Photosynthetic isotope biosignatures in laminated micro-stromatolitic andnon-laminated nodules associated with modern, freshwater microbialites inPavilion Lake, B.C.
A.L. Brady h, G.F. Slater a,⁎, C.R. Omelon b, G. Southam b, G. Druschel c, D.T. Andersen d,I. Hawes e, B. Laval f, D.S.S. Lim d,g
a School of Geography and Earth Sciences, McMaster University, 1280 Main Street West, Hamilton, ON, Canada L8S 4K1b Department of Earth Sciences, Biology and Geological Sciences Building, The University of Western Ontario 1151 Richmond Street, London, ON, Canada N6A 5B7c Department of Geology, Delehanty Hall, University of Vermont, Trinity Campus, 180 Colchester Avenue, Burlington, VT 05405-1758, United Statesd Carl Sagan Center for the Study of Life in the Universe, SETI Institute, 515 North Whisman Road, Mountain View, CA 94043, United Statese Aquatic Research Solutions Ltd., Cambridge, New Zealandf Department of Civil Engineering, University of British Columbia, Civil and Mechanical Engineering Building, 2002-6250 Applied Science Lane, Vancouver, B.C. Canada V6T 1Z4g NASA Ames Research Center, Moffett Field, CA 94035, United Statesh Department of Biological Sciences, University of Calgary, 2500 University Drive N.W., Calgary, Alberta, Canada T2N 1N4
a b s t r a c ta r t i c l e i n f o
Article history:Received 13 October 2009Received in revised form 19 March 2010Accepted 19 March 2010
The influence of microbial activity on carbonate precipitation was investigated within micro-stromatoliticnodules associatedwith modern, freshwatermicrobialites located in Pavilion Lake, B.C. Observed carbonate δ13Cvalues enrichedbyup to+3.6‰ as compared topredicted abiotic carbonate δ13C values frommeasureddissolvedinorganic carbon (mean −1.2‰, n=13) were consistent with microbial photosynthetic influence on in situprecipitation within the nodule microenvironment. Estimated carbonate precipitation temperatures within thenodules based on δ18O were consistent with recorded summertime temperatures, indicative of precipitationduring the period of highest levels of photosynthetic activity. Low δ13C values of organic matter within thenodules (−30.6 to −21.1‰) and an average inorganic to organic carbon Δδ13C value of 26.8‰ reflected thepreferential uptake of 12C during non-CO2 limitedphotosynthesis, supporting the generation of 13C-enrichedDIC.Microelectrode profiles through the nodules showed oxygen supersaturation of up to ∼275%, elevated pHcompared to ambient water and a lack of any observable dissolved sulphide, Mn or Fe further indicated thatphotosynthetic activity was the predominant metabolic process within the nodule during light exposure.Microbial phospholipid fatty acid profiles of the nodule communities were indicative of bacteria rather thaneukaryotes and PLFA δ13C values were depleted relative to the bulk cell by 2.6–6.6‰, consistent with apredominance of photosynthetic microbes. Scanning electron microscopy images of the relationship betweencarbonate minerals and filaments indicated that carbonate precipitation had occurred in situ due to microbialinfluences on the geochemistry within the nodule microenvironment rather than due to cell surface effects ortrapping and binding. The observation of photosynthetically induced 13C-enrichment of in situ precipitatedcarbonatewithin thenodulemicroenvironment is thus a biosignatureof the activity of these surface communitiesand is consistent with the hypothesized role of biology in the formation of microbialites.
Resolving biological signatures (biosignatures) from abiotic signa-tures of carbonate formation remains an important component ofinterpretation of the early rock record and the debate concerning thetimingof the rise of life (Kempeet al., 1991; Awramik, 1992; Lowe, 1994;Walter, 1996; Merz-Preiß and Riding, 1999). Differentiating betweenbiotic and abiotic processes of carbonate formation on the basis of
mineralogical and morphological data remains challenging and un-doubtedly some systems comprise both biotic and abiotic mechanisms(Chafetz andGuidry, 1999). Putativemicrofossils and stromatolites havebeen cited as evidence for early life (Tyler and Barghoorn, 1954; Schopf,1993), however clear evidence for the biologic origin of such fossils isdebated (Grotzinger and Rothman, 1996; Brasier et al., 2002).Stromatolites are laminated lithified structures that are consideredamongst the earliest purported evidence for life on Earth (see Riding,2000 for overview). “Microbialite” is a more general term used todescribe organo-sedimentary structures, including stromatolites,formed through the trapping and binding of sediment and/or calcifica-tion of microbes resulting in a layered fabric (Burne and Moore, 1987).
The three main hypotheses regarding biogenic mechanisms of micro-bialite formation are; 1) trapping and binding of sediment by microbialcommunities (Burne and Moore, 1987; Reid et al., 2000); 2) microbialcell surfaces acting as nucleation sites for crystal growth (Schultze-Lamet al., 1996; Bosak and Newman, 2003) and/or 3) promotion ofcalcification via alteration of the local geochemical environment throughmetabolic activity (Thompson and Ferris, 1990; Merz-Preiß and Riding,1999; Merz-Preiß, 2000; Stolz et al., 2001). Alternatively, abioticprocesses resulting from changes in geochemical conditions and/orsedimentation patterns have also been proposed as mechanisms for theformation of microbialites (Council and Bennett, 1993; Grotzinger andRothman, 1996; Walter, 1996).
Modern microbialites have been discovered in a number ofdifferent environments and locations around the world includingAustralia (Logan, 1961; Awramik and Riding, 1988), the Bahamas(Reid et al., 1995; Visscher et al., 1998; Andres et al., 2006), Mexico(Breitbart et al., 2009), Lake Van (Kempe et al., 1991) and Canada(Ferris et al., 1997; Laval et al., 2000). These modern microbialiteshave been suggested as analogues to ancient structures andopportunities to gain insight into mechanisms of their formationand identification of microbial biosignatures. The discovery ofmicrobialites in association with photosynthetic microbial communi-ties in freshwater Pavilion Lake, British Columbia, Canada (50°51′ N,121°44′ W) presents one such opportunity. Our study focuses onunderstanding the potential for biological influences on carbonateprecipitation and associated isotope biosignatures within laminatedmicro-stromatolitic and non-laminated nodular microbial communi-ties that occur on the surface of the Pavilion Lake microbialites andthat may be preserved in the geologic record.
1.1. Microbial influences on carbonate precipitation and isotopiccomposition
If biology is playing a role in the formation of microbialitestructures, it is often unclear whether this is through trapping andbinding of externally precipitated material or through directmicrobial influence causing in situ precipitation. Of the threeproposed mechanisms of biological formation of microbialites, thepotential for microbial metabolic activity to change local aqueousgeochemistry and promote the precipitation of calcium carbonateminerals (Merz, 1992; Shiraiwa et al., 1993; Shiraishi et al., 2008)represents the greatest potential for the generation of a biosignature.These microbially induced changes in geochemistry, and anyassociated biosignatures, can be substantial enough to result insystem-wide effects in lacustrine environments (Hollander andMcKenzie, 1991; Thompson et al., 1997). Alternatively, microbialeffects on geochemical conditions may be restricted to microenvir-onments associated with the cells, particularly if these cells aregenerating a distinct microenvironment, as occurs in microbial mats(Jørgensen et al., 1983; Revsbech et al., 1983; de Beer et al., 1997;Andres et al., 2006). Microbial influences on the concentration andisotopic composition of the dissolved inorganic carbon (DIC) andprecipitated carbonate occur during both autotrophic and hetero-trophic metabolisms. Both of these processes can generate biosigna-tures that may be incorporated into carbonate and preserved in thegeologic record (Merz, 1992; Andres et al., 2006; Breitbart et al.,2009).
In most natural environments, the pH is high enough (∼7–8) thatmost dissolved CO2 is in the form of bicarbonate (HCO3
−). Duringphotosynthesis,HCO3
− is taken into the cell, converted intoCO2 andOH−
by the enzyme carbonic anhydrase for use in the ribulose-1,5-bispho-sphate carboxylase enzyme (RUBISCO) (Miller and Colman, 1980;Paneth and O'Leary, 1985; Thompson and Ferris, 1990). The CO2 isincorporated into the cell biomass while the OH− is expelled, resultingin an increased pH in the microenvironment around the cell andtherefore a shift towards a higher CO3
2− concentration. This shift can
result in a corresponding increase in the calcium carbonate saturationindex (SI) leading to precipitation (Revsbechet al., 1983; Thompson andFerris, 1990;Merz, 1992; Schultze-Lam et al., 1992; Ludwig et al., 2005).
Heterotrophic metabolic activity may also have significant butcontrasting influence on carbonate precipitation. Sulfate-reducingbacteria (SRB) oxidize 13C-depleted organic matter to hydrogensulphide and CO2. Input of this CO2 into the medium leads to anincrease in total DIC and an increase in the SI (Visscher et al., 2000;Altermann et al., 2006; Baumgartner et al., 2006). Other heterotrophicactivity may have similar effects on ion concentration duringoxidation of organic matter. Furthermore, Ca2+ and Mg2+ ions storedin cyanobacteria extracellular polymeric substance (EPS) are releasedduring decomposition, increasing the SI and promoting precipitationof CaCO3 (Paerl et al., 2001; Altermann et al., 2006).
