Article pubs.acs.org/est
Direct Linkage between Dimethyl Sulfide Production and Microzooplankton Grazing, Resulting from Prey Composition Change under High Partial Pressure of Carbon Dioxide Conditions Ki-Tae Park,† Kitack Lee,*,† Kyoungsoon Shin,‡ Eun Jin Yang,§ Bonggil Hyun,‡ Ja-Myung Kim,∥ Jae Hoon Noh,⊥ Miok Kim,† Bokyung Kong,† Dong Han Choi,⊥ Su-Jin Choi,† Pung-Guk Jang,‡ and Hae Jin Jeong# †
School of Environmental Science and Engineering, Pohang University of Science and Technology, Pohang 790-784, Korea Korea Institute of Ocean Science and Technology/South Sea Institute, Jangmok 656-830, Korea § Korea Polar Research Institute, Incheon 406-840, Korea ∥ Department of Geosciences, Princeton University, Princeton, New Jersey 08544, United States ⊥ Korea Institute of Ocean Science and Technology, Ansan 425-600, Korea # School of Earth and Environmental Sciences, Seoul National University, Seoul 151-747, Korea ‡
S Supporting Information *
ABSTRACT: Oceanic dimethyl sulfide (DMS) is the enzymatic cleavage product of the algal metabolite dimethylsulfoniopropionate (DMSP) and is the most abundant form of sulfur released into the atmosphere. To investigate the effects of two emerging environmental threats (ocean acidification and warming) on marine DMS production, we performed a large-scale perturbation experiment in a coastal environment. At both ambient temperature and ∼2 °C warmer, an increase in partial pressure of carbon dioxide (pCO2) in seawater (160−830 ppmv pCO2) favored the growth of large diatoms, which outcompeted other phytoplankton species in a natural phytoplankton assemblage and reduced the growth rate of smaller, DMSP-rich phototrophic dinoflagellates. This decreased the grazing rate of heterotrophic dinoflagellates (ubiquitous micrograzers), resulting in reduced DMS production via grazing activity. Both the magnitude and sign of the effect of pCO2 on possible future oceanic DMS production were strongly linked to pCO2-induced alterations to the phytoplankton community and the cellular DMSP content of the dominant species and its association with micrograzers.
■
INTRODUCTION In the atmosphere, dimethyl sulfide (DMS) is oxidized to nonsea-salt sulfate aerosols, which can act as cloud condensation nuclei and, thereby, exert a cooling effect by increasing planetary albedo.1 The CLAW hypothesis suggests that a negative feedback loop may exist between oceanic phytoplankton and climate through the production of oceanic DMS.2 This pioneering hypothesis has been challenged by studies reevaluating the occurrence of climate feedback.3 However, understanding the consequences of climate change on oceanic DMS production remains of considerable interest to the scientific community because oceanic DMS production contributes to approximately 50% of global biogenic sulfur emission4 and DMS is involved in the oxidation and production of important climate gases, such as ozone, methane, and methanesulfonic acid.5,6 The past 20 years of research have revealed that the interaction of multiple processes (i.e., phytoplankton community structure, grazing by zooplankton, and bacterial activity), © 2014 American Chemical Society
acting either simultaneously or in sequence, regulate oceanic DMS production.1,7 DMS is mostly produced by intra- or extracellular enzymatic cleavage of dimethylsulfoniopropionate (DMSP) by DMSP-lyase, which is synthesized by algae and bacteria following DMSP secretion from producer cells or release following viral attack or grazing by zooplankton.7−9 Although the Earth’s climate system may be affected by DMS, the impact of future oceanic conditions (e.g., acidification and warming) on DMS production is poorly understood and warrants investigation. According to recent modeling simulations, ocean acidification may accelerate anthropogenic warming because of decreased emission of oceanic DMS.10 An integrated understanding of the sensitivity of marine biota to human-induced ocean acidification and warming can be Received: Revised: Accepted: Published: 4750
July 29, 2013 March 24, 2014 April 1, 2014 April 1, 2014 dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
through a 30 m flexible tube wrapped around the lower parts of the seawater mixer; the target temperature then was maintained during the experimental period (see Figure S1C of the Supporting Information). To initiate a phytoplankton bloom, on day 0, nutrients were added to each enclosure to yield standardized initial concentrations of 15.6 ± 0.8 μmol kg−1 nitrate (N), 0.93 ± 0.05 μmol kg−1 phosphate (P), and 13.4 ± 0.8 μmol kg−1 silicate (Si). The initial concentration ratio of N/P was approximately set to the Redfield value of 16.8. Each enclosure was sampled daily at a depth of ∼1 m. In a previous experiment conducted at the same site, we confirmed that all enclosures received approximately the same intensity of solar radiation (within ±5% of daily mean values) during experiments (see Figure S3 of the Supporting Information by Kim et al.). The highest values of simulated pCO2 (∼840 ppmv) and temperature (∼2 °C warmer than ambient) approximate the predicted conditions in the year 2100 based on model projections of the A2 Scenario of the Intergovernmental Panel on Climate Change Special Report on Emissions Scenarios.20 In contrast to experiments conducted using either excessively simplified planktonic food chains (e.g., on the basis of single phytoplankton species) or densities that are unrealistically high relative to those naturally occurring in ocean environments, mesocosm experiments using natural communities enable more comprehensive hypothesis testing under natural ocean conditions and will provide insights into the functioning of this ecosystem in the future. The experiment was carried out in the coastal waters of Korea (Jangmok, 34.6° N and 128.5° E) over 19 days from May 2 to 20, 2012. The experiment involved nine enclosures each of approximately 2400 L and included acidification and greenhouse treatments. For the acidification treatment, six of the enclosures were used to simulate a pCO2 range of 160−830 ppmv (average pCO2 values over the experimental period of 160, 230, 420, 610, 700, and 830 ppmv). For the greenhouse treatment, the remaining three enclosures were operated at 2 °C warmer than ambient, at pCO2 concentrations of 290, 690, and 790 ppmv, respectively (see Figure S1 of the Supporting Information). Analysis of DMS, DMSP, and DMSP-lyase Activity (DLA). DMS analysis was conducted using DMS extraction and trapping devices and a gas chromatograph equipped with a flame photometric detector (GC−FPD).21 Gravity filtration was used to filter 50 mL of sample through a 47 mm glass-fiber filter (Whatman GF/F). After filtration, 5−10 mL of filtrate was injected into a sparging chamber for DMS extraction. The sample volume for DMS analysis was adjusted to accommodate a range of DMS concentrations encountered in our experiment. Samples for DMSP analysis were collected and preserved following the widely accepted procedure detailed by Kiene and Slezak.22 To determine the particulate DMSP concentration, we measured the total DMSP (dissolved and particulate forms) and dissolved DMSP; the particulate DMSP concentration was calculated by the difference. A small volume of 50% H2SO4 (5 μL/mL of sample) was added to each sample (∼10 mL) to ensure preservation until DMSP analysis. For each sample, analysis of the dissolved DMSP was initiated by gravity filtering of the sample through a 47 mm glass-fiber filter (Whatman GF/ F) using the small-volume gravity drip filtration procedure. The sample was then preserved by the addition of 50% H2SO4 (5 μL addition/mL of sample) until analyzed. The preserved DMSP sample was hydrolyzed to DMS by the addition of 10 N
gained from controlled in situ experiments using natural communities of phytoplankton in large enclosures equipped to simulate predicted future partial pressure of carbon dioxide (pCO2) concentration and temperature conditions. Several experiments of this type have been performed in various marine environments to investigate the effect of ocean acidification on DMSP and DMS production.11−16 However, differences in the conclusions drawn from those studies have not been adequately explained, and we are unaware of any other previous studies that have reported changes in DMS production associated with changes in major phytoplankton species induced by increased pCO2 and their association with micrograzers. Such sequential interactions of multiple biological processes may be key to marine DMS production under the present and future high pCO2 conditions.7 In this study, we evaluated the community-scale effects of increased levels of pCO2 on marine DMS production at ambient temperature and ∼2 °C warmer conditions (greenhouse) using a controlled mesocosm facility. We considered our results in the context of comparable studies to evaluate the significance of the effects of increased pCO2 on planktonic DMS production and in terms of the strong correlation between observed responses and underlying biological community interactions.
■
MATERIALS AND METHODS Mesocosm Facility and Operational Sequence. The mesocosm facility consists of a floating raft supporting nine impermeable cylindrical enclosures. Each enclosure is 3 m in height, has a volume of approximately 2400 L, and is equipped with a pCO2 regulation unit and a bubble-mediated seawater mixer. Each enclosure was filled two-thirds with seawater and capped with a transparent dome that transmits incoming radiation.17 The target levels for seawater pCO2 in the enclosures were achieved by adding CO2-saturated seawater to the enclosures while filling them with ambient seawater, which was passed through a filter (100 μm pore size). Such use of CO2-saturated seawater is more suitable for large CO2 perturbation facilities equipped with large sample enclosures needing large amounts of CO2. The addition of CO2-saturated seawater into the enclosures increases the dissolved inorganic carbon concentration while leaving the alkalinity unchanged, thereby approximately mimicking the ocean acidification process. In addition, the effects of this CO2 perturbation method on marine organisms have been shown to be less significant than the effect of direct target CO2 aeration.18,19 Dependent upon the target pCO2 level, 2−15 L of CO2-saturated seawater was added to each enclosure. This CO2-gradient approach carries a low risk of damage to the enclosure relative to a replication approach (e.g., each treatment with three replicates) and provides a greater chance of detecting a threshold level for any of the CO2-sensitive processes.18 For 20 min/day, air containing the target CO2 concentration was flowed (approximately 0.5 L min−1) into the seawater mixer to generate a convective flow within the enclosure to ensure homogeneity. This procedure resulted in an even distribution of biogenic particles (particulate organic matter) throughout the enclosures and prevented particles settling to the bottom.17 Immediately following completion of the 20 min seawater aeration period, the enclosures were sampled. The seawater temperature in the greenhouse treatments reached the target value (2 °C warmer than ambient) within 7 h by circulating warm water (∼10 °C warmer than ambient) 4751
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
period were used to calculate the apparent phytoplankton growth rate and microzooplankton grazing. Apparent growth rates were calculated as r = 1/t ln(Ct/C0), where t is the time and C0 and Ct are the initial and final chl a concentrations, respectively. Microzooplankton grazing rates were calculated from the slopes of the regressions of total chl a change during the incubation period as a function of the percentage of dilution.
