SCOPE 48 - Sulphur Cycling on the Continents

3.1 INTRODUCTION

Estimates of man-made sulphur emissions to the atmosphere fall into a relatively narrow range of a 93 ± 15 Tg (S) a^-1 (2.9 ± 0.47 Tmol a^-1) (Cullis and Hirschler, 1980). In contrast, the characteristics of the natural biogeochemical sulphur cycle in the atmosphere-biosphere-ocean system are much less well known. The existence of natural sources of atmospheric sulphur compounds has long been known, particularly for such source processes as volcanic emissions of SO₂ and emission of H₂S from anaerobic marshlands and estuaries (Jaeschke et al., 1978). In the last 15 years, a number of additional sulphides have been discovered in the atmosphere: dimethyl sulphide (DMS; Hill, Aneja and Felder, 1978; Andreae et al., 1985), carbonyl sulphide (COS; Hanst et al.,1975), and carbon disulphide (CS₂; Sandalls and Penkett, 1977). Other mercaptans have been identified over cattle feed lots, where they arise from rotting manure (Burnett, 1969; Stephens, 1971). These discoveries have shifted the emphasis away from H₂S and have led towards a much better understanding of biogenic sulphur emissions.

3.2 BIOLOGICAL PROCESSES LEADING TO BIOGENIC SULPHUR FLUXES TO THE ATMOSPHERE

Reduced volatile sulphur compounds, which are released to the oxygen-rich atmosphere, are chemically oxidized during their atmospheric lifetime and end up finally as sulphur dioxide (oxidation state +4) and sulphuric acid and particulate sulphate (oxidation state +6) and methane sulphonate (oxidation state +6; Andreae et al., 1988; Saltzman et al., 1986). It is mainly these compounds that are removed from the atmosphere and brought back to the earth by dry and wet deposition.

Since the oxidation state +6 is the most stable under oxic conditions, sulphate is the predominant form of sulphur in oxic waters and soils. Thus, the reduction of sulphate to a more reduced sulphur species is a necessary prerequisite for the formation of volatile sulphur compounds and their emission to the atmosphere. Biochemical processes which lead to this reduction can be considered as the driving force of the atmospheric sulphur cycle. A simplified, conceptual overview of the global biogeochemical cycle of sulphur is shown in Figure 3.1. The global environment is subdivided into four compartments: atmosphere, biosphere, hydrosphere, and lithosphere, the latter including sediments and rocks of the earth's crust. Two types of biochemical pathways of sulphate reduction are important in the global cycles: dissimilatory and assimilatory sulphate reduction. The influence of these two pathways on sulphur cycling between biosphere and atmosphere is shown in Figure 3.2.

Figure 3.1 Interactions in the global biogeochemical sulphur cycle

Figure 3.2 Scheme of the microbiological cycle of sulphur and its possible influence on the atmosphere

3.2.1 DISSIMILATORY SULPHATE REDUCTION

Dissimilatory reduction of sulphate is a strictly anaerobic process which takes place only in anoxic environments. Sulphate-reducing bacteria reduce sulphate and other sulphur oxides to support respiratory metabolism, using sulphate as a terminal electron acceptor instead of molecular oxygen (see chapters by Howarth and Stewart, Cooke and Kelly, and Giblin and Wieder , this volume). Since the process is strictly anaerobic, dissimilatory sulphate reduction occurs largely in stratified, anoxic water basins and in sediments of wetlands, lakes, and coastal marine ecosystems. The process is particularly important in marine ecosystems, including salt marshes, because sulphate is easily available due to its high concentration in seawater (28 mM; 900 g (S) l^-1). The interface between oxic and anoxic regimes is called the redoxcline and is indicated in Figure 3.1 by a dashed line through the biosphere and hydrosphere compartments. The redoxcline is maintained by physical barriers to oxygen movement and by rapid rates of oxygen consumption. As can be seen from Figure 3.2, dissimilatory sulphate reduction is the major pathway for the production of H₂S globally.

Under favourable conditions, the rate of dissimilatory sulphate reduction to H₂S in anoxic environments can be quite high, ranging up to 100 mmol m^-2 day^-1 (3 mg (S) m^-2 day^-1 ) and more in shallow near-shore marine sediments and in salt marshes (Jørgensen, 1982; Howarth, 1984; Ivanov et al., 1989). However, since the occurrence of this process is dependent on the existence of a mixing barrier which prevents oxygen from entering the system, the escape of H₂S from the system will be limited by the same barrier. Most of the sulphate reduced is re-oxidized before ever leaving the sediment (Howarth, 1984; Giblin and Wieder, this volume; Luther and Church, this volume). In the presence of oxygen, H₂S and other reduced sulphur compounds provide excellent substrates for microbial oxidation from which certain bacteria can obtain a substantial amount of energy (Howarth, 1984, and references therein). Such mircoorganisms tend therefore to be present in high numbers at the oxic-anoxic interface. They are very efficient in removing H₂S and can completely oxidize this compound in a layer only a fraction of a millimetre thick (Jørgensen and Revsbech, 1983). Consequently, the very large amounts of H₂S which are produced in coastal marine ecosystems cannot usually be transferred to the atmosphere (Andreae, 1986, and references therein; Giblin and Wieder, this volume) , but are either re-oxidized at the oxic/anoxic interface, or precipitated in the form of iron sulphides. In spite of this limitation, near-shore environments like estuaries and salt marshes are the major source of H₂S emissions from the oceans.

3.2.2 ASSIMILATORY SULPHATE REDUCTION

In contrast to animals, which are dependent on organosulphur compounds in their food to supply their sulphur requirement, other biota (bacteria, cyano bacteria, fungi, eukaryotic algae, and vascular plants) can obtain sulphur from assimilatory sulphate reduction for synthesis of organosulphur compounds (see review by Anderson, 1980). Sulphate is assimilated from the environment, reduced inside the cell, and fixed into sulphur-containing amino acids and other organic compounds. The process is ubiquitous in both oxic and anoxic environments. Most of the reduced sulphur is fixed by the intracellular assimilation process and only a minor fraction of the reduced sulphur is released as volatile gaseous compounds as long as the organisms are alive. However, after death of the organisms, microbial degradation liberates reduced sulphur compounds (mainly in the form of H₂S, but also as organic sulphides) to the environment. During this stage, volatile sulphur compounds may escape to the atmosphere. However, as with sulphides formed from dissimilatory sulphate reduction, the sulphides released during decomposition are chemically unstable in an oxic environment and are re-oxidized to sulphate by a variety of microorganisms (Figure 3.2; Schoenau and Germida, this volume).

The biochemistry of assimilatory sulphate reduction has been studied mostly in the green alga Chlorella. Therefore, most of the following discussion refers specifically to this organism. It is not altogether clear at this time how far these conclusions can be generalized to other organisms.

The assimilation of sulphate to cysteine, the first organosulphur metabolite produced, is a complex, multi-step process (Figure 3.3). Sulphate is taken up into the cell by an active transport mechanism, and inserted into an energetically activated molecule, APS (adenosine-5'-phosphosulphate), which can be further activated at the expense of one more ATP molecule to PAPS (3'- phosphoadenosine-5'-phosphosulphate). It is then transferred to a thiol carrier (RSH) and reduced to the -2 oxidation state. In contrast to nitrate assimilation, where the various intermediates are present free in the cytoplasm, sulphur remains attached to a carrier during the reduction sequence. In a final step, the carrier-bound sulphide reacts with O-acetyl-serine to form cysteine. Wilson, Bressan and Filner (1978) have suggested that under conditions when availability of this or other endogenous sulphide acceptors is limiting the rate of cysteine synthesis, the volatilization of H₂S could serve as a mechanism for removing excess reduced sulphur. Such volatilization has been observed from vascular plants (Winner et al., 1981), but its possible occurrence in marine algae has yet to be investigated.

 Figure 3.3  Major metabolic pathways of sulphur in algae and plants. The percentages represent the approximate distribution of the organosulphur compounds in Chlorella

Cysteine serves as the starting compound for the biosynthesis of all other sulphur metabolites, especially the sulphur-containing amino acids homocysteine and methionine (Figure 3.3). Cysteine and methionine are the major sulphur amino acids in plants and usually represent a very large fraction of the sulphur content of biological materials (Giovanelli, Mudd and Datko, 1980). Glutathione (L-glutamyl-L-cysteyl-L-glycine) plays a variety of biochemical roles, including redox transfer reactions and the removal of H₂O₂ in chloroplasts. Methionine reacts with ATP to form S-adenosyl-methionine (SAM), the most important methyl group donor in methyl group transfer reactions in plants and algae. Transfer of a methyl group from SAM to methionine yields S-methyl-methionine, the precursor of dimethylsulphide in plants. In some marine algae, dimethylsulphonium propionate (DMSP) is formed in a multi-step process from methionine.

Information about the biochemical precursors of volatile sulphur compounds has been obtained from artificial culture studies (reviewed by Kadota and Ishida, 1972), and from incubation studies of natural and amended soils (reviewed by Bremner and Steele, 1978). A compilation of the most prominent biochemical precursors of volatile sulphides produced in soils by microbial degradation of organic matter under aerobic and anaerobic conditions is given in Table 3.1.

Table 3.1 Biochemical origin of volatile sulphides produced in soils by the microbial degradation of organic matter under aerobic and anaerobic conditions (Warneck, 1988)

3.3 BIOGENIC SULPHUR EMISSIONS FROM THE OCEANS

In  recent years, the knowledge about biogenic sulphur emissions from the oceans has improved considerably and now there is a reasonably consistent picture of the types and amounts of sulphur gases emitted from the sea and their fate in the atmosphere. The main gases which are released from the ocean surface are dimethylsulphide (DMS), H₂S, carbon disulphide (CS₂), and carbonyl sulphide (COS). In this section, we concentrate on emissions from open, oceanic waters. Biogenic sulphur fluxes from salt marshes are discussed by Giblin and Wieder (this volume) .

3.3.1 DIMETHYLSULPHIDE

In open ocean waters, DMS is the predominant volatile sulphur compound, a result of its association with phytoplankton production (Barnard et al., 1982; Nguyen, Bergeret and Lambert, 1984; Bates et al., 1987; Turner et al., 1988; Turner, Malin and Liss, 1989). The main precursor of DMS in algae is DMSP, a ternary sulphonium compound that is thought to be generated from methionine (Challenger, 1951). Although DMSP is known to undergo enzymatic cleavage to DMS in the living algal cell, not all DMS appears to be released directly. Many observations suggest that DMS in aquatic environments derives to a significant extent from the bacterial decomposition of DMSP leaked from aged cells, or from zooplankton grazing on these cells. For example, as discussed by Bremner and Steele (1978), very high concentrations of DMS have been observed during the bacterial putrefraction of algae following algal blooms. 