Either of the microbial influences above has the potential to inducedeviations in measured δ13C carbonate values from the predictedequilibrium values thereby providing a biosignature. The direction ofthese deviations indicates whether autotrophic or heterotrophicprocesses are dominant in affecting the isotopic composition of thelocal DIC pool (McConnaughey, 1989; Merz, 1992; Ferris et al., 1997;McConnaughey et al., 1997; Thompson et al., 1997; Hodell et al., 1998).Biological preference for 12C during photosynthesis leads to incorpora-tion of the lighter isotope into cell biomass and a correspondingenrichment in 13C of the residual DIC (O'Leary, 1988). In contrast,heterotrophic activity results in an input of 13C-depleted CO2 fromdegradation of organic matter, leading to a 13C-depletion in the residualDIC. When carbonate precipitates from DIC that has been affected byeither of these processes, the precipitated carbonates record the δ13Cvalue, including any microbial effects on the isotopic composition(Burne and Moore, 1987; Guo et al., 1996; Thompson et al., 1997;Sumner, 2001). Carbonates enriched in 13C have been reported from avariety of environments including saline ponds, shallow lakeswith highmethane production, hot spring travertines (Guo et al., 1996; Valero-Garcés et al., 1999; Gu et al., 2004), and small carbonate precipitatesfrom freshwater lakes with high levels of photosynthetic activity(Hollander and McKenzie, 1991; Thompson et al., 1997). Carbonatewith δ13C values lower than expected for equilibrium precipitation hasbeen used in recent studies to infer influence of heterotrophicmetabolisms on carbonate precipitation in freshwater microbialitesfrom Cuatro Ciénegas, Mexico (Breitbart et al., 2009) and in modernmarine microbialites from Highborne Cay, Bahamas (Andres et al.,2006).
In addition to the isotopic composition of the inorganic component,examination of the isotopic composition of the organic materialassociated with microbialites provides insight into dominant microbialmetabolic processes. The isotopic composition of microbial phospho-lipid fatty acids (PLFA) offers insight into the metabolisms of the activein situ microbial community in an environmental system (Sakata et al.,1997; Abraham et al., 1998). Phospholipids are membrane componentsthat are known to degrade rapidly, within days to weeks, upon death(White et al., 1979) therefore they represent the viable microbialcommunity at a site. Further, specific PLFAhave been shown to be linkedto certain microbial groups and may be used to identifying changes inmicrobial community (Vestal and White, 1989; Rajendran et al., 1995;Zelles, 1999; Green and Scow, 2000). The PLFA profiles and associatedisotopic signatures provide additional information about the activity ofthe dominant microbial community present within the nodules.Biosynthesis of PLFA in heterotrophic microbes leads to PLFA withδ13C values that are depleted relative to the bulk cell biomass by 3–4‰(Blair et al., 1985; Monson and Hayes, 1982) while cyanobacteria lipidδ13C values are depleted relative to bulk biomass by 7–9‰ (Sakata et al.,1997).
This study focused on the laminated micro-stromatolitic and non-laminated nodules located on the surface of Pavilion Lake micro-bialites. These nodules were targeted due to their potential for highlevels of microbial activity and for the occurrence of a geochemical
57A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
microenvironment within the nodule that might allow resolution ofbiogeochemical processing not easily recognized in the thin (b5 mm)surface biofilms covering the majority of the remainder of themicrobialite surfaces. This study combined stable isotope analysis,geochemical analysis and imaging of associated microbial communi-ties to investigate the potential for identifying a biosignature ofphotosynthetically influenced carbonate precipitation within thesesurface nodules.
2. Sampling and analytical methods
2.1. Study site
Pavilion Lake is located in south-central British Columbia, Canadaapproximately 450 km north-east of Vancouver at an altitude of 823 m
above sea level. It is a small (5.7 km×0.8 km and 65 m deep)freshwater, ultra-oligotrophic lake with a pH of ∼8.3 and hostsmicrobialites ranging from several centimetres to meters in heightwith varying morphologies (Laval et al., 2000; Lim et al., 2009). Themicrobialites are estimated to be younger than 12,000 years based onuraniumseries dating and their location above silts that bury postglacialclastic sediments (Laval et al., 2000). Photosynthetic cyanobacteria suchas Synechococcus sp. andOscillatoria sp. are known to be associatedwiththese freshwater microbialites (Laval et al., 2000).
2.2. Microbialite and water chemistry collection and characterization
Sampling of microbialites was performed along transects runningperpendicular to the shoreline from 7 to 24 m depth as describedfurther in (Lim et al., 2009). Microbialite samples were collected alongthese transects by SCUBA divers during field seasons conducted twoto four times a year from spring to autumn between 2006 and 2008(Table 1). Microbialite pieces were frozen on-site and transported toMcMaster University on dry ice for further analysis. Green laminatedmicro-stromatolitic and pink/purple non-laminated nodules that areassociated with the surface of many of the microbialites are the focusof the current study (Fig. 1). Selected samples for investigation usingscanning electron microscopy (SEM) were preserved in a 2.5%glutaraldehyde solution with 0.45 μm pore size filtered lake water.
Water samples for 13C analysis were collected in crimp sealed glassserum bottles with no headspace and fixed with mercuric chloride toprevent further microbial activity. In addition to bulk water samples,syringe pull samples were collected via SCUBA divers as close to themicrobialite surface as possible, however no differences in δ13C valueswere found at this sampling resolution. Water samples for 18O analysiswere also collected in crimpsealedglass serumbottleswithnoheadspace.
In situ water temperature was measured along the aforemen-tioned transects 10 cm above the lake bed at nominal water depths of10 m and 18 m (lake level varies by about 1 m over seasonal time
Table 1δ13C value of DIC, predicted carbonate and δ18O values of water in Pavilion Lake from2005 to 2008.
Fig. 1. Representative images of the samples examined in this study. a) Pavilion Lake microbialite approximately 1 m in height (image courtesy D. Reid). b) Green (G) and purple (P)nodules on the surface of a microbialite showing range in size (image courtesy D. Reid). c) Cross-section of green nodule illustrating internal dark, green organic-rich laminations offilamentous cyanobacteria alternating with light coloured carbonate rich lamination at a distance of approximately 1 mm apart. d) Cross-section of purple nodules showing randomdistribution of carbonate among filaments.
58 A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
scales) using SCUBA diver deployed Onset WaterTempPro v2 watertemperature data loggers (resolution and calibrated accuracy 0.2 °C).The sampling interval was 30 min.
2.3. Phospholipid fatty acid analysis of microbialite nodules
Due to the small size/mass of the individual nodules, multiple greennodules were combined into a representative sample from each of thetwo field seasons in which sufficient sample was collected for PLFAanalysis (April 2006 and April 2008). Microbial PLFA were extractedfrom micro-stromatolitic nodules according to a modified Bligh andDyer method (Bligh and Dyer, 1959) and purified using silica gelchromatography to separate lipids into non-polar, neutral and polarfractions. Phospholipids recovered from the polar fraction weresubjected to a mild alkaline methanolysis and converted to fatty acidmethyl esters (FAMEs) (Guckert et al., 1985). Microbial FAMEs wereseparated using gas chromatographymass spectrometry (GC/MS) on anAgilent GC–MS (Agilent Technologies Inc., Santa Clara, California, USA)with DB-XLB capillary column (30 m×0.32 mm I.D.×0.25 μm filmthickness) using a temperature program of 40 °C (1 min.), 20 °C/min to130 °C, 4 °C/min to 160 °C, 8 °C/min to 300 °C (5 min). Identification ofPLFAwasmade based on the retention time andmass spectra of knownreference standards (Bacterial Acid Methyl Esters Mix, Matreya Inc.,Pleasant Gap, Pennsylvania, USA). PLFA are identified according to thenumber of carbon atoms present and the number of double bonds.
2.4. Stable isotope analysis
DIC isotopic composition was determined by acidification andconversion to CO2 analyzed by an automated continuous flow isotoperatio mass spectrometer at the G.G. Hatch Laboratory in Ottawa (St-Jean,2003). All δ13CDIC values are reported in standard delta notation inreference to PeeDee Belemnite (PDB). Water oxygen values weredetermined by CO2–water equilibration at 25 °C prior to analysis on aGasbench and Finnigan MAT DeltaPlus XP. δ18O values are reported instandard delta notation in reference to Vienna Standard Mean OceanWater (VSMOW).
Surface nodules were removed from microbialite samples usingsolvent rinsed tweezers. Carbonate stable isotope analyses wereperformed on anOptima isotope ratiomass spectrometerwith an Isocarbcommon acid bath at 90 °C at McMaster University. Triplicate analysisof carbonate samples gave precisions of less ±0.7‰ (1 σ) for δ13Ccarband ±0.2‰ (1 σ) for δ18O analyses. Samples for bulk organic carbonanalysis were dried and treated with 1 M HCl to remove carbonate. Bulkorganic isotopic analyseswere conducted on an EA-Delta XL atMcMasterUniversity. Triplicate analyses gave a precision of less ±1.2‰ (1 σ) forδ13Corg values of all nodules. All carbonate δ13C and δ18O values arereported in standard delta notation relative to PeeDee Belemnite (PDB).
