Read the paper: Trubl et al., 2025
Permafrost, Carbon, and the Viral Unknown
Permafrost is frozen ground that can be imagined as a vast natural freezer. Locked within it is an amount of carbon roughly double that currently in Earth’s atmosphere, preserved as incompletely decomposed organic matter. The upper portion of permafrost, known as the active layer, thaws each summer and supports thriving microbial and plant communities. As the planet warms, the active layer deepens and remains thawed for longer periods, triggering dramatic environmental shifts. These changes prevent the ground from re-freezing and accelerate permafrost thaw, exposing once-stable carbon to microbial decomposition and releasing carbon dioxide and methane; thereby intensifying a powerful climate feedback loop.
We know that microbial communities change rapidly as thaw progresses, but the viral roles in this transformation are still largely unknown. Viruses shape microbial ecology across every biome on Earth, yet in soils their diversity, functions, and ecological impacts remain poorly understood. Are viruses helping retain carbon by keeping microbial growth in check, or accelerating its release by stimulating microbial turnover? These questions guided our investigation.
Why Stordalen Mire?
Our fieldwork centered on Stordalen Mire, a long-term ecological research site that naturally spans a permafrost thaw gradient—from intact permafrost (palsa), to partially thawed bog, to fully thawed fen. This natural mosaic allowed us to examine viruses across stages of thaw while minimizing environmental disturbance. Nearby, the Abisko Scientific Research Station provided essential support through laboratory facilities, on-site housing, and a vibrant research community.
Collecting samples across this gradient was both breathtaking and physically demanding. The terrain felt unlike anything I had walked on before, somewhere between a sponge, a bouncy mattress, and shifting ground. Each habitat required different coring strategies. The palsa is firm yet fragile, the bog is a moss-held mat that can break apart as you cut, and the fen is waterlogged, with dense sediments that make deep coring strenuous. Temperature swings were dramatic, ranging from near freezing to unexpectedly hot days reaching 29°C, all while giant flies buzzed around us. A curious moose even wandered close one afternoon, reminding us that the mire is very much alive. We were not alone in other ways, as the site is popular among researchers and locals, thanks to its boardwalks and the delicious cloudberries that ripen during our summer sampling months.
From Field to Viromes: Illuminating the Unseen
Studying viruses in soil requires peeling back layers of biological complexity. Soil is a structured and heterogeneous matrix of microbial cells, DNA fragments, plant debris, minerals, and countless other components that readily obscure viral signals. Previous metagenomic studies at Stordalen Mire had detected hints of viral activity (Emerson et al., 2018), but these “bulk” datasets were dominated by microbial DNA, offering only a partial view of viral diversity. To truly reveal the viral communities, we turned to an optimized viromics approach that physically removed particles larger than viruses prior to sequencing. Early in the project, we conducted multiple experiments to test how we biased our samples, one example being some samples were chilled on ice, while others were flash-frozen in liquid nitrogen (Trubl et al., 2016, 2018, 2019). Successfully shipping our carefully collected samples back to the lab felt like an early victory. The next came months later, after extensive troubleshooting and many long hours in the lab, when we finally recovered high-quality viral DNA that captured both ssDNA and dsDNA viruses from the peat.
What Thousands of Viruses Revealed
Once the viromes were digitized, we were awarded a Department of Energy SCGSR fellowship to develop and refine a bioinformatics pipeline at the Joint Genome Institute. This opportunity allowed me to sharpen my computational skills (JGI podcast) and test a suite of tools designed to improve viral genome detection and characterization (cite bioinformatics paper link). The resulting dataset uncovered nearly 10,000 viral populations—a tenfold improvement compared to our previous metagenomic efforts—and revealed a rich mixture of dsDNA viruses, commonly captured in environmental sequencing, and ssDNA viruses, which are widespread but historically underrepresented due to methodological constraints and their small genomes.
Across the thaw gradient, viral communities shifted markedly in both composition and abundance. Thawed habitats harbored more active and diverse viral populations, mirroring microbial changes as the ecosystem transitions toward faster nutrient turnover. Earlier long-term studies at Stordalen Mire had hinted at habitat-specific viral patterns, with bogs often hosting the richest viral communities and fens containing the highest abundances (Sun et al., 2024), and our viromics dataset revealed similar trends. Habitat consistently emerged as a stronger driver of viral community structure than depth or sampling year, highlighting the strong influence of thaw stage on viral ecology.
What truly stood out, however, was the unprecedented view of ssDNA viruses in these peatland soils. Historically underrepresented in environmental studies due to methodological constraints and their small genomes, ssDNA viruses have remained largely invisible in permafrost research. By combining optimized laboratory workflows with targeted bioinformatics, we recovered hundreds of ssDNA viral populations, many belonging to the phylum Cressdnaviricota. These viruses were especially enriched in the fully thawed fen and displayed striking taxonomic diversity, including numerous lineages with no close cultured relatives. Their strong habitat specificity and dominance in the fen suggest that ssDNA viruses may be far more ecologically important in permafrost-affected soils than previously recognized. Together, these patterns show that the accelerated ecological tempo of thawed soils, defined by rapid microbial growth and decay, is not only reflected in but may be amplified by viral dynamics.
Viruses often infect the most abundant and active microbial hosts, initiating cycles of infection, lysis, and nutrient release. These processes can increase microbial diversity, stimulate horizontal gene transfer, and mobilize both carbon and nutrients. In thawing permafrost systems, such viral-driven turnover may accelerate the decomposition of ancient carbon stores, ultimately enhancing greenhouse gas emissions.
Among the most intriguing discoveries were viral genes predicted to encode enzymes involved in carbon degradation. Similar to auxiliary metabolic genes observed in marine viruses, these soil viral genes may augment host metabolism during infection and boost the breakdown of complex organic matter. If expressed in situ, they could represent a previously overlooked biological mechanism influencing carbon flux from thawing permafrost.
Why This Matters and What Comes Next
Understanding how viruses function across different stages of permafrost thaw is critical. These ecosystems are changing rapidly, and predictive climate models require accurate representations of all major biological drivers. Viruses are not merely pathogens but potential key agents in global nutrient and carbon cycles, capable of amplifying or moderating microbial activity. Incorporating viral processes into ecosystem models may therefore be essential for accurately predicting carbon feedbacks in a warming Arctic.
Our findings indicate that as permafrost thaws, viruses increasingly shape how carbon moves through the ecosystem and how much ultimately reaches the atmosphere. Quantifying the magnitude of viral contributions to carbon cycling is an important next step. We also need to explore viral roles in other soil systems worldwide, as viruses often respond to environmental change before their hosts and may serve as early indicators of ecosystem transformation.
Northern peatlands, especially those underlain by permafrost, harbor astonishing biological complexity. As we continue to uncover the hidden interactions between viruses and microbes, we not only gain scientific insights but also deepen our understanding of the processes shaping Earth’s climate.