Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Oxic water column methanogenesis as a major component of aquatic CH4 fluxes

Oxic water column methanogenesis as a major component of aquatic CH4 fluxes ARTICLE Received 29 May 2014 | Accepted 22 Sep 2014 | Published 30 Oct 2014 DOI: 10.1038/ncomms6350 Oxic water column methanogenesis as a major component of aquatic CH fluxes 1 1 1 1 1 Matthew J. Bogard , Paul A. del Giorgio , Lennie Boutet , Maria Carolina Garcia Chaves , Yves T. Prairie , 1 1 Anthony Merante & Alison M. Derry Methanogenesis has traditionally been assumed to occur only in anoxic environments, yet there is mounting, albeit indirect, evidence of methane (CH ) production in oxic marine and freshwaters. Here we present the first direct, ecosystem-scale demonstration of methanogenesis in oxic lake waters. This methanogenesis appears to be driven by acetoclastic production, and is closely linked to algal dynamics. We show that oxic water methanogenesis is a significant component of the overall CH budget in a small, shallow lake, and provide evidence that this pathway may be the main CH source in large, deep lakes and open oceans. Our results challenge the current global understanding of aquatic CH dynamics, and suggest a hitherto unestablished link between pelagic CH emissions and surface-water primary production. This link may be particularly sensitive to widespread and increasing human influences on aquatic ecosystem primary productivity. Groupe de Recherche Interuniversitaire en Limnologie, De´partement des Sciences Biologiques, Universite´ du Que´bec a` Montre´al, Case Postale 8888, Succursale Centre-Ville, Montre´al, Quebec, Canada H3C 3P8. Correspondence and requests for materials should be addressed to M.J.B. (email: [email protected]). NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 1 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ethane (CH ) emissions currently contribute B20% The observed surface CH concentrations in the mesocosms 4 4 to the planet’s greenhouse effect, with a large, but represent the net balance between CH production, CH 4 4 Mpoorly defined, fraction derived from freshwater oxidation (MOX) and CH exchange with the atmosphere. In 1–4 ecosystems . Accurately placing freshwaters in the global CH the absence of internal inputs of CH , the initial CH super- 4 4 4 budget requires a better understanding of the controls and saturation in the enclosures would have declined to atmospheric 2–5 contributions of CH from different sources . In this regard, the equilibrium within approximately 1 week (Fig. 1a, dotted lines), surface waters of lakes and rivers are systematically as estimated on the basis of our own empirical measurements of supersaturated with CH , and it has been traditionally assumed the gas exchange coefficient (see Methods section). Thus, an that this CH is derived from anoxic environments, via vertical internal CH source was necessary to sustain the systematic CH 4 4 4 3,6–8 and lateral transport from profundal and littoral sediments . supersaturation observed in all mesocosms throughout the This assumption certainly holds for small, shallow ecosystems experiment. Further, there was an overall significant increase in where the surface layers are in relatively close contact with CH concentrations through time across treatments (Fig. 1a; sediments. Yet CH supersaturation is also prevalent in large, repeated measures analysis of variance (RM-ANOVA) time effect: 3,5,7–12 13–17 deep lakes and oceanic surface waters , where deeper P ¼ 0.007), with greater increases in nutrient-amended enclosures water columns result in significantly reduced surface water (nitrogen and phosphorus (NP), dissolved organic carbon exchange with anoxic sediments. Studies of CH dynamics in (DOC)-NP), and weaker increases in the DOC and control surface waters of oceans and large lakes have indeed concluded enclosures (RM-ANOVA treatment  time effect: P ¼ 0.04). As a that pelagic CH supersaturation cannot be sustained either result, all enclosures emitted CH to the atmosphere for the 4 4 by lateral inputs from the littoral or from benthic inputs duration of the experiment, and the estimated water/air CH 9,10,13–15,17–19 alone . fluxes (based on our measurements of gas transfer coefficients in A pelagic source of CH would thus be required to sustain the the mesocosms) ranged from 0.07 to 0.36 mmol m per day observed CH supersaturation, and multiple lines of evidence (Fig. 1b). Net CH production in the enclosures, calculated as the 4 4 suggest that CH is produced in oxic water columns of freshwater sum of the observed net increase in concentration and the and marine systems. In vitro experiments point to CH calculated CH fluxes to the atmosphere, were on average highest 4 4 production in both microanoxic habitats in metazoan guts and in the nutrient-enriched enclosures, whereas the DOC-only 14,17–23 15,16,18,19,23–25 on particles , and in particle-free oxic water , addition generated the lowest production rates (Fig. 1b; analysis 19,22,26–28 via multiple biochemical pathways. The presence of variance (ANOVA): P ¼ 0.003). We assessed the robustness of 19,22,27 and activity of hydrogenotrophic and acetoclastic chamber-based calculations of CH emissions by comparing methanogenic archaea have been confirmed at the molecular chamber-derived gas exchange coefficients (k ) to estimates CH4 level in diverse oxic environments. At the same time, marine based on wind speed (detailed in Supplementary Note 1). studies suggest that the microbial decomposition of methylated There was good agreement between the two approaches, and 15 16,23–25 compounds such as methanethiol and methylphosphonate therefore we conclude that chamber-derived results yield reliable could be an important source of CH in surface waters of the open estimates of k and CH fluxes for both the enclosures and 4 CH4 4 ocean. Further, metalimnetic peaks of CH are a recurrent feature the lake. in many ecosystems, which correlate positively with dissolved O 8–10,14,15,18,19 (DO), algal biomass and production .However, despite past efforts, the ecological and biogeochemical significance of oxic Linking CH production to algal dynamics. The differences water column methanogenesis remain speculative, and we have among treatments in water column CH concentrations and yet to determine the dominant biochemical pathway, its controls, fluxes were strongly linked to pelagic gross primary production and its contribution to diffusive CH emissions from aquatic (GPP; Fig. 2a) and net ecosystem production (NEP ¼ GPP-R; ecosystems. Fig. 2b). There was a 20-fold range in GPP across treatments In this study, we experimentally assess pelagic CH production (Fig. 2a), whereas there was only a 5-fold range in ecosystem in the oxygenated water column of Lac Cromwell, a typical respiration (R), therefore nutrient additions resulted in a strong Canadian temperate Shield lake, using floating mesocosms open shift in NEP (Fig. 2b). The average CH fluxes at day 7 of the to the atmosphere, but closed at the bottom and therefore experiment were strongly positively related to average GPP rates uncoupled from non-pelagic sources of CH . The results in the enclosures (Fig. 2a), and to NEP, whereas the DOC addi- presented herein represent the first direct, ecosystem-scale tion had no positive influence relative to the control. Increases in demonstration of oxic water CH production, its drivers and CH flux appear to have been driven mostly by increases in GPP 4 4 source pathway, and the potential widespread importance of this rather than heterotrophic metabolism. poorly considered process. Other potential sources of CH in the enclosures were also considered. The enclosures contained very little particulate organic C because of the initial filtration of the surrounding lake Results water before filling, and there was little or no accumulation of CH dynamics in experimental enclosures. The initial CH particulate organic C in the bottom of enclosures at the end of the 4 4 concentrations in the enclosures were four- to tenfold lower than experiment. Moreover, we estimate that zooplankton-derived surrounding lake waters, because of degassing during the initial CH , calculated from the measured zooplankton biomass in the filtration and filling process, but the enclosures were nevertheless enclosures and published production rates (Supplementary supersaturated relative to the atmosphere at the onset of the Table 1, Supplementary Note 2), contributed o10% to the experiment (Fig. 1a). Vertical profiles of the enclosures estimated gross CH production rates in ambient enclosures, a 10,18,19,29 (Supplementary Fig. 1) showed a range in DO between 45.6 and result that is consistent with previous studies . Membrane 128.6% saturation, indicating that the entire water column permeability and the resulting exchange with the surrounding remained oxic in all treatments for the duration of the experi- lake waters were also considered as a potential cause for CH ment. Despite oxic conditions, CH concentrations ranged supersaturation. However, the estimated contributions of CH 4 4 between 0.10 and 0.53mM throughout the entire experiment, from the surrounding lake environment were on the order of across all treatments (Fig. 1a), which represents B50- to 265-fold o1% of gross CH production (Supplementary Methods and super-saturation relative to the atmosphere. Supplementary Table 2). Finally, there was no significant 2 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. –2 –1 Evasion (μmol CH m d ) –1 Net production (μM CH d ) NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE 0.6 0.25 400 DOC 0.20 DOC-NP 0.4 NP 0.15 0.10 0.2 0.05 CH at atmospheric equilibrium 0.0 0.00 01724 1 28 Day of experiment Treatment Figure 1 | Surface water CH dynamics in experimental enclosures. Treatments include ambient untreated controls (C), addition of dissolved organic carbon (DOC), nitrogen and phosphorus (NP) or a combination of both (DOC-NP). (a)CH concentrations (error bars 1 s.e.m., n ¼ 3) generally increased through time, but differed between treatments (RM-ANOVA, treatment: P ¼ 0.004, F ¼ 10.71; time: P ¼ 0.007, F ¼ 7.67; treatment* time: P ¼ 0.04, F ¼ 3.04; Tukey’s HSD post-hoc grouping ¼ NP, DOC-NP, C4C, DOC). Dashed lines show the expected declines in dissolved CH because of diffusion at the air–water interface, had no methane been produced in situ. Without internal production of CH , concentrations would have equilibrated with the atmosphere (mean concentration at equilibrium ¼ 0.002mM) in all treatments within 1–2 weeks. (b) Collectively, net CH production (evasion plus water column accumulation per enclosure) and evasion rates (second y axis) displayed large differences among treatments (ANOVA, P ¼ 0.003; post-hoc groups ¼ NP & DOC-NP & C4C4C & DOC). Thick and thin horizontal lines are mean and median, respectively. Identifying the biochemical source of CH . Several methanogenic pathways exist, but as acetoclastic methanogenesis is less isotopically discriminatory than CO reduction or 19,27,28 methylotrophy , apparent fractionation factors (a see app; Methods for calculation details) can be used to qualitatively distinguish whether CH is produced via methanogenesis from acetate, or CO reduction . Our measured a values were well 2 app within the range indicating dominance of the acetoclastic pathway (Fig. 3a), suggesting that acetate was the dominant source material for pelagic methanogenesis in all treatments. As we only had one estimate of water column MOX per 0  3  1 treatment (range ¼ 0.003–0.008 mmol m h ), we used a 0 4 8 12 scenario analysis with two extreme MOX rates (0 and –1 GPP (μM O h ) 3  1 0.05 mmol m h ; detailed in Methods section) to conservatively constrain the possible a values based on 300 app variable MOX rates and associated isotopic fractionation. In all cases, a values remained well within the acetoclastic range, that app is, o1.055 (ref. 30; Fig. 3a). Acetoclastic methanogenesis was 13 13 likely dominant through time, because d C-CO and d C-CH 2 4 showed little temporal variability (Fig. 3b). Rates of oxic water-column methanogenesis. The CH accu- mulation and outgassing that we measured in the enclosures (Fig. 1) represent the net result of pelagic methanogenesis, gas exchange and MOX. In order to derive a first-order estimate of –2 0 2 4 6 8 pelagic methanogenesis, we carried out a detailed CH mass –1 NEP (μM O h ) 2 balance (described in the Methods section; summarized in Supplementary Table 2) for day 7 for the control enclosures. Here Figure 2 | Linking pelagic ecosystem metabolism and CH dynamics. we combined the measured CH evasion, the change in storage Metabolic estimates from day 7 of the experiment revealed CH evasion corrected for cross-membrane inputs from surrounding lake was strongly positively correlated to (a) ecosystem gross primary water and measured MOX. This exercise yielded rates of apparent 2  1.6x production (GPP; P ¼ 0.003; r ¼ 0.93; y ¼ 215.3[1  e ]), and pelagic methanogenesis on the order of 0.21–0.24 mmol m per (b) shifts to increased net ecosystem production (NEP), which is different day (Table 1), of which B60% was apparently oxidized and the between GPP and community respiration (P ¼ 0.02; r ¼ 0.99; y ¼ 111– remainder evaded to the atmosphere. 2.6x 106[1  e ]). Symbols with error bars ( 1 s.e.m., n ¼ 3) denote the different treatments, including ambient untreated controls (red circles), addition of dissolved organic carbon (blue squares), nitrogen and Oxic water methanogenesis in a whole-lake perspective. Mean phosphorus (grey triangles) or a combination of both (yellow diamonds). pelagic CH evasion from control (that is, non-manipulated) enclosures (0.15 0.03 mmol m per day; Table 1) represented development of biofilm on the walls of the enclosures during B20% of the average diffusive CH fluxes measured in the the experiment, and this is unlikely to have been a significant ± lake over the summer (0.78 0.35 mmol m per day), and at source of CH . the whole-lake scale, were similar to CH ebullitive fluxes NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 3 & 2014 Macmillan Publishers Limited. All rights reserved. DOC DOC- NP NP –2 –1 –2 –1 CH evasion (μmol m d)CH evasion (μmol m d ) 4 4 Dissolved CH (μM) 4 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ± ± (15.2 32.5 mol per day or 0.31 0.66 mmol m per day in gas emissions (CO þ CH , expressed as CO equivalents; 2 4 2 lake regions o3 m deep; Table 1). Further, oxic pelagic metha- Table 1). It was possible to compare fluxes directly between the nogenesis contributed on the order of 4% to total lake greenhouse enclosures and the lake because the physical conditions shaping gas exchange dynamics were quantitatively similar for each. As shown in Supplementary Fig. 2, floating-chamber-derived gas 1.08 H /CO or DOM-CH exchange coefficients (k ) for the lake and the enclosures varied 2 2 3 CH4 dominance on a diurnal basis, but averaged 0.65 and 0.69 m per day and 1.06 ranged between 0.10–1.91 and 0.24–1.44 m per day for the Acetate-dominance enclosures and the lake, respectively. 