Figure 3.4 shows a typical vertical distribution of particulate (intracellular) DMSP, dissolved DMSP and DMS, and chlorophyll (an indicator of phytoplankton biomass) in the marine water column for the example of data from the northwestern Atlantic. The vertical distribution of DMS and DMSP in seawater as shown in Figure 3.4 is typical for these compounds as well as for a number of other phytoplankton metabolites, e.g. dimethylsulphoxide (DMSO) and the methylarsenates (Andreae, 1979, 1980). The characteristic features of this distribution are a maximum at or a few metres below the sea surface, and a sharp decrease in DMS concentration near the level of 1% light transmission. This depth represents the lower limit of growth for most phytoplankton species.

Figure 3.4 Typical vertical distribution of particulate DMSP, dissolved DMSP and DMS during the April/May 1986 cruise of R/V Columbus Iselin in the northwestern Atlantic Ocean. The vertical distribution of chlorophyll a is also included because it indicates the stratification of phytoplankton biomass 

Volatile substances like DMS are transferred across the air/sea interface by a combination of molecular and turbulent diffusion processes, which are still poorly understood and for which no entirely satisfactory physical and mathematical models are available. For a discussion of the state of the art in this field, see the review by Liss and Merlivat (1986). The sea-to-air flux is proportional to the air/sea concentration difference. The atmospheric concentration of DMS is several orders of magnitude below the value in equilibrium with seawater predicted by Henry's law (Andreae et al., 1985). It can therefore be ignored for the purpose of estimating the sea-to-air concentration gradient and only the concentration of DMS in seawater is required to estimate the emission flux of DMS.

Table 3.2 summarizes the distribution of DMS in the world's oceans as measured on numerous cruises conducted by several groups (Barnard et al., 1982; Andreae, Barnard and Ammons, 1983; Cline and Bates, 1983; Andreae and Barnard, 1984; Nguyen, Bergeret and Lambert, 1984; Bingemer, 1984; Bates et al., 1987). Table 3.2 is based on the compilation of Andreae (1986). The data are organized by biogeographical regions as defined by Koblentz- Mishke, Volkevinsky and Kabanova (1970); averages for each of these regions are used together with an estimate of their areal extent for the prediction of the flux of DMS from each region. The data base used for Table 3.2 contains no measurements from the Southern Ocean; recent work by Berresheim (1987) has shown, however, that oceanic emissions of DMS in this region are similar to those found in temperate regions.

  Table 3.2  Concentrations and fluxes of DMS, CS₂ and COS for the world oceans

To obtain the DMS transfer velocities used in the flux calculations in Table 3.2, the radon transfer velocities of Peng et at. (1979) and of Smethie et al. (1985) were adjusted by assuming that the transfer velocity is proportional to the square root of the diffusivity. If we use the global average ¹⁴CO₂ transfer velocity (c. 21 cm h^-1; Liss and Merlivat, 1986), we obtain a significantly higher flux: a mean estimate of 51 instead of 38 Tg (S) a^-1 (1.6 instead of 1.2 Tmol a^-1). This is probably due to the fact that the radon transfer velocity is based almost entirely on summer data, when wind speeds are lower. In view of the extensive data on DMS concentrations in the surface ocean (Table 3.2), it appears that the major error in estimating the sea-to-air flux of DMS rests in the uncertainties associated with the use of the 'stagnant-film' model, and in particular with the assumption of the transfer velocities. This uncertainty may be as large as a factor of two.

There are only relatively small differences in total DMS fluxes from different oceanic regions (Table 3.2). One reason is that the tropical regions have relatively high DMS concentrations year-round, whereas in temperate regions, especially the temperate coastal areas, there is a pronounced seasonality with quite low values during the cold season. Secondly, the large areas of low and moderate biological productivity contribute amounts of DMS to the atmosphere comparable to those from the relatively small regions of high productivity in the upwelling regions and the coastal areas. This is in contrast to earlier views (Graedel, 1979), which had assumed that the biogenic sulphur flux from the oceans would be dominated by localized 'hot spots' of biological productivity.

We find that the estimate for the global ocean DMS flux has by now become very stable, in the sense that the addition of new data even from a number of different groups does not change it significantly. Whereas the number of data points in Table 3.2 is about 2.5 times greater than in a comparable table in Andreae and Raemdonck (1983), the estimate for the global mean DMS concentration has only changed from 3.2 to 3.1 nmol l^-1 (102 to 99 ng l^-1), and that for the global flux remains unchanged at c. 38 Tga^-1 (1.2 Tmol a^-1). Using a data set from the Pacific Ocean only, Bates et al. (1987) obtain a lower mean DMS concentration (c. 1.8 nmoll-1; 58 ng l^-1), and a correspondingly lower flux of 16 Tg a^-1 (0.5Tmol a^-1) with an estimated uncertainty of a factor of two. In view of the large uncertainties associated with the 'stagnant-film' model, it seems very important that independent methods should be developed to test the predictions based on this model. However, alternative methods to determine the flux, e.g. the eddy- correlation or gradient techniques, still face large experimental difficulties. No rapid-response sensor which would make the eddy-correlation technique possible is available for DMS, or in fact any of the reduced sulphur gases. The gradient method has been used on board ship by Bingemer (1984) and by Nguyen, Bergeret and Lambert (1984) by sampling at different levels above the waterline. While the results compare well to predictions from gas transfer calculations, they may contain substantial error due to the influence of the ship on the air flow characteristics. Due to the difficulty of simulating a realistic wave climate inside a flux chamber, direct measurements of sulphur gas fluxes across the air-sea interface by the chamber technique have not been attempted.

3.3.2 HYDROGEN SULPHIDE

There are few data on the concentration of dissolved H₂S in surface seawater , and a small number of reliable measurements of H₂S in the marine atmosphere. Therefore, the air-sea exchange flux of this compound is difficult to estimate. H₂S is oxidized rapidly in oxygenated seawater; chemical half-lives on the order of a few hours are reported (Almgren and Hagstrom, 1974); other workers have found values as high as 50 hours (Chen and Morris, 1972). The most reliable measurements appear to be those of Millero et al. (1987), who found a half-life of 26 hours at 25° C. Dissolved sulphide concentrations of 0.1 to 1.0 nmol l^-1 (3 to 32 ng l^-1) have been found in surface seawater from the Atlantic Ocean (T.W. Andreae et al., 1991).

The production mechanism of this H₂S remains unclear, but its vertical distribution in the ocean suggests that bacterial reduction in microbial micro- environments may play an important role (Jørgensen, Hansen and Ingvorsen, 1978). It must, however, be remembered that biological processes in plants can result in the production and release of substantial amounts of H₂S even in the presence of oxygen. This is especially true in the presence of high ambient sulphate concentrations, as is the case in seawater. The hydrolysis of COS may also result in the production of significant amounts of H₂S in seawater (T.W. Andreae et al., 1991).

 H₂S has been observed in the marine atmosphere at levels of a few pptv to a few tens of pptv (Slatt et al., 1978; Delmas and Servant, 1982; Herrmann and Jaeschke, 1984; Cooper and Saltzman, 1987; T. W. Andreae et al., 1991). All authors used the same method of trapping the H₂S on AgNO₃-impregnated filters and subsequent determination of the enriched Ag₂S by the quenching of the fluorescence of fluorescein mercuric acetate (Jaeschke and Herrmann, 1981). However, Cooper and Saltzman (1987) found a positive interference in the determination of H₂S by this method and suggested that the mean concentration of H₂S in the marine boundary layer does not exceed 10 pptv :Saltzman and Cooper, 1988; T.W. Andreae et al., 1990). At these levels, H₂S in the atmosphere is near thermodynamical equilibrium with the concentrations in surface seawater (T. W. Andreae et al., 1991) .

 To obtain an estimate of the rate of H₂S oxidation in the marine atmosphere, we can simply use an average concentration of 6 pptv (Saltzman and Cooper, 1988; T .W .Andreae et al., 1990, 1991) with a scale height of 2 km,.a diurnal average OH concentration of 8 X 10⁵ mol cm^-3, and the measured reaction rate for the oxidation of H₂S by OH (5 X 10^-12 cm^-3 mol^-1 S^-1; Cox and Sheppard, 1980). The resulting estimate, 0.5 Tg (S) a^-1 (0.015 Tmol a^-1), is an upper limit for the sea-to-air flux of H₂S, and is much smaller than the DMS flux of c. 38 Tg (S) a^-1 (1.2 Tmol a^-1). Based on their measurements in the Caribbean and the Gulf of Mexico, Saltzman and Cooper (1988) suggest that the oxidation of H₂S accounts for only 11% of the production of biogenic non-seasalt sulphate in the remote marine boundary layer, the rest being due to the oxidation of DMS. It is not clear, however, if the source of the H₂S found in the marine troposphere is necessarily the ocean surface or if other processes could be responsible for its presence. For example, advection from coastal regions, where H₂S is emitted from salt marshes (Giblin and Wieder, this volume), may supply some of this H₂S. This is supported by a correlation between H₂S and radon observed over the northern Atlantic (T.W. Andreae et al., 1991). Based on simultaneous measurements of H₂S in seawater and air, T. W. Andreae et al. (1991) estimate that the oceanic emission of H₂S is less than 0.3 Tg (S) a^-l (0.01 Tmol a^-l). On the other hand, McElroy, Wofsy and Sze (1980) have speculated that atmospheric reactions of COS and CS₂ with OH radical could produce the necessary amounts of H₂S. However, this suggestion has not yet been verified experimentally.

3.3.3 CARBON DISULPHIDE

The presence of CS₂ in seawater was first observed by Lovelock (1974), who measured an average concentration of 14 pmol CS₂ l^-1 (0.45 ng (S) l^-1) in 35 samples taken in the open Atlantic Ocean. Inshore values were about an order of magnitude higher. Turner and Liss (1985) also report the presence of high levels of CS₂ in coastal waters of England, but they give quantitative information for only a few samples with values near 300 pmol CS₂ l^-1 (9.6 ng (S) l^-1). They found substantially higher concentrations in the low salinity region of an estuary (upto c. 2 nmol CS₂ l^-1; 64 ng (S) l^-1). It is possible that much of the CS₂ found in coastal waters is the result of the diffusion of this substance from the porewater of the underlying sediments. This would be consistent with the relatively high concentrations and fluxes of CS₂ observed in coastal marsh environments (Adams et al., 1981a,b; Steudler and Peterson, 1984). CS₂ could be formed there either by fermentation reactions of organosulphur compounds or by 'pulp-mill' type reactions of terrestrial plant matter , including dead salt-marsh grasses with dissolved polysulphides originating from bacterial dissimilatory sulphate reduction (see Luther and Church, this volume). Kim and Andreae (1987) have recently determined CS₂ in open ocean and coastal seawater from the North Atlantic, and have observed mean concentrations of 16 ± 8 and 33 ± 19 pmol CS₂ l^-1 respectively (0.51 ± 0.25 and 1.1 ± 0.6 ng (S) l^-1) (Table 3.2), somewhat higher than Lovelock's results.

From these data, we estimate a flux of c. 0.23 Tg (S)a^-1 (0.007 Tmol a^-1) in the form of CS₂ from the world ocean surface, about 0.6% of the DMS flux. The photochemical oxidation of CS₂ produces one molecule each of SO₂ and COS per molecule of CS₂ oxidized. Thus, the marine emission of CS₂ provides a significant indirect source of COS  whereas it is clearly inconsequential as a source of tropospheric SO₂ or sulphate.

