We studied CO 2 and CH 4 fluxes from two boreal lakes with differing trophic status (chlorophyll a 17.8 vs. 48.7 mg m 22 ) and water color (100 vs. 20 mg Pt L 21 ) throughout an open-water period when summer precipitation doubled, using both floating chambers and concentration gradients. Fluxes measured in chambers were higher, but irrespective of the method, both lakes were heterotrophic and were annual sources of carbon gases to the atmosphere. However, with the annual CO 2 flux of 6.85 (chambers) or 5.43 mol m 22 (gradients), the humic lake had notably higher emissions than the clear-water lake, where the fluxes were 3.97 and 3.38 mol m 22 , respectively. The annual CH 4 flux from the clear-water lake was 28.5 (chambers) or 20.5 mmol m 22 (gradients) and from the humic lake 20.7 or 16.2 mmol m 22 , respectively. There were interlake differences in seasonal patterns, but the most obvious changes in fluxes occurred during or just after the rains. In the humic lake, the resulting peak in CO 2 and CH 4 flux was responsible for 46% and 48% of the annual flux, respectively. Before the rains, the clear-water lake was a small sink of CO 2 or had near-zero efflux but afterwards became a source of CO 2 . In the humic lake, biological mineralization explained the majority of the fluxes, whereas in the clear-water lake the association between the biological processes and fluxes was less pronounced.

We studied the concentrations and diffusive fluxes of carbon dioxide (CO 2 ) and methane (CH 4 ) in 12 lakes in a size range of 0.004-35 km 2 and mean total organic carbon (TOC) range of 6-25 mg L 21 , located in the southern boreal area of Finland. TOC, mainly originating from the catchment, was the best predictor of concentrations and areal CO 2 effluxes in the whole lake-area range, but among the lakes with an area , 1 km 2 having an anoxic hypolimnion during summer, the areal effluxes were also related to lake size and an index of turbulence. The concentration and areal effluxes of CH 4 from the lakes' pelagic zones were related more closely to lake-sizerelated water column stability and turbulent mixing than directly to TOC. The total regional flux estimate from 619 lakes in the region of 1600 km 2 showed that seven lakes with an area of 10-50 km 2 were responsible for 57-69% of the pelagic CO 2 effluxes and about half of the sum of pelagic and littoral CH 4 effluxes from the lakes. The proportion accounted for by the small lakes (area , 1 km 2 ) in the gas effluxes was 1.4-2.1 times higher, 18-26% for CO 2 and 21-26% for CH 4 , than their areal proportion (12.5%) in the landscape, although numerically these small lakes predominate in the whole lake population (96%).Most inland waters worldwide have been shown to be supersaturated with the greenhouse gases (GHG) carbon dioxide (CO 2 ) and methane (CH 4 ), and thus release these gases to the atmosphere (Cole et al. 1994;Bastviken et al. 2004). The GHG supersaturation indicates that freshwater ecosystems play an important role in the global carbon cycle by mineralizing and/or accumulating terrestrial organic carbon (Tranvik et al. 2009). By applying Pareto distribution analysis, Downing et al. (2006) estimated that . 99% of the world's lakes have an area , 1 km 2 and these small lakes form , 43% of the global lake area. Thus, in many recent analyses of lake-rich landscapes, the role of small lakes has received much attention, because the small headwater lakes are more typically characterized by higher concentrations of allochthonous organic carbon than the larger lowland lakes (Xenopoulos et al. 2003;Hanson et al. 2007;Einola et al. 2011). In the boreal region, high fluxes of total organic carbon (TOC), originating mainly from peatland and forested catchments (Kortelainen 1993;Humborg et al. 2010), are among the most important factors contributing to CO 2 supersaturation and oxygen consumption in the lake water columns and sediments (Salonen et al. 1983;Larsen et al. 2011). As a result of anaerobic decomposition of organic matter, high concentrations of CH 4 are typical in eutrophic lakes with an anoxic hypolimnion but also in sheltered forest lakes, which likewise develop marked and prolonged hypolimnetic anoxia. It has been suggested that spring and fall mixing periods are generally the periods of the greatest effluxes of GHG into the atmosphere. However, rainy periods may also be followed by efflux peaks , and thus precipitation influences the inte...

