15:00 〜 15:15
[AHW22-17] 気候変動下における一回循環湖・琵琶湖の貧酸素化とメタン栄養
キーワード：メタン、メタン酸化細菌、メタン栄養食物網、貧酸素化、部分循環、一回循環
1. Introduction
The increasing number of researches have reported warming trends in the world’s lakes over the last half century but we have poorly understood their ecosystem responses to the climate warming. A large deep ancient Lake Biwa is classified into temperate monomictic lake in which complete mixing takes place in winter. It has been a matter of great concern that the climate warming prevents from the complete mixing and consequently causes hypoxia in the profundal waters, resulting in mass mortality of endemic zoobenthos. Since the hypoxia also enhances methanogenesis, there may be a positive feedback loop between the climate warming and CH4 emission from the lakes. This scenario, however, could be an overestimate if methane oxidizing bacteria (MOB) efficiently assimilate the CH4 and then their CH4-derived carbon is embedded in consumer biomass through trophic interactions, resulting in reduction of atmospheric CH4 emission.
In Lake Biwa, the previous researches reported that the methanogenesis is active in the sediment but CH4 concentration drastically decreases to nano molar level in the hypolimnion due to CH4 oxidation (Miyajima et al. 1997; Murase et al. 2005; Tsunogai et al. 2020). In addition, the MOB were found only from the lake sediment by means of PCR-DGGE analyses and real-time PCR (Tsutsumi et al. 2012; Kojima et al. 2012), suggesting that a hot spot of methane cycling exists in an oxidoreduction boundary layer of lake sediment.
In 2019 which is abnormally warm year, the winter vertical mixing was incomplete. This is the first to observe meromixis through the long-term monitoring by LBERI over more than four decades. Here we report interannual variation in vertical profiles of dissolved DO, CH4 and methanotrophy under different mixing regimes in Lake Biwa through four-year monitoring.
2. Materials & Methods
We conducted monitoring around the deepest point of the north basin in 2016 and at the LBERI’s monitoring station (35°23′41″N, 136°7′57″E; near to the deepest point) in 2017-2019. The monitoring was done at the end (December of 2016-2018) or middle (October of 2019) of thermal stratification period. Water samples were collected at 8 depths from the surface to near-bottom. After measurement of dissolved oxygen (DO) and water temperature, their sub-samples were prepared for measurement of CH4 concentration and its δ13C, and amplicon analysis of bacteria and archaea communities. Particulate organic matter (POM) was collected from surface and bottom layers. Zooplankton was also collected in the epilimnion and hypolimnion separately. These samples were prepared for fatty acid analysis (FAA).
3. Results & Discussion
Profundal waters became hypoxic (1.44 mg-O2/L) only in 2019 when the vertical mixing was incomplete due to warm winter. CH4 concentration showed a maximum in the oxic subsurface with a small peak near the bottom, except for 2016, in which it was the higher in the profundal waters. There was no negative correlation between the profundal DO and CH4 concentrations. In 2016 when the sampling was conducted at the deepest point, a small thermal stratification was detected around 90m at depth. Since hydrothermal vents were recently found to be scattered around the deepest lake bottom, the profundal CH4 peak may be attributed to CH4 emission from the vents rather than methanogenesis in the water column (M. Kumagai unpublished data).
Microbial amplicon analysis revealed that a variety of taxa involved in methanogenesis or methanotrophy, including three types of MOB (Type I and II, and NC10), were distributed broadly in the hypolimnion (50m to the near-bottom) though the amplicon analysis does not necessarily provide the abundance data. FAA revealed that hypolimnetic POM contained a small portion (at most 2.8%) of MOB-specific fatty acids and hypolimnetic zooplankton at most 1.9%, suggesting that CH4-derived carbon is trophically transferred to pelagic food webs. Contrary to our prediction, however, the relationship between profundal hypoxia and the development of methanotrophic food webs was unclear.
Based on isotopic evidence that δ13C-CH4 increased in the hypolimnion, we conclude that CH4 generated from the sediment is substantially consumed by the MOB, resulting in the negligibly low profundal CH4 concentration, except for the limited area of hydrothermal vents. Due to inhibitory effect of light on CH4 oxidation (Murase & Sugimoto 2005), by contrast, the subsurface CH4 may be diffused to the air, accounting for the greater part of atmospheric CH4 emission in Lake Biwa. In conclusion, there are still some scope for methanotrophy to alleviate enlargement of profundal CH4 storage under the warming climate.
