Tromsø - A Summary of the Presentations

Published:June 2025

This post summarises presentations from a conference in Tromsø focused on environmental DNA (sedaDNA), highlighting key developments in methods, interpretation and applications in palaeoecology.

Key themes and insights from the Tromsø conference

Lacustrine sedaDNA 1

1. This study uses sedimentary DNA from Lake Monticchio (Italy) to reconstruct vegetation and environmental change over ~33,000 years, comparing results with pollen data.Lake Monticchio sedaDNA over ~33 ka shows DNA gives more local, precise vegetation signals than pollen, but both together provide the most robust reconstructions. This opening talk sets out the broader goals of environmental DNA research in the context of climate change and ecosystem dynamics. It emphasises how sedimentary DNA allows researchers to reconstruct past plant and animal communities and assess key questions such as whether climate or biotic interactions drive ecosystem change. The central message is that eDNA provides a powerful tool to link molecular data to large-scale ecological patterns and to understand how ecosystems respond to environmental change over time.
2. DNA provides more localised and taxonomically precise signals, particularly for lake-edge and aquatic vegetation, while pollen captures broader regional patterns. The findings show that combining both proxies improves resolution and highlights differences in how each records past ecosystems. Overall, sedimentary DNA is shown to be a powerful complement to traditional paleoecological methods.
3. Southeast Alaska sedaDNA reveals postglacial plant succession driven by major climate events (e.g. Younger Dryas, mid-Holocene wet phase). This talk reconstructs plant community development in Southeast Alaska following deglaciation using sedimentary DNA. The results show clear ecological shifts linked to major climate events, including the Younger Dryas and mid-Holocene wet period, which strongly influenced vegetation composition. Early communities were dominated by pioneer species, followed by increasing diversity and stabilisation into modern forest systems. The study demonstrates how climate transitions drive long-term vegetation dynamics.
4. This study uses sedimentary DNA from a Swedish lake to investigate postglacial environmental change and species colonisation. Early results are highly preliminary and reveal challenges with data quality, non-specific taxonomic assignments, and database limitations. Sedimentation patterns and lithology appear to influence DNA preservation and signal strength. Despite these issues, the work aims to refine methods and ultimately reconstruct ecosystem development and possible marine influence following deglaciation.
5. Long-term lake sedaDNA shows increasing plant diversity linked to grazing pressure, especially cattle, alongside climate influences. This presentation examines long-term changes in plant diversity and grazing dynamics using sedimentary DNA from multiple lakes. The results show increases in plant richness over the past 4,000–2,000 years, aligning with pollen records but offering finer resolution. Changes in vegetation are closely linked to the appearance of domesticated animals and grazing pressure, with cattle identified as a major driver of plant diversity, followed by large wild herbivores. The study suggests that both climate and herbivore activity shape vegetation patterns, with open habitats supporting higher diversity. It highlights the combined influence of human activity and natural processes on ecosystem development.

DNA Taphonomy

1. DNA preservation is strongly controlled by mineral binding, with pH and salinity governing adsorption and long-term stability. This talk explores how environmental DNA (eDNA) is preserved in sediments, focusing on interactions between DNA and minerals—particularly goethite (an iron oxide). The speaker shows that DNA binding is strongly influenced by pH and salinity, with greater adsorption at low pH and high salinity, especially in the presence of magnesium ions. Experiments demonstrate that DNA remains tightly bound under acidic conditions and is more easily released at higher pH. Using chemical force microscopy, the study confirms that molecular interactions are stronger in acidic, saline environments. Overall, the work suggests that mineral composition and environmental chemistry play a critical role in long-term DNA preservation and distribution in sediments.
2. Different DNA fractions (intra- vs extracellular) vary in preservation, and combining them improves reconstruction of past ecosystems. This talk examines how different fractions of sedimentary DNA (extracellular vs intracellular) affect ecological interpretation. By analysing lake sediment cores, the study shows that DNA preservation varies across layers due to environmental factors such as oxygen exposure, erosion, and human activity. Combining multiple DNA fractions provides a more complete picture of past ecosystems and improves understanding of taphonomic processes and historical environmental change.
3. Lake morphology and environmental conditions strongly influence plant DNA preservation and distribution in sediments. This study investigates what controls plant DNA preservation in lake sediments across Wisconsin. By analysing sediment cores from 20 lakes, the researcher finds that lake morphology (depth, shape, fetch) and water quality explain about 55% of DNA variation. Most DNA originates from aquatic plants, likely due to their proximity to deposition sites. Contrary to expectations, deeper lakes sometimes contain less DNA, possibly due to transport dynamics. A key finding is that older (deeper) sediment layers contain more DNA than surface layers, potentially reflecting historical land use and erosion. The work highlights that both environmental factors and historical processes shape sedimentary DNA signals.
4. Environmental DNA distribution reflects animal presence and behaviour, showing spatial clustering linked to activity patterns. This talk investigates whether environmental DNA can reflect animal presence and behaviour in modern environments. By sampling different locations, the study shows that DNA distribution is uneven and often clustered around areas where specific animals spend time, such as individual dogs. The results demonstrate that DNA signals correlate with physical presence and movement patterns, suggesting that eDNA can be used not just for species detection but also for understanding spatial behaviour. The work highlights the potential of eDNA as a tool for tracking activity patterns in real-world settings.
5. Sedimentary DNA signals are shaped by transport and deposition processes, requiring integration with sedimentology for correct interpretation. This talk connects sedimentology with sedimentary DNA, arguing that DNA must be understood within the context of sediment transport and deposition processes. Using case studies (e.g. tsunami deposits), the speaker shows that eDNA can help identify sediment origins—even when traditional indicators are lost. DNA can survive extreme conditions and be transported across environments, making it a powerful tracer. However, interpretation is complex because signals may reflect transported rather than local ecosystems. The talk emphasises the need for integrating sedimentology and DNA analysis to correctly interpret environmental records.

