Doctoral thesis: Drought-influenced greenhouse gas fluxes and their emission pathways of the subsaline reed ecosystem of Lake Neusiedl

Wetlands are often spatially dynamic ecosystems and provide various ecosystem services, such as biodiversity, purification, livelihood, and storage of carbon and water. Wetlands dominated by the reed plant (Phragmites australis) can store carbon through photosynthetic assimilation of carbon dioxide (CO2) and sequestration of organic matter in sediments. For this reason, they are often known as strong CO2 sinks. However, these wetlands naturally release methane (CH4) produced in the waterlogged sediments. In the context of climate change, more frequent or severe droughts increase the likelihood of periods in which wetland ecosystems act as CO2 sources. Little is known about the contribution of central European reed wetlands as a source of greenhouse gases (GHG) due to drought. The main objective of this thesis is to enhance our comprehension of the influence of drought on the spatial and temporal dynamics of reed wetlands, particularly with regards to GHG fluxes, emission pathways, sediment microbial communities, and land cover. The studies were conducted in a wetland of international importance, the reed belt of Lake Neusiedl on the Austrian-Hungarian border, which is characterized by exceptional properties such as subsalinity and high sulfate content. In order to deepen and improve our understanding of subsaline reed ecosystems, their processes, and their interactions, a combination of methodologies was applied. These included techniques for measuring GHGs, such as eddy covariance, chambers, and bubble traps; stable carbon isotope measurements; laboratory analyses of water and sediment properties; remote sensing using drone and satellite imagery; and image and data analyses, including deep learning and vegetation indices. Since mid-2015, Lake Neusiedl’s catchment area has experienced drought, with mostly negative SPEI (standardized precipitation evapotranspiration index) values, leading to a decrease in water levels. Under these conditions, the GHG exchange between this reed ecosystem and the atmosphere was continuously measured with an eddy covariance tower over 4.5 years (mid-2018 to 2022). From 2019 to 2022, annual CH4 emissions decreased by 76% from 9.2 to 2.2 g CH4-C m-2 a-1. Initially, annual CO2 emissions decreased by 85% from 181 to 27 g CO2-C m-2 a-1 due to reed growth from 2019 to 2021. However, by 2022, they had risen to twice the 2019 level (391 g CO2-C m-2 a-1). This was the result of a sharp drop in sediment water content (SWC) from about 65 to 32 Vol-% in mid-2022, suggesting that SWC is a good proxy of the magnitude of CO2 emissions from wetlands. In general, this study challenges the assumption that all reed wetlands are CO2 sinks. Another study in this thesis explored the intra-annual dynamics of the land cover mosaic of reed, water, and sediment patches within the reed ecosystem and its phenology within one year. The results showed that within the reed ecosystem in 2021, the reed area increased by 10%, the open sediment areas increased by 27%, and water areas decreased by 23% from May to November. Nevertheless, the reed area remained dominant throughout, accounting for 61–71%. At lakes, and especially shallow lakes like Lake Neusiedl, it is often assumed that ebullition is the dominant CH4 emission pathway, although ebullition is difficult to study and is therefore often not measured. Thus, one study in this thesis investigated whether ebullition or diffusion is the more important CH4 emission pathway in three subsystems of the Lake Neusiedl ecosystem: the Reed belt, the Channel, and the Open water/Lake. The results show that at the Reed belt ebullition accounted for 48% of the sum of the two CH4 emission pathways on a cumulative basis, while at the other two subsystems it was responsible for only about 1%. The Reed belt with 17 ± 28 mg CH4 m-2 d-1 showed 340 times higher mean CH4 ebullition rates than the other two subsystems. In this reed wetland, all assessable emission pathways of CH4 from the ecosystem to the atmosphere were investigated individually with chamber and ebullition trap measurements, in particular the contribution and isotopic signature of ebullition, but also the contribution of the plant-mediated transport and diffusion pathways at the water–air and sediment–air interfaces in each season for one year. In addition, the research aimed to investigate whether the reed ecosystem exhibits a diel cycle (24 h) of CH4 emissions depending on the season and emission pathway. To improve system understanding of subsaline reed wetlands, the physicochemical properties and the microbial communities such as methanogens, methanotrophs, and sulfate reducers in the sediments were investigated and how they changed due to desiccation in one year. The plant-mediated transport showed the highest median CH4 emission rates in all seasons, not only in summer but also in winter. A pronounced diel pattern of the CH4 was found only for the plant-mediated transport in summer. Both approaches, the stable carbon isotope values of the ebullition gases from March to July 2021 and the seasonal source signatures of the Keeling plots, demonstrated that at Lake Neusiedl the dominant methanogenic pathway in the sediment is acetoclastic methanogenesis yearlong. Overall, this thesis provides insights into the interplay of various dynamics in reed wetlands in relation to land cover, GHG fluxes, emission pathways, microbial diversity, and sediment properties, both within and between years, especially under the influence of drought.