Drilled Extreme Events from the Past: Unravelling a long-term history of giant earthquakes and tsunamis for geohazards assessment

Project Information

DEEP

Grant agreement ID: 101155517

DOI

10.3030/101155517

EC signature date 6 March 2024

Start date 3 June 2024

End date 2 June 2027

Funded under

• Marie Skłodowska-Curie Actions (MSCA)

Total cost

No data

EU contribution

€ 268 025,28

Coordinated by

INSTITUT FRANCAIS DE RECHERCHE POUR L'EXPLOITATION DE LA MER
France

Periodic Reporting for period 1 - DEEP (Drilled Extreme Events from the Past: Unravelling a long-term history of giant earthquakes and tsunamis for geohazards assessment)

Reporting period: 2024-06-03 to 2026-06-02

Summary of the context and overall objectives of the project

Large subduction earthquakes and the tsunamis they generate are among the most destructive natural hazards affecting coastal populations. Their recurrence times can span centuries to millennia, meaning that instrumental and historical records capture only a small part of their full behaviour. Geological archives preserved in marine sediments therefore provide a unique opportunity to reconstruct the long-term history of extreme earthquakes and tsunamis and improve our understanding of their recurrence, magnitude and spatial extent.

The DEEP project focuses on the Japan Trench, one of the world’s most active subduction zones and the source of the 2011 Mw 9.0 Tohoku-Oki earthquake and tsunami. The project builds on the exceptionally long sedimentary sequences recovered during International Ocean Discovery Program Expedition 386, which sampled more than 800 m of sediment at 15 sites in the deep Japan Trench. These sediments contain a long-term record of past disturbance events that cannot be reconstructed from historical observations alone. The project particularly investigates homogenite–turbidite (HmTu) deposits, sedimentary layers that can preserve the imprint of large earthquakes and associated tsunami processes over large distances.

DEEP aims to improve the identification and interpretation of these event deposits through an interdisciplinary, multi-proxy approach combining X-ray computed tomography (XCT), micro-XCT, sedimentological observations, grain-size analysis, physical properties, geochemical measurements and chronological constraints. The project has three interconnected objectives: (1) to establish quantitative and reproducible criteria for identifying and characterising HmTu deposits and improve understanding of their depositional processes; (2) to correlate event deposits through time and space in order to extend the long-term earthquake and tsunami catalogue of the Japan Trench; and (3) to lay the foundations for integrating offshore geological records with onshore earthquake and tsunami information in support of improved natural-hazard assessment.

The pathway to impact is based on progressively transforming geological observations into quantitative, transferable and reusable information. High-resolution imaging and multi-proxy analyses will improve the identification of earthquake-related deposits; stratigraphic and chronological correlations will help reconstruct the timing and recurrence of past events; and the resulting information will ultimately contribute to integrated geohazard databases and improved assessments of seismic and tsunami risk. In the longer term, these outputs can support researchers and organisations involved in hazard assessment and disaster-risk reduction by providing evidence from geological timescales that complements instrumental monitoring and historical records.

The project therefore addresses a fundamental scientific challenge while contributing, in the longer term, to better understanding and management of seismic and tsunami hazards in exposed coastal regions.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

During the reporting period, substantial progress was achieved towards the first two scientific objectives of DEEP.

For Objective 1, a quantitative and reproducible workflow was developed for the identification and characterisation of homogenite–turbidite deposits. A complete XCT processing workflow was implemented using Fiji, including image pre-processing, automated extraction of CT-number profiles, normalisation and statistical analysis. The approach was tested on representative core sections containing homogenites, turbidites, hemipelagic sediments and mass-transport deposits. At a larger scale, 841 XCT scans from 58 sediment cores, corresponding to approximately 832 m of sediment and 881 core sections, were processed and organised into standardised high-resolution 3D datasets and visualisation plates. This provides a consistent basis for large-scale comparative analysis and future automation.

High-resolution micro-XCT analyses were also initiated on 12 core sections from the Japan Trench, representing approximately 3.42 m of sediment. These analyses are being used to investigate grain-size distributions, three-dimensional sedimentary structures and depositional interfaces at a much finer scale than conventional XCT imaging. Additional training was undertaken to address the complexity of the datasets and optimise their analysis. The project also integrated complementary physical-property and geochemical observations, including MSCL, SCOPIX and XRF data, strengthening the multi-proxy characterisation of the sedimentary record.

For Objective 2, chronological information from radiocarbon dating, 210Pb, paleomagnetism and tephrochronology was compiled and integrated with sedimentological observations to establish a robust chronostratigraphic framework for the Japan Trench cores. A key result was the identification of a sedimentary sequence interpreted as being associated with the historically documented 869 CE Jōgan megathrust earthquake. This deposit is currently being investigated using high-resolution sedimentology, grain-size extraction from micro-XCT, three-dimensional reconstruction and integration with XCT and physical-property data. IODP Expedition 405 material and associated SCOPIX and XRF observations were also incorporated to support targeted analyses.

Preparatory work for Objective 3 was initiated through the standardisation of XCT-derived parameters and the development of reproducible workflows for large datasets. Initial machine-learning approaches for automated lithofacies classification were explored. These activities provide the methodological foundations required for the subsequent integration of offshore geological records into a broader database for paleoseismic and tsunami hazard assessment.

Overall, the project has established the main methodological basis for quantitative HmTu characterisation, generated a substantial standardised XCT dataset, developed a multi-proxy analytical framework and identified a key geological record linked to a known historical megathrust earthquake.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

DEEP advances the state of the art in submarine paleoseismology by moving the identification of earthquake-related sedimentary deposits towards a more quantitative, reproducible and multi-scale approach.

A major methodological advance is the development of a workflow that combines conventional XCT and micro-XCT imaging with sedimentological, physical-property and geochemical observations. While XCT has already demonstrated its value for visualising sediment cores, DEEP extends its application by extracting quantitative CT-number information and combining this with high-resolution micro-XCT observations and other independent proxies. This provides a stronger basis for distinguishing event deposits from background sedimentation and for characterising their internal architecture and depositional processes.

The project also contributes to overcoming an important limitation of previous studies: the difficulty of comparing large numbers of sediment cores using consistent criteria. The processing and standardisation of 841 XCT scans from 58 sediment cores demonstrates the feasibility of applying reproducible analytical procedures across a very large dataset. The development of automated image-processing workflows, standardised 3D datasets and initial machine-learning approaches creates a pathway towards more efficient and scalable lithofacies classification.

A further advance lies in linking detailed sedimentary observations to regional chronostratigraphic information. The identification of a deposit associated with the 869 CE Jōgan earthquake provides an important reference point for evaluating the sedimentary expression of a known major earthquake and for testing the transferability of event-recognition criteria to older deposits. More broadly, the project contributes to extending the geological record of earthquakes and tsunamis beyond the short timescale covered by instrumental and historical observations.

The longer-term potential impact is to transform these methodological and geological advances into an integrated source of information for marine paleoseismology and geohazard assessment. The original project envisaged the integration of offshore and onshore earthquake and tsunami records and the possible extension of the approach to other high-risk subduction zones. The work completed so far provides the scientific and technical foundations for this next stage.

Further research will be required to complete the high-resolution analyses, finalise the regional correlations and develop the integrated database and hazard-assessment component. The main requirement for further uptake is therefore continued integration of datasets, validation of the methodological framework across additional sites and transfer of the resulting workflows and data products to the wider scientific community.