Revised Storegga Slide reconstruction reveals two major submarine landslides 12,000 years apart

The 1929 Grand Banks earthquake triggered the most recent large-scale landslide on open continental slopes, incorporating 200 km3 of seafloor sediments and causing a tsunami with up to 13 m run-up on Newfoundland. Morphological analyses of North Atlantic continental margins reveal dozens of late-Quaternary slide scars of even greater dimensions. With an estimated volume of 2400–3200 km3, the Storegga Slide is the largest known continental margin slope failure, extending ~300 km along the mid-Norwegian shelf-break and >800 km into the Norway Basin. Prior work concluded Storegga developed in multiple phases: an initial phase starting >100 km downslope and regressing upslope, removing up to 50 m of poorly consolidated glacial sediments from the northern half of the scar and depositing debris flows and turbidity currents far into the basin; a second phase of lateral spreading along a well-defined glide plane with minor local failures; and a final failure of the central segment incising several hundred meters. Seismically constrained reconstructions suggested the first phase mobilized ~1300–1600 km3 (about 50% of total), lateral spreading and minor failures added ~600–900 km3, and the central failure involved ~500–800 km3, considered the main tsunami source. Radiocarbon dating indicates Storegga failed ~8150 years BP, coincident with tsunami deposits showing >20 m run-up on the Shetland Islands and impacts as far as Denmark, Greenland, and the Barents Sea. Despite intensive study, we present new geophysical and sedimentological data from the north-eastern region showing this part of the scar records a previously unknown large-scale slope failure at the end of the Last Glacial Maximum (~20,000 BP), termed the Nyegga Slide. This finding has major implications for understanding Storegga and for geohazard assessment of continental margin slope failures.

Previous studies established Storegga as a multi-phase mega-slide, with initial evacuation of up to 50 m of glacial sediments in the northern scar, followed by lateral spreading and a deeply incised central failure. Tsunami deposits and modeling linked the ~8150 BP central segment failure to a basin-wide tsunami. Numerous works on preconditioning processes highlight roles for gas hydrate dissociation, rapid glacial sediment loading causing elevated pore pressures, and ice-sheet dynamics (tilting and excess pore pressure from unloading). Analogous large slides in the region (e.g., Trænadjupet, Hinlopen) provide context for expected volumes and behaviors. Earlier reconstructions associated essentially the entire present-day NE slide scar evacuation with the ~8150 BP event, inferring removal of all sediments younger than ~30 ka, but lacked datable material beneath key horizons to test this assumption.

Geophysical data: Sub-bottom profiler (Teledyne Parasound) data were acquired during RV Meteor cruise M87/2 (2012). Bathymetry is a compilation of regional grids (Seabed project, Norsk Hydro/Equinor), high-resolution Mareano data (Norwegian Mapping Authority Hydrographic Service), and 3D seismic seafloor picks. Interpretation used IHS Kingdom Suite and Schlumberger Petrel. Seismostratigraphic framework: Key reflections R1–R8 were tied to a high-resolution chronostratigraphy from multiple sediment cores on the southern Vøring Plateau. R8 marks the Storegga Slide glide plane; R1–R7 span late-glacial to Holocene units. Profiles were flattened on R8 to compare intact sidewall and slide scar stratigraphy and quantify evacuated thickness. Identification of mass-transport deposits (MTDs) relied on acoustically transparent bodies bounded by stratified reflectors and correlation to dated horizons. Sediment cores: 89 cores from prior Seabed/Ormen Lange projects provided radiocarbon ages; seven were re-analyzed in detail. Radiocarbon ages were corrected using the Marine20 calibration. Key cores that penetrated the Storegga discontinuity (SD) were used to assess the presence and depth of post-LGM sediments beneath SD. Integration approach: (1) Map a transparent MTD between R6 (base) and R3 (top) in the northern sidewall indicating a ~20 ka failure (Nyegga Slide MTD) up to ~15 m thick; (2) Compare thickness between R8 and dated horizons in intact sidewall versus within the scar to estimate evacuated sediment thickness for the ~20 ka event; (3) Use cores C72 and C73 age–depth data to show post-LGM sediments occur beneath SD, incompatible with a single ~8150 BP evacuation; (4) Infer an internal boundary within scar infill (Nyegga discontinuity, ND) separating two transparent subunits, consistent with a two-stage failure history; (5) Synthesize observations into a revised failure sequence and estimate Nyegga Slide volume by scaling the proportion of evacuation attributed to ~20 ka across the initial Storegga evacuation volume.

