Search

Question Measurements of ecosystem carbon (C) exchange are usually labour-intensive and expensive. In peatlands, these temporally and spatially limited measurements are often up-scaled over comparable ecosystems, such as ombrotrophic bogs, to provide an estimate for ecosystem level carbon dioxide (CO2) fluxes. Peatlands typically have moisture variations reflected in the presence of microforms (e.g. hummocks and hollows), each with characteristic plant life forms. However, so far the applicability of peatland type (e.g. bog, fen) or microforms for up-scaling has not been assessed. Does the vegetation composition or function of associated species differ so greatly between the same types of peatland that up-scaling is impossible? Location Five ombrotrophic bogs in Northern Europe; in Ireland, Finland and western Russia. Methods We described the variation in vegetation of microforms within and between ombrotrophic bogs using multivariate analyses. Thereafter, we measured CO2 exchange at different microforms and evaluated the relationship between vegetation structure and the light response of photosynthesis. Results Our results show that the community composition of hummocks, lawns and hollows was rather uniform at the plant life-form level. The photosynthetic capacity per leaf area unit was quite similar within microform classes over the different bogs. The observed differences between sites in capacity were mainly related to variation in leaf area. Conclusions A reliable estimate of ecosystem-level photosynthesis requires knowledge of the proportion of different microforms in an area and the leaf area characteristics for each microform in the year(s) in question. Assessments of ecosystem-level photosynthesis are important with regard to current and future changes in climate, as the most dramatic changes in peatlands involve water level drawdown, which in turn is likely to lead to changes in the relative proportions of microforms within peatlands.

Hot spots of CH4 emissions are a typical feature of pristine peatlands at the microsite and landscape scale. To determine whether rewetting and lake construction in a cutaway peatland would result in the re-creation of hot spots, we first measured CH4 fluxes over a 2-year period with static chambers and estimated annual emissions. Second, to assess whether rewetting and lake creation would produce hot spots at the landscape level, we hypothesized a number of alternative land use scenarios for the peatland following the cessation of peat extraction. Using the results from this study and other studies from literature, we calculated the global warming potential (GWP) of each scenario and the respective contribution of CH4. The results showed that hot spots of CH4 fluxes were observed as a consequence of microsite-specific differences in water table (WT) position and plant productivity. CH4 fluxes were closely related to peat temperature at 10 cm depth and WT position. Annual emissions ranged from 4.3 to 38.8 g CH4 m−2 yr−1 in 2002 and 3.2 to 28.8 g CH4 m−2 yr−1 in 2003. The scenario results suggest that lake creation is likely to result in the re-creation of a hot spot at the landscape level. However, the transition from cutaway to wetland ecosystem may lead to a reduction in the GWP of the peatland.

Peatland buffer zones with sedimentation ponds are established with the intention of capturing solids and nutrients liberated in drained forestry catchments. As noted in earlier fertilization experiments, added nitrogen (N) immediately increases nitrous oxide (N2O) emissions in such buffers, and we expected the same to happen after disturbances in the catchment caused by clear-cutting, soil preparation, and ditch cleaning. We measured N2O fluxes, water table dynamics, and vegetation cover from a wetland one year before and two years after the clear-cut and buffer establishment. The low pre-harvest emissions did not increase, but N2O emissions from the sedimentation pond exceeded those from humic lakes with a high N load. In the soil profile, N2O concentrations were high, indicating a potential to produce N2O in the buffer. In one sub-site the soil N2O concentration was below the atmospheric level, which was in accordance with the high concentrations of carbon dioxide (CO2) and methane (CH4). The change in vegetation along the overland flow paths could be explained by a shift in the species thriving in wet conditions but not in those requiring higher nutrient levels. In spite of the apparent potential of soil to produce N2O, the fluxes to the atmosphere remained low. Transformation of N2O to unobserved N2 may explain some of the low N emissions, together with the low concentrations entering the buffer.

