6. DISCUSSION
6.1 TEMPERATURE AND OXYGEN CONCENTRATIONS
Summer stratification in Lake Kuivajärvi developed late in summer 2016, and the water column was oxygenated still in July. The rainy summer could have affected water column stability by enabling water column mixing. In August, hypoxic conditions were detected under 6 m depth. In the beginning of September, the hypolimnion below 9 m was below the detection limit of the O2-sensor. Previously in Lake Kuivajärvi a duration of hypolimnetic anoxia has varied from 3 weeks in 2012 to over 2 months in 2011 (Miettinen et al. 2015). Since anoxia did not begin until the beginning of September, it most likely lasted only few weeks before the autumnal turnover of water masses.
One possibility for the short and late period of bottom anoxia is that during the warm spring Kuivajärvi began to thermally stratify and rapidly formed a warm epilimnetic layer due to humic waters absorbing solar radiation, while the bottom waters remained cold and oxygenated. Since the average temperatures during spring months have clearly increased in Finland due to climate change (Mikkonen et al. 2015), this kind of early thermal stratification with well-oxygenated water column can be quite common in humic lakes.
6.2 NUTRIENT CONCENTRATIONS
The concentrations of SO42- mainly corresponded average concentrations in freshwaters (Wetzel 2001). The hypolimnetic concentrations of SO42- peaked in July. These increased concentrations in July may indicate SO42- release from the sediment. In oxygenated conditions, reoxidation of reduced S2- to SO42- may turn the freshwater sediments from a sink to a source of SO42- to the overlying water (Holmer and Storkholm 2001). This kind of hypolimnetic increase in SO42- is pronounced in the early phases of summer stratification, when there is still oxygen available in the bottom (Wetzel 2001).
The NH4+ concentrations were quite similar to other studies that have determined NH4+ in boreal lakes during summer (e.g. Rissanen et al. 2013). In July, NH4+ was detected in the epilimnion in the absence of NO3-, which is interesting, considering that in freshwater systems NH4+ uptake by algae is expected to be dominative compared to NO3- uptake (e.g. Présing et al. 2001). The NO3
-concentrations were generally low compared to usual -concentrations in Finnish lakes (e.g. Arvola et al. 1996), but the results were in good agreement with previous measurements in Kuivajärvi
(Miettinen et al. 2015). The increased concentrations of nitrogen in the poorly oxygenated hypolimnion indicate that decomposition of organic matter has consumed oxygen and produced NH4+. In addition, nitrification (oxidation of NH4+ to NO2- and further to NO3-) has resulted in the increased NO3- concentrations at the oxic-anoxic interface (Rantakari and Kortelainen 2005).
The measured DOC concentrations were typical for a boreal humic lake (Lopez Bellido 2013; Rasilo et al. 2012) and comparable with those previously measured in Kuivajärvi (11.8-14.1 mg l-1; Miettinen et al. 2015). The DOC concentrations were highest in the surface water, which indicates allochthonous carbon loading from the drainage basin. DOC may have originated from peatlands or forest soil in the catchment area.
Total Fe concentrations were relatively high in the hypolimnion. However, it is common that in humic waters the concentrations of Fe increase during the period of bottom anoxia (Wetzel 2001).
6.3 CH4 AND CO2 CONCENTRATIONS
The epilimnetic concentrations of CH4 were rather similar to those previously measured from boreal lakes in Finland (see comparison in Table 7), but the CH4 concentrations in the hypolimnion were relatively low in summer 2016. Although the CH4 concentrations did not correlate with oxygen conditions, the statistical relationships between CH4 and other variables regulating the water column stratification and productivity (e.g. temperature and nutrient concentrations) were clear.
According to the study of 207 Finnish boreal lakes by Juutinen et al. (2009), the highest concentrations of CH4 usually appear in the water layers closest to sediment during late summer. In September, the hypolimnetic concentrations of CH4 were indeed ~10 times higher than in the epilimnion and metalimnion, most likely due to the ongoing period of bottom anoxia that allowed favourable conditions for methanogenesis. However, from May to August the highest concentrations of CH4 occurred in the upper water layers, and the lowest concentrations in the hypolimnion. These concentration profiles are somewhat unexpected, because the CH4 concentrations usually decrease in the well-oxygenated water column due to oxidation by methanotrophic bacteria. Hence, it seems that the CH4 in the epilimnion and metalimnion originated from outside the lake. For example, Miettinen et al. (2015) and Lopez Bellido et al. (2013) reported a similar phenomenon. They both suggested that since the CH4 concentrations were lowest in the hypolimnion, the source of CH4 in the epilimnion was not sediment, but instead, laterally transported CH4 from the littoral zone or surrounding
peatlands (Lopez Bellido et al. 2013; Miettinen et al. 2015; Ojala et al. 2011). However, there were no extreme rainfall episodes during the sampling periods, and without extreme rainfalls the lateral transport would not be very efficient. Hence, some internal source of carbon could be more likely.
