In the North Atlantic Ocean, a mid-ocean ridge bisects the Icelandic mantle plume, and provides a window into its temporal evolution1,2,3. V-shaped ridges of thick oceanic crust observed south of Iceland are thought to record pulses of upwelling within the plume4,5,6,7. Specifically, excess crust is thought to form during the quasi-periodic generation of hot solitary waves triggered by thermal instabilities in the mantle8. Here we use seismic reflection data to show that V-shaped ridges have formed over the past 55 million years—providing the longest record of plume periodicity of its kind. We find evidence for minor, but systematic, asymmetric formation of crust, due to migration of the mid-ocean ridge with respect to the underlying plume. We also find changes in periodicity: from 55 to 35 million years ago, the V-shaped ridges form every 3 million years or so and reflect small fluctuations in plume temperature of about 5–10 °C. From 35 million years ago, the periodicity changes to about 8 million years and reflects changes in mantle temperature of 25–30 °C. We suggest that this change in periodicity is probably caused by perturbations in the thermal state at the plume source, either at the mantle-transition zone or core–mantle boundary.
Spatial and temporal patterns of convective circulation beneath lithospheric plates cause regional elevation changes at the Earth’s surface that have important—but poorly understood—implications for the development of dynamic topography on geologic timescales. As the Rayleigh number of convecting mantle is 106–108, this circulation is expected to be transient, varying on timescales of 1–100 Myr and on length scales of hundreds to thousands of kilometres3,9. A global network of mid-ocean ridges provides a useful means of estimating the temperature of underlying asthenospheric mantle10,11. At spreading mid-ocean ridges, accretion of oceanic crust is sensitive to small temperature fluctuations that change the thickness of newly formed crust by kilometres2.
In the North Atlantic Ocean, the Reykjanes Ridge crosses the Icelandic plume, a hot convective upwelling with a radius of at least 1,200 km (ref. 12). Within the region influenced by this plume, average thickness of oceanic crust increases from 7 to 14 km and the seabed is anomalously shallow by up to 2 km. Both observations are consistent with an average temperature anomaly of 150 °C (ref. 2). Several different timescales of transient behaviour are sampled by the mid-ocean ridge’s interaction with this plume. On the shortest timescale, the most obvious and best-known features are diachronous V-shaped ridges (VSRs) that are visible on either side of the ridge axis where sedimentary cover is minimal (Fig. 1). These VSRs probably reflect minor changes in the thickness and composition of oceanic crust and are generated when hotter than average parcels of plume material travel radially away from the plume’s conduit13,14. On a much longer timescale, there is a transition from smooth crust without fracture zones, accreted over hotter asthenosphere, to rough crust with fracture zones, accreted over colder asthenosphere (Fig. 1). This observation suggests that the plume’s planform has changed through time. Today, the plume’s thermal (as opposed to chemical) influence extends to the intersection between the mid-ocean ridge and the Bight Fracture Zone at 57° N and 33° W (refs 2, 5, 15). Despite their importance in providing otherwise inaccessible insights into convective processes, the structure and extent of these VSRs are poorly known and their origin is still debated16,17. It is especially unclear how many VSRs exist and how far back in time their history can be traced.
To address these general issues, we acquired two regional (>1,200 km) seismic reflection profiles that traverse the oceanic basin south of Iceland (Fig. 1). Crucially, these profiles provide conjugate images of the Iceland and Irminger basins, because each one of them is oriented parallel to plate-spreading flowlines18. Acquisition and processing details are provided in the Methods. We have two significant findings. First, we have mapped the sediment–basement interface, which demonstrates that VSR activity can be continuously traced back to 55 Myr ago (Ma). Second, this activity has been used to build a detailed chronology of asthenospheric potential temperature, Tp. This continuous record provides a reference frame for analysing relationships between plume activity and other geologic observations.