NaOH (0.25 mL/mL of sample) and allowed to react overnight in the dark. DMS that was subsequently evolved was measured using a GC−FPD. The detection limit was 0.2 nM S (DMS), and the analytical precision was better than 3%. The response of the GC−FPD was calibrated against standard DMS solutions of known concentration (1−100 nmol L−1). Measurements of DLA were carried out following the method described by Steinke et al.23 An aliquot of seawater sample (150−250 mL) was filtered through a 47 mm polycarbonate membrane filter (Millipore; 3 μm pore size). The filter were placed into a Petri dish and stored at −80 °C until analyzed. The filter was transferred into a 10 mL bottle containing 300 mmol L−1 sterile 1,3-bis[tris(hydroxymethyl)methylamino]propane (BTP) buffer solution that was amended with 0.5 mol L−1 NaCl at pH 8.2. Analysis was conducted on 100−200 μL of the extract. The linear production of DMS was quantified at 20 °C for 15−60 min, following the addition of 10 μL of 1 mol L−1 DMSP stock (the final DMSP concentration was 1 mmol L−1, and the final pH was 8.2). Phytoplankton and Microzooplankton Abundance. To count phytoplankton and microzooplankton cell abundance, 500 mL of seawater sample was preserved with Lugol’s iodine solution. The samples were pre-concentrated to 50 mL by sedimentation for at least 48 h and analyzed immediately by a light microscope, together with a Sedgwick-Rafter chamber and a sedimentation chamber. This enabled identification of the plankton groups and calculation of their concentrations. The plankton identification and cell-counting procedure was repeated 2 or 3 times for each sample. For each species, the mean cell concentration was determined by averaging the 2 or 3 separate cell counts. The average uncertainty in cell counting was less than 10% of the total population of each species. To estimate the carbon biomass of microzooplankton, the cell volume was calculated by measuring cell dimensions with an ocular micrometer in the microscope. Microzooplankton assemblages were classified as ciliates and heterotrophic dinoflagellates (HDFs). Conversion factors and equations were used to translate cell volume into carbon biomass, including 0.19 μg of C μm−3 for naked ciliates,24 carbon (pg) = 444.5 + 0.053(lorica volume, μm3) for loricate ciliates,25 and carbon (pg) = 0.216(cell volume, μm3)0.939 for HDFs.26 Incubation Experiments Measuring Grazing Activity. Microzooplankton grazing rates over a 24 h period were estimated every 2−3 days using the dilution method based on chlorophyll a (chl a) concentration data.27 A dilution series was prepared in 1.2 L glass bottles that had previously been cleaned with 10% HCl and thoroughly rinsed with distilled water. Four dilutions (0, 25, 50, and 75%) were prepared in particle-free seawater, which was prepared by gravity filtration through a 0.2 μm cellulose acetate filter, to minimize cell damage. Excess nitrate (15 μmol kg−1) and phosphate (1 μmol kg−1) were added to each dilution bottle to ensure that nutrients would not be a limiting factor in algal growth, because insufficient nutrients would bias the results.27 The experimental bottles containing samples from the nine enclosures were incubated in a chamber receiving a continuous flow of ambient seawater, whereas the bottles containing samples from the greenhouse enclosures were incubated in a chamber receiving a flow of seawater 2 °C higher than ambient. Each experimental bottle was covered with a neutral density screen matching the light intensity at a depth of 1 m. Samples for chl a analysis were removed from the dilution bottles immediately and at 24 h. The chl a concentrations measured during the incubation
■
RESULTS AND DISCUSSION In all enclosures, the phytoplankton biomass (measured as the chl a concentration) reached a maximum level on days 4−6 and decreased thereafter (Figure 1A). Regardless of the pCO2 and
Figure 1. Concentrations of (A) chl a, (B) DMS, and (C) DMSPP during the experimental period. The vertical dashed lines in panels A− C indicate the boundaries of the three periods of increasing (period 1), decreasing (period 2), and no change (period 3) in cell population. The boundaries were arbitrarily chosen on the basis of the trends in chl a concentration. Panels D, E, and F show the total accumulated concentration (integrated over the period of days 1−18) for chl a, DMS, and DMSP, respectively, as a function of the initial pCO2 concentration. The solid and dashed lines in D−F indicate the best fits of the six ambient treatments and the three elevated (2 °C warmer than ambient) temperature treatments, respectively. The colored filled and open symbols indicate the ambient and 2 °C warmer than ambient treatments, respectively.
temperature treatment, the total accumulated chl a concentration (integrated during the study period) was similar (98.3 ± 5.2 μg L−1) among all enclosures (Figure 1D). However, the trend in the DMS concentration differed from that of chl a. In each treatment, the DMS concentration was low (∼1.4 nM) during the bloom period (period 1; days 0−5) and did not differ significantly among treatments. During the post-bloom period (period 2, days 6−13), the DMS concentration increased in each enclosure, reached a maximum level that varied widely among treatments (10−100 nmol L−1) on days 11−15 and decreased thereafter (Figure 1B). The trend in 4752
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
period of increasing DMS concentration (days 4, 7, 9, 11, and 13; Figure 3; r2 = 0.60; p < 0.01; n = 36). These results indicate
particulate DMSP (DMSPP) concentration was broadly consistent with that of DMS, although the maximum levels occurred approximately 2 days earlier (Figure 1C). The total DMS and DMSPP concentrations clearly showed the dependence upon pCO2 (panels E and F of Figure 1). As shown in Figure 1E, the total integrated DMS concentration decreased at a rate of 58 ± 17 nmol L−1 per 100 ppmv pCO2 increase. Relative to the total integrated DMS concentration in the 160 ppmv pCO2 enclosure, the concentration was 82% lower in the highest pCO2 treatment (830 ppmv pCO2), indicating an adverse effect of increased pCO2 on oceanic DMS production. At all three pCO2 conditions, a 2 °C increase in temperature reduced the total integrated DMS concentration compared to that at ambient temperature (Figure 1E). Moreover, the DMS concentration decreased at a rate of 24 ± 24 nmol L−1 per 100 ppmv increase in pCO2. The difference in the integrated DMS concentration between ambient temperature and 2 °C warmer appeared to decrease as pCO2 levels increased (Figure 1E); however, because measurements were made at only two temperatures, the effect of pCO2 on DMS production as a function of the temperature remains unclear. Grazing activity is the most likely explanation for the difference in net DMS production among the pCO2 treatment enclosures in our experiment. In all enclosures, the overall grazing activity was largely governed by the HDF Protoperidinium spp. (cell volume of ∼28 000 μm3), followed by ciliates. The production of DMS was highest during days 7, 9, 11, and 13 of period 2, and the measured grazing rates were inversely correlated with the pCO2 levels (Figure 2A; r2 = 0.70; p =
Figure 3. Correlation between DMS concentrations and corresponding grazing rates measured at days 4, 7, 9, 11, and 13. The solid line indicates the best fit.
that grazing activity was the dominant factor in DMS production in our experiment, regardless of the pCO2 level. The grazer ciliate appeared to account for a large proportion of the microzooplankton biomass at high pCO2, but the abundance of ciliate showed no correlation with pCO2 at levels lower than 800 ppmv (see Figure S4 of the Supporting Information). Not all species have the same ability to form DMSP; for example, diatoms generally produce only small amounts of DMSP, whereas prymnesiophytes and dinoflagellates produce considerably greater amounts.1,7 For grazing activity to be a major cause of the difference in net DMS production among the enclosures, DMSP-containing prey species must be involved. In our experiment, the large diatom Cerataulina pelagica (cell volume of ∼24 100 μm3), which was the most abundant species during period 1, decreased in abundance during the transition from period 1 to period 2. This coincided with a shift to the phototrophic dinoflagellate Alexandrium spp. (cell volume of ∼2200 μm3) as the dominant species; the population of HDFs rapidly increased immediately following the population peaks of Alexandrium spp. (Figure 4). In contrast to other prey groups (dinoflagellates, autotrophic picoplankton, nanoflagellates, and diatoms), only Alexandrium spp. (high cellular DMSP content and DLA) showed a negative growth response to increasing pCO2 levels (Figure 5A; r2 = 0.80; p = 0.02), whereas C. pelagica (low cellular DMSP content and no DLA) showed a positive response to increasing pCO2 levels (Figure 5B; r2 = 0.72; p = 0.03). Biomarker pigment data showing changes in phytoplankton species composition are also consistent with this finding (Figure 6). In all enclosures, the diatom-specific pigment fucoxanthin was dominant on day 0, accounting for 90.5 ± 1.0% of the phytoplankton. However, during the DMS producing period (day 12), the combination of fucoxanthin and peridinin, a dinoflagellate-specific pigment, accounted for 79.9 ± 2.9% of the six major pigments found, but the proportions of fucoxanthin and peridinin showed opposite trends in response to pCO2 levels (Figure 6B). These results further indicate that the major dinoflagellate species found in our experiments negatively responded to increasing pCO2, whereas the major diatom species responded positively to increasing pCO2. Some studies have also shown that elevated pCO2 enhanced the growth rate of diatoms,16,28−30 which may be due to energy savings caused by the downregulation of the CO2 concentrating mechanism.30 The cellular DMSP content and DLA of Alexandrium spp. range from 180 to 290 mmol
Figure 2. (A) Mean grazing rate during the DMS production period (days 7, 9, 11, and 13) as a function of pCO2. The error bars indicate 1 standard deviation (1σ) from the 4 day mean values. Daily grazing rates are presented in Figure S3 of the Supporting Information. (B) Total accumulated populations of HDFs during the experimental period (days 1−18) as a function of pCO2. The solid and dashed lines indicate the best fits of the six ambient and three elevated (2 °C warmer than ambient) temperature treatments, respectively.