3.3.4 CARBONYL SULPHIDE

Carbonyl sulphide (COS) is the most abundant atmospheric sulphur species in the remote troposphere, with an average concentration near 500pptv .Due to its low reactivity in the troposphere, and its correspondingly long residence time (of the order of one year), it is the only sulphur compound which can enter the stratosphere (with the exception of SO₂ injections during violent volcanic eruptions). The input of COS is considered responsible for the maintenance of the sulphate aerosol layer in the stratosphere during volcanically quiescent periods (Crutzen, 1976; Jaeschke, Schmidt and Georgii, 1976). Therefore, even a relatively small COS source flux can be of considerable importance in atmospheric chemistry.

COS is present in surface seawater at concentrations of about 0.03 to 1.0 nmol l^-1 (1 to 32 ng l^-1) (Rasmussen, Khalil and Hoyt, 1982; Ferek and Andreae, 1983,1984; Turner and Liss, 1985; Andreae, 1991b; T. W. Andreae et al., 1991). The observed concentrations are almost always higher than the equilibrium concentration relative to the overlying atmosphere, so that a net sea-to-air flux exists essentially from the entire ocean surface. Johnson (1981) has speculated that the ocean should be a sink for COS due to its hydrolysis at the slightly alkaline pH of seawater. This suggestion is clearly not supported by the measured COS supersaturation ratios across the air-sea interface.  Pronounced diurnal variations of the COS concentration in surface seawater (Figure 3.5) suggest that COS is produced there by photochemical reactions (Ferek and Andreae, 1984). Laboratory experiments with seawater and with solutions of organosulphur compounds in distilled water showed that seawater sulphate does not participate in the reaction, and that only the presence of dissolved organic sulphur compounds, dissolved oxygen, and light are necessary to produce COS (Ferek and Andreae, 1984; Zepp and Andreae, 1989, 1990). Carbonyl sulphide is formed by irradiation of a variety of organic sulphur compounds commonly found in biological materials, e.g. cysteine, methionine, glutathione, and dimethylsulphonium propionate. The mechanism to this reaction is not known at this time, but it is likely that short-lived, photochemically produced radicals (e.g. OH) are involved. The presence or absence of living microorganisms like planktonic algae or bacteria is of no influence on the rate of formation of COS in seawater. It appears that the role of organisms in the production of COS in seawater is limited to the synthesis of dissolved organic sulphur compounds which are then abiotically photolysed to COS. The dependence of the rate of COS formation on the concentration of dissolved organic sulphur in seawater is reflected by the difference between the COS supersaturation measured in coastal and open ocean waters (Figure 3.5).

Figure 3.5  Mean diurnal variation of COS in surface seawater during a cruise of R/V Bellows in November 1983. The concentration of COS is indicated as a saturation ratio, i.e. the ratio between the measured concentration and the concentration in equilibrium with ambient air with about 500 pptv COS

An attempt to obtain a representative estimate of the sea-to-air flux of COS is also presented in Table 3.2, where the ocean surface is divided into the same biogeographic regions as used for DMS. Then, based on the diurnally averaged data on the supersaturation of COS in the surface seawater relative to the overlying atmosphere, and the average temperature of the surface ocean in these regions, the flux of COS across the air-sea interface for these regions was calculated (the piston velocities for COS are a factor of 1.3 higher than for DMS, due to the higher diffusivity of COS). We see that, in contrast to DMS, the flux of COS is dominated by the high productivity regions, especially the coastal and shelf areas. Due to the low levels of COS in oligotrophic areas, they contribute little to the global flux, which is estimated to be c. 0.35 Tg (S) a^-1 (0.11 Tmol a^-1), similar to previous flux estimates (Rasmussen, Khalil and Hoyt, 1982; Ferek and Andreae, 1983).

3.4 BIOGENIC EMISSION FROM FRESHWATER WETLANDS AND INLAND SOILS

In contrast to the oceanic emissions, our knowledge about biogenic sources of volatile sulphur compounds from the continents is extremely limited. At this time, the terrestrial biogenic sources represent the largest uncertainty in the global atmospheric sulphur budget (Andreae, 1985a, 1986). This uncertainty is related to several problems: (1) the diversity of the sulphur species emitted, which include hydrogen sulphide (H₂S), dimethylsulphide (DMS), dimethyldisulphide (DMDS), methylmercaptan (MeSH), carbon disulphide (CS₂), carbonyl sulphide (COS), and others; (2) the diversity of terrestrial ecosystems, including freshwater wetlands; (3) the patchiness in time and space of determined sulphur gas fluxes from terrestrial ecosystems; (4) analytical problems with some of the most important species, especially H₂S, as well as problems with flux measurement methodology; and (5) the logistical problem of obtaining representative measurements in several important regions, especially with respect to variations caused by climatical and microbiological circumstances at the chosen measuring site.

As an example, in Figure 3.6 the diurnal variation of measured H₂S flux is shown, as determined by Jaeschke, Claude and Herrmann (1980) in a freshwater marshland region in northern Germany. After sunset the H₂S flux rose and reached a maximum shortly before sunrise. This variation was observed on several occasions during intensive studies of H₂S fluxes in German wetlands. In the example shown in Figure 3.6, the H₂S flux varied during the 24 h measuring period by a factor of 1000. If the flux had been determined only during one or two hours, it would have been highly misleading.

The concentrations of dissolved volatile sulphur compounds in freshwater bodies tend to be much lower than those in the surface layer of the oceans (Adams et al., 1981a,b; Froelich et al., 1985; Iverson, Nearhoof and Andreae, 1989). However, based on measurements of DMS concentrations in waters from freshwater wetlands in Ontario, Nriagu, Holdway and Coker (1987) recently proposed a DMS emission rate of 150 ng (S) m^-2 min^-l (4.7 nmol m^-2 min^-l) for such ecosystems, and suggested that H₂S emissions of a similar magnitude may be occurring. These fluxes may, however, have been overestimated since the exchange coefficient appropriate for the hydrodynamic regime of a bog is not known, and a rather high value was selected. It is also possible that the very high sulphate loading from anthropogenic pollution may have enhanced sulphur fluxes from these wetlands. Recently, Staubes (1986) has studied the temperature dependence of biogenic sulphur exhalation at several sites over Germany. These studies suggest extremely low emissions of sulphur volatiles from bog and wetlands in cold northerly regions.

Figure 3.6  Diurnal variation of the H₂S flux measured at marshland soils in the Ems River region of northern Germany

Most studies of biogenic sulphur emissions from terrestrial ecosystems have concentrated on H₂S. Delmas et al. (1980) measured an average flux of 134 ng (S) m^-2 min^-1 (4.2 nmol m^-2 min^-1 ) from oxic lawn soils in France. Jaeschke, Claude and Herrmann (1980) found fluxes between 2 and 2000 ng (S) m^-2 min^-1 (0.06 to 63 nmol m^-2 min^-1 ) from marshland soils in the Ems River region of northern Germany (Figure 3.6). Somewhat higher average fluxes were determined from tropical soils in the Ivory Coast by Delmas et al. (1980). They estimated the mean annual flux from these soils to be in the range of 600 to 1700 ng (S) m^-2 min^-1 (19 to 53 nmol m^-2 min^-1 ).

In the first comprehensive study of sulphur gas emissions from inland soils, Adams et al. (1981a,b) determined the fluxes of H₂S, COS, MeSH, OMS, CS₂, and dimethyldisulphide (DMDS) from soils at 35 sites in the eastern and southeastern United States. Their flux data showed large variability between sites and between sampling periods at the same site (several orders of magnitude). For non-saline soils, fluxes ranged from 3.8 to 640 ng (S) m^-2 min^-1 (0.12 to 20 nmol m^-2 min^-1 ). The average flux from inland soils was c. 38 ng (S) m^-2 min^-1 (1.2 nmol m^-2 min^-1 ) for the latitude zone from 25° N to 47° N. Most of the flux from inland soils was in the form of H₂S (66%); the rest was accounted for by COS (13%), CS₂ (13%), DMS (7%), and a trace of DMDS (2%). MeSH was not found to be emitted from non-saline soils.

Most recent studies have shown much lower flux estimates from soils. In a large study of soil emissions conducted partly at the same sites as had been used by Adams and co-workers, values about ten times lower than those reported previously were found by Goldan et al. (1987) and Lamb et al. (1987). The differences between these values and those reported previously are largely attributed to experimental problems in the older studies. Staubes, Ockelmann and Georgii (1986) measured DMS and CS₂ fluxes from soils in Germany. For soil temperatures above 21°C, they observed DMS fluxes in the range of 6.4 to 16 ng (S) m^-2 min^-1 (0.2 to 0.5 nmol m^-2 min^-1) and CS₂ fluxes in the range of 3.2 to 6.4 ng (S) m^-2 min^-1 (0.1 to 0.2 nmol m^-2 min^-1). In the rainforest at La Selva, Costa Rica, Haines et al. (1984, pers. comm. quoted by Delmas and Servant, 1988) measured a mean flux of 4 ng (S) m^-2 min^-1 (0.1 nmol m^-2 min^-1) for DMS, with possibly some contribution by methylmercaptan.

Andreae and Andreae (1988) made some preliminary measurements of soil emissions in wet tropical forests using a chamber technique at the Ducke Reserve near Manaus, Brazil during the Amazon Boundary Layer Experiment (ABLE-2). During the 1985 dry season (ABLE-2A), fluxes of DMS were 12.8 ± 1.3 ng (S) m^-2 min^-1 (0.4 ± 0.04 nmol m^-2 min^-1 ), fluxes of MeSH were 1.3 ± 0.3 ng (S) m^-2 min^-1 (0.04 ± 0.009 nmol m^-2 min^-1 ), and fluxes of H₂S were 2.2 ± 1.0 ng (S) m^-2 min^-1 (69 ± 0.03 nmol m^-2 min^-1). These values combine to a total flux of short-lived sulphur species (ignoring CS₂, COS, etc.) of 16 ± 2.6 ng (S) m^-2 min^-1 (0.5 ± 0.08 nmol m^-2 min^-1), in good agreement with the soil emission fluxes obtained by the authors cited above. During the 1987 wet season (ABLE-2B), even lower values were observed. However, due to the small number of measurements and sites involved, our ability to extrapolate from these data is very limited.