Profile 2 resolves the detailed structure of the Iceland and Irminger basins (Fig. 2a). Away from a prominent mid-ocean ridge, the top of oceanic crust is clearly imaged beneath layered sediments, which thicken in either direction. A sediment–basement interface can easily be traced, despite being cut by minor faults (Fig. 2b). The sedimentary pile is dominated by contourite drift deposits that record the history of deep-water overflow across the Greenland–Scotland Ridge. For example, Eirik Drift records 7 Myr of overflow through the Denmark Strait and is visible at the northwestern end of the profile14. This overflow caused incision of older contourite deposits northwest of the mid-ocean ridge. The sediment–basement interface is deformed into a series of prominent ridges and troughs that are imaged out to ~500 km on either side of the mid-ocean ridge. These ridges and troughs are 20–40 km wide with amplitudes of up to 1 km and they correlate with small free-air gravity anomalies (Fig. 2c, d). Detailed interpretation shows that these ridges and troughs are broken up, but not controlled, by normal faulting (Fig. 2e, f).
Average crustal thickness along the Reykjanes Ridge is primarily controlled by asthenospheric temperature within the plume head13. V-shaped ridges and troughs are maintained by minor changes in oceanic crustal thickness, which in turn are generated by temperature fluctuations within the plume18. Changes in the composition of basaltic rocks and in the geometry of active faults along the Reykjanes Ridge suggest that these temperature fluctuations are ±25 °C (refs 13, 14, 19). Here, we exploit residual depth anomalies as a proxy for tracking crustal thickness and asthenospheric temperature fluctuations (Fig. 3a). Residual depth is the water-loaded depth to oceanic crust that has been corrected for sediment loading, plate age and present-day dynamic support6. South of Iceland, residual depth varies by ±400 m and is controlled by changes in crustal thickness. If crust is generated at the mid-ocean ridge by isentropic decompression of anhydrous mantle11,13, Tp can be estimated from residual depth measurements using
where tc = 8.4 km is a reference crustal thickness18, dr is residual depth, ρa = 3.2 Mg m−3 is density of asthenospheric mantle, ρc = 2.8 Mg m−3 is density of oceanic crust, and ρw = 1 Mg m−3 is the density of sea water. We have projected our Tp estimates into age–distance space and combined them with satellite gravity observations (Fig. 3b). There is excellent agreement between these estimates and free-air gravity anomalies on young, smooth oceanic crust (<20 Ma). On the oldest crust, VSRs are also visible and correlate with weak linear gravity anomalies whose significance was not previously recognized owing to variable thicknesses of sedimentary cover (Fig. 1). Some of the oldest VSRs are manifested by resolvable crustal thickness changes of ±1 km (ref. 20). At radial distances of >500 km from the plume centre, symmetric lobes of rough crust are intersected by profile 1 between 20 and 35 Ma (Fig. 3b). Within these highly fractured lobes, coherent VSRs are not clearly observed and legacy seismic refraction data suggest that the crust is only 6.1 km thick5,21. This observation suggests that the rough–smooth boundary represents the lateral extent of the plume.
VSRs are not exactly symmetric about the Reykjanes Ridge. For example, an old and prominent VSR occurs at a distance of 350 km from the ridge axis on the eastern side of profile 2 (Fig. 2b). On the western side of the same profile, this VSR occurs at 370 km, which corresponds to a cumulative offset of 20 km. Over the past 30 Myr, a systematic pattern of increasing offset is consistent with a history of asymmetric crustal accretion documented using magnetic anomaly profiles located closer to Iceland16,17. At distances of <250 km from the ridge axis, estimates of asymmetry made from magnetic anomalies and VSRs agree within error (Fig. 3c). Increasing asymmetry corresponds to a series of well-known eastward ridge jumps on Iceland that reflect the fact that the mid-ocean ridge gradually drifts westward with respect to the plume centre, periodically relocating itself at the centre of the plume22. VSR asymmetry between 300 and 500 km corresponds to a much older westward switching in seafloor spreading from the now-extinct Aegir Ridge to the active Kolbeinsey Ridge located north of Iceland15,23.
Growth and decay of asymmetry enables us to synchronize VSR chronology on either side of the Reykjanes Ridge (Fig. 3a, d). The optimal result is obtained by adjusting a subset of poorly constrained magnetic anomaly picks (that is, magnetic chrons 8–17 corresponding to 20–40 Ma) along profile 2 by ±1 Myr, which is within the range of uncertainty15. The resultant match between eastern and western portions of this profile implies that VSRs were generated by radially expanding temperature fluctuations that were generated deep within the plume’s conduit7,14,16. Growth and decay of asymmetric spreading correlate with plume activity, which suggests a causal relationship. In the North Atlantic Ocean, it is known that the mid-ocean ridge drifts northwestward with respect to the centre of the plume15. Our results indicate that the cumulative amount of drift increases when the plume is quiescent (compare Fig. 3c and d). Greater plume activity substantially increases the distal radial force that probably acts to inhibit plate spreading, encouraging the mid-ocean ridge to switch back to the plume centre. If elevation at the plume centre goes up by 200 m, the distal radial force increases by a minimum of 2 × 108 N m−1.