[1] The greatest gas loss from dimictic lakes occur during spring and autumn mixing periods. Thus, we measured daily concentration gradients of carbon gases (CO 2 and CH 4 ) in mesohumic Lake Pääjärvi during the mixing periods in autumn 2004 and spring 2005 and calculated and compared the fluxes using three different methods: the boundary layer diffusion model (DCO 2 and DCH 4 ), floating static chambers (FC), and changes in gas storage. CO 2 fluxes were higher in autumn than in spring, whereas CH 4 fluxes were lower in autumn than in spring. The method based on changes in storage underestimated the fluxes whereas the floating chambers and the boundary layer diffusion models resulted in similar estimates. However, the chambers always yielded somewhat higher fluxes. Total DCO 2 flux in autumn was 883 mmol m À2 and in spring, 666 mmol m À2, whereas total DCH 4 fluxes were 0.60 mmol m À2 and 0.80 mmol m À2 in autumn and spring, respectively. We calculated gas transfer velocities (k 600 ) to explain the near surface exchange mechanism and the difference between the results based on diffusion models and chambers. Wind speed and k 600 showed significant correlation. In spring the transfer velocity at similar wind speed was higher compared to the autumn. Weekly measurements of algal primary production and community respiration revealed that the lake was net heterotrophic in autumn as well as in spring. Our study showed that the excess CO 2 from the lake metabolism contributed significantly to the CO 2 fluxes during the mixing periods, violating the primary assumption used in the storage method.Citation: López Bellido, J., T. Tulonen, P. Kankaala, and A. Ojala (2009), CO 2 and CH 4 fluxes during spring and autumn mixing periods in a boreal lake (Pääjärvi, southern Finland),

Changes in water pH and colour since the late 1980s were studied in 35 small boreal lakes of varying hydrological and landscape settings but similar climate and acid deposition. The data was collected during the autumnal overturn on the annual basis except in lake with weekly sampling during the ice-free period. In addition to the deposition data information about catchment soil types as well as local meteorological and hydrological conditions were used for the long-term data interpretation. The lakes are situated in a small area in southern Finland, 130 km north from Helsinki, where sulphate deposition declined by [60% in one decade since the mid1980s. The results showed that water colour increased in most lakes while pH did not. In lakes dominated by surface runoff there was a distinct upward shift in colour, with an initial increase after the mid-1990s and a second increase in 2004. The first shift appeared when the sulphate deposition reached a level ca. 25% of that in 1988. However, the upward shift in colour also coincided with a change in hydrological conditions after several dry summers. In contrast, the second shift in colour clearly coincided with a switch in hydrology due to the abnormally wet summer of 2004 after severe drought in 2002-2003. Although the hydrological conditions indisputably had a key role in determining the annual variability in colour, a distinct negative relationship between acid deposition and water colour in 90% of the lakes strongly suggested that reduction in sulphate deposition fostered the leaching of coloured organic substances from the catchment soils. Increase in colour, in turn, strongly influenced lake water pH, and the present day higher organic matter concentrations seemingly depress pH values more than in the 1980s, before the reduction in acid deposition.

Effects of different molecular size fractions (< 1000 MW, < 10 000 MW, < 100 000 MW and < 0.1 m) of dissolved organic matter (DOM) on the growth of bacteria, algae and protozoa from a highly humic lake were investigated. DOM from catchment drainage water as well as from the lake consisted mostly (59-63 %) of high molecular weight (HMW) compounds (> 10 000 MW). With excess inorganic nutrients, the growth rate and yield of bacteria were almost identical in all size fractions. However, in < 1000 MW fractions and with glucose added, a longer lag phase occurred. Without added nutrients both the growth rates and biomasses of bacteria decreased towards the smaller size fractions and the percentage of dissolved organic carbon (DOC) used during the experiment and the growth efficiency of bacteria were lower than with excess nutrients. The growth efficiency of bacteria was estimated to vary between 3-66% in different MW fractions, largely depending on the nutrient concentrations, but the highest growth efficiencies were observed in HMW fractions and with glucose. The growth of algae was clearly lowest in the < 1000 MW fraction. In dim light no net growth of algae could be found. In contrast, added nutrients substantially enhanced algal growth and in deionized water with glucose, algae achieved almost the same growth rate and biomass as in higher MW fractions of DOM. The results suggested that bacteria and some algae were favoured by DOM, but protozoans seemed to benefit only indirectly, through bacterial grazing. The utilization of DOM by bacteria and algae was strongly affected by the availability of phosphorus and nitrogen.