Environmental DNA signatures distinguish between tsunami and storm deposition in overwash sand

The increasing coastal populations heighten the risk of coastal hazards like tsunamis and storm surges. Assessing these risks is vital for coastal development. While modern events are well-documented, understanding the frequency and magnitude of past events requires examining geological records. Tsunami and storm surge deposits are often indistinguishable based on traditional sedimentary characteristics, leading to ambiguity in hazard assessments. Previous methods, such as grain size analysis, microfossil analysis, and chemical analysis, have limitations due to factors like preservation issues, post-depositional modification, and ambiguous elemental sources. This study explores the potential of microbial eDNA analysis as a novel approach to distinguish between tsunami and storm deposits. Microbial communities are known to respond to abrupt environmental changes caused by catastrophic events, offering a potential unique 'fingerprint' for different events.

Existing literature demonstrates microbial community changes in response to pulse disturbances from catastrophic events like coastal flooding, wildfires, and earthquakes. Previous studies on tsunami deposits have used both culture-based and metabarcoding techniques, but these have limitations, including focusing on culturable fractions of microbial communities, neglecting detailed signatures, and failing to directly compare tsunami and storm deposits within the same geological record. While some studies used metabarcoding to compare pre- and post-tsunami microbial communities or to contrast tsunami deposits with unflooded sediments, none have directly compared tsunami and storm deposits at the same site using this technique.

Sediment samples were collected from Phra Thong Island, Thailand, and Cuddalore, India, both affected by the 2004 Indian Ocean Tsunami and subsequent storm events. Phra Thong Island had the 2004 tsunami deposits preserved in a swale, with a 2007 storm overwash deposited behind the modern berm. Cuddalore had 2004 tsunami deposits behind a beach dune, inundated by Cyclone Thane in 2011. The sites offered a comparison of the same tsunami impacting different morphologies and different storm events affecting the same coastlines. Samples included sediment cores, beach sand, intertidal sand, and marine sediments. A high-throughput metabarcoding approach was employed, targeting the SSU rRNA gene (both 16S for prokaryotes and 18S for eukaryotes). Environmental DNA was extracted and amplified, followed by sequencing and bioinformatic analysis. Diversity indices (Shannon and Simpson) were calculated. Principal Coordinate Analysis (PCoA) and distance-based redundancy analysis (dbRDA) were used to analyze microbial community composition and dissimilarities between sample types. Geochemical analyses (TOC, TN, TS) and grain-size analyses were also conducted. Differential analysis using DESeq2 was performed to identify ASVs significantly differentiating between tsunami and storm deposits.

The study revealed higher microbial diversity on Phra Thong Island than Cuddalore. Marine sediments had the highest diversity at both locations. Overwash deposits showed the lowest diversity compared to other sediment types. PCoA analysis demonstrated significant differences in microbial communities between marine sediments and onshore samples (beach, intertidal, backdune, swale). Overwash deposits showed high similarity to overlying and underlying soils. dbRDA showed that storm deposits were significantly different from tsunami deposits and surrounding sediments at both sites. While the differences were more pronounced for storm deposits, the analysis also differentiated tsunami deposits from the surrounding sediments. Differential abundance analysis revealed specific ASVs associated with storm deposits (e.g., families Blastocatellaceae, Holophagae, and Chitinophagaceae) at both locations. Tsunami deposits showed less distinct unique ASVs, suggesting more complex interactions with surrounding materials.

The findings confirm that microbial community composition differs between tsunami and storm deposits, even within the same geographic location. This difference reflects distinct depositional mechanisms and the varying sources of sediments transported. The success of the method in identifying relatively minor storm deposits indicates its potential for historical coastal hazard studies. The subtle differences in tsunami deposits compared to storm deposits and surrounding sediments may be due to mixing of underlying sediments during deposition and backflow. The study suggests that, despite the ubiquity of many microbial taxa, specific ASVs can differentiate between tsunami and storm events. Although some ASVs might indicate storm events, more research is needed to confirm the presence of global signatures for distinguishing different overwash events. The eukaryotic community analysis did not show as strong a signal of distinction, possibly due to limitations in primer design and sequencing length.

This study provides compelling evidence that metabarcoding of eDNA can effectively discriminate between tsunami and storm deposits. This method offers a powerful tool for improving coastal hazard assessments and understanding past events. Future research should focus on extending this method to older deposits, exploring the environmental factors shaping microbial community differences, and searching for robust global microbial signatures for different overwash events.

The study focused on modern tsunami and storm deposits, limiting the ability to generalize findings to older events. The analysis of eukaryotic communities was less conclusive, requiring improvements in methodologies. Further research is needed to investigate the factors driving the observed microbial community differences between tsunami and storm deposits. The specific ASVs identified as differentiating markers require validation across a wider range of sites and events to establish their general applicability.