SCOPE 21 -The Major Biogeochemical Cycles and Their Interactions  

	Abstract
	15.1 Introduction
	15.2 Estimation of AirSea Gas Fluxes using the KC Approach
	15.3 Gas Fluxes Across the AirSea Interface
	15.4 Hydrogen
		15.4.1 Molecular Hydrogen (H2)
	15.5 Carbon
		15.5.1 Formaldehyde (HCHO)
		15.5.2 Methane (CH4)
		15.5.3 Carbon Monoxide (CO)
		15.5.4 Methyl Chloride (CH3Cl)
		15.5.5 Carbon Dioxide (CO2)
	15.6 Nitrogen
		15.6.1 Nitrous Oxide (N2O)
		15.6.2 Nitric Oxide (NO)
		15.6.3 Nitrogen Dioxide (NO2)
		15.6.4 Ammonia (NH3)
	15.7 Oxygen
		15.7.1 Ozone (O3)
	15.8 Sulphur
		15.8.1 Total Volatile Sulphur
		15.8.2 Hydrogen Sulphide (H2S)
		15.8.3 Dimethyl Sulphide (DMS, (CH3)2S)
		15.8.4 Carbon Disulphide (CS2) and Carbonyl Sulphide (COS)
		15.8.5 Sulphur Dioxide (SO2)
	15.9 Iodine
		15.9.1 Methyl Iodide (CH3I)
	References

ABSTRACT

One method for deriving the flux of gases across the sea surface is to multiply the airsea water concentration difference by an appropriate rate constant. Provided suitably time and space averaged values of these two terms are available, it is possible to calculate airsea fluxes on a global basis. Results of such calculations for a variety of biogeochemically active gases are presented and discussed in terms of the importance of airsea transfer in overall element cycling and interaction in the environment.

15.1 INTRODUCTION 

Any study of interactions of the biogeochemical cycles must inevitably concern itself with not only processes within the various environmental reservoirs (soil, air, fresh- and salt-water, etc.) but also material transfer across the boundaries separating the reservoirs. The principal aim of the present paper is to gather together estimates of chemical fluxes across one such boundary, the airsea interface. Consideration of these airsea fluxes, together with those for other interfaces given elsewhere in this volume, should help in formulating the overall environmental budgets of the substances investigated and in identifying points where quantitative information and qualitative understanding are lacking. Since here the main concern is with natural biogeochemical cycles, compounds introduced into the environment solely by man's activities (e.g. Freons, DDT, PCBs) will not be considered; although some natural substances, whose concentration or rate of cycling are being affected by anthropogenic activity (e.g. CO₂), will be discussed. Elements whose compounds are clearly involved in biogeochemical cycling include H, C, N, O, P, S and I.

Although chemical fluxes across the airsea interface can take place via transfer in gas, liquid or solid states, attention here will be confined to exchange of gaseous forms; transfer in other states is dealt with by Duce (Chapter 16, this volume).

Wherever possible, air-sea fluxes given in the main part of the paper are calculated from direct environmental measurements. Such numbers should be more reliable than fluxes derived from budget exercises, where the air-sea flux is obtained by difference in order to balance the budget, e.g. most attempts at estimating the flux of gaseous sulphur compounds from sea to air, and in some studies of photochemical processes in the marine atmosphere (Graedel, 1979a).

The majority of the airsea gas fluxes given here have been calculated from the product of an air-sea concentration difference and a term which quantifies the rate of transfer across the interface. Since it has been used so widely, the principles of the approach will be briefly outlined in the next section.

15.2 ESTIMATION OF AIRSEA GAS FLUXES USING THE KC APPROACH

The processes governing the transfer of gases across the airsea interface have been reviewed by several authors (e.g. Bolin, 1960; Kanwisher, 1963; Broecker and Peng, 1974; Liss and Slater, 1976; Deacon, 1977, Slinn et al., 1978). Basically, any net gas flux (F) across the interface must be driven by an airsurface water concentration difference (C); the magnitude and direction of the flux being proportional to the numerical value and sign of C. The constant of proportionality (K) linking the flux and concentration difference has dimensions of a velocity and is variously known as an exchange constant/coefficient or a piston/transfer velocity; the lattermost term will be used here. It is often more convenient to think in terms of the reciprocal of the transfer velocity, which is a measure of the resistance (R) to interfacial gas exchange. The following equations show how the total resistance to gas transfer (R) can be split into its gas (r_g) and liquid (r₁) phase components.

