Diatoms are photosynthetic unicellular algae that inhabit almost all aquatic environments and are also found in moist terrestrial habitats. These organisms have several forensic applications, ranging from post-mortem interval (PMI) estimation to drowning diagnosis and locating drowning sites. Due to their high sensitivity to environmental conditions, diatom communities are locally shaped by physicochemical and biological factors providing potential information on the location of the drowning site. Traditionally, diatoms have been analysed morphologically, but in recent years there has been an effort to employ molecular techniques such as metabarcoding, which do not require the high taxonomic expertise needed for microscopic species identification. In this project, we investigate the spatial and temporal distribution of diatom communities using metabarcoding, with a specific focus on a section of the river Aare catchment in Switzerland, comprising two lakes (lake Thun and lake Biel). Most fatal drownings in Swiss waters occur in this river, where the victim is often dragged away for long distances by the strong current. This makes it difficult to identify the exact location where the drowning occurred, a piece of information that can nevertheless be highly relevant to the case. 

For this study, a total of 450 environmental samples were collected over the course of a year. From these, high-throughput sequencing data of the rbcl gene allowed us to identify 11,000 different amplicon sequence variants (ASVs) classified in the Bacillariophyta phylum comprising 73 genera and 203 distinct diatom species. By comparison, a water quality assessment of this river stretch in 2018, identified only 73 different diatom taxa using light microscopy. This highlights the usefulness of metabarcoding for identifying the vast diversity of diatom assemblages. A clear spatial structure in the composition of diatom communities was observed among connected water bodies (lake Thun, river Aare, lake Biel), indicating the potential use of diatom communities in distinguishing between drowning in lotic (flowing freshwater habitats) or lentic (standing freshwater bodies) environments. Spatial and temporal information on diatom communities will allow us to generate a comprehensive diatomological map of this catchment and, ideally, infer drowning sites by comparing diatoms identified in tissues of drowning victims to this map.

Authors

• Elia Rossini (Institute of Forensic Medicine, University of Bern and Institute of Ecology and Evolution, University of Bern, Switzerland)
• Gerald Heckel (Institute of Ecology and Evolution, University of Bern, Switzerland)
• Martin Zieger (Institute of Forensic Medicine, University of Bern, Switzerland)