Aliquots of microbial FAMEs were injected into a split/splitlessinjector set to splitless mode at 300 °C prior to separation using gaschromatography mass spectrometry (GC/MS) on an Agilent GC–MSwith DB-XLB capillary column (30 m×0.32 mm I.D.×0.25 μm filmthickness) and a temperature program of 80 °C (1 min.), 4 °C/min to280 °C, 10 °C/min to 320 °C (20 min). Individual FAMEs werecombusted to CO2 as they eluted from the column via a combustionoven set at 960 °C. Evolved CO2 was analyzed using a DeltaPlus XPcontinuous flow isotope ratio mass spectrometer (IRMS).
The methanol used during methanolysis was characterized for 13Cand FAME δ13C values were corrected for the addedmethyl carbon viathe relationship:
δ13CFAME = N + 1ð Þ⁎δ13Cmeasured−δ13CMeOH
h i=N
where N is the number of carbon atoms. All δ13C values are reportedin standard delta notation relative to PeeDee Belemnite (PDB).
Individual samples were analyzed in triplicate and data are reportedas mean±one standard deviation (s.d.). Precision on triplicateδ13CPLFA values for individual PLFA was less than ±0.7‰.
2.5. Voltammetry
Microsensor measurements were conducted ex situ on freshlycollected nodules and on nodules in situ. Ex situ measurements onfreshly collected nodules incubated in Pavilion lake water wereperformed using glass Au–Hg amalgam electrodes (with tips drawn to∼500 µm diameter) constructed in the lab according to methodspublished in Brendel and Luther (1995) and lowered vertically in 1–5 mm increments using a micromanipulator. A sequence of ten cyclicvoltammograms (−0.1 to−1.8 V vs. Ag/AgCl at 1 V/swith 2 s depositionat−0.1 V)was obtained from each electrode at each depth using DLK-60(AIS Instruments) software. The current response for signals of the last 5scans of each sequence were measured and averaged (AIS InstrumentsDLK-60 Analysis program). The instrumental variability between mea-surements is extremely small (typically less than 1%). Oxygen peaks existat−1.3 V and−0.3 V (O2 and H2O2), Mn2+ at−1.6 V, Fe(II) at−1.4 V,Fe(III) at−0.6 V, andHS−at−0.8 V. Theelectrodeswere calibratedusing2-point O2 calibrations (air-saturated and N2 purged), standard additionsof freshly prepared Na2S 9H2O, and standard additions of MnCl to N2
purged water, with calibration for other ions, relative to Mn2+,accomplished using the pilot ion method (Brendel and Luther, 1995). Insitu oxygen and pHmeasurements were conducted on green and purplenodules at depths of 17–19 m using a Unisense oxygen and pH electrodeconnected to a Unisense underwater picoammeter/mVmeter. Electrodeswere 100 µm in diameter and lowered in 200 µm steps via amicromanipulator fixed to a stand. The pH electrode had an externalreference electrode. Profiling was performed from the bulk solution intoand through any boundary layer that may have existed at the surface ofthe nodules. Placement of the microelectrode above the nodular surfacewas aided by visual observation. Green nodules were penetrated to amaximum depth of 3 mm below the surface of the nodule, crossinglaminations present within these sampleswhile the purple noduleswerepenetrated to a maximum depth of 5 mm. In all cases, great care wastaken to minimize disruption to the boundary layer.
2.6. Scanning electron microscopy
To examine the internal structure of the microbialites, subsamplespreserved in 2.5% glutaraldehyde were prepared following the proce-dure of Omelon et al. (2006). Specifically, nodule were sectionedvertically using sterile razor blades and washed in 0.1 M sodiumcacodylate buffer (pH 7.3), postfix stained in 0.1% osmium tetroxide/0.1 M sodium cacodylate buffer, washed in 0.1 M sodium cacodylatebuffer, dehydrated through a graded alcohol series (70%, 90%, 95%, and100% ETOH), and embedding in LR White acrylic resin in 1″ diameterplasticmoulds. Thesewere subsequently ground, polished, andplatinumcoated. All samples were viewed on a LEO (Zeiss) 1540XB scanningelectron microscope equipped with both quadrant back scattering andX-ray spectroscopic (EDS) detectors.
3. Results
As noted, this study focused on the nodularmicrobial communitiesfound on the surface of the Pavilion Lake microbialites. Light andfluorescence microscopy determined that these nodules weredominated by filamentous bacterial communities that ranged in size(up to 1 cm in diameter) and were either green or pink/purple incolour (Fig. 1b). Sectioning of the green spherical nodules revealedinternal laminations of dark green organic-rich bands with lightercoloured carbonate rich bands of up to 1 mm thickness that wereobservable with the naked eye (Fig. 1c). The purple nodules had a lessorganized internal structure composed of microbial filaments and
59A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
randomly distributed carbonate precipitates (Fig. 1d). The two noduletypes were included in this study as they represented the two primarytypes of nodular microbial communities found in association with thesurface of the microbialites.
3.1. Isotopic composition of DIC and nodule carbonate
DIC and predicted carbonate δ13C and H2O δ18O values for PavilionLake surfacewater samples thatwere collected at between2005and2008are presented in Table 1. DIC δ13C values average −1.2±1.3‰ (n=13)and water δ18O values average −11.1±0.2‰ (n=10). No difference inδ13CDIC was observed for any syringe pull samples collected in the localenvironment of themicrobialites. Pavilion Lake has a mean pH of 8.3 thatremains stable on a yearly basis and throughout the water column andresults in bicarbonate (HCO3
−) being the dominant DIC species (Lim et al.,2009). Based upon the fractionation factor of Mook et al. (1974) and themeasured δ13Cof atmosphericCO2at PavilionLake (δ13C=−9.6±0.2‰),a temperature range of 0–20 °C would result in equilibrium DIC δ13Cvalues ranging from +1.1‰ to −1.3‰. DIC δ13C values average −1.2±1.3‰ (n=13) and generally fall within this predicted seasonal rangeconfirming that the lake DIC is consistently at or near isotopic equilibriumwith atmospheric CO2, in agreement with previous results based on 14Canalysis (Brady et al., 2009). Previous studies have shown that calcite isenriched in 13C by +1.0±0.2‰ above the bicarbonate from which itprecipitates (Romanek et al., 1992). Thus, abiotic precipitation ofmicrobialites would result in carbonate δ13C values generally within 1‰of the bulk measured DIC δ13C values. Since the precise timing ofcarbonate precipitationwas not known, the average δ13C value of DIC andthe expected 1‰ enrichment during precipitation were used to generatean average predicted carbonate equilibrium δ13C value for Pavilion Lakethat was −0.2±1.3‰ (Fig. 2). In addition to this averaged prediction,Fig. 2 also shows the maximum range of predicted carbonate δ13C valuesbased on the range of DIC δ13C values observed in individual samplescollected over 4 years.
Nodule carbonate δ13C and δ18O values are listed in Table 2. Themeasured δ13C values from greenmicro-stromatolitic and purple non-laminated nodules range from +1.1 to +3.4‰ (mean +2.3±0.5‰,n=27). These values were 13C-enriched above the prediction basedon equilibrium precipitation for both the average and maximumranges of observed DIC values with the exception of three samples.However, these three samples are enriched or identical relative to thepredicted carbonate δ13C values based on the corresponding DICsampled during the same season (Fig. 2). δ18O values ranged from−10.7 to −9.2‰ (mean −9.5±0.3‰, n=27). No correlation wasobserved between carbonate δ13C and δ18O values for either the green(R2=0.13) or purple nodules (R2=0.00) (Fig. 3).
3.2. Organic matter carbon isotope composition of microbial communities
Bulk organic δ13C values of themicro-stromatolitic nodules are listedin Table 2. All of the nodule samples combined have an average δ13Cvalue of−28.0±2.1‰ (n=19) with a range of−30.6 to−21.1‰. Thegreen nodules averaged −27.5±2.0‰ and the purple nodules had anaverage δ13C value of −30.0±1.1‰. The difference between thesemeans was statistically significant (Mann–Whitney Rank Sum test,p=0.011, α=0.05). The average offset between mean DIC and organicδ13C values (Δδ13C) for all nodule samples is−26.8±1.1‰ (n=21). Forgreen nodules the average offset was −26.3±2.0‰ (n=15) and thepurple nodules had a mean offset of−28.8±1.1‰ (n=4).
3.3. Microbial phospholipid fatty acid analysis
The microbial PLFA composition of the micro-stromatoliticnodule communities are listed in Table 3. Saturated straight chainPLFA ranged from C14 to C18, comprised 27–36% of the total with themost abundant, C16, comprising 28.1 and 22.5% in April 2006 andApril 2008. Monounsaturated PLFA comprised the majority of thetotal PLFAs present in both samples (totalN45%) with C16:1 and C18:1
Fig. 2. Relationship between carbonate δ13C values of individual green (◆) and purple (▲) nodules to the predicted carbonate δ13C values precipitated under abiotic conditions withno biological influence from the mean DIC δ13C value in Pavilion Lake. The shaded area illustrates the total range in predicted abiotic precipitation δ13C values based on all measuredDIC values. The solid line represents the mean predicted carbonate δ13C values while the dashed line represents one standard deviation of the mean DIC values. Green and purplenodules carbonate δ13C are elevated above the range expected for abiotic precipitation.