  In equation (1), k₁ and k_g are the transfer vellocities for chemically unreactive gases in the liquid and gas phases, respectively. H is the dimensionless Henry's law constant, expressed as the ratio of the gas concentration in the gas phase to the concentration of un-ionized gas dissolved in the liquid phase at equilibrium. Although experimentally derived values of H should be valid over a wide range of concentrations, it is desirable, in the present context, to use values measured at typical environmental concentration levels (Liss, 1976). is a factor that quantifies any enhancement in the value of k₁ due to chemical reactivity of the gas in the aqueous phase. Theoretical treatments for the computation of have been given by Hoover and Berkshire (1969), Quinn and Otto (1971) and Emerson (1975). For unreactive gases (e.g. N₂, inert gases) = 1.00. Under oceanic conditions CO₂ is about 1.02 1.03, indicating that, for this gas, chemical reaction increases the value of k₁ but only about 23%. However, for a highly reactive gas like SO₂ the value of is of the order of 10³ (Liss, 1971).

By substituting appropriate values for the various terms in equation (1), it is possible to calculate the magnitude of r₁ and r_g for any particular gas and thus to identify the phase whose resistance controls its airsea transfer. When this is done it turns out that many gases (but see discussion on HCHO (section 15.5.1) for an important exception) fall into two distinct groups: i.e. those for which r_g » r₁ (e.g. H₂O, SO₂, SO₃, NH₃) and those for which r₁ » r_g (e.g. O₂, N₂, the inert gases, CO₂, CO, CH₄, CH₃I, and (CH₃)₂S). Gases whose exchange is under liquid-phase control generally have low solubility and are chemically unreactive in the aqueous phase, whereas for gases of high solubility or rapid aqueous phase chemistry it is the gas phase that controls their interfacial transfer.

Thus, in principle, all that needs to be done in order to calculate airsea gas fluxes is to multiply the value of the transfer velocity of the gas in the rate controlling phase by the airsea concentration difference driving the flux across the interface. On a world-wide time-averaged basis the distribution of radioactive isotopes (such as ¹⁴C: Broecker and Peng, 1974) and measurements of mean evaporation rates over the oceans can be used to obtain mean values for k₁ and k_g, respectively. Such global estimates have generally been used, faute de mieux, by the authors whose flux calculations are presented in this paper.

Attempts have been made to specify transfer velocities on smaller space and time scales. This has often been done by trying to relate the transfer velocity to some easily measured meteorological parameter, such as wind speed (u). For gases for which r_g is the controlling resistance, it is known from both field and wind-tunnel studies that k_g varies quasi-linearly with wind speed, and it has also been possible to incorporate the effects of atmospheric stability (Hicks and Liss, 1976). The situation for gases whose exchange is under liquid phase control (which includes many of the gases of interest) is much less satisfactory. Different wind-tunnel studies do not give an unequivocal dependence of k₁ on wind speed (k₁ uⁿ where n varies between 1 and 2; Broecker et al., 1978). Furthermore, the only technique which has been used extensively in the field (the Radon Deficiency Method; Broecker, 1965) seems to indicate little or no relationship between k₁ and u (Peng et al., 1979). This rather unsatisfactory state of affairs has been discussed recently by Hasse and Liss (1980) and Deacon (1981). Intuitively, gas exchange should increase with wind speed, not only due to the enhanced turbulent mixing but also from the production of bubbles that constitute an additional transfer mechanism. Thorpe (1982) has recently predicted that bubbles may be the dominant process for airsea gas transfer at wind speeds greater than 12 m s^-1.

If, due to these difficulties, globally averaged transfer velocities are used in airsea flux computations, then they need to be combined with similarly averaged values for the airsea concentration difference. However, for many gases the magnitude (and sometimes even the sign) of the concentration difference is changing both spatially and on various temporal scales (diurnal, seasonal). Such variability is often the factor that limits the accuracy of the technique. The greater the variability, the greater the number of concentration observations that are required in order to obtain a representative average value. In some cases (e.g. anthropogenic CO₂: see Section 15.5.5) this means that the approach cannot be used with the presently available number of observations, and fluxes have had to be calculated by other methods. In other instances the calculated fluxes have had to be substantially revised as more representative concentration data have become available (e.g. downwards for N₂O; upwards for (CH₃)₂S). Another source of uncertainty is whether concentrations measured in `surface' waters, which are often taken from several metres below the interface, are the correct ones to use in computing C. Work by Nguyen et al. (1978) indicates that, at least for dimethyl sulphide, water concentrations close to the interface may differ significantly from those several metres down (see Section 15.8.3).

Some attention has been paid to the possibility that natural and man-made films at the water surface might affect airsea gas exchange (see, for example, Garrett, 1972; Liss, 1977). This is of interest in the present context since such an effect would be a potentially important example of the interaction of biogeochemical cycles, in this case that of carbon (in the form of organic matter constituting the film) with those of other elements.