3.5 EMISSIONS BY LIVING PLANTS, AND COMBINED MEASUREMENTS OF FLUXES FROM PLANTS AND SOILS

The emission of sulphur gases from vegetation, including trees, is well established. The emission of H₂S from plant leaves is light-dependent: it increases strongly with the intensity of light flux, and drops to very low values in the absence of light (Wilson, Bressan and Filner, 1978; Filner et al., 1984). H₂S emission also increases as a result of root injury, high levels of atmospheric SO₂, and increased levels of soil-water sulphate or bisulphite. Based on studies with various crop plants, Filner et al. (1984) have estimated a world-wide emission of volatile sulphur from plants of the order of 7.4 Tg (S) a ^-1 (0.23 Tmol a^-1), whereas Winner et al. (1981) gave an estimate of 54 Tg (S)a^-1 (1.7 Tmol a^-1). Adjusting these values to the leaf areas and biomass values characteristic of tropical forests, fluxes of about 160 to 1600 ng (S) m^-2 min^-1 (5 to 50 nmol m^-2 min^-1 ) can be predicted. Filner et al. (1984) also report the emission of MeSH by plants under some conditions. The emission of DMS from trees was observed by Lovelock, Maggs and Rasmussen (1972); from their data a flux of 32 to 640 ng (S) m^-2 min^-1 (1 to 20 nmol m^-2 min^-1) can be estimated for a forest with a leaf mass of c. 2 kg (dry weight) m^-2 . In a recent study, Lamb et al. (1987) found emission fluxes in the range of 6.4 to 64 ng (S) m^-2 min^-1 (0.2 to 2 nmol m^-2 min^-1 ) from various crops and trees. The mean sulphur flux from deciduous trees was 29 ng (S) m^-2 min^-1 (0.9 nmol m^-2 min^-1 ) during summer conditions (mean sample temperature: 29.5° C). Emissions from soybeans and corn were dominated by DMS, while deciduous trees emitted H₂S and DMS in similar amounts. These fluxes are comparable to or greater than sulphur gas fluxes which have been reported from vegetated inland soils.

Several authors have used micrometeorological or budget techniques which estimate a total flux from the soil-vegetation system. In temperate latitudes, Jaeschke, Claude and Herrmann (1980) measured an H₂S flux of 5.2 ng (S) m^-2 min^-1 (0.17nmol m^-2 min^-1) from wetland areas near the German coast. In the Ivory Coast, Delmas et al. (1980) obtained H₂S fluxes of 208 and 5120 ng (S) m^-2 min^-1 (6.5 and 160 nmol m^-2 min^-1) from forests on oxic and flooded anoxic soils, respectively. Delmas and Servant (1983) reported fluxes of 960 ng (S) m^-2 min^-1 (30 nmol m^-2 min^-1) in the dry season and 4480 ng (S) m^-2 min^-l (140 nmol m^-2 min^-1) in the wet season from equatorial Africa, using aircraft data and a budget method.

Recent laboratory studies on corn, alfalfa, and wheat plants showed DMS to be the main sulphur species emitted, followed by variable amounts of MeSH, H₂S, and CS₂. The emission rates for all sulphur gases increased with temperature and illumination (Fall et al., 1988). Comparison of the emission of COS from bare and vegetated soils showed that the vegetation canopy was a sink for this gas, reducing the flux of COS from soils to the atmosphere (Goldan et al., 1987; Fall et al.,1988). The uptake of COS by vegetation has been proposed as the major global sink for this compound (Brown and Bell, 1986).

The fluxes of DMS, H₂S, and MeSH from the soil/plant system of the Amazon forest were determined during the 1985 dry season by a gradient/flux technique (Andreae and Andreae, 1988). The mean fluxes were 26 and 13 ng (S) m^-2 min^-1 (0.8 and 0.4 nmol m^-2 min^-1) for DMS and MeSH, respectively. For H₂S, an upper limit of 50 ng (S) m^-2 min^-1 (1.6 nmol m^-2 min^-1 ) was found. During the wet season a similar approach resulted in estimates of 6.4 and 35 ng (S) m^-2 min^-1 (0.2 and 1.1 nmol m^-2 min^-1) for the emission fluxes of DMS and H₂S respectively (Andreae et al., 1990a). Since these fluxes are much larger than the observed soil fluxes, most of the sulphur emissions must come from the plant canopy. In the case of DMS, the vertical profiles through the forest canopy suggest that emission takes place only during the day-time, and that the forest canopy acts as a sink for this species during the night.

In Figure 3.7, the DMS fluxes obtained from the gradient measurements are compared with fluxes estimated by a model which incorporates transport and photochemical oxidation of DMS. Both model and measurements show the same diel variation of the DMS flux, and suggest that the fluxes at night are low. In view of the uncertainties  of both the measurements and the model, the observed and predicted values agree reasonably well. Sulphur gas measurements in and over the Congo rainforest suggest that similar emission fluxes prevail in the African humid tropics (Bingemer et al., 1991).

Assuming that a sulphur emission flux of about 30 to 100 ng (S) m^-2 min^-1 (0.9 to 3.1 nmol m^-2 min^-1 ) is representative of wet tropical South America and that this flux can be extrapolated to the wet tropics world-wide, we can assess the importance of the wet tropics in the global atmospheric sulphur cycle. Using 14.5 x 10¹² m² as an estimate for the area covered by tropical forests (Cicerone et al., 1984), we predict a sulphur emission flux of 0.23 to 0.76 Tg (S) a^-1(0.007 to 0.024 Tmol a^-1 ) from thiS region. If the rest of the vegetated land surface (88 X 10¹² m²) has an emission rate as estimated by Guenther, Lamb and Westberg (1989) for the United States (including wetlands), the sulphur emissions from this area are about 3.5 Tg (S) a^-1 (0.11 Tmol a^-1). For the total land area, this corresponds to 3.8 to 4.3 Tg (S) a^-1 (0.12 to 0.13 Tmol a^-1), considerably less than the total oceanic flux of about 38 Tg (S) a^-1 (1.2 Tmol a^-1). It is also small relative to the anthropogenic emissions from fossil fuel burning (about 96 Tg (S) a^-1), so that increasing development can be expected to significantly perturb the atmospheric sulphur cycle in tropical regions (McDowell, 1987).

Figure 3.7  Observed DMS fluxes from the Amazon forest near Manaus, Brazil, during the ABLE-2B wet season experiment in July/August 1987, obtained by gradient/flux measurements compared with fluxes estimated by a model incorporating transport and photochemical oxidation of DMS (solid line)

If these observations are representative for the sulphur cycle over remote continents world-wide, an assumption which is supported by the existing data for sulphate deposition (Galloway, 1985; Andreae et al., 1990b, and references therein), the biogenic sulphur emission from continents must be near the low end of the range given in Table 3.4, and emissions from land biota must play a subordinate role in the continental sulphur cycle over the continents world-wide, being overshadowed by marine and anthropogenic sources essentially everywhere.

3.6 SULPHUR EMISSIONS FROM BIOMASS BURNING

Biomass burning cannot be classified clearly as either a natural or an anthropogenic source. Most of the biomass burning takes place as a result of agricultural practices (including tropical deforestation) and of fuel wood burning (totalling c. 8700 Tg dry matter per year; Andreae, 1991a). Since most of the burning occurs in relatively undeveloped regions, especially in the tropics, we briefly discuss its importance as a sulphur source.

The average sulphur content of land plants is c. 0.2% (Bowen, 1979), which corresponds to a potential sulphur flux of c. 17 Tg (S) a^-1 (0.53 Tmol a^-1) at a burning rate of 8700Tg (dry matter) a^-1. Not all of this sulphur enters the atmosphere, since a significant fraction is retained in the ash (35-67% in the case of the savanna fires described by Delmas (1982)). The chemical forms in which sulphur is emitted from fires and their relative proportions are still quite uncertain. It is to be assumed that SO₂ is a major product, but H₂S, COS, and sulphate aerosol have also been observed. SO₂ was found to be more abundant than particulate sulphur in emissions from a wood-burning stove (Cooper, 1980) .There are no studies at this time in which sulphur gases and aerosols have been measured together in biomass burning emissions except from highly polluted regions. The data from controlled burns in the western United States cannot be considered globally representative due to the high anthropogenic and marine sulphur deposition in this region, much of which appears to be remobilized in the burning process (Hegg et al., 1987).

The global emission rate for COS from biomass burning has been estimated to be 0.09 Tg (S) a^-1 (0.003 Tmol a^-1), only a minor fraction of the sulphur flux from fires, but a significant component in the atmospheric cycle of COS (Crutzen et al., 1979; Andreae, 1991a). Preliminary studies from our laboratory show that only a very small fraction of the fuel sulphur is emitted in the form of H₂S during biomass burning. The global emission of SO₂ from biomass burning can be estimated based on the total amount of biomass burned and an emission factor, i.e. the amount of SO₂ emitted per unit of fuel burned. Based on our studies of biomass burning in Brazil, we have obtained an emission factor of 0.32 X10^-3 (mol SO₂/mol C). From Delmas' (1982) data on the sulphur yield in savanna fires, an emission factor of 0.16 x 10^-3 can be calculated, somewhat lower than our estimate.

The rate of sulphur emission from biomass burning, derived from measurements over Amazonia, corresponds to a global source flux of 2.8 Tg (S) a^-1 (0.088 Tmol a^-1). This estimate is lower than previous ones, e.g. that of Andreae (1985a), who had estimated a global flux of 7 Tg (S) a^-1 (0.2 Tmol a^-1), based on a global burning rate of 6.8 Tg dry matter per year, a sulphur content of 0.2 wt. %, and a burning yield of 50%. The value of 2.8 Tg (S) a^-1 appears to be more reliable, however, since it is based on actual field measurements.

In comparison to the global emissions of sulphur from the land and ocean biota, c. 4 and 38 Tg (S) a^-1 (0.13 and 1.2 Tmol a^-1) respectively, and from fossil fuel burning, biomass burning may be considered a minor source of atmospheric sulphur on a global scale. However, the fact that biomass burning occurs predominantly in tropical regions during a defined burning season suggests that it may be the dominant sulphur source to the atmosphere in the remote tropics during a significant part of the year. Observations from Africa (Lacaux et al., 1988; Bingemer et al., 1991) and from Amazonia (Andreae and Andreae, 1988; Andreae et al., 1990a) appear to substantiate this hypothesis.

3.7 EXHALATION OF SULPHUR GASES BY VOLCANOES

Although volcanism is a geological phenomenon, sulphur gases emitted by volcanoes are ultimately also a product of biogenic sulphur reduction. As pointed out earlier, large quantities of reduced sulphur gases, especially H₂S, are not able to escape to the atmosphere, because they are precipitated by metal ions forming sulphide minerals in the sediment. Due to tectonical movement, sulphide-containing sedimentary rocks are subducted into the upper mantle of the earth from where volatile sulphur compounds are re-emitted to the atmosphere by volcanic activity. The exhalations contain sulphur mainly in the form of SO₂ and H₂S in the gas phase and sulphate in the particle phase. Thermodynamic equilibrium calculations by Heald, Naughton and Barnes (1963) indicate that in the anoxic environment of the magma, SO₂ is dominant at high temperatures, whereas H₂S is preferred at low temperatures. Accordingly, most active volcanoes are expected to release sulphur mainly as SO₂. Upon entering the atmosphere, both compounds are oxidized as they react with oxygen; H₂S is partially converted to SO₂, and SO₂ to sulphuric acid and sulphate.

