Paint Particles in Marine Sediment: Interactions with Microbiota and Effects on Sediment Processes

• Applicant: Alexander Tagg, Ph.D.
• Subject Area: Microbial Ecology and Applied Microbiology ; Hydrogeology, Hydrology, Limnology, Urban Water Management, Water Chemistry, Integrated Water Resources Management ; Oceanography
• Term: since 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 452841212

This proposal will investigate how paint particles affect the microbial communities in marine sediment. The goal is to build and validate a model capable of determining paint pollution levels in sediment from 16S microbial community data. In order to produce and validate this model, 3 objectives need to be achieved. Paints vary considerably in chemical composition. Therefore, the first objective is to determine which specific chemical attributes of paints are responsible for key microbial changes in marine sediment exposed to paint particles. This will be addressed using a laboratory exposure experiment. Sediment containing paint particles of varying chemistry (including antifouling ingredients) will be incubated over time and the microbial community of this sediment will be sequenced. Supervised machine learning approaches, such as random forests, will be used to determine which specific paint chemistry has the greatest effect on microbial communities, especially with regard to changes in taxa important in sediment biogeochemical processes. Once the chemical attributes are known, the second experiment will proceed. The second experiment aims to address the second objective; to determine how paint particles affect surrounding sediment microbial communities and to model how this effect scales with paint particle concentration in sediment, ultimately determining at what level paint particle contamination in sediment causes changes critical enough to implicate biogeochemical processes. Specially-designed chambers will be deployed into the Baltic Sea containing sediment and varying amounts of paint particles (the chemistry of which has been pre-determined by the prior experiment). After an exposure period, the chambers will be collected and the sediment microbial communities sequenced. Random forests will be used to construct a predictive model for paint pollution levels as a function of the microbial community composition. Additionally, phylogenetic distance trees of key taxa will be combined with available literature to estimate changes in microbially-mediated biogeochemical cycles. Estimates of how environmental parameters (i.e. hydrogen sulphide or iron levels) might change will be included into the model.The final objective is to validate the model. To do this, a number of locations on the German Baltic Sea coast will be sampled. Sediment will be tested for both paint pollution levels and environmental parameters. The sediment microbial community will also be sequenced and the model will be used to predict the paint pollution levels based on the microbial community compositions. This prediction will be compared to real pollution and environmental data. In this way the model can be assessed, adjusted and ultimately validated. The final outcome of the work will be a working model which can predict paint pollution levels based on 16S microbial community data, available as a python script package.

• DFG Programme: Research Grants