Films can potentially affect air-sea gas exchange by either constituting a source of additional resistance to transfer (i.e., adding a third term in equation (2)), or interfering with some mechanism that in the non-film case aids gas exchange (e.g. damping capillary waves that are known to be important for gas transfer). However, unless the film is continuous and coherent, its ability to reduce the rate of gas exchange is probably small or non-existent. In the open oceans conditions are generally against formation of coherent films since there is insufficient film-forming material and what is present tends to be broken up by wind action. Exceptions may occur in coastal and other biologically productive regions and in areas of oil spillage. The absence of capillary waves on the water surface is probably a good indication of areas where retardation of gas transfer might take place, since a coherent film appears to be necessary for the damping of capillaries to occur.

15.3 GAS FLUXES ACROSS THE AIRSEA INTERFACE

In Table 15.1 are gathered together the results of the best estimates to date of gas fluxes across the sea surface. Compounds are tabulated under one of their constituent elements. The choice of which element is somewhat arbitrary, but generally the central atom or the most biogeochemically active one is used, e.g. CH₃I is given under I in view of the well established biogeochemical importance of this compound in the cycling of iodine, whereas CH₃Cl is listed under carbon. Comments on how the flux numbers have been obtained and their biogeochemical importance are given in the following sections.

15.4 HYDROGEN 

15.4.1 Molecular Hydrogen (H₂)

Surface sea-water is generally 23 times supersaturated in H₂ with respect to the overlying atmosphere, according to Schmidt and Kulessa (1981), who calculate a total global sea-to-air flux of about 10¹² g H₂ yr^-1. This is in rather good agreement with the earlier estimate for the northern hemisphere oceanic source strength of 0.33 x 10¹² g yr^-1 (Herr and Barger, 1978). However, since the concentration data are rather limited geographically, the tabulated flux estimate will have to be revised in the future. A total ocean source strength of 10¹² g yr^-1 accounts for only a few percent of estimates of the total global H₂ production rate, most of which is thought to be anthropogenic (Schmidt, 1974).

Table 15.1 Net global gas fluxes across the airsea interface

Element	Compound	Controlling	Global AirSea Flux		Method‡	Reference
		resistance
			Direction*	Magnitude†
H	H2	r1	+	1012	Ö	Schmidt and Kulessa (I981)
C	HCHO	rg ~ r1		1013                     Ö		Zafiriou et al. (1980a)
	CH3	r1	+	10121013	Ö	Ehhalt and Schmidt (1978)
	CO	r1	+	1014	Ö	Seiler (1974)
	CH3Cl	r1	+	38 x 1012	Ö	Singh et al. (I979)
						Watson et al. (I980)
	CO2	r1		6 x 1015	X	Broecker et al. (I979)
N	N20	r1	+	6 x 1012	Ö	Cohen and Gordon (1979)
						Seiler (1981)
	NO	r1	+(?)			Zafiriou et al. (I9806)
	NO2	?	+(?)			Helas et al. (1981)
	NH,	rg	+(?)	?	Ö	Georgii and Gravenhorst (1977)
O	O3	r1		37 x 1014	X	Fabian and Junge (I970)
						Tiefenau and Fabian (1972)
S	Total volatile		+	34170 x 1012		Granat et al. (1976)
	S			(As S)	X	Eriksson (1963)
	H2S	r1	+	15 x 1012	X	Graedel (19796)
	(CH3)2S	r1	+	3050 x 1012	Ö	Barnard et al. (1982)
						Nguyen et al. (1978)
	CS2	r1	+	0.3 x 1012	Ö	This work
	COs	r1	+ (?)
	SO2	rg		5 x 1012	Ö	This work
I	CH3I	r1	+	3 x I011	Ö	Liss and Slater (1974)
*+, SeaAir; , Air Sea
†Units are g (of the compound) yr-1, except as indicated
‡ Öcalculated by KC technique;
X calculated by other technique details in text.

15.5 CARBON

15.5.1 Formaldehyde (HCHO)

The physical chemistry of the airsea transfer of HCHO is interesting because this appears to be one of the few compounds for which r_g and r₁ are of similar magnitude, although a similar situation may exist for some man-made organic chemicals (Hunter and Liss, 1977). this is brought out clearly by the two attempts to estimate r_g and r₁ for HCHO. Zafiriou et al. (1980a) calculate the total resistance to be partitioned 2/3 in the liquid phase and 1/3 in the gas phase, using H = 0.07 and = 7. In contrast, Thompson (1980) finds the gas phase just dominant (r_g: r₁ = 2/3:1/3), taking H = 0.008 and = 2.2. Both calculations are based on theory and the situation clearly requires experimental verification. However, the resistance of both phases must obviously be included in any calculation of the airsea flux of this gas.