A high-resolution plant macrofossil analysis was applied to investigate wetness dynamics in a southern Finnish boreal bog, Kontolanrahka, during the last 5000 years. The chronological control and the age—depth model were based on 40 AMS radiocarbon dates. Pollen analysis provided information on regional-scale vegetation changes. Macrofossil analysis revealed prominent changes in vegetation assemblages during the late Holocene, indicating fluctuations in water-table. The reconstruction suggests that at the coring point, which nowadays is a relatively wet lawn, habitat type has repeatedly varied between transient communities similar to those currently represented in dry hummocks, very wet lawns and even hollows. In order to quantify historical changes in water-table, Generalized Additive Models (GAM) were used to investigate the current relationships between surface plant species and water-table depth. Modern water-table measurements and a survey of associated plant communities along moisture gradients provided data for GAM-analyses. The plant species showed unimodal distributions with apparent optima and narrow tolerances along the water-table gradient. A transfer function for water-table reconstruction was created by calibrating plant macrofossil records against the modern vegetation/water-table relationship using the weighted averaging partial least squares (WA-PLS) regression method. The quantitative water-table reconstruction for the late Holocene showed that the water-table depth had varied between 38 and 2.5 cm, the root mean square error of prediction being 3 cm. The detected historical wet and dry shifts were compared with other similar data from Finland, Sweden and Estonia, and from Western Europe. Despite some similarities, especially during the last c. 1000 years, noticeable differences in timing and duration occur, suggesting they may not have been driven only by climate, but also by local factors, including surface fires.

Summary 1. We studied the patterns of litter decomposition in lake littoral habitats and investigated whether decay rates, as an integrating proxy for environmental conditions in the sediment, would co-vary with net carbon dioxide (CO2) exchange and methane (CH4) efflux. These gas fluxes are known to be sensitive to environmental conditions. Losses in the mass of cellulose, root, rhizome and moss litter were measured during 2 years in boreal littoral wetlands in Finland and compared with published data on concurrently measured gas fluxes. Four study sites covered a range of sediment types and hydrological conditions. 2. Decomposition was not linearly related to the duration of flooding but depended on sediment type. Readily decomposable litter fractions, such as cellulose and rhizome litter, lost mass at a faster rate in marshes with a longer period of flooding but wide water level fluctuations that hinder establishment of a Sphagnum cover, than in peat-forming fens. In marshes, the mean first-year mass losses were 83–99% and 19–62% for cellulose and rhizomes, respectively. In fens, the respective losses were 40–53% and 33%. In the first year, the loss in the mass of the more recalcitrant root litter did not differ between sites (mean 19–30%) and moss litter lost no mass. 3. The estimated first-year carbon loss from belowground litter was about 0.1–0.3 times ecosystem respiration and roughly similar to net carbon gas (CO2, CH4) efflux, suggesting that vascular plants and recent plant residues contribute substantially to ecosystem release of carbon gases. On the other hand, at least 40% of the mass of the belowground litter remained on a littoral site after the first 2 years of decomposition. Slow decomposition may indicate the accumulation of organic-rich sediments. The accumulated carbon could explain the excess CO2 release found in most littoral sites. In continuously inundated sites decomposition rates were similar to those in periodically flooded sites, but ecosystem-atmosphere CO2 exchange fell to close to zero. This discrepancy implies that the released CO2 is dissolved in water and may be exported into the pelagic zone of the lake.

We show that sediment respiration is one of the key factors contributing to the high CO2 supersaturation in and evasion from Finnish lakes, and evidently also over large areas in the boreal landscape, where the majority of the lakes are small and shallow. A subpopulation of 177 randomly selected lakes (o100km 2 ) and 32 lakes with the highest total phosphorus (Ptot) concentrations in the Nordic Lake Survey (NLS) data base were sampled during four seasons and at four depths. Patterns of CO2 concentrations plotted against depth and time demonstrate strong CO2 accumulation in hypolimnetic waters during the stratification periods. The relationship between O2 departure from the saturation and CO2 departure from the saturation was strong in the entire data set (r 2 50.79, n 52 740, Po0.0001). CO2 concentrations were positively associated with lake trophic state and the proportion of agricultural land in the catchment. In contrast, CO2 concentrations negatively correlated with the peatland percentage indicating that either input of easily degraded organic matter and/or nutrient load from agricultural land enhance degradation. The average lake-area-weighted annual CO2 evasion based on our 177 randomly selected lakes and all Finnish lakes 4100km 2 (Rantakari & Kortelainen, 2005) was 42gCm � 2 LA (lake area), approximately 20% of the average annual C accumulation in Finnish forest soils and tree biomass (covering 51% of the total area of Finland) in the 1990s. Extrapolating our estimate from Finland to all lakes of the boreal region suggests a total annual CO2 evasion of about 50TgC, a value upto 40% of current estimates for lakes of the entire globe, emphasizing the role of small boreal lakes as conduits for transferring terrestrially fixed C into the atmosphere.