A first estimate for the global rate of sulphur emissions from volcanoes was made by Kellog et al. (1972). Their assessment was based on the total volume of lava and ashes ejected during the time period from 1500 to 1914 according to Sapper (1927), the estimate by MacDonald (1955) of the weight of the gases evolved from fresh lava on Hawaii, and the analysis of Shepherd (1938), who found 10% by weight of the gases to be SO₂. The average flux of volcanic SO₂ derived in this manner is 0.75 Tg (S) a^-1 (0.023 Tmol a^-1). The result actually is a lower limit to the real value, because it considers only SO₂ released during eruptions, whereas the much longer quiescent periods of volcanic activity are not taken into account. The same deficiency is inherent in later attempts by Friend (1973) and Cadle (1975) to improve on the estimate of Kellog et al. (1972).

In the meantime, more direct information has been obtained on volcanic SO₂ emissions by means of aircraft sampling of volcanic plumes (Jaeschke, Berresheim and Georgii, 1982), and by remote measurement techniques. Table 3.3 presents a compilation of observational data prepared by Berresheim and Jaeschke (1983). To determine average emission rates from these data requires a correlation with the intensity of volcanic activity at the time of observation. Berresheim and Jaeschke (1983) have performed the analysis and thereby achieved a more realistic assessment of the situation than was hitherto available. Five categories of volcanic activities were distinguished, namely eruptive, pre-eruptive, intra-eruptive, post-eruptive, and extra-eruptive phases. The first category was further subdivided into nine groups of differing intensities. According to Tsuya (1955) the intensity can be expressed by the total volume of ejected material, often reported in connection with the eruptive events. The nine categories of the intensity scale represent the following ranges: I £ 10^-5 k m³ , II = 10^-5 to 10-4 k m³, etc. , up to VIII = 10¹ to l0² k m³ and IX ³ 10² k m³ . After this categorization, the world-wide frequency of eruptions in the period between 1961 and 1979 was considered and individual eruptions were classified by this intensity scale. On average, about 55 volcanoes undergo eruptions each year. This is about one- tenth of all active volcanoes currently in existence. By summing over the entire amount of SO₂ released within each intensity class, Berresheim and Jaeschke (1982) found an average emission rate of 0.5 Tg (S) a^-1 (0.015 Tmola^-1) for the time period considered. The value differs little from that of Kellog et al. (1972).

A much greater total emission rate was obtained, however, for the steadier , non-eruptive activities. Among some 520 live volcanoes, there are about 365 that may be classified as post-eruptive, and an additional 100 in the extra-eruptive phase. Volcanoes in the latter category may be neglected to a first approximation because they emit only a small fraction of the sulphur released by the others. The pre-eruptive and intra-eruptive phases are relatively short and may also be disregarded. According to Table 3.3, the average SO₂ flux for a single volcano in the post-eruptive activity class is roughly 3 x 10⁵ kg (SO₂) day^-1 . The use of this latter value leads to a global sulphur emission rate of 20 Tg (S) a^-1 (0.63 Tmol a^-1).

Berresheim and Jaeschke (1983) adopted a more conservative attitude. On the basis of MacDonald's (1972) description of 516 active volcanoes, they assigned a source strength of the magnitude suggested by the average of the data in Table 3.3 to not more than 90 volcanoes. The remaining 275 volcanoes with post-eruptive activity were considered to produce only 5 x 10⁴ kg (SO₂) day^-1 each. The global sulphur flux from non-erupting volcanoes is then reduced to 7 Tg (S) a^-1 (0.22 Tmol a^-1). The rate is still an order of magnitude greater than that associated with erupting volcanoes. Based on their own measurements of H₂S and sulphate in the plume of Mt Etna,

Table 3.3. Measured or estimated rates of SO₂ emissions from volcanoes

Berresheim and Jaeschke (1982) also estimated the rate of H₂S and sulphate emissions from all volcanoes with post- and extra-eruptive activities as in the order of 1 ± 1 Tg (S) a^-1 (0.03 ± 0.03 Tmol a^-1) and 3 ± 1 Tg (S)a^-1 (0.1 ± 0.03 Tmol a^-1), respectively. Measurements at the southern Italian volcanoes Etna, Stromboli, and Vulcano also showed that the contribution of COS and CS₂ is negligible compared to the other volcanic sulphur exhalations.

3.8 SUMMARY

A summary of the most important naturally emitted sulphur compounds from all sources is given in Table 3.4. This table presents our best estimates of these fluxes based on current information. However, it must be emphasized that most of these estimates are rather uncertain. This applies especially to the emissions of particulate sulphur in the form of dust and seaspray and to the emissions from soil and plants on the continents. The main reasons for the uncertainty regarding continental emissions of biogenic sulphur compounds are: (1) the difficulty of accurately determining the various biogenic sulphur species, particularly hydrogen sulphide (H₂S), at the low levels found in unpolluted environments; (2) the technical problems of measuring emission fluxes from forest and brush ecosystems; and (3) the inadequate geographical coverage of existing data.

As can be seen from Table 3.4, the total flux of sulphur compounds into the atmosphere, including anthropogenic sources, is of the order of 160 to 500 Tg (S) a^-1 (5 to 16 Tmol a^-1). The flux which is most uncertain is the emission of  particulate sulphate as seaspray. Without this uncertain source, the flux of sulphur compounds into the atmosphere would be in the range of 125 to 185 Tg (S) a^-1 (3.9 to 5.8 Tmol a^-1).

 Table 3.4 Estimates of natural sulphur emissions (Tg (S) a^-1)  

In order to consider an atmospheric sulphur budget, the emissions of sulphur compounds can be compared with the global sulphur deposition fluxes estimated by several authors. Since the main part of the emitted reduced sulphur gases undergo atmospheric oxidation before they are deposited, only the finally deposited oxidation products SO₂ and sulphate are taken into account in the budget estimation. A summary of the most recent estimates of sulphur deposition fluxes is given in Table 3.5 (Jaeschke, 1988, and references therein). The fluxes are subdivided with respect to dry and wet deposition and to continental and oceanic regions. Moreover, the fluxes of seaspray sulphate and excess sulphate are considered separately. In this context, excess sulphate is defined as that part of sulphate which is generated by the oxidation of reduced sulphur gases in the atmosphere. Therefore, excess sulphate is mainly occurring in the particle size range below 1 µm radius.

 A diagram of the total global atmospheric sulphur budget is presented in Figure 3.8. The thickness of the arrows represents the rate of the respective fluxes. It can be seen from this graph that the emission and deposition of seasalt-sulphate is a rather independent, closed cycle. Therefore, the seasalt-sulphate can be ignored when analysing the rest of the global biogeochemical cycle of sulphur. A comparison of the global deposition flux of sulphur compounds of 190 ± 40 Tg (S) a^-1 (5.9 ± 1.3 Tmol a^-1) in Table 3.5 with the global emission rates of 155 ± 30 Tg (S) a^-1 (4.8 ± 0.94 Tmol a^-1) in Table 3.4 shows that ignoring the emission and deposition of seasalt-sulphate, the budget of global emission and deposition fluxes is balanced within the accuracy of these estimates.

 Table 3.5. Estimates of global sulphur depositions (Tg (S) a^-1)

Figure 3.8  Scheme of the atmospheric sulphur cycle, together with numbers of the following annual fluxes: Emissions: 1. S(II)- oceans; 2. S(II)- soil and wetlands; 3. S(II)- volcanoes; 4. S(IV) volcanoes; 5. S(VI) volcanoes; 6. S(II)- anthropogenic; 7. S(IV) anthropogenic; 8. S(IV) biomass burning; 9. S(VI) dust; 10. S(VI) sea salt. Depositions: 11. S(VI) sea salt dry on the oceans; 12. sea salt dry and wet on the continents; 13. S(VI) wet on the oceans; 14. S(IV) dry; 15. S(IV) wet; 16. S(VI) excess sulphate wet; 17. S(VI) excess sulphate dry; 18. S(II)- dry and wet. The budget of the fluxes is nearly balanced when the cycle of the seasalt sulphate is ignored (see text) 

REFERENCES

Adams, D. F., Farwell, S. O., Robinson, E., Pack, M. R. and Bamesberger, W. L. (1981a). Biogenic sulfur source strengths. Environ. Sci. Technol., 15, 1493-8.

Adams, D. F., Farwell, S. O, Pack, M. R. and Robinson, E. (1981b). Biogenic gas emissions from soils in Eastern and Southeastern United States. J. Air Pollut. Control Assoc., 31, 1083-9.

Almgren, T. and Hagstrom, I. (1974). The oxidation rate of sulphide in sea water. Water Res., 8, 395-400.

Anderson, J. W. (1980). Assimilation of inorganic sulfate into cysteine. In: Stumpf, P . K. and Conn, E. E. (Eds). The Biochemistry of Plants, Vol. 5, Academic Press, New York, pp. 203-23.

Andreae, M. O. (1979). Arsenic speciation in seawater and interstitial waters: the influence of biological-chemical interactions on the chemistry of a trace element. Limnol. Oceanogr., 24, 440-52.

Andreae, M. O. (1980). Dimethylsulfoxide in marine and freshwaters. Limnol. Oceanogr., 25, 1054-63.

Andreae, M. O. (1985). The emission of sulfur to the remote atmosphere. In: Galloway, J. N., Charlson, R. J., Andreae, M. O. and Rodhe, H. (Eds), The Biogeochemical Cycling of Sulfur and Nitrogen in the Remote Atmosphere, Reidel, Dordrecht, pp. 5-25.

Andreae, M. O. (1986). The ocean as a source of atmospheric sulfur compounds. In: Buat-Ménard, P. (Ed.). The Role of Air-Sea Exchange in Geochemical Cycling, Reidel, Dordrecht, pp. 331-62.

Andreae, M. O. (1991a). Biomass burning: Its history, use and distribution and its impact on environmental quality and global climate. In: Levine, J. S. (Ed.). Global Biomass Burning, MIT Press, Boston, Mass. (in press).

Andreae, M. O. (1991b ). Photochemical production of carbonyl sulfide in open ocean waters. J. Geophys. Res. (in press).

Andreae, M. O. and Andreae, T. W. (1988). The cycle of biogenic sulfur compounds over the Amazon Basin. I. Dry season. J. Geophys. Res., 93, 1487-97.

Andreae, M. O. and Barnard, W. R. (1984). The marine chemistry of dimethylsulfide. Mar. Chem., 14, 267-79.

Andreae, M. O., Barnard, W. R. and Ammons, J. M. (1983). The biological production of dimethylsulfide in the ocean and its role in the global atmospheric sulfur budget. Ecol. Bull. (Stockholm), 35, 167-77.

Andreae, M. O. and Raemdonck, H. (1983). Dimethylsulfide in the surface ocean and the marine atmosphere: a global view. Science, 221,744-7.

Andreae, M. O., Ferek, R. J., Bermond, F., Byrd, K. P., Engstrom, R. T., Hardin, S., Houmere, P. D., LeMarrec, F., Raemdonck, H. and Chatfield, R. B. (1985). Dimethyl sulfide in the marine atmosphere. J. Geophys. Res., 90, 12891-900.