The flux given in Table 15.1 for HCHO is calculated from the results given by Zafiriou et al. (1980a) since their concentration measurements are from a remote oceanic area (Enewetak Atoll in the central equatorial Pacific). Thompson (1980), using similar data from a coastal site (Woods Hole, Mass.) predicts an air-sea flux about 50% greater. In both of these papers fluxes are given per unit area of sea surface. For present purposes the concentration data of Zafiriou et al. are assumed to be representative of the whole of the oceans, so that by multiplying the flux per unit area by the total oceanic area a global air-sea flux is obtained. This may well not be too gross an extension since recent measurements in the North and South Atlantic give an HCHO air concentration of 0.22 ppbv (Lowe and Schmidt, 1981), which is reasonably close to the value of 0.4 ppbv obtained by Zafiriou et al. at Enewetak. The work of Lowe and Schmidt also lends credance to the assumption made by Zafiriou et al. that surface sea-water contains no HCHO, since although the former authors were just able to detect HCHO in surface waters, they found them vastly undersaturated with respect to the overlying air.

Both Zafiriou et al. (1980a) and Thompson (1980) found some transfer of HCHO to the sea surface by wet deposition. In the case of the open Pacific, precipitation accounted for only about 20% of the total flux, whereas at the Woods Hole site gaseous and wet desposition were of roughly equal magnitude.

15.5.2 Methane (CH₄)

Ehhalt and Schmidt (1978) calculate the sea-to-air flux of CH₄ to be in the range of 1.316.6 x 10¹² g yr^-1. Most open-ocean estimates are near the lower end of this range, e.g. Weiss (1981); Scranton and Brewer (1977), whose calculated flux for the western sub-tropical North Atlantic extrapolate to a global value of about 5 x 10¹² g yr^-1; Liss and Slater (1974) give 3 x 10¹² g yr^-1. Supersaturations are often considerably greater in shallow coastal areas, leading to substantially higher fluxes from such waters. In spite of this, production by the oceans is probably, at most, only a few percent of the total amount of CH₄ formed by natural terrestrial sources (0.51.0 x 10¹⁵ g yr^-1, Ehhalt and Schmidt, 1978).

15.5.3 Carbon Monoxide (CO)

Carbon monoxide shows supersaturations in surface sea-waters of several tens of times and this leads to the large flux shown in Table 15.1. Although the figure given was published some years ago, more recent measurements of the gas in both surface sea-waters (Conrad and Seiler, 1982) and the marine atmosphere (Heidt et al., 1980) are in rather good agreement with the data used by Seiler in his 1974 paper, so that his calculated flux still stands. There is evidence that the CO found in surface sea-water is formed by photo-oxidative as well as microbiological processes (Wilson et al., 1970; Conrad and Seiler, 1982).

The sea-to-air flux of CO is about the same as the amount produced by man through the incomplete combustion of fossil fuels. However, the efflux of CO from the oceans appears to be one to two orders of magnitude less than the amount formed in the atmosphere by the photochemical oxidation of CH₄ by OH radicals. When this oxidation does not go to completion it produces much of the formaldehyde that was calculated earlier to transfer from the atmosphere into the oceans.

15.5.4 Methyl Chloride (CH₃Cl)

Since few measurements have been made of the concentration of CH₃Cl in sea-water, and these show rather variable results, the sea-to-air flux in Table 15.1 must be regarded as tentative. Recent atmospheric measurements made during project GAMETAG (Rasmussen et al., 1980), which give global tropospheric CH₃Cl levels in the range 755815 ppt (compared with the value of 613 ppt used by Singh et al., 1979), do not significantly alter the calculated flux.

The mechanism for the production of CH₃Cl in seawater is not known, although it is assumed to be natural; CH₃Cl may be formed by reaction between the abundant Cl ions in sea-water and methyl iodide produced by algae (Zafiriou, 1975). As will be discussed later (section 15.9.1), surface sea-water also appears to be a source of methyl iodide to the atmosphere (3 x 10¹¹ g yr^-1) so that if the mechanism proposed by Zafiriou is to work then the algae (or some inorganic processsee section 15.7.1) must be producing at least an order of magnitude more CH₃l than is required to account for the sea-to-air flux of this compound alone.

The sea appears to be the largest source of CH₃Cl to the atmosphere, with biomass burning as the next most prolific contributor (Watson et al., 1981). The reason for estimating the flux of this compound to the atmosphere is not for its role in the carbon cycle, but because it is probably comparable to the man-made Freons as a natural source of chlorine atoms to the stratosphere, and may also be of importance for the behaviour of gases in the troposphere (Rasmussen et al., 1980).

15.5.5 Carbon Dioxide (CO₂)

The net flux of anthropogenic CO₂ into the oceans given in Table 15.1 is equivalent to only about 2% of the natural rates of exchange of this gas across the sea surface. Averaged globally and annually the natural exchange is in balance with about 330 x 10¹⁵ g yr^-1 of atmospheric CO₂ entering the oceans and an equal amount returning to the atmosphere. In view of this very large natural cycling, it is presently not possible to use the KC technique to estimate the uptake of man-made CO₂ by the oceans, and other techniques have to be used instead (Siegenthaler and Oeschger, 1978; Broecker et al., 1979).

