UC Irvine

Lakes are widespread in permafrost regions. Across these landscapes, lakes have the unique ability to thaw permafrost, and large quantities of ancient organic matter contained within it, tens-to-hundreds of meters below the surface. Once thawed, the organic material is vulnerable to decomposition by microorganisms and emission to the atmosphere as the greenhouse gases methane (CH4) and carbon dioxide (CO2). Because northern regions currently store more carbon (C) in permafrost than what is found in the atmosphere or all living biomass on Earth, there is a great need to understand how lakes may mobilize this vast C reservoir. Despite this need, our current knowledge of this phenomenon is insufficient to fully characterize the role of northern lakes in the global CH4 budget, C cycle, or earth system models. Thus, our predictability of future permafrost-C emissions remains highly uncertain. To address this uncertainty, this dissertation aims to understand CH4 and CO2 emission sources on the whole-lake level, constrain regional estimates of ancient permafrost-C emissions from northern lakes, and determine the regional-scale environmental controls that regulate these processes.

A major challenge of this dissertation was to understand CH4 and CO2 emission sources on the whole-lake level, a spatial scale relevant for lake-to-lake comparisons and regional estimates. To do this, I approached the dissolved CH4 and CO2 pools as integrators of whole-lake emission pathways and sources. This required novel methods to sample and analyze relatively low-concentration, dissolved CH4 and CO2, especially in open waters during summer. In Chapter 2, I demonstrated the accuracy and precision of the sampling methods that were developed to collect low-concentration dissolved CH4 and CO2 for stable carbon isotope (13C) and radiocarbon (14C) analysis in Arctic and Boreal lakes in Alaska. Relative to common sample yields for CH4 and CO2 (> 0.3 mg C), the consistent isotopic contaminations of ≤ 2.0 μg C-CH4 and ≤ 25.0 μg C-CO2 were determined to be negligible. In conclusion, this study verified and enabled, for the first time, the simultaneous collection and analysis of dissolved 14CH4 and 14CO2 in Arctic lakes, which improved our ability to characterize whole-lake emission sources and pathways on landscape scales.

Our current understanding of CH4 and CO2 emissions from northern lakes is based on limited measurements, mostly of CH4 ebullition (bubbling), in lakes residing in the C- and-ice rich eolian sediment, yedoma. While these observations are insightful for determining the global impact of lake-CH4 emissions, they only provide a partial understanding of intra-lake processes, and on a broader scale, the vast environmental diversity across the circumpolar region. Chapter 3 aims to scale and integrate heterogeneous emission sources and pathways of CH4 and CO2 to the whole-lake level, facilitating lake-to-lake comparisons, and thus elucidating the broad scale environmental controls permafrost-C emissions. Here I showed, through the use of 14C and 13C mass balance approaches, that winter lake-ice mixes heterogeneous ebullition-CH4 into the dissolved CH4 pool, thereby integrating whole-lake CH4 emission sources below ice. Furthermore, I quantified that of the CH4 accumulated below ice, approximately 50% is oxidized to CO2, reducing the global warming impact of lake-C emissions. Conclusively, I demonstrated that a relatively simplistic Keeling plot approach can estimate the integrated whole-lake mean 14CH4-ebullition within 10% of a more comprehensive, but time-intensive bubble survey approach, improving our ability to understand lakes on larger scales. These results verify the below-ice dissolved CH4 and CO2 pools as spatial and temporal integrators of whole-lake emission sources and pathways, enabling whole- lake scaling.

To forecast, and potentially mitigate permafrost-C emissions from climate-sensitive northern lakes, we must first understand their variability on annual and regional scales. Once whole-lake emission sources are known, many lakes can be inter-compared to elucidate the key environmental divers regulating permafrost-C emission sources. Chapter 4 reveals the predominant influence of general surficial geology on whole-lake CH4 and CO2 emission sources in a 40-lake regional survey on the North Slope of Alaska. This study showed that lakes residing in sandy surficial geology categories only emit CH4 and CO2 from young C sources (≤ 300 YBP), whereas lakes residing in finer-grained geology categories can access and emit ancient C (≥ 11,500 YBP), albeit in small proportions (< 20%) relative to total lake-C emissions. Additionally, this study demonstrated that almost 100% of lake-C emissions to the atmosphere are in the form of CO2, implying a lesser role of CH4 in the global warming potential of Arctic lake emissions. Furthermore, 14CH4 and 14CO2 were highly correlated across the North Slope, implying that large quantities of CH4 were oxidized to CO2 prior atmospheric emission. In conclusion, this 40-lake survey revealed the regional-scale control of general surficial geology on whole-lake CH4 and CO2 emission sources, and that these emissions are predominantly sourced from relatively young, late-Holocene C, rather than ancient permafrost-C.

My PhD research developed the methodology to monitor C emissions from climate- sensitive lakes in the northern circumpolar permafrost region and provides critical baseline data for future monitoring efforts. My work also demonstrated that, in addition to climate, surficial geological substrate should be considered in forecasting the role of thaw lakes in the permafrost-C climate feedback, since it effectively integrates thermokarst behavior and the C retention properties of the landscape.