Andreae, M. O., Berresheim, H., Andreae, T. W., Kritz, M. A., Bates, T. S. and Merrill, J. T. (1988). Vertical distribution of dimethylsulfide, sulfur dioxide, aerosol ions, and radon over the northwest Pacific ocean. J. Atmos. Chem., 6, 149-73.

Andreae, M. O., Berresheim, H., Bingemer, H., Jacob, D. J. and Talbot, R. W. (1990a). The atmospheric sulfur cycle over the Amazon Basin, 2. Wet Season. J. Geophys. Res., 95, 16813-24.

Andreae, M. O., Talbot, R. W., Berresheim, H. and Beecher, K. M. (1990b). Precipitation chemistry in central Amazonia. J. Geophys. Res., 95, 16987-99.

Andreae, T. W., Andreae, M. O., Bingemer, H. G. and Leck, C. (1990). Measurements of DMS and H₂S over the western North Atlantic and the Tropical Atlantic. Eos Trans. Am. Geophys. Union, 71, 1254.

Andreae, T. W., Cutter, G. A., Hussain, N., Radford-Knoery, J. and Andreae, M. O. (1991). Hydrogen sulfide and radon in and over the western North Atlantic ocean. J. Geophys. Res. (in press).

Barnard, W. R., Andreae, M. O., Watkins, W. E., Bingemer, H. and Georgii, H. W. (1982). The flux of dimethylsulfide from the oceans to the atmosphere. J. Geophys. Res., 87, 8787-93.

Bates, T. S., Cline, J. D., Gammon, R. H. and Kelly-Hansen, S. R. (1987). Regional and seasonal variations in the flux of oceanic dimethylsulfide to the atmosphere. J. Geophys. Res., 92, 2930-8.

Berresheim, H. (1987). Biogenic sulfur emissions from the subantarctic and Antarctic Oceans. J. Geophys. Res., 92, 13245-62.

Berresheim, H. and Jaeschke, W. (1982). Sulfur emissions from volcanoes. In: Georgii, H. W. and Jaeschke, W. (Eds), Chemistry of the Unpolluted and Polluted Troposphere. Reidel, Dordrecht, pp. 325-37.

Berresheim, H. and Jaeschke, W. (1983). The contribution of volcanoes to the global atmospheric sulfur budget. J. Geophys. Res., 88, 3732-40.

Bingemer, H. G. (1984). Dimethylsulfid in Ozean und mariner Atmosphäre. Experimentelle Untersuchung einer natürlichen Schwefelquelle für die Atmosphäre. Ph.D. Dissertation, J. W. Goethe Universität, Frankfurt-am-Main.

Bingemer, H. G., Andreae, M. O., Andreae, T. W., Artaxo, P. Helas, G., Mihalopoulos, N. and Nguyen, B. C. (1991). Sulfur gases and aerosols in and above the equatorial African rainforest. J. Geophys. Res. (in press).

Bowen, H. J. M. (1979). Environmental Chemistry of the Elements .Academic Press, New York.

Bremner, J. M. and Steele, C. G. (1978). Role of microorganisms in the atmospheric sulfur cycle. Adv. Microb. Ecol., 2, 155-201.

Brown, K. A. and Bell, J. N. B. (1986). Vegetation-the missing link in the global cycle of carbonyl sulphide (COS). Atmos. Environ., 20, 537-40.

Burnett, W. E. (1969). Air pollution from animal wastes. Environ. Sci. Technol., 3, 744-9.

Cadle, R. D. (1975). Volcanic emissions of halides and sulfur compounds to the troposphere and stratosphere. J. Geophys. Res., 80, 1650-2.

Casadevall, T. J., Johnston, D. A., Harris, D. M., Rose, W. I., Malinconico, L. L., Stoiber, R. E., Bornhorst, T. J., Williams, S. N., Woodruff, L. and Thompson, J. M. (1981). SO₂ emission rates at Mount St Helens from March 29 through December, 1980. USGS Prof Pap., 1250, 193-200.

Challenger, F. (1951). Biological methylation. Adv. Enzymol., 12, 429-91.

Chen, K. Y. and Morris, J. C. (1972). Kinetics of oxidation of aqueous sulfide by 02. Environ. Sci. Technol., 6, 529-37.

Cicerone, R. J., Delwiche, C. C., Harriss, R. and Dickinson, R. (1984). Critical processes affecting the distribution of chemical species: Biological and surface sources. In: Global Tropospheric Chemistry: A Plan for Action. National Academy Press, Washington, DC, pp. 55-68.

Cline, J. D. and Bates, T. S. (1983). Dimethyl sulfide in the equatorial Pacific Ocean: a natural source of sulfur to the atmosphere. Geophys. Res. Lett., 10, 949-52.

Cooke, R. B. and Kelly, C. A. Chapter 7 of this volume.

Cooper, J. A. (1980). Environmental impact of residential wood combustion emissions and its implications. J. Air Pollut. Control Assoc., 30, 855-61.

Cooper, D. J. and Saltzman, E. S. (1987). Uptake of carbonyl sulfide by silver nitrate impregnated filters: implications for the measurement of low level atmospheric H₂S. Geophys. Res. Lett., 14, 206-9.

Cox, R. A. and Sheppard, D. (1980). Reactions of OH radicals with gaseous sulphur compounds. Nature, 284, 330-1.

Crafford, T. C. (1975). SO₂-emission of the 1974 eruption of Volcan Fuego, Guate-mala. Bull. Volcanol., 39, 536-56.

Crutzen, P. (1976). The possible importance of CSO for the sulfate layer of the stratosphere. Geophys. Res. Lett., 3, 73-6.

Crutzen, P. J., Heidt, L. E., Krasnec, J. P., Pollock, W. H. and Seiler, W. (1979). Biomass burning as a source of the atmospheric gases CO, H₂, N₂O, CH₃Cl, and COS. Nature, 282, 253-6.

 Cullis, C. F. and Hirschler, M. M. (1980). Atmospheric sulfur: natural and man-made sources. Atmos. Environ., 14, 1263-78.

 Delmas, R. (1982). On the emission of carbon, nitrogen and sulfur in the atmosphere during bushfires in intertropical savannah zones. Geophys. Res. Lett., 9, 761-764.

Delmas, R. and Servant, J. (1982). The origins of sulfur compounds in the atmosphere of a zone of high productivity (Gulf of Guinea). J. Geophys. Res., 87, 11019-26.

 Delmas, R. and Servant, J. (1983). Atmospheric balance of sulphur above an equatorial forest. Tellus, 358, 110-20.

Delmas, R. and Servant, J. (1988). The atmospheric sulfur cycle in the tropics. In: Rodhe, H. and Herrera, R. (Eds). Acidification in Tropical Countries. Wiley, New York, pp. 43-72.

 Delmas, R., Baudet, J., Servant, J. and Baziard, Y. (1980). Emissions and concentrations of hydrogen sulfide in the air of the tropical forest of the Ivory Coast and of temperate regions in France. J. Geophys. Res., 85, 4468-74.

Fall, R., Albritton, D. L., Fehsenfeld, F. C., Kuster, W. C. and Goldan, P. D. (1988). Laboratory studies of some environmental variables controlling sulfur emissions from plants. J. Atmos. Chem., 6, 341-62.

Ferek, R. J. and Andreae, M. O. (1983). The supersaturation of carbonyl sulfide in surface waters of the Pacific Ocean off Peru. Geophys. Res. Lett., 10, 393-6.

Ferek, R. J. and Andreae, M. O. (1984). Photochemical production of carbonyl sulfide in marine surface waters. Nature, 307, 148-50.

Filner, P., Rennenberg, H., Sekiya, J., Bressan, R. A., Wilson, L. G., Le Cureux, L. and Shimel, T. (1984). Biosynthesis and emission of hydrogen sulfide by higher plants. In: Koziol, M. J. and Whatley, F. R. (Eds). Gaseous Air Pollutants and Plant Metabolism. Butterworth, Stoneham, MA, pp. 291-312.

Friend, J. P. (1973). The global sulfur cycle. In: Rasool, S. I. (Ed.). Chemistry of the Lower Atmosphere. Plenum Press, New York, pp. 117-201.

Froelich, P. N., Kaul, L. W., Byrd, J. T., Andreae, M. O. and Roe, K. K. (1985). Arsenic, barium, geranium, tin, dimethylsulfide, and nutrient biogeochemistry in Charlotte Harbor, Florida, a phosphorus-enriched estuary. Estuarine, Coastal, Shelf Sci., 20, 239-64.

Galloway, J. N. (1985). The deposition of sulfur and nitrogen from the remote atmosphere. In: Galloway, J. N., Charlson, R. J., Andreae, M. O. and Rodhe, H. (Eds). The Biogeochemical Cycling of Sulfur and Nitrogen in the Remote Atmosphere. Reidel, Dordrecht, pp. 143-75.

Giblin, A. E. and Wieder, R. K. This volume.

Giovanelli, J., Mudd, S. H. and Datko, A. H. (1980). Sulfur aminoacids in plants. In: Stumpf, P. K.and Conn, E. E. (Eds), The Biochemistry of Plants, Vol. 5. Academic Press, New York, pp. 453-505.

Goldan, P. D., Kuster, W. C., Albritton, D. L. and Fehsenfeld, F. C. (1987). The measurement of biogenic sulfur emissions from soils and vegetation: three sites in the eastern United States revisited. J. Atmos. Chem., 5, 439-67.

Graedel, T. E. (1979). Reduced sulfur emission from the open oceans. Geophys. Res. Lett., 6, 329-31.

Guenther, J. , Lamb, B. and Westberg, H. (1989). U. S. National biogenic sulfur emissions inventory. In: Saltzman, E. S. and Cooper, W. J. (Eds). Biogenic Sulfur in the Environment. American Chemical Society Symposium Series No.393. American Chemical Society, pp. 14-31.

Hanst, P. L., Spiller, L. L., Watts, D. M., Spence, J. W. and Miller, M. F. (1975). Infrared measurement of fluorocarbons, carbontetrachloride, carrbonylsulfide and other atmospheric trace gases. J. Air Pollut. Control Assoc., 25, 1220-6.

Haulet, R., Zettwoog, P. and Sarbroux, J. C (1977). Sulphar dioxide discharge from  Mount Etna. Nature, 268, 715-17.

Heald, E. F., Naughton, J. J. and Barnes, J. L. (1963). The chemistry of volcanic gases. 2. Use of equilibrium calculations in the interpretation of volcanic gas samples. J. Geophys. Res., 68, 545-57.

Hegg, D. A., Radke, L. R., Hobbs, P. V., Brock, C. A. and Riggan, P. J. (1987). Nitrogen and sulfur emissions from the burning of forest products near large urban areas. J. Geophys. Res., 92, 14701-9.

Herrmann, J. and Jaeschke, W. (1984). Measurements of H₂S and SO₂ over the Atlantic Ocean. J. Atmos. Chem., 1, 111-23.

Hill, F. B., Aneja, V. P. and Felder, R. M. (1978). A technique for measurements of biogenic sulfur emission fluxes. J. Environ. Sci. Health A, 13, 199-225.