Observations of the partial pressure of CO₂ in both air and surface sea-water show that the two phases are rarely at equilibrium. However, as expected, the size and sign of the airsea CO₂ concentration difference is found to be highly variable both geographically and on various time scales (diurnal, seasonal). Thus, in order to use the KC approach to extract the net anthropogenic input signal from the much larger natural two-way flux, a very large number of observations would be required. As well as this, knowledge of the appropriate transfer velocity for each set of measurements would also be needed. As pointed out earlier, although the global mean value for k_l is thought to be well known, it is presently not possible to specify its value at any particular time and place in the oceans.

It is hardly necessary to stress the importance of being able to accurately quantify the amount of man-mobilised CO₂ that is taken up by the oceans. The global budget for such CO₂ is at present either difficult or nearly impossible (depending on your view concerning the importance of land-use change in affecting atmospheric CO₂) to balance (cf. Melillo and Gosz, Chapter 6, this volume; and Houghton and Woodwell, Chapter 11 this volume). However, the oceans are well accepted to be the major sink for anthropogenic CO₂ released into the atmosphere, so that prediction of the time scale of any climatic warming, due to elevated atmospheric levels of the gas, is clearly dependent on an accurate assessment of the role of the oceanic sink.

15.6 NITROGEN

15.6.1 Nitrous Oxide (N₂O)

Earlier estimates of the flux of (N₂O) from the sea to the atmosphere are about an order of magnitude greater than the present best value given in Table 15.1. This decrease comes about due to poor geographical coverage of the early surface water data that were largely confined to the North Atlantic. The extent of supersaturation of surface waters is only 12% over large parts of the oceans, but with substantially higher concentrations in coastal water and areas of upwelling and vertical mixing (Weiss, 1981). The flux given in the table corresponds to an average saturation for all surface sea-waters of approximately 104%. The sea-to-air flux is about an order of magnitude less than the probable input of (N₂O) to the atmosphere from terrestrial sources (Hahn, 1979).

15.6.2 Nitric Oxide (NO)

No one has yet used the KC approach to calculate the flux of NO across the sea surface. However, NO can probably be formed in near-surface sea-waters by photolysis of nitrite ions, and Zafiriou et al. (1980b) present results of direct measurements made in the equatorial Pacific that appear to show that this is the case. These authors calculate that in the area where their measurements were performed the sea is clearly a net source of NO for the atmosphere. Extrapolation of this result to other oceanic areas is difficult because formation of NO requires both light in the correct wavelength range (295410 nm) and the presence of precursor NO₂ ions, whose concentration tends to be very variable. Further measurements of NO in surface waters are needed in order to resolve the matter. Since NO is rather insoluble in water (H = 33, Zafiriou and McFarland, 1980), its airwater transfer is probably under liquid phase control.

In the photolytic production of NO from NO₂, hydroxyl radicals are also formed and these are likely to react with many different species in surface sea-water. Although this possibility has apparently received little attention, it is potentially of considerable importance in the context of interactions between biogeochemical cycles.

15.6.3 Nitrogen Dioxide (NO₂)

As with NO, no attempt has been made to calculate an airsea flux, although Helas et al. (1981), from their measurements of diurnal variations in NO₂ concentrations in marine air, suggest that the ocean acts as a source.

15.6.4 Ammonia (NH₃)

Georgii and Gravenhorst (1977) have tried to use the KC approach to calculate the airsea transfer of ammonia (they obtain an upward flux of 5 x 10^-2 µg m^-2 hr^-1). However, the result must be treated with caution since the calculation is based on very few observations and is also probably incorrect because the authors appear to have assumed NH₃ exchange is under liquid phase control (which is unlikely since H = 3 x 10^-4). Further uncertainty is introduced because water concentrations are obtained by calculation from measurements of pH and total dissolved ammonia (NH₃ and NH₄+), which is quite variable. All that can really be said is that marine air and surface sea-water do not appear to be greatly out of equilibrium with respect to NH₃, a conclusion that is supported by the recent measurements of ammonia gas over the Southern Ocean by Ayers and Gras (1980).

15.7 OXYGEN

15.7.1 Ozone (O₃)

Various indirect techniques have been used to estimate the flux of O₃ into the oceans (as discussed by Balls, 1980) and a range of values is given in Table 15.1. Even higher (2) fluxes are possible if the rates of deposition onto sea-water surfaces given by Regener (1974) and recently by Garland et al. (1980) are used.

In principle it should be possible to obtain the flux directly by use of the KC technique, since O₃ concentrations in the marine troposphere are reasonably well known and the sea can probably be taken as a perfect sink due to the high reactivity of sea-water constituents with O₃. However, it is this reactivity that presents a problem in establishing the appropriate value of K for O₃. The gas itself is relatively insoluble in water (H = 3) but rapid reaction between it and sea-water constituents means that is substantially greater than 1.0 (probably 5 < _O3 < 20). Even with as high as 20, exchange will still be under liquid phase control.