We have combined analysis of both profiles with the results of ref. 20 to calculate an average history of plume temperature fluctuations (Fig. 3d). Between 55 and 35 Ma, small (~5–10 °C) fluctuations of plume temperature have a periodicity of ~3 Myr. These fluctuations are superimposed on a rapidly cooling temperature structure that is also manifested by a northward shift in the transition from smooth to rough crust. Both observations are consistent with wholesale shrinkage of the plume20. After 35 Ma, the radius of the convective planform rapidly expanded from 400 to at least 1,200 km. This growth was accompanied by larger (~25–30 °C) fluctuations of plume temperature that have a periodicity of up to 8 Myr (Fig. 3d). This changing periodicity is probably caused by boundary layer perturbations within the convecting mantle4,5,20,24. Scaling analysis suggests that VSR activity is compatible with perturbations that form either at the 670 km mantle discontinuity or at the core–mantle boundary (Supplementary Information).
Using values from Supplementary Table 1, the geometry of the youngest VSR confirms that the present-day buoyancy flux of the plume is B = 18 ± 7 Mg s−1 if plume material flows radially away from the plume centre within an asthenospheric layer that is 125 ± 25 km thick with an excess temperature of ΔT = 150 ± 50 °C (refs 6, 7). Independent values of B can be obtained by exploiting two separate observations. First, the changing boundary between smooth and rough crust, d, is controlled by a combination of plate-spreading rate, u, and B where
which yields B = 26 ± 9 Mg s−1 for the past 2 Myr. Second, the present-day planform of the plume swell constrains its excess volume25. If a plume radius of 1,200 ± 100 km grew over the past 25–35 Myr, B = 17 ± 5 Mg s−1. All three estimates of buoyancy flux are consistently large, indicating that the Iceland plume is probably the biggest convective upwelling on Earth. In contrast, the Hawaiian plume has a buoyancy flux of only 8.7 Mg s−1 (ref. 26).
Our seismic reflection interpretations suggest that buoyancy flux has changed through time. For example, the oldest VSRs within the Irminger and Iceland basins have weak linear gravity anomalies that yield B = 73 ± 15 Mg s−1 and 66 ± 14 Mg s−1, respectively. Such high values are consistent with the oldest smooth lobes of crust that extend at least 1,400 km away from the centre of the plume, implying that B ≥ 70 Mg s−1 (Fig. 3b).
Finally, our observations help to bound the dimensions of solitary waves that are generated at putative thermal boundary layers and travel up deformable conduits of plumes (Fig. 4)8. In the plate-spreading direction, the youngest VSRs are 25–30 km wide whereas older ones are 15–20 km wide. The youngest VSR is ~730 km from the plume centre and has a width x = 25 ± 3 km. Assuming a present-day plume flux of ~18 ± 4 Mg s−1, the width of the VSR in the direction of the mid-ocean ridge, ΔR, is expected to be 244 ± 44 km. This value is consistent with a 250-km-long segment of increased volcanism and reduced seismicity along the ridge crest near 60° N (ref. 19).
We have presented observations that document a continuous record of transient behaviour of the Icelandic plume between 55 Ma and the present day. Transient thermal anomalies occur every 3–8 Myr and are generated by boundary layer instabilities. Present-day buoyancy flux of the Iceland plume suggests that it is the largest convective upwelling on Earth. Fluctuating dynamic support during the Cenozoic era provides a general mechanism for proposed changes in deep-water oceanic circulation14, for sedimentary drift accumulation27, and for the carving of ancient ephemeral landscapes28. Establishing these connections between convective chronologies and surface observations has helped to yield new insights into the coupled nature of Earth’s deep and surficial realms.
Seismic data acquisition and processing.