It is not completely exceptional to observe the water column CH4 maxima in the presence of oxygen.
Several studies have previously confirmed CH4 supersaturation also in the well-oxygenated water column, which is often explained by in situ methanogenesis within microanoxic zones (Grossart et al. 2011). Methanogens might function in well-oxygenated environments if they are associated with possible microanoxic habitats, such as digestive tracts of zooplankton or fecal pellets (De Angelis and Lee 1994; Oremland 1979). Grossart et al. (2011) also discovered direct attachment of potentially methanogenic archaea to photoautotrophic cells in the oxygenated epilimnion, which may enable anaerobic growth and transport of substrates (H2 and acetate) for CH4 production.
On the other hand, in marine waters the same paradox has been explained by CH4 production through aerobic bacterial degradation of phosphonate esters in semi-labile DOM (Repeta et al. 2016). Repeta et al. (2016) suggested that since freshwaters accumulate chemically corresponding semi-labile DOM polysaccharides that most likely incorporate methylphosphonate esters, similar mechanism could be feasible also in freshwater lakes.
Table 7. Mean and median CH4 concentrations (µmol l-1) in surface (1 m below the surface) and bottom water (11.5-11.75 m) during spring (May) and late summer (August-September) compared to corresponding results of Juutinen et al. (2009) from 207 Finnish lakes.
Depth Season Mean
There was a positive correlation between the CH4 and NH4+ concentrations. Both CH4 and NH4+
production occur primarily in anoxic conditions, which may explain this relationship, but it is also possible that methanotrophic activity was affected by the presence of NH4+ (Dunfield and Knowles 1995). Since methane-oxidizing and ammonia-oxidizing bacteria are able to switch substrates due to evolutionarily related enzymes responsible for CH4 and NH4+ oxidation, CH4 oxidation might be inhibited by the high concentrations of nitrogen (Dunfield and Knowles 1995; Hanson and Hanson 1996), thus increasing the concentrations of CH4. However, it is uncertain whether the measured concentrations of NH4+ in Lake Kuivajärvi would have been sufficient for the inhibition of CH4
oxidation.
The concentrations of CO2 were in good agreement with previous measurements from Kuivajärvi (Miettinen et al. 2015). Based on the results of the correlation analysis, the CO2 concentration seems to increase with depth and decreasing temperature, pH and oxygen concentration. The negative correlation between CO2 and O2 indicates consumption of oxygen and production of CO2 through in-situ decomposition of organic matter in the hypolimnion (Miettinen et al. 2015). The decomposition of OM also releases nutrients, which can thus explain the positive correlation for CO2 with NO3- and SO42- (Rantakari and Kortelainen 2005).
The positive correlation for CO2 with Fe3+ is quite unexpected, since some previous studies have suggested that Fe3+ addition may suppress CO2 production. According to Karvinen et al. (2014), Fe can form organoferric complexes with organic matter, which may inhibit the microbial degradation of organic compounds and decrease the availability of substrates for CH4 and CO2 production. On the other hand, the availability of Fe3+ may also lead to enhanced anaerobic respiration of OM (Lalonde et al. 2012; Lovley and Phillips 1986), thus releasing CO2. However, since Fe was measured only in September, the observed positive correlation may be one-time event and not a permanent feature.
6.4 STABLE ISOTOPE COMPOSITION (δ13C-CH4 AND δ13C-DIC) AND THE POTENTIAL CH4 OXIDATION RATES
The measured δ13C-CH4 values are in good agreement with previous studies in freshwaters (Bastviken et al. 2002; Kankaala et al. 2007; Milucka et al. 2015; Nykänen et al. 2014). Correlation analysis showed a negative correlation for δ13C-CH4 not only with the CH4 concentration, but also with factors possibly affecting CH4 production: temperature and DOC (Bastviken et al. 2008; Bogard et al. 2014).