1.04 Discussion Here we present an ecosystem-scale, experimental demonstration 1.02 of significant CH production and evasion from the oxic water column of Lac Cromwell that is closely linked to algal dynamics, and which contributes a baseline flux of CH that is likely present 1.00 in all lakes. The relationship between CH and phytoplankton C DOC DOC- NP 4 NP observed here (Fig. 2) has been hypothesized before to explain Treatment both the presence of metabolically active methanogens, and the recurrent metalimnetic and near-surface CH peaks in oxic 9,10,18,19 13–17 lake and marine environments. We confirm this link α = 1.06 1.04 1.02 1.00 app experimentally, and further show that it generates a significant –5 out flux of CH from the mesocosms to the atmosphere (Fig. 1b). These results in turn imply that factors influencing –10 phytoplankton standing stock and GPP, such as grazing, nutrient availability and the physical structure of the water –15 column, will have a strong bearing on pelagic CH dynamics and –20 resulting CH emissions. To our knowledge, our results represent the first ecosystem- –25 level estimates of methanogenesis in oxic freshwaters. For all treatments, intense CH production was needed to sustain the –30 observed patterns in CH concentrations and to offset atmo- –90 –70 –50 –30 –10 4 spheric losses. In the absence of methanogenesis, the decreases in C–CH (‰) enclosure CH concentrations due solely to atmospheric evasion Figure 3 | Assessing the biochemical source of pelagic CH . (a) Here would have rapidly depleted the CH pool (dashed lines, Fig. 1a). patterns in the apparent fractionation (a ; see methods for calculation These calculations are potentially conservative, as we did not app details) during methanogenesis suggest that the acetoclastic pathway consider MOX. Considering all potential sources and sinks, we strongly dominated in all treatments, and increased in importance with the estimate an average rate of methanogenesis of 0.23 mmol m addition of nutrients. White and black circles represent values estimated per day (Table 1; Supplementary Table 2). This rate is higher than using measured CH oxidation rates (MOX) and a maximum and minimum previous estimates from in vitro incubations for oligotrophic Lake 3 18,19 associated isotopic fractionation factor, respectively. To constrain the effect Stechlin, Germany (0.04–0.09 mmol m per day) , but it of error introduced by MOX estimates, a was estimated from a highly should be noted that this estimate is very sensitive to the values of app conservative scenario analysis, summarized here by bars and error bars MOX used. In-situ MOX can vary dramatically across systems , 18,19 (midpoint and upper potential range, respectively). Values of a were with depth, diurnally and with changes in light exposure . app estimated for day 7 of the experiment, although isotopic signatures of CO Therefore, it is possible that our estimates of MOX and and ambient CH displayed little change through time for each treatment methanogenesis may be biased by the fact that water from only (b). Red, blue, grey and yellow symbols indicate control, DOC, DOC þ NP one depth was used, and incubations were run in the dark (see and NP treatments, respectively, whereas circles, squares and triangles, Methods section for details). Consequently, we consider our respectively, indicate days 0, 7 and 28 of the experiment. One enclosure estimates of net CH production (excluding MOX; Fig. 1b) as a was followed for each treatment, and sample points represent one lower bound for epilimnetic methanogenesis, as these estimates unreplicated measurement. are analogous to methanogenesis where MOX equals zero. In the Table 1 | Greenhouse gas dynamics in Lac Cromwell. CH and CO dynamics 4 2 3  2  2 (mmol m (mmol m (mol per lake (mmol CO eq m (mol CO eq per 2 2 per day) per day) per day) per day) lake per day) Pelagic production in enclosures 0.23 0.01 ± ± ± ± Pelagic diffusive fluxes from enclosures 0.15 0.03 15.5 2.6 1.4 0.2 140 24 ± ± ± ± Pelagic diffusive fluxes from lake 0.78 0.35 79.5 31.1 7.1 3.2 722 325 ± ± ± ± Ebullition from lake 0.31 0.66 15.2 32.5 2.8 6.0 138 295 ± ± CO diffusion from lake 33.4 12.4 3,408 1,262 Both oxic water column production and diffusion of CH in ambient (that is, control) experimental enclosures were estimated. Mean ( 1 s.d.) summertime (June–August inclusive) areal estimates of surface emissions and ebullition were determined for Lac Cromwell alongside the experiment. In all cases, CH emissions were converted to CO equivalents (where 1 kg CH ¼ 25 kg CO for a 100-year 4 2 4 2 period) to compare with areal and whole-lake CO emissions. Ebullition was not detected at depths 43 m, thereby restricting contributions to whole-lake flux. See Methods section for a detailed description of both enclosure and whole-lake calculations. 4 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. 13 Apparent fractionation ( ) C–CO (‰) app 2 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE ambient enclosures, net production averaged 0.09 and ranged zone may potentially contribute B20% of mean summertime from 0.04 to 0.12 mmol m per day, and is in close agreement CH diffusive fluxes, approximately the same magnitude as with estimates of methanogenesis in Lake Stechlin, where MOX ebullitive fluxes measured for the entire lake. Careful extrapola- 18,19 was shown to be extremely low throughout the epilimnion . tion of our absolute emissions rates (Fig. 1b, Table 1) to other The fact that our approximation of pelagic methanogenesis based ecosystems should be made, as previously overlooked, system- on mass balance (Fig. 1b; Table 1) is in good agreement with specific differences in gas exchange coefficients (k ) and non- CH 18,19 in vitro estimates is very promising, and suggests that both Fickian fluxes will introduce error to conventional cross-system approaches could be incorporated into future detailed studies of flux comparisons (see Supplementary Note 3 for extended CH dynamics for other systems. coverage of this potential problem). Although oxic water Our conclusion that acetate consumption supplied most CH methanogenesis contributed B4% to Lac Cromwell’s summer- in the enclosures, based on estimates of a , is particularly time greenhouse gas footprint, it is conceivable that this pathway app sensitive to changes in the rate of MOX. Given the potential could have greater relative importance in more nutrient-rich, variability in MOX discussed above, estimates of a may be productive ecosystems, and in hardwater environments where app biased by our limited MOX measurement. However, scenario low to negative (that is, net CO uptake) CO emissions are 2 2 42–44 analyses estimating a from extreme hypothetical MOX rates common . app (see Methods for calculations) were all well within the range of Within Lac Cromwell, it is interesting to note that oxic water acetoclastic dominance (Fig. 3a). This supports our conclusion methanogenesis contributed a significant component of whole- that acetate supplied the majority of CH across our treatments, lake emissions, despite the fact that it is a small (0.1 km ), likely for the duration of the experiment (Fig. 3b). Consistent relatively shallow (B3.5 m mean depth) lake with an extensive with this conclusion, observations of enriched surface d C-CH littoral area, and a CH -rich anoxic hypolimnion, all of which are 4 4 8,10,18,29,32 in numerous lakes have been hypothesized to typically major sources of CH . A corollary is that in large, deep indicate dominance of acetoclastic methanogenesis in oxic lakes and open ocean sites where surface waters are increasingly freshwater pelagic zones , and both known acetoclastic genera decoupled from benthic or littoral sources of CH , surface CH 4 4 (Methanosaeta and Methanosarcina) have recently been detected concentrations should be mostly driven by pelagic methanogen- 19,26 27,28 in oxic freshwater and terrestrial environments. esis and therefore a function of algal biomass and metabolism. In Furthermore, acetoclastic dominance was enhanced with the this regard, whereas chlorophyll a does not explain the large-scale addition of nutrients and increased algal production (Fig. 3a). patterns in surface CH across lakes in general , the relationship This pattern parallels that from anoxic environments, where we observed between surface water CH and chlorophyll a in increased abundance of biologically young, high-quality algal our isolated enclosures (log [CH ] ¼ 0.46log [chl a] þ 0.68, 10 4 10 material promotes acetoclastic over hydrogenotrophic r ¼ 0.68, n ¼ 4) generally agrees with observations specifically methanogenesis, the latter instead supported by lower quality, from large lakes, extending all the way to ultraoligotrophic open aged terrestrial DOC . Finally, cyanobacteria were near absent in ocean regions (Fig. 4; log [CH ] ¼ 0.99log [chl a]  1.63, 10 4 10 the mixed layer of our enclosures (o3.5% relative abundance, r ¼ 0.73). This continuity suggests that algal-linked pelagic as biomass, estimated from one enclosure per treatment on CH production may explain much of the ambient CH 4 4 day 7 in all treatments), ruling out major contributions of dynamics observed in large and deep aquatic ecosystems, diazotroph-derived water-column H for hydrogenotrophic regardless of the fact that different mechanisms have been methanogenesis , and cyanobacterial methanogenesis during identified as potentially important in marine (that is, 23 15,16,23–25 DOM demethylation . Taken together, it appears that methylphosphonate or methanethiol catabolism) acetoclastic methanogenesis linked to in-situ, algal DOC versus freshwater (hydrogenotrophic or acetoclastic methano- 18,19 production, potentially plays a central role in supporting water genesis; Fig. 3) . There are clearly a number of potential column CH production in oxic freshwaters. sources of CH in oxic pelagic waters, and although it remains 4 4 The potential importance of acetate for oxic water methanogen- unknown how each pathway differs in relative importance across esis is intriguing, as acetate concentrations are typically extremely these diverse aquatic environments, our results highlight a 18,33–35 low in oxic surface waters , and the presence of oxygen may consistent relationship between open-water phytoplankton inhibit acetate consumption . Yet past reports based on dynamics and the regulation of oxic pelagic CH . radioactive substrate additions have unequivocally demonstrated Taken together, our findings suggest that pelagic CH production extremely rapid turnover of acetate in oxic environments across in oxic environments may be widespread, and likely supports a 33–36 fresh and marine surface waters , suggesting active production baseline evasion of CH from all oxic water columns. In small and of acetate from multiple metabolic pathways in oxic shallow ecosystems, such as Lac Cromwell, this pathway will be 33–36 environments, as well as very efficient uptake .Fermentative quantitatively significant but secondary relative to anoxic littoral and metabolism by diverse and widespread facultative anaerobic benthic sources. In large and deep aquatic environments, however, bacteria in particle-associated microanoxic zones could be the this pathway could become the single dominant source fueling CH 38–40 main source of water column acetate , and such microhabitats supersaturation and fluxes to the atmosphere. It should be could be directly associated to algae or to particles ,yet in vitro emphasized that consideration of oxic water column methanogenesis work as well as our own results further suggest that methanogenesis does not necessarily lead to higher flux estimates for lakes, as the 18,19 proceeds even when these potential sites are removed .One CH derived from this process is already included in routine 2– possibility that should be further explored is that fermentative measurements of surface water pCH and CH surface fluxes 4 4 7,32,46 bacteria themselves create the conditions for anoxic . Incorporating the origin of the CH that outfluxes to the methanogenesis through syntrophic interactions with acetoclastic atmosphere from these lakes, however, is critical to our methanogens in mixed cell aggregates . understanding of the regulation of these fluxes and our capacity to Is oxic water methanogenesis a significant component of the predict their future change. In this regard, there are potentially large CH budget of Lac Cromwell? We have combined our estimated global implications for algal-driven oxic-water methanogenesis. CH 4 4 2–4 rates of oxic water methanogenesis and CH emissions in the emissions from surface waters, particularly from freshwaters ,are ambient enclosures with measurements of diffusive and ebullitive a major element of the global atmospheric CH budget, and we CH fluxes from the surrounding lake waters (summarized in suggest here that the algal-mediated baseline flux is not only a major Table 1), and conclude that methanogenesis in the oxic pelagic contributor to these overall aquatic CH emissions, but also one that NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 5 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 1,000 Enclosures - NP Enclosures - C Enclosures - DOC Enclosures - DOC+NP Lake Baikal (Southern basin) Lake Biwa Lake Constance (centre) Lake Constance (narrows) Canadian Boreal Lakes North Pacific Sub-Tropical Gyre Western North Pacific Ocean Arabian Sea Arctic Ocean 1 10 100 1,000 10,000 100,000 –1 Chlorophyll a (ng l ) Figure 4 | Linking dissolved CH and algal biomass across diverse open-water aquatic ecosystems. Here we combine experimental data from this study with literature-derived measurements from large (410 km ), deep (410 m maximum depth) lakes and marine environments where both CH and chlorophyll a samples were simultaneously measured. A strong positive relationship between CH and chlorophyll a exists across systems: (a) averaged data ([log x] ¼ 0.993[log y]  1.630, r ¼ 0.73, Po0.0001), where error bars represent 1 s.e.m.; (b) raw data ([log x] ¼ 0.855[log y]  1.166, 10 10 10 10 r ¼ 0.72, Po0.0001). Guidelines for meta-analysis are discussed in the Methods section, and raw data are available in Supplementary Table 3. is particularly sensitive to environmental change. As such, constant corrected for temperature. On day 7, dark incubations to estimate MOX using unfiltered surface water were conducted in the laboratory at room widespread and intensifying human- and