Seismic profiles were acquired onboard the RRS James Cook during July–August 2010 by the universities of Cambridge, Southampton and Birmingham. This cruise, JC50, was financially supported by the Natural Environmental Research Council. Acoustic energy was generated using a single generator–injector airgun with a total volume of 5.82 l (generator pulse = 4.1 l, injector pulse = 1.72 l) and a frequency bandwidth of 10–400 Hz. The airgun was towed at a depth of 5.5 m behind the vessel, which steamed at ~9.3 km h−1. This airgun was primed with compressed air (20.7 MPa) and fired every 15 s (~40 m). Reflected acoustic energy was recorded on a 1,600-m-long streamer towed at 7 m depth. This streamer consisted of 132 groups of hydrophones located every 12.5 m. Distance from the airgun to the first group (that is, near-trace offset) was 163 m. The digital sampling interval of recorded signals was 1 ms. During the survey, impulses of acoustic energy are transmitted and reflected at discontinuities within the Earth, where changes in acoustic impedance are generated by density and velocity contrasts. The geometry of this survey was designed to repeatedly record signals every 6.25 m along the profile. This redundancy improves signal to noise because reflections from different shotpoint–receiver pairs can be stacked together. Before stacking, acoustic velocity is carefully picked as a function of two-way travel time to correct for the travel-time delay (that is, normal move-out) of different raypaths within a single common midpoint gather. Here, velocity functions were hand-picked every 100 common midpoints (that is, every 625 m). The resultant 21-fold stacked image has a vertical and horizontal resolution of 10–20 m. Signal processing techniques also included application of a 12 Hz high-pass filter with a roll-off of 24 dB per octave, and a post-stack Stolt migration with a constant acoustic velocity of 1,500 m s−1.
Vogt, P. R. Asthenosphere motion recorded by the ocean floor south of Iceland. Earth Planet. Sci. Lett. 13, 153–160 (1971).
White, R. S. Rift-plume interaction in the North Atlantic. Phil. Trans. R. Soc. Lond. A 355, 319–339 (1997).
Schubert, G., Turcotte, D. L. & Olson, P. Mantle Convection in the Earth and Planets (Cambridge Univ. Press, 2001).
Ito, G. Reykjanes ‘V’-shaped ridges originating from a pulsing and dehydrating mantle plume. Nature 411, 681–684 (2001).
Jones, S. M., White, N. J. & Maclennan, J. V-shaped ridges around Iceland: Implications for spatial and temporal patterns of mantle convection. Geochem. Geophys. Geosyst. 3, 1–23 (2002).
Poore, H. R., White, N. J. & Jones, S. M. A Neogene chronology of Iceland plume activity from V-shaped ridges. Earth Planet. Sci. Lett. 283, 1–13 (2009).
Jones, S. M. et al. A joint geochemical-geophysical record of time-dependent mantle convection south of Iceland. Earth Planet. Sci. Lett. 386, 86–97 (2014).
Schubert, G., Olson, P., Anderson, C. & Goldman, P. Solitary waves in mantle plumes. J. Geophys. Res. 94, 9523–9532 (1989).
Loper, D. E. & Stacey, F. D. The dynamical and thermal structure of deep mantle plumes. Phys. Earth Planet. Inter. 33, 304–317 (1983).
Klein, E. M. & Langmuir, C. H. Global correlations of ocean ridge basalt chemistry with axial depth and crustal thickness. J. Geophys. Res. 92, 8089–8115 (1987).
McKenzie, D. P. & Bickle, M. J. The volume and composition of melt generated by extension of the lithosphere. J. Petrol. 29, 625–679 (1988).
Rickers, F., Fichtner, A. & Trampert, J. The Iceland Jan Mayen plume system and its impact on mantle dynamics in the North Atlantic region: Evidence from full-waveform inversion. Earth Planet. Sci. Lett. 367, 39–51 (2013).
White, R. S., Bown, J. & Smallwood, J. R. The temperature of the Iceland plume and origin of outward-propagating V-shaped ridges. J. Geol. Soc. 152, 1039–1045 (1995).
Poore, H. R., White, N. J. & Maclennan, J. Ocean circulation and mantle melting controlled by radial flow of hot pulses in the Iceland plume. Nature Geosci. 4, 558–561 (2011).
