DMS and climate

Dimethyl sulphide, or DMS, was first observed to be present in the ocean in considerable quantity in the early 1970s. Since the mid-1980s there has been

Table 4.1. Gaseous sulphur emission rates Units are Tg S yr-1.

Emission source Northern Hemisphere Southern Hemisphere Global

Marine DMS and H2S 11 13 24

Volcanoes 6.3 3.0 9.3

Terrestrial biogenic 0.6 0.4 1.0

Anthropogenic sulphur 69.2 9.0 78.2

Total 87.1 25.4 112.5

Source: After Table 5.2 of Houghton et al., 2001.

considerable interest in this gas as a major source for oceanic sulphate aerosols, which are now thought to form the majority of sub-micron particles in the troposphere. It has even been proposed that climatic feedbacks between algal production, sulphate aerosol levels, and cloud albedo may have exerted a strong climatic control in the past, and be contributing to an amelioration of global warming due to enhanced atmospheric concentrations of greenhouse gases.

DMSis of biological origin, as noted in §4.2.4. It is an oxidation or breakdown product of dimethylsulphonium propionate (DMSP). This latter substance is thought to assist in the limitation of osmotic loss of algal cell material to sea water. However, probably more importantly, it has recently been established that DMS and DMSP are significant anti-oxidant chemicals in planktonic cells, reacting with free radicals such as the hydroxyl radical. DMStends to be released from aged or dead cells, and released during grazing by zooplankton. DMS is also produced by terrestrial plants. Its impact on the atmospheric aerosol load is less over land, however, as there are many additional terrestrial sources of sulphur.

Table 4.1 shows estimates of the size of the global sources for sulphur emission. Marine DMS contributes about 20% of global emissions (60% of pre-industrial emissions, however), but 80% of those with a marine origin. Long range tropospheric transport of sulphate, the eventual oxidation product of sulphur gases (see reactions (3.11) and (4.8)), is limited because of the hygroscopic nature of the particles and their consequent active participation in cloud, and rain, formation. Measurements over mid-oceanic sites in both the North and South Pacific Oceans suggest that the massive Northern Hemisphere anthropogenic input of sulphur to the atmosphere has only a limited impact on remote marine tropospheric sulphate concentrations, with most of the sulphur being deposited locally or regionally. While emission ratios imply that they should be more like four times Southern Hemisphere values, North Pacific atmospheric levels of sulphate are often substantially less than double those in the South Pacific. DMS may therefore provide about 80% of sub-micron marine sulphate in the Southern Hemisphere, and about 50% of that in the Northern Hemisphere.

There are, however, some significant uncertainties in these figures. The production of sea salt aerosols by breaking waves, discussed in §2.9.2 and §3.5.2, will add sulphate particles to the atmosphere, as well as sodium chloride. The mass proportion of sea-salt-derived sulphate can be high in some circumstances,

Fig. 4.11. Schematic illustration of the sources and sinks of DMS in the marine boundary layer of the atmosphere and the oceanic mixed layer.

Fig. 4.11. Schematic illustration of the sources and sinks of DMS in the marine boundary layer of the atmosphere and the oceanic mixed layer.

Marine Atmosphere Boundary Layer

but the particles will generally be Giant nuclei, because of their mode of formation, and therefore much fewer in number. The sub-micron sulphate we are considering is often called non-sea-salt, or nss, sulphate because of its distinctly different origin. The quantity of nss sulphate is derived by measuring the sodium or chloride ion concentration in the aerosols and deducting the seawater sulphate:sodium/chloride ratio (see Table 1.4) of the sodium/chloride concentration from the total sulphate level. In addition, it has been found that in the marine boundary layer significant enhancement of the number of larger (supermicron) nss sulphate aerosols occurs. These may be produced by oxidation by ozone within the very moist atmosphere of this near sea zone. These are then believed to be lost by deposition back into the ocean, thus diminishing the potential for DMS to contribute to mid-tropospheric cloud condensation nuclei. A schematic of the sources and sinks of DMS within the oceanic and atmospheric boundary layers is shown in Fig. 4.11.