climate-driven changes 47–49 50,51 temperature, lasting 96 h (Supplementary Fig. 4). Incubations began at in pelagic nutrient availability , terrestrial DOC inputs and approximately noon on 18 June, day 6 of the experiment; samples for measurement 47,48,51 physical structure of the water column , which strongly shape of MOX were collected from one mesocosm per treatment, at 25 cm depth, into 47,51–54 aquatic algal dynamics , may have major, but previously 12 ml precombusted vials. Samples were capped to exclude any air with a gastight, 3-mm-thick, butyl rubber-lined plastic cap, then incubated in the dark at 20 C. unconsidered, consequences for oxic water methanogenesis and Ambient CH measured in the mesocosms was used as a time 0 concentration, and aquatic CH emissions. samples were subsequently killed every 24 h by injection of HgCl , and stored at 4 C in the dark until analysis. Upon analysis, 5 ml of water was displaced with ambient air, and the remaining sample was equilibrated with the air by physically Methods shaking the vials. Headspace equilibration and gas chromatography analysis were Study site. Lac Cromwell is located at the Station de Biologie des Laurentides, performed as for ambient CH . Respiration rates were determined by dark in situ, a field research facility of the Universite´ de Montre´al. The lake turns over in 24 h incubations on experiment days 6 and 7 in all enclosures, using unfiltered spring and fall, is relatively shallow and small (mean depth ¼ 3.5 m, maximum 2 5 3 water in 4-l cubitainers. Initial and final samples of DO were collected and depth ¼ 9.5 m, surface area of 0.11 km , volume of 3  10 m ), and is oligo- 55 measured by membrane inlet mass spectrometry . mesotrophic, with little phytoplankton biomass (B6.0mgof chl a per litre) . CH chamber flux estimates. Air–water CH fluxes in the mesocosms and the 4 4 Experimental design. Here, 12, 1 m diameter, 6 m deep, 4712-L polyethylene surrounding lake were calculated by combining the measured surface water pCH enclosures were attached to floating wooden frames anchored in the middle of the and the CH gas exchange coefficient (k ), the latter derived from the measured 4 CH4 lake (B8 m depth). On 5 June 2012, bags were filled by pumping epilimnetic water gas exchange coefficient for CO (k ) in mesocosms supersaturated in CO . k 2 CO2 2 CO2 sequentially across two mesh screens of 110 and 45 mm to remove zooplankton, was measured in one mesocosm per treatment, 3–5 times daily during days 6 and 7 and three mesocosms each received N þ P, DOC, N þ P þ DOC additions, or no and 27–28 of the experiment, using the floating chamber method following Vachon addition (control). Both KH PO and NaNO additions were made to reach target 58 2 4 3 et al. Briefly, chambers were connected via a closed loop system to an EGM-4 concentrations of 50 and 700 mg P and N per litre, respectively. For DOC, infrared gas monitor (PP Systems). Fluxes were calculated as the rate of change of Superhume brand humic slurry was added to reach a target concentration of 15 mg chamber pCO per min during a 15-min interval. Diel sampling periods were of DOC per litre. For a full review of Superhume properties and applicability to spaced to obtain flux estimates from the morning, afternoon and nighttime periods, freshwater experimentation, see Lennon et al. . The mesocosms were allowed to and mean values were used for daily flux estimates. Dissolved CO samples were stabilize for 1 week before initial sampling and re-stocking zooplankton at ambient taken before all chamber measurements using the same headspace technique as for lake concentrations. Zooplanktons were collected from Lac Cromwell by vertical CH , but with direct injection of the equilibrated gas into the EGM-4 in the field. tows using a 54-mm Nitex net, and immediately released into mesocosms after The gas exchange coefficient for CH was then calculated from k using 4 CO2 each haul. equation (1), following Jahne et al. : 2=3 Limnological measurements. All limnological samples were gathered at B1200 k ¼ k  Sc Sc ð1Þ CH4 CO2 CH4 CO2 hours on days 0, 7 and 28 of the experiment (13, 19 June 2012 and 10 July 2012, respectively; Supplementary Fig. 3). Dissolved oxygen (DO) and temperature were where Sc is the Schmidt number of CO and CH , respectively . To calculate CH 2 4 4 measured at 0.5 m depth in all bags, whereas water column profiles were taken fluxes, we used these average k values, the measured pCH and the temperature- CH 4 from one enclosure per treatment using an Yellow Springs Instruments (YSI) Pro dependent solubility of the gas following equation (2): Plus multiparameter sonde (Supplementary Fig. 1). On all three dates, water was collected at 0.5 m depth, filtered to 0.45mm and chlorophyll a was measured F ¼ k K D ð2Þ gas h gas spectrophotometrically in ethanol filter extracts. Dissolved CH concentrations at 25 cm depth were measured in triplicate for each mesocosm following Prairie and where F is flux of CH (mmol m per day), K is the temperature-corrected 4 h del Giorgio . Partial pressures were converted to concentrations using Henry’s Henry’s constant and D is the difference in partial pressures between the air and gas 6 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. CH (nM) 4 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE estimates of a , based on this analysis, as upper and lower bounds for a the water partial pressures of CH . We assumed an atmospheric partial pressure of app app estimated using our actual measurements of MOX. CH of 1.75 p.p.m. Mesocosm mass balances. To quantify methanogenesis in the oxic mixed Isotopic analyses. Isotope compositions are reported in delta notation following layer of ambient enclosures, mass balances from the ambient control treatment equation (3): enclosures were constructed (Supplementary Table 2) for comparison with other 13 18 d C or d O ¼ R =R  1 1; 000 ð3Þ lake-based CH fluxes using equation (7): sample standard 13 13 where R is the ratio of heavy to light isotope, d C represents d C–CH or CH ¼ D þ E þ O  M ð7Þ 4 4gross CH4 13 18 18 18 C of dissolved inorganic carbon (DIC), d O represents d O–O or d O–H O 2 2 13 18 and R is the d C or d O signature of Vienna Pee Dee Belemnite or stan- standard Here rates of methanogenesis (CH ) are estimated by summing the rate of 4gross dard mean ocean water, respectively. All samples were taken at B25 cm depth. change in storage (D ), defined as the rate of increase in the ambient CH pool, CH4 4 18 13 Samples for d O–O and d C–CH were stored in 12-ml precombusted bor- 2 4 with the rates of CH evasion (E) and oxidation (O), minus the horizontal influx of osilicate vials, preserved with HgCl , and air-free samples capped with a gastight external CH via diffusion across the walls of the enclosures (M; detailed in the 2 4 rubber-lined plastic cap. d O–O samples were analysed at the University of Supplementary Methods). In all cases, mass balance terms were first standardized Ottawa Stable Isotopes Laboratory, and d C–CH samples were measured at the to the entire mixed layer or surface of the enclosure (as mmol per enclosure per University of Waterloo, both using standard laboratory methods. A single sample day) in order to combine aerial (E and M) and volumetric (D and O) processes CH4 for d O–H O was collected from the lake before filling enclosures, analysed with a on each sampling date. Picarro L230i isotopic water analyser, and assumed representative for all enclosures The enclosure mixed layer average depth of 1.68 m (Supplementary Fig. 1) was for the duration of the experiment. d C-DIC samples were collected in acid- used to calculate both mixed layer volume, and the surface area of the enclosure washed, 40 ml vials with both Teflon and rubber-lined gastight plastic caps, and wall that contributed to CH influx from the surrounding lake environment. The measured at the Colorado Plateau Stable Isotopes Laboratory using standard change in storage (D ) was calculated as the rate of change in ambient CH CH 4 18 18 methods. Day 7 respiration (R), d O–O and d O–H O were used to estimate concentrations in each bag as a function of time using the slope of the regression 2 2 GPP and NEP following Quin˜ones-Rivera et al. line as determined by least-squares regression analysis, then averaged and adjusted volumetrically to the enclosure scale (mmol per enclosure per day; Supplementary Table 2). The rate of MOX (O) for day 7 of the experiment was Numerical methods. We used RM-ANOVA to compare the effects of treatments, used for all calculations. Finally, estimates of CH were converted to units 4gross and treatment by time interactions, on ambient CH . One-way ANOVA was used 4  3 of mmol m per day for comparison with other lake-based measurements to assess differences between treatments for grouped data, and log (x þ 1) (Table 1). transformations applied to maintain the homogeneity of variances, followed by Bonferroni post-hoc pairwise comparisons of treatment means. All regression analyses were performed using Sigmaplot version 12, and ANOVAs were com- Whole-lake CH dynamics. Enclosure measurements and mass balance results puted with SPSS version 16. We used ordinary least squares regression to quantify were compared with the surrounding Lac Cromwell environment. To quantify the link between GPP or NEP and CH evasion. To estimate rates of MOX, the loss 4 summertime rates of ebullition in Lac Cromwell (Table 1), bubble traps were fixed of CH through time was fitted with a polynomial function, the derivative was at five permanent sampling sites along a transect from the littoral to pelagic at B1, taken and the slope at time 0 was calculated (Supplementary Fig. 4). This approach 2, 3, 5 and 7 m depths. Bubble traps were left in place over the entire sampling was deemed more accurate in quantifying the slope at time 0, as traditional period and collected monthly in June, July and August 2012. Traps consisted of an methods, such as log transformations followed by visual selection of the linear inverted funnel (63.5 cm diameter) suspended at 0.5 m below the surface of the portion of the data set , can introduce large errors in the slope of the regression water (0.25 m at the shallowest sample site). A graduated 1-l glass bottle was line when deciding which points to fit and which to exclude. Although MOX attached to the funnel by gluing the sides of the bottle cap to the neck of the funnel was estimated only once, our estimates are within the range of pelagic MOX to make a gastight seal. The bottles were covered with aluminum foil to prevent 3 31,32,62 (0.170–0.250 mmol m per day) previously reported , particularly for overheating or light exposure of the collected gas. Bottles were filled with water boreal lakes of similar trophic condition . upon deployment, so the gas accumulated by displacing water in the bottle. Bottles were collected once a month or until a volume greater than 250 ml of gas was detected. All gas was carefully removed by displacement with ambient water using Apparent isotopic fractionation of methanogenesis. We estimated the apparent a stopcock cap fitted with two syringes. A subset of the gas was injected in 30 ml fractionation factor (a ) during methanogenesis following Conrad . The app saline vials, and analysed upon return to the laboratory by gas chromatographic apparent fractionation factor is defined using equation (4) as analysis as detailed for enclosure samples. As the majority of samples settled in the 13 3 13 3 bottles for up to a month before collection, we assumed that the composition of a ¼ d C-CO þ 10 = d C-CH þ 10 ð4Þ app 2 4source CH in bubbles was 0.6 atm or 60%, based on a subset of samples analysed that 13 13 where d C–CH is the isotopic signature of source CH ,and d C–CO was were collected within days of sedimentary release. This assumption is consistent 4source 4 2 13 63 with literature estimates of bubble CH concentration . calculated from measured d C-DIC following the study by Stumm and Morgan , 4 using fractionation factors from the study by Mook et al. . As we measured Diffusive evasion of lake CH was estimated by first measuring concentrations 13 13 13 of CH on a monthly basis near each bubble trap site along the transect. ambient d C–CH (d C–CH ), we estimated d C–CH by correcting 4 4 4ambient 4source Concentrations ranged between 0.63 and 2.09 with an average of 1.21mM, and did for the isotopic fractionation effects of MOX and evasion to the atmosphere using not vary across sites (ANOVA, P ¼ 0.98, F ¼ 0.10). As detailed above for the an open-system, steady-state model for isotope fractionation , as defined in enclosure mass balance, atmospheric fluxes were estimated by applying the average equation (5): 24 h gas exchange coefficient measured at 7 m depth along the transect (mean 13 13 d C-CH ¼f d C-CH  D k ¼ 0.65 m per day, Supplementary Fig. 2). 4source 4ambient evasion CH ð5Þ Whole-lake CH and CO diffusive fluxes were calculated by applying average 4 2 þðÞ 1  f d C-CH  D 4ambient MOX areal estimates of diffusion to the entire lake surface area (102,000 m ). As ebullition was only detected at depths o3 m, average rates of ebullition where f and 1  f are the fractions of CH loss, standardized to daily rates for the (0.31 0.66 mmol m per day) were applied to the lake surface o3 m in depth entire enclosure, estimated, respectively, as evasion and MOX rates divided by the sum of each. D is the isotopic effect of each loss pathway. In cases such as ours (49,166 m ). This area was estimated by determining the area of the lake o3m in depth from a hypsographic curve for Lac Cromwell , and existing bathymetric where Do100%, D can be approximated using equation (6): information for Lac Cromwell . D ¼ðÞ 1 a 10 ð6Þ where a is the fractionation factor of a given reaction. Literature-derived values of a Data compilation for meta-analysis. To assess the relationship between algal were used for both loss pathways. For evasion, an a value of 0.9992 was used , and dynamics and CH in open water environments, we compared the results from our 62 2 for MOX, the range reported by Bastviken et al. (0.9816–0.9792) was used. experimental enclosures with directly obtained results from large (410 km ), deep To better constrain the potential range of a estimated on day 7 of our (410 m maximum depth) boreal lakes and other freshwater and marine data app experiment, we performed a scenario analysis by varying combinations of MOX from the literature (raw data and selection methods tabulated in Supplementary rates and isotopic fractionation values, to determine the sensitivity of our estimates Table 3). Literature data include only studies where CH and chlorophyll a were of a to potential variability in MOX. For this analysis, we chose two very measured simultaneously, except for Lake Baikal, where detailed studies of CH app 4 conservative, extreme (for oxic pelagic freshwaters) values in MOX rates (0 and and chlorophyll a were measured independently, but overlapped temporally and 3  1 0.05 mmol m h ) and paired each rate with both minimum and maximum thus were included. To remain consistent with the sampling approach from our values reported for isotopic fractionation during MOX (18.4 and 20.8%). study, and sampling of boreal lakes, only near-surface, open-water results from the Because the MOX rate of zero did not have associated isotopic fractionation, we centre of the lake or furthest from the coastal environment (in marine studies) were calculated three different combinations of extreme parameters: no oxidation with used. Summertime results were taken if seasonal data were presented. If surface no fractionation, high oxidation with low fractionation and high oxidation rates water results were given as a range, the midpoint of that range was chosen. If with high associated fractionation. We then took the maximum and minimum multiple samples were taken at one site, then results were averaged. NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 7 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 References 30. Conrad, R. Quantification of methanogenic pathways using stable carbon 1. Solomon, S. et al. in Climate Change 2007: The Physical Science Basis. isotopic signatures: A review and a proposal. Org. Geochem. 