Smallwood, J. R. & White, R. S. Ridge-plume interaction in the North Atlantic and its influence on continental breakup and seafloor spreading. Geol. Soc. Spec. Publ. 197, 15–37 (2002).
Hey, R., Martinez, F., Höskuldsson, Á. & Benediktsdóttir, Á. Propagating rift model for the V-shaped ridges south of Iceland. Geochem. Geophys. Geosyst. 11, Q03011 (2010).
Benediktsdóttir, Á., Hey, R., Martinez, F. & Höskuldsson, Á. Detailed tectonic evolution of the Reykjanes Ridge during the past 15 Ma. Geochem. Geophys. Geosyst. 13, Q02008 (2012).
Smallwood, J. R. & White, R. S. Crustal accretion at the Reykjanes Ridge. J. Geophys. Res. 103, 5185–5201 (1998).
Parnell-Turner, R. E. et al. Crustal manifestations of a hot transient pulse at 60° N beneath the Mid-Atlantic Ridge. Earth Planet. Sci. Lett. 363, 109–120 (2013).
Parkin, C. J. & White, R. S. Influence of the Iceland mantle plume on oceanic crust generation in the North Atlantic. Geophys. J. Int. 173, 168–188 (2008).
Whitmarsh, R. B. Seismic anisotropy of the uppermost mantle absent beneath the east flank of the Reykjanes Ridge. Bull. Seismol. Soc. Am. 61, 1351–1368 (1971).
Hardarson, B., Godfrey Fitton, J., Ellam, R. M. & Pringle, M. S. Rift relocation — a geochemical and geochronological investigation of a palaeo-rift in northwest Iceland. Earth Planet. Sci. Lett. 153, 181–196 (1997).
Jung, W-Y. & Vogt, P. R. A gravity and magnetic anomaly study of the extinct Aegir Ridge, Norwegian Sea. J. Geophys. Res. 102, 5065–5089 (1997).
Olson, P., Schubert, G. & Anderson, C. Plume formation in the D′′-layer and the roughness of the core–mantle boundary. Nature 327, 409–413 (1987).
Crosby, A. G. & McKenzie, D. P. An analysis of young ocean depth, gravity and global residual topography. Geophys. J. Int. 178, 1198–1219 (2009).
Sleep, N. H. Hotspots and mantle plumes: Some phenomenology. J. Geophys. Res. 95, 6715–6736 (1990).
Wold, C. N. Cenozoic sediment accumulation on drifts in the northern North Atlantic. Paleoceanography 9, 917–941 (1994).
Hartley, R. A., Roberts, G. G., White, N. J. & Richardson, C. Transient convective uplift of an ancient buried landscape. Nature Geosci. 4, 562–565 (2011).
Sandwell, D. T. & Smith, W. H. F. Global marine gravity from retracked Geosat and ERS-1 altimetry: Ridge segmentation versus spreading rate. J. Geophys. Res. 114, B01411 (2009).
Shorttle, O., Maclennan, J. & Jones, S. M. Control of the symmetry of plume-ridge interaction by spreading ridge geometry. Geochem. Geophys. Geosyst. 11, Q0AC05 (2010).
This research was supported by NERC Grant NE/G007632/1 and by the Girdler Fund, University of Cambridge. We thank the Master, crew and technicians of RRS James Cook Cruise JC50 for their dedicated professionalism. We are grateful to J. Rudge for assisting with boundary layer scaling analysis and to M. Falder for pointing out an error. Earth Sciences contribution esc.3118.
The authors declare no competing financial interests.
About this article
Cite this article
Parnell-Turner, R., White, N., Henstock, T. et al. A continuous 55-million-year record of transient mantle plume activity beneath Iceland. Nature Geosci 7, 914–919 (2014). https://doi.org/10.1038/ngeo2281
Nature Reviews Earth & Environment (2021)
Transient mantle cooling linked to regional volcanic shut-down and early rifting in the North Atlantic Igneous Province
Bulletin of Volcanology (2020)
Large Igneous Province thermogenic greenhouse gas flux could have initiated Paleocene-Eocene Thermal Maximum climate change
Nature Communications (2019)
Acta Oceanologica Sinica (2019)
Surface Couplings for Subdomain-Wise Isoviscous Gradient Based Stokes Finite Element Discretizations
Journal of Scientific Computing (2018)