Garland et al. (1980) have tried to identify the constituents in sea-water that are most reactive with O₃ and conclude that I ions and unidentified organic surfactants are probably of greatest importance. Variations in the distribution of both I^- and organcs in surface sea-water probably mean that the O₃ transfer velocity is itself highly variable. These authors have also raised the intriguing possibility that reaction between O₃ and I may play a role in transferring volatile iodine species from the sea to the atmosphere. They suggest that this reaction will produce I₂ that can either transfer directly across the sea surface or be rapidly hydrolysed to hypoiodous acid. The HIO could then react with organic molecules near the sea surface to form organo-iodine compounds such as the volatile methyl iodide. As Garland et al. point out, even if only a small part of the O₃ taken up by the oceans reacts in this way, it is probably enough to account for the estimated flux of CH₃I across the sea surface (see Table 15.1 and section 15.9.1). The proposed reaction represents a potentially important example of interaction between biogeochemical cycles, in this case those of oxygen, carbon and iodine.

15.8 SULPHUR

15.8.1 Total Volatile Sulphur

Over the last 20 years several global sulphur budgets have been published; a review of these is presented in Chapter 2 (this volume). Each of them has a flux of volatile sulphur from the oceans (and land surfaces) to the atmosphere whose magnitude is calculated by difference, i.e. to make the budget balance. Since such an approach is inherently unreliable, it is not surprising that the size of the flux varies substantially between budgets (see Table 15.1). Furthermore, we do not know at all well how the total flux is partitioned between the various possible volatile sulphur compounds (H₂S, (CH₃)₂S, COS, CS₂, etc.) Present knowledge (or lack of it) concerning the transfer of these individual compounds is discussed in subsequent sections.

However, balancing the global cycle is not the only reason why the sea-to-air flux of volatile sulphur is important. For example, once in the atmosphere these compounds will be subject to photochemical oxidation leading to the production of SO₂ (Logan et al., 1979). A fraction of the SO₂ may be re-absorbed by the ocean (Section 15.8.5), but much will probably remain in the atmosphere. There it can have at least two environmentally important fates: (i) incorporation into raindrops whose acidity will thereby increase, thus playing a potentially important role in controlling the pH and excess sulphate of marine precipitation; (ii) formation of stratospheric sulphate aerosol, with implications for the radiation balance and hence the climate of the Earth (Sze and Ko, 1979; Turco et al., 1980).

15.8.2 Hydrogen Sulphide (H₂S)

Until recently it was assumed that H₂S was the compound that carried the whole of the sea-to-air flux of volatile sulphur. However, this is unlikely since the compound is thermodynamically unstable in oxygenated waters and even if formed is very rapidly oxidized in such environments. Thus, the only areas where it is at all likely to be a transferring species are shallow coastal waters overlying anoxic sediments. Release of H₂S has indeed been shown in such environments (e.g. Hansen et al. ,1978) but, as expected, it does not seem to have been found in open ocean surface waters.

However, H₂S does occur in remote marine air and this had led Graedel (1979b) to compute, from a knowledge of its tropospheric photochemistry, the size of the sea-to-air flux required to maintain the observed air concentration. This value is given in Table 15.1 but, for the reasons outlined, it should be regarded as unconfirmed.

15.8.3 Dimethyl Sulphide (DMS, (CH₃)₂S)

Using the concentrations of DMS measured by Lovelock et al. (1972) in the Atlantic, Liss and Slater (1974) calculated the global sea-to-air flux of this gas to be 7 x 10¹² g yr^-1. More recent concentration measurements have led to a substantial upward revision of the flux and these new values are the ones given in Table 15.1

The value of 50 x 10¹² g yr^-1 calculated by Nguyen et al. (1978) is based on sea surface microlayer (top 400 µm) samples that they found to be up to five times enriched in DMS compared to conventional `surface' waters.

15.8.4 Carbon Disulphide (CS₂) and Carbonyl Sulphide (COS)

Using the data given by Lovelock (1974) for the concentration of CS₂ in the open Atlantic (50^°N. to 65^°S.) it is possible to calculate the global sea-to-air flux of the gas as approximately 0.3 x 10¹² g yr^-1. This implies that CS₂ is about 100 times less effective than DMS in transferring vapour phase sulphur from the oceans to the atmosphere.

Another gaseous sulphur compound that may transfer across the airsea interface is carbonyl sulphide. Although it occurs at relatively high concentrations in the atmosphere (about 500 pptv), it is often assumed that transfer of COS is from the sea into the air.