36, 739–752 Contribution of Working Group I to the Fourth Assessment Report of the (2005). Intergovernmental Panel on Climate Change. (eds Solomon, S. et al.) 31. Utsumi, M. et al. Oxidation of dissolved methane in a eutrophic, shallow lake: (Cambridge Univ., 2007). Lake Kasumigaura, Japan. Limnol. Oceanogr. 43, 471–480 (1998). 2. Bastviken, D., Tranvik, L. J., Downing, J. A., Crill, P. M. & Enrich-Prast, A. 32. Bastviken, D., Cole, J. J., Pace, M. L. & Van de Bogert, M. C. Fates of methane Freshwater methane emissions offset the continental carbon sink. Science 331, from different lake habitats: Connecting whole-lake budgets and CH 50 (2011). emissions. J. Geophys. Res. 113, G02024 (2008). 3. Michmerhuizen, C. M., Striegl, R. G. & Mcdonald, M. E. Potential methane 33. Wright, R. R. & Hobbie, J. E. Use of glucose and acetate by bacteria and algae in emission from north-temperate lakes following ice melt. Limnol. Oceanogr. 41, aquatic ecosystems. Ecology 47, 447–464 (1966). 985–991 (1996). 34. Allen, H. L. Acetate in fresh water: Natural substrate concentrations determined 4. Kirschke, S. et al. Three decades of global methane sources and sinks. Nat. by dilution bioassay. Ecology 49, 346–349 (1968). Geosci. 6, 813–823 (2013). 35. Ho, T.-Y. et al. Acetate cycling in the water column of the Cariaco Basin: 5. Prairie, Y. T. & del Giorgio, P. A. A new pathway of freshwater methane Seasonal and vertical variability and implication for carbon cycling. Limnol. emissions and the putative importance of microbubbles. Inl. Waters 3, 311–320 Oceanogr. 47, 1119–1128 (2002). (2013). 36. Sepers, A. The utilization of dissolved organic compounds in aquatic 6. Rudd, J. W. M. & Hamilton, R. D. Methane cycling in a eutrophic shield lake and environments. Hydrobiologia 52, 39–54 (1977). its effects on whole lake metabolism. Limnol. Oceanogr. 23, 337–348 (1978). 37. Thauer, R. K., Mo¨ller-Zinkhan, D. & Spormann, A. M. Biochemistry of acetate 7. Bastviken, D., Cole, J., Pace, M. & Tranvik, L. Methane emissions from lakes: catabolism in anaerobic chemotrophic bacteria. Annu. Rev. Microbiol. 43, dependence of lake characteristics, two regional assessments, and a global 43–67 (1989). estimate. Global Biogeochem. Cycles 18, GB4009 (2004). 38. Riemann, L. & Azam, F. Widespread N-acetyl-d-glucosamine uptake among 8. Murase, J., Sakai, Y., Sugimoto, A., Okubo, K. & Sakamoto, M. Sources of pelagic marine bacteria and its ecological implications. Appl. Environ. dissolved methane in Lake Biwa. Limnology 4, 91–99 (2003). Microbiol. 68, 5554–5562 (2002). 9. Schmidt, U. & Conrad, R. Hydrogen, carbon monoxide, and methane dynamics 39. Alonso, C. & Pernthaler, J. Incorporation of glucose under anoxic conditions by in Lake Constance. Limnol. Oceanogr. 38, 1214–1226 (1993). bacterioplankton from coastal North Sea surface waters. Appl. Environ. 10. Schulz, M., Faber, E., Hollerbach, A., Schroder, H. G. & Gude, H. The methane Microbiol. 71, 1709–1716 (2005). cycle in the epilimnion of Lake Constance. Arch. Hydrobiol. 151, 157–176 (2001). 40. Tada, Y. & Grossart, H.-P. Community shifts of actively growing lake bacteria 11. Hofmann, H. Spatiotemporal distribution patterns of dissolved methane in after N-acetyl-glucosamine addition: Improving the BrdU-FACS method. ISME lakes: How accurate are the current estimations of the diffusive flux path? J. 8, 1–14 (2013). Geophys. Res. Lett. 40, 2779–2784 (2013). 41. Stams, A. J. M. & Plugge, C. M. Electron transfer in syntrophic 12. Schmid, M. et al. Sources and sinks of methane in Lake Baikal: A synthesis of communities of anaerobic bacteria and archaea. Nat. Rev. Microbiol. 7, measurements and modeling. Limnol. Oceanogr. 52, 1824–1837 (2007). 568–577 (2009). 13. Scranton, M. I. & Brewer, P. G. Occurrence of methane in the near-surface 42. Finlay, K., Leavitt, P. R., Wissel, B. & Prairie, Y. T. Regulation of spatial and waters of the western subtropical North-Atlantic. Deep. Sea Res. 24, 127–138 temporal variability of carbon flux in six hard-water lakes of the northern Great (1977). Plains. Limnol. Oceanogr. 54, 2553–2564 (2009). 14. Owens, N. J. P., Law, C. S., Mantoura, R. F. C., Burkill, P. H. & Llewellyn, C. A. 43. Raymond, P. A. et al. Global carbon dioxide emissions from inland waters. Methane flux to the atmosphere from the Arabian Sea. Nature 354, 293–296 Nature 503, 355–359 (2013). (1991). 44. McDonald, C. P., Stets, E. G., Striegl, R. G. & Butman, D. Inorganic carbon 15. Damm, E. et al. Methane production in aerobic oligotrophic surface water in loading as a primary driver of dissolved carbon dioxide concentrations in the the central Arctic Ocean. Biogeosciences 7, 1099–1108 (2010). lakes and reservoirs of the contiguous United States. Global Biogeochem. Cycles 16. Karl, D. M. et al. Aerobic production of methane in the sea. Nat. Geosci. 1, 27, 285–295 (2013). 473–478 (2008). 45. Rasilo, T., Prairie, Y. T. & del Giorgio, P. A. Large-scale patterns in summer 17. Tilbrook, B. D. & Karl, D. M. Methane sources, distributions and sinks from diffusive CH fluxes across boreal lakes, and contribution to diffusive C California coastal waters to the oligotrophic North Pacific gyre. Mar. Chem. 49, emissions. Glob. Chang. Biol. (In the press). 51–64 (1995). 46. Ortiz-Llorente, M. J. & Alvarez-Cobelas, M. Comparison of biogenic methane 18. Tang, K. W., McGinnis, D. G., Frindte, K., Bru¨chert, V. & Grossart, H.-P. emissions from unmanaged estuaries, lakes, oceans, rivers and wetlands. Atmos. Paradox reconsidered: Methane oversaturation in well-oxygenated lake waters. Environ. 59, 328–337 (2012). Limnol. Oceanogr. 59, 275–284 (2014). 47. Schindler, D. W. Recent advances in the understanding and management of 19. Grossart, H.-P., Frindte, K., Dziallas, C., Eckert, W. & Tang, K. W. Microbial eutrophication. Limnol. Oceanogr. 51, 356–363 (2006). methane production in oxygenated water column of an oligotrophic lake. Proc. 48. Moss, B. Cogs in the endless machine: Lakes, climate change and nutrient Natl Acad. Sci. USA 108, 19657–19661 (2011). cycles: a review. Sci. Total Environ. 434, 130–142 (2012). 20. De Angelis, M. A. & Lee, C. Methane production during zooplankton grazing 49. Carpenter, S. R. et al. Nonpoint pollution of surface waters with phosphorus on marine phytoplankton. Limnol. Oceanogr. 39, 1298–1308 (1994). and nitrogen. Ecol. Appl. 8, 559–568 (1998). 21. Oremland, R. S. Methanogenic activity in plankton samples and fish intestines: 50. Monteith, D. T. et al. Dissolved organic carbon trends resulting from changes A mechanism for in situ methanogenesis in oceanic surface waters. Limnol. in atmospheric deposition chemistry. Nature 450, 537–540 (2007). Oceanogr. 24, 1136–1141 (1979). 51. Schindler, D. W. et al. Climate-induced changes in the dissolved organic carbon 22. Ditchfield, A. et al. Identification of putative methylotrophic and budgets of boreal lakes. Biogeochemistry 36, 9–28 (1997). hydrogenotrophic methanogens within sedimenting material and copepod 52. Taranu, Z. E., Zurawell, R. W., Pick, F. & Gregory-Eaves, I. Predicting faecal pellets. Aquat. Microb. Ecol. 67, 151–160 (2012). cyanobacterial dynamics in the face of global change: The importance of scale 23. Beversdorf, L. J., White, A. E., Bjo¨rkman, K. M., Letelier, L. M. & Karl, D. M. and environmental context. Glob. Chang. Biol. 18, 3477–3490 (2012). Phosphonate metabolism by Trichodesmium IMS101 and the production of 53. Carpenter, S. R., Cole, J. J., Kitchel, J. F. & Pace, M. L. Impact of dissolved greenhouse gases. Limnol. Oceanogr. 55, 1768–1778 (2010). organic carbon, phosphorus, and grazing on phytoplankton biomass and 24. Carini, P., White, A. E., Campbell, E. O. & Giovannoni, S. J. Methane production in experimental lakes. Limnol. Oceanogr. 43, 73–80 (1998). production by phosphate-starved SAR11 chemoheterotrophic marine bacteria. 54. Jones, R. I. The influence of humic substances on lacustrine planktonic food Nat. Commun. 5, 4346 (2014). chains. Hydrobiologia 229, 73–91 (1992). 25. del Valle, D. A. & Karl, D. M. Aerobic production of methane from dissolved 55. Den Heyer, C. & Kalff, J. Organic matter mineralization rates in sediments: methylphosphonate in the water column and sinking particles collected in the A within- and among-lake study. Limnol. Oceanogr. 43, 695–705 (1998). North Pacific Subtropical Gyre. Aquat. Microb. Ecol 73, 93–105 (2014). 56. Lennon, J. T. et al. A source of terrestrial organic carbon to investigate the 26. Paganin, P. et al. Vertical distribution of bacterioplankton in Lake Averno in browning of aquatic ecosystems. PLoS One 8, e75771 (2013). relation to water chemistry. FEMS Microbiol. Ecol. 84, 176–188 (2013). 57. Bouvier, T. C. & del Giorgio, P. A. Compositional changes in free-living 27. Angel, R., Matthies, D. & Conrad, R. Activation of methanogenesis in arid bacterial communities along a salinity gradient in two temperate estuaries. biological soil crusts despite the presence of oxygen. PLoS One 6, e20453 (2011). Limnol. Oceanogr. 47, 453–470 (2002). 28. Angel, R., Claus, P. & Conrad, R. Methanogenic archaea are globally ubiquitous 58. Vachon, D., Prairie, Y. T. & Cole, J. J. The relationship between near-surface in aerated soils and become active under wet anoxic conditions. ISME J. 6, turbulence and gas transfer velocity in freshwater systems and its implications 847–862 (2012). for floating chamber measurements of gas exchange. Limnol. Oceanogr. 55, 29. Miyajima, T. et al. Distribution of greenhouse gases, nitrite, and Cof 1723–1732 (2010). dissolved inorganic carbon in Lake Biwa: Implications for hypolimnetic 59. Jahne, B. et al. On the parameters influencing air-water gas exchange. metabolism. Biogeochemistry 36, 205–221 (1997). J. Geophys. Res. 92, 1937–1949 (1987). 8 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE ´ ´ biologie des Laurentides de l’Universite de Montreal provided logistical support. Terhi 60. Wanninkhof, R. Relationship between wind speed and gas exchange over the Rasilo and Tonya Del Sontro revised an early version of the manuscript. M.J.B. was ocean. J. Geophys. Res. 97, 7373–7382 (1992). supported by a doctoral scholarship from the National Science and Engineering Research 61. Quin˜ones-Rivera, Z. J., Wissel, B. & Justic´, D. Development of productivity Council of Canada (NSERC). This project is part of the programme of the NSERC/HQ models for the northern Gulf of Mexico based on oxygen concentrations and Industrial Research Chair in Carbon Biogeochemistry in Boreal Aquatic Systems stable isotopes. Estuaries Coasts 32, 436–446 (2009). (CarBBAS), and was co-funded by grants from NSERC (to P.A.d.G. and A.M.D.), Hydro- 62. Bastviken, D., Ejlertsson, J. & Tranvik, L. Measurement of methane oxidation in Que´bec (to P.A.d.G.) and a Fonds de recherche Que´bec–Nature et Technologie award lakes: A comparison of methods. Environ. Sci. Technol. 36, 3354–3361 (2002). from the government of Quebec (to A.M.D.). 63. Stumm, W. & Morgan, J. J. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters (Wiley, 1996). 64. Mook, K. A., Bommerson, J. C. & Staverman, W. H. Carbon isotope Author Contributions fractionation between dissolved bicarbonate and gaseous carbon dioxide. Earth M.J.B, P.A.d.G. and A.M.D. designed the study; M.J.B., L.B., M.C.G.C. and A.M. per- Planet. Sci. Lett. 22, 169–176 (1974). formed field and lab work; M.J.B. and Y.T.P. analysed the data; M.J.B. and P.A.d.G. wrote 65. Fry, B. Stable Isotope Ecology (Springer, 2006). the paper; all authors revised the paper; A.M.D. and P.A.d.G. contributed materials. 66. Knox, M., Quay, P. D. & Wilbur, D. Kinetic isotopic fractionation during air-water gas transfer of O ,N ,CH , and H . J. Geophys. Res. 97, 335–343 2 2 4 2 (1992). Additional information 67. Livingstone, D. M. & Imboden, D. M. The prediction of hypolimnetic oxygen Supplementary Information accompanies this paper at http://www.nature.com/ profiles: a plea for a deductive approach. Can. J. Fish. Aquat. Sci. 53, 924–932 naturecommunications (1996). Competing financial interests: The authors declare no competing financial ´ ´ 68. Carignan, R Station de biologie des Laurentides–Universite de Montreal http:// interests. www.sbl.umontreal.ca/territoire-cartes/cartes/index.html (2010). Reprints and permission information is available online at http://npg.nature.com/ Acknowledgements reprintsandpermissions/ Simon Gauthier-Fautaux, Juan Pablo Nino Garcia, Cynthia Soued, Marilyne Robidoux and Ryan Hutchins provided field and laboratory assistance. Jean-Francois Lapierre, How to cite this article: Bogard, M. J. et al. Oxic water column methanogenesis Dominic Vachon, Adam Heathcote and David Bastviken provided conceptual and as a major component of aquatic CH fluxes. Nat. Commun. 5:5350 methodological advice. Annick St. Pierre, Alice Parks and the employees of the Station de doi: 10.1038/ncomms6350 (2014). NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 9 & 2014 Macmillan Publishers Limited. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Oxic water column methanogenesis as a major component of aquatic CH4 fluxes

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/oxic-water-column-methanogenesis-as-a-major-component-of-aquatic-ch4-RSGbvMhzFs

References (76)

Publisher
Springer Journals
Copyright
Copyright © 2014 by Nature Publishing Group, a division of Macmillan Publishers Limited. All Rights Reserved.
