A 1500-year record of mercury isotopes in seal feces documents sea ice changes in the Antarctic

Mercury (Hg) is a global pollutant capable of long-range atmospheric transport to remote regions, including Antarctica, where elevated Hg levels have been observed. After deposition to the ocean, part of Hg is reduced and re-emitted, while a fraction is converted to toxic, bioavailable methylmercury (MeHg), which bioaccumulates in marine food webs and threatens top predators. In surface oceans, MeHg is photodegraded by sunlight, making photodemethylation a key sink for MeHg. Prior work indicates sea ice extent in the Arctic modulates solar radiation penetration, altering MeHg photodegradation and imprinting characteristic mass-independent Hg isotope fractionation (MIF; Δ199Hg, Δ201Hg) in marine biota. Hg isotopes are thus a promising proxy for reconstructing sea-ice-driven changes in MeHg cycling. The Antarctic Peninsula has experienced rapid climate variability with marked changes in glaciers and sea ice, potentially affecting regional Hg cycling. Ice-free coastal zones host abundant seals and penguins whose biological sediments archive past environmental conditions. This study analyzes Hg isotope compositions from a seal-affected sediment core (HF4) from Fildes Peninsula, King George Island, to identify Hg sources and reconstruct historical variability in MeHg photodemethylation and its linkage to sea ice. The authors find that after ~470 CE, Hg input is dominated by seal feces and that Δ199Hg variations after ~750 CE record changes in ocean photodemethylation strongly controlled by sea ice.

- Photochemical reactions impart odd-mass Hg MIF (Δ199Hg, Δ201Hg), whereas most other processes yield mass-dependent fractionation (MDF, δ202Hg). Aquatic photoreduction of Hg(II) and photodemethylation of MeHg enrich odd Hg isotopes in the residual species. These MIF signals are conserved during trophic transfer, allowing biological tissues to record pre-food-web photochemistry. - Arctic studies using seabird eggs and ringed seal tissues showed sea ice extent modulates MeHg photodegradation, reflected by Δ199Hg increases with reduced ice cover. Prior Antarctic work on ornithogenic deposits (Ross Island) suggested enhanced photodemethylation during the Little Ice Age due to polynya expansion. - Regional climate of the Antarctic Peninsula exhibits warm (e.g., Medieval Climate Anomaly, MCA) and cold intervals (e.g., Little Ice Age, LIA), alongside significant sea ice and glacial variability. Biological sediments from predators provide unique archives for reconstructing past environmental and Hg cycle changes. - Together, these studies underpin the hypothesis that Hg MIF in seal-feces-dominated sediments can serve as a proxy for historical sea ice variability via its control on MeHg photodemethylation.