0.04). Results obtained from all 42 incubation experiments are presented in Table S1 and Figures S2 and S3 of the Supporting Information. The populations of HDF were also inversely correlated with pCO2 levels (Figure 2B), thereby yielding a significant correlation between the variation in the DMS concentration and the change in the grazing rate during the 4753
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
Figure 6. Relative composition of six major biomarker pigments of phytoplankton on (A) day 0 and (B) day 12. Peridinin (Per), 19′hexanoyloxyfucoxanthin (19-hex), alloxanthin (Allo), chlorophyll b (Chl-b), zeaxanthin (Zea), and fucoxanthin (Fuc) are markers for dinoflagellates, prymnesiophytes, cryptophytes, green flagellates, cyanobacteria, and diatoms, respectively.
Figure 4. Cell abundance of the three key plankton species in response to pCO2 change during the experimental period. The colored lines indicate C. pelagica (red), Alexandrium spp. (blue), and HDFs (green). The vertical dashed lines show the boundaries of the three periods, as defined in Figure 1
The ratio of the DMSPP concentration to total biomass [expressed as particulate organic carbon (POC)] provided another compelling line of evidence supporting our explanation. During period 1 (days 1 and 2), the ratio of DMSPP/ POC remained approximately unchanged over the pCO2 range investigated (Figure 5D) but decreased with increasing pCO2 levels during period 2 (days 5 and 6) (Figure 5E). The rate of decrease in the ratio became more pronounced as the experiment proceeded (from Figure 5E to Figure 5F). The results of the measurements of DLA were also consistent with our explanation. The DLA decreased with increasing pCO2 levels during 3 days of period 2 (days 6, 8, and 10; Figure 5C). In particular, the DLA was highly correlated with the change in populations of dinoflagellate during this period. Hence, the mixing of planktonic DMSP and DMSP-lyase during grazing (resulting in DMS production) contributed considerably to DMS production during period 2. The higher values of DMSPP/POC (greater proportions of DMSP-containing phytoplankton) and higher DLA in the low pCO2 treatments, in conjunction with the higher grazing rates in those treatments, indicate that the higher abundance of DMSP-rich prey under the low pCO2 treatment conditions may lead to higher HDF grazing activity compared to that observed under high pCO2 conditions. This pCO2 dependence of grazing activity eventually resulted in greater DMS production in the former treatments than in the latter treatments. Our results indicate a direct link between reduced DMS production under high pCO2 conditions and the interactions of phytoplankton species and their association with grazers. A previous mesocosm study conducted at the same site but in a different season16 (December for the 2008 experiment versus April for the present study) also found that the growth of a major diatom, Skeletonema costatum, was enhanced at high pCO2 and that grazing activity associated with this species had an impact on DMS production. Although the major diatom groups found in these studies responded positively to increased pCO2, the total DMS production in these studies had opposite signs and was critically dependent upon the interactions of the plankton community (i.e., competition between non-DMSPcontaining and other DMSP-containing species and their associations with potential predators). In particular, the
Figure 5. Total cell abundance of (A) Alexandrium spp. and (B) C. pelagica integrated over the experimental period (days 1−18) as a function of pCO2. (C) Mean DLA measured during 3 days (6, 8, and 10) of the DMS production period as a function of pCO2. Daily DLA is shown in Figure S5 of the Supporting Information. (D−F) Mean ratios of DMSPP/POC during 2 days as a function of pCO2: (D) days 1 and 2, (E) days 5 and 6, and (F) days 11 and 12. The error bars indicate 1 standard deviation (1σ) from the (C) 3 day means and (D− F) 2 day means. The solid and dashed lines indicate the best fits of the six ambient treatments and three elevated (2 °C warmer than ambient) temperature treatments, respectively.
Lcell−1 and from 230 to 1620 mmol Lcell−1 h−1, respectively,31,32 and Alexandrium spp. accounted for more than 60% of the total dinoflagellate population in our experiment, except for pCO2 treatments of >700 ppmv, where its contribution decreased to 30%. In all enclosures, Alexandrium spp. were largely fed upon by Protoperidinium spp. (Figure 4). 4754
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
more importantly, their association with potential micrograzers. In the context of increasing global ocean acidification, the key implication of our results is that increased seawater pCO2 could change the species composition of the phytoplankton community and possibly change the dominant phytoplankton species. If so, DMS production in the ocean will be critically dependent upon whether the dominant species contains DMSP and is a suitable prey for marine micrograzers, because this will determine the grazing activity and, thereby, grazing-mediated DMS production.
conversion of DMSP to DMS was considerably accelerated in the 2012 experiment compared to the 2008 experiment, resulting in higher DMS production in 2012. In the 2008 experiment, the high pCO2 conditions increased the growth rate of DMSP-containing small diatoms (i.e., S. costatum; cell volume of ∼70 μm3; DMSP = 0.2−0.4 mmol Lcell−1), which led to an increase in the rate of grazing by microzooplankton, and, thereby, produced higher levels of DMS. In both experiments, the mean grazing rates were similar during the DMS production period (0.73 ± 0.13 day−1 for 2008 versus 0.78 ± 0.27 day−1 for 2012), but the DMS/DMSP ratio (0.03 ± 0.01) observed in the 2008 experiment was only 25% of the level found in the 2012 experiment (0.11 ± 0.2). The low DMS/DMSP ratio indicates a lower level of conversion of cellular DMSP to DMS. This suggests that some DMSP released into seawater during micrograzer consumption of S. costatum (which contains DMSP but lacks DLA) may be converted into DMS through bacterial DLA in the seawater, while most of the cellular DMSP is degraded to other sulfur compounds through a bacterial demethylation process. In contrast to the 2008 experiment, in the 2012 experiment, the micrograzers largely fed on Alexandrium spp., which contain both high DMSP concentration and DLA. This grazing activity resulted in more complete mixing of cellular DMSP and DMSP-lyase and, thereby, led to greater conversion of DMSP to DMS (indicating more DMS production). This enhancement of DMS production during grazing depends upon the cellular DMSP content and the cellular DLA of the prey. This highlights the complexity of oceanic DMS production in response to elevated pCO2 levels. The effects of increased pCO2 on marine DMS production have been examined in several studies.11−16,33 With the exception of the work by Kim et al.,16 the results of these studies have been largely consistent and have generally pointed to a reduction in DMS production. In several mesocosm studies carried out in Norwegian coastal waters, the reduction in the prymnesiophyte Emiliania huxleyi in response to increasing pCO2 was found to be responsible for decreases in DMS and DMSP production.11,15 A similar conclusion was also drawn from laboratory experiments with E. huxleyi.33 In marine environments, microzooplankton grazing is the main cause of phytoplankton mortality34 and can either directly or indirectly contribute to DMS production. Therefore, the role of grazing in marine DMS production has been assessed under a wide range of experimental conditions.7,35,36 Grazingmediated DMS production per cell has been reported to be an order of magnitude greater than the combination of direct algal release and cell death by viral infection.9,37,38 Although not yet widely confirmed, these results indicate that DMS production associated with grazing activity may be a key mechanism of DMS production in present and future marine environments. In particular, our findings may be important in relation to DMS production in productive areas, including highlatitude oceans, coastal waters, and upwelling regions, where diatom and other DMSP-producing phytoplankton (i.e., dinoflagellates and prymnesiophytes) blooms frequently occur together or in sequence over time.39 Analysis of experiments conducted at diverse scales indicates that marine DMS production is affected by the interactions of multiple factors acting simultaneously and sequentially over time. Nonetheless, we conclude that the oceanic DMS production under high pCO2 conditions primarily depends upon the growth of DMSP-rich phytoplankton species and,
■
ASSOCIATED CONTENT
S Supporting Information *
Pigments analysis, statistical analysis, grazing rates obtained from the dilution experiments carried out at days 4, 7, 9, 11, and 13 (Table S1), (A) seawater pCO2, (B) pH, and (C) temperature during the experimental period (Figure S1), apparent growth rates determined by chl a changes during the dilution incubation as a function of the fraction whole water in nine mesocosm enclosures at days 7, 9, 11, and 13 (Figure S2), grazing rate at days (A) 4, (B) 7, (C) 9, (D) 11, and (E) 13 as a function of pCO2 (Figure S3), total accumulated populations of ciliates during the experimental period (days 1− 18) as a function of pCO2 (Figure S4), and DLA measured at days (A) 6, (B) 8, and (C) 10 as a function of pCO2 (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Telephone: 82-54-279-2285. Fax: 82-54-279-8299. E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the Midcareer Researcher Program (2012R1A2A1A01004631) and the Global Research Project funded by the National Research Foundation (NRF) of the Ministry of Science, Information and Communication Technology (ICT), and Future Planning (MSIP). Partial support was provided by “Long-Term Change of Structure and Function in Marine Ecosystems of Korea” and “Management of Marine Organisms Causing Ecological Disturbance and Harmful Effects” funded by the Ministry of Oceans and Fisheries.
■
REFERENCES
(1) Liss, P. S.; Hatton, A. D.; Malin, G.; Nightingale, P. D.; Turner, S. M. Marine sulphur emissions. Philos. Trans. R. Soc. London, Ser. B 1997, 352, 159−169. (2) Charlson, R. J.; Lovelock, J. E.; Andreae, M. O.; Warren, S. G. Oceanic phytoplankton, atmospheric sulphur, cloud albedo, and climate. Nature 1987, 326, 655−661. (3) Quinn, P. K.; Bates, T. S. The case against climate regulation via oceanic phytoplankton sulphur emissions. Nature 2011, 480, 51−56. (4) Lana, A.; Bell, T. G.; Simó, R.; Vallina, S. M.; Ballabrera-Poy, J.; Kettle, A. J.; Dachs, J.; Bopp, L.; Salzman, E. S.; Stefels, J.; Johnson, J. E.; Liss, P. S. An updated climatology of surface dimethlysulfide concentrations and emission fluxes in the global ocean. Global Biogeochem. Cycles 2011, DOI: 10.1029/2010GB003850. (5) Chen, T.; Jang, M. Secondary organic aerosol formation from photooxidation of a mixture of dimethyl sulfide and isoprene. Atmos. Environ. 2012, 46, 271−278.