15.8.5 Sulphur Dioxide (SO₂)

Unlike the sulphur compounds discussed so far, the direction of the net airsea flux for SO₂ is into the oceans. Any possibility of a reverse flux can be ruled out in view of the high pH of sea-water and its ability to oxidize sulphite to sulphate.

The magnitude of the flux from air to sea was calculated by Liss and Slater (1974) to be 1.5 x 10¹⁴ g yr^-1, based on a concentration of SO₂ in marine air of 3 µg m^-3. More recent measurements give substantially lower SO₂ concentrations over the oceans, generally close to 0.1 µg m^-3 (Meszaros, 1978; Maroulis et al., 1980; Ockelmann et al., 1981). This leads to a much reduced flux, as given in Table 1.

Much of the SO₂ taken up by the oceans is probably derived from sulphur compounds evolved from sea-water and oxidized in the atmosphere, although some may be from combustion sources, especially near heavily industrialized land areas.

15.9 IODINE

15.9.1 Methyl Iodide (CH₃I)

The global geochemical budget for iodine, as for sulphur, requires a flux of some volatile species from the ocean to the atmosphere in order to achieve balance; both I₂ and CH₃I have been suggested as possibilities. Laboratory studies seem to show that I₂ can be generated in sea-water by either ultraviolet irradiation (Miyake and Tsunogai, 1963) or the action of O₃ (Garland and Curtis, 1981), although the applicability of such laboratory results to the real oceans is unknown. If methyl iodide is the transferring species then field data are available for marine air and surface sea-water concentrations in the Atlantic (Lovelock et al., 1973), and Liss and Slater (1974) have used these to calculate the global flux of CH₃I out of the oceans as 3 x 10¹¹ g yr^-1. This is about half the amount needed to balance the global iodine cycle. As well as its probable importance in global budgeting of iodine, Chameides and Davis (1980) have pointed out the important role that methyl iodide may play in tropospheric photochemical reactions.

15.10 REFERENCES

Balls, P. W. (1980) Gas transfer across airwater interfaces. Ph.D. Thesis, University of East Anglia.

Bolin, B. (1960) On the exchange of carbon dioxide between the atmosphere and the sea, Tellus, 12, 274-281.

Broecker, W. S. (1965) An application of natural radon to problems in ocean circulation. In T. Ichiye (ed.) Symposium on Diffusion in Oceans and Fresh Waters, Palisades, N.Y., Columbia University Press, 116-145.

Broecker, W. S., Takahashi, T., Simpson, H. J., and Peng, T.-H. (1979) Fate of fossil fuel carbon dioxide and the global carbon budget, Science, 206, 409-418.

 Chameides, W. L., and Davis, D. D. (1980) Iodine: its possible role in tropospheric photochemistry, J. Geophys. Res., 85, 7383-7389.

Conrad, R., and Seiler, W. (1982) Carbon monoxide in seawater (Atlantic Ocean), J. Geophys. Res., 87, 8839-8852.

Deacon, E. L. (1981) Seaair gas transfer: The wind speed dependence, BoundaryLayer Meteorology, 21, 31-37.

Ehhalt, D. H., and Schmidt, U. (1978) Sources and sinks of atmospheric methane, Pageoph., 116, 452-464.

Eriksson, E. (1963) The yearly circulation of sulfur in nature, J. Geophys. Res., 68, 4001-4008.

Garland, J. A., and Curtis, H. (1981) Emission of iodine from the sea surface in the presence of ozone, J. Geophys. Res., 86, 3183-3196.

Garrett, W. D. (1972) Impact of natural and man-made surface films on the properties of the airsea interface, in (Dyrssen, D. and Jagner, D. (eds) The Changing Chemistry of the Oceans, New York, Wiley-Interscience, 75-91.

Graedel, T. E. (1979a) The kinetic photochemistry of the marine atmosphere, J. Geophys. Res., 84, 273-286.

Granat, L., Rodhe, H., and Hallberg, R. O. (1976) The global sulphur cycle, in Svensson, B. H., and Söderlund, R. (eds) Nitrogen, Phosphorus and SulphurGlobal Cycles, SCOPE Report No. 7, Ecol. Bull., (Stockholm) 22, 89-134.

Hansen, M. H., Ingvorsen, K., and Jørgensen, B. B. (1978) Mechanisms of hydrogen sulfide release from coastal marine sediments to the atmosphere, Limnol. Oceanogr., 23, 68-76.

Heidt, L. E., Krasnec, J. P., Lueb, R. A., Pollock, W. H., Henry, B. E., and Crutzen, P. J. (1980) Latitudinal distributions of CO and CH₄ over the Pacific, J. Geophys. Res., 85, 7329-7336.

Herr, F. L., and Barger, W. R. (1978) Molecular hydrogen in the near-surface atmosphere and dissolved in waters of the Tropical North Atlantic, J. Geophys. Res., 83, 6199-6205.