60 A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
as the major components detected in the two samples (10.7–19.8%and 31.4–34.6% respectively). Branched-chain saturated PLFA werealso present, specifically iso- and anteiso-C15:0 that represented onaverage 1.8 and 1.9% respectively of the total PLFA. Iso- and anteiso-C17:0 PLFA were also observed in the nodule sample from April 2008(b2%). Cyclopropyl PLFA C17:0Δ was also detected in small amounts(1.5–1.6%) in both samples. Polyunsatured PLFA C18:3 was identifiedin the April 2006 sample at 4.1 mol% butwas not detected in the April2008 sample.
The small nature of the nodules necessitated the combination ofmultiple nodules to have sufficient material for GC/MS and δ13Canalysis of individual PLFA. Bulk δ13Corg values of individual noduleswere consistent, indicating that combined nodules would give a
representative PLFA profile. PLFA δ13C values from nodules collectedin April 2006 and April 2008 are listed in Table 3. The δ13C values fromApril 2006 ranged from −34.0 to −31.5‰ and from April 2008ranged from −35.0 to −31.4‰. PLFA were depleted relative to themean bulk organic δ13C values by on average −5.0 and −4.9‰respectively. The largest offsets are seen in saturated straight chainPLFA (C14 and C16) with values ranging from −5.9 to −6.6‰.Branched-chain PLFA iso- and anteiso-C15:0 from the April 2008sample were depleted relative to the average April 2008 bulk organicδ13C value of −28.0‰ by −4.0‰.
3.5. Voltammetry
Microelectrode profiling of oxygen, sulphide, Mn, Fe and pH wereused to characterize the chemical composition of the boundary layerand within the nodular structures during the day (Fig. 4). In the greennodules, oxygen showed an increase in concentration as compared tothe bulk solution, reaching the maximum concentration observed in
Fig. 3. Carbon versus oxygen isotope composition for green (◆) and purple (▲) nodulecarbonate. The lack of correlation between carbonate δ13C and δ18O values for green andpurple nodules suggests no significant non-biological enrichment effects.
Table 3PLFA distribution inmol% and δ13C values for April 2006 and April 2008 nodule samples.
PLFA I.D. April 2006 δ13C‰ PDB April 2008 δ13C‰ PDB
N1 Green, depth 18 m 04/2006 2.7 −9.5 −27.1 25.9N2 Green, depth 18 m 04/2006 1.9 −9.6 −28.0±1.2 26.8N3 Green, depth 18 m 04/2006 2.3±0.4 −9.6±0.1 −27.1 25.9N4 Green, depth 21 m 04/2006 2.4 −9.2 – –
N5 Green, depth 18 m 06/2007 2.2±0.1 −9.6±0.2 −26.6 25.4N6 Green, depth 18 m 06/2007 3.4 −9.3 – –
N7 Green, depth 18 m 06/2007 1.1 −9.8 −28.5±0.7 27.3N8 Green, depth 10 m 02/2008 1.7 −10.7 – –
N9 Green, depth 18 m 04/2008 2.4 −9.4 −28.5±0.7 27.3N10 Green, depth 18 m 04/2008 2.5±0.1 −9.2±0.1 – –
N11 Green, depth 18 m 04/2008 2.8±0.3 −9.3±0.1 −28.3±0.6 27.1N12 Green, depth 18 m 04/2008 2.8 −9.2 −29.6 28.4N13 Green, depth 18 m 04/2008 2.0 −9.6 – –
N14 Green, depth 18 m 04/2008 2.3 −9.2 – –
N15 Green, depth 18 m 04/2008 2.6 −9.2 – –
N16 Green, depth 18 m 07/2008 2.7±0.4 −9.4±0.0 −27.3±1.2 26.1N17 Green, depth 18 m 07/2008 1.6±0.3 −9.7±0.1 −28.1±0.4 26.1N18 Green, depth 18 m 07/2008 1.3 −9.5 −27.1 25.9N19 Green, depth 18 m 07/2008 2.2 −9.7 −27.5 26.3N20 Green, depth 18 m 07/2008 2.2 −9.9 −29.7 28.5N21 Green, depth 18 m 07/2008 2.8 −9.4 −28.1±0.8 26.8N22 Green, depth 10 m 10/2008 2.7 −10.3 −21.1±0.7 19.9N23 Purple, depth 18 m 04/2008 2.2±0.7 −9.3±0.0 −30.6 29.4N24 Purple, depth 18 m 04/2008 2.2±0.1 −9.5±0.1 – –
N25 Purple, depth 18 m 04/2008 2.3 −9.4 −28.4 27.2N26 Purple, depth 18 m 04/2008 2.7 −9.4 −30.4±0.9 29.2N27 Purple, depth 18 m 04/2008 2.4 −9.6 −30.6 29.4
61A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
all nodules of 35.2 mg/L (275% saturation) within the region of strongestgreen/white banding in the initial 1.5–2 mm from the surface of thenodule. O2 concentrations generally began to decline at depths of 2 mmwithin the green nodules reaching a low of 7.1 mg/L or 78% saturation at3 mm in one nodule. Within the purple nodules, oxygen concentrationswere at saturation or supersaturated through the nodule with oneexception. Generally, oxygen increased from an average saturation of14.2 mg/L (112% saturation) at the nodule surface to an observedmaximumof 34.2 mg/L (267% saturation) at a depth of 2.8 mm.However,onepurple nodule showed amaximumoxygen level of 100% saturation atadepthof1 mmfromthe surfacewith theconcentrationdeclining to0%atadepthof 3.5 mm(Fig. 4a).Differences inoxygenconcentrationsbetweennodules profiled ex situ on the surface and those profiled in situ likelyresulted from differences in nodule size, temperature, and correspondingoxygen solubility, and/or ambient light levels during profiling. Nosulphide, Fe2+ or Mn2+ was detected in either the green or purplenodules during ex situ profiling, similar measurements were not possiblefor in situ measurements.
The average pH value for the ambient water at a maximum distanceof 3 mm from the surface of the nodules was 8.9 (n=3), slightly higherthan previously measured values from Pavilion (Lim et al., 2009) butpossibly resulting from photosynthetic influences within the nodulesextending from the nodule surface into the boundary layer (Jørgensenand Des Marais, 1990). All nodules showed an increase in pH withincreased depth from the surface of the nodule (Fig. 4b). The purplenodules reached a peak of 9.5 at a depth of 1.5–2 mm from the surface
with one sample reaching9.6 and the greennodule reached amaximumof 9.5 at a depth of 1.8 mm.
3.6. Scanning electron microscopy
SEM examination of the internal structure of the green nodulesshowed filamentous cyanobacteria oriented perpendicular to the micro-bialite surface with carbonate aggregates present within this biomass aslaminated bands (Fig. 5a). Purple nodules also had filamentouscyanobacteria oriented perpendicular to the microbialite surface butcarbonate precipitates were not organized in distinct laminations(Fig. 5b). Consistent with light and florescence microscopy, coccoidalcyanobacteria were observed to be a minor component of thecommunities. Cyanobacteria filaments were embedded in the carbonatematrix and within the green nodules crossed through the carbonatelaminations, maintaining a vertical growth orientation. The carbonatepresentwithinboth thegreenandpurplenodules showednoevidence forprecipitation within filamentous cyanobacteria sheaths or in directassociation with filamentous cyanobacterial cell surfaces, but ratherwithin the surrounding EPSmatrix. Smaller, coccoidal cyanobacteria cellswere observed in the samples however there was no evidence of directassociationbetweenprecipitates and cell surfaces. Structureswithin thesecarbonates include void spaces of similar shape andmorphology to extantmicrobes, as well as secondary infilling of void spaces within the mineralmatrix (Fig. 6).
4. Discussion
4.1. Carbonate isotope biosignatures
All of the nodules sampled from Pavilion Lakemicrobialites had δ13Cvalues enriched above the average predicted abiotic precipitation value
Fig. 4. Oxygen and pH microelectrode profiles through green and purple nodules.a) Oxygen profiles demonstrating increased oxygen concentrations within the nodules,green 1 and purple 2 represent samples profiled on the surface in Pavilion Lake waterwhile green 2.3 and purple 2.3 were profiled in situ. b) pH profiles demonstratingincreased pH values within the nodules consistent with photosynthetic influences onthe nodule geochemistry. Error bars represent one standard deviation on triplicatemeasurements in the purple nodules.
Fig. 5. SEM back-scattered electron (BSE) micrographs of microbial nodules in cross-section. (a) Green-pigmented nodule, with the outermost (exposed) surface of thenodule at top of image. Filamentous phototrophic bacteria grow outward from a centralpoint or region, observed as cell trichomes with distinct organized orientation (seearrow). Note carbonates in horizons perpendicular to microbial growth within thebiomass framework. (b) Purple-pigmented nodule showing filamentous bacteria in thepresence of carbonates. Filaments retain their growth orientation despite the presenceof carbonate minerals, suggesting in situ nucleation and crystal growth.