Hydrology controls the physical, chemical and biological processes in peatlands and hence could be the most important process regulating their function, development and characteristic biogeochemistry. Models describing hydrological processes and soil heat exchange phenomena are an important tool in understanding the peatland biogeochemical cycles of C and N. Presented in this paper is a peat soil climate model that uses weather data, mire site and peat characteristics as input data. The model is related to the heterogeneous peat characteristics (e.g. bulk density, degree of humification, remains of Sphagnum , Carex , Eriophorum and woody tissues), which in turn vary with depth. Evapotranspiration is related to weather parameters and tree stand characteristics. The model simulations were validated against the observed data collected during 1993 and 1994 growing seasons at a fen and bog sites at the Lakkasuo mire complex located in central Finland. The observed and simulated day-to-day variations in water table levels and soil temperatures during both seasons appeared overall to be in phase. As the model lacks the data needed to characterize the peat matrix hysteresis, the model was found lacking in response to wetting and drying cycles. The mean differences between simulated and observed water table levels during 1994 were −0.3 ± 1.3 and −0.5 ± 2.3 cm for the fen and bog sites, respectively. The model was found to be sensitive to mire surface characteristics and evapotranspiration, particularly for the bog site with an unsaturated zone deeper than at the fen. The absolute mean differences between the simulated and measured peat temperatures from 5 to 150 cm were less than 1.0 °C with a maximal standard deviation of 1.6 °C. The model deviations for the upper layers showed larger variations compared to deeper layers, implying a greater accuracy in defining the lower boundary of the thermal regime within the peat column.

Summary 1 In order to study the dynamics of primary production and decomposition in the lake littoral, an interface zone between the pelagial, the catchment and the atmosphere, we measured ecosystem/atmosphere carbon dioxide (CO2) exchange in the littoral zone of an eutrophic boreal lake in Finland during two open water periods (1998–1999). We reconstructed the seasonal net CO2 exchange and identified the key factors controlling CO2 dynamics. The seasonal net ecosystem exchange (NEE) was related to the amount of carbon accumulated in plant biomass. 2 In the continuously inundated zones, spatial and temporal variation in the density of aerial shoots controlled CO2 fluxes, but seasonal net exchange was in most cases close to zero. The lower flooded zone had a net CO2 uptake of 1.8–6.2 mol m−2 per open water period, but the upper flooded zone with the highest photosynthetic capacity and above-ground plant biomass, had a net CO2 loss of 1.1–7.1 mol m−2 per open water period as a result of the high respiration rate. The excess of respiration can be explained by decomposition of organic matter produced on site in previous years or leached from the catchment. 3 Our results from the two study years suggest that changes in phenology and water level were the prime cause of the large interannual difference in NEE in the littoral zone. Thus, the littoral is a dynamic buffer and source for the load of allochthonous and autochthonous carbon to small lakes.