Plenary – Eske Willerslev

• Ancient environmental DNA provides a blueprint of past ecosystem adaptations, with major potential to inform climate-resilient agriculture and biodiversity conservation. This plenary talk outlines a large-scale research project using environmental DNA to address global challenges in food security and biodiversity under climate change. The central idea is to use ancient DNA records as a “blueprint” of how ecosystems adapted to past environmental changes, including species composition, genetic traits, and microbial interactions. These insights can be applied to develop more resilient crops and sustainable agricultural systems. The talk also highlights broader advances in environmental DNA, including ecosystem reconstruction, genome recovery from sediments, and the importance of reference databases and sediment integrity for accurate interpretation.

Lacustrine sedaDNA 2

1. Swedish sediment unexpectedly yielded a near-complete ancient brown bear genome, revealing a distinct lineage and highlighting sediment DNA potential. This study uses sedimentary DNA from a ~9,600-year-old site in Sweden to reconstruct early postglacial environments and unexpectedly recovers a near-complete brown bear genome. Genetic analysis suggests the presence of a previously unrecognised lineage in northern Scandinavia. The finding demonstrates that sediments can preserve high-quality DNA from large animals and provide new insights into early species colonisation and population history, even from unexpected “bycatch” data.
2. SedaDNA shows postglacial colonisation in waves, with gradual environmental change driving biodiversity and megafaunal decline. This talk reconstructs long-term ecological change using sedimentary DNA, focusing on vegetation and mammal communities. The results show multiple waves of species colonisation following glaciation, increasing biodiversity over time, and the eventual disappearance of megafauna such as mammoths. These changes appear to be driven more by gradual environmental shifts than sudden events. The study highlights how sedimentary DNA can track both biodiversity trends and large-scale ecological transitions.
3. Beaver activity reconstructed over millennia shows strong ecosystem engineering effects, increasing biodiversity and habitat complexity. This study investigates beavers as ecosystem engineers using sedimentary DNA to reconstruct their long-term presence and ecological impact. The results suggest that beavers have persisted for thousands of years and are associated with increased plant diversity and habitat complexity. Their dam-building and landscape modification create conditions that support biodiversity. The work emphasises the importance of understanding historical baselines to inform modern ecosystem restoration efforts.
4. Ecological networks reconstructed from DNA show declining long-distance interactions and increasing ecosystem fragmentation over time. This talk reconstructs ecological interaction networks between plants and mammals using fossil and DNA data. Over time, the study finds a decline in long-distance ecological connections and an increase in local trophic interactions, leading to greater ecosystem fragmentation. These structural changes suggest that ecosystems have become more locally constrained, with implications for resilience and function. The work highlights the importance of interaction networks, not just species lists, in understanding ecosystem change.