4755
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
space analysis of dimethylsulphide (DMS). J. Sea Res. 2000, 43, 233− 244. (24) Putt, M.; Stoecker, D. K. An experimentally determined carbon: volume ratio for marine “oligotrichous” ciliates from estuarine and coastal waters. Limnol. Oceanogr. 1989, 34 (6), 1097−1103. (25) Verity, P. G.; Langdon, C. Relationships between lorica volume, carbon, nitrogen and ATP content of tintinnids in Narragansett Bay. J. Plankton Res. 1984, 6 (5), 859−868. (26) Menden-Deuer, S.; Lessard, E. J. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnol. Oceanogr. 2000, 45 (3), 569−579. (27) Landry, M. R.; Hassett, R. P. Estimating the grazing impact of marine micro-zooplankton. Mar. Biol. 1982, 67, 283−288. (28) Kim, J.-M.; Lee, K.; Shin, K.; Kang, J.-H.; Lee, H.-W.; Kim, M.; Jang, P.-G.; Jang, M.-C. The effect of seawater CO2 concentration on growth of a natural phytoplankton assemblage in a controlled mesocosm experiment. Limnol. Oceanogr. 2006, 51 (2), 1629−1636. (29) Wu, Y.; Gao, K.; Riebesell, U. CO2-induced seawater acidification affects physiological performance of the marine diatom Phaeodactylum tricornutum. Biogeosciences 2010, 7, 2915−2923. (30) Tortell, P. D.; Payne, C. D.; Li, Y.; Trimborn, S.; Rost, B.; Smith, W. O.; Riesselman, C.; Dunber, R. B.; Sedwick, P.; DiTullio, G. R. CO2 sensitivity of southern ocean phytoplankton. Geophys. Res. Lett. 2008, 35 (4), L04605. (31) Caruana, A. DMS and DMSP production by marine dinoflagetllates. Ph.D. Dissertation, University of East Anglia, Norwich, U.K., 2010. (32) Wolfe, G. V.; Strom, S. L.; Holmes, J. L.; Radzio, T.; Olson, M. B. Dimethylsulfoniopropionate cleavage by marine phytoplankton in response to mechanical, chemical, or dark stress. J. Phycol. 2002, 38 (5), 948−960. (33) Arnold, H. E.; Kerrison, P.; Steinke, M. Interacting effects of ocean acidification and warming on growth and DMS-production in the haptophyte coccolithophore Emiliania huxleyi. Global Change Biol. 2013, 19 (4), 1007−1016. (34) Calbet, A.; Landry, M. R. Phytoplankton growth, microzooplankton grazing, and carbon cycling in marine systems. Limnol. Oceanogr. 2004, 49 (1), 51−57. (35) Archer, S. D.; Widdicombe, C. E.; Tarran, G. A.; Rees, A. P.; Burkill, P. H. Production and turnover of particulate dimethylsulphoniopropionate during a coccolithophore bloom in the northern North Sea. Aquat. Microb. Ecol. 2001, 24 (3), 225−241. (36) Wolfe, G. V.; Steinke, M. Grazing-activated production of dimethylsulfide (DMS) by two clones of Emiliania huxleyi. Limnol. Oceanogr. 1996, 41 (6), 1151−1160. (37) Lee, H.; Park, K.-T.; Lee, K.; Jeong, H. J.; Yoo, Y. D. Preydependent retention of dimethylsulfoniopropionate (DMSP) by mixotrophic dinoflagellates. Environ. Microb. 2012, 14 (3), 605−616. (38) Evans, C.; Kandner, S. V.; Darroch, L. J.; Wilson, W. H.; Liss, P. S.; Malin, G. The relative significance of viral lysis and microzooplankton grazing as pathways of dimethylsulfoniopropionate (DMSP) cleavage: An Emiliania huxleyi culture study. Limnol. Oceanogr. 2007, 52 (3), 1036−1045. (39) Degerlund, M.; Eilertsen, H. C. Main species characteristics of phytoplankton spring blooms in NE Atlantic and Arctic waters (68− 80° N). Estuaries Coasts 2010, 33, 242−269.
(6) Barnes, I.; Hjorth, J.; Mihalopoulos, N. Dimethyl sulfide and dimethyl sulfoxide and their oxidation in the atmosphere. Chem. Rev. 2006, 106 (3), 940−975. (7) Stefels, J.; Steinke, M.; Turner, S.; Malin, G.; Belviso, S. Environmental constraints on the production and removal of the climatically active gas dimethylsulphide (DMS) and implications for ecosystem modelling. Biogeochemistry 2007, 83, 245−275. (8) Zubkov, M. V.; Fuchs, B. M.; Archer, S. D.; Kiene, R. P.; Amann, R.; Burkill, P. A. Linking the composition of bacterioplankton to rapid turnover of dissolved dimethylsulfoniopropionate in an algal bloom in the North Sea. Environ. Microb. 2001, 3 (5), 304−311. (9) Dacey, J. W. H.; Wakeham, S. G. Oceanic dimethylsulfide: Production during zooplankton grazing on phytoplankton. Science 1986, 233, 1314−1316. (10) Six, K. D.; Kloster, S.; Ilyina, T.; Archer, S. D.; Zhang, K.; MaierReimer, E. Global warming amplified by reduced sulphur fluxes as a result of ocean acidification. Nat. Clim. Change 2013, 3, 975−978. (11) Hopkins, F. E.; Turner, S. M.; Nightingale, P. D.; Steinke, M.; Bakker, D.; Liss, P. S. Ocean acidification and marine trace gas emissions. Proc. Natl. Acad. Sci. U. S. A. 2010, 107 (2), 760−765. (12) Vogt, M.; Steinke, M.; Turner, S.; Paulino, A.; Meyerhöfer, M.; Riebesell, U.; LeQuéré, C.; Liss, P. S. Dynamics of dimethylsulphoniopropionate and dimethylsulphide under different CO2 concentrations during a mesocosm experiment. Biogeosciences 2008, 5, 407− 419. (13) Wingenter, O. W.; Haase, K. B.; Zeigler, M.; Blake, D. R.; Rowland, F. S.; Sive, B. C.; Paulino, A.; Thyrhaug, R.; Larsen, A.; Schulz, K.; Meyerhöfer, M.; Riebesell, U. Unexpected consequences of increasing CO2 and ocean acidity on marine production of DMS and CH2ClI: Potential climate impacts. Geophys. Res. Lett. 2007, 34, L05710. (14) Archer, S. D.; Kimmance, S. A.; Stephens, J. A.; Hopkins, F. E.; Bellerby, R. G.; Schulz, K. G.; Piontek, J.; Engel, A. Contrasting responses of DMS and DMSP to ocean acidification in Arctic waters. Biogeosciences 2013, 10, 1893−1908. (15) Avgoustidi, V.; Nightingale, P. D.; Joint, I.; Steinke, M.; Turner, S. M.; Hopkins, F. E.; Liss, P. S. Decreased marine dimethyl sulphide production under elevated CO2. Environ. Chem. 2012, 9, 399−404. (16) Kim, J.-M.; Lee, K.; Yang, E. J.; Shin, K.; Noh, J. H.; Park, K.-T.; Hyun, B.; Jeong, H.-J.; Kim, J.-H.; Kim, K. Y.; Kim, M.; Kim, H.-C.; Jang, P.-G.; Jang, M.-C. Enhanced production of oceanic dimethylsulfide resulting from CO2-induced grazing activity in a high CO2 world. Environ. Sci. Technol. 2010, 44 (21), 8140−8143. (17) Kim, J.-M.; Shin, K.; Lee, K.; Park, B.-K. In situ ecosystem-based carbon dioxide perturbation experiments: Design and performance evaluation of a mesocosm facility. Limnol. Oceanogr.: Methods 2008, 6, 208−217. (18) Riebesell, U.; Czerny, J.; von Bröckel, K.; Boxhammer, T.; Büdenbender, J.; Deckelnick, M.; Fischer, M.; Hoffmann, D.; Krug, S. A.; Lentz, U.; Ludwig, A.; Muche, R.; Schulz, K. G. Technical note: A mobile sea-going mesocosm system: New opportunities for ocean change research. Biogeosciences 2013, 10, 1835−1874. (19) Schulz, K. G.; Barcelos e Ramos, J.; Zeebe, R. E.; Riebesell, U. CO2 perturbation experiments: Similarities and differences between dissolved inorganic carbon and total alkalinity manipulations. Biogeosciences 2009, 6, 2145−2153. (20) Intergovernmental Panel on Climatic Change (IPCC). IPCC Fourth Assessment Report: Climate Change 2007. The Physical Science Basis; Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L., Eds.; Cambridge University Press: New York, 2007. (21) Park, K.-T.; Lee, K. High-frequency, accurate measurement of dimethylsulfide in surface marine environments using a microporous membrane contactor. Limnol. Oceanogr.: Methods 2008, 6, 548−557. (22) Kiene, R. P.; Slezak, D. Low dissolved DMSP concentrations in seawater revealed by small-volume gravity filtration and dialysis sampling. Limnol. Oceanogr.: Methods 2006, 4, 80−95. (23) Steinke, M.; Malin, G.; Turner, S. M.; Liss, P. S. Determinations of dimethylsulphoniopropionate (DMSP) lyase activity using head4756
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Direct Linkage between Dimethyl Sulfide Production and Microzooplankton Grazing, Resulting from Prey Composition Change under High Partial Pressure of Carbon Dioxide Conditions Ki-Tae Park,† Kitack Lee,*,† Kyoungsoon Shin,‡ Eun Jin Yang,§ Bonggil Hyun,‡ Ja-Myung Kim,∥ Jae Hoon Noh,⊥ Miok Kim,† Bokyung Kong,† Dong Han Choi,⊥ Su-Jin Choi,† Pung-Guk Jang,‡ and Hae Jin Jeong# †
School of Environmental Science and Engineering, Pohang University of Science and Technology, Pohang 790-784, Korea Korea Institute of Ocean Science and Technology/South Sea Institute, Jangmok 656-830, Korea § Korea Polar Research Institute, Incheon 406-840, Korea ∥ Department of Geosciences, Princeton University, Princeton, New Jersey 08544, United States ⊥ Korea Institute of Ocean Science and Technology, Ansan 425-600, Korea # School of Earth and Environmental Sciences, Seoul National University, Seoul 151-747, Korea ‡
S Supporting Information *
ABSTRACT: Oceanic dimethyl sulfide (DMS) is the enzymatic cleavage product of the algal metabolite dimethylsulfoniopropionate (DMSP) and is the most abundant form of sulfur released into the atmosphere. To investigate the effects of two emerging environmental threats (ocean acidification and warming) on marine DMS production, we performed a large-scale perturbation experiment in a coastal environment. At both ambient temperature and ∼2 °C warmer, an increase in partial pressure of carbon dioxide (pCO2) in seawater (160−830 ppmv pCO2) favored the growth of large diatoms, which outcompeted other phytoplankton species in a natural phytoplankton assemblage and reduced the growth rate of smaller, DMSP-rich phototrophic dinoflagellates. This decreased the grazing rate of heterotrophic dinoflagellates (ubiquitous micrograzers), resulting in reduced DMS production via grazing activity. Both the magnitude and sign of the effect of pCO2 on possible future oceanic DMS production were strongly linked to pCO2-induced alterations to the phytoplankton community and the cellular DMSP content of the dominant species and its association with micrograzers.
