The quantitative role of microzooplankton grazing in dimethylsulfide (DMS) production in the NW Mediterranean

The ubiquitous, biogenic trace gas dimethylsulfide (DMS) represents the largest natural source of atmospheric sulfur. Given DMS involvement in cloud formation and climate, understanding and parameterizing the oceanic DMS source and cycling processes is a necessary challenge. We report DMS cycling rates from microzooplankton dilution grazing experiments conducted monthly during 1 year in coastal northwestern Mediterranean waters. Concentrations of DMS, its algal precursor dimethylsulfoniopropionate (DMSPt) and chlorophyll a (Chla) ranged 0.9–11 nmol L^?1, 10–71 nmol L^?1, and 0.2–1.5 µg L^?1, respectively. By comparing the growth and stock production rates of the DMSP-producing algae to those of total phytoplankton, we estimated that 3?±?4% (range 0.4–12%) of the carbon primary production was invested in DMSP biosynthesis. Microzooplankton grazing rates on DMSP-producing phytoplankton (0.46–1.45 day^?1) were generally higher than those on the bulk assemblage (0.08–0.99 day^?1), except in midsummer months. This could have been due to the smaller size of most DMSP producers. There was no indication of micrograzer selection against DMSP-containing phytoplankton, since they were not grazed at lower rates than the bulk phytoplankton assemblage. A proportion of 6–20% of the grazed DMSP was converted into DMS, and this grazing-derived production accounted for 32–96% of dark gross DMS production by the total community. Bacteria consumed daily?≤?14–100% of the gross DMS production, which resulted in biological DMS turnover times of 1 to?≥?10 days. Throughout the year, grazing-mediated DMS production explained 73% of the variance in the DMS concentration, implying that microzooplankton grazing plays a major role in controlling DMS concentration in surface waters across a broad range of environmental and productivity conditions in the Mediterranean Sea. These findings should help improve the representation of herbivore grazing in prognostic models to predict the distribution and dynamics of the global DMS emission and its feedback response to changing climate.