In every sampling month, δ13C-CH4 slightly increased after 6 m depth simultaneously with the decreasing CH4 concentrations, and in August and September δ13C-CH4 decreased again in the hypolimnion. As in the study of Nykänen et al. (2014), the lowest δ13C-CH4 was generally measured at the lake bottom where CH4 was formed, and this hypolimnetic decrease of δ13C-CH4 was substantial particularly in September with maximum CH4 concentrations. This is consistent with the fact that biogenic CH4 is strongly 13C-depleted due to fractionation (Whiticar 1999). The fractionation factors (αCO2–CH4) between average δ13C-CO2 and δ13C-CH4 in the bottom water were under a threshold limit of 1.055, suggesting that acetoclastic methanogenesis was the dominant source of CH4
throughout the study period (Whiticar et al. 1986). This conclusion is in agreement with the current knowledge of methanogenic pathways, because acetatoclastic methanogenesis is considered to be the dominant pathway for methanogenesis in freshwater lakes (Bouchard et al. 2015; Wik 2016).
However, there was no hypolimnetic decrease of δ13C-CH4 in July, but instead, CH4 remained enriched in 13C at the bottom. This result suggests that during July CH4 was already oxidized in the sediment, because the CH4 oxidation leaves a residue of CH4 enriched in 13C. In other words, the lighter carbon isotope was consumed, while the proportion of the heavier carbon isotope increased (Bastviken et al. 2002). Conversely, during August and September the most 13C-enriched values were detected in the vicinity of the oxycline, thus indicating the transition of CH4 oxidation from the sediment to the water column.
The increased δ13C-CH4 values during July occurred simultaneously with the elevated SO42-
concentrations in the hypolimnion. For example, Yoshinaga et al. (2014) claimed that residual CH4
in marine sediments became enriched in 13C at SO42- concentrations above 0.5 mmol l-1, and depleted in 13C below this threshold limit. They suggested that sulphate-limited AOM can lead to equilibrium isotope exchange and thus 13C-depletion in sulphate-methane transition zones. Our results show a similar positive correlation between δ13C-CH4 and SO42-, but it has to be considered that the SO42-
concentrations are much lower in freshwaters (in Kuivajärvi max. 0.094 mmol l-1) and so far the known anaerobic methanotrophs have not been detected in anoxic freshwaters (Milucka et al. 2015;
Oswald et al. 2015).
Furthermore, there was a negative correlation between the concentrations of SO42- and CH4, which could imply that in July methanogenesis was inhibited by the high SO42-concentrations, because sulphate-reducing bacteria are able to use same substrates than acetoclastic methanogens and use H2
as an electron source (Rudd and Hamilton 1978). The lower SO42-concentrations in August and September could have led to stronger CH4 build-up due to more favourable conditions for
methanogenesis. After all, both of these above-mentioned correlations could simply be explained by oxidation of S2- to SO42-simultaneously with the CH4 oxidation leaving the residual methane 13 C-enriched.
The variations of δ13C-DIC with depth were most likely connected to biological processes such as photosynthesis and organic matter oxidation. In the deeper water column the decrease of δ13C-DIC, O2, pH and temperature was accompanied by slightly increased nutrient concentrations. This is consistent with the fact that photosynthetic activity decreases from the euphotic zone to the metalimnion, while the role of respiration and organic matter oxidation becomes more dominant in deeper water layers (Assayag et al. 2008).
The negative correlation between δ13C-DIC and CO2 can also be explained by organic matter oxidation. In the euphotic zone the slight preference for 12CO2 uptake (isotopic fractionation) by photosynthesis leads to 13C-enrichment of epilimnetic DIC (Nykänen et al. 2014). In contrast, in the hypolimnion the accumulated organic matter is oxidized into DIC, which leads to increase of the 13 C-depleted CO2 and decrease of δ13C-DIC (Assayag et al. 2008; Hanson et al. 2006). The influence of organic matter oxidation can be seen especially in August and September measurements, where the CO2 concentration peaked and δ13C-DIC decreased at the depths of 6 m and 7 m.
The slight decrease of δ13C-DIC from summer to autumn indicates the effects of seasonal variation in biological activity. In July, the higher epilimnetic δ13C-DIC refers to enhanced photosynthesis due to more favourable light conditions compared to autumn. Furthermore, during fall the photosynthetic activity decreases due to lower light intensity, thus leading to decrease of δ13C-DIC in the epilimnion (Assayag et al. 2008), which is consistent with δ13C-DIC values measured in September.
In September, CH4 oxidation was detected in the deep water column. In September sampling, the estimated proportion of all CH4 consumed in the water column of Lake Kuivajärvi was 61 %. Thus, 39 % of produced CH4 could have been released to the atmosphere. This estimation is in agreement with previous studies: e.g. on an annual basis, 79 % was oxidized in the water column of Lake Valkea-Kotinen (Kankaala et al. 2006a).
