Inland waters (lakes, reservoirs, streams, and rivers) are often substantial methane (CH4) sources in the terrestrial landscape. They are, however, not yet well integrated in global greenhouse gas (GHG) budgets. Data from 474 freshwater ecosystems and the most recent global water area estimates indicate that freshwaters emit at least 103 teragrams of CH4 year−1, corresponding to 0.65 petagrams of C as carbon dioxide (CO2) equivalents year−1, offsetting 25% of the estimated land carbon sink. Thus, the continental GHG sink may be considerably overestimated, and freshwaters need to be recognized as important in the global carbon cycle.Original Publication:David Bastviken, Lars J. Tranvik, John A. Downing, Patrick M. Crill and Alex Enrich-Prast, Freshwater Methane Emissions Offset the Continental Carbon Sink, 2011, Science, (331), 6013, 50-50.http://dx.doi.org/10.1126/science.1196808Copyright: American Association for the Advancement of Sciencehttp://www.aaas.org

Wetlands are the largest global source of atmospheric methane (CH), a potent greenhouse gas. However, methane emission inventories from the Amazon floodplain, the largest natural geographic source of CH in the tropics, consistently underestimate the atmospheric burden of CH determined via remote sensing and inversion modelling, pointing to a major gap in our understanding of the contribution of these ecosystems to CH emissions. Here we report CH fluxes from the stems of 2,357 individual Amazonian floodplain trees from 13 locations across the central Amazon basin. We find that escape of soil gas through wetland trees is the dominant source of regional CH emissions. Methane fluxes from Amazon tree stems were up to 200 times larger than emissions reported for temperate wet forests and tropical peat swamp forests, representing the largest non-ebullitive wetland fluxes observed. Emissions from trees had an average stable carbon isotope value (δC) of -66.2 ± 6.4 per mil, consistent with a soil biogenic origin. We estimate that floodplain trees emit 15.1 ± 1.8 to 21.2 ± 2.5 teragrams of CH a year, in addition to the 20.5 ± 5.3 teragrams a year emitted regionally from other sources. Furthermore, we provide a 'top-down' regional estimate of CH emissions of 42.7 ± 5.6 teragrams of CH a year for the Amazon basin, based on regular vertical lower-troposphere CH profiles covering the period 2010-2013. We find close agreement between our 'top-down' and combined 'bottom-up' estimates, indicating that large CH emissions from trees adapted to permanent or seasonal inundation can account for the emission source that is required to close the Amazon CH budget. Our findings demonstrate the importance of tree stem surfaces in mediating approximately half of all wetland CH emissions in the Amazon floodplain, a region that represents up to one-third of the global wetland CH source when trees are combined with other emission sources.

Freshwater environments contribute 75% of the natural global methane (CH(4)) emissions. While there are indications that tropical lakes and reservoirs emit 58-400% more CH(4) per unit area than similar environments in boreal and temperate biomes, direct measurements of tropical lake emissions are scarce. We measured CH(4) emissions from 16 natural shallow lakes in the Pantanal region of South America, one of the world's largest tropical wetland areas, during the low water period using floating flux chambers. Measured fluxes ranged from 3.9 to 74.2 mmol m(-2) d(-1) with the average from all studied lakes being 8.8 mmol m(-2) d(-1) (131.8 mg CH(4) m(-2) d(-1)), of which ebullition accounted for 91% of the flux (28-98% on individual lakes). Diel cycling of emission rates was observed and therefore 24-h long measurements are recommended rather than short-term measurements not accounting for the full diel cycle. Methane emission variability within a lake may be equal to or more important than between lake variability in floodplain areas as this study identified diverse habitats within lakes having widely different flux rates. Future measurements with static floating chambers should be based on many individual chambers distributed in the various subenvironments of a lake that may differ in emissions in order to account for the within lake variability.

15N and microsensor techniques. Anammox rates estimated with microsensors were less than 22% of the rates measured with isotopes. It is suggested that this discrepancy was due to the presence of fauna, because the applied 15 N technique captures total N 2 production while the microsensor technique only captures diffusion-controlled N 2 production at the sediment surface. This hypothesis was verified by consistent agreement between the methods when applied to defaunated sediments.