Lake littoral zones have a transitional nature and dynamic conditions, which are reflected in their CH4 emissions. Thus, detailed studies are needed to assess the littoral CH4 emissions in a regional scale. In this study, CH4 fluxes were followed during the ice-free seasons in 1998 and 1999 by using the static chamber method in the littoral zone of two lakes in Finland. An exceptionally high water level in 1998 caused an unusually long inundation in otherwise ephemerally flooded zone. The flooding was normal in year 1999. The factors controlling CH4 emissions were examined and statistical response functions were constructed. Further, the effect of extended flooding on the littoral CH4 budged was estimated. The methane flux was primarily regulated by the water level in grass and sedge dominated eulittoral zone, but not in infralittoral reed and water lily stands. Methane emissions in the sedge dominated zone decreased significantly, when the flood was high enough to submerge the venting structures of the plants. Besides water level, sediment temperature determined CH4 emission. The cumulative CH4 emissions from the whole littoral wetlands in wet year were 1.1 times (L. Kevaton), or 0.61 and 0.79 times (L. Mekrijarvi) those in dry year. The crucial factor was the discrepancy between the exceptional and the average water level. The extension of inundated area does not necessarily increase CH4 emissions if the flood reaches infrequently inundated areas, which apparently have low CH4 production potential. This is the case especially, if the emissions in lower zones simultaneously decrease due to high water level. Our study analyses these complex responses between CH4 emissions and water level.

The springtime methane (CH4) emission from a small, eutrophied boreal lake was assessed during the winter ice-cover by measurement of gas ebullition and CH4 accumulation in the water column in association with the development of oxygen depletion after ice formation. The winter CH4 production was estimated to result in a loss of 3.6-7.9 g CH4 m(-2) from the lake to the atmosphere during the short period of ice melt. This could account for 22-48% of the annual CH4 emission from the pelagic zone of the lake. The contribution of winter to the annual CH4 release can be similar or even higher in seasonally ice-covered northern aquatic ecosystems than in northern terrestrial wetlands, thus winter must be considered in any studies into the aquatic CH4 emissions. The trophic state and wintertime oxygen conditions, linked to the changes in land-use in the catchments and climate, are important factors controlling the springtime lake CH4 emissions.

CO2 and CH4 fluxes during the winter were measured at natural and drained bog and fen sites in eastern Finland using both the closed chamber method and calculations of gas diffusion along a concentration gradient through the snowpack. The snow diffusion results were compared with those obtained by chamber, but the winter flux estimates were derived from chamber data only. CH4 emissions from a poor bog were lower than those from an oligotrophic fen, while both CO2 and CH4 fluxes were higher in the Carex rostrata-occupied marginal (lagg) area of the fen than in the slightly less fertile centre. Average estimated winter CO2-C losses from virgin and drained forested peatlands were 41 and 68 g CO2-C m- 2, respectively, accounting for 23 and 21% of the annual total CO2 release from the peat. The mean release of CH4-C was 1.0 g in natural bogs and 3.4 g m-2 in fens, giving rise to winter emissions averaging to 22% of the annual emission from the bogs and 10% of that from the fens. These wintertime carbon gas losses in Finnish natural peatlands were even greater than reported average long-term annual C accumulation values (less than 25 g C m-2). The narrow range of 10–30% of the proportion of winter CO2 and CH4 emissions from annual emissions found in Finnish peatlands suggest that a wider generalization in the boreal zone is possible. Drained forested bogs emitted 0.3 g CH4-C m-2 on the average, while the effectively drained fens consumed an average of 0.01 g CH4-C m- 2. Reason for the low CH4 efflux or net oxidation in drained peatlands probably lies in low substrate supply and thus low CH4 production in the anoxic deep peat layers. N2O release from a fertilized grassland site in November–May was 0.7 g N2O m-2, accounting for 38% of the total annual emission, while a forested bog released none and two efficiently drained forested fens 0.09 (28% of annual release) and 0.04 g N2O m- 2 (27%) during the winter, respectively.

