JOURNAL OF GEOPHYSICAL RESEARCH, VOL.95,NO. C3,PAGES 3353-3363, MARCH15,1990 Dimethyl Sulfide in the BalticSea: Annual Variability in Relationto Biological Activity CAROLINE LECK, 1ULF LARSSON, 2 LARS Emac B•GANDER, 3 SIF JOHANSSON, 4 AND SUSANNA HAJDU 2 University of Stockholm, Stockholm, Sweden Thein situ variation ofdimethyl sulfide (CH3SCH3, DMS) ata fixed station in a coastal area of the Baltic Sea has been studied for a period of time (January 1987 to June 1988) covering the annual biologicalcycle. DMS in the surfacewaters of the brackishBaltic Sea showed a clear seasonal variation, ranging from 2to 200 ng SL -1 . Lowest concentrations were in winter, peak values followed the spring bloom, and a pronounced maximum was found during the summer (July-August). Concentrations above lowwinter levels occurred only when thetrophogenic layer was depleted of inorganic nitrogen.From our data it is clearthat the seasonal variation in DMS concentration is related to biological activity.However, we did not findany correlations between DMS concentration and gross parameters such as chlorophyll a, totalphytoplankton biomass, or primary production on an annual basis.Further, we were not able to relate high DMS concentrations to any particular phytoplankton species or species assemblages. It appears that DMSproduction isprimarily associated withphytoplankton growth under nitrogen-limited conditions and notwithcertain species. We found a significant correlation of ambient DMSconcentration with copepod and total zooplankton biomass, suggesting zooplankton grazing pressure as the major factor responsible for the liberationof /3-dimethylsulfoniopropionate (DMSP) fromphytoplankton cells andthus for the DMS production. The turnover time of DMS in the watercolumn was calculated to be of the orderof 2 days, andthe most effective sink process seems to be of microbiological and/orchemical origin. Previousto this study, no in situ data set has beenavailable to test the relativeimportance of the variousfactors responsible for the DMS production in seawaters. We have demonstrated that variations in DMS concentration must be lookedupon as the result of complex physiological as well as ecological interactions. INTRODUCTION The ocean surface layer plays an important role in the global biogeochemical sulfur cycle. Dimethyl sulfide(DMS) constitutes about 90% of the reduced volatile sulfur in surface seawaters [Wakeham et al., 1987; Leck and Bggan- der, 1988] and is far in excess of the concentration expected at equilibrium with atmospheric concentration [Lovelock et al., 1972; Maroulis and Bandy, 1977;Barnard et al., 1982; Clineand Bates, 1983;Dacey and Wakeham, 1984].This gives riseto a flux of DMS from the ocean to the atmosphere estimated to be in the range30-50 Tg S a-1 globally [Andreae, !986]. The total naturalflux of gaseous sulfur to the atmosphere originating from oceans, continents, and volcanoes isestimated to bearound 80TgS a-• . This figure is ofthesame order of magnitude astheanthropogenic sulfur flux from fossil fuel combustion [Cullis andHirschler, 1980; Mbller, 1984]. In theboundary layerof the marine atmosphere, DMS is photochemically oxidized to non-sea-salt sulfate (NSS- SO42-) viaintermediates such assulfur dioxide (SO2) and methane sulfonic acid (CH3SO3 H) [Andreae, !986].The NSS-SO42- aerosol in the marine troposphere is the major contributor to the acidityof naturalprecipitation [Chadson •Department ofMeteorology. 2 A , sk6 Laboratory, Section Marine Ecology. 'Department of Geology, Section Microbial Geochemistry. 4Ask6 Laboratory, Section Marine Ecology and Department of Zoology. Copyright 1990 by the American Geophysical Union. Paper number 89JC03150. 0148-0227/90/89 J C-03150505.00 and Rodhe, !982] and probably also to the number of cloud condensation nuclei (CCN), which are critical to cloud formation and therefore important to the Earth's radiation budget [Chadson et al., 1987]. Phytoplankton is generally believed to be the major source of DMS in the oceans [Andteac et al., 1983;Barnard et al., 1984; Holligan et al., 1987; Keller et al., 1989], but the physiology of its production and release as well as its removal processes is still poorly understood. Most likely, DMS is produced by enzymatic cleavageof/5-dimethylsul- foniopropionate(DMSP) originating from phytoplankton [Challengerand Simpson, 1948;Ackman et al., 1966;Ishida, 1968; White, 1982; Granroth and Hattula,. 1976] and mac- roalgae [Dacey et al., 1987]. It has been suggestedthat DMSP increases in nitrate- limited situations by replacingits nitrogenanalogue,glycine betaine [Andteac, !986]. Both substancesare believed to be involved in cellular osmoregulation, although the mecha- nisms are less well known [Reed, 1983; Vairavamurthy et al., 1985; Wyn Jones and Storey, 1981; Dickson and Kirst, 1986]. A commonassumption has been that phytoplankton cells excrete DMS directly [Vairavamurthy et al., 1985]. More recently, however, it has been experimentally shown that processes such as lysis of the phytoplankton, zooplank- ton grazing, and bacterial attack on the phytoplankton greatly increase the release of DMS [Zinder et al., 1977; Dacey and Wakeham, 1986; Nguyen et al., 1988]. Ventila- tion to the atmosphere [Bates et al., 1987], photochemical oxidation [Brimblecombe and Shooter, 1986], and microbial degradation [Sivelti and Sundman, 1975; Kanagawa and Kelly, !986; Zeyer et at., 1987] have been suggested as prominentremoval processes in surfaceseawaters. 3353