■
INTRODUCTION In the atmosphere, dimethyl sulfide (DMS) is oxidized to nonsea-salt sulfate aerosols, which can act as cloud condensation nuclei and, thereby, exert a cooling effect by increasing planetary albedo.1 The CLAW hypothesis suggests that a negative feedback loop may exist between oceanic phytoplankton and climate through the production of oceanic DMS.2 This pioneering hypothesis has been challenged by studies reevaluating the occurrence of climate feedback.3 However, understanding the consequences of climate change on oceanic DMS production remains of considerable interest to the scientific community because oceanic DMS production contributes to approximately 50% of global biogenic sulfur emission4 and DMS is involved in the oxidation and production of important climate gases, such as ozone, methane, and methanesulfonic acid.5,6 The past 20 years of research have revealed that the interaction of multiple processes (i.e., phytoplankton community structure, grazing by zooplankton, and bacterial activity), © 2014 American Chemical Society
acting either simultaneously or in sequence, regulate oceanic DMS production.1,7 DMS is mostly produced by intra- or extracellular enzymatic cleavage of dimethylsulfoniopropionate (DMSP) by DMSP-lyase, which is synthesized by algae and bacteria following DMSP secretion from producer cells or release following viral attack or grazing by zooplankton.7−9 Although the Earth’s climate system may be affected by DMS, the impact of future oceanic conditions (e.g., acidification and warming) on DMS production is poorly understood and warrants investigation. According to recent modeling simulations, ocean acidification may accelerate anthropogenic warming because of decreased emission of oceanic DMS.10 An integrated understanding of the sensitivity of marine biota to human-induced ocean acidification and warming can be Received: Revised: Accepted: Published: 4750
July 29, 2013 March 24, 2014 April 1, 2014 April 1, 2014 dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
through a 30 m flexible tube wrapped around the lower parts of the seawater mixer; the target temperature then was maintained during the experimental period (see Figure S1C of the Supporting Information). To initiate a phytoplankton bloom, on day 0, nutrients were added to each enclosure to yield standardized initial concentrations of 15.6 ± 0.8 μmol kg−1 nitrate (N), 0.93 ± 0.05 μmol kg−1 phosphate (P), and 13.4 ± 0.8 μmol kg−1 silicate (Si). The initial concentration ratio of N/P was approximately set to the Redfield value of 16.8. Each enclosure was sampled daily at a depth of ∼1 m. In a previous experiment conducted at the same site, we confirmed that all enclosures received approximately the same intensity of solar radiation (within ±5% of daily mean values) during experiments (see Figure S3 of the Supporting Information by Kim et al.). The highest values of simulated pCO2 (∼840 ppmv) and temperature (∼2 °C warmer than ambient) approximate the predicted conditions in the year 2100 based on model projections of the A2 Scenario of the Intergovernmental Panel on Climate Change Special Report on Emissions Scenarios.20 In contrast to experiments conducted using either excessively simplified planktonic food chains (e.g., on the basis of single phytoplankton species) or densities that are unrealistically high relative to those naturally occurring in ocean environments, mesocosm experiments using natural communities enable more comprehensive hypothesis testing under natural ocean conditions and will provide insights into the functioning of this ecosystem in the future. The experiment was carried out in the coastal waters of Korea (Jangmok, 34.6° N and 128.5° E) over 19 days from May 2 to 20, 2012. The experiment involved nine enclosures each of approximately 2400 L and included acidification and greenhouse treatments. For the acidification treatment, six of the enclosures were used to simulate a pCO2 range of 160−830 ppmv (average pCO2 values over the experimental period of 160, 230, 420, 610, 700, and 830 ppmv). For the greenhouse treatment, the remaining three enclosures were operated at 2 °C warmer than ambient, at pCO2 concentrations of 290, 690, and 790 ppmv, respectively (see Figure S1 of the Supporting Information). Analysis of DMS, DMSP, and DMSP-lyase Activity (DLA). DMS analysis was conducted using DMS extraction and trapping devices and a gas chromatograph equipped with a flame photometric detector (GC−FPD).21 Gravity filtration was used to filter 50 mL of sample through a 47 mm glass-fiber filter (Whatman GF/F). After filtration, 5−10 mL of filtrate was injected into a sparging chamber for DMS extraction. The sample volume for DMS analysis was adjusted to accommodate a range of DMS concentrations encountered in our experiment. Samples for DMSP analysis were collected and preserved following the widely accepted procedure detailed by Kiene and Slezak.22 To determine the particulate DMSP concentration, we measured the total DMSP (dissolved and particulate forms) and dissolved DMSP; the particulate DMSP concentration was calculated by the difference. A small volume of 50% H2SO4 (5 μL/mL of sample) was added to each sample (∼10 mL) to ensure preservation until DMSP analysis. For each sample, analysis of the dissolved DMSP was initiated by gravity filtering of the sample through a 47 mm glass-fiber filter (Whatman GF/ F) using the small-volume gravity drip filtration procedure. The sample was then preserved by the addition of 50% H2SO4 (5 μL addition/mL of sample) until analyzed. The preserved DMSP sample was hydrolyzed to DMS by the addition of 10 N
gained from controlled in situ experiments using natural communities of phytoplankton in large enclosures equipped to simulate predicted future partial pressure of carbon dioxide (pCO2) concentration and temperature conditions. Several experiments of this type have been performed in various marine environments to investigate the effect of ocean acidification on DMSP and DMS production.11−16 However, differences in the conclusions drawn from those studies have not been adequately explained, and we are unaware of any other previous studies that have reported changes in DMS production associated with changes in major phytoplankton species induced by increased pCO2 and their association with micrograzers. Such sequential interactions of multiple biological processes may be key to marine DMS production under the present and future high pCO2 conditions.7 In this study, we evaluated the community-scale effects of increased levels of pCO2 on marine DMS production at ambient temperature and ∼2 °C warmer conditions (greenhouse) using a controlled mesocosm facility. We considered our results in the context of comparable studies to evaluate the significance of the effects of increased pCO2 on planktonic DMS production and in terms of the strong correlation between observed responses and underlying biological community interactions.
■
MATERIALS AND METHODS Mesocosm Facility and Operational Sequence. The mesocosm facility consists of a floating raft supporting nine impermeable cylindrical enclosures. Each enclosure is 3 m in height, has a volume of approximately 2400 L, and is equipped with a pCO2 regulation unit and a bubble-mediated seawater mixer. Each enclosure was filled two-thirds with seawater and capped with a transparent dome that transmits incoming radiation.17 The target levels for seawater pCO2 in the enclosures were achieved by adding CO2-saturated seawater to the enclosures while filling them with ambient seawater, which was passed through a filter (100 μm pore size). Such use of CO2-saturated seawater is more suitable for large CO2 perturbation facilities equipped with large sample enclosures needing large amounts of CO2. The addition of CO2-saturated seawater into the enclosures increases the dissolved inorganic carbon concentration while leaving the alkalinity unchanged, thereby approximately mimicking the ocean acidification process. In addition, the effects of this CO2 perturbation method on marine organisms have been shown to be less significant than the effect of direct target CO2 aeration.18,19 Dependent upon the target pCO2 level, 2−15 L of CO2-saturated seawater was added to each enclosure. This CO2-gradient approach carries a low risk of damage to the enclosure relative to a replication approach (e.g., each treatment with three replicates) and provides a greater chance of detecting a threshold level for any of the CO2-sensitive processes.18 For 20 min/day, air containing the target CO2 concentration was flowed (approximately 0.5 L min−1) into the seawater mixer to generate a convective flow within the enclosure to ensure homogeneity. This procedure resulted in an even distribution of biogenic particles (particulate organic matter) throughout the enclosures and prevented particles settling to the bottom.17 Immediately following completion of the 20 min seawater aeration period, the enclosures were sampled. The seawater temperature in the greenhouse treatments reached the target value (2 °C warmer than ambient) within 7 h by circulating warm water (∼10 °C warmer than ambient) 4751
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
period were used to calculate the apparent phytoplankton growth rate and microzooplankton grazing. Apparent growth rates were calculated as r = 1/t ln(Ct/C0), where t is the time and C0 and Ct are the initial and final chl a concentrations, respectively. Microzooplankton grazing rates were calculated from the slopes of the regressions of total chl a change during the incubation period as a function of the percentage of dilution.