62 A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
of−0.2±1.3‰ from bulk Pavilion LakeDIC (mean δ13C=−1.2±1.3‰(n=13) over 4 years) with values ranging from +1.1 to 3.4‰ (+1.3to3.6‰ above expected) (Fig. 2). While three samples were not 13C-enriched with respect to the maximum range of predicted carbonateδ13C values (Fig. 2), when these samples were compared to predictedcarbonate δ13C values based on DIC at the time of sampling, all but onewas observed to be enriched. Thesefindings are consistentwith shifts incarbonate isotopic composition typically on the order of 2–5‰ inducedby microbial photosynthetic activity (Merz, 1992; Guo et al., 1996;Thompsonet al., 1997) andare similar to, albeit lower than, enrichmentsof 4.6–5.2‰observed in small stromatolites and thrombolites recoveredin nearby Kelly Lake (Ferris et al., 1997). The observation of nodulecarbonate with δ13C values enriched over values predicted forprecipitation from the bulk lakewaterDICwas consistentwithmicrobialmetabolic influences on the geochemistry and isotopic composition ofthe DIC within the local zone of influence (microenvironment) of themicrobes, either surrounding the microbial cells or within theenvironment of the nodule as a whole (Beveridge, 1988; Schultze-Lamet al., 1992). It indicated that the rate of preferential 12C utilization bymicrobial photosynthetic activity within the microenvironment of thenodule was sufficiently greater than the rate of exchange with the bulkwater such that it resulted in a 13C-enrichment of the DIC within thisenvironment but did not affect the bulk lake water. This observedisotopic enrichment of the carbonate is therefore a biosignature ofmicrobial photosynthetic, rather than heterotrophic, activity beingdominant in affecting the local geochemistry within the nodule. Ifheterotrophic activity was the predominant process, the carbonate δ13Cvalues would be expected to be 13C-depleted with respect to thepredicted equilibrium values as opposed to the enrichment that wasobserved in the nodules. And while the relative predominance ofautotrophic and heterotrophic respiration is expected to shift duringday/night cycles, the observation of 13C-enriched carbonate indicatedthat the photosynthetic effects on the 13C content of the DIC weredominating and creating an isotopic biosignature preserved within thenodule carbonate.
This signature is consistent with previous observations thatcyanobacteria blooms and extensive photosynthetic activity in thewater column are known to be capable of resulting in precipitation of13C-enriched carbonate in whiting events such as those observed inGreen Lake, NY (Thompson et al., 1997). These whiting eventssignificantly affect water clarity and induced positive shifts in theGreen Lake DIC δ13C values of ∼3‰ over the period of increasedphotosynthetic activity. In contrast to the Green Lake system, the factthat over a period of 4 years Pavilion Lake δ13C DIC fell within the rangepredicted for equilibrium with the atmosphere (predicted range ofδ13C=+1.1 to−1.3 see Table 1), coupled with the low sedimentationrate within the lake (Lim et al., 2009) and the lack of any observation ofwhiting events or other changes inwater clarity by year round residentsof the lake indicates that similar lake wide effects cannot explain theobserved isotopic enrichments and that these enrichments are confinedto the microenvironment within the nodules.
Concurrent with this observation of photosynthetic effects on thenodule carbonate, nodule bulk organic δ13C values were 13C-depleted(mean δ13C=−28.0±2.1‰) as compared to DIC, consistent withmicrobial photosynthesis and the uptake of 12C. The discriminationbetween the average bulk Pavilion Lake DIC δ13C value (−1.2±1.3‰,n=13) and the mean bulk organic δ13C value for all nodules was−26.8±2.1‰ (n=19) close to the maximum isotopic discriminationexpected for C3-photosynthesis under non-limiting CO2 conditions(O'Leary, 1988). The purple nodules had a slightly greater offset thanthe green nodules, likely resulting from differences in the microbialcommunity (Estep et al., 1978). Expression of this isotopic discriminationduring DIC uptake from the local DIC pool within the nodulemicroenvironment explains a 13C-enrichment in the DIC and thereforethe observed enrichment of the carbonates.
The lack of correlation betweenmicro-stromatolitic nodule carbonateδ13C and δ18Ovalues (R2=0.13 for greennodules andR2=0.00 for purplenodules) (Fig. 3) further supports that the observed 13C-enrichment wasdue to microbial activity rather than physical processes such asevaporation. Microbial metabolic (photosynthetic) effects on carbon
Fig. 6. SEM–BSE micrographs of microbialite nodules in cross-section. Predominant growth orientation shown by arrows. Evidence for in situ carbonate precipitation: (a) filamentousmicroorganisms entombed in carbonateminerals, suggestingmineral growth around individualfilaments; (b) void spaceswithin carbonateswith similar size andmorphology to adjacentextantmicroorganisms; (c) in addition to having coccoidmorphologies, void spaceswithin carbonates resemblefilamentous shapes thatmimic themorphologyof thedominantmicrobialpopulation; (d) carbonate minerals also reveal secondary infilling of void spaces, further suggesting active carbonate precipitation within the microbialite nodules.
63A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
isotope values have been shown to have no significant influence onoxygen values (McConnaughey, 1989). In contrast, correlation betweenδ13C and δ18O values has been observed in previous studies wherephysical processes such as evaporation affect both isotope systems andcan lead to 13C- and 18O-enrichment in precipitates (Andrews et al., 1993;Valero-Garcés et al., 1999; Léveillé et al., 2007). Therefore, a lack ofcorrelation between δ13C and δ18O values can be used to infer biologicalinfluences on carbonate precipitation (Burne and Moore, 1987; Lojen etal., 2004; Kremer et al., 2008).
Since they are not affected by biological factors, the δ18O values of thenodule carbonates can also be used to estimate the temperature offormation as they are a function of the δ18O value in the water and thetemperature of precipitation (Epstein and Mayeda, 1953). Pavilion Lakewater δ18O values show minor variation over 3 years (−11.1±0.2‰)suggesting that the majority of the observed variation within thecarbonate δ18O values is the result of temperature fluctuations duringprecipitation. Using the measured δ18O values of the ambient water, andestablished temperature-dependent fractionation factors and equations(Kim and O'Neil, 1997), the δ18O values of the carbonates within thenodules correspond to an estimated temperature range of 8.2 to 13.6 °C(Fig. 7). The estimated temperatures of formation were higher for thesamples from 10m than those from 18m, consistent with the observeddifferences in annual mean water temperature at those depths (Fig. 7).The estimated temperatures of formation were also higher than therecorded water temperatures during the winter months and for bothdepths and tended to corresponded to summer and fall temperatures. Inorder to account for potential effects of the observed increased pHwithinthe nodules and associated changes in carbonate speciation on thistemperature estimate (Zeebe, 1999), temperature estimates wererecalculated using the maximum observed pH increase of 0.7 units(Fig. 4b). These recalculated estimateswere on average 3.6±0.1 °C lowerthan the previous estimates (Fig. 7) and were now consistent withcarbonate precipitation at temperatures recorded circa June and Octoberat 10 m and the period circa May to October at 18 m depth. Theimplication of these estimates is that there appears to be a seasonal periodof carbonate precipitation within the nodules that is consistent withsummertime temperatures in Pavilion Lake. Since this is also the time
whenphotosynthetic activity is expected tobehighest (McConnaugheyetal., 1994; Fritsen and Priscu, 1998), these observations are consistentwithnodule carbonate precipitation being a photosynthetically driven process.
4.2. Predominance of photosynthetic activity within the nodules
The 13C enrichment in the nodule carbonate, the depleted δ13Cvalues of organic matter and the evidence for summer precipitationwhen photosynthesis rates are highest indicate that biological photo-synthetic activity was the predominant net influence on DIC geochem-istry within the microenvironment of the nodules. In addition, theobservation within the nodules of elevated oxygen (up to 275%saturation) and pH levels that increased by 0.7 pH units as comparedto the ambient water demonstrated that microbial photosynthesis wasstrongly affecting the local geochemistry during the day (Fig. 4) as hasbeen observed in previous studies of microbial mats (Jørgensen et al.,1983; Revsbech et al., 1983; Stal et al., 1985; Ludwig et al., 2005;Vasconcelos et al., 2006).
PLFA analysis further supported bacterial photosynthetic activitywithin the nodules as the results indicated that nodular communitieswere dominated by prokaryotic microbes, particularly cyanobacteria,and that eukaryotic organisms contribute little (below detection) ofthe community. PLFA profiles of the two green micro-stromatolicnodule samples contained biomarkers common to many microbialgroups including iso-C15:0, anteiso-C15:0, monoenoics and straightchain saturates ranging from C14 to C18 (Vestal and White, 1989). Inparticular, iso-C15:0, anteiso-C15:0 are characteristic for Gram-negativeand Gram-positive bacteria but are less common in cyanobacteriaindicating that, as expected, heterotrophic organisms are present thatcould be contributing to carbonate precipitation (Jahnke et al., 2004).However, specific biomarkers for sulfate-reducing bacteria such as10me16:0 were not observed suggesting that sulfate-reduction wasnot a dominant process within the nodules (Londry et al., 2004). Highconcentrations of monoenoic PLFA are typically found in Gram-negative bacteria (Guckert et al., 1985; Vestal and White, 1989).However, cyanobacteria have likewise been shown to producemonoenoic fatty acids in significant quantities (Nichols and Wood,
Fig. 7. Measured water temperature at a) 10 m and b) 18 m depth along transect in Pavilion Lake. Temperature of nodule carbonate formation predicted from δ18O are plotted astriangles. Circles represent predicted temperature of formation adjusted for pH effects.
