `chemosynthetic secondary production' of organic carbon (Jannasch, Chapter 19, this volume). This recycled portion of `photosynthetic primary production' has not yet been sufficiently quantified.

Due to the relatively large quantities of reduceable sulphur contained in sea-water, the microbial oxidation of HS may well account for most of the chemosynthesis in marine sedimentwater interfaces. In the case of geothermically produced HS at hydrothermal deep sea vents, the resulting chemosynthesis must be described as primary production and has been observed to support substantial populations of invertebrates (Jannasch, Chapter 19, this volume). The ratio of biological versus chemical oxidation of HS is apt to vary greatly, depending on the chemical/biochemical conditions of the medium and on the types of micro-organisms present.

Intracellular and extracellular elemental sulphur is produced in considerable quantities by photosynthetic and some chemosynthetic bacteria. The accumulation of elemental sulphur in anoxic environments (Jannasch, Chapter 19, this volume) has long been an enigma, but has recently been explained by observations of anaerobic sulphur reduction by a number of newly described micro-organisms (Widdel, 1980). A short-cut sulphur cycle (S² = S⁰) with a concomitant carbon cycle (CO₂ = acetate) has been found in a two-species mixed culture of Chlorobium sp. and Desulfuronomas acetoxidans (Pfennig and Biebl, 1976). In the absence of oxygen, certain strains of Beggiatoa have also been observed to reduce elemental sulphur to hydrogen sulphide (Nelson and Castenholz, 1981).

Ammonia is either assimilated in plants or micro-organisms (NH₄⁺ NH₂⁺) or oxidized by nitrifying bacteria (NH₄⁺ NO₂ NO₃) with the concomitant reduction of CO₂ to organic carbon. Nitrification is of considerable importance with respect to: (a) the utilization of inorganic nitrogen in the reduced or oxidized form for plant and microbial metabolism; (b) certain chemical characteristics, e.g., solubility and mobility, of the two forms of nitrogen; and (c) the loss of combined nitrogen by denitrification that has to be preceded by nitrification. So far there is only one report (Horrigan, 1981) where the secondary production of biomass by nitrification is believed to be significant.

The oxidation of methane by methylotrophic bacteria is a common occurrence in soil, fresh-water and marine environments (Rudd and Taylor, 1980). Reduced organic carbon products of fermentation and readily oxidized aerobically by a high diversity of heterotrophic bacteria. Some of the compounds, however, may be very stable in the absence of free oxygen and may become part of the geopolymer formation.

REFERENCES

Baxter, M. S., Stenhouse, M. J., and Drdarski, N. (1980) Fossil carbon in coastal sediments, Nature, 287, 35-36.

Bemer, R. A. (1974) Kinetic models for the early diagenesis of nitrogen, sulphur, phosphorus and silicon in anoxic marine sediments, in Goldberg, E. D. (ed.) The Sea, Vol. 5, New York, Wiley, 427-450.

Chanton, J. P., Martens, C. S., and Kipphut, G. W. (1983) Lead-210 sediment geochronology in a rapidly changing coastal environment, Geochim. Cosmochim. Acta (in press).

Hedges, J. I., and Mann, D. C. (1979) The lignin geochemistry of marine sediments from the southern Washington coast, Geochim. Cosmochim. Acta, 43, 1809-1818. 

Hedges, J. I., and Parker, P. L. (1976) Land-derived organic matter in surface sediments from the Gulf of Mexico, Geochim. Cosmochim. Acta, 40, 1019-1029.

Herrera, H. (1981) Evolution of gaseous phosphorus from the soil of a tropical rain forest in Amazonia, Paper presented at 5th International Symposium on Environmental Biogeochemistry, Stockholm, June 1-5,1981.

Jannasch, H. W. Interactions between the carbon and sulphur cycles in the marine environment, Chapter 19, this volume.

Martens, C. S. (1983) Radiocarbon tracing of organic matter degradation in coastal sediments, (in preparation).

Martens, C. S., and Klump, J. V. (1980) Biogeochemical cycling in an organic-rich coastal marine basin-1. Methane sediment-water exchange processes, Geochim. Cosmochim. Acta, 44, 471-490.

Murray, J. W., Grundmanis, V., and Smethie, Jr, W. M. (1978) Interstitial water chemistry in the sediments of Saanich Inlet, Geochim. Cosmochim. Acta, 42, 1011-1026.

Pfennig, N. and Biebl, H. (1976) Desulphuromonas acetoxidans gen. nov. sp. nov., a new anaerobic, sulphur-reducing, acetate oxidizing bacterium, Arch. Microbiol., 110, 3-12.

Sansone, F. J., and Martens, C. S. (1981) Methane production from acetate and associated methane fluxes from anoxic coastal sediments, Science, 211, 707-709. 

Scharpenseel, H. W. (1979) Soil fraction dating, in Berger, R. and Suess, H. E. (eds) Radiocarbon Dating, Proc. Ninth Int. Conf., 1976, Berkeley, University of California Press, 277-305.

Turekian, K. K., Benoit, G. J., and Benninger, L. K. (1980) The mean residence time of plankton-derived carbon in a Long Island Sound sediment core: a correction, Estuarine Coastal. Mar. Sci, 11, 583.

Widdel, F. (1980) Anaerober Abbau von Fettsaureu and Benzeosaure durch neu isolierte Arten Sulfat-reduzierender Bakterien, Dissertation, Universität Gottingen, FRG.