NaOH (0.25 mL/mL of sample) and allowed to react overnight in the dark. DMS that was subsequently evolved was measured using a GC−FPD. The detection limit was 0.2 nM S (DMS), and the analytical precision was better than 3%. The response of the GC−FPD was calibrated against standard DMS solutions of known concentration (1−100 nmol L−1). Measurements of DLA were carried out following the method described by Steinke et al.23 An aliquot of seawater sample (150−250 mL) was filtered through a 47 mm polycarbonate membrane filter (Millipore; 3 μm pore size). The filter were placed into a Petri dish and stored at −80 °C until analyzed. The filter was transferred into a 10 mL bottle containing 300 mmol L−1 sterile 1,3-bis[tris(hydroxymethyl)methylamino]propane (BTP) buffer solution that was amended with 0.5 mol L−1 NaCl at pH 8.2. Analysis was conducted on 100−200 μL of the extract. The linear production of DMS was quantified at 20 °C for 15−60 min, following the addition of 10 μL of 1 mol L−1 DMSP stock (the final DMSP concentration was 1 mmol L−1, and the final pH was 8.2). Phytoplankton and Microzooplankton Abundance. To count phytoplankton and microzooplankton cell abundance, 500 mL of seawater sample was preserved with Lugol’s iodine solution. The samples were pre-concentrated to 50 mL by sedimentation for at least 48 h and analyzed immediately by a light microscope, together with a Sedgwick-Rafter chamber and a sedimentation chamber. This enabled identification of the plankton groups and calculation of their concentrations. The plankton identification and cell-counting procedure was repeated 2 or 3 times for each sample. For each species, the mean cell concentration was determined by averaging the 2 or 3 separate cell counts. The average uncertainty in cell counting was less than 10% of the total population of each species. To estimate the carbon biomass of microzooplankton, the cell volume was calculated by measuring cell dimensions with an ocular micrometer in the microscope. Microzooplankton assemblages were classified as ciliates and heterotrophic dinoflagellates (HDFs). Conversion factors and equations were used to translate cell volume into carbon biomass, including 0.19 μg of C μm−3 for naked ciliates,24 carbon (pg) = 444.5 + 0.053(lorica volume, μm3) for loricate ciliates,25 and carbon (pg) = 0.216(cell volume, μm3)0.939 for HDFs.26 Incubation Experiments Measuring Grazing Activity. Microzooplankton grazing rates over a 24 h period were estimated every 2−3 days using the dilution method based on chlorophyll a (chl a) concentration data.27 A dilution series was prepared in 1.2 L glass bottles that had previously been cleaned with 10% HCl and thoroughly rinsed with distilled water. Four dilutions (0, 25, 50, and 75%) were prepared in particle-free seawater, which was prepared by gravity filtration through a 0.2 μm cellulose acetate filter, to minimize cell damage. Excess nitrate (15 μmol kg−1) and phosphate (1 μmol kg−1) were added to each dilution bottle to ensure that nutrients would not be a limiting factor in algal growth, because insufficient nutrients would bias the results.27 The experimental bottles containing samples from the nine enclosures were incubated in a chamber receiving a continuous flow of ambient seawater, whereas the bottles containing samples from the greenhouse enclosures were incubated in a chamber receiving a flow of seawater 2 °C higher than ambient. Each experimental bottle was covered with a neutral density screen matching the light intensity at a depth of 1 m. Samples for chl a analysis were removed from the dilution bottles immediately and at 24 h. The chl a concentrations measured during the incubation
■
RESULTS AND DISCUSSION In all enclosures, the phytoplankton biomass (measured as the chl a concentration) reached a maximum level on days 4−6 and decreased thereafter (Figure 1A). Regardless of the pCO2 and
Figure 1. Concentrations of (A) chl a, (B) DMS, and (C) DMSPP during the experimental period. The vertical dashed lines in panels A− C indicate the boundaries of the three periods of increasing (period 1), decreasing (period 2), and no change (period 3) in cell population. The boundaries were arbitrarily chosen on the basis of the trends in chl a concentration. Panels D, E, and F show the total accumulated concentration (integrated over the period of days 1−18) for chl a, DMS, and DMSP, respectively, as a function of the initial pCO2 concentration. The solid and dashed lines in D−F indicate the best fits of the six ambient treatments and the three elevated (2 °C warmer than ambient) temperature treatments, respectively. The colored filled and open symbols indicate the ambient and 2 °C warmer than ambient treatments, respectively.
temperature treatment, the total accumulated chl a concentration (integrated during the study period) was similar (98.3 ± 5.2 μg L−1) among all enclosures (Figure 1D). However, the trend in the DMS concentration differed from that of chl a. In each treatment, the DMS concentration was low (∼1.4 nM) during the bloom period (period 1; days 0−5) and did not differ significantly among treatments. During the post-bloom period (period 2, days 6−13), the DMS concentration increased in each enclosure, reached a maximum level that varied widely among treatments (10−100 nmol L−1) on days 11−15 and decreased thereafter (Figure 1B). The trend in 4752
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
period of increasing DMS concentration (days 4, 7, 9, 11, and 13; Figure 3; r2 = 0.60; p < 0.01; n = 36). These results indicate
particulate DMSP (DMSPP) concentration was broadly consistent with that of DMS, although the maximum levels occurred approximately 2 days earlier (Figure 1C). The total DMS and DMSPP concentrations clearly showed the dependence upon pCO2 (panels E and F of Figure 1). As shown in Figure 1E, the total integrated DMS concentration decreased at a rate of 58 ± 17 nmol L−1 per 100 ppmv pCO2 increase. Relative to the total integrated DMS concentration in the 160 ppmv pCO2 enclosure, the concentration was 82% lower in the highest pCO2 treatment (830 ppmv pCO2), indicating an adverse effect of increased pCO2 on oceanic DMS production. At all three pCO2 conditions, a 2 °C increase in temperature reduced the total integrated DMS concentration compared to that at ambient temperature (Figure 1E). Moreover, the DMS concentration decreased at a rate of 24 ± 24 nmol L−1 per 100 ppmv increase in pCO2. The difference in the integrated DMS concentration between ambient temperature and 2 °C warmer appeared to decrease as pCO2 levels increased (Figure 1E); however, because measurements were made at only two temperatures, the effect of pCO2 on DMS production as a function of the temperature remains unclear. Grazing activity is the most likely explanation for the difference in net DMS production among the pCO2 treatment enclosures in our experiment. In all enclosures, the overall grazing activity was largely governed by the HDF Protoperidinium spp. (cell volume of ∼28 000 μm3), followed by ciliates. The production of DMS was highest during days 7, 9, 11, and 13 of period 2, and the measured grazing rates were inversely correlated with the pCO2 levels (Figure 2A; r2 = 0.70; p =
Figure 3. Correlation between DMS concentrations and corresponding grazing rates measured at days 4, 7, 9, 11, and 13. The solid line indicates the best fit.
that grazing activity was the dominant factor in DMS production in our experiment, regardless of the pCO2 level. The grazer ciliate appeared to account for a large proportion of the microzooplankton biomass at high pCO2, but the abundance of ciliate showed no correlation with pCO2 at levels lower than 800 ppmv (see Figure S4 of the Supporting Information). Not all species have the same ability to form DMSP; for example, diatoms generally produce only small amounts of DMSP, whereas prymnesiophytes and dinoflagellates produce considerably greater amounts.1,7 For grazing activity to be a major cause of the difference in net DMS production among the enclosures, DMSP-containing prey species must be involved. In our experiment, the large diatom Cerataulina pelagica (cell volume of ∼24 100 μm3), which was the most abundant species during period 1, decreased in abundance during the transition from period 1 to period 2. This coincided with a shift to the phototrophic dinoflagellate Alexandrium spp. (cell volume of ∼2200 μm3) as the dominant species; the population of HDFs rapidly increased immediately following the population peaks of Alexandrium spp. (Figure 4). In contrast to other prey groups (dinoflagellates, autotrophic picoplankton, nanoflagellates, and diatoms), only Alexandrium spp. (high cellular DMSP content and DLA) showed a negative growth response to increasing pCO2 levels (Figure 5A; r2 = 0.80; p = 0.02), whereas C. pelagica (low cellular DMSP content and no DLA) showed a positive response to increasing pCO2 levels (Figure 5B; r2 = 0.72; p = 0.03). Biomarker pigment data showing changes in phytoplankton species composition are also consistent with this finding (Figure 6). In all enclosures, the diatom-specific pigment fucoxanthin was dominant on day 0, accounting for 90.5 ± 1.0% of the phytoplankton. However, during the DMS producing period (day 12), the combination of fucoxanthin and peridinin, a dinoflagellate-specific pigment, accounted for 79.9 ± 2.9% of the six major pigments found, but the proportions of fucoxanthin and peridinin showed opposite trends in response to pCO2 levels (Figure 6B). These results further indicate that the major dinoflagellate species found in our experiments negatively responded to increasing pCO2, whereas the major diatom species responded positively to increasing pCO2. Some studies have also shown that elevated pCO2 enhanced the growth rate of diatoms,16,28−30 which may be due to energy savings caused by the downregulation of the CO2 concentrating mechanism.30 The cellular DMSP content and DLA of Alexandrium spp. range from 180 to 290 mmol
Figure 2. (A) Mean grazing rate during the DMS production period (days 7, 9, 11, and 13) as a function of pCO2. The error bars indicate 1 standard deviation (1σ) from the 4 day mean values. Daily grazing rates are presented in Figure S3 of the Supporting Information. (B) Total accumulated populations of HDFs during the experimental period (days 1−18) as a function of pCO2. The solid and dashed lines indicate the best fits of the six ambient and three elevated (2 °C warmer than ambient) temperature treatments, respectively.