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/ncomms6350
Publisher site
See Article on Publisher Site

Abstract

ARTICLE Received 29 May 2014 | Accepted 22 Sep 2014 | Published 30 Oct 2014 DOI: 10.1038/ncomms6350 Oxic water column methanogenesis as a major component of aquatic CH fluxes 1 1 1 1 1 Matthew J. Bogard , Paul A. del Giorgio , Lennie Boutet , Maria Carolina Garcia Chaves , Yves T. Prairie , 1 1 Anthony Merante & Alison M. Derry Methanogenesis has traditionally been assumed to occur only in anoxic environments, yet there is mounting, albeit indirect, evidence of methane (CH ) production in oxic marine and freshwaters. Here we present the first direct, ecosystem-scale demonstration of methanogenesis in oxic lake waters. This methanogenesis appears to be driven by acetoclastic production, and is closely linked to algal dynamics. We show that oxic water methanogenesis is a significant component of the overall CH budget in a small, shallow lake, and provide evidence that this pathway may be the main CH source in large, deep lakes and open oceans. Our results challenge the current global understanding of aquatic CH dynamics, and suggest a hitherto unestablished link between pelagic CH emissions and surface-water primary production. This link may be particularly sensitive to widespread and increasing human influences on aquatic ecosystem primary productivity. Groupe de Recherche Interuniversitaire en Limnologie, De´partement des Sciences Biologiques, Universite´ du Que´bec a` Montre´al, Case Postale 8888, Succursale Centre-Ville, Montre´al, Quebec, Canada H3C 3P8. Correspondence and requests for materials should be addressed to M.J.B. (email: [email protected]). NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 1 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ethane (CH ) emissions currently contribute B20% The observed surface CH concentrations in the mesocosms 4 4 to the planet’s greenhouse effect, with a large, but represent the net balance between CH production, CH 4 4 Mpoorly defined, fraction derived from freshwater oxidation (MOX) and CH exchange with the atmosphere. In 1–4 ecosystems . Accurately placing freshwaters in the global CH the absence of internal inputs of CH , the initial CH super- 4 4 4 budget requires a better understanding of the controls and saturation in the enclosures would have declined to atmospheric 2–5 contributions of CH from different sources . In this regard, the equilibrium within approximately 1 week (Fig. 1a, dotted lines), surface waters of lakes and rivers are systematically as estimated on the basis of our own empirical measurements of supersaturated with CH , and it has been traditionally assumed the gas exchange coefficient (see Methods section). Thus, an that this CH is derived from anoxic environments, via vertical internal CH source was necessary to sustain the systematic CH 4 4 4 3,6–8 and lateral transport from profundal and littoral sediments . supersaturation observed in all mesocosms throughout the This assumption certainly holds for small, shallow ecosystems experiment. Further, there was an overall significant increase in where the surface layers are in relatively close contact with CH concentrations through time across treatments (Fig. 1a; sediments. Yet CH supersaturation is also prevalent in large, repeated measures analysis of variance (RM-ANOVA) time effect: 3,5,7–12 13–17 deep lakes and oceanic surface waters , where deeper P ¼ 0.007), with greater increases in nutrient-amended enclosures water columns result in significantly reduced surface water (nitrogen and phosphorus (NP), dissolved organic carbon exchange with anoxic sediments. Studies of CH dynamics in (DOC)-NP), and weaker increases in the DOC and control surface waters of oceans and large lakes have indeed concluded enclosures (RM-ANOVA treatment  time effect: P ¼ 0.04). As a that pelagic CH supersaturation cannot be sustained either result, all enclosures emitted CH to the atmosphere for the 4 4 by lateral inputs from the littoral or from benthic inputs duration of the experiment, and the estimated water/air CH 9,10,13–15,17–19 alone . fluxes (based on our measurements of gas transfer coefficients in A pelagic source of CH would thus be required to sustain the the mesocosms) ranged from 0.07 to 0.36 mmol m per day observed CH supersaturation, and multiple lines of evidence (Fig. 1b). Net CH production in the enclosures, calculated as the 4 4 suggest that CH is produced in oxic water columns of freshwater sum of the observed net increase in concentration and the and marine systems. In vitro experiments point to CH calculated CH fluxes to the atmosphere, were on average highest 4 4 production in both microanoxic habitats in metazoan guts and in the nutrient-enriched enclosures, whereas the DOC-only 14,17–23 15,16,18,19,23–25 on particles , and in particle-free oxic water , addition generated the lowest production rates (Fig. 1b; analysis 19,22,26–28 via multiple biochemical pathways. The presence of variance (ANOVA): P ¼ 0.003). We assessed the robustness of 19,22,27 and activity of hydrogenotrophic and acetoclastic chamber-based calculations of CH emissions by comparing methanogenic archaea have been confirmed at the molecular chamber-derived gas exchange coefficients (k ) to estimates CH4 level in diverse oxic environments. At the same time, marine based on wind speed (detailed in Supplementary Note 1). studies suggest that the microbial decomposition of methylated There was good agreement between the two approaches, and 15 16,23–25 compounds such as methanethiol and methylphosphonate therefore we conclude that chamber-derived results yield reliable could be an important source of CH in surface waters of the open estimates of k and CH fluxes for both the enclosures and 4 CH4 4 ocean. Further, metalimnetic peaks of CH are a recurrent feature the lake. in many ecosystems, which correlate positively with dissolved O 8–10,14,15,18,19 (DO), algal biomass and production .However, despite past efforts, the ecological and biogeochemical significance of oxic Linking CH production to algal dynamics. The differences water column methanogenesis remain speculative, and we have among treatments in water column CH concentrations and yet to determine the dominant biochemical pathway, its controls, fluxes were strongly linked to pelagic gross primary production and its contribution to diffusive CH emissions from aquatic (GPP; Fig. 2a) and net ecosystem production (NEP ¼ GPP-R; ecosystems. Fig. 2b). There was a 20-fold range in GPP across treatments In this study, we experimentally assess pelagic CH production (Fig. 2a), whereas there was only a 5-fold range in ecosystem in the oxygenated water column of Lac Cromwell, a typical respiration (R), therefore nutrient additions resulted in a strong Canadian temperate Shield lake, using floating mesocosms open shift in NEP (Fig. 2b). The average CH fluxes at day 7 of the to the atmosphere, but closed at the bottom and therefore experiment were strongly positively related to average GPP rates uncoupled from non-pelagic sources of CH . The results in the enclosures (Fig. 2a), and to NEP, whereas the DOC addi- presented herein represent the first direct, ecosystem-scale tion had no positive influence relative to the control. Increases in demonstration of oxic water CH production, its drivers and CH flux appear to have been driven mostly by increases in GPP 4 4 source pathway, and the potential widespread importance of this rather than heterotrophic metabolism. poorly considered process. Other potential sources of CH in the enclosures were also considered. The enclosures contained very little particulate organic C because of the initial filtration of the surrounding lake Results water before filling, and there was little or no accumulation of CH dynamics in experimental enclosures. The initial CH particulate organic C in the bottom of enclosures at the end of the 4 4 concentrations in the enclosures were four- to tenfold lower than experiment. Moreover, we estimate that zooplankton-derived surrounding lake waters, because of degassing during the initial CH , calculated from the measured zooplankton biomass in the filtration and filling process, but the enclosures were nevertheless enclosures and published production rates (Supplementary supersaturated relative to the atmosphere at the onset of the Table 1, Supplementary Note 2), contributed o10% to the experiment (Fig. 1a). Vertical profiles of the enclosures estimated gross CH production rates in ambient enclosures, a 10,18,19,29 (Supplementary Fig. 1) showed a range in DO between 45.6 and result that is consistent with previous studies . Membrane 128.6% saturation, indicating that the entire water column permeability and the resulting exchange with the surrounding remained oxic in all treatments for the duration of the experi- lake waters were also considered as a potential cause for CH ment. Despite oxic conditions, CH concentrations ranged supersaturation. However, the estimated contributions of CH 4 4 between 0.10 and 0.53mM throughout the entire experiment, from the surrounding lake environment were on the order of across all treatments (Fig. 1a), which represents B50- to 265-fold o1% of gross CH production (Supplementary Methods and super-saturation relative to the atmosphere. Supplementary Table 2). Finally, there was no significant 2 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. –2 –1 Evasion (μmol CH m d ) –1 Net production (μM CH d ) NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE 0.6 0.25 400 DOC 0.20 DOC-NP 0.4 NP 0.15 0.10 0.2 0.05 CH at atmospheric equilibrium 0.0 0.00 01724 1 28 Day of experiment Treatment Figure 1 | Surface water CH dynamics in experimental enclosures. Treatments include ambient untreated controls (C), addition of dissolved organic carbon (DOC), nitrogen and phosphorus (NP) or a combination of both (DOC-NP). (a)CH concentrations (error bars 1 s.e.m., n ¼ 3) generally increased through time, but differed between treatments (RM-ANOVA, treatment: P ¼ 0.004, F ¼ 10.71; time: P ¼ 0.007, F ¼ 7.67; treatment* time: P ¼ 0.04, F ¼ 3.04; Tukey’s HSD post-hoc grouping ¼ NP, DOC-NP, C4C, DOC). Dashed lines show the expected declines in dissolved CH because of diffusion at the air–water interface, had no methane been produced in situ. Without internal production of CH , concentrations would have equilibrated with the atmosphere (mean concentration at equilibrium ¼ 0.002mM) in all treatments within 1–2 weeks. (b) Collectively, net CH production (evasion plus water column accumulation per enclosure) and evasion rates (second y axis) displayed large differences among treatments (ANOVA, P ¼ 0.003; post-hoc groups ¼ NP & DOC-NP & C4C4C & DOC). Thick and thin horizontal lines are mean and median, respectively. Identifying the biochemical source of CH . Several methanogenic pathways exist, but as acetoclastic methanogenesis is less isotopically discriminatory than CO reduction or 19,27,28 methylotrophy , apparent fractionation factors (a see app; Methods for calculation details) can be used to qualitatively distinguish whether CH is produced via methanogenesis from acetate, or CO reduction . Our measured a values were well 2 app within the range indicating dominance of the acetoclastic pathway (Fig. 3a), suggesting that acetate was the dominant source material for pelagic methanogenesis in all treatments. As we only had one estimate of water column MOX per 0  3  1 treatment (range ¼ 0.003–0.008 mmol m h ), we used a 0 4 8 12 scenario analysis with two extreme MOX rates (0 and –1 GPP (μM O h ) 3  1 0.05 mmol m h ; detailed in Methods section) to conservatively constrain the possible a values based on 300 app variable MOX rates and associated isotopic fractionation. In all cases, a values remained well within the acetoclastic range, that app is, o1.055 (ref. 30; Fig. 3a). Acetoclastic methanogenesis was 13 13 likely dominant through time, because d C-CO and d C-CH 2 4 showed little temporal variability (Fig. 3b). Rates of oxic water-column methanogenesis. The CH accu- mulation and outgassing that we measured in the enclosures (Fig. 1) represent the net result of pelagic methanogenesis, gas exchange and MOX. In order to derive a first-order estimate of –2 0 2 4 6 8 pelagic methanogenesis, we carried out a detailed CH mass –1 NEP (μM O h ) 2 balance (described in the Methods section; summarized in Supplementary Table 2) for day 7 for the control enclosures. Here Figure 2 | Linking pelagic ecosystem metabolism and CH dynamics. we combined the measured CH evasion, the change in storage Metabolic estimates from day 7 of the experiment revealed CH evasion corrected for cross-membrane inputs from surrounding lake was strongly positively correlated to (a) ecosystem gross primary water and measured MOX. This exercise yielded rates of apparent 2  1.6x production (GPP; P ¼ 0.003; r ¼ 0.93; y ¼ 215.3[1  e ]), and pelagic methanogenesis on the order of 0.21–0.24 mmol m per (b) shifts to increased net ecosystem production (NEP), which is different day (Table 1), of which B60% was apparently oxidized and the between GPP and community respiration (P ¼ 0.02; r ¼ 0.99; y ¼ 111– remainder evaded to the atmosphere. 2.6x 106[1  e ]). Symbols with error bars ( 1 s.e.m., n ¼ 3) denote the different treatments, including ambient untreated controls (red circles), addition of dissolved organic carbon (blue squares), nitrogen and Oxic water methanogenesis in a whole-lake perspective. Mean phosphorus (grey triangles) or a combination of both (yellow diamonds). pelagic CH evasion from control (that is, non-manipulated) enclosures (0.15 0.03 mmol m per day; Table 1) represented development of biofilm on the walls of the enclosures during B20% of the average diffusive CH fluxes measured in the the experiment, and this is unlikely to have been a significant ± lake over the summer (0.78 0.35 mmol m per day), and at source of CH . the whole-lake scale, were similar to CH ebullitive fluxes NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 3 & 2014 Macmillan Publishers Limited. All rights reserved. DOC DOC- NP NP –2 –1 –2 –1 CH evasion (μmol m d)CH evasion (μmol m d ) 4 4 Dissolved CH (μM) 4 ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ± ± (15.2 32.5 mol per day or 0.31 0.66 mmol m per day in gas emissions (CO þ CH , expressed as CO equivalents; 2 4 2 lake regions o3 m deep; Table 1). Further, oxic pelagic metha- Table 1). It was possible to compare fluxes directly between the nogenesis contributed on the order of 4% to total lake greenhouse enclosures and the lake because the physical conditions shaping gas exchange dynamics were quantitatively similar for each. As shown in Supplementary Fig. 2, floating-chamber-derived gas 1.08 H /CO or DOM-CH exchange coefficients (k ) for the lake and the enclosures varied 2 2 3 CH4 dominance on a diurnal basis, but averaged 0.65 and 0.69 m per day and 1.06 ranged between 0.10–1.91 and 0.24–1.44 m per day for the Acetate-dominance enclosures and the lake, respectively. 