With the above reservations, primary production in the ocean is thus responsible for a considerable proportion of the tropospheric aerosols, and thus cloud droplets, that contribute to the climate system in various ways (see §3.5). We have seen that primary production is highly variable in both space and time (§4.1.2); an additional complication with DMS production is its strong dependence on species. The reasons for this, and a good knowledge of the emission rates of different plankton, are presently elusive. The causes are presumably linked to the osmotic processes across cell boundaries, and hence may partially depend on salinity and temperature. The differences in species and seasonal

Fig. 4.12. Diagram of the feedback loop involving climate and planktonic production of DMS.

Cloud formation

Increased concentration of cloud droplets

♦-Increased cloud albedo

Cloud condensation nuclei

Loss of solar radiation

Particle formation


Lower surface temperature

Less near-surface solar flux


Lower surface temperature


DMS (in solution)

Biological production of DMS (±?)

cycles are sufficiently large that the weak productivity of nutrient-limited, or eutrophic, tropical waters produces a similar emission (about 2.2 x 10-3 molm-2yr-1) to more productive temperate localities. Upwelling regions produce slightly more DMS, and the highly productive coastal zones several times as much (5-6 x 10-3 molm-2yr-1). These latter regions, because of their small surface area relative to the global ocean, will contribute rather little to the net global flux of 24 x 1012 gyr-1.

Not all the DMS released by cell decay of algae escapes into the atmosphere. Some is photolysed within the sea (see §3.6). A large proportion is absorbed by bacteria, and oxidized to allow the sulphur to be made available to these organisms. A by-product of this oxidation is hydrogen sulphide, H2S. Recent observations in tropical waters suggest that this may be a significant (30-90%) sink for DMS, allowing much less to escape into the atmosphere than suggested in the preceding paragraph. Of course, it is possible that the production of H2S, and its consequent escape to the atmosphere, may partially off-set the atmospheric implications of this bacterial DMS sink. The activity of bacteria in extra-tropical environments may also differ radically from the tropics. Nonetheless, uncertainties about the size of the DMS flux to the atmosphere make present estimates of its climatic significance problematical.

What is this climatic role for DMS, assuming sufficiently large fluxes to the atmosphere to account for the majority of nss sulphate aerosols? Taking a mass-weighted average radius of a cloud droplet to be r, then for a given cloud liquid water amount, L, and droplet number concentration, N, volume arguments show that where p is the density of water. Thus, if there is a given amount of liquid water in a cloud a change in number of aerosols, and thus droplets, leads to an inverse change in radius. More droplets lead to tinier droplets. A greater number of such

droplets tends to increase the net surface area of the droplets, and hence the cloud albedo by reflecting more solar radiation. Thus if DMS were to increase, the net effect might be a decrease in the input of energy to the climate system, and a consequent global cooling.

This mechanism could be part of a feedback process, illustrated in Fig. 4.12. If climatic warming led to greater oceanic productivity, more nss sulphate aerosols would be produced, leading to more reflective clouds and global cooling. The cloudiness reduces light, and with the cooling, lowers productivity, hence reducing nss sulphate aerosols and allowing the planet to warm again. The climate would be in a state of constant planktonic-aerosol adjustment.

Such a feedback mechanism has an appealing simplicity. However, within the climate system any one feedback loop is only part of a much more complex whole. For instance, the processes linking changing temperatures andplanktonic population size and distribution are probably not well understood because we have never consciously observed such a link. In Chapter 7 we will investigate in some detail the various interacting components that may contribute to climatic change over the next century or two.

Further reading

A complete reference list is available at the end of the book but the following is a selection of the best books or articles to follow up particular topics within this chapter. Full details of each reference are to be found in the Bibliography.

Andrews et al. (1996): A good introductory to medium level text on environmental biogeochemistry. Broecker and Peng (1982): An invaluable guide to ocean chemistry. Well written with a very comprehensive list of pre-1982 references. Discusses biological processes where appropriate in text. Lalli and Parsons (1997): A thorough and readable account of oceanic biological processes, used by the Open University. Manahan (1990): A comprehensive guide to environmental chemistry with substantial sections on aquatic and atmospheric chemistry with considerable biological discussion. More advanced reading. Mann and Lazier (1991): An excellent synthesis of marine biology and its interaction with the physical environment.

Was this article helpful?

0 0


  • jens
    What effect will warming oceans have on dms production?
    2 years ago
  • Teresio
    How is dms deposited into atmosphere?
    3 years ago