0.04). Results obtained from all 42 incubation experiments are presented in Table S1 and Figures S2 and S3 of the Supporting Information. The populations of HDF were also inversely correlated with pCO2 levels (Figure 2B), thereby yielding a significant correlation between the variation in the DMS concentration and the change in the grazing rate during the 4753
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
Figure 6. Relative composition of six major biomarker pigments of phytoplankton on (A) day 0 and (B) day 12. Peridinin (Per), 19′hexanoyloxyfucoxanthin (19-hex), alloxanthin (Allo), chlorophyll b (Chl-b), zeaxanthin (Zea), and fucoxanthin (Fuc) are markers for dinoflagellates, prymnesiophytes, cryptophytes, green flagellates, cyanobacteria, and diatoms, respectively.
Figure 4. Cell abundance of the three key plankton species in response to pCO2 change during the experimental period. The colored lines indicate C. pelagica (red), Alexandrium spp. (blue), and HDFs (green). The vertical dashed lines show the boundaries of the three periods, as defined in Figure 1
The ratio of the DMSPP concentration to total biomass [expressed as particulate organic carbon (POC)] provided another compelling line of evidence supporting our explanation. During period 1 (days 1 and 2), the ratio of DMSPP/ POC remained approximately unchanged over the pCO2 range investigated (Figure 5D) but decreased with increasing pCO2 levels during period 2 (days 5 and 6) (Figure 5E). The rate of decrease in the ratio became more pronounced as the experiment proceeded (from Figure 5E to Figure 5F). The results of the measurements of DLA were also consistent with our explanation. The DLA decreased with increasing pCO2 levels during 3 days of period 2 (days 6, 8, and 10; Figure 5C). In particular, the DLA was highly correlated with the change in populations of dinoflagellate during this period. Hence, the mixing of planktonic DMSP and DMSP-lyase during grazing (resulting in DMS production) contributed considerably to DMS production during period 2. The higher values of DMSPP/POC (greater proportions of DMSP-containing phytoplankton) and higher DLA in the low pCO2 treatments, in conjunction with the higher grazing rates in those treatments, indicate that the higher abundance of DMSP-rich prey under the low pCO2 treatment conditions may lead to higher HDF grazing activity compared to that observed under high pCO2 conditions. This pCO2 dependence of grazing activity eventually resulted in greater DMS production in the former treatments than in the latter treatments. Our results indicate a direct link between reduced DMS production under high pCO2 conditions and the interactions of phytoplankton species and their association with grazers. A previous mesocosm study conducted at the same site but in a different season16 (December for the 2008 experiment versus April for the present study) also found that the growth of a major diatom, Skeletonema costatum, was enhanced at high pCO2 and that grazing activity associated with this species had an impact on DMS production. Although the major diatom groups found in these studies responded positively to increased pCO2, the total DMS production in these studies had opposite signs and was critically dependent upon the interactions of the plankton community (i.e., competition between non-DMSPcontaining and other DMSP-containing species and their associations with potential predators). In particular, the
Figure 5. Total cell abundance of (A) Alexandrium spp. and (B) C. pelagica integrated over the experimental period (days 1−18) as a function of pCO2. (C) Mean DLA measured during 3 days (6, 8, and 10) of the DMS production period as a function of pCO2. Daily DLA is shown in Figure S5 of the Supporting Information. (D−F) Mean ratios of DMSPP/POC during 2 days as a function of pCO2: (D) days 1 and 2, (E) days 5 and 6, and (F) days 11 and 12. The error bars indicate 1 standard deviation (1σ) from the (C) 3 day means and (D− F) 2 day means. The solid and dashed lines indicate the best fits of the six ambient treatments and three elevated (2 °C warmer than ambient) temperature treatments, respectively.
Lcell−1 and from 230 to 1620 mmol Lcell−1 h−1, respectively,31,32 and Alexandrium spp. accounted for more than 60% of the total dinoflagellate population in our experiment, except for pCO2 treatments of >700 ppmv, where its contribution decreased to 30%. In all enclosures, Alexandrium spp. were largely fed upon by Protoperidinium spp. (Figure 4). 4754
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
more importantly, their association with potential micrograzers. In the context of increasing global ocean acidification, the key implication of our results is that increased seawater pCO2 could change the species composition of the phytoplankton community and possibly change the dominant phytoplankton species. If so, DMS production in the ocean will be critically dependent upon whether the dominant species contains DMSP and is a suitable prey for marine micrograzers, because this will determine the grazing activity and, thereby, grazing-mediated DMS production.
conversion of DMSP to DMS was considerably accelerated in the 2012 experiment compared to the 2008 experiment, resulting in higher DMS production in 2012. In the 2008 experiment, the high pCO2 conditions increased the growth rate of DMSP-containing small diatoms (i.e., S. costatum; cell volume of ∼70 μm3; DMSP = 0.2−0.4 mmol Lcell−1), which led to an increase in the rate of grazing by microzooplankton, and, thereby, produced higher levels of DMS. In both experiments, the mean grazing rates were similar during the DMS production period (0.73 ± 0.13 day−1 for 2008 versus 0.78 ± 0.27 day−1 for 2012), but the DMS/DMSP ratio (0.03 ± 0.01) observed in the 2008 experiment was only 25% of the level found in the 2012 experiment (0.11 ± 0.2). The low DMS/DMSP ratio indicates a lower level of conversion of cellular DMSP to DMS. This suggests that some DMSP released into seawater during micrograzer consumption of S. costatum (which contains DMSP but lacks DLA) may be converted into DMS through bacterial DLA in the seawater, while most of the cellular DMSP is degraded to other sulfur compounds through a bacterial demethylation process. In contrast to the 2008 experiment, in the 2012 experiment, the micrograzers largely fed on Alexandrium spp., which contain both high DMSP concentration and DLA. This grazing activity resulted in more complete mixing of cellular DMSP and DMSP-lyase and, thereby, led to greater conversion of DMSP to DMS (indicating more DMS production). This enhancement of DMS production during grazing depends upon the cellular DMSP content and the cellular DLA of the prey. This highlights the complexity of oceanic DMS production in response to elevated pCO2 levels. The effects of increased pCO2 on marine DMS production have been examined in several studies.11−16,33 With the exception of the work by Kim et al.,16 the results of these studies have been largely consistent and have generally pointed to a reduction in DMS production. In several mesocosm studies carried out in Norwegian coastal waters, the reduction in the prymnesiophyte Emiliania huxleyi in response to increasing pCO2 was found to be responsible for decreases in DMS and DMSP production.11,15 A similar conclusion was also drawn from laboratory experiments with E. huxleyi.33 In marine environments, microzooplankton grazing is the main cause of phytoplankton mortality34 and can either directly or indirectly contribute to DMS production. Therefore, the role of grazing in marine DMS production has been assessed under a wide range of experimental conditions.7,35,36 Grazingmediated DMS production per cell has been reported to be an order of magnitude greater than the combination of direct algal release and cell death by viral infection.9,37,38 Although not yet widely confirmed, these results indicate that DMS production associated with grazing activity may be a key mechanism of DMS production in present and future marine environments. In particular, our findings may be important in relation to DMS production in productive areas, including highlatitude oceans, coastal waters, and upwelling regions, where diatom and other DMSP-producing phytoplankton (i.e., dinoflagellates and prymnesiophytes) blooms frequently occur together or in sequence over time.39 Analysis of experiments conducted at diverse scales indicates that marine DMS production is affected by the interactions of multiple factors acting simultaneously and sequentially over time. Nonetheless, we conclude that the oceanic DMS production under high pCO2 conditions primarily depends upon the growth of DMSP-rich phytoplankton species and,
■
ASSOCIATED CONTENT
S Supporting Information *
Pigments analysis, statistical analysis, grazing rates obtained from the dilution experiments carried out at days 4, 7, 9, 11, and 13 (Table S1), (A) seawater pCO2, (B) pH, and (C) temperature during the experimental period (Figure S1), apparent growth rates determined by chl a changes during the dilution incubation as a function of the fraction whole water in nine mesocosm enclosures at days 7, 9, 11, and 13 (Figure S2), grazing rate at days (A) 4, (B) 7, (C) 9, (D) 11, and (E) 13 as a function of pCO2 (Figure S3), total accumulated populations of ciliates during the experimental period (days 1− 18) as a function of pCO2 (Figure S4), and DLA measured at days (A) 6, (B) 8, and (C) 10 as a function of pCO2 (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Telephone: 82-54-279-2285. Fax: 82-54-279-8299. E-mail: [email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the Midcareer Researcher Program (2012R1A2A1A01004631) and the Global Research Project funded by the National Research Foundation (NRF) of the Ministry of Science, Information and Communication Technology (ICT), and Future Planning (MSIP). Partial support was provided by “Long-Term Change of Structure and Function in Marine Ecosystems of Korea” and “Management of Marine Organisms Causing Ecological Disturbance and Harmful Effects” funded by the Ministry of Oceans and Fisheries.
■
REFERENCES
(1) Liss, P. S.; Hatton, A. D.; Malin, G.; Nightingale, P. D.; Turner, S. M. Marine sulphur emissions. Philos. Trans. R. Soc. London, Ser. B 1997, 352, 159−169. (2) Charlson, R. J.; Lovelock, J. E.; Andreae, M. O.; Warren, S. G. Oceanic phytoplankton, atmospheric sulphur, cloud albedo, and climate. Nature 1987, 326, 655−661. (3) Quinn, P. K.; Bates, T. S. The case against climate regulation via oceanic phytoplankton sulphur emissions. Nature 2011, 480, 51−56. (4) Lana, A.; Bell, T. G.; Simó, R.; Vallina, S. M.; Ballabrera-Poy, J.; Kettle, A. J.; Dachs, J.; Bopp, L.; Salzman, E. S.; Stefels, J.; Johnson, J. E.; Liss, P. S. An updated climatology of surface dimethlysulfide concentrations and emission fluxes in the global ocean. Global Biogeochem. Cycles 2011, DOI: 10.1029/2010GB003850. (5) Chen, T.; Jang, M. Secondary organic aerosol formation from photooxidation of a mixture of dimethyl sulfide and isoprene. Atmos. Environ. 2012, 46, 271−278.
