Giant offshore pumice deposit records a shallow submarine explosive eruption of ancestral Santorini

The processes and impacts of submarine explosive eruptions are poorly understood compared to their terrestrial equivalents. However, submarine calderas are common on island arcs, and shallow submarine eruptions can be very violent, as shown by the 2022 Hunga Tonga-Hunga Ha'apai Volcano eruption. Pyroclastic currents from such eruptions enter the sea, entrain water, and transform into water-supported gravity flows. While resulting deposits can be studied in ancient successions, those around modern island volcanoes are difficult to access except by deep drilling. The South Aegean Volcanic Arc presents a significant hazard due to its submarine volcanoes. While the eruptive history of the arc has been investigated through onland mapping and marine tephrachronology, the record of submarine volcanism has been poorly constrained. IODP Expedition 398 drilled the marine rifts of the central island arc to depths of up to 900 m below the seafloor to investigate the Neogene-Quaternary volcanism and find deposits from past submarine eruptions. The twelve drill sites are in and around the Christiana Santorini-Kolumbo Volcanic Field (CSKVF), which includes Santorini caldera. The CSKVF is in a 100-km-long rift system with three basins containing up to 1400 m of sediments and volcanics. Santorini's activity can be divided into old (>650–550 ka; ‘Early Centres of Akrotiri’) and young (~530 ka to present day) periods. Kolumbo Volcano and its submarine cones are located NE of Santorini. Deep drilling provided a unique opportunity to generate a complete eruptive time series of the CSKVF, particularly given a 2011–12 caldera unrest period and the presence of two shallow magma reservoirs. The discovery of the submarine pumice deposit discussed in this paper resulted from a combination of IODP deep drilling, shipboard datasets, laboratory analysis, and marine seismic profiles.

Previous research on the South Aegean Volcanic Arc focused primarily on onland geological mapping and tephrochronology, providing a good understanding of the subaerial eruptive history of Santorini and other volcanoes in the arc. However, the knowledge of submarine volcanism remained limited, mostly inferred from offshore seismic imagery and limited sampling. Studies on submarine explosive eruptions have explored the processes of magma-water interaction, the formation of pyroclastic currents and their transformation into submarine flows, and the characteristics of the resulting deposits. While some studies have focused on specific events and locations, a comprehensive understanding of the magnitude, frequency, and impact of submarine explosive eruptions, particularly in complex volcanic arcs like the South Aegean, was lacking. This paper builds upon previous work by providing direct evidence from deep-sea drilling, offering a much more complete record of submarine volcanism in this region, filling a critical gap in our understanding of this hazardous geological process.

IODP Expedition 398 drilled twelve sites in and around the Christiana-Santorini-Kolumbo Volcanic Field (CSKVF). The Archaeos Tuff was sampled at seven of these sites with core recoveries ranging from <1 to 88%. The deposit's thickness varied across sites, being thickest in the Christiana Basin and north of Santorini, and thinnest at the distal end of the Anhydros Basin. The deposit is composed of pumice and ash with lesser lithic components. Grain-size analysis was challenging due to drilling disturbance, but samples showing minimal disturbance had median diameters of −1.9 to 3.3 phi, Inman sorting coefficients of 1.4 to 2.9 phi, and <20% sub-63 µm ash. Chemical analyses of high-silica rhyolite glasses from pumice samples showed identical major element, trace element, and incompatible trace element ratios, distinct from products of Christiana, Kolumbo, other volcanic fields, and young Santorini. They are most similar to the Early Akrotiri centres (>650–550 ka) of Santorini. Biostratigraphic constraints using foraminifer and calcareous nannofossil assemblages in sediment layers above the Archaeos Tuff constrained the eruption age to 520 ± 10 ka. Palaeowater depths were inferred from benthic foraminifer assemblages, ranging from 200–1000 m. Seismic stratigraphy and core-seismic correlation allowed mapping of the deposit, yielding an observed bulk volume of 89 ± 8 km³. Shipboard density measurements allowed conversion to a Dense Rock Equivalent (DRE) volume of 30 ± 3 km³. Onland outcrops of a geochemically similar rhyolitic tuff on Christiani, Santorini, and Anafi islands were correlated with the submarine Archaeos Tuff based on chemistry, mineralogy, and pumice characteristics.

The study discovered the Archaeos Tuff, a giant rhyolitic pumice deposit emplaced 520 ± 10 ka ago at water depths of 200–1000 m. This deposit resulted from a high-intensity, shallow submarine eruption. The eruption generated pyroclastic currents that transformed into turbidity currents and slurries, forming a >89 ± 8 km³ (30 ± 3 km³ DRE) volcaniclastic megaturbidite. The eruption column also breached the sea surface, depositing ignimbrite veneers on three islands. The pumice clasts are highly vesicular, indicating magma fragmentation driven mainly by exsolution of magmatic gases. The uniform melt chemistry, distinctive mineral assemblage, and lack of depositional breaks suggest a single volcanic event. The Archaeos Tuff's characteristics – great thickness, thickening into rift basins, and locally erosional base – indicate emplacement by gravity flows. The moderate to good sorting of the deposit is consistent with transport in water-supported gravity flows. The study concludes that the eruption likely originated from shallow submarine vents, with the upper part of the eruption column breaching the sea surface to produce subaerial pyroclastic currents and ignimbrite deposits on nearby islands. The Archaeos Tuff's chemical composition is most similar to that of the Early Akrotiri centres of Santorini, suggesting the eruption culminated the development of the submarine Akrotiri complex. The volume of the Archaeos Tuff makes it the largest pyroclastic-current-derived deposit of the CSKVF, possibly comparable in size to the Kos Plateau Tuff.

The findings significantly advance our understanding of submarine explosive volcanism in the South Aegean Volcanic Arc. The discovery of the Archaeos Tuff demonstrates the arc's capacity for highly hazardous submarine eruptions, exceeding previously recognized magnitudes. The large volume and extensive distribution of the deposit highlight the significant potential hazards posed by such events, including the generation of tsunamis and impacts on marine ecosystems. The age of the eruption lies near the transition between the Akrotiri and younger Santorini magmatic periods, suggesting crustal stress changes may have triggered the eruption by tapping new magma batches. The under-representation of the Archaeos Tuff in the subaerial geological record emphasizes the importance of deep-sea drilling for understanding the complete eruptive history of island arcs.

This study documented the offshore and onshore deposits of a large, shallow submarine explosive eruption, well-constrained by volume, age, bathymetry, field, and geochemical data. The pyroclastic currents were larger than those of the 2022 Hunga Tonga-Hunga Ha'apai eruption. The findings highlight the importance of deep drilling in densely populated regions like the Mediterranean to unravel the full secrets of island arcs and assess the hazards from submarine volcanism. Future research should focus on precisely locating the source caldera and further investigating the eruption dynamics and associated hazards.

While the study provides a comprehensive analysis of the Archaeos Tuff, some limitations exist. The grain-size analysis was affected by drilling disturbances, necessitating careful sample selection. The total volume of the eruption is a minimum estimate because distal deposits and other eruption products outside the study area are not included. The exact location of the source caldera remains unclear.