Site and sampling: A 42.5 cm sediment core (HF4) was taken during the 18th Chinese Antarctic Research Expedition (Nov 2001–Mar 2002) from a depositional basin (62°11′57″S, 59°58′48″W; 8 m a.s.l.) on the west coast of Fildes Peninsula, King George Island. A 12 cm diameter PVC gravity corer was driven to bedrock and retrieved. The core was sectioned at 0.5 cm intervals for the upper 18 cm and 1.0 cm thereafter. Seal hair was abundant in the top 18 cm, sparse from 18–23 cm, and absent below 23 cm. Subsamples were air-dried, homogenized, and a portion was dried at 105 °C for 24 h for seal hair counting. Chronology: 137Cs and conventional 14C dating were applied. To minimize marine reservoir effects, four samples with minimal seal hair were used. A linear extrapolation of 14C age vs depth yielded a modern surface intercept of 114 ± 4 years (validated by 137Cs), adopted as the reservoir correction and subtracted from measured 14C ages. Core chronology was established by linear interpolation between corrected ages. Hg concentration and extraction: Hg concentrations followed prior protocols. For isotope analysis, Hg was extracted by thermal combustion of powdered samples (<2 g) at 800 °C in a Hydra IIc Direct Mercury Analyzer under O2 flow (250 mL min−1). Evolved Hg0 was trapped in 0.2% KMnO4 (w/w) + 5% H2SO4 (v/v), oxidizing to Hg(II). Blanks, and standard reference material CCRMP TILL-1, were processed in parallel. Recoveries: TILL-1, 96.1 ± 2% (1 SD, n=3); samples, 94.4 ± 8% (1 SD, n=24). Procedural blanks <0.04 ng g−1 (<1% of sample Hg). Isotope measurements: Hg isotopes were measured by MC-ICP-MS (Thermo Neptune Plus, University of Toronto) with online cold vapor generation. Traps were neutralized with NH2OH·HCl and diluted to 1–2 ng g−1 Hg in matrix-matched solution. Instrumental mass bias was corrected using 205Tl/203Tl (NIST 997) and strict standard-sample-standard bracketing with NIST SRM 3133 matched in matrix and signal intensity (±10%). 204Pb interference was corrected using 206Pb and was negligible. On-peak zero from a matrix-matched blank was applied. Isotopic notation: MDF as δ202Hg; MIF as ΔxHg = δxHg − βx·δ202Hg with βx = 0.2520, 0.5024, 0.7520, 1.4930 for 199, 200, 201, 204, respectively. Each sample was run in duplicate; an in-house JTBaker Hg standard was run 5–7 times per session. Uncertainties (2σ) reported as the larger of 2SE of sample replicates or 2SD of JTBaker. JTBaker results: δ202Hg = −0.63 ± 0.05‰; δ199Hg = 0.02 ± 0.02‰; δ200Hg = 0.01 ± 0.03‰ (n=12). TILL-1: δ202Hg = −1.11 ± 0.01‰; δ199Hg = −0.11 ± 0.01‰; Δ200Hg = 0.01 ± 0.01‰ (n=3), consistent with published values. Data sources for context: Sea ice data from Hadley Centre, SAM index from NOAA NCEI. Analytical focus was on δ202Hg and Δ199Hg, as Δ200Hg was near zero throughout.

- Strong seal input and Hg accumulation: Seal hair counts averaged 1359 ± 1059 g−1 (1 SD, n=38) above 23 cm, indicating major biological input. Hg concentrations correlate with seal hair counts (Pearson r = 0.66, two-sided t-test p < 0.001, n = 38). Mean sediment Hg increased from 20.2 ± 6.0 ng g−1 (1 SD, n=14) before ~470 CE to 203.0 ± 80.4 ng g−1 (1 SD, n=34) after ~470 CE, far exceeding local soil background (~<13 ng g−1) and atmospheric-deposition-dominated sediments (~2.5 ng g−1). - Source apportionment via Hg isotopes: Period I (before ~470 CE) exhibits low δ202Hg (−0.76 ± 0.19‰, n=10) and Δ199Hg (0.12 ± 0.11‰), consistent with geogenic/background atmospheric inputs and minimal biological input. Period II (after ~470 CE) shows markedly higher δ202Hg (~1.79 ± 0.24‰, n=14) and Δ199Hg (~1.75 ± 0.19‰), distinct from atmospheric aerosols, snow, volcanic, vegetation, and soil/bedrock sources (all with near-zero or negative Δ199Hg; two-sided t-tests p < 0.01), and aligning with seal-feces-dominated sediments and modern seal feces. This indicates Hg is dominantly from seal feces in Period II. - Δ199Hg variability decoupled from seal abundance after ~750 CE: While Hg content and Δ199Hg rose with seal population from ~470–~750 CE, thereafter (∼750–1300 CE) Δ199Hg peaked when seal counts declined, and post-1300 CE Δ199Hg varied little despite population fluctuations. Thus Δ199Hg changes reflect the isotopic signature of seal diet/MeHg at the food-web base rather than varying fecal input fractions. - Mechanism: The Δ199Hg–Δ201Hg slope in sediments is ~1.17, close to values characteristic of MeHg photodemethylation in seawater (~1.2), and higher than that of Hg(II) photoreduction (~1.0). Open-ocean surface seawater Δ199Hg (<~0.4‰) is too low to explain observed values (mean ~1.81‰ after ~750 CE). Given MeHg dominance in seal prey and conservation of MIF during trophic transfer, odd-MIF in sediments primarily records seawater MeHg photodemethylation. - Sea ice control: Elevated Δ199Hg during the Medieval Climate Anomaly (~1000–1300 CE) indicates enhanced photodemethylation due to reduced sea ice and increased light penetration. Lower Δ199Hg during colder intervals (~750–1000 CE; ~1300–1750 CE) implies inhibited photodemethylation under greater ice cover. Alternative drivers (increased snow/ice melt with negative Δ199Hg inputs, stronger upwelling delivering low-Δ199Hg MeHg from depth, or shifts in atmospheric deposition) are inconsistent with the observed increases. - Magnitude: A Δ199Hg difference of ~0.39‰ between warm and cold periods corresponds, by analogy to Arctic seabird egg data, to roughly a 52% change in sea ice coverage and about a 5% change in photodemethylation extent between periods. - Regional comparison: Timing of Δ199Hg increases differs from Ross Island records influenced by polynya dynamics and katabatic winds, underscoring regional controls on sea ice and MeHg cycling. - Modern trend: A weak increase in Δ199Hg toward recent times suggests ongoing amplification of photodemethylation with warming/ice loss, though core resolution is insufficient to robustly resolve modern changes.