Computational Methods 1

1. New method estimates ages of ancient DNA sequences directly from molecular data, improving chronological resolution. This talk presents a new method for directly estimating the age of ancient DNA sequences using molecular data rather than relying on external dating (e.g. radiocarbon or stratigraphy). The approach addresses limitations of previous methods by being computationally efficient, able to handle fragmented DNA, and explicitly accounting for DNA damage. Simulations show that the method can recover accurate age estimates, even for degraded sequences. The broader aim is to build more reliable timelines for environmental DNA and better link molecular signals to real chronological history.
2. Large collaborative databases enable global-scale sedaDNA analyses but require strong standardisation and data integration. This talk focuses on building large, community-driven databases of sedimentary DNA and paleoecological data (e.g. Neotoma) to enable global-scale analyses. With thousands of sites now available, the challenge is to standardise, harmonise, and integrate diverse datasets across regions and methods. Such databases allow researchers to study long-term biodiversity and climate change patterns, compare rates of ecological change, and identify global trends. The key message is that collaborative data infrastructure is essential for advancing the field beyond individual case studies.
3. Integrating sedaDNA with process-based models improves predictions of vegetation change by including biotic interactions. This study examines how to improve predictions of vegetation change under climate change by combining process-based ecological models with sedimentary DNA data. Traditional climate-envelope models often fail because they ignore factors like species interactions, competition, herbivory, and demographic processes. By incorporating these into a dynamic model, the study better explains observed vegetation changes over time. The results suggest that biotic interactions and ecosystem processes are critical for accurate forecasting, and that integrating DNA-based reconstructions with mechanistic models can significantly improve predictions of future landscapes.
4. Ecological modelling of sedaDNA data is sensitive to degradation and bias, but can still reconstruct past environments if carefully applied. This talk explores the application of ecological niche modelling and transfer functions to sedimentary ancient DNA data. While these approaches can reconstruct past environments, they are challenged by DNA degradation, noise, and uneven diversity in ancient datasets. The speaker shows that method choice strongly affects outcomes, particularly in estimating diversity and environmental parameters. Despite these challenges, carefully calibrated approaches can still extract meaningful ecological signals, highlighting both the potential and limitations of modelling with ancient DNA.

Molecular Methods

1. Species identification depends heavily on bioinformatic pipelines and reference databases, affecting accuracy and bias. This study evaluates how different bioinformatic pipelines and reference databases affect species identification in sedimentary DNA, particularly for marine microbial communities. Results show that performance varies widely depending on database completeness and pipeline design, with trade-offs between sensitivity (detecting more species) and precision (avoiding false identifications). The work highlights the importance of well-curated reference databases and careful methodological choices to ensure reliable biodiversity reconstructions.
2. Poor filtering and database issues can inflate biodiversity estimates, highlighting the need for strict quality control.This talk investigates the problem of false positives and inflated species richness in sedimentary DNA analyses. It shows that results are highly sensitive to filtering thresholds, read counts, and especially the choice of reference database. Without careful processing, datasets can overestimate biodiversity by including spurious matches. The study emphasises the need for robust filtering strategies and database validation to produce accurate ecological interpretations.
3. Automation enables high-throughput ancient DNA processing, improving efficiency and reproducibility at scale. This presentation describes the development of a large-scale automated laboratory facility for ancient DNA processing. Automation enables high-throughput, standardised workflows, allowing thousands of samples to be processed efficiently with reduced human error. The system integrates sample handling, extraction, and sequencing preparation within a modular infrastructure. While highly effective, it requires significant investment and specialised expertise. Overall, the work demonstrates how automation can transform the scale and reproducibility of ancient DNA research.
4. Optimised capture probe design improves recovery of informative DNA regions, especially for low-yield samples. This talk introduces new computational methods for designing capture probes to improve recovery of ancient DNA from low-yield samples. The approach targets informative genomic regions and species-specific markers, enabling both single-species and whole-community analyses. By optimising probe design and filtering problematic sequences, the method improves capture efficiency and standardisation. This has important applications for biodiversity reconstruction and population genetics from challenging samples.
5. Different sequencing approaches (shotgun vs targeted) yield contrasting biodiversity results, requiring careful method selection. This talk compares shotgun metagenomics and targeted capture (metabarcoding) approaches for analysing sedimentary DNA from marine cores. The results show that different methods—and especially different reference databases—can produce contrasting taxonomic profiles and depth patterns. Shotgun data may detect broader diversity, while targeted approaches can be more sensitive to specific groups. The key takeaway is that method choice strongly influences ecological interpretation, and combining approaches may provide a more reliable reconstruction of past marine biodiversity.
6. Advances in metagenomics allow reconstruction of ancient microbial genomes and functions, moving beyond species identification. This talk explores how advances in ancient DNA, particularly long-read sequencing and metagenome assembly, allow reconstruction of entire ancient microbial genomes and their functions. The speaker highlights the potential to recover and “resurrect” lost biomolecules and metabolic pathways, which may have disappeared from modern ecosystems. These reconstructed functions could have applications in biotechnology, such as improving crop resilience or restoring beneficial microbiomes. The work points to a shift from simply identifying species to understanding and utilising ancient biological functions.
7. Ancient proteins extend molecular analysis beyond DNA limits, enabling study of deeper time despite technical challenges.