64 A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
1968; Grimalt et al., 1992) suggesting that they may have contributedto the significant proportion of these PLFA. Cyclopropyl PLFA havebeen shown to indicate aerobic bacteria (Parkes and Taylor, 1983) andcould result from cyanobacteria or aerobic heterotrophs present in thenodules. Long chain and PLFA characteristic of eukaryotic organisms(Volkman et al., 1980) were not detected but polyunsatured C18:3 wasdetected within the April 2006 sample. Polyenoic PLFA are typicallyassociated with eukaryotes but have been shown to be produced bycyanobacteria (Kenyon, 1972; Kenyon et al., 1972). These findings areconsistent with previous characterizations of the Pavilion Lakemicrobialites that found surface communities composed of diatomsand cyanobacteria including Synechococcus sp., Pseudoanabaena sp.and purple-pigmented Oscillatoria sp. in addition to heterotrophicorganisms (Laval et al., 2000).
PLFA δ13C values likewise supported the presence of a photosyn-thetically dominated community within the nodules with someevidence of heterotrophic bacteria that are using photosyntheticallyproduced organic matter as their substrate. PLFA δ13C values fromboth April 2006 and April 2008 samples ranged from −34.0 to−31.5‰ and −35.0 to −31.4‰ respectively (Table 3). SaturatedPLFA from the two samples, in particular C14:0 and C16:0 which are themost abundant are depleted relative to the bulk cell δ13C value by 5.9to 6.6‰, offsets of this size are more characteristic of cyanobacteria(Sakata et al., 1997). Other PLFA such as iso- and anteiso-C15:0 that areindicative of heterotrophic microbes had smaller offset (∼4‰) thatare characteristic of heterotrophic metabolisms (Blair et al., 1985;Abraham et al., 1998).
4.3. Evidence for in situ carbonate precipitation
Microbial photosynthetic influence on carbonate precipitationwithin the nodule microenvironment was further supported by SEManalysis that indicated carbonate build-up occurred via in situprecipitation rather than through processes of trapping and bindingof allochthonous particles. Filaments in both the green and purplenodules showed an organized direction of growth that did not appearto be affected by the presence of carbonate precipitates (Fig. 5).Within the nodules the filaments were present within the mineralmatrix suggesting precipitation around individual filaments, a stateunlikely to occur via trapping and binding but that would occur ifprecipitation occurred within the nodule. Void spaces within thematrix of similar size and morphology to adjacent extant microbesare consistent with in situ precipitation and infilling of void spacesindicated active precipitation within the nodules (Fig. 6). In situcarbonate precipitation could be the result of microbial influences onthe local geochemical environment and/or microbial cells andassociated EPS acting as nucleation sites for mineral growth.Cyanobacteria are known to produce the EPS for attachment andprotection (Dittrich and Obst, 2004) and there is strong evidence inthe literature for microbial cells acting as nucleation sites forheterogeneous precipitation of carbonate (Thompson and Ferris,1990; Bosak and Newman, 2003; Chekroun et al., 2004). Howevercarbonate precipitates were not observed within the sheaths ofcyanobacteria filaments, suggesting that direct surface contact withcells was not a prerequisite for carbonate formation in thisenvironment. And while these findings do not discount EPS presentwithin the nodules as potentially contributing nucleation sites forcarbonate forming within a biologically influenced microenviron-ment, the observation of a 13C-enriched carbonate biosignaturedemonstrates that photosynthetic effects are the predominantcontrol on the geochemistry of this microenvironment.
In situ precipitation due to microbial influences on nodulemicroenvironment geochemistry is also consistent with the lack ofsignificant sedimentation within the lake. Although Pavilion Lake issaturated with respect to calcite, dolomite and aragonite (0.72, 1.39and 0.57 respectively) (Lim et al., 2009), sedimentation rates within
the lake were low compared to surrounding lakes with a maximum of0.07 g of sediment (estimated to be ∼2–5% CaCO3) collected fromsediment traps after a period of one year (Lim et al., 2009). Given thislow rate of sedimentation, it is unlikely that trapping and binding ofcarbonate precipitates is the dominant mechanism of accretion,however this mechanism cannot be wholly discounted on this basisalone. In particular, the effect of benthic sediment dynamics and slopecharacteristics on microbialite development has yet to be explored.
4.4. Potential contributions to carbonate precipitation from heterotrophicprocesses
While there is extensive evidence for the predominance ofphotosynthetic activity within the nodules creating the observedisotopic biosignature, heterotrophic organisms are present withinthese nodules as well. In fact, the activity of heterotrophic organismscould account for the decrease in oxygen observed within some of thenodules. Aerobic and anaerobic heterotrophic activity also has thepotential to be a contributing mechanism in carbonate precipitation asobserved in other studies (Visscher et al., 1998;Visscher et al., 2000). Thefact that no sulphide,Mn2+or Fe2+wasdetected in either type of noduleindicated that heterotrophic activity using these electron acceptors wasnot occurring at a significant level during the day. This is consistentwithpreviousmicroelectrode characterizations ofmicrobialmat systems thatfound that oxygen dominated during the day when photosynthesis wasactivewith no sulphide detection until the dark periodwhen respirationdominated (Revsbech et al., 1983). Heterotrophic activity occurringduring the night could influenceprecipitation by raising the carbonate SIthrough addition of DIC and also Ca2+ released during respiration oforganic material. However, if this were the case such heterotrophicinputs would also be associated with inputs of 13C-depleted carbongenerated from the bulk organic matter with an average δ13C value of−28.0±2.1‰ leading to a 13C-depleted DIC pool within the nodules.The observation of 13C-depleted carbonates has been used as one of themajor points of evidence of heterotrophic inputs and influence onprecipitation in othermicrobialite systems (Andres et al., 2006; Breitbartet al., 2009). In thePavilion Lakenodules investigated in this study,wedonot see any resolvable evidence of such 13C-depleted inputs. Because theisotopic composition of the DIC and therefore precipitated carbonate inany system records inputs from both autotrophic and heterotrophicprocesses, this implies that photosynthetic effects on the local isotopegeochemistry are predominant. While the potential for contributionsfrom heterotrophic metabolism on the level of carbonate saturationcannot be discounted, based on the extensive evidence of thepredominance of photosynthetic effects we propose that not only isphotosynthetic activity creating an isotopic biosignature in the DIC pooland associated carbonate, but it is the most likely explanation for thegeochemical changes leading to precipitation within the nodules.
5. Conclusions
Elevated carbonate δ13C values within green and purple nodulespresent on the surface of Pavilion Lake microbialites represent biosigna-tures formed via a predominance of photosynthetic influences on theisotopic geochemistry of the nodule microenvironment over heterotro-phic influences as has been found in other microbialite environments.High levels of oxygen saturation and elevated pH values within thenodules demonstrated that microbial photosynthetic activity influencedthegeochemistryof themicroenvironmentwithout affecting thebulk lakewater. Biological uptake of 12Cwas recorded by the low δ13C values of thebulk organic material and PLFA profiles and δ13C values supported thedominance of photosynthetic activity within the nodules. Thecorresponding 13C-enrichmentof the residualDIC resulted inprecipitationof carbonatewithin the noduleswith δ13C values on average 1–2‰higherthan expected for abiotic precipitation. The lack of correlation withcarbonate δ18O values further supported that the observed 13C-
65A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
enrichment was due to microbial activity rather than physical processes.Estimates of formation temperatures based on carbonate δ18O valuesfurther supported summer as opposed to winter precipitation consistentwith precipitation during the period of highest photosynthetic activity.SEM imaging showing cyanobacteria filaments entombed within themineral matrix supports in situ carbonate precipitation and microbialinfluences onmicroenvironment geochemistry and isotopic composition.The observation ofmicrobial influences on carbonate precipitationwithinthe nodules suggests that microbes are in some part playing a role in theprecipitation of the carbonate in Pavilion Lake microbialites.
These findings represent the first evidence of photosyntheticinfluences on micro-environmental geochemistry and isotopic composi-tion associated with large-scale modern freshwater microbialites. Resultsare consistent with a hypothesized role of biology in the formation ofmicrobialites and the observed 13C-enrichment in photosyntheticallyinfluenced carbonates represents a biosignature of microbial activity forrecent carbonate precipitation within these nodules on the surface of themicrobialites. However, research is ongoing to determine the extent towhich these surface processes are responsible for overall microbialitegrowth and the potential for the biosignatures of photosynthetic activityobserved in this study to be preserved within the bulk structures. Thesefindings give insight into potential formation mechanisms of modernmicrobialites and the recognition of potential biosignatures that may bepreserved throughout geologic time increasing our understanding ofmicrobialite genesis.
Acknowledgements
Thank you to the members of the PLRP, with particular mention toDonnie Reid for his assistance with SCUBA based sample collections andunderwater photography. Thank you also to Jennie Kirby andMartin Knyffor invaluable assistance in laboratory analysis. Infrastructure support forour field research was provided by a Canadian Space Agency (CSA)Canadian Analogue Research Network (CARN) PLRP contract.We are alsograteful to the Ts'Kw'aylaxw First Nation, Linda andMickeyMacri and thePavilion Community, and British Columbia Parks for their continuedsupport of our research. General research funding was provided by aNatural Sciences and Engineering Research Council (NSERC) of CanadaDiscovery Grant and a CSA CARN grant to GFS. We also acknowledge thesupport fromNASA's ExobiologyProgram(DTA). This is PLRP contribution#09-04.
References
Abraham, W.-R., Hesse, C., Pelz, O., 1998. Ratios of carbon isotopes in microbial lipids as anindicator of substrate usage. Applied and Environmental Microbiology 64 (11),4202–4209.