4755
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
Environmental Science & Technology
Article
space analysis of dimethylsulphide (DMS). J. Sea Res. 2000, 43, 233− 244. (24) Putt, M.; Stoecker, D. K. An experimentally determined carbon: volume ratio for marine “oligotrichous” ciliates from estuarine and coastal waters. Limnol. Oceanogr. 1989, 34 (6), 1097−1103. (25) Verity, P. G.; Langdon, C. Relationships between lorica volume, carbon, nitrogen and ATP content of tintinnids in Narragansett Bay. J. Plankton Res. 1984, 6 (5), 859−868. (26) Menden-Deuer, S.; Lessard, E. J. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnol. Oceanogr. 2000, 45 (3), 569−579. (27) Landry, M. R.; Hassett, R. P. Estimating the grazing impact of marine micro-zooplankton. Mar. Biol. 1982, 67, 283−288. (28) Kim, J.-M.; Lee, K.; Shin, K.; Kang, J.-H.; Lee, H.-W.; Kim, M.; Jang, P.-G.; Jang, M.-C. The effect of seawater CO2 concentration on growth of a natural phytoplankton assemblage in a controlled mesocosm experiment. Limnol. Oceanogr. 2006, 51 (2), 1629−1636. (29) Wu, Y.; Gao, K.; Riebesell, U. CO2-induced seawater acidification affects physiological performance of the marine diatom Phaeodactylum tricornutum. Biogeosciences 2010, 7, 2915−2923. (30) Tortell, P. D.; Payne, C. D.; Li, Y.; Trimborn, S.; Rost, B.; Smith, W. O.; Riesselman, C.; Dunber, R. B.; Sedwick, P.; DiTullio, G. R. CO2 sensitivity of southern ocean phytoplankton. Geophys. Res. Lett. 2008, 35 (4), L04605. (31) Caruana, A. DMS and DMSP production by marine dinoflagetllates. Ph.D. Dissertation, University of East Anglia, Norwich, U.K., 2010. (32) Wolfe, G. V.; Strom, S. L.; Holmes, J. L.; Radzio, T.; Olson, M. B. Dimethylsulfoniopropionate cleavage by marine phytoplankton in response to mechanical, chemical, or dark stress. J. Phycol. 2002, 38 (5), 948−960. (33) Arnold, H. E.; Kerrison, P.; Steinke, M. Interacting effects of ocean acidification and warming on growth and DMS-production in the haptophyte coccolithophore Emiliania huxleyi. Global Change Biol. 2013, 19 (4), 1007−1016. (34) Calbet, A.; Landry, M. R. Phytoplankton growth, microzooplankton grazing, and carbon cycling in marine systems. Limnol. Oceanogr. 2004, 49 (1), 51−57. (35) Archer, S. D.; Widdicombe, C. E.; Tarran, G. A.; Rees, A. P.; Burkill, P. H. Production and turnover of particulate dimethylsulphoniopropionate during a coccolithophore bloom in the northern North Sea. Aquat. Microb. Ecol. 2001, 24 (3), 225−241. (36) Wolfe, G. V.; Steinke, M. Grazing-activated production of dimethylsulfide (DMS) by two clones of Emiliania huxleyi. Limnol. Oceanogr. 1996, 41 (6), 1151−1160. (37) Lee, H.; Park, K.-T.; Lee, K.; Jeong, H. J.; Yoo, Y. D. Preydependent retention of dimethylsulfoniopropionate (DMSP) by mixotrophic dinoflagellates. Environ. Microb. 2012, 14 (3), 605−616. (38) Evans, C.; Kandner, S. V.; Darroch, L. J.; Wilson, W. H.; Liss, P. S.; Malin, G. The relative significance of viral lysis and microzooplankton grazing as pathways of dimethylsulfoniopropionate (DMSP) cleavage: An Emiliania huxleyi culture study. Limnol. Oceanogr. 2007, 52 (3), 1036−1045. (39) Degerlund, M.; Eilertsen, H. C. Main species characteristics of phytoplankton spring blooms in NE Atlantic and Arctic waters (68− 80° N). Estuaries Coasts 2010, 33, 242−269.
(6) Barnes, I.; Hjorth, J.; Mihalopoulos, N. Dimethyl sulfide and dimethyl sulfoxide and their oxidation in the atmosphere. Chem. Rev. 2006, 106 (3), 940−975. (7) Stefels, J.; Steinke, M.; Turner, S.; Malin, G.; Belviso, S. Environmental constraints on the production and removal of the climatically active gas dimethylsulphide (DMS) and implications for ecosystem modelling. Biogeochemistry 2007, 83, 245−275. (8) Zubkov, M. V.; Fuchs, B. M.; Archer, S. D.; Kiene, R. P.; Amann, R.; Burkill, P. A. Linking the composition of bacterioplankton to rapid turnover of dissolved dimethylsulfoniopropionate in an algal bloom in the North Sea. Environ. Microb. 2001, 3 (5), 304−311. (9) Dacey, J. W. H.; Wakeham, S. G. Oceanic dimethylsulfide: Production during zooplankton grazing on phytoplankton. Science 1986, 233, 1314−1316. (10) Six, K. D.; Kloster, S.; Ilyina, T.; Archer, S. D.; Zhang, K.; MaierReimer, E. Global warming amplified by reduced sulphur fluxes as a result of ocean acidification. Nat. Clim. Change 2013, 3, 975−978. (11) Hopkins, F. E.; Turner, S. M.; Nightingale, P. D.; Steinke, M.; Bakker, D.; Liss, P. S. Ocean acidification and marine trace gas emissions. Proc. Natl. Acad. Sci. U. S. A. 2010, 107 (2), 760−765. (12) Vogt, M.; Steinke, M.; Turner, S.; Paulino, A.; Meyerhöfer, M.; Riebesell, U.; LeQuéré, C.; Liss, P. S. Dynamics of dimethylsulphoniopropionate and dimethylsulphide under different CO2 concentrations during a mesocosm experiment. Biogeosciences 2008, 5, 407− 419. (13) Wingenter, O. W.; Haase, K. B.; Zeigler, M.; Blake, D. R.; Rowland, F. S.; Sive, B. C.; Paulino, A.; Thyrhaug, R.; Larsen, A.; Schulz, K.; Meyerhöfer, M.; Riebesell, U. Unexpected consequences of increasing CO2 and ocean acidity on marine production of DMS and CH2ClI: Potential climate impacts. Geophys. Res. Lett. 2007, 34, L05710. (14) Archer, S. D.; Kimmance, S. A.; Stephens, J. A.; Hopkins, F. E.; Bellerby, R. G.; Schulz, K. G.; Piontek, J.; Engel, A. Contrasting responses of DMS and DMSP to ocean acidification in Arctic waters. Biogeosciences 2013, 10, 1893−1908. (15) Avgoustidi, V.; Nightingale, P. D.; Joint, I.; Steinke, M.; Turner, S. M.; Hopkins, F. E.; Liss, P. S. Decreased marine dimethyl sulphide production under elevated CO2. Environ. Chem. 2012, 9, 399−404. (16) Kim, J.-M.; Lee, K.; Yang, E. J.; Shin, K.; Noh, J. H.; Park, K.-T.; Hyun, B.; Jeong, H.-J.; Kim, J.-H.; Kim, K. Y.; Kim, M.; Kim, H.-C.; Jang, P.-G.; Jang, M.-C. Enhanced production of oceanic dimethylsulfide resulting from CO2-induced grazing activity in a high CO2 world. Environ. Sci. Technol. 2010, 44 (21), 8140−8143. (17) Kim, J.-M.; Shin, K.; Lee, K.; Park, B.-K. In situ ecosystem-based carbon dioxide perturbation experiments: Design and performance evaluation of a mesocosm facility. Limnol. Oceanogr.: Methods 2008, 6, 208−217. (18) Riebesell, U.; Czerny, J.; von Bröckel, K.; Boxhammer, T.; Büdenbender, J.; Deckelnick, M.; Fischer, M.; Hoffmann, D.; Krug, S. A.; Lentz, U.; Ludwig, A.; Muche, R.; Schulz, K. G. Technical note: A mobile sea-going mesocosm system: New opportunities for ocean change research. Biogeosciences 2013, 10, 1835−1874. (19) Schulz, K. G.; Barcelos e Ramos, J.; Zeebe, R. E.; Riebesell, U. CO2 perturbation experiments: Similarities and differences between dissolved inorganic carbon and total alkalinity manipulations. Biogeosciences 2009, 6, 2145−2153. (20) Intergovernmental Panel on Climatic Change (IPCC). IPCC Fourth Assessment Report: Climate Change 2007. The Physical Science Basis; Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L., Eds.; Cambridge University Press: New York, 2007. (21) Park, K.-T.; Lee, K. High-frequency, accurate measurement of dimethylsulfide in surface marine environments using a microporous membrane contactor. Limnol. Oceanogr.: Methods 2008, 6, 548−557. (22) Kiene, R. P.; Slezak, D. Low dissolved DMSP concentrations in seawater revealed by small-volume gravity filtration and dialysis sampling. Limnol. Oceanogr.: Methods 2006, 4, 80−95. (23) Steinke, M.; Malin, G.; Turner, S. M.; Liss, P. S. Determinations of dimethylsulphoniopropionate (DMSP) lyase activity using head4756
dx.doi.org/10.1021/es403351h | Environ. Sci. Technol. 2014, 48, 4750−4756
