Permafrost, defined as ground that has been frozen for more than two consecutive years, underlies 15% of the land area of the Northern Hemisphere, storing vast amounts of organic carbon accumulated over millennia. Over 30% of global coastlines are shaped by permafrost, which is particularly sensitive to climate change. The rapid reduction in sea ice, rising air and sea temperatures, and more intense storm activity amplify erosion along these ice-rich coasts, leading to land loss of up to 25 meters per year at the Beaufort Sea. Coastal erosion taps lakes and drained lake basins formed during permafrost thaw, transforming them into so-called thermokarst lagoons. These lagoons serve as transition zones between terrestrial and subsea permafrost. Entering seawater accelerates thawing, making previously frozen or lacustrine organic carbon available for microorganisms to decompose into greenhouse gases (GHGs) such as carbon dioxide and methane, which contributes to further climate warming. So far, only a few studies have focused on the geology and evolution of these Arctic lagoons, and little is known about the role of thermokarst lagoons in the permafrost carbon cycle. To address this research gap, the key research questions of this thesis are: • What is the spatial pan-Arctic extent of thermokarst lagoons, how are they distributed and how can they be classified according to their development stages? • What is the effect of increasing seawater influence on GHG production and microbial community composition? • Does GHG production of inundated Arctic coastal lowlands differ between landscape features and regions? To answer these research questions, a combination of remote sensing and laboratory analyses was conducted. High-resolution satellite imagery and geographic information system tools were used for mapping and classifying thermokarst lagoons, providing a pan- Arctic context. The effect of seawater inundation on GHG production in thawed permafrost sediments was studied through four long-term anaerobic incubation experiments (at 4°C for up to 415 days) using permafrost, active layer, thermokarst lake, and lagoon sediment from three distinct Arctic coastal regions under various saline conditions to simulate increasing seawater influence. Microbial analyses were conducted before and after two of the experiments to understand how microorganisms respond to changing seawater influence. Along the Arctic coast between the Taimyr Peninsula (Siberia) and the Tuktoyaktuk Peninsula (Canada), 520 lagoons originating from former lakes and lake basins were identified and mapped with remote sensing and cover an area of 3,457 km², which is only a fraction of the area occupied by thermokarst lakes. Based on their connectivity to the sea, the lagoons were categorized into five classes, with the majority (55%) in early transition stages (very low and low connected). Incubation studies consistently revealed that under the prevailing brackish conditions, methane production is highest in these low-connected lagoons and decreases during the ongoing transition into a marine environment. Due to the higher global warming potential of methane, the climate impact of these low-connected lagoons is up to 18 times greater than that of open lagoons, where CO₂ is the dominant GHG produced. The microbial diversity is found to be higher in lagoon sediments than in terrestrial sediments, underlining the uniqueness of these transitional systems. However, the shift from terrestrial to marine conditions involves changes in oxygen availability and salinity, both of which disrupt the initial terrestrial microbial community, leading to a decrease in GHG production in the short term. Over time, as salt-tolerant, anaerobic microbial communities establish, CO₂ production increases, reaching levels as high as eight times those found in terrestrial permafrost. Combining data from all incubation experiments revealed that GHG production varies significantly during the transition from newly thawed terrestrial permafrost to established lagoons. GHG production is highest in lagoon sediments compared to the terrestrial landscape features (permafrost, active layer, lake). Although regional variations in GHG production exist, they are less pronounced than the variations between different landscape features. The combination of incubation and mapping results enabled the first rough estimate of potential carbon release from all mapped pan-Arctic lagoons, showing that, on average, 3 Tg CO₂eq could be released per year from all mapped thermokarst lagoons by 2100. In conclusion, these results show that, while the total thermokarst lagoon area is relatively small (3,457 km²), thermokarst lagoons release more than four times the CO₂- equivalent per unit area compared to thermokarst lakes, and even more than gradually thawing permafrost. This indicates that thermokarst lagoons are hotspots of carbon cycling and may play a more significant role in the carbon budget of rapidly thawing Arctic landscapes than previously anticipated. This role may become even more critical, as current climate scenarios suggest accelerating permafrost coastal erosion and rising sea levels. The incubation experiments were useful for understanding the impact of changing salinities on microbial dynamics and GHG production in seawater-inundated terrestrial and lagoon sediments, but it is not possible to reproduce all processes occurring in natural environments. To confirm and verify the presented incubation results, in situ measurements of GHG fluxes during the land-sea transition are needed.