Inland water sediments receive large quantities of terrestrial organic matter [1][2][3][4][5] and are globally important sites for organic carbon preservation [5][6] . Sediment organic matter mineralization is positively related with temperature across a wide range of high-latitude ecosystems [6][7][8][9][10] , but the situation in the tropics remains unclear. Here we assessed temperature effects on the biological production of CO 2 and CH 4 in anaerobic sediments of tropical lakes in the Amazon and boreal lakes in Sweden. Based on conservative regional warming projections until 2100 11 , we estimate that sediment CO 2 and CH 4 production will increase 9-61 % above present rates. Combining the CO 2 and CH 4 as CO 2 equivalents (CO 2eq ) 11 , the predicted increase is 2.4 -4.5 times higher in tropical than boreal sediments. Although the estimated lake area in low latitudes is 3.2 times smaller than that of the boreal zone, we estimate that the increase in gas production from tropical lake sediments would be on an average 2.4 times higher for CO 2 and 2.8 times higher for CH 4 . The exponential temperature response of organic matter mineralization, coupled with higher increases in the proportion of CH 4 relative to CO 2 upon warming, suggests that the production of GHGs in tropical sediments will increase substantially. This represents a potential large-scale positive feedback to climate change. 2 Main textTropical and boreal biomes harbour approximately 50 % of the lakes on Earth 12 . These inland waters emit substantial amounts of carbon dioxide (CO 2 ; in the order of 0.5 Pg yr -1 ) 1,4,13,14 and methane (CH 4 ; 70 Tg yr -1 ) 15 . Organic matter escapes mineralization via burial in lake sediments, representing a global carbon (C) sink [13][14][15] . Cold conditions are favouring organic carbon (OC) preservation in lakes at northern latitudes [8][9][10]16 , whereas warm inland waters show intense organic degradation supporting high C emissions to the atmosphere 4,5,17,18 .Temperature and OC mineralization were recently shown to be strongly positively related in boreal lake sediments overlain by oxic water 9 . However, the majority of freshwater sediments below the uppermost layer (typically a few mm) are anoxic 19 , where the anaerobic biological degradation of OC releases not only CO 2 but also significant amounts of CH 4 15 . Although higher temperatures are also expected to increase metabolic responses 20 , the effects of changing temperatures on OC mineralization can depend on several factors including organic matter characteristics (e.g. the Carbon-QualityTemperature hypothesis) 21 . Thus, the temperature sensitivity of OC stocks at high latitudes previously reported [6][7][8]16 , may not be valid in the tropics where temperature sensitivity data is much more scarce 22 . We compared the anaerobic OC mineralization to CO 2 and CH 4 in tropical and boreal lake sediments along a temperature gradient. We simultaneously sampled a wide range of lake sediments from both tropical and boreal zones (see Suppl...

[1] On the basis of a broad compilation of data on pCO 2 in surface waters, we show tropical lakes to be, on average, far more supersaturated and variable in CO 2 (geometric mean ± SE pCO 2 = 1804 ± 35 matm) than temperate lakes (1070 ± 6 matm). There was a significant negative relationship between pCO 2 and latitude, resulting in an average decrease of pCO 2 by 2.8 ± 0.5% per degree latitude. In addition, we found a general positive relationship between pCO 2 and water temperature across lakes involving an average increase (±SE) in 6.7 ± 0.8% per°C. A conservative annual efflux from global lakes to the atmosphere was reestimated to 0.44 Gt C. Our results show tropical lakes maintain large CO 2 disequilibria with the atmosphere, playing a disproportionate and variable role in the flux of CO 2 between lakes and the atmosphere, thereby being a significant component of the global C cycle.

Methane is an important end product of degradation of organic matter in anoxic lake sediments. Methane is mainly produced by either reduction of CO₂ or cleavage of acetate involving different methanogenic archaea. The contribution of the different methanogenic paths and of the diverse bacteria and archaea involved in CH₄ production exhibits a large variability that is not well understood. Lakes in tropical areas, e.g. in Brazil, are wetlands with high potential impact on the global CH₄ budget. However, they have hardly been studied with respect to methanogenesis. Therefore, we used samples from 16 different lake sediments in the Pantanal and Amazon region of Brazil to measure production of CH₄, CO₂, analyze the content of ¹³C in the products and in intermediately formed acetate, determine the abundance of bacterial and archaeal microorgansisms and their community composition and diversity by targeting the genes of bacterial and archaeal ribosomal RNA and of methyl coenzyme M reductase, the key enzyme of methanogenic archaea. These experiments were done in the presence and absence of methyl fluoride, an inhibitor of acetoclastic methanogenesis. While production rates of CH₄ and CO₂ were correlated to the content of organic matter and the abundance of archaea in the sediment, values of ¹³C in acetate, CO₂, and CH₄ were related to the ¹³C content of organic matter and to the path of CH₄ production with its intrinsic carbon isotope fractionation. Isotope fractionation was small (average 10‰) for conversion of C_org to acetate-methyl, which was hardly further fractionated during CH₄ production. However, fractionation was strong for CO₂ conversion to CH₄ (average 75‰), which generally accounted for >50% of total CH₄ production. Canonical correspondence analysis did not reveal an effect of microbial community composition, despite the fact that it exhibited a pronounced variability among the different sediments

Natural lakes and reservoirs are important yet not well‐constrained sources of greenhouse gasses to the atmosphere. In particular for N2O emissions, a huge variability is observed in the few, observation‐driven flux estimates that have been published so far. Recently, a process‐based, spatially explicit model has been used to estimate global N2O emissions from more than 6,000 reservoirs based on nitrogen (N) and phosphorous inflows and water residence time. Here we extend the model to a data set of 1.4 million standing water bodies comprising natural lakes and reservoirs. For validation, we normalized the simulated N2O emissions by the surface area of each water body and compared them against regional averages of N2O emission rates taken from the literature or estimated based on observed N2O concentrations. We estimate that natural lakes and reservoirs together emit 4.5 ± 2.9 Gmol N2O‐N year−1 globally. Our global‐scale estimate falls in the far lower end of existing, observation‐driven estimates. Natural lakes contribute only about half of this flux, although they contribute 91% of the total surface area of standing water bodies. Hence, the mean N2O emission rates per surface area are substantially lower for natural lakes than for reservoirs with 0.8 ± 0.5 versus 9.6 ± 6.0 mmol N·m−2·year−1, respectively. This finding can be explained by on average lower external N inputs to natural lakes. We conclude that upscaling‐based estimates, which do not distinguish natural lakes from reservoirs, are prone to important biases.