Substantial trace metal input from the 2022 Hunga Tonga-Hunga Haʼapai eruption into the South Pacific

This study investigates how the January 2022 Hunga Tonga–Hunga Ha'apai (HTHH) eruption affected the biogeochemistry of the oligotrophic South Pacific Gyre (SPG). Volcanic ash deposited to surface oceans can rapidly release bioactive trace metals such as iron (Fe) and manganese (Mn), potentially fertilizing phytoplankton directly in macronutrient-replete waters or indirectly by stimulating nitrogen fixation where nitrate is depleted and phosphate is available. Because the Pacific is surrounded by the volcanically active Ring of Fire, airborne volcanic ash has been proposed as a major Fe source to its surface waters, with estimated ash inputs comparable to mineral dust. The SPG is characterized by extremely low nitrate, relatively high phosphate, and likely Fe limitation of N2 fixation. The HTHH eruption (VEI ~6) injected vast quantities of ash and pumice over a broad area, with rapid sinking of ash but prolonged rafting of buoyant pumice. Tracing volcanic inputs is challenging given other external sources (continental dust, island weathering, hydrothermal inputs). Radiogenic neodymium isotopes (εNd) and dissolved rare earth element (REE) patterns provide source fingerprints. The authors therefore combine ash dispersion/trajectory modeling with surface seawater εNd, REEs, trace metals (Al, Mn, Fe), and chlorophyll-a to test the hypothesis that the HTHH eruption delivered trace metals to the SPG surface ocean, measurably altered εNd–REE signatures, and potentially stimulated a biological response.

Prior work shows that subduction-zone volcanic ash can be an important atmospheric source of soluble Fe to the Pacific, with ash deposition comparable to continental dust inputs and contributing 3–75 × 10^6 mol Fe per year to the Pacific. Pelagic clays of the South Pacific often contain >50% dispersed ash, recording episodes of Southern Hemisphere volcanism. The SPG exhibits oligotrophic conditions with low nitrate, relatively elevated phosphate, and likely Fe limitation of N2 fixation. Volcanic events (e.g., Eyjafjallajökull 2010; Kasatochi 2008) have demonstrated both natural Fe fertilization and biological responses, though the magnitude varies with eruption size and ash load. Long-distance transport of ash by prevailing winds can spread impacts thousands of kilometers; floating pumice can disperse over months to years via surface currents, further extending biogeochemical effects. εNd and REE patterns in seawater are established tools to trace lithogenic sources (dust, island weathering, volcanic inputs) because endmember εNd values vary with lithology and age, and PAAS-normalized REE patterns (HREE/LREE enrichment, Ce anomalies) distinguish seawater signatures from fresh lithogenic inputs. Baseline surface εNd in the region from prior studies showed less radiogenic values than those observed after the HTHH event, providing a comparative framework for detecting volcanic signals.

