Sedimentological storm and tsunami record of Loch Flugarth, Shetland Islands (UK)

Abstract

This dataset comprises the geochemical and sedimentological results of both publications Hess et al., 2023 and Engel et al., 2023. Hess et al., 2023: Severe storm flooding poses a major hazard to the coasts of north-western Europe. However, the long-term recurrence patterns of extreme coastal flooding and their governing factors are poorly understood. Therefore, high-resolution sedimentary records of past North Atlantic storm flooding are required. This multi-proxy study reconstructs storm-induced overwash processes from coastal lake sediments on the Shetland Islands using grain-size and geochemical data, and the re-analysis of historical data. The chronostratigraphy is based on Bayesian age–depth modelling using accelerator mass spectrometry 14C and 137Cs data. A high XRF-based Si/Ti ratio and the unimodal grain-size distribution link the sand layers to the beach and thus storm-induced overwash events. Periods with more frequent storm flooding occurred 980–1050, 1150–1300, 1450–1550, 1820–1900 and 1950–2000 ce, which is largely consistent with a positive North Atlantic Oscillation mode. The Little Ice Age (1400–1850 ce) shows a gap of major sand layers suggesting a southward shift of storm tracks and a seasonal variance with more storm floods in spring and autumn. Warmer phases shifted winter storm tracks towards the north-east Atlantic, indicating a possible trend for future storm-track changes and increased storm flooding in the northern North Sea region. Engel et al., 2023: Tsunami deposits around the North Sea basin are needed to assess the long-term hazard of tsunamis. Here, we present sedimentary evidence of the youngest tsunami on the Shetland Islands from Loch Flugarth, a coastal lake on northern Mainland. Three gravity cores show organic-rich background sedimentation with many sub-centimetre-scale sand layers, reflecting recurring storm overwash and a sediment source limited to the active beach and uppermost subtidal zone. A basal 13-cm-thick sand layer, dated to 426–787 cal. a CE based on 14C, 137Cs and Bayesian age–depth modelling, was found in all cores. High-resolution grain-size analysis identified four normally graded or massive sublayers with inversely graded traction carpets at the base of two sublayers. A thin organic-rich ‘mud’ drape and a ‘mud’ cap cover the two uppermost sublayers, which also contain small rip-up clasts. Grain-size distributions show a difference between the basal sand layer and the coarser and better sorted storm layers above. Multivariate statistical analysis of X-ray fluorescence core scanning data also distinguishes both sand units: Zr, Fe and Ti dominate the thick basal sand, while the thin storm layers are high in K and Si. Enriched Zr and Ti in the basal sand layer, in combination with increased magnetic susceptibility, may be related to higher heavy mineral content reflecting an additional marine sediment source below the storm-wave base that is activated by a tsunami. Based on reinterpretation of chronological data from two different published sites and the chronostratigraphy of the present study, the tsunami seems to date to c. 1400 cal. a BP. Although the source of the tsunami remains unclear, the lack of evidence for this event outside of the Shetland Islands suggests that it had a local source and was smaller than the older Storegga tsunami (8.15 cal. ka BP), which affected most of the North Sea basin.