Potential CH4 oxidation rates gradually increased from the oxic-anoxic interface (8 m) to the anoxic hypolimnion (11.5 m), which was seen simultaneously with a strong decrease of δ13C-CH4 while δ13C-DIC remained stable. These findings are quite unexpected, since highest CH4 oxidation rates are typically observed at the oxic-anoxic interfaces with high gradients of O2 and CH4 (Oswald et al.
2015) and oxygen limitation generally decreases methanotrophic activity (Kankaala 2006a).
Nevertheless, the CH4 depth profile showed peaked concentrations in 11.5 m, which may have enhanced consumption of CH4 in the hypolimnion (Bastviken et al. 2008).
These CH4 oxidation measurements hint that CH4 was oxidized anaerobically with an alternative electron acceptor, but the concentrations of SO42- or NO3- during September are not necessarily high enough to enable anaerobic oxidation in the hypolimnion. Fe3+-dependent AOM, however, cannot be completely excluded, because the total Fe concentrations at the bottom were rather high and Fe3+ as an electron acceptor may be more significant than previously thought (Beal et al. 2009).
Nevertheless, it is more likely that there was still some oxygen available in sub-micromolar concentrations below the detection limit of the O2-sensor (Blees et al. 2014). Even small concentrations of oxygen might be sufficient to drive aerobic CH4 oxidation in the water column.
According to Kalyuzhnaya et al. (2013), aerobic methanotrophs only require oxygen for the activation of CH4 oxidation, whereas the oxidation products, such as formaldehyde, could be utilized through anaerobic respiration or fermentation. It is also known that nitrite-reducing bacteria can provide oxygen to CH4 oxidation through O2-producing intra-aerobic methanotrophy pathway (Ettwig et al.
2010; Padilla et al. 2016).
Taken together, it seems that some temporal supply of oxygen might have occurred in seemingly anoxic zone, thus allowing aerobic CH4 oxidation. Different weather and wind events, such as internal waves, can sometimes introduce oxygenated water into otherwise anoxic layers (Oswald et al. 2015), and that might have been the case in Lake Kuivajärvi.
7 CONCLUSIONS
The summer stratification and hypolimnetic anoxia developed late in summer 2016, which affected CH4 dynamics in Kuivajärvi. The well-oxygenated water column decreased hypolimnetic CH4
concentrations, until the bottom anoxia in September resulted in maximum concentrations of CH4 in the hypolimnion. In early summer, higher CH4 concentrations in the epilimnion and metalimnion compared to the hypolimnetic concentrations could indicate lateral transport of CH4 from the catchment or littoral zone, or in situ CH4 production within microanoxic zones.
Stable isotope analyses showed seasonal variation in biological processes, such as photosynthesis in the epilimnion and OM decomposition in the deeper water layers. These processes are strongly regulating the production of greenhouse gases in the lake. The seasonal changes in δ13C-CH4
indicated that the zone of CH4 oxidation ascended from the sediment to the hypolimnion in the late summer.
Our results revealed that in September the CH4 oxidation potential was highest in the anoxic hypolimnion. Since anaerobic oxidation of CH4 taking place in Lake Kuivajärvi seems unlikely, it is possible that temporal micro-oxic zone in otherwise anaerobic conditions could have allowed CH4
oxidation.
In September, approximately 60 % of produced CH4 was estimated to be oxidized in the water column, while 40 % possibly escaped methanotrophs and was lost to the atmosphere. Even though boreal lakes can be a notable source of atmospheric CH4 particularly during the period of bottom anoxia, methanotrophic bacteria are able to significantly reduce CH4 emissions in changing climate.
REFERENCES
Algesten, G., Sobek, S., Bergström, A.K., Ågren, A., Tranvik, L.J. and Jansson, M., 2004. Role of lakes for organic carbon cycling in the boreal zone. Global change biology 10: 141-147.
Arvola, L., Kankaala, P., Tulonen, T. and Ojala, A., 1996. Effects of phosphorus and allochthonous humic matter enrichment on metabolic processes and community structure of plankton in a boreal lake (Lake Pääjärvi). Canadian Journal of Fisheries and Aquatic Sciences 53: 1646-1662.
Aronson, E.L., Allison, S.D. and Helliker, B.R., 2013. Environmental impacts on the diversity of methane-cycling microbes and their resultant function. Frontiers in microbiology 4: 225.
Assayag, N., Jézéquel, D., Ader, M., Viollier, E., Michard, G., Prévot, F. and Agrinier, P., 2008.
Hydrological budget, carbon sources and biogeochemical processes in Lac Pavin (France):
constraints from δ18O of water and δ13C of dissolved inorganic carbon. Applied Geochemistry 23:
2800-2816.