- Cruise and study region: GEOTRACES GP21 (RV SONNE SO289) transected 23 Feb–4 Apr 2022 (39–79 days post-eruption) along 26–32.5°S from Chile to New Caledonia, crossing 170°E–90°W and the southern limb of the SPG. - Atmospheric modeling: The NOAA HYSPLIT volcanic ash dispersion model was run for 72 h from 15 Jan 2022, 09:00 UTC, with three altitude layers (6, 12, 18 km) to map depositional mass loading. Forward air mass trajectories (maximum 315 h) at multiple altitudes were also computed to assess eastward transport across the SPG. - Sampling: 28 surface seawater samples for εNd were collected (uppermost 3–5 m). Stations 1–14 used 10 L Niskins on a stainless steel CTD rosette (20 L per station); from station 16 onward, a tow-fish trace-metal-clean system collected ~40 L per sample. Between stations 39–44, additional 1–2 L surface samples from the tow-fish were taken for REEs. Floating pumice was observed and collected near the Tonga–Kermadec Ridge on 30 March 2022. - Sample processing for εNd and REEs: Large-volume samples were filtered (0.45 µm), acidified (pH 1.9), spiked with FeCl3 and co-precipitated as Fe hydroxides after pH adjustment (7.8–8.2), allowed to settle, then transferred for shore-based analysis. Nd was purified via cation exchange (AG 50W-X8) and REE separation (Eichrom LN-Spec). εNd was measured on a Neptune Plus MC-ICP-MS with mass-bias corrections and external normalization (JNdi-1). Procedural blanks for Nd were <20 pg, contributing ≤0.3%. - REE concentrations: Samples were preconcentrated using a SeaFAST M5 system with NOBIAS PA-1 resin and measured by Element XR ICP-MS with desolvating nebulizer. External reproducibility (2 SD) for surface REEs was <10% for most REEs (Ce 14.9%). Calculated PAAS-normalized metrics included Ce anomaly (Ce/Ce*PAAS = 2×CePAAS/(LaPAAS+PrPAAS)) and HREE/LREEPAAS ((YbPAAS+LuPAAS)/(PrPAAS+NdPAAS)). Eu anomalies (2×EuPAAS/(GdPAAS+SmPAAS)) were also computed for vertical profiles. - Pumice analyses: Pumice was rinsed, dried, homogenized, digested by alkaline fusion, and εNd measured with the same protocol; REEs measured by quadrupole ICP-MS. Reference materials verified accuracy. - Trace metals: Surface dissolved Mn and Fe were preconcentrated online (SeaFAST SC-4 DX) and measured by Element XR ICP-MS; accuracy verified via standard addition and certified reference materials (WMAPE <5.5% for dFe, <11% for dMn). Detection limits: dMn 0.005 nmol/L, dFe 0.05 nmol/L. Dissolved Al was measured shipboard by lumogallion fluorimetry with calibration and blanks correction; replicates provided 1 SD errors. - Chlorophyll-a: Total chlorophyll-a (TChl-a) in the euphotic layer was quantified by HPLC from depth-resolved sampling (PAR-based) and integrated via trapezoidal method. - Helium isotopes: 3He/4He measurements were made at the University of Bremen to assess potential hydrothermal contributions; excess 3He was calculated after background corrections. - Data analysis: Along-transect maps and cross-plots examined relationships between εNd, REE ratios, trace metals, and TChl-a. Comparisons with historical εNd data at nearby stations contextualized anomalies. Budgets estimated volcanic Nd and Fe releases using pumice Nd content (4.18 µg/g), total ejecta mass (2.9 × 10^15 g), modeled depositional area bounds (5.4 × 10^6 to 1.2 × 10^7 km^2), mixed-layer depth (~50 m), and literature Fe solubility from ash (200 ± 50 nmol Fe/g).

- Modeled ash deposition: HYSPLIT indicated dominant ash deposition south of HTHH in the western SPG (20°–50°S, 175°E–150°W) within 3 days post-eruption; forward trajectories showed eastward transport overlapping the cruise track. - εNd and REEs: Surface εNd varied from −0.9 at the Chilean coast to −6.2 at ~90°W, then increased to +0.9 near the Tonga–Kermadec Arc. Western SPG εNd (−0 to +1) was more radiogenic than historical values nearby (e.g., −1.9, −2.3 at 170°W in 2005; −1.7 at 174°E in 2012), evidencing recent external inputs. Westward, Ce anomalies weakened and HREE/LREEPAAS increased, consistent with volcanogenic dissolution. A positive correlation existed between Ce anomaly and HREE enrichment (R² = 0.56; p < 0.001). Collected pumice had εNd = +7.5 ± 0.2. - Vertical structure: Near Tonga–Kermadec, εNd increased toward the surface within the upper 600 m, and Eu anomalies and 3He profiles did not support a dominant hydrothermal source to the surface, indicating inputs from above. - Trace metals and biomass: From east to west, surface dAl and dMn increased; strong correlation between dAl and εNd (n = 19, R² = 0.88, p < 0.001) indicates a shared volcanic aluminosilicate source. dMn correlated more weakly with εNd (n = 18, R² = 0.53, p < 0.001). In contrast, dFe reached very low concentrations (~0.1 nmol/L) in the main ash area and showed no significant correlation with εNd (n = 18, R² = 0.11, p = 0.19), consistent with rapid scavenging and/or biological uptake. TChl-a inventory in the euphotic layer showed a weak positive correlation with εNd (n = 18, R² = 0.32; p < 0.05), and satellite data indicated positive Chl-a anomalies in March 2022 in the western transect. - Rapid release and scavenging: Despite evidence for volcanogenic Nd addition (estimated 0.9–1.9 pmol/kg needed to shift εNd to −0 to +1), surface Nd concentrations remained ~4 pmol/kg due to rapid particle scavenging (preferential LREE removal), implying short mixed-layer REE residence times post-deposition. - Budgets and significance: Estimated volcanic Nd release to the SPG mixed layer is 0.4–1.6 × 10^8 g (up to 0.16 kt), comparable to the annual global dust-borne Nd input. Potential Fe release from the eruption corresponds to ~32 kt, comparable to the annual dust-borne Fe flux to the entire SPG. The geochemical signal spatially overlaps modeled ash deposition and observed pumice locations, with likely eastward extension via atmospheric transport and surface currents.