Hicks, B. B., and Liss, P. S. (1976) Transfer of SO₂ and other reactive gases across the airsea interface, Tellus, 38,. 348-354.

Hunter, K. A., and Liss, P. S. (1977) The input of organic material to the oceans: Airsea interactions and the organic chemical composition of the sea surface, Marine Chemistry, 5, 361-379.

Liss, P. S. (1971) Exchange of SO₂ between the atmosphere and natural waters, Nature, 233, 327-329.

Liss, P. S. (1977) Effect of surface films on gas exchange across the airsea interface. Rapports et Procés Verbaux des RéunionsConseil International pour l'Exploration de la mer, 171, 120-124.

Liss, P. S., and Slater, P. G. (1976) Mechanism and rate of gas transfer across the airsea interface, in Engelmann, R. J., and Sehmel, G. A. (eds) AtmosphereSurface Exchange of Praticulate and Gaseous Pollutants (1974) Washington, D.C., U.S.E.R.D.A., 354-366.

Lovelock, J. E. (1974) CS₂ and the natural sulphur cycle, Nature, 248, 625-626. 

Lovelock, J. E., Maggs, R. J., and Rasmussen, R. A. (1972) Atmospheric dimethyl sulphide and the natural sulphur cycle, Nature, 237, 452-453.

Lowe, D. C., and Schmidt, U. (1981) Formaldehyde in marine air, Abstracts ROAC Symposium, IAMAP Meeting, Hamburg, August 1981.

Meszaros, E. (1978) Concentration of sulfur compounds in remote continental and oceanic areas, Atmos. Environ., 12, 699-705.

Nguyen, B. C., Gaudry, A., Bonsang, B., and Lambert, G. (1978) Re-evaluation of the role of dimethyl sulphide to the sulphur budget, Nature, 275, 637-639.

Ockelmann, G., Meixner, F. X., and Georgii, H. W. (1981) SO₂ measurements over the Arctic and Atlantic Ocean, Abstracts ROAC Symposium, IAMAP Meeting, Hamburg, August, 1981.

Quinn, J. A., and Otto, N. C. (1971) Carbon dioxide exchange at the airsea interface: Flux augmentation by chemical reaction, J. Geophys. Res., 76,1539-1549.

Rasmussen, R. A., Rasmussen, L. E., Khalil, M. A. K., and Dalluge, R. W. (1980) Concentration distribution of methyl chloride in the atmosphere, J. Geophys. Res., 85,7350-7356.

Schmidt, U. (1974) Molecular hydrogen in the atmosphere, Tellus, 26, 78-90. 

Schmidt, U., and Kulessa, G. (1981) The oceans as a source of atmospheric hydrogen (H₂), Abstracts ROAC Symposium, IAMAP Meeting, Hamburg, August 1981. 

Scranton, M. L, and Brewer, P. G. (1977) Occurrence of methane in the near-surface waters of the western subtropical North-Atlantic, Deep-Sea Res., 24, 127-138. 

Seiler, W. (1974) The cycle of atmospheric CO, Tellus, 26,116-135.

Siegenthaler, U., and Oeschger, H. (1978) Predicting future atmospheric carbon dioxide levels, Science, 199, 388-395.

Slinn, W. G. N., Hasse, L., Hicks, B. B., Hogan, A. W., Lal, D., Liss, P. S., Munnich, K. O., Sehmel, G. A. and Vittori, O. (1978) Some aspects of the transfer of atmospheric trace constituents past the airsea interface, Atmos. Environ., 12, 2055-2087.

Thompson, A. M. (1980) Wet and dry removal of tropospheric formaldehyde at a coastal site, Tellus, 32, 376-383.

Tiefanau, H., and Fabian, P. (1972) The specific ozone destruction of the ocean surface and its dependence on horizontal wind velocity from profile measurement, Arch. Met. Geoph. Biokl. Ser. A., 21, 399-412.

Watson, A. J., Lovelock, J. E., and Stedman, D. H. (1980) The problem of atmospheric methyl chloride, in Aikin, A. C. (ed.) Proc. Nato Advanced Study Institute on Atmospheric Ozone, Washington, U.S. Federal Aviation Administration, 365-372.

Wilson, D. F., Swinnerton, J. W., and Lamontagne, R. A. (1970) Production of carbon monoxide and gaseous hydrocarbons in seawater: Relation to dissolved organic carbon, Science, 168, 1577-1579.

Zafiriou, O. C. Alford, J., Herrerra, M., Peltzer, E., Thompson, A. M., and Gagosian, R. B. (1980a) Formaldehyde in remote marine air and rain: Flux measurements and estimates, Geophys. Res. Lett., 7, 341-344.

Zafiriou, O. C., McFarland, M., and Bromund, R. H. (1980b) Nitric oxide in seawater, Science, 207, 637-639.