Plenary – Ulrike Herzschuh

• Ecosystem change is best understood through species interaction networks, showing how biodiversity and relationships shift through time. This talk discusses the use of environmental DNA and network-based approaches to reconstruct past ecosystems and species interactions over time. Using time-sliced data and sequencing approaches, the speaker explores how ecological networks (e.g. plant–animal interactions) can be inferred and how they change across climatic periods such as interglacials. The work suggests that shifts in biodiversity—including megafaunal extinctions—can be understood through changes in interaction networks rather than just species loss. It also highlights the integration of bioinformatics, shotgun sequencing, and co-occurrence analyses to uncover complex ecological dynamics, though methodological challenges and interpretation limits remain.

Computational Methods 2

1. Improved probe design and filtering strategies enhance DNA capture for population genetics from challenging samples. This talk addresses challenges in obtaining sufficient DNA for population genetics, especially for rare species. It presents a method for designing and filtering capture probes to improve the recovery of informative genetic regions while reducing bias. The approach focuses on selecting functional genomic regions, optimising probe sets, and validating results through benchmarking. The goal is to enhance the reliability and efficiency of DNA capture for population-level analyses, particularly in difficult or low-yield samples.
2. Standardised pipelines for genome selection and assembly improve consistency in large genomic datasets. This talk discusses challenges in managing rapidly growing genomic databases and proposes a pipeline for standardising genome selection and assembly. The method involves filtering genomes, re-mapping reads, estimating coverage, and ensuring consistent quality control. By creating a more controlled and reproducible workflow, the approach aims to improve the reliability of large-scale genomic analyses.
3. Arctic sedaDNA analysis uses classification and mapping tools to track species turnover, though accuracy remains a challenge. This study focuses on identifying species and population turnover using environmental DNA across Arctic samples. It uses Kraken-based screening and genome mapping to filter and classify reads, followed by validation steps to avoid misclassification. The approach aims to track biodiversity and population changes, though computational demands and classification accuracy remain key challenges.

Lacustrine sedaDNA 3

1. Combining pollen and sedaDNA reveals Holocene vegetation shifts, with strongest changes in the last 300 years. This study reconstructs environmental change in the Stoltenberg region using both pollen and sedimentary DNA. The results show largely continuous sedimentation and significant ecological shifts during the middle to late Holocene, with especially strong vegetation changes in the last 300 years. Differences between proxies highlight biases in each method, but overall agreement strengthens confidence in the findings. The key message is that combining multiple proxies provides the most reliable reconstruction of past environments.
2. SedaDNA captures winter microbial communities, showing seasonal ecological dynamics in lake sediments. This study explores winter ecology in Siberian lakes using sedimentary DNA, focusing on microbial communities such as diatoms, fungi, and bacteria. By comparing winter, summer, and sediment samples, the research shows that distinct winter communities are preserved in sediments, capturing seasonal ecological variation. The work demonstrates that sedimentary DNA can provide insights into under-ice ecosystems and seasonal dynamics, which are often poorly represented in traditional ecological records.
3. Multi-proxy lake study links vegetation, monsoon strength, and human activity over the past ~2–3 ka. This study uses lake sediments as archives to reconstruct past climate, vegetation, and human activity using a combination of geochemical proxies and sedimentary DNA. Indicators such as alkanes (vegetation), sediment input (monsoon strength), and combustion markers (human activity) are combined with DNA evidence to build a multi-proxy reconstruction. The results show shifts in vegetation and monsoon intensity over time, along with clear evidence of human presence and landscape impact over the last ~2,000–3,000 years.
4. Tibetan lake records show hydrological and ecological shifts driven by monsoon variability over ~2 ka.This talk reconstructs past environmental and hydrological changes in a large Tibetan Plateau lake (Namtso) using ostracods and sedimentary DNA. Changes in species composition reflect shifts in lake level, salinity, and precipitation, linked to monsoon dynamics over the past ~2,000 years. The study highlights strong environmental variability and demonstrates how combining biological proxies with DNA can improve reconstructions of climate-driven ecosystem change.

Closing Address

• The closing address for the conference, expressing thanks to organisers, volunteers, and participants. The collaborative nature of the research community is emphasised and continued sharing of ideas, methods, and results to advance the field is encouraged. Practical information about post-conference activities is also provided.

Final Call for Publication

• Combining conference announcements with a publishing proposal, encouraging participants to contribute to a special issue in a Quaternary science journal. The benefits of publishing in reputable journals are outlined, including editorial support and open-access options. A key message is that early-career researchers are encouraged to participate, both as contributors and potential editors, highlighting the importance of community-led publication initiatives.