Distal records of Katla’s explosive past: Ocean-rafted pumice found in archaeological contexts and raised shorelines in Norway

Ocean-rafted pumice is found on modern and paleo-beaches from the North Atlantic to the South Pacific. Correlation of ocean-rafting events to their volcanic source can help date sedimentary records and landforms, while geochemical fingerprinting of artefacts made from ocean-rafted pumice provides age constraints for pumice-bearing archaeological sites. These pumices also record large explosive eruptions that might not be preserved near the volcano due to extensive erosion or a submarine volcanic source and as such can be used for petrological and geochemical investigations. This study explores links between spatiotemporal patterns of Holocene pumice deposition along the Norwegian coastline and prehistoric human use of this versatile resource. We first focused on a suite of pumice pieces from archaeological contexts of variable isostatic uplift in Northern Norway with most showing evidence of use as abrasive tools. These samples were geochemically correlated to individual Holocene eruptions or groups of tephras (SILK) from the Katla Volcanic System in Iceland. Our age data revealed that the estimated eruption dates typically predate the contexts by several hundred and up to 2-3,000 years, probably reflecting abundance and availability of certain pumice types at the time. To investigate how distal resource availability is influenced by geological processes like eruption frequency, ocean-currents, and deposition/preservation we conducted field surveys of two coastal stretches with different climates, geomorphic settings and uplift histories. On Varanger Peninsula in Northern Norway strong Holocene uplift rates and sea-level changes have built a unique record of raised shorelines that provide windows into fossil beach ridges up to the marine limit, covered in little vegetation. We found that pumice was abundant on specific paleo-shorelines and in defined geomorphic settings but absent from older beach ridges. Most samples correlated with the <7 ka Katla record and the distinct mid-Holocene transgression high-stand accumulated the largest variety of pumice types and clast sizes. One sample cluster overlaps with SILK tephra compositions but does not correlate to any known Katla units while a single pumice plots outside the field defined by Katla eruptives, resembling compositions known for Jan Mayen. In contrast, the Trøndelag and Nordland coast in Southern Norway is heavily vegetated and pumice was only found at isolated sites that displayed the right conditions, important factors being: a) the paleo-setting of the beach (e.g., currents, orientation, morphology) favouring accumulation in the first place, b) rapid uplift and limited erosion enabling preservation of the stranded pumice and c) exposure of pumice-bearing beach ridges (sections cut by rivers/erosion, roads/construction) as subsequent burial by sediment, soil and vegetation further reduces access to previously available pumice resources. This supports our hypothesis of pumice only being readily available for limited periods of time following eruption, rafting and onshore deposition. We will further test this assumption by integrating compositional information of archaeological pumice pieces from a recent archaeological excavation in Lofoten into our data-set. Overall, our study provides a better understanding of the nature and frequency of Holocene silicic eruptions from Katla while also improving age control for existing relative sea-level curves and archaeological contexts in Norway.