Research ArticleGEOLOGY

Extensive marine-terminating ice sheets in Europe from 2.5 million years ago

See allHide authors and affiliations

Science Advances  13 Jun 2018:
Vol. 4, no. 6, eaar8327
DOI: 10.1126/sciadv.aar8327
  • Fig. 1 Marine oxygen isotope climate proxy record and modern and Early Pleistocene North Sea bathymetry.

    (A) Composite benthic δ18O stack (42) climate proxy record, showing the transition from 41-ka obliquity to 100-ka eccentricity forcing between ~1.2 and 0.6 Ma ago. The solid black line indicates the currently accepted timing for major glaciation of the North Sea. The red dashed line shows the earliest glaciation identified from this study, when ice margins were terminating in water depths of ~250 m. The solid red line shows earliest age for grounded ice in deepest part of North Sea, at ~300 m water depths, identified from this study. E, Early Pleistocene; M, Middle Pleistocene; L, Late Pleistocene. (B) Base Pleistocene bathymetry shown as depth below modern sea level. The white circles are the fluvial inputs used for the iceberg trajectory modeling, and the light blue line is the −70 m contour representing the Early Pleistocene glacial coastline. This uses the same bathymetric color scale as in (C), with contours every 50 m. The inset shows a section along the line A–A′ illustrating the bathymetric lip south of 60°N shown as depth below modern sea level (−70 m for Early Pleistocene glacial maxima). MegaSurvey three-dimensional (3D) seismic data courtesy of PGS. (C) North Sea, present day with the North Viking Graben (NVG), South Viking Graben (SVG), and Central Graben (CG) marked (gray dashed lines). It should be noted that the Norwegian Channel (directly east of NVG and SVG) had not been eroded during the Early Pleistocene timeframe of interest in this paper. The locations of other figures are shown. The white circles are drill sites, and the light blue line delimits the 3D seismic coverage. Bathymetric contours are every 100 m. A, Aviat; J, Josephine; A15, well A15-03.

  • Fig. 2 Clinoform geometry, chronology, and iceberg scours.

    (A) Lines show the semiautomatically picked reflectors from the 3D seismic data. Blue lines show the seismic horizons that are used as a dating control and that define the six seismic units (see Methods and fig. S6). The red horizons are those with iceberg scours present, and the white horizons are those where glacial landforms have not been identified. The seismic line location is shown in Fig. 1C. The yellow dotted line shows the stratigraphic location of the surface mapped and displayed in (B). Mapping from picked reflectors/surfaces provides more robust chronological and depth/TWT (two-way travel time) control than mapping from time slices (22) (horizontal planes of equal TWT). (B) Surface created from the 3D seismic with the root mean square amplitude attribute displayed. White arrows show examples of curvilinear iceberg scours that were mapped on this surface. MegaSurvey 3D seismic data courtesy of PGS.

  • Fig. 3 Early Pleistocene pan–North Atlantic modeled iceberg trajectories and inferred ice sheet geometries around the North Sea.

    (A) Schematic showing the modeled iceberg trajectories for all studied release sites (see fig. S3 for individual experiment iceberg density fields). Approximate areas where icebergs were seeded into the Early Pleistocene ocean simulations are shown by the red dots. (B) Hypothetical geometries of the BIIS and FIS during the Early Pleistocene from MIS 100 (2.53 Ma ago) with the dashed ice sheet margins indicating that they are unconstrained. The western margin of the FIS, between 61°N and 67°N, is believed to have encroached out toward the shelf break as it prograded westward, but no evidence suitable for constraining the exact ice margin geometry has yet been identified in the earliest Pleistocene section (6). The iceberg scours in the North Sea, mapped in this study, required an ice sheet margin/s to have been present south of the bathymetric lip at ~60°N (Fig. 1B), grounded in water depths on the order of 250 m (Fig. 1B, figs. S1 to S3, and table S1). The back dashed line represents the −70-m Early Pleistocene glacial shoreline determined from the seismic data, and the black arrows indicate the freshwater inputs based on modern locations (extended outward to the paleoshoreline) and paleodrainage patterns on continental Europe. Water depths are relative to modern sea level. Compared to previous interpretations, the hypothetical FIS has a greater southward extension, and the BIIS is greatly expanded (18, 58). MegaSurvey 3D seismic data courtesy of PGS.

  • Fig. 4 Evidence for grounded ice streams, chronology, and landform metrics.