Altermann,W.,Kazmierczak, J.,Oren,A.,Wright,D.T., 2006.Cyanobacterial calcificationand itsrock-building potential during 3.5 billion years of Earth history. Geobiology 4, 147–166.
Andrews, J.E., Riding, R., Dennis, P.F., 1993. Stable isotopic compositions of Recentfreshwater cyanobacterial carbonates from the British Isles: local and regionalenvironmental controls. Sedimentology 40, 303–314.
Awramik, S.M., 1992. The oldest records of photosynthesis. Photosynthesis Research 33,75–89.
Awramik, S.M., Riding, R., 1988. Role of algal eukaryotes in subtidal columnar stromatoliteformation. Proceedings of the National Academy of Sciences 85, 1327–1329.
Beveridge, T.J., 1988. The bacterial surface: general considerations toward design andfunction. Canadian Journal of Microbiology 34, 363–372.
Blair, N., Leu, A., Munoz, E., Olsen, J., Kwong, E., Des Marais, D., 1985. Carbon isotopicfractionation in heterotrophic microbial metabolism. Applied and EnvironmentalMicrobiology 50, 996–1001.
Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification.Canadian Journal of Biochemistry and Physiology 37, 911–917.
Bosak, T., Newman, D.K., 2003. Microbial nucleation of calcium carbonate in thePrecambrian. Geology 31 (7), 577–580.
Brady, A.L., Slater, G., Laval, B., Lim, D.S., 2009. Constraining carbon sources and growthrates of freshwater microbialites in Pavilion Lake using 14C analysis. Geobiology 7,544–555.
Brasier, D.M., Green, O.R., Jephcoat, A.P., Kleppe, A.K., Kranendonk, M.J.V., Lindsay, J.F.,Steele, A., Grassineau, N.V., 2002. Questioning the evidence of Earth's oldest fossils.Nature 416, 76–81.
Breitbart, M., Hoare, A., Nitti, A., Siefert, J., Haynes, M., Dinsdale, E., Edwards, R., Souza, V.,Rohwer, F., Hollander, D., 2009. Metagenomic and stable isotopic analysis of modernfreshwatermicrobialites inCuatroCiénegas,Mexico. EnvironmentalMicrobiology11 (1),16–34.
Brendel, P.J., Luther, G.W., 1995. Development of a gold amalgam voltammetricmicroelectrode for the determination of dissolved Fe, Mn, O2 and S(-II) in porewatersof marine and fresh-water sediments. Environmental Science and Technology 29,751–761.
Chafetz, H.S., Guidry, S.A., 1999. Bacterial shrubs, crystal shrubs, and ray-crystal shrubs:bacterial vs. abiotic precipitation. Sedimentary Geology 126, 57–74.
Chekroun, K.B., Rodriguez-Navarro, C., Gonzalez-Munoz, M.T., Arias, J.M., Cultrone, G.,Rodriguez-Gallego, M., 2004. Precipitation and growth morphology of calciumcarbonate induced byMyxococcus xanthus: implications for recognition of bacterialcarbonates. Journal of Sedimentary Research 74, 868–876.
Council, T.C., Bennett, P.C., 1993. Geochemistry of ikaite formation at Mono Lake,California: implications for the origin of tufa mounds. Geology 21, 971–974.
de Beer, D., Glud, A., Epping, E., Kühl, M., 1997. A fast-responding CO2 microelectrodefor profiling sediments, microbial mats, and biofilms. Limnology and Oceanography42, 1590–1600.
Dittrich, M., Obst, M., 2004. Are picoplankton responsible for calcite precipitation inlakes? Ambio 33 (8), 559–564.
Epstein, S., Mayeda, T., 1953. Variation of 18O content of waters from natural sources.Geochimica et Cosmochimica Acta 4, 213–224.
Estep, M.F., Tabita, F.R., Parker, P.L., Van Baalen, C., 1978. Carbon isotope fractionation byribulose-1, 5-bisophosphate carboxylase from various organisms. Plant Physiology61 (4), 680–687.
Ferris, F.G., Thompson, J.B., Beveridge, T.J., 1997. Modern freshwater microbialites fromKelly Lake, British Columbia, Canada. Palaios 12, 213–219.
Fritsen, C.H., Priscu, J.C., 1998. Cyanobacterial assemblages in permanent ice covers onAntarctic lakes: distribution, growth rate, and temperature response of photosyn-thesis. Journal of Phycology 24, 587–897.
Green, C.T., Scow, K.M., 2000. Analysis of phospholipid fatty acids (PLFA) to characterizemicrobial communities in aquifers. Hydrogeology Journal 8, 126–141.
Grimalt, J.O., de Wit, R., Teixidor, P., Albaiges, J., 1992. Lipid biogeochemistry ofPhormidium and Microcoleus mats. Organic Geochemistry 19, 509–530.
Grotzinger, J.P., Rothman, D.H., 1996. An abiotic model for stromatolite morphogenesis.Nature 383, 423–425.
Gu, B., Schelske, C.L., Hodell, D.A., 2004. Extreme 13C enrichments in a shallowhypereutrophic lake: implications for carbon cycling. Limnology and Oceanography49 (4), 1152–1159.
Guckert, J.B., Antworth, C.P., Nichols, P.D., White, D.C., 1985. Phospholipid, ester-linked fatty acid profiles as reproducible assays for changes in prokaryoticcommunity structure of estuarine sediments. FEMS Microbiology Ecology 31,147–158.
Guo, L., Andrews, J., Riding, R., Dennis, P., Dresser, Q., 1996. Possible microbial effects onstable carbon isotopes in hot-spring travertines. Journal of Sedimentary Research66 (3), 468–473.
Hodell, D.A., Schelske, C.L., Fahnenstiel, G.L., Robbins, L.L., 1998. Biologically inducedcalcite and its isotopic composition in Lake Ontario. Limnology and Oceanography43 (2), 187–199.
Hollander, D.J., McKenzie, J.A., 1991. CO2 control on carbon-isotope fractionationduring aqueous photosynthesis: a paleo-pCO2 barometer. Geology 19, 929–932.
Jahnke, L.L., Embaye, T., Hope, J., Turk, K.A., Van Zuilen, M., Des Marais, D.J., Farmer, J.D.,Summons, R.E., 2004. Lipid biomarker and carbon isotopic signatures forstromatolite-forming, microbial mat communities and Phormidium cultures fromYellowstone National Park. Geobiology 2 (1), 31–47.
Jørgensen, B.B., Des Marais, D.J., 1990. The diffusive boundary layer of sediments:oxygen microgradients over a microbial mat. Limnology and Oceanography 35 (6),1343–1355.
Jørgensen, B.B., Revsbech, N.P., Cohen, Y., 1983. Photosynthesis and structure of benthicmicrobial mats: microelectrode and SEM studies of four cyanobacterial commu-nities. Limnology and Oceanography 28 (6), 1075–1093.
Kempe, S., Kazmierczak, J., Landmann, G., Konuk, T., Reimer, A., Lipp, A., 1991. Largestknown microbialites discovered in Lake Van, Turkey. Nature 349 (6310),605–608.
Kenyon, C.N., 1972. Fatty acid composition of unicellular strains of blue-green algae.Journal of Bacteriology 109, 827–834.
Kenyon, C.N., Rippka, R., Stanier, R.Y., 1972. Fatty acid composition and physiologicalproperties of some filamentous blue-green algae. Archives of Microbiology 83,216–236.
Kim, S.-T., O'Neil, J.R., 1997. Equilibrium and nonequilibrium oxygen isotope effects insynthetic carbonates. Geochimica et Cosmochimica Acta 61, 3461–3475.
Kremer, B., Kazmierczak, J., Stal, L.J., 2008. Calcium carbonate precipitation incyanobacterial mats from sandy tidal flats of the North Sea. Geobiology 6 (1),46–56.
Laval, B., Cady, S.L., Pollack, J.C., McKay, C.P., Bird, J.S., Grotzinger, J.P., Ford, D.C., Bohm,H.R., 2000. Modern freshwater microbialite analogues for ancient dendritic reefstructures. Nature 407, 626–629.
66 A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
Léveillé, R.J., Longstaffe, F.J., Fyfe, W.S., 2007. An isotopic and geochemical study ofcarbonate-clay mineralization in basaltic caves: abiotic versus microbial processes.Geobiology 5, 235–249.
Lim, D.S.S., Laval, B.E., Slater, G.F., Antoniades, D., Forrest, A., Pike,W., Pieters, R., Saffari,M.,Reid, D., Andersen, D., McKay, C.P., 2009. Limnology of Pavilion Lake B.C., Canada —characterization of a microbialite forming environment. Fundamental and AppliedLimnology 173, 329–351.
Logan, B.W., 1961. Cryptozoan and associated stromatolites from the Recent of SharkBay, Western Australia. Journal of Geology 69, 517–533.
Lojen, S., Dolenec, T., Vokal, B., Cukrov, N., Michelčić, G., Papesch, W., 2004. C and Ostable isotope variability in recent freshwater carbonates (River Krka, Croatia).Sedimentology 51, 361–375.
Londry, K.L., Jahnke, L.L., Des Marais, D.J., 2004. Stable carbon isotope ratios of lipidbiomarkers of sulfate-reducing bacteria. Applied and Environmental Microbiology70 (2), 745–751.