Northern peatlands are important terrestrial carbon stores, and they show large spatial and temporal variation in the atmospheric exchange of CO2 and CH4. Thus, annual carbon balance must be studied in detail in order to predict the climatic responses of these ecosystems. Closed-chamber methods were used to study CO2 and CH4 in hollow, Sphagnum angustifolium lawn, S. fuscum lawn, and hummock microsites within an om- brotrophic S. fuscum bog. Micrometeorological tower measurements were used as a ref- erence for the CH4 efflux from the bog. Low precipitation during May-August in 1994 (84 mm below the long-term average for the same period) and a warm July-August period caused the water table to drop by more than 15 cm below the peat surface in the hollows and to 48 cm below the surface in high hummocks. Increased annual total respiration exceeded gross production and resulted in a net C loss of 4-157 g/m2 in the different microsites. Drought probably caused irreversible desiccation in some lawns of S. angus- tifolium and S. balticum and in S. fuscum in the hummocks, while S. balticum growing in hollows retained its moisture and even increased its photosynthetic capacity during the July-August period. Seasonal (12 May-4 October) CH4 emissions ranged from 2 g CH4- C/M2 in drier S. fuscum hummocks and lawns to 7 and 14 g/m2 in wetter S. angustifolium- S. balticum lawns and hollows, respectively. Aerodynamic gradient measurements at the tower showed slightly higher CH4 flux rates than the average estimates for the whole bog obtained by closed-chamber methods. Winter C efflux comprised 30 g C02-C/m2 and 1 g CH4-C/m2 out of a total loss of 90 g C/M2 on average in the bog, and there was an estimated annual loss of 7 g C/M2 by leaching. This study shows how delicately the boreal bog's C balance in different microsites depends on climatic variations, especially the distribution of precipitation. It also confirms that severe C losses can occur in boreal bogs during extended summer droughts, even in years with annual temperatures close to the long-term average and with precipitation clearly greater than the long-term average.

Methane (CH4) fluxes were measured at 17 peatland sites with different nutritional and hydrological characteristics in the southern and middle boreal zones in Finland by a static chamber technique. Many of the natural peatlands also had counterparts drained for forestry 30–50 years ago. The mean emissions from May to September were 8.0 g CH4 m−2 for natural ombrogenous bogs and 19.0 g CH4 m−2 for natural minerogenous fens thus being higher than the 2 g CH4 m−2 yr−1 estimated for the Canadian peatlands. Change in the mean water table level had greater effect on CH4 fluxes on natural fens than on natural bogs. The mean CH4 emissions on drained bogs and fens were 3.9 g CH4 m−2 and 0.3 g CH4m−2, respectively. Some drained fens even had CH4 uptake from the atmosphere. The change in the mean water table had the lowest effect on CH4 fluxes on drained peatlands. The CH4 fluxes on peatlands (natural fens and bogs and drained peatlands) are associated with peat aeration, nutrient level, vegetation cover, peat compaction, peat temperature, and finally with microbial processes responsible for the net release of CH4. We could explain 67% of the variation in mean CH4 fluxes on Finnish peatlands by measuring the mean water table, peat bulk density, and peat pH. The present results can be used to predict the possible changes in CH4 emissions on peatlands if the climate is drying in north. For example, lowering of the present water table by 10 cm would induce a 70% reduction in the CH4 emissions from fens and a 45% reduction from bogs.

The effects of temperature, water table, and precipitation on the methane fluxes from a boreal low-sedge Sphagnum papillosum pine fen were analyzed with statistical cross correlations of daily data. The six measurement sites represented different vegetation surfaces of the mire (hummocks, lawns, and flarks) with increasing moisture. The dynamics were analyzed separately for the early summer (May-July) and the late summer (August-October) periods in addition to the whole summer (May-October) period. Methane emissions increased with increasing peat temperature. During the late summer period, changes in peat temperatures at depths of 20 and 50 cm were reflected in methane emissions within 2 days. The persistently high water tables during the measurement period probably did not reveal the dynamics between water table fluctuations and methane emissions very clearly. Methane emission levels correlated negatively with depths of the water tables, that is, high methane emissions were associated with low water tables and vice versa. The suppression of methane emissions by filling the unsaturated gas space during precipitation and the increased release rate caused by a declining water table could explain the result. Methane emissions correlated positively with changes in water tables, that is, a rise in water table increased methane emissions during the early and whole summer periods. Precipitation increased emissions with a lag from zero to several days throughout the summer. Generally, the estimated responses of methane fluxes to precipitation and changes in water table indicated similar time lags. Methane flux from the flark surfaces seemed to respond rapidly to rainfall and changes in water table with a lag of zero or 1 day. In the lawn-low hummock, the lawn and one hummock site, methane flux showed a slow response with several days lag. This study strongly indicates that temperature, water table, and precipitation affect methane emissions with complex interactions.