    (A) Mapped flowsets of MSGLs with color coding indicating estimated minimum ages. The inset rose diagram indicates the average bidirectional orientation of each flowset, and the inset box indicates the location of the mapped surface shown in (B). (B) Surface extracted from 3D seismic data, displaying instantaneous amplitude, highlighting MSGLs over Aviat. The inset rose diagram indicates the bidirectional orientation of MSGLs mapped in the flowset. (C) All data from the mapped flowsets compare favorably with metrics from a benchmark data set (27). MegaSurvey 3D seismic data courtesy of PGS.

  • Fig. 5 Advance and retreat of grounded ice, sediment flux into the bottom of the basin from 1.87 Ma ago, and possible ice sheet geometries.

    (A) Massive distal glacimarine silty muds with occasional angular clasts interpreted as dropstones (sample location i on fig. S4A). Note that the subhorizontal lines are saw cuts. (B) Rippled, very fine sands deposited from hyperpycnal flows during ice advance. They are deformed compressively by subsequent ice stream overriding with the vergence indicating compression from the north/west (sample location ii on fig. S4A). (C) Cross-stratified to rippled very fine sand deposited from hyperpycnal flows during ice stream retreat (sample location iii on fig. S4A). Sediment transport appears to be toward the north/west (left), but some cross-sets build toward the right, indicating variable flow directions. (D) SEM micrograph illustrating the fresh texture of the grains, noting the lack of solution features (specifically on the feldspars and biotite) or overgrowths (sample location iv on fig. S4A). (E) Sediment thickness determined from seismic and borehole log data indicating the extent of the subglacial till. The thickness map is overlaid onto the basin structure at ~1.8 Ma ago. Contours are measured every 50 ms. Sediment flux from the glaciers and the major European rivers (clinoform progradation) are indicated. (F) Hypothetical geometries of the BIIS and FIS during the Early Pleistocene from MIS 70 (1.87 Ma ago) onward, indicating near or complete coalescence (the configurations from MIS 100 shown in Fig. 3B are also provided by way of comparison). The dashed ice sheet margin indicates that these limits are unconstrained. The MSGLs directions are indicated by the black arrows, with the single arrowheads indicating the inferred ice flow direction based on paleogeography and paleobathymetry. The dual arrowheads in the center of the basin indicate repeated incursions/ice flows from the NE and SW. The back dashed line represents the −70 m Early Pleistocene glacial shoreline determined from the seismic data. Water depths are relative to modern sea level. Compared to previous interpretations, the hypothetical FIS has a greater southward extension, and the BIIS is greatly expanded (18, 58). MegaSurvey 3D seismic data courtesy of PGS.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/6/eaar8327/DC1

    fig. S1. The modern North Sea, bathymetry, classification, seismic data coverage, well locations, and sites of reported Early Pleistocene iceberg scours.

    fig. S2. Temporal and spatial patterns of iceberg scours and their relationship with IRD and sea level.

    fig. S3. Iceberg trajectory modeling experiments.

    fig. S4. Core stratigraphy, images, and interpretation.

    fig. S5. Theoretical ice sheet surface profiles demonstrating the effects of basal resistance and topography.

    fig. S6. A section showing the basis for the seismic stratigraphic framework tied to the magnetic reversal and palynology-based chronology from A15-03 in the Dutch sector of the North Sea (blue lines), which is corroborated by ages from Josephine and Aviat in the UK sector (16).

    fig. S7. Downhole gamma ray logs from the Dutch sector of the southern North Sea.

    table S1. Details for sites where Early Pleistocene iceberg scours have previously been identified.

    table S2. Summary information for the biostratigraphical evaluation of Aviat cores 22-7a-5z and 22-7a-6z.

    References (6671)

  • Supplementary Materials

    This PDF file includes:

    • fig. S1. The modern North Sea, bathymetry, classification, seismic data coverage, well locations, and sites of reported Early Pleistocene iceberg scours.
    • fig. S2. Temporal and spatial patterns of iceberg scours and their relationship with IRD and sea level.
    • fig. S3. Iceberg trajectory modeling experiments.
    • fig. S4. Core stratigraphy, images, and interpretation.
    • fig. S5. Theoretical ice sheet surface profiles demonstrating the effects of basal resistance and topography.
    • fig. S6. A section showing the basis for the seismic stratigraphic framework tied to the magnetic reversal and palynology-based chronology from A15-03 in the Dutch sector of the North Sea (blue lines), which is corroborated by ages from Josephine and Aviat in the UK sector (16).
    • fig. S7. Downhole gamma ray logs from the Dutch sector of the southern North Sea.
    • table S1. Details for sites where Early Pleistocene iceberg scours have previously been identified.
    • table S2. Summary information for the biostratigraphical evaluation of Aviat cores 22-7a-5z and 22-7a-6z.
    • References (66–71)

    Download PDF

    Files in this Data Supplement:

Navigate This Article