The CO( 2)flux (FCO2) from lakes to the atmosphere is a large component of the global carbon cycle anddepends on the air-water CO2concentration gradient (Delta CO2) and the gas transfer velocity (k). Both Delta CO2 and k can vary on multiple timescales and understanding their contributions toFCO(2)is important for explaining var-iability influxes and developing optimal sampling designs. We measuredFCO2 and Delta CO(2 )and derivedkforone full ice-free period in 18 lakes usingfloating chambers and estimated the contributions of Delta CO2 and k to FCO2 variability. Generally, kcontributed more than Delta CO2to short-term (1-9d) FCO2 variability. With in creased temporal period, the contribution of k to FCO2 variability decreased, and in some lakes resulted in Delta CO2 contrib-uting more thank to FCO2 variability over the full ice-free period. Increased contribution of Delta CO2 to FCO2 vari-ability over time occurred across all lakes but was most apparent in large-volume southern-boreal lakes and indeeper (>2m) parts of lakes, whereaskwas linked to FCO(2 )variability in shallow waters. Accordingly, knowing the variability of bothk and Delta CO(2 )over time and space is needed for accurate modeling of F CO2 from these vari-ables. We conclude that priority in FCO(2 )assessments should be given to direct measurements of FCO2 at multiplesites when possible, or otherwise from spatially distributed measurements of Delta CO(2 )combined with k- models that incorporate spatial variability of lake thermal structure and meteorology.

Methane (CH₄) is a potent greenhouse gas which is emitted to the atmosphere from both natural and anthropogenic sources. Current evidence indicates that lakes account for a large part of the global emissions of CH₄, but their contribution is difficult to quantify because of large temporal and spatial variability in processes leading to CH₄ fluxes from lakes to the atmosphere. Making sense of the complexity and variability of CH₄ emissions from lakes requires observations covering the range of temporal and spatial scales at which these processes occur, both within and between lakes. Northern regions are of particular interest for such studies because they contain a larger number of lakes than any other region in the world and they are disproportionately affected by climate change, with possible consequences for future CH₄ emissions.

The aim of this thesis was to investigate patterns of CH₄ dynamics and emissions in several lakes distributed in different climatic regions of Sweden, paying particular attention to spatial and temporal variability of CH₄ fluxes and concentrations. Fluxes, concentrations, carbon stable isotope signature of CH₄, and a range of commonly monitored lake characteristics were measured several times during one year at multiple locations in each lake. The measurements provided an extensive set of observations of CH₄ concentrations and fluxes in lakes, together with possible environmental drivers. These observations were then used to investigate patterns of CH₄ dynamics in northern lakes and to assess the ability of empirical and process-based models to predict CH₄ concentrations and fluxes in lakes.

The results indicate that simple empirical models, consisting of linear regressions between explanatory variables and CH₄ fluxes and concentrations averaged over the lake surface and ice-free period of the year, can be useful in some specific cases (for example describing ebullitive fluxes from total phosphorus or chlorophyll a concentrations). However, it was also noted that using such models for extrapolation can lead to large errors, especially if the observations do not account for temporal and spatial variability of CH₄ fluxes and concentrations. An example of high variability was seen in day-night measurements of CH₄ fluxes in four lakes over several months. To try to compensate for some of the shortcomings of empirical models, an established process-based and one-dimensional lake model was used to simulate CH₄ concentration in the water column of the studied lakes. Predictions were in good agreement with observations in several of the investigated lakes, considering that the model was not pre-calibrated for any of the lake specifically. However, it was also clear that there can be key processes that require specific consideration in process-based models, and some degree of simplification is needed, especially when detailed information on the modelled systems is not available. The simplifications and assumptions that need to be made can be informed by the study and observation of relevant processes in situ. For example, groundwater was found to potentially contribute a major part of CH₄ stored in one small boreal lake using measurements of stable isotope signature of CH₄ in littoral sediment and deep water of that lake, as well as in the groundwater in the mire next to it. Stable isotope measurements in five other lakes also revealed consistent differences in CH₄ sources to the surface and deep zones of lakes when they are separated by thermal stratification of the water column. Such knowledge could be used in the design of numerical models of lakes with the objective to improve predictions of current and future emissions of CH₄ from these environments.

Overall, this thesis contributes to the current knowledge on assessment of CH₄ emissions from lakes at several temporal and spatial scales. It also emphasizes critical aspects which must be considered to reduce bias in future empirical and process-based models of CH₄ in lakes.

Lakes are generally supersaturated in carbon dioxide (CO2) and emitters of CO2 to the atmosphere. However, estimates of CO2 flux (FCO2) from lakes are seldom based on direct flux measurements and usually do not account for nighttime emissions, yielding risk of biased assessments. Here, we present direct FCO2 measurements from automated floating chambers collected every 2-3 hr and spanning 115 24 hr periods in three boreal lakes during summer stratification and before and after autumn mixing in the most eutrophic lake of these. We observed 40%-67% higher mean FCO2 in daytime during periods of surface water CO2 supersaturation in all lakes. Day-night differences in wind speed were correlated with the day-night FCO2 differences in the two larger lakes, but in the smallest and most wind-sheltered lake peaks of FCO2 coincided with low-winds at night. During stratification in the eutrophic lake, CO2 was near equilibrium and diel variability of FCO2 insignificant, but after autumn mixing FCO2 was high with distinct diel variability making this lake a net CO2 source on an annual basis. We found that extrapolating daytime measurements to 24 hr periods overestimated FCO2 by up to 30%, whereas extrapolating measurements from the stratified period to annual rates in the eutrophic lake underestimated FCO2 by 86%. This shows the importance of accounting for diel and seasonal variability in lake CO2 emission estimates.

Methane (CH₄) is an important component of the carbon (C) cycling in lakes. CH₄ production enables carbon in sediments to be either reintroduced to the food web via CH₄ oxidation or emitted as a greenhouse gas making lakes one of the largest natural sources of atmospheric CH₄. Large stable carbon isotopic fractionation during CH₄ oxidation makes changes in ¹³C:¹²C ratio (δ¹³C) a powerful and widely used tool to determine the extent to which lake CH₄ is oxidized, rather than emitted. This relies on correct δ¹³C values of original CH₄ sources, the variability of which has rarely been investigated systematically in lakes. In this study, we measured δ¹³C in CH₄ bubbles in littoral sediments and in CH₄ dissolved in the anoxic hypolimnion of six boreal lakes with different characteristics. The results indicate that δ¹³C of CH₄ sources is consistently higher (less ¹³C depletion) in littoral sediments than in deep waters across boreal and subarctic lakes. Variability in organic matter substrates across depths is a potential explanation. In one of the studied lakes available data from nearby soils showed correspondence between δ¹³C-CH₄ in groundwater and deep lake water, and input from the catchment of CH₄ via groundwater exceeded atmospheric CH₄ emissions tenfold over a period of 1 month. It indicates that lateral hydrological transport of CH₄ can explain the observed δ¹³C-CH₄ patterns and be important for lake CH₄ cycling. Our results have important consequences for modelling and process assessments relative to lake CH₄ using δ¹³C, including for CH₄ oxidation, which is a key regulator of lake CH₄ emissions.