1.04 Discussion Here we present an ecosystem-scale, experimental demonstration 1.02 of significant CH production and evasion from the oxic water column of Lac Cromwell that is closely linked to algal dynamics, and which contributes a baseline flux of CH that is likely present 1.00 in all lakes. The relationship between CH and phytoplankton C DOC DOC- NP 4 NP observed here (Fig. 2) has been hypothesized before to explain Treatment both the presence of metabolically active methanogens, and the recurrent metalimnetic and near-surface CH peaks in oxic 9,10,18,19 13–17 lake and marine environments. We confirm this link α = 1.06 1.04 1.02 1.00 app experimentally, and further show that it generates a significant –5 out flux of CH from the mesocosms to the atmosphere (Fig. 1b). These results in turn imply that factors influencing –10 phytoplankton standing stock and GPP, such as grazing, nutrient availability and the physical structure of the water –15 column, will have a strong bearing on pelagic CH dynamics and –20 resulting CH emissions. To our knowledge, our results represent the first ecosystem- –25 level estimates of methanogenesis in oxic freshwaters. For all treatments, intense CH production was needed to sustain the –30 observed patterns in CH concentrations and to offset atmo- –90 –70 –50 –30 –10 4 spheric losses. In the absence of methanogenesis, the decreases in C–CH (‰) enclosure CH concentrations due solely to atmospheric evasion Figure 3 | Assessing the biochemical source of pelagic CH . (a) Here would have rapidly depleted the CH pool (dashed lines, Fig. 1a). patterns in the apparent fractionation (a ; see methods for calculation These calculations are potentially conservative, as we did not app details) during methanogenesis suggest that the acetoclastic pathway consider MOX. Considering all potential sources and sinks, we strongly dominated in all treatments, and increased in importance with the estimate an average rate of methanogenesis of 0.23 mmol m addition of nutrients. White and black circles represent values estimated per day (Table 1; Supplementary Table 2). This rate is higher than using measured CH oxidation rates (MOX) and a maximum and minimum previous estimates from in vitro incubations for oligotrophic Lake 3 18,19 associated isotopic fractionation factor, respectively. To constrain the effect Stechlin, Germany (0.04–0.09 mmol m per day) , but it of error introduced by MOX estimates, a was estimated from a highly should be noted that this estimate is very sensitive to the values of app conservative scenario analysis, summarized here by bars and error bars MOX used. In-situ MOX can vary dramatically across systems , 18,19 (midpoint and upper potential range, respectively). Values of a were with depth, diurnally and with changes in light exposure . app estimated for day 7 of the experiment, although isotopic signatures of CO Therefore, it is possible that our estimates of MOX and and ambient CH displayed little change through time for each treatment methanogenesis may be biased by the fact that water from only (b). Red, blue, grey and yellow symbols indicate control, DOC, DOC þ NP one depth was used, and incubations were run in the dark (see and NP treatments, respectively, whereas circles, squares and triangles, Methods section for details). Consequently, we consider our respectively, indicate days 0, 7 and 28 of the experiment. One enclosure estimates of net CH production (excluding MOX; Fig. 1b) as a was followed for each treatment, and sample points represent one lower bound for epilimnetic methanogenesis, as these estimates unreplicated measurement. are analogous to methanogenesis where MOX equals zero. In the Table 1 | Greenhouse gas dynamics in Lac Cromwell. CH and CO dynamics 4 2 3  2  2 (mmol m (mmol m (mol per lake (mmol CO eq m (mol CO eq per 2 2 per day) per day) per day) per day) lake per day) Pelagic production in enclosures 0.23 0.01 ± ± ± ± Pelagic diffusive fluxes from enclosures 0.15 0.03 15.5 2.6 1.4 0.2 140 24 ± ± ± ± Pelagic diffusive fluxes from lake 0.78 0.35 79.5 31.1 7.1 3.2 722 325 ± ± ± ± Ebullition from lake 0.31 0.66 15.2 32.5 2.8 6.0 138 295 ± ± CO diffusion from lake 33.4 12.4 3,408 1,262 Both oxic water column production and diffusion of CH in ambient (that is, control) experimental enclosures were estimated. Mean ( 1 s.d.) summertime (June–August inclusive) areal estimates of surface emissions and ebullition were determined for Lac Cromwell alongside the experiment. In all cases, CH emissions were converted to CO equivalents (where 1 kg CH ¼ 25 kg CO for a 100-year 4 2 4 2 period) to compare with areal and whole-lake CO emissions. Ebullition was not detected at depths 43 m, thereby restricting contributions to whole-lake flux. See Methods section for a detailed description of both enclosure and whole-lake calculations. 4 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. 13 Apparent fractionation ( ) C–CO (‰) app 2 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE ambient enclosures, net production averaged 0.09 and ranged zone may potentially contribute B20% of mean summertime from 0.04 to 0.12 mmol m per day, and is in close agreement CH diffusive fluxes, approximately the same magnitude as with estimates of methanogenesis in Lake Stechlin, where MOX ebullitive fluxes measured for the entire lake. Careful extrapola- 18,19 was shown to be extremely low throughout the epilimnion . tion of our absolute emissions rates (Fig. 1b, Table 1) to other The fact that our approximation of pelagic methanogenesis based ecosystems should be made, as previously overlooked, system- on mass balance (Fig. 1b; Table 1) is in good agreement with specific differences in gas exchange coefficients (k ) and non- CH 18,19 in vitro estimates is very promising, and suggests that both Fickian fluxes will introduce error to conventional cross-system approaches could be incorporated into future detailed studies of flux comparisons (see Supplementary Note 3 for extended CH dynamics for other systems. coverage of this potential problem). Although oxic water Our conclusion that acetate consumption supplied most CH methanogenesis contributed B4% to Lac Cromwell’s summer- in the enclosures, based on estimates of a , is particularly time greenhouse gas footprint, it is conceivable that this pathway app sensitive to changes in the rate of MOX. Given the potential could have greater relative importance in more nutrient-rich, variability in MOX discussed above, estimates of a may be productive ecosystems, and in hardwater environments where app biased by our limited MOX measurement. However, scenario low to negative (that is, net CO uptake) CO emissions are 2 2 42–44 analyses estimating a from extreme hypothetical MOX rates common . app (see Methods for calculations) were all well within the range of Within Lac Cromwell, it is interesting to note that oxic water acetoclastic dominance (Fig. 3a). This supports our conclusion methanogenesis contributed a significant component of whole- that acetate supplied the majority of CH across our treatments, lake emissions, despite the fact that it is a small (0.1 km ), likely for the duration of the experiment (Fig. 3b). Consistent relatively shallow (B3.5 m mean depth) lake with an extensive with this conclusion, observations of enriched surface d C-CH littoral area, and a CH -rich anoxic hypolimnion, all of which are 4 4 8,10,18,29,32 in numerous lakes have been hypothesized to typically major sources of CH . A corollary is that in large, deep indicate dominance of acetoclastic methanogenesis in oxic lakes and open ocean sites where surface waters are increasingly freshwater pelagic zones , and both known acetoclastic genera decoupled from benthic or littoral sources of CH , surface CH 4 4 (Methanosaeta and Methanosarcina) have recently been detected concentrations should be mostly driven by pelagic methanogen- 19,26 27,28 in oxic freshwater and terrestrial environments. esis and therefore a function of algal biomass and metabolism. In Furthermore, acetoclastic dominance was enhanced with the this regard, whereas chlorophyll a does not explain the large-scale addition of nutrients and increased algal production (Fig. 3a). patterns in surface CH across lakes in general , the relationship This pattern parallels that from anoxic environments, where we observed between surface water CH and chlorophyll a in increased abundance of biologically young, high-quality algal our isolated enclosures (log [CH ] ¼ 0.46log [chl a] þ 0.68, 10 4 10 material promotes acetoclastic over hydrogenotrophic r ¼ 0.68, n ¼ 4) generally agrees with observations specifically methanogenesis, the latter instead supported by lower quality, from large lakes, extending all the way to ultraoligotrophic open aged terrestrial DOC . Finally, cyanobacteria were near absent in ocean regions (Fig. 4; log [CH ] ¼ 0.99log [chl a]  1.63, 10 4 10 the mixed layer of our enclosures (o3.5% relative abundance, r ¼ 0.73). This continuity suggests that algal-linked pelagic as biomass, estimated from one enclosure per treatment on CH production may explain much of the ambient CH 4 4 day 7 in all treatments), ruling out major contributions of dynamics observed in large and deep aquatic ecosystems, diazotroph-derived water-column H for hydrogenotrophic regardless of the fact that different mechanisms have been methanogenesis , and cyanobacterial methanogenesis during identified as potentially important in marine (that is, 23 15,16,23–25 DOM demethylation . Taken together, it appears that methylphosphonate or methanethiol catabolism) acetoclastic methanogenesis linked to in-situ, algal DOC versus freshwater (hydrogenotrophic or acetoclastic methano- 18,19 production, potentially plays a central role in supporting water genesis; Fig. 3) . There are clearly a number of potential column CH production in oxic freshwaters. sources of CH in oxic pelagic waters, and although it remains 4 4 The potential importance of acetate for oxic water methanogen- unknown how each pathway differs in relative importance across esis is intriguing, as acetate concentrations are typically extremely these diverse aquatic environments, our results highlight a 18,33–35 low in oxic surface waters , and the presence of oxygen may consistent relationship between open-water phytoplankton inhibit acetate consumption . Yet past reports based on dynamics and the regulation of oxic pelagic CH . radioactive substrate additions have unequivocally demonstrated Taken together, our findings suggest that pelagic CH production extremely rapid turnover of acetate in oxic environments across in oxic environments may be widespread, and likely supports a 33–36 fresh and marine surface waters , suggesting active production baseline evasion of CH from all oxic water columns. In small and of acetate from multiple metabolic pathways in oxic shallow ecosystems, such as Lac Cromwell, this pathway will be 33–36 environments, as well as very efficient uptake .Fermentative quantitatively significant but secondary relative to anoxic littoral and metabolism by diverse and widespread facultative anaerobic benthic sources. In large and deep aquatic environments, however, bacteria in particle-associated microanoxic zones could be the this pathway could become the single dominant source fueling CH 38–40 main source of water column acetate , and such microhabitats supersaturation and fluxes to the atmosphere. It should be could be directly associated to algae or to particles ,yet in vitro emphasized that consideration of oxic water column methanogenesis work as well as our own results further suggest that methanogenesis does not necessarily lead to higher flux estimates for lakes, as the 18,19 proceeds even when these potential sites are removed .One CH derived from this process is already included in routine 2– possibility that should be further explored is that fermentative measurements of surface water pCH and CH surface fluxes 4 4 7,32,46 bacteria themselves create the conditions for anoxic . Incorporating the origin of the CH that outfluxes to the methanogenesis through syntrophic interactions with acetoclastic atmosphere from these lakes, however, is critical to our methanogens in mixed cell aggregates . understanding of the regulation of these fluxes and our capacity to Is oxic water methanogenesis a significant component of the predict their future change. In this regard, there are potentially large CH budget of Lac Cromwell? We have combined our estimated global implications for algal-driven oxic-water methanogenesis. CH 4 4 2–4 rates of oxic water methanogenesis and CH emissions in the emissions from surface waters, particularly from freshwaters ,are ambient enclosures with measurements of diffusive and ebullitive a major element of the global atmospheric CH budget, and we CH fluxes from the surrounding lake waters (summarized in suggest here that the algal-mediated baseline flux is not only a major Table 1), and conclude that methanogenesis in the oxic pelagic contributor to these overall aquatic CH emissions, but also one that NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 5 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 1,000 Enclosures - NP Enclosures - C Enclosures - DOC Enclosures - DOC+NP Lake Baikal (Southern basin) Lake Biwa Lake Constance (centre) Lake Constance (narrows) Canadian Boreal Lakes North Pacific Sub-Tropical Gyre Western North Pacific Ocean Arabian Sea Arctic Ocean 1 10 100 1,000 10,000 100,000 –1 Chlorophyll a (ng l ) Figure 4 | Linking dissolved CH and algal biomass across diverse open-water aquatic ecosystems. Here we combine experimental data from this study with literature-derived measurements from large (410 km ), deep (410 m maximum depth) lakes and marine environments where both CH and chlorophyll a samples were simultaneously measured. A strong positive relationship between CH and chlorophyll a exists across systems: (a) averaged data ([log x] ¼ 0.993[log y]  1.630, r ¼ 0.73, Po0.0001), where error bars represent 1 s.e.m.; (b) raw data ([log x] ¼ 0.855[log y]  1.166, 10 10 10 10 r ¼ 0.72, Po0.0001). Guidelines for meta-analysis are discussed in the Methods section, and raw data are available in Supplementary Table 3. is particularly sensitive to environmental change. As such, constant corrected for temperature. On day 7, dark incubations to estimate MOX using unfiltered surface water were conducted in the laboratory at room widespread and intensifying human- and climate-driven