CombiMo
CombiMo: Combining the sediment geochemistry of molybdenum with machine learning to identify anoxic-sulfidic seafloor environments along the German Baltic Sea coast
The Baltic Sea is currently undergoing drastic environmental changes driven by climate change and anthropogenic nutrient inputs from the surrounding countries. As a result of warming and eutrophication, hypoxic conditions (oxygen <63 µmol L-1) occur frequently in shallow (<30 m) coastal waters with severe consequences for ecosystems and the regional economies. Furthermore, in many coastal areas of the Baltic Sea, hypoxic events culminate in the release of hydrogen sulfide (H2S) from the surface sediment in the bottom water. Even though H2S is toxic to higher marine organism, it is currently unknown how widespread intermittently sulfidic seafloor environments are along the German Baltic Sea coast. This lack of knowledge is partly related to the inability of current water column monitoring programs to detect short-lived (order of days) sulfidic events. Furthermore, sulfidic water masses are often limited to the lowermost layer of the water column (i.e., a few meters or less from the bottom), which cannot be captured by conventional sampling approaches. To overcome this problem, we intend to establish sedimentary concentrations and isotope compositions of molybdenum (Mo) as an environmental indicator. Our preliminary data from Kiel Bight suggest that sedimentary Mo concentrations closely reflect the presence and concentration of H2S at the sediment-water interface. Based on this finding and by adding Mo isotopes as an additional constraint, we aim to develop a quantitative predictor for sulfidic conditions at the seafloor. Additional Mo concentration and isotope data will be collected in other areas of the German Baltic Sea, which are known to be affected by the seasonal occurrence of hypoxic conditions (Flensburg Firth, Schlei Fjord, Mecklenburger Bight). The Mo-based environmental indicator will be then used along with other predictors to train a machine learning algorithm for the spatial prediction of near-bottom sulfidic conditions in coastal areas. With this project we aim for a better understanding of the controls on and spatial distribution of anoxic-sulfidic seafloor environments along the German Baltic Sea coast.
Duration: 2026-2029
Funding source: DFG
Project partner: Dr. René Friedland, Leibniz-Institute for Baltic Sea Research Warnemünde (IOW)
Information about the project: Prof. Dr. Florian Scholz, Sarima Vahrenkamp