Hobbs, P. V., Radke, L. F., Eltgroth, M. W. and Hegg, D. A. (1981). Airborne studies of the emissions from the volcanic eruptions of Mount St Helens. Science, 211, 816-18.

Howarth, R. W. (1984). The ecological significance of sulfur in the energy dynamics of salt marsh and marine sediments. Biogeochemistry, 1, 5-27.

Howarth, R. W. and Stewart, J. W. B. This volume.

Ivanov, M. V., Lein, A. Yu., Reeburgh, W. S. and Skyring, G. W. (1989). Interaction of sulphur and carbon cycles in marine sediments. In: Brimblecombe, P. and Lein,  A. Yu (Eds). Evolution of the Global Biogeochemical Sulphur Cycle. Wiley and   Sons, Chichester, pp. 125-79.

Iverson, R. L., Nearhoof, F. L. and Andreae, M. O. (1989). Production of dimethylsulfonium propionate and dimethylsulfide by phytoplankton in estuarine and coastal waters. Limnol. Oceanogr., 34, 53-67.

Jaeschke, W. (1988). Untersuchungen zur Chemie des Schwefels in der Atmosphäre. Habilitationsschrift, Univ. Frankfurt, Ber. d. ZUF, 24pp.

Jaeschke, W., Berresheim, H. and Georgii, H. W. (1982). Sulfur emissions from Mt. Etna. J. Geophys. Res., 87, 7253-61.

Jaeschke, W., Claude, H. and Herrmann, J. (1980). Sources and sinks of atmospheric  H₂S. J. Geophys. Res., 85, 5639-44.

Jaeschke, W. and Herrmann, J. (1981). Measurements of H₂S in the atmosphere. Int. J. Environ. Anal. Chem., 10, 107-20.

Jaeschke, W., Schmidt, R. and Georgii, H. W. (1976). Preliminary results of stratospheric SO₂ measurements. Geophys. Res. Lett., 3, 517-19.

Jaeschke, W., Georgii, H. W., Claude, H. and Malewski, H. (1978). Contributions of H₂S to the atmospheric sulfur cycle. Pageoph., 116, 465-75.

Johnson, J. E. (1981). The lifetime of carbonyl sulfide in the troposphere. Geophys. Res. Lett., 8, 938-40.

Jørgensen, B. B. (1982). The sulfur cycle of a coastal marine sediment (Limfjorden, Denmark). Limnol. Oceanogr., 22, 814-32.

Jørgensen, B. B., Hansen, M. H. and Ingvorsen, K. (1978). Sulfate reduction in coastal sediments and the release of H₂S to the atmosphere. In: Krumbein, W. E. (Ed.). Environmental Biogeochemistry and Geomicrobiology, Vol. 1. Ann Arbor
Science, Michigan, pp. 245-53.

Jørgensen, B. B. and Revsbech, N. P. (1983). Colorless sulfur bacteria, Beggiatoa sp. and Thiovulum sp., in O₂ and H₂S microgradients. Appl. Environ. Microbiol., 45, 1261-70.

Kadota, H. and Ishida, Y. (1972). Production of volatile sulfur compounds by microorganisms. Annu. Rev. Microbiol., 26, 127-63.

Kellog, W. W., Cadle, R. D., Allen, E. R., Lazrus, A. L. and Martell, E. A. (1972).  The sulfur cycle. Science, 175, 587-96.

Kim, K. H. and Andreae, M. O. (1987). Carbon disulfide in seawater and the marine atmosphere over the North Atlantic. J. Geophys. Res., 92, 14733-8.

Koblentz-Mishke, O. J., Volkevinsky, V. V. and Kabanova, J. G. (1970). Plankton primary production of the world ocean. in: Wooster, W. S. (Ed.). Scientific Exploration of the South Pacific. National Academy Science, Washington DC, pp. 183-93.

Lacaux, J. P., Servant, J., Huertas, M. L., Cros, B., Delmas, R., Loemba-Ndembi, J. and Andreae, M. O. (1988). Precipitation chemistry from remote sites in the African equatorial forest. Eos Trans. Am. Geophys. Union, 69, 1069.

Lamb, B., Westberg, H., Allwine, G., Bamesberger, L. and Guenther, A. (1987). Measurement of biogenic sulfur emissions from soils and vegetation: application of dynamic enclosure methods with Natusch filter and GC/FPD analysis. J. Atmos. Chem., 5, 469-91.

Liss, P. S. and Merlivat, L. (1986). Air-sea exchange rates: introduction and synthesis. In: Buat-Menard, P. (Ed.). The Role of Air-Sea Exchange in Geochemical Cycling. Reidel, Dordrecht, pp. 113-27.

Lovelock, J. E. (1974). CS₂ and the natural sulphur cycle. Nature, 248, 625-6. 

Lovelock, J. E., Maggs, R. J. and Rasmussen, R. A. (1972). Atmospheric dimethylsulphide and the natural sulphur cycle. Nature, 237, 452-3.

Luther, G. and Church, T. Chapter 6 of this volume.

MacDonald, G. A. (1955). Hawaiian volcanoes during 1952. US Geol. Surv. Bull. , 1021-B, 15-108.

MacDonald, G. A. (1972). Volcanoes. Prentice-Hall, Englewood Cliffs, New Jersey.

McDowell, W. H. (1987). Potential effects of acidic atmospheric deposition on tropical terrestrial ecosystems. In: Rodhe, H. (Ed.). Effects of Acidification on Tropical Ecosystems. Wiley-Interscience, New York.

McElroy, M. B., Wofsy, S. C. and Sze, N. D. (1980). Photochemical sources for atmospheric H₂S. Atmos. Environ., 14, 159-63.

Malinconico, L. L. (1979). Fluctuations in SO₂ emissions during recent eruptions of Etna. Nature, 278, 43-5.

Millero, J. J., Hubinger, S., Fernandez, M. and Garnett, S. (1987). Oxidation of H₂S in seawater as a function of temperature, pH, and ionic strength. Environ. Sci. Technol., 21, 439-43.

Naughton, J. J., Lewis, V. A., Thomas, D. and Finlayson, J. B. (1975). Fume compositions found at various stages of activity at Kilauea volcano, Hawaii. J. Geophys. Res., 80, 2963-6.

Nguyen, B. C., Bergeret, C. and Lambert, G. (1984). Exchange rates of dimethyl sulfide between ocean and atmosphere. In: Brunsaert, W. and Jirka, G. H. (Eds). Gas Transfer at Water Surfaces. Reidel, Dordrecht, pp. 539-45.

Nriagu, J. O., Holdway, D. A. and Coker, R. D. (1987). Biogenic sulfur and the acidity of rainfall in remote areas of Canada. Science, 237, 1189-92.

Okita, T. (1971). Detection of SO₂ and NO₂ gas in the atmosphere by Barringer spectrometer. ASCO, Pep. 8/7.

Okita, T. and Shimozuru, D. (1975). Remote sensing measurements of mass flow of sulfur dioxide gas from volcanoes. Volcanol. Soc. Jpn. Bull., 19-3, 153-7.

Peng, T. H., Broecker, W. S., Mathieu, G. G. and Li, Y. H. (1979). Radon evasion rates in the Atlantic and Pacific Oceans as determined during the Geosecs program. J. Geophys. Res., 84, 2471-86.

Rasmussen, R. A., Khalil, M. A. K. and Hoyt, S. D. (1982). The oceanic source of carbonyl sulfide (COS). Atmos. Environ., 16, 1591-4.

Rose, W. I. , Bonis, S. , Stoiber, R. E. , Keller, M. and Bickford, T. (1973). Studies of volcanic ash from two recent Central American eruptions. Bull. Volcanol., 37, 338-64.

Saltzman, E. S. and Cooper, D. J. (1988). Shipboard measurements of atmospheric dimethylsulfide and hydrogen sulfide in the Caribbean and Gulf of Mexico. J. Atmos. Chem., 7,191-209.

Saltzman, E. S., Savoie, D. L., Prospero, J. M. and Zika, R. G. (1986). Methanesulfonic acid and non-sea-salt sulfate in Pacific air: Regional and seasonal variations. J. Atmos. Chem., 4, 227-40.

Sandalls, F. J. and Penkett, S. A. (1977). Measurements of carbonyl sulphide and carbon disulphide in the atmosphere. Atmos. Environ., 11, 197-199.

Sapper, K. (1927). Vulkankunde. Engelhorn Verlag. Stuttgart. 

Schoenau, J. J. and Germida, J. J. Chapter 10 of this volume.

Shepherd, E. S. (1938). The gases in rocks and some related problems. Am. J. Sci. 5th Ser., 235a, 311-51.

Slatt, B. J., Natusch, D. F. S., Prospero, J. M. and Savoie, D. L. (1978). Hydrogen sulfide in the atmosphere of the northern equatorial Atlantic Ocean and its relation to the global sulfur cycle. Atmos. Environ., 12, 981-91.

Smethie, W. M., Jr, Takahashi, T., Chipman, D. W. and Ledwell, J. R. (1985). Gas exchange and CO₂ flux in the tropical Atlantic Ocean determined from ²²²Rn and pCO₂ measurements. J. Geophys. Res., 90, 7005-22.

Staubes, R. (1986). Untersuchungen der Bodenexhalation von Carbonylfulfid, Dimethylsulfid und Schwefelkohlenstoff. Diplomarbeit, Institut fur Meterologie und Geophysik, Johann-Wolfang-Geothe-Universitat, Frankfurt-am-Main.

Staubes, R., Ockelmann, G. and Georgii, H. W. (1986). Emissions of biogenic sulfur compounds from various soils. Paper presented at 2nd International Symposium on Biosphere-Atmosphere Exchange, Mainz, Federal Republic of Germany, 16-21 March, 1986.

Stephens, E. R. (1971). Identification of odors from cattle feed lots. Calif. Agric., 25, 10-11.

Steudler, P. A. and Peterson, B. J. (1984). Contribution of the sulfur form salt marshes to the global sulfur cycle. Nature, 311, 455-7.

Stith, J. L., Hobbs, P. V. and Radke, L. F. (1978). Airborne particle and gas measurements in the emissions from six volcanoes. J. Geophys. Res., 83, 4009-17.

Stoiber, R. E. and Bratton, G. (1978). Airborne correlation spectrometer measurements of SO₂ in eruption clouds form Guatemalan Volcanoes. Eos Trans. Am. Geophys. Union, 59, 1222.

Stoiber, R. E. and Jepsen, A. (1973). Sulfur dioxide contributions to the atmosphere by volcanoes. Science, 182, 577-8.

Stoiber, R. E. and Malone, G. B. (1975). SO₂-emission at the crater of Kilauea, at Mauna Ulu and at Sulfur Banks, Hawaii. Eos Trans. Am. Geophys. Union, 56,461.

Taylor, P. S. and Stoiber, R. E. (1973). Soluble material on ash from active Central American volcanoes. Geol. Soc. Am. Bull., 84, 1031-42.

Tsuya, H. (1955). Geological and petrological studies of vulcano Fuji, 5. Tokyo Daigaku Jinshin Kenkyusho Iho, 33, 341-182.