Lowe, D.R., 1994. Abiological origin of described stromatolites older than 3.2 Ga.Geology 22, 387–390.
Ludwig, R., Al-Horani, F.A., de Beer, D., Jonkers, H.M., 2005. Photosynthesis-controlledcalcification in a hypersaline microbial mat. Limnology and Oceanography 50 (6),1836–1843.
McConnaughey, T.A., 1989. 13C and 18O isotopic disequilibrium in biological carbonates:I. Patterns. Geochimica et Cosmochimica Acta 53 (1), 151–162.
McConnaughey, T.A., LaBaugh, J.W., Rosenberry,D.O., Striegl, R.G., Reddy,M.M., Schuster, P.F.,Carter, V., 1994. Carbon budget for a groundwater-fed lake: calcification supportssummer photosynthesis. Limnology and Oceanography 39, 1319–1332.
Merz, M.U.E., 1992. The biology of carbonate precipitation by cyanobacteria. Facies 26,81–102.
Merz-Preiß, M., 2000. Calcification in cyanobacteria. In: Riding, R.E., Awramik, S.M.(Eds.), Microbial Sediments. Springer-Verlag, Berlin, pp. 50–56.
Merz-Preiß, M., Riding, R., 1999. Cyanobacterial tufa calcification in two freshwater streams:ambient environment, chemical thresholds and biological processes. SedimentaryGeology 126, 103–124.
Miller, A.G., Colman, B., 1980. Evidence for HCO3− transport by the blue-green alga
(cyanobacterium) Coccochloris peniocystis. Plant Physiology 65 (2), 397–402.Monson, K.D., Hayes, J.M., 1982. Carbon isotopic fractionation in the biosynthesis of bacterial
fatty acids. Ozonolysis of unsaturated fatty acids as a means of determining theintramolecular distribution of carbon isotopes. Geochimica et Cosmochimica Acta 46,139–149.
Mook,W.G., Bommerson, J.C., Staverman,W.H., 1974. Carbon isotope fractionation betweendissolved bicarbonate and gaseous carbon dioxide. Earth and Planetary Science Letters22, 169–176.
Nichols, B.W., Wood, B.J.B., 1968. The occurrence and biosynthesis of gamma-linolenicacid in a blue-green alga, Spirulina platensis. Lipids 3, 46–50.
O'Leary, M.H., 1988. Carbon isotopes in photosynthesis. BioScience 38 (5), 328–336.Omelon, C.R., Pollard,W.H., Ferris, F.G., 2006. Chemical andultrastructural characterization
of high Arctic cryptoendolithic habitats. Geomicrobiology Journal 23, 189–200.Paerl, H.W., Steppe, T.F., Reid, R.P., 2001. Bacterially mediated precipitation in marine
stromatolites. Environmental Microbiology 3 (2), 123–130.Paneth, P., O'Leary, M.H., 1985. Carbon isotope effect on dehydration of bicarbonate ion
catalyzed by carbonic anhydrase. Biochemistry 24, 5143–5147.Parkes, R.J., Taylor, J., 1983. The relationship between fatty acid distributions and
Rajendran, N., Matsuda, O., Imamura, N., Urushigawa, Y., 1995. Microbial communitystructure analysis of euxinic sediments using phospholipid fatty acid biomarkers.Journal of Oceanography 51, 21–38.
Reid, R.P., Macintyre, I.G., Browne, K.M., Steneck, R.S., iller, T., 1995. Modern marinestromatolites in the Exuma Cays, Bahamas: uncommonly common. Facies 33, 1–18.
Reid, R.P., Visscher, P.T., Decho, A.W., Stolz, J.F., Bebout, B.M., Dupraz, C., Macintyre, I.G.,Paerl, H.W., Pinckney, J.L., Prufert-Bebout, L., Steppe, T.F., Des Marais, D.J., 2000. Therole of microbes in accretion, lamination and early lithification of modern marinestromatolites. Nature 406, 989–992.
Revsbech, N.P., Jørgensen, B.B., Blackburn, T.H., Cohen, Y., 1983. Microelectrode studiesof the photosynthesis and O2, H2S, and pH profiles of a microbial mat. Limnologyand Oceanography 28 (6), 1062–1974.
Riding, R., 2000. Microbial carbonates: the geological record of calcified bacterial–algalmats and biofilms. Sedimentology 47, 179–214.
Romanek, C.S., Grossman, E.L., Morse, J.W., 1992. Carbon isotopic fractionation insynthetic aragonite and calcite: effects of temperature and precipitation rate.Geochimica et Cosmochimica Acta 56, 419–430.
Sakata, S., Hayes, J.M., McTaggart, A.R., Evans, R.A., Leckrone, K.J., Togasaki, R.K., 1997.Carbon isotopic fractionation associated with lipid biosynthesis by a cyanobacte-rium: relevance for interpretation of biomarker records. Geochimica et Cosmochi-mica Acta 61, 5379–5389.
Schopf, J.W., 1993. Microfossils of the early Archean Apex chert: new evidence of theantiquity of life. Science 260, 640–646.
Schultze-Lam, S., Harauz, G., Beveridge, T.J., 1992. Participation of a cyanobacterial Slayer in fine-grain mineral formation. Journal of Bacteriology 174, 7971–7981.
Schultze-Lam, S., Fortin, D., Davis, B.S., Beveridge, T.J., 1996. Mineralization of bacterialsurfaces. Chemical Geology 132, 171–181.
Shiraishi, F., Reimer, A., Bissett, A., de Beer, D., Arp, G., 2008. Microbial effects on biofilmcalcification, ambient water chemistry and stable isotope records in a highlysupersaturated setting (Westerhöfer Bach, Germany). Palaeogeography, Palaeocli-matology, Palaeoecology 262 (1–2), 91–106.
Shiraiwa, Y., Goyal, A., Tolbert, N.E., 1993. Alkalization of the medium by unicellulargreen algae during uptake of dissolved inorganic carbon. Plant and Cell Physiology34 (5), 649–657.
Stal, L.J., van Gemerden, H., Krumbein, W.E., 1985. Structure and development of abenthic marine microbial mat. FEMS Microbiology Ecology 31, 111–125.
St-Jean,G., 2003. Automatedquantitative and isotopic (13C) analysis of dissolved inorganiccarbon and dissolved organic carbon in continuous-flow using a total organic carbonanalyser. Rapid Communications in Mass Spectrometry 17, 419–428.
Stolz, J.F., Feinstein, T.N., Salsi, J., Visscher, P.T., Reid, R.P., 2001. TEM analysis ofmicrobial mediated sedimentation and lithification in modern marine stromato-lites. American Mineralogist 86, 826–833.
Sumner, D.Y., 2001. Microbial influences on local carbon isotopic ratios and theirpreservation in carbonate. Astrobiology 1, 57–70.
Thompson, J.B., Ferris, F.G., 1990. Cyanobacterial precipitation of gypsum, calcite, andmagnesite from natural alkaline lake water. Geology 18, 995–998.
Thompson, J.B., Schultze-Lam, S., Beveridge, T.J., Des Marais, D.J., 1997. Whiting events:biogenic origin due to the photosynthetic activity of cyanobacterial picoplankton.Limnology and Oceanography 42 (1), 133–141.
Tyler, S.A., Barghoorn, E.S., 1954. Occurrence of structurally preserved plants in Pre-Cambrian rocks of the Canadian Shield. Science 119, 606–608.
Valero-Garcés, B.L., Delgado-Huertas, A., Ratto, N., Navas, A., 1999. Large 13C enrichmentin primary carbonates from Andean Altiplano lakes, northwest Argentina. Earthand Planetary Science Letters 171, 253–266.
Vasconcelos, C., Warthmann, R., McKenzie, J.A., Visscher, P.T., Bittermann, A.G., van Lith,Y., 2006. Lithifying microbial mats in Lagoa Vermelha, Brazil: modern Precambrianrelics? Sedimentary Geology 185 (3–4), 175–183.
1998. Formation of lithified micritic laminae in modern marine stromatolites(Bahamas): the role of sulfur cycling. American Mineralogist 83, 1482–1493.
Visscher, P.T., Reid, R.P., Bebout, B.M., 2000. Microscale observations of sulfatereduction: correlation of microbial activity with lithified micritic laminae inmodern marine stromatolites. Geology 28, 919–922.
Volkman, J.K., Johns, R.B., Gillan, F.T., Perry, G.J., Bavor, H.J., 1980. Microbial lipids of anintertidal sediment— I. Fatty acids and hydrocarbons. Geochimica et CosmochimicaActa 44, 1133–1143.
Walter, M., 1996. Old fossils could be fractal frauds. Nature 383, 385–386.White, D.C., Davis, W.M., Nickels, J.S., King, J.D., Bobbie, R.J., 1979. Determination of the
sedimentary microbial biomass by extractible lipid phosphate. Oecologia 40, 51–62.Zeebe, R.E., 1999. An explanation of the effect of seawater carbonate concentration on
foraminiferal oxygen isotopes. Geochimica et Cosmochimica Acta 63, 2001–2007.Zelles, L., 1999. Fatty acid patterns of phospholipids and lipopolysaccharides in the
characterisation of microbial communities in soil: a review. Biology and Fertility ofSoils 29, 111–129.
67A.L. Brady et al. / Chemical Geology 274 (2010) 56–67
