The study system is remarkable in its simplicity and its power. Two bays separated by roughly 1.5 kilometers of open Baltic Sea experience nearly identical geography, salinity, and water chemistry, with one critical difference. The heated bay has absorbed thermal discharge for over fifty years, raising its average water temperature by about 5.2 degrees Celsius relative to the control bay, a difference comparable to the warming projected for the Baltic Sea by 2100 under the IPCC’s high-emissions SSP5-8.5 scenario. Continuous temperature logging from December 2017 to November 2023 confirmed that during the experiment itself, from August 2021 to September 2022, the heated bay averaged 16.1 degrees Celsius against 10.7 degrees in the control. This makes the paired bays a rare real-world window into what decades of climate warming do to seafloor ecosystems, and what might happen if that warming were somehow undone.

In August 2021, the team collected three sediment cores from each of sixteen sites, eight per bay, using a gravity corer. One core from each site was analyzed immediately to establish a baseline, designated T0. A second core was returned to its home bay as a resident control, while the third was translocated to a depth-matched site in the opposite bay, sealed in a specially designed incubation cage, and left in place for approximately one year. The cores were open at the top, allowing exchange with the overlying seawater, but their walls prevented horizontal mixing with surrounding sediment. In September 2022, fifteen of the sixteen cages were recovered, yielding thirty cores for the endpoint analysis. The researchers then sliced the cores in the field, measuring porewater geochemistry, methane concentrations, and bacterial diversity through 16S rRNA gene amplicon sequencing, which produced more than 21 million reads across the experiment.

The geochemical results revealed how profoundly temperature shapes the vertical architecture of coastal sediments. In the heated bay, the biogeochemical zones where different electron acceptors are consumed were compressed and shallowed, a pattern consistent with earlier work at the site. After the year of translocation, linear mixed-effect modeling identified a significant three-way interaction among treatment, bay, and depth for the first principal component of geochemical variation, indicating that the direction and magnitude of change depended on where a core ended up and how deep within it one looked. Cores moved from the control bay into the heated bay developed shallower geochemical zones that began to resemble those of the heated bay’s resident sediments. Cores moved in the opposite direction deepened their profiles, but the convergence was only partial in both cases, and one year proved insufficient for complete equilibration with local conditions.

Methane and sulfate profiles added a dynamic dimension to the story. Before translocation, the heated bay sediments held methane closer to the seafloor than the control bay, where elevated concentrations appeared only at greater depth, reflecting a deeper sulfate-methane transition zone, the horizon where sulfate-consuming microorganisms deplete methane before it can escape. After a year in the control bay, translocated sediments from the heated bay showed reduced methane at shallow depths, likely because cooler temperatures slowed sulfate consumption, allowing sulfate to penetrate deeper and push methane oxidation downward. Conversely, control-bay cores transplanted into the heated bay accumulated more methane at depth, consistent with a shallowing transition zone driven by intensified sulfate reduction at higher temperatures. The researchers caution that the sealed base of the cores may have interrupted upward methane diffusion, so some of these asymmetries could partly reflect methodological effects, although comparisons of resident cores over time argued against enclosure artifacts dominating the results.

The bacterial data told a parallel tale with a twist of persistence. Before translocation, heated-bay sediments were significantly more diverse than control-bay sediments, with the difference fading at depth where warming penetrates less. Sequencing yielded 18,309 amplicon sequence variants at the baseline and 37,926 a year later, and rarefaction curves confirmed that sequencing depth captured the majority of diversity. After twelve months, both resident and translocated cores in the heated bay maintained diversity comparable to the long-term warmed community, and crucially, cores translocated from the heated bay back into the control bay retained their elevated diversity rather than reverting. One year of heating was enough to raise diversity in previously cool sediments, but one year of cooling was not enough to reverse it, a textbook example of an asymmetric response with what the authors describe as legacy effects persisting in the sediment.

Community composition, assessed through principal coordinate analysis of robust Aitchison distances and permutational tests, reinforced this picture. Bay, depth, and translocation treatment all significantly structured the communities, and shallow assemblages converged more readily toward their new environment, while deeper communities, at or below the sulfate-methane transition zone, retained signatures of their original home even after a full year. Redundancy analysis identified temperature, depth, phosphate, and organic carbon as significant environmental correlates of community structure. Family-level analysis sharpened the biological interpretation: anaerobic specialists such as the candidate phylum Atribacteria family 34-128, Desulfocapsaceae, and UBA4823 dominated the long-warmed sediments, whereas the control bay favored aerobic and facultative taxa including Cyanobiaceae, Flavobacteriaceae, Ilumatobacteraceae, and Thiobacillaceae. Tellingly, the three most abundant heated-bay families lost their dominance after a year in the control bay, yet several control-bay families maintained their abundances even after residence in the heated bay, hinting at partial resistance to warming in some lineages and delayed recovery in others.

The authors are careful about the limits of their design. The cores experienced a temperature shock when moved, larger than most natural fluctuations and even marine heatwaves, although the translocation occurred in summer when the temperature contrast between bays was at its minimum. The enclosures prevented horizontal exchange of microbes and electron acceptors, which could in principle exaggerate the translocation effect or, conversely, underestimate the communities’ capacity to converge. Because the study relied on total DNA rather than RNA, the sequence data may include remnant cells from inactive organisms, potentially overstating active community responses. Distinguishing among slow recovery, non-linear responses, transient shifts, and genuine conversion to an alternative stable state would require longer experiments with intermediate sampling. Still, the authors note that enclosure effects alone cannot explain why responses differed depending on the direction of translocation or why individual bacterial families responded so differently, which strengthens the case that the asymmetry is a real biological phenomenon.

The broader implications ripple outward from a quiet corner of the Baltic. Coastal sediments are hotspots of carbon, nitrogen, and sulfur cycling, and their microbial engineers sit at the base of processes that regulate oxygen consumption, nutrient regeneration, and even methane emissions to the atmosphere. The observation that warming compresses these biogeochemical zones, bringing anaerobic metabolism and methane production closer to the seafloor, suggests that continued warming could accelerate the delivery of reduced gases and nutrients to coastal waters, compounding the eutrophication and hypoxia that already plague the Baltic. More unsettling is the implication for reversal: if humanity were to cool the climate, these sediments would not simply snap back to their former state on human timescales. A partial return appears possible, but the experiment indicates that recovery, if it occurs at all, would take well beyond a single year in the natural environment, a reminder that some ecological debts incurred by warming are paid back far more slowly than they were borrowed.

Subject of Research: Reversibility of warming impacts on coastal sediment bacterial communities and biogeochemistry in the Baltic Sea

Article Title: Baltic Sea sediment bacterial communities show asymmetric and incomplete recovery after reciprocal temperature change translocation

Article References: Baltic Sea sediment bacterial communities show asymmetric and incomplete recovery after reciprocal temperature change translocation. (n.d.). https://doi.org/10.1038/s43247-026-04147-x

Image Credits: AI Generated

DOI: 10.1038/s43247-026-04147-x

Keywords: Baltic Sea, sediment bacteria, climate change, warming, biogeochemistry, methane, sulfate, 16S rRNA, microbial ecology, translocation experiment, legacy effects, coastal ecosystems