• Discovery of the Nyegga Slide (~20,000 BP): Echosounder profiles across the northern sidewall reveal an acoustically transparent MTD between horizons R6 (base) and R3 (top), up to ~15 m thick, indicating failure of material deposited after 27,600 BP and failure at ~20,000 BP.
• Evacuation thickness: Sidewall–scar comparison (flattened on R8) and core chronologies show sediments dated ~18,600 BP occur 55–57 m above R8 in the intact sidewall but at ~22 m above R8 within the scar. This implies post-Nyegga sediments are 35–37 m deeper in the slide scar, indicating >35 m of sediments were removed during the ~20 ka Nyegga Slide. The previously inferred 50–70 m evacuation for the NE scar is therefore largely attributable to Nyegga (~70% of the 50 m).
• Internal stratigraphy of scar infill: The NE scar infill is transparent with a thin veneer at the top and an internal boundary (Nyegga discontinuity, ND) that correlates with the top of an intact stratified block, separating deposits emplaced after the ~20 ka event from younger units.
• Storegga Slide (~8150 BP) revision: Cores that penetrate the Storegga discontinuity (SD) contain post-LGM sediments beneath SD (e.g., C72 with ~10 ka, C73 with ~18 ka), incompatible with a single-event evacuation at 8150 BP. The Storegga event primarily removed portions of the post-LGM (meltwater plume) cover and induced lateral spreading along R8; the deeply incised central segment remained the main tsunami source.
• Volume implications: If ~70% of the initial NE evacuation occurred at ~20 ka, applying this proportion to the previously estimated initial Storegga evacuation of ~1300–1600 km3 yields a Nyegga Slide volume of ~900–1100 km3. Consequently, the total Storegga Slide volume is revised downward to ~1300–2300 km3 (from 2400–3200 km3 previously), without affecting the central, tsunami-generating failure.
• Tsunami hazard: The existence of at least two major failures within one glacial cycle implies large submarine landslides are more frequent than previously assumed, increasing tsunami hazard for the North Atlantic. Despite a smaller Storegga volume, observations of the 8150 BP tsunami (run-up >20 m in Shetland; effects to Denmark, Greenland, Barents Sea) are consistent with the central segment being the dominant source.
• Contextual data points: LGM sea level ~130 m lower than present; rapid LGM sedimentation (>12 m/kyr) and glacial debris lobe emplacement near the shelf edge likely contributed to overpressure and weak layers; the Nyegga Slide MTD is distinct from local glacial debris lobes and indicates evacuation within the scar area.

The findings directly challenge the prevailing model that attributed the NE Storegga scar evacuation solely to the ~8150 BP event. Seismostratigraphic and core chronostratigraphic evidence demonstrates that a substantial portion (>35 m) of sediment was evacuated at ~20 ka (Nyegga Slide), with Storegga later removing only parts of the post-LGM cover and inducing lateral spreading along a deeper glide plane (R8). This resolves the presence of post-LGM sediments beneath the Storegga discontinuity and necessitates a two-stage reconstruction with an internal Nyegga discontinuity within the scar infill. The revised history preserves the role of the deeply incised central Storegga segment as the tsunami source, reconciling observed run-ups with tsunami models that emphasize the central failure. The study underscores amplified preconditioning at or after the LGM—gas hydrate redistribution and fluid flow, rapid sediment loading and resulting excess pore pressures, ice-sheet dynamics, and possible failure of glacial debris lobes at the shelf edge—as plausible drivers of the Nyegga Slide. However, the available data cannot distinguish whether lateral spreading initiated during Nyegga and was reactivated at 8150 BP, or began with Storegga, nor whether Nyegga comprised one large event or multiple smaller failures. The recognition of an additional major failure implies that large slope failures on the mid-Norwegian margin recur more frequently than assumed, elevating regional tsunami hazard and compelling re-parameterization of hazard and tsunami models that have used Storegga as a benchmark. The reinterpretation also shifts scientific focus toward processes controlling the deeply incised central segment, including the likely role of focused fluid flow near the Ormen Lange gas field.

This work identifies the previously unrecognized ~20,000 BP Nyegga Slide in the Storegga Slide Complex and presents a revised failure history: (1) Nyegga Slide evacuated >35 m of late-glacial sediments (likely 900–1100 km3), forming an internal Nyegga discontinuity within the scar infill; (2) post-LGM plumites (~25–30 m) accumulated; (3) the ~8150 BP Storegga Slide removed part of this cover and induced lateral spreading along R8, while the central segment’s deep failure generated the observed tsunami. The total Storegga Slide volume is revised downward (~1300–2300 km3) without contradicting tsunami evidence. These results imply at least two major landslides per glacial cycle in the area, increasing the assessed tsunami hazard for the North Atlantic and necessitating revisions to risk assessments and tsunami models that treat Storegga as a reference case. Future research should extend high-resolution echosounder and coring coverage across the entire scar to constrain the spatial extent and volume of the Nyegga Slide, determine whether it was a single or multiple events, investigate triggers (e.g., debris lobe failures, fluid flow, gas hydrate dynamics), and refine tsunami source modeling focusing on the central segment’s mechanics and fluid migration pathways.

• Spatial coverage: Echosounder data cover only the northern boundary of the Storegga Slide Complex; glacial debris lobes hinder correlation along the headwall, limiting basin-wide reconstruction.
• Event characterization: Unable to determine whether lateral spreading initiated during the Nyegga Slide or only during Storegga; inconclusive whether the ~20 ka failure was a single large event or multiple smaller events.
• Extent: Difficult to constrain whether Nyegga was local to the northern margin or regional; southern segment cores suggest LGM failure but evidence is not continuous along large stretches of the scar.
• Stratigraphic and depositional constraints: Lack of datable material beneath some key horizons; tsunami deposits from ~20 ka are unlikely to be preserved due to lower sea level and glaciated coasts, limiting validation of potential tsunami impacts.