The CO2 released in soil respiration is formed from organic matter which differs in age and stability, ranging from soluble root exudates to more persistent plant remains. The contribution of roots, a relatively fast component of soil cycling, was studied in three experiments. (1) Willows were grown in a greenhouse and CO2 fluxes from the substrate soil (milled peat) and from control peat were measured. (2) CO2 fluxes from various peatland sites were measured at control points and points where the roots were severed from the plants. (3) CO2 fluxes in cultivated grassland established on peatland were measured in grassy subsites and in subsites where the growth of grass was prevented by regular tilling. The root-derived respiration followed the typical annual phenology of the vegetation, being at its maximum in the middle and late summer. All the experiments gave similar results, root-derived respiration accounting for 35–45% of total soil respiration in the middle and late summer at sites with an abundant vegetation. The root-derived respiration from the virgin peatland sites correlated well with the tree biomass, and also partly with the understorey vegetation, but in the drained sites the root effect was greater, even in the presence of less understorey vegetation than at virgin subsites.

[1] Diurnal variation in methane fluxes can cause systematic errors in flux estimates, particularly at places with aquatic vegetation. Closed chamber measurements were made during the ice-free period in central Finland. Methane fluxes were quantified at an interval of a few hours for several boreal lakeshore communities consisting of Phragmites australis (Cav.) Trin. ex Steud. and of wetland sedges, grasses, and herbs. The automated measurement system was operated for 4 months in a P. australis stand and for 2 months in a site with mixed wetland vegetation. Additional manual measurements lasted for a few days. In P. australis stands the CH4 fluxes were typically highest around noon. On the average, efflux of CH4 measured during office hours (0800–1600 local time (LT)) should be corrected using a factor 0.68 to obtain a more reliable estimate of daily CH4 release. The fluctuation magnitude varied between different locations and over the study period, being the greatest in the most productive parts of the stands in August. A harmonic regression model, scaled with temperature and stand biomass, was successfully applied to reconstruct CH4 flux in P. australis stand. In the sedge, grass and herb stands the diurnal fluctuation in CH4 flux was generally small. Commonly, daily maximum in efflux occurred during evening or night and large (threefold) diurnal differences were occasionally found. Daytime measurements alone can result in a slight or moderate underestimate of the total flux for the wetland communities. These results recommend that spatial and seasonal differences in diurnal CH4 flux dynamics should be considered when planning measuring or modeling CH4 fluxes.

We have examined how some major catchment disturbances may affect the aquatic greenhouse gas fluxes in the boreal zone, using gas flux data from studies made in 1994-1999 in the pelagic regions of seven lakes and two reservoirs in Finland. The highest pelagic seasonal average methane (CH(4)) emissions were up to 12 mmol x m(-2) x d(-1) from eutrophied lakes with agricultural catchments. Nutrient loading increases autochthonous primary production in lakes, promoting oxygen consumption and anaerobic decomposition in the sediments and this can lead to increased CH(4) release from lakes to the atmosphere. The carbon dioxide (CO(2)) fluxes were higher from reservoirs and lakes whose catchment areas were rich in peatlands or managed forests, and from eutrophied lakes in comparison to oligotrophic and mesotrophic sites. However, all these sites were net sources of CO(2) to the atmosphere. The pelagic CH(4) emissions were generally lower than those from the littoral zone. The fluxes of nitrous oxide (N(2)O) were negligible in the pelagic regions, apparently due to low nitrate inputs and/or low nitrification activity. However, the littoral zone, acting as a buffer for leached nitrogen, did release N(2)O. Anthropogenic disturbances of boreal lakes, such as increasing eutrophication, can change the aquatic greenhouse gas balance, but also the gas exchange in the littoral zone should be included in any assessment of the overall effect. It seems that autochthonous and allochthonous carbon sources, which contribute to the CH(4) and CO(2) production in lakes, also have importance in the greenhouse gas emissions from reservoirs.