The increasing number of researches have reported warming trends in the world’s lakes over the last half century but we have poorly understood their ecosystem responses to the climate warming. A large deep ancient Lake Biwa is classified into temperate monomictic lake in which complete mixing takes place in winter. It has been a matter of great concern that the climate warming prevents from the complete mixing and consequently causes hypoxia in the profundal waters, resulting in mass mortality of endemic zoobenthos. Since the hypoxia also enhances methanogenesis, there may be a positive feedback loop between the climate warming and CH4 emission from the lakes. This scenario, however, could be an overestimate if methane oxidizing bacteria (MOB) efficiently assimilate the CH4 and then their CH4-derived carbon is embedded in consumer biomass through trophic interactions, resulting in reduction of atmospheric CH4 emission.
In Lake Biwa, the previous researches reported that the methanogenesis is active in the sediment but CH4 concentration drastically decreases to nano molar level in the hypolimnion due to CH4 oxidation (Miyajima et al. 1997; Murase et al. 2005; Tsunogai et al. 2020). In addition, the MOB were found only from the lake sediment by means of PCR-DGGE analyses and real-time PCR (Tsutsumi et al. 2012; Kojima et al. 2012), suggesting that a hot spot of methane cycling exists in an oxidoreduction boundary layer of lake sediment.
In 2019 which is abnormally warm year, the winter vertical mixing was incomplete. This is the first to observe meromixis through the long-term monitoring by LBERI over more than four decades. Here we report interannual variation in vertical profiles of dissolved DO, CH4 and methanotrophy under different mixing regimes in Lake Biwa through four-year monitoring.
2. Materials & Methods
We conducted monitoring around the deepest point of the north basin in 2016 and at the LBERI’s monitoring station (35°23′41″N, 136°7′57″E; near to the deepest point) in 2017-2019. The monitoring was done at the end (December of 2016-2018) or middle (October of 2019) of thermal stratification period. Water samples were collected at 8 depths from the surface to near-bottom. After measurement of dissolved oxygen (DO) and water temperature, their sub-samples were prepared for measurement of CH4 concentration and its δ13C, and amplicon analysis of bacteria and archaea communities. Particulate organic matter (POM) was collected from surface and bottom layers. Zooplankton was also collected in the epilimnion and hypolimnion separately. These samples were prepared for fatty acid analysis (FAA).
3. Results & Discussion
Profundal waters became hypoxic (1.44 mg-O2/L) only in 2019 when the vertical mixing was incomplete due to warm winter. CH4 concentration showed a maximum in the oxic subsurface with a small peak near the bottom, except for 2016, in which it was the higher in the profundal waters. There was no negative correlation between the profundal DO and CH4 concentrations. In 2016 when the sampling was conducted at the deepest point, a small thermal stratification was detected around 90m at depth. Since hydrothermal vents were recently found to be scattered around the deepest lake bottom, the profundal CH4 peak may be attributed to CH4 emission from the vents rather than methanogenesis in the water column (M. Kumagai unpublished data).
Microbial amplicon analysis revealed that a variety of taxa involved in methanogenesis or methanotrophy, including three types of MOB (Type I and II, and NC10), were distributed broadly in the hypolimnion (50m to the near-bottom) though the amplicon analysis does not necessarily provide the abundance data. FAA revealed that hypolimnetic POM contained a small portion (at most 2.8%) of MOB-specific fatty acids and hypolimnetic zooplankton at most 1.9%, suggesting that CH4-derived carbon is trophically transferred to pelagic food webs. Contrary to our prediction, however, the relationship between profundal hypoxia and the development of methanotrophic food webs was unclear.
Based on isotopic evidence that δ13C-CH4 increased in the hypolimnion, we conclude that CH4 generated from the sediment is substantially consumed by the MOB, resulting in the negligibly low profundal CH4 concentration, except for the limited area of hydrothermal vents. Due to inhibitory effect of light on CH4 oxidation (Murase & Sugimoto 2005), by contrast, the subsurface CH4 may be diffused to the air, accounting for the greater part of atmospheric CH4 emission in Lake Biwa. In conclusion, there are still some scope for methanotrophy to alleviate enlargement of profundal CH4 storage under the warming climate.