Turner, S. M. and Liss, P. S. (1985). Measurements of various sulphur gases in a coastal marine environment. J. Atmos. Chem., 2, 223-32.

Turner, S. M., Malin, G. and Liss, P. S. (1989). Dimethylsulfide and (dimethylsulfonio) propionate in European coastal and shelf waters. In: Saltzman, E. S. and Cooper, W. J. (Eds). Biogenic Sulfate in the Environment. ACS Symposium Series No.393, American Chemical Society, Washington DC.

Turner, S. M., Malin, G., Liss, P. S., Harbour, D. S. and Holligan, P. M. (1988). The seasonal variation of dimethyl sulfide and dimethylsulfoniopropionate concentrations in nearshore waters. Limnol. Oceanogr., 33, 364-75.

Warneck, P. (1988). Chemistry of the Natural Atmosphere. Academic Press, San Diego, New York, p. 499.
Wilson, L. G., Bressan, R. A. and Filner, P. (1978). Light-dependent emission of hydrogen sulfide from plants. Plant Physiol., 61, 184-9.

Winner, W. E., Smith, C. L., Koch, G. W., Mooney, H. A., Bewley, J. D. and Krouse, H. R. (1981). Rates of emission of H₂S form plants and patterns of stable sulphur isotope fractionation. Nature, 289, 672-3.

Zepp, R. G. and Andreae, M. O. (1989). Factors affecting the photochemical formation of carbonyl sulfide in sea water. Eos Trans. Am. Geophys. Union, 70, 1023.

Zepp, R. G. and Andreae, M. O. (1990). Photosensitized formation of carbonyl sulfide in sea water. In: Blough, N. V. and Zepp, R. G. (Eds). Effects of Solar Ultraviolet Radiation on Biogeochemical Dynamics in Aquatic Environments . Woods Hole Oceanographic Inst, Woods Hole.

Zettwoog, P. and Haulet, R. (1978). Experimental results on the SO₂ transfer in the Mediterranean obtained with remote sensing devices. Atmos. Environ., 12, 795-6.

APPENDIX TO CHAPTER 3: SULPHUR ISOTOPE EVIDENCE FOR EMISSION OF BIOGENIC SULPHUR

The direct measurement of sulphur gas fluxes from waters and terrestrial ecosystems to the atmosphere has proven problematical (Andreae and Jaeschke, this volume). The fluxes themselves can be highly variable in space and time, and many approaches for measuring the fluxes, such as the use of chambers, may alter flux rates. Controversy over the magnitude of these emissions continues, but estimates have tended to decrease over the past decade due mainly to better instrumentation and measurement.

The analysis of the natural abundance of stable isotopes in gases and atmospheric particles provides another approach for assessing biogenic sulphur gas fluxes. The approach is discussed in detail in another SCOPE volume (Krouse and Gorlenko, 1990). Here we briefly outline the use of stable isotope data and illustrate the potential power of the approach.

Sulphur isotope abundances are expressed on a d³⁴S scale defined by:

  
 where ³⁴S/³²S is the number of ³⁴S atoms to ³²S atoms in the sample. Meteoritic troilite from Canyon Diablo has been chosen as an international standard because of its isotopic homogeneity and composition close to that of the mean value for terrestrial sulphur.

A number of biological processes can affect the sulphur isotopic abundances in compounds, and the d³⁴S ratio can be used to trace the origin of sulphur in these compounds, including atmospheric gases and particles. For example, sulphide generated by bacterial sulphate reduction is characterized by large kinetic isotope effects whereby ³²SO₄^2- is reduced faster than ³⁴SO₄^2-. The product sulphide is relatively depleted in ³⁴S, which usullay means that such sulphide has quite negative d³⁴S values. This has been demonstrated in the laboratory for the classical sulphate-reducing bacteria such as Desulphovibrio desulphuricans (see review by Chambers and Trudinger, 1972) as well as for a novel sequential reduction by bacillus reducing sulphate to sulphite and a clostridium reducing sulphite to sulphide (Hunt, 1979). The isotopic selectivity is quite variable and tends to be larger at lower reduction rates. In nature, the sulphide may be depleted in ³⁴S by as much as 70%_º as compared to the initial sulphate (Weyer, Krouse and Horwood, 1979). The dissolved sulphides in salt marsh porewaters are depleted by some 30 to 40%_º compared to the sulphate (Peterson and Howarth, 1987).

Living plants under sulphur stress release H₂S and perhaps other reduced sulphur compounds. This process is considered to involve reduction. This is confirmed by sulphur isotope data from laboratory experiments. The evolved H₂S was found to be as much as 15%_º depleted in ³⁴S as compared to foliar sulphur and SO₄ available to the roots (Winner et al., 1981).

Information on the natural abundance of oxygen-stable isotopes can also prove useful in studying the sulphur cycle. Dissolved oxygen in near-surface water re-oxidizes much of the aqueous sulphide that was generated in anoxic environments (Giblin and Wieder, this volume; Andreae and Jaeschke, this volume). The oxygen isotope composition of sulphate provides evidence of this process, particularly at higher latitudes (van Everdingen, Shakur and Krouse, 1982).

Two studies, one in the Great Lakes Basin of North America and the other near sour gas (H₂S-rich) processing plants in Alberta, Canada, illustrate the power of using the natural abundances of sulphur-stable isotopes to elucidate sources of atmospheric sulphur. In the Great Lakes Basin study, sulphur in bulk precipitation samples was examined (Nriagu and Coker, 1978). Whereas the mean d³⁴S value for rural stations was about 20%o lower than for urban centres, the striking observation was that at all sampling sites, the d³⁴S values during winter were about 4%_o higher than during summer despite the fact that concentrations at remote locations were lower. Further, the seasonable isotopic trend did not correlate with concentration changes. Fossil fuel consumption in urban areas for winter heating cannot explain the trend in remote areas. Incorporation of sulphur and associated isotope fractionation during rainfall and snowfall are not well understood but preliminary assessment showed that these phenomena could not be the dominant mechanism which determined the isotope behaviour. Although other possibilities could not be entirely excluded, the most likely explanation was that the ratio of biogenic to anthropogenic emission was higher in the warmer months.

The sour gas processing industry of Alberta, Canada, emits sulphur oxides which have d³⁴S values ranging from + 15 to +30%_º. In contrast, the preindustrial soils of Alberta had d³⁴S values ranging from -30 to 0%_º. On the basis of this isotopic difference, it was demonstrated that epiphytic lichens aquired sulphur rather directly from the atmosphere. In contrast, leaves and needles incorporated sulphur from two sources, atmospheric sulphur oxides and SO₄ transported from the soil by the root system (Krouse, 1977).

In contrast to the Great Lakes wetfall study, sulphur isotope data for atmospheric SO₂ were obtained in the Alberta study. At high SO₂ concentrations near sour gas processing facilities, d³⁴ values converged towards that of  the stack gas (Krouse, 1980). At lower concentrations, d³⁴S values were interpreted as reflecting mixtures of emissions from the industry and background sources. The background may contain emissions from distant processing plants.

For data from a number of sites at different times around sour gas plant operations, two important observations have emerged. In the warmer months, the d³⁴S values for ambient SO₂ are lower and SO₂ concentration are higher, particularly at sites where the plume does not impact continuously. This must reflect an increase in biogenic emissions, but what are the sources? An answer seems evident in the second observation, namely that  two peaks are found in the distribution of d³⁴S values for SO₂. The major  peak is that of the industrial plume whereas the other coincides with the d³⁴S values found for vegetation. Data in Figure 3A.1 show how the secondary peak in the distribution of d³⁴S values of SO₂ occurs near that found for conifer needles in a forested environment. Any oxidized H₂S from sulphate reduction should have possessed lower d³⁴S values and a greater spread. It is more likely that this secondary peak reflects SO₂ generated by the oxidation of sulphide produced during decay of the litter, which mainly comprises fallen needles. Kinetic isotope effects during removal of sulphur from large organic molecules have not been measured but are expected to be relatively small, consistent with this interpretation.

Figure 3A.1. Distributions in d³⁴S values for atmospheric SO₂ and vegetation near the Ram River sour gas processing plant, Alberta, Canada. With the exception of the control area, each distribution curve is based upon 30 to 60 samples taken at various times and locations over a three-year period. The d³⁴S value for the soil below a few centimetres depth is close to 0%_º at all sites whereas that of the emissions is near + 22%_º

 In conclusion, stable sulphur isotope data with emphasis on the seasonal shifts in d³⁴S values for wetfall and dryfall SO₂ appear to be sensitive indicators of biogenic sulphide emissions. The technique permits source assignments more readily than concentration measurements. Of course simultaneous measurements of both parameters is preferred along with meteorological parameters such as wind direction. In view of the range of estimates for biogenic sulphide fluxes, further sulphur isotope studies relevant to the problem are warranted.

REFERENCES

Andreae, M. O. and Jaeschke, W. This volume.

Chambers, L. A. and Trudinger, P. A. (1972). Microbiological fractionation of stable sulfur isotopes: A review and critique. Geomicrobiol. J., 1, 249-93.

Giblin, A. and Wieder, R. K. This volume.

Hunt, R. N. (1974). oxygen isotope studies on sulphates. Ph.D. Thesis, University of Alberta, Edmonton.

Krouse, H. R. (1977). Sulphur isotope abundances elucidate uptake of atmospheric sulphur emission by vegetation. Nature, 265, 45-6.

Krouse, H. R. (1980). Sulphur isotopes in our environment. In: Fritz, P. and J. Ch. Fontes (Eds). Handbook of Environmental Isotope Geochemistry. Elsevier, Amsterdam, pp. 435-71.

Krouse, H. R. and Grinenko, V. A. (in press). Stable Isotopes in the Assessment of Natural and Anthropogenic Sulphur in the Environment. Wiley, Chichester .

Nriagu, J. O. and Coker, R. D. (1978). Isotopic composition of sulfur in precipitation within the Great Lakes Basin. Tellus, 30, 365-75.

Peterson, B. J. and Howarth, R. W. (1987). Sulfur, carbon, and nitrogen isotopes used to trace organic matter flow in the salt-marsh estuaries of Sapelo Island, Georgia, Limnology and Oceanography, 32, 1195-213.

van Everdingen, R. O., Shakur, M. A. and Krouse, H. R. (1982). Isotope geochemistry of dissolved, precipitated airborne and fallout sulfur species associated with springs near Paige Mountain, Norma Range, NWT. Can. J. Earth Sci., 19, 1395- 407.

Weyer, K. O., Krouse, H. R. and Horwood, W. C. (1979). Investigation of regional geohydrology south of great Slave Lake, Canada, utilizing natural sulphur and hydrogen isotope variations. In: Isotope Hydrology 1978, International Atomic Energy Agency, Vienna, pp. 251-64.

Winner, W. E., Smith, C. L., Koch, G. W., Mooney, H. A., Bewley, J. E. and Krouse, H. R. (1981). Rates of emission of H₂S from plants and patterns of stable isotope fractionation. Nature, 289, 672-3.