Northern peatlands accumulate atmospheric CO2 thus counteracting climate warming. However, CH4 which is more efficient as a greenhouse gas than CO2, is produced in the anaerobic decomposition processes in peat. When peatlands are taken for forestry their water table is lowered by ditching. We studied long-term effects of lowered water table on the development of vegetation and the annual emissions of CO2, CH4 and N2O in an ombrotrophic bog and in a minerotrophic fen in Finland. Reclamation of the peat sites for forestry had changed the composition and coverage of the field and ground layer species, and increased highly the growth of tree stand at the drained fen. In general, drainage increased the annual CO2 emissions but the emissions were also affected by the natural fluctuations of water table. In contrast to CO2, drainage had decreased the emissions of CH4, the drained fen even consumed atmospheric CH4. CO2 and CH4 emissions were higher in the virgin fen than in the virgin bog. There were no N2O emissions from neither type of virgin sites. Drainage had, however, highly increased the N2O emissions from the fen. The results suggest that post-drainage changes in gas fluxes depend on the trophy of the original mires.

Greenhouse gas fluxes (CH4, N20 and C02) at a managed Finnish peat soil site were compared with those in a virgin fen in 1991 and 1992. The field site was drained about 60 years ago, and as it was used for pasturing and fodder production, the CH4 released by the cattle was estimated on the basis of data in the literature. The annual budgets of the gases were estimated and their GWP values calculated to assess the climatic impact of the current agricultural activity. Methane emission from the field site, 1 kg CH4-C ha- 1 yr- 1, was low, compared with that from the natural fen site, 260 kg CH4-C ha- 1 yr- 1. Annual N20 emission at the field site was 8 and 9 kg N20-N ha - 1 yr - 1 in 1991 and 1992, respectively. Summer N20 emission in 1992 from vegetationless soil was five times higher than that from actual grassland. The N20

[1] The surface-atmospheric exchange of nitrous oxide (N2O) was investigated in the vegetated littoral zone of a eutrophied midboreal lake (Lake Kevaton, Finland) with a static chamber technique. During a dry summer (three to six samplings per site), the meadow site and two marsh sites in the temporarily flooded eulittoral zone and the Phragmites australis-dominated site in the continuously flooded infralittoral zone had mean daytime N2O-N emissions from 11 ± 7 to 22 ± 7 μg m−2 h−1, whereas the Nuphar lutea-dominated site in the infralittoral zone had a mean N2O flux close to zero. During a wet summer (13–14 samplings per site), the mean daytime N2O-N fluxes ranged from 4 ± 1 to 15 ± 5 μg m−2 h−1 at the three eulittoral sites and were negligible at the two infralittoral sites. The littoral zone occupied 26% of the lake area but was estimated to account for most of the N2O emissions from the lake. The studied eulittoral zone, which did not have adjacent nitrogen fertilization, exhibited higher N2O emissions during the summer than seen in northern natural ecosystems in general, including peatlands, forests, and the pelagic regions of lakes. Thus in lake-rich landscapes the littoral zone and other lake-associated wetlands must be considered as potential sources of atmospheric N2O. An assessment of their atmospheric importance requires further data on the N2O fluxes and their regulation in different littoral areas and on the total littoral coverage, neither of which is yet available.

A three-dimensional model, based on numerous vertical and horizontal 14C datings and GIS simulation from a concentrically domed mire in southern Finland, demonstrates considerable variation in actual (net) rate of carbon accumulation (ARCA) through time. Lateral growth between 9000 and 3000 yr ago accounts for only 25% of the total carbon sequestration, whereas bog formation after ombrotrophication is responsible for the remaining 75%. On the other hand, the most rapid increase in landscape CH4 flux occurred between 4500 and 3000 yr ago, the period of the fastest horizontal growth. In addition to autecological factors, a climatic shift towards cooler and wetter conditions is a plausible mechanism for maintaining accelerated carbon accumulation.