Bastviken, D., Ejlertsson, J. and Tranvik, L., 2002. Measurement of methane oxidation in lakes: a comparison of methods. Environmental science & technology 36: 3354-3361.
Bastviken, D., Ejlertsson, J., Sundh, I. and Tranvik, L., 2003. Methane as a source of carbon and energy for lake pelagic food webs. Ecology 84: 969-981.
Bastviken, D., Cole, J., Pace, M. and Tranvik, L., 2004. Methane emissions from lakes: Dependence of lake characteristics, two regional assessments, and a global estimate. Global biogeochemical cycles 18.
Bastviken, D., Cole, J.J., Pace, M.L. and Van de Bogert, M.C., 2008. Fates of methane from different lake habitats: connecting whole-lake budgets and CH4 emissions. Journal of Geophysical Research:
Biogeosciences 113.
Bastviken, D., Tranvik, L.J., Downing, J.A., Crill, P.M. and Enrich-Prast, A., 2011. Freshwater methane emissions offset the continental carbon sink. Science 331: 50-50.
Beal, E.J., House, C.H. and Orphan, V.J., 2009. Manganese-and iron-dependent marine methane oxidation. Science 325: 184-187.
Bianchi, T.S. and Canuel, E.A., 2011. Chemical biomarkers in aquatic ecosystems. Princeton University Press.
Bigeleisen, J., 1965. Chemistry of isotopes. Science 147: 463-471.
Birge, E.A. and Juday, C., 1927. The organic content of the water of small lakes. Proceedings of the American Philosophical Society 66: 357-372.
Blees, J., Niemann, H., Wenk, C.B., Zopfi, J., Schubert, C.J., Kirf, M.K., Veronesi, M.L., Hitz, C.
and Lehmann, M.F., 2014. Micro-aerobic bacterial methane oxidation in the chemocline and anoxic water column of deep south-Alpine Lake Lugano (Switzerland). Limnology and Oceanography 59:
311-324.
Bogard, M.J., Del Giorgio, P.A., Boutet, L., Chaves, M.C.G., Prairie, Y.T., Merante, A. and Derry, A.M., 2014. Oxic water column methanogenesis as a major component of aquatic CH4 fluxes. Nature communications 5.
Borrel, G., Jézéquel, D., Biderre-Petit, C., Morel-Desrosiers, N., Morel, J.P., Peyret, P., Fonty, G. and Lehours, A.C., 2011. Production and consumption of methane in freshwater lake ecosystems.
Research in microbiology 162: 832-847.
Borrel, G., O’Toole, P.W., Harris, H.M., Peyret, P., Brugère, J.F. and Gribaldo, S., 2013.
Phylogenomic data support a seventh order of methylotrophic methanogens and provide insights into the evolution of methanogenesis. Genome biology and evolution 5: 1769-1780.
Bouchard, F., Laurion, I., Preskienis, V., Fortier, D., Xu, X. and Whiticar, M.J., 2015. Modern to millennium-old greenhouse gases emitted from freshwater ecosystems of the eastern Canadian Arctic.
Biogeosciences Discuss 12: 11661-11705.
Bousquet, P., Ciais, P., Miller, J.B., Dlugokencky, E.J., Hauglustaine, D.A., Prigent, C., Van der Werf, G.R., Peylin, P., Brunke, E.G., Carouge, C. and Langenfelds, R.L., 2006. Contribution of anthropogenic and natural sources to atmospheric methane variability. Nature 443: 439-443.
Caldwell, S.L., Laidler, J.R., Brewer, E.A., Eberly, J.O., Sandborgh, S.C. and Colwell, F.S., 2008.
Anaerobic oxidation of methane: mechanisms, bioenergetics, and the ecology of associated microorganisms. Environmental science & technology 42: 6791-6799.
Capone, D.G. and Kiene, R.P., 1988. Comparison of microbial dynamics in marine and freshwater sediments: Contrasts in anaerobic carbon catabolism. Limnology and Oceanography 33: 725-749.
Carlson, R.E. 1977. A trophic state index for lakes. Limnology and oceanography 22: 361-369.
Chaudhary, P.P., Brablcová, L., Buriánková, I. and Rulík, M., 2013. Molecular diversity and tools for deciphering the methanogen community structure and diversity in freshwater sediments. Applied microbiology and biotechnology 97: 7553-7562.
Cole, J.J., Pace, M.L., Carpenter, S.R. and Kitchell, J.F., 2000. Persistence of net heterotrophy in
Cole, J.J., Pace, M.L., Carpenter, S.R. and Kitchell, J.F., 2000. Persistence of net heterotrophy in