The observations demonstrate that the HTHH eruption delivered volcanogenic material to the SPG surface ocean, imprinting a distinct radiogenic εNd signal, weakened Ce anomaly, and enhanced HREE/LREE patterns in the western gyre. These source-sensitive tracers enable discrimination of volcanic inputs from continental dust and island weathering. The strong coupling between εNd and dAl indicates volcanic aluminosilicate dissolution as the dominant source during the cruise period. In contrast, the rapid post-deposition scavenging and biological demand for Fe likely led to depletion of surface dFe in the primary ash deposition region by 9–10 weeks post-eruption, explaining the lack of a positive εNd–dFe correlation despite substantial initial input. A weak but significant increase in euphotic TChl-a with more radiogenic εNd, together with satellite chlorophyll anomalies, suggests a biological response, plausibly mediated by Fe supplied from volcanic material (including potentially prolonged release from floating pumice) and subsequent stimulation of N2 fixation in this Fe-limited region. The progressive east–west gradients in εNd and REE ratios imply that volcanic influence extended beyond the immediate deposition area via eastward air mass transport and surface currents. Overall, the findings support the hypothesis that episodic, large volcanic eruptions can exert substantial and measurable impacts on SPG surface biogeochemistry on basin scales and weekly-to-monthly timescales.

This work provides direct geochemical evidence that the 2022 HTHH eruption delivered substantial trace metal inputs to the SPG surface ocean, measurably altering εNd and REE signatures and elevating trace metal levels (especially Al and Mn), with indications of enhanced phytoplankton biomass. Volcanic input was identified as the dominant external trace metal source in the western SPG during the observation period. Budget estimates indicate up to 0.16 kt of Nd and ~32 kt of Fe released to the SPG—amounts comparable to major annual dust-borne inputs. These results underscore the need to account for episodic volcanic sources and rapid post-deposition scavenging/biological processes in biogeochemical and climate models of the Pacific. Future research should constrain the temporal evolution of trace metal speciation and residence times after eruptions, quantify the long-term Fe release efficiency from pumice, refine depositional area and flux estimates, and disentangle volcanic signals from other regional sources using combined tracers and time-series observations.

- Temporal gap: Sampling occurred 9–10 weeks after the eruption, likely missing peak dissolved Fe concentrations and immediate biological responses in the deposition area. - Source attribution: Multiple potential sources (e.g., island weathering, hydrothermal inputs, continental dust) complicate unambiguous attribution, although tracers favor volcanic dominance during the cruise. - Rapid scavenging: Intense particle scavenging after ash deposition obscured initial dissolved Nd and Fe concentration increases, requiring indirect inference from εNd shifts and REE patterns. - Modeling and spatial coverage: HYSPLIT dispersion and trajectory simulations provide bounds but carry uncertainties; observational coverage was along a single zonal transect, limiting two-dimensional mapping of impacts. - Pumice Fe release: The efficiency and timescale of Fe release from floating pumice remain uncertain and require dedicated quantification.