The study shows that Hg mass-independent isotope signals preserved in seal-feces-dominated sediments track historical changes in seawater MeHg photodemethylation modulated by sea ice. After establishing that post-~470 CE Hg in the HF4 core derives predominantly from seal feces, the authors interpret Δ199Hg variability as reflecting the isotopic composition of MeHg at the base of the food web. The Δ199Hg–Δ201Hg relationship, prey MeHg dominance, and open-ocean Δ199Hg constraints collectively indicate that MeHg photodemethylation, rather than Hg(II) photoreduction or dietary depth shifts, is the principal control. Elevated Δ199Hg during the MCA aligns with reduced sea ice and greater irradiance, while suppressed Δ199Hg during colder intervals reflects increased sea ice and diminished photodegradation. Alternative mechanisms (snow/ice melt with strongly negative Δ199Hg, enhanced upwelling of low-Δ199Hg MeHg, or major atmospheric deposition shifts) are inconsistent with the data and climate proxies (e.g., SAM index, diatom sea-ice taxa, TOC). Comparison with Ross Sea records highlights the role of regional ocean-atmosphere dynamics (katabatic winds, polynyas) in mediating photochemical MeHg cycling. Overall, the findings validate Hg isotopes in biological sediments as sensitive proxies for Antarctic sea ice variability and provide insight into the sensitivity of marine MeHg cycling to cryospheric change.

Hg isotopes from a 1.5-kyr seal-affected sediment core on King George Island reveal that post-~470 CE Hg inputs are dominated by seal feces and that Δ199Hg variations primarily record seawater MeHg photodemethylation controlled by sea ice. Higher Δ199Hg during the MCA indicates enhanced photodemethylation under reduced sea ice, while lower values during colder periods reflect inhibited photodegradation with greater sea ice cover. These results demonstrate that biological sediments can serve as valuable archives of past sea ice and marine MeHg cycling in Antarctica. Future work should target higher-temporal-resolution biological sediment cores and modern biological samples across regions to better resolve recent trends, quantify spatial heterogeneity, and clarify how changes in photodemethylation propagate to Hg concentrations and risks in marine organisms considering co-varying factors (productivity, trophic structure, and ocean-atmosphere processes).

- Temporal resolution: The HF4 core resolution is insufficient to robustly resolve modern (recent decades) changes in Δ199Hg and sea ice. - Proxy-to-process linkage: While Δ199Hg tracks photodemethylation, the study cannot determine how changes in photodemethylation quantitatively affect Hg burdens in organisms due to confounding ecological and biogeochemical factors. - Regional and individual variability: Comparisons to modern seal feces show variability linked to sampling locations and individual seals; regional environmental differences (e.g., sea ice, upwelling, polynyas) can modulate signals. - Assumptions: Interpretation assumes negligible post-depositional alteration of Hg isotopes and that trophic transfer does not alter MIF; although supported by prior work, minor unrecognized processes cannot be entirely excluded.