changes 47–49 50,51 temperature, lasting 96 h (Supplementary Fig. 4). Incubations began at in pelagic nutrient availability , terrestrial DOC inputs and approximately noon on 18 June, day 6 of the experiment; samples for measurement 47,48,51 physical structure of the water column , which strongly shape of MOX were collected from one mesocosm per treatment, at 25 cm depth, into 47,51–54 aquatic algal dynamics , may have major, but previously 12 ml precombusted vials. Samples were capped to exclude any air with a gastight, 3-mm-thick, butyl rubber-lined plastic cap, then incubated in the dark at 20 C. unconsidered, consequences for oxic water methanogenesis and Ambient CH measured in the mesocosms was used as a time 0 concentration, and aquatic CH emissions. samples were subsequently killed every 24 h by injection of HgCl , and stored at 4 C in the dark until analysis. Upon analysis, 5 ml of water was displaced with ambient air, and the remaining sample was equilibrated with the air by physically Methods shaking the vials. Headspace equilibration and gas chromatography analysis were Study site. Lac Cromwell is located at the Station de Biologie des Laurentides, performed as for ambient CH . Respiration rates were determined by dark in situ, a field research facility of the Universite´ de Montre´al. The lake turns over in 24 h incubations on experiment days 6 and 7 in all enclosures, using unfiltered spring and fall, is relatively shallow and small (mean depth ¼ 3.5 m, maximum 2 5 3 water in 4-l cubitainers. Initial and final samples of DO were collected and depth ¼ 9.5 m, surface area of 0.11 km , volume of 3  10 m ), and is oligo- 55 measured by membrane inlet mass spectrometry . mesotrophic, with little phytoplankton biomass (B6.0mgof chl a per litre) . CH chamber flux estimates. Air–water CH fluxes in the mesocosms and the 4 4 Experimental design. Here, 12, 1 m diameter, 6 m deep, 4712-L polyethylene surrounding lake were calculated by combining the measured surface water pCH enclosures were attached to floating wooden frames anchored in the middle of the and the CH gas exchange coefficient (k ), the latter derived from the measured 4 CH4 lake (B8 m depth). On 5 June 2012, bags were filled by pumping epilimnetic water gas exchange coefficient for CO (k ) in mesocosms supersaturated in CO . k 2 CO2 2 CO2 sequentially across two mesh screens of 110 and 45 mm to remove zooplankton, was measured in one mesocosm per treatment, 3–5 times daily during days 6 and 7 and three mesocosms each received N þ P, DOC, N þ P þ DOC additions, or no and 27–28 of the experiment, using the floating chamber method following Vachon addition (control). Both KH PO and NaNO additions were made to reach target 58 2 4 3 et al. Briefly, chambers were connected via a closed loop system to an EGM-4 concentrations of 50 and 700 mg P and N per litre, respectively. For DOC, infrared gas monitor (PP Systems). Fluxes were calculated as the rate of change of Superhume brand humic slurry was added to reach a target concentration of 15 mg chamber pCO per min during a 15-min interval. Diel sampling periods were of DOC per litre. For a full review of Superhume properties and applicability to spaced to obtain flux estimates from the morning, afternoon and nighttime periods, freshwater experimentation, see Lennon et al. . The mesocosms were allowed to and mean values were used for daily flux estimates. Dissolved CO samples were stabilize for 1 week before initial sampling and re-stocking zooplankton at ambient taken before all chamber measurements using the same headspace technique as for lake concentrations. Zooplanktons were collected from Lac Cromwell by vertical CH , but with direct injection of the equilibrated gas into the EGM-4 in the field. tows using a 54-mm Nitex net, and immediately released into mesocosms after The gas exchange coefficient for CH was then calculated from k using 4 CO2 each haul. equation (1), following Jahne et al. : 2=3 Limnological measurements. All limnological samples were gathered at B1200 k ¼ k  Sc Sc ð1Þ CH4 CO2 CH4 CO2 hours on days 0, 7 and 28 of the experiment (13, 19 June 2012 and 10 July 2012, respectively; Supplementary Fig. 3). Dissolved oxygen (DO) and temperature were where Sc is the Schmidt number of CO and CH , respectively . To calculate CH 2 4 4 measured at 0.5 m depth in all bags, whereas water column profiles were taken fluxes, we used these average k values, the measured pCH and the temperature- CH 4 from one enclosure per treatment using an Yellow Springs Instruments (YSI) Pro dependent solubility of the gas following equation (2): Plus multiparameter sonde (Supplementary Fig. 1). On all three dates, water was collected at 0.5 m depth, filtered to 0.45mm and chlorophyll a was measured F ¼ k K D ð2Þ gas h gas spectrophotometrically in ethanol filter extracts. Dissolved CH concentrations at 25 cm depth were measured in triplicate for each mesocosm following Prairie and where F is flux of CH (mmol m per day), K is the temperature-corrected 4 h del Giorgio . Partial pressures were converted to concentrations using Henry’s Henry’s constant and D is the difference in partial pressures between the air and gas 6 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. CH (nM) 4 NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE estimates of a , based on this analysis, as upper and lower bounds for a the water partial pressures of CH . We assumed an atmospheric partial pressure of app app estimated using our actual measurements of MOX. CH of 1.75 p.p.m. Mesocosm mass balances. To quantify methanogenesis in the oxic mixed Isotopic analyses. Isotope compositions are reported in delta notation following layer of ambient enclosures, mass balances from the ambient control treatment equation (3): enclosures were constructed (Supplementary Table 2) for comparison with other 13 18 d C or d O ¼ R =R  1 1; 000 ð3Þ lake-based CH fluxes using equation (7): sample standard 13 13 where R is the ratio of heavy to light isotope, d C represents d C–CH or CH ¼ D þ E þ O  M ð7Þ 4 4gross CH4 13 18 18 18 C of dissolved inorganic carbon (DIC), d O represents d O–O or d O–H O 2 2 13 18 and R is the d C or d O signature of Vienna Pee Dee Belemnite or stan- standard Here rates of methanogenesis (CH ) are estimated by summing the rate of 4gross dard mean ocean water, respectively. All samples were taken at B25 cm depth. change in storage (D ), defined as the rate of increase in the ambient CH pool, CH4 4 18 13 Samples for d O–O and d C–CH were stored in 12-ml precombusted bor- 2 4 with the rates of CH evasion (E) and oxidation (O), minus the horizontal influx of osilicate vials, preserved with HgCl , and air-free samples capped with a gastight external CH via diffusion across the walls of the enclosures (M; detailed in the 2 4 rubber-lined plastic cap. d O–O samples were analysed at the University of Supplementary Methods). In all cases, mass balance terms were first standardized Ottawa Stable Isotopes Laboratory, and d C–CH samples were measured at the to the entire mixed layer or surface of the enclosure (as mmol per enclosure per University of Waterloo, both using standard laboratory methods. A single sample day) in order to combine aerial (E and M) and volumetric (D and O) processes CH4 for d O–H O was collected from the lake before filling enclosures, analysed with a on each sampling date. Picarro L230i isotopic water analyser, and assumed representative for all enclosures The enclosure mixed layer average depth of 1.68 m (Supplementary Fig. 1) was for the duration of the experiment. d C-DIC samples were collected in acid- used to calculate both mixed layer volume, and the surface area of the enclosure washed, 40 ml vials with both Teflon and rubber-lined gastight plastic caps, and wall that contributed to CH influx from the surrounding lake environment. The measured at the Colorado Plateau Stable Isotopes Laboratory using standard change in storage (D ) was calculated as the rate of change in ambient CH CH 4 18 18 methods. Day 7 respiration (R), d O–O and d O–H O were used to estimate concentrations in each bag as a function of time using the slope of the regression 2 2 GPP and NEP following Quin˜ones-Rivera et al. line as determined by least-squares regression analysis, then averaged and adjusted volumetrically to the enclosure scale (mmol per enclosure per day; Supplementary Table 2). The rate of MOX (O) for day 7 of the experiment was Numerical methods. We used RM-ANOVA to compare the effects of treatments, used for all calculations. Finally, estimates of CH were converted to units 4gross and treatment by time interactions, on ambient CH . One-way ANOVA was used 4  3 of mmol m per day for comparison with other lake-based measurements to assess differences between treatments for grouped data, and log (x þ 1) (Table 1). transformations applied to maintain the homogeneity of variances, followed by Bonferroni post-hoc pairwise comparisons of treatment means. All regression analyses were performed using Sigmaplot version 12, and ANOVAs were com- Whole-lake CH dynamics. Enclosure measurements and mass balance results puted with SPSS version 16. We used ordinary least squares regression to quantify were compared with the surrounding Lac Cromwell environment. To quantify the link between GPP or NEP and CH evasion. To estimate rates of MOX, the loss 4 summertime rates of ebullition in Lac Cromwell (Table 1), bubble traps were fixed of CH through time was fitted with a polynomial function, the derivative was at five permanent sampling sites along a transect from the littoral to pelagic at B1, taken and the slope at time 0 was calculated (Supplementary Fig. 4). This approach 2, 3, 5 and 7 m depths. Bubble traps were left in place over the entire sampling was deemed more accurate in quantifying the slope at time 0, as traditional period and collected monthly in June, July and August 2012. Traps consisted of an methods, such as log transformations followed by visual selection of the linear inverted funnel (63.5 cm diameter) suspended at 0.5 m below the surface of the portion of the data set , can introduce large errors in the slope of the regression water (0.25 m at the shallowest sample site). A graduated 1-l glass bottle was line when deciding which points to fit and which to exclude. Although MOX attached to the funnel by gluing the sides of the bottle cap to the neck of the funnel was estimated only once, our estimates are within the range of pelagic MOX to make a gastight seal. The bottles were covered with aluminum foil to prevent 3 31,32,62 (0.170–0.250 mmol m per day) previously reported , particularly for overheating or light exposure of the collected gas. Bottles were filled with water boreal lakes of similar trophic condition . upon deployment, so the gas accumulated by displacing water in the bottle. Bottles were collected once a month or until a volume greater than 250 ml of gas was detected. All gas was carefully removed by displacement with ambient water using Apparent isotopic fractionation of methanogenesis. We estimated the apparent a stopcock cap fitted with two syringes. A subset of the gas was injected in 30 ml fractionation factor (a ) during methanogenesis following Conrad . The app saline vials, and analysed upon return to the laboratory by gas chromatographic apparent fractionation factor is defined using equation (4) as analysis as detailed for enclosure samples. As the majority of samples settled in the 13 3 13 3 bottles for up to a month before collection, we assumed that the composition of a ¼ d C-CO þ 10 = d C-CH þ 10 ð4Þ app 2 4source CH in bubbles was 0.6 atm or 60%, based on a subset of samples analysed that 13 13 where d C–CH is the isotopic signature of source CH ,and d C–CO was were collected within days of sedimentary release. This assumption is consistent 4source 4 2 13 63 with literature estimates of bubble CH concentration . calculated from measured d C-DIC following the study by Stumm and Morgan , 4 using fractionation factors from the study by Mook et al. . As we measured Diffusive evasion of lake CH was estimated by first measuring concentrations 13 13 13 of CH on a monthly basis near each bubble trap site along the transect. ambient d C–CH (d C–CH ), we estimated d C–CH by correcting 4 4 4ambient 4source Concentrations ranged between 0.63 and 2.09 with an average of 1.21mM, and did for the isotopic fractionation effects of MOX and evasion to the atmosphere using not vary across sites (ANOVA, P ¼ 0.98, F ¼ 0.10). As detailed above for the an open-system, steady-state model for isotope fractionation , as defined in enclosure mass balance, atmospheric fluxes were estimated by applying the average equation (5): 24 h gas exchange coefficient measured at 7 m depth along the transect (mean 13 13 d C-CH ¼f d C-CH  D k ¼ 0.65 m per day, Supplementary Fig. 2). 4source 4ambient evasion CH ð5Þ Whole-lake CH and CO diffusive fluxes were calculated by applying average 4 2 þðÞ 1  f d C-CH  D 4ambient MOX areal estimates of diffusion to the entire lake surface area (102,000 m ). As ebullition was only detected at depths o3 m, average rates of ebullition where f and 1  f are the fractions of CH loss, standardized to daily rates for the (0.31 0.66 mmol m per day) were applied to the lake surface o3 m in depth entire enclosure, estimated, respectively, as evasion and MOX rates divided by the sum of each. D is the isotopic effect of each loss pathway. In cases such as ours (49,166 m ). This area was estimated by determining the area of the lake o3m in depth from a hypsographic curve for Lac Cromwell , and existing bathymetric where Do100%, D can be approximated using equation (6): information for Lac Cromwell . D ¼ðÞ 1 a 10 ð6Þ where a is the fractionation factor of a given reaction. Literature-derived values of a Data compilation for meta-analysis. To assess the relationship between algal were used for both loss pathways. For evasion, an a value of 0.9992 was used , and dynamics and CH in open water environments, we compared the results from our 62 2 for MOX, the range reported by Bastviken et al. (0.9816–0.9792) was used. experimental enclosures with directly obtained results from large (410 km ), deep To better constrain the potential range of a estimated on day 7 of our (410 m maximum depth) boreal lakes and other freshwater and marine data app experiment, we performed a scenario analysis by varying combinations of MOX from the literature (raw data and selection methods tabulated in Supplementary rates and isotopic fractionation values, to determine the sensitivity of our estimates Table 3). Literature data include only studies where CH and chlorophyll a were of a to potential variability in MOX. For this analysis, we chose two very measured simultaneously, except for Lake Baikal, where detailed studies of CH app 4 conservative, extreme (for oxic pelagic freshwaters) values in MOX rates (0 and and chlorophyll a were measured independently, but overlapped temporally and 3  1 0.05 mmol m h ) and paired each rate with both minimum and maximum thus were included. To remain consistent with the sampling approach from our values reported for isotopic fractionation during MOX (18.4 and 20.8%). study, and sampling of boreal lakes, only near-surface, open-water results from the Because the MOX rate of zero did not have associated isotopic fractionation, we centre of the lake or furthest from the coastal environment (in marine studies) were calculated three different combinations of extreme parameters: no oxidation with used. Summertime results were taken if seasonal data were presented. If surface no fractionation, high oxidation with low fractionation and high oxidation rates water results were given as a range, the midpoint of that range was chosen. If with high associated fractionation. We then took the maximum and minimum multiple samples were taken at one site, then results were averaged. NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 7 & 2014 Macmillan Publishers Limited. All rights reserved. ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 References 30. Conrad, R. Quantification of methanogenic pathways using stable carbon 1. Solomon, S. et al. in Climate Change 2007: The Physical Science Basis. isotopic signatures: A review and a proposal. Org. Geochem. 36, 739–752 Contribution of Working Group I to the Fourth Assessment Report of the (2005). Intergovernmental Panel on Climate Change. (eds Solomon, S. et al.) 31. Utsumi, M. et al. Oxidation of dissolved methane in a eutrophic, shallow lake: (Cambridge Univ., 2007). Lake Kasumigaura, Japan. Limnol. Oceanogr. 43, 471–480 (1998). 2. Bastviken, D., Tranvik, L. J., Downing, J. A., Crill, P. M. & Enrich-Prast, A. 32. Bastviken, D., Cole, J. J., Pace, M. L. & Van de Bogert, M. C. Fates of methane Freshwater methane emissions offset the continental carbon sink. Science 331, from different lake habitats: Connecting whole-lake budgets and CH 50 (2011). emissions. J. Geophys. Res. 113, G02024 (2008). 3. Michmerhuizen, C. M., Striegl, R. G. & Mcdonald, M. E. Potential methane 33. Wright, R. R. & Hobbie, J. E. Use of glucose and acetate by bacteria and algae in emission from north-temperate lakes following ice melt. Limnol. Oceanogr. 41, aquatic ecosystems. Ecology 47, 447–464 (1966). 985–991 (1996). 34. Allen, H. L. Acetate in fresh water: Natural substrate concentrations determined 4. Kirschke, S. et al. Three decades of global methane sources and sinks. Nat. by dilution bioassay. Ecology 49, 346–349 (1968). Geosci. 6, 813–823 (2013). 35. Ho, T.-Y. et al. Acetate cycling in the water column of the Cariaco Basin: 5. Prairie, Y. T. & del Giorgio, P. A. A new pathway of freshwater methane Seasonal and vertical variability and implication for carbon cycling. Limnol. emissions and the putative importance of microbubbles. Inl. Waters 3, 311–320 Oceanogr. 47, 1119–1128 (2002). (2013). 36. Sepers, A. The utilization of dissolved organic compounds in aquatic 6. Rudd, J. W. M. & Hamilton, R. D. Methane cycling in a eutrophic shield lake and environments. Hydrobiologia 52, 39–54 (1977). its effects on whole lake metabolism. Limnol. Oceanogr. 23, 337–348 (1978). 37. Thauer, R. K., Mo¨ller-Zinkhan, D. & Spormann, A. M. Biochemistry of acetate 7. Bastviken, D., Cole, J., Pace, M. & Tranvik, L. Methane emissions from lakes: catabolism in anaerobic chemotrophic bacteria. Annu. Rev. Microbiol. 43, dependence of lake characteristics, two regional assessments, and a global 43–67 (1989). estimate. Global Biogeochem. Cycles 18, GB4009 (2004). 38. Riemann, L. & Azam, F. Widespread N-acetyl-d-glucosamine uptake among 8. Murase, J., Sakai, Y., Sugimoto, A., Okubo, K. & Sakamoto, M. Sources of pelagic marine bacteria and its ecological implications. Appl. Environ. dissolved methane in Lake Biwa. Limnology 4, 91–99 (2003). Microbiol. 68, 5554–5562 (2002). 9. Schmidt, U. & Conrad, R. Hydrogen, carbon monoxide, and methane dynamics 39. Alonso, C. & Pernthaler, J. Incorporation of glucose under anoxic conditions by in Lake Constance. Limnol. Oceanogr. 38, 1214–1226 (1993). bacterioplankton from coastal North Sea surface waters. Appl. Environ. 10. Schulz, M., Faber, E., Hollerbach, A., Schroder, H. G. & Gude, H. The methane Microbiol. 71, 1709–1716 (2005). cycle in the epilimnion of Lake Constance. Arch. Hydrobiol. 151, 157–176 (2001). 40. Tada, Y. & Grossart, H.-P. Community shifts of actively growing lake bacteria 11. Hofmann, H. Spatiotemporal distribution patterns of dissolved methane in after N-acetyl-glucosamine addition: Improving the BrdU-FACS method. ISME lakes: How accurate are the current estimations of the diffusive flux path? J. 8, 1–14 (2013). Geophys. Res. Lett. 40, 2779–2784 (2013). 41. Stams, A. J. M. & Plugge, C. M. Electron transfer in syntrophic 12. Schmid, M. et al. Sources and sinks of methane in Lake Baikal: A synthesis of communities of anaerobic bacteria and archaea. Nat. Rev. Microbiol. 7, measurements and modeling. Limnol. Oceanogr. 52, 1824–1837 (2007). 568–577 (2009). 13. Scranton, M. I. & Brewer, P. G. Occurrence of methane in the near-surface 42. Finlay, K., Leavitt, P. R., Wissel, B. & Prairie, Y. T. Regulation of spatial and waters of the western subtropical North-Atlantic. Deep. Sea Res. 24, 127–138 temporal variability of carbon flux in six hard-water lakes of the northern Great (1977). Plains. Limnol. Oceanogr. 54, 2553–2564 (2009). 14. Owens, N. J. P., Law, C. S., Mantoura, R. F. C., Burkill, P. H. & Llewellyn, C. A. 43. Raymond, P. A. et al. Global carbon dioxide emissions from inland waters. Methane flux to the atmosphere from the Arabian Sea. Nature 354, 293–296 Nature 503, 355–359 (2013). (1991). 44. McDonald, C. P., Stets, E. G., Striegl, R. G. & Butman, D. Inorganic carbon 15. Damm, E. et al. Methane production in aerobic oligotrophic surface water in loading as a primary driver of dissolved carbon dioxide concentrations in the the central Arctic Ocean. Biogeosciences 7, 1099–1108 (2010). lakes and reservoirs of the contiguous United States. Global Biogeochem. Cycles 16. Karl, D. M. et al. Aerobic production of methane in the sea. Nat. Geosci. 1, 27, 285–295 (2013). 473–478 (2008). 45. Rasilo, T., Prairie, Y. T. & del Giorgio, P. A. Large-scale patterns in summer 17. Tilbrook, B. D. & Karl, D. M. Methane sources, distributions and sinks from diffusive CH fluxes across boreal lakes, and contribution to diffusive C California coastal waters to the oligotrophic North Pacific gyre. Mar. Chem. 49, emissions. Glob. Chang. Biol. (In the press). 51–64 (1995). 46. Ortiz-Llorente, M. J. & Alvarez-Cobelas, M. Comparison of biogenic methane 18. Tang, K. W., McGinnis, D. G., Frindte, K., Bru¨chert, V. & Grossart, H.-P. emissions from unmanaged estuaries, lakes, oceans, rivers and wetlands. Atmos. Paradox reconsidered: Methane oversaturation in well-oxygenated lake waters. Environ. 59, 328–337 (2012). Limnol. Oceanogr. 59, 275–284 (2014). 47. Schindler, D. W. Recent advances in the understanding and management of 19. Grossart, H.-P., Frindte, K., Dziallas, C., Eckert, W. & Tang, K. W. Microbial eutrophication. Limnol. Oceanogr. 51, 356–363 (2006). methane production in oxygenated water column of an oligotrophic lake. Proc. 48. Moss, B. Cogs in the endless machine: Lakes, climate change and nutrient Natl Acad. Sci. USA 108, 19657–19661 (2011). cycles: a review. Sci. Total Environ. 434, 130–142 (2012). 20. De Angelis, M. A. & Lee, C. Methane production during zooplankton grazing 49. Carpenter, S. R. et al. Nonpoint pollution of surface waters with phosphorus on marine phytoplankton. Limnol. Oceanogr. 39, 1298–1308 (1994). and nitrogen. Ecol. Appl. 8, 559–568 (1998). 21. Oremland, R. S. Methanogenic activity in plankton samples and fish intestines: 50. Monteith, D. T. et al. Dissolved organic carbon trends resulting from changes A mechanism for in situ methanogenesis in oceanic surface waters. Limnol. in atmospheric deposition chemistry. Nature 450, 537–540 (2007). Oceanogr. 24, 1136–1141 (1979). 51. Schindler, D. W. et al. Climate-induced changes in the dissolved organic carbon 22. Ditchfield, A. et al. Identification of putative methylotrophic and budgets of boreal lakes. Biogeochemistry 36, 9–28 (1997). hydrogenotrophic methanogens within sedimenting material and copepod 52. Taranu, Z. E., Zurawell, R. W., Pick, F. & Gregory-Eaves, I. Predicting faecal pellets. Aquat. Microb. Ecol. 67, 151–160 (2012). cyanobacterial dynamics in the face of global change: The importance of scale 23. Beversdorf, L. J., White, A. E., Bjo¨rkman, K. M., Letelier, L. M. & Karl, D. M. and environmental context. Glob. Chang. Biol. 18, 3477–3490 (2012). Phosphonate metabolism by Trichodesmium IMS101 and the production of 53. Carpenter, S. R., Cole, J. J., Kitchel, J. F. & Pace, M. L. Impact of dissolved greenhouse gases. Limnol. Oceanogr. 55, 1768–1778 (2010). organic carbon, phosphorus, and grazing on phytoplankton biomass and 24. Carini, P., White, A. E., Campbell, E. O. & Giovannoni, S. J. Methane production in experimental lakes. Limnol. Oceanogr. 43, 73–80 (1998). production by phosphate-starved SAR11 chemoheterotrophic marine bacteria. 54. Jones, R. I. The influence of humic substances on lacustrine planktonic food Nat. Commun. 5, 4346 (2014). chains. Hydrobiologia 229, 73–91 (1992). 25. del Valle, D. A. & Karl, D. M. Aerobic production of methane from dissolved 55. Den Heyer, C. & Kalff, J. Organic matter mineralization rates in sediments: methylphosphonate in the water column and sinking particles collected in the A within- and among-lake study. Limnol. Oceanogr. 43, 695–705 (1998). North Pacific Subtropical Gyre. Aquat. Microb. Ecol 73, 93–105 (2014). 56. Lennon, J. T. et al. A source of terrestrial organic carbon to investigate the 26. Paganin, P. et al. Vertical distribution of bacterioplankton in Lake Averno in browning of aquatic ecosystems. PLoS One 8, e75771 (2013). relation to water chemistry. FEMS Microbiol. Ecol. 84, 176–188 (2013). 57. Bouvier, T. C. & del Giorgio, P. A. Compositional changes in free-living 27. Angel, R., Matthies, D. & Conrad, R. Activation of methanogenesis in arid bacterial communities along a salinity gradient in two temperate estuaries. biological soil crusts despite the presence of oxygen. PLoS One 6, e20453 (2011). Limnol. Oceanogr. 47, 453–470 (2002). 28. Angel, R., Claus, P. & Conrad, R. Methanogenic archaea are globally ubiquitous 58. Vachon, D., Prairie, Y. T. & Cole, J. J. The relationship between near-surface in aerated soils and become active under wet anoxic conditions. ISME J. 6, turbulence and gas transfer velocity in freshwater systems and its implications 847–862 (2012). for floating chamber measurements of gas exchange. Limnol. Oceanogr. 55, 29. Miyajima, T. et al. Distribution of greenhouse gases, nitrite, and Cof 1723–1732 (2010). dissolved inorganic carbon in Lake Biwa: Implications for hypolimnetic 59. Jahne, B. et al. On the parameters influencing air-water gas exchange. metabolism. Biogeochemistry 36, 205–221 (1997). J. Geophys. Res. 92, 1937–1949 (1987). 8 NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications & 2014 Macmillan Publishers Limited. All rights reserved. NATURE COMMUNICATIONS | DOI: 10.1038/ncomms6350 ARTICLE ´ ´ biologie des Laurentides de l’Universite de Montreal provided logistical support. Terhi 60. Wanninkhof, R. Relationship between wind speed and gas exchange over the Rasilo and Tonya Del Sontro revised an early version of the manuscript. M.J.B. was ocean. J. Geophys. Res. 97, 7373–7382 (1992). supported by a doctoral scholarship from the National Science and Engineering Research 61. Quin˜ones-Rivera, Z. J., Wissel, B. & Justic´, D. Development of productivity Council of Canada (NSERC). This project is part of the programme of the NSERC/HQ models for the northern Gulf of Mexico based on oxygen concentrations and Industrial Research Chair in Carbon Biogeochemistry in Boreal Aquatic Systems stable isotopes. Estuaries Coasts 32, 436–446 (2009). (CarBBAS), and was co-funded by grants from NSERC (to P.A.d.G. and A.M.D.), Hydro- 62. Bastviken, D., Ejlertsson, J. & Tranvik, L. Measurement of methane oxidation in Que´bec (to P.A.d.G.) and a Fonds de recherche Que´bec–Nature et Technologie award lakes: A comparison of methods. Environ. Sci. Technol. 36, 3354–3361 (2002). from the government of Quebec (to A.M.D.). 63. Stumm, W. & Morgan, J. J. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters (Wiley, 1996). 64. Mook, K. A., Bommerson, J. C. & Staverman, W. H. Carbon isotope Author Contributions fractionation between dissolved bicarbonate and gaseous carbon dioxide. Earth M.J.B, P.A.d.G. and A.M.D. designed the study; M.J.B., L.B., M.C.G.C. and A.M. per- Planet. Sci. Lett. 22, 169–176 (1974). formed field and lab work; M.J.B. and Y.T.P. analysed the data; M.J.B. and P.A.d.G. wrote 65. Fry, B. Stable Isotope Ecology (Springer, 2006). the paper; all authors revised the paper; A.M.D. and P.A.d.G. contributed materials. 66. Knox, M., Quay, P. D. & Wilbur, D. Kinetic isotopic fractionation during air-water gas transfer of O ,N ,CH , and H . J. Geophys. Res. 97, 335–343 2 2 4 2 (1992). Additional information 67. Livingstone, D. M. & Imboden, D. M. The prediction of hypolimnetic oxygen Supplementary Information accompanies this paper at http://www.nature.com/ profiles: a plea for a deductive approach. Can. J. Fish. Aquat. Sci. 53, 924–932 naturecommunications (1996). Competing financial interests: The authors declare no competing financial ´ ´ 68. Carignan, R Station de biologie des Laurentides–Universite de Montreal http:// interests. www.sbl.umontreal.ca/territoire-cartes/cartes/index.html (2010). Reprints and permission information is available online at http://npg.nature.com/ Acknowledgements reprintsandpermissions/ Simon Gauthier-Fautaux, Juan Pablo Nino Garcia, Cynthia Soued, Marilyne Robidoux and Ryan Hutchins provided field and laboratory assistance. Jean-Francois Lapierre, How to cite this article: Bogard, M. J. et al. Oxic water column methanogenesis Dominic Vachon, Adam Heathcote and David Bastviken provided conceptual and as a major component of aquatic CH fluxes. Nat. Commun. 5:5350 methodological advice. Annick St. Pierre, Alice Parks and the employees of the Station de doi: 10.1038/ncomms6350 (2014). NATURE COMMUNICATIONS | 5:5350 | DOI: 10.1038/ncomms6350 | www.nature.com/naturecommunications 9 & 2014 Macmillan Publishers Limited. All rights reserved.

Journal

Nature Communications – Springer Journals

Published: Oct 30, 2014

There are no references for this article.