Roadmap | The review process | Home
   Tectonic origin for Hawaii
Speculations on tectonic origin of the Hawaii hotspot

Ian O. Norton

ExxonMobil Upstream Research, Houston, Texas, USA


The Hawaii-Emperor island and seamount chain is the product of a long-lived volcanic center that has moved in various directions relative to the Pacific plate. The volcanic center is usually referred to as a "hotspot" and the seamount-island chain as a "hotspot track". Age data on the Hawaii-Emperor chain indicates a remarkably linear trend of age vs. distance from Hawaii along the chain. Age progression continues to the oldest seamount at the northern end of the Emperor Chain, Meiji, 85 m.y. old. This seamount is close to the Aleutian trench. An unanswered question is whether the Emperor Chain once continued to the north, with seamounts older than Meiji that have been subducted into the trench. A corollary to this question is the ultimate age of the hotspot – when did it initiate? In this note I suggest that Meiji is in fact the oldest seamount in the chain and that the hotspot initiated as a result of significant reorganizations on the spreading boundaries of the Pacific plate, i.e., as a result of tectonic processes. It is probable that other topographic features in the North Pacific such as the Shatsky and Hess Rises were also generated by tectonic processes. I propose that hotspots like Hawaii are initially formed by tectonic (i.e., shallow) processes, although the mechanisms for their longevity remain unknown.

This posting is a summary of a paper submitted to the proceedings of the AGU Chapman Conference “The Great Plume Debate: The Origin and Impact of LIPs and Hot Spots”


Tectonic Evolution of the North Pacific

To examine initiation of the Hawaii hotspot we need to understand the tectonic evolution of the North Pacific. Tectonic information from the area comes from the mapping of oceanic spreading anomalies, fracture zones and other topographic features (Figures 1 and 2). Topographic features (except Shatsky Rise) show up clearly on the satellite gravity data (Sandwell, 2005) used in these figures. The most prominent feature is the Hawaii-Emperor chain, but for now we will concentrate on other tectonic information. Figure 1 is a view of the satellite gravity. Image processing used to produce the figure, with artificial illumination from the northeast, is designed to enhance subtle features in the data; fracture zones are especially clear. Magnetic lineations, interpreted fracture zones and other topographic features are included in Figure 1, with named features identified. In Figure 2 the magnetic lineations are identified, with a more subdued gravity background. The thick black lines in Figure 2 are tracks of triple junctions where they can be mapped from magnetic lineations; lines are dashed where tracks are inferred within the Cretaceous normal polarity interval.

Figure 1. Free air gravity map of the northern Pacific (Sandwell 13.1, 2005). Thin magenta lines are mapped fracture zones, thin yellow lines are identified magnetic lineations. SR = Shatsky Ridge; HR = Hess Rise; ET = Emperor Trough; CT = Chinook Trough; SFZ = Surveyor Fracture Zone; MFZ = Mendocino Fracture Zone; PFZ = Pioneer Fracture Zone; UFZ = Murray Fracture Zone; OFZ = Molokai Fracture Zone; AFZ = Amlia Fracture Zone; LR = Liliuokalani Ridge; MS = Musicians Seamounts; JP = Japanese Group seamounts; MWC = Marcus Wake chain; MPM = Mid Pacific Mountains; NFZ = Nosappu fracture zone; NR = Necker Ridge; KU = Kruzenstern fracture zone; NS = Non Surveyor feature, HT = Hokkaido Trough. Magnetic lineations (identified in Figure 2) are from compilatioms maintained by Larry Lawver and Lisa Gahagan at the Plates Project, Univerity of Texas at Austin, and my own updates digitized from Nakashini et al. (1989) and Atwater (1989). Mercator projection; scale bar is for approximately the latitude of Hess Rise. Click here or on Figure for enlargement.

Figure 2. Identified magnetic lineations in the North Pacific. Heavy black lines track triple junctions, dashed where inferred. M-sequence lineations are west of the Hawaii-Emperor chain; they are numbered without the M prefix. Click here or on Figure for enlargement.

Previous interpretations of the North Pacific include those of Larson & Chase (1972), Hilde et al. (1976, 1977), Woods & Davies (1982), Rea & Dixon (1983), Mammerickx & Sharman (1988), Atwater (1989) and Smith (2003). As discussed in Norton (2006), differences between these interpretations lie mostly in how tectonic evolution during the Cretaceous quiet zone is interpreted. One important difference lies in what plate(s) moved generally northward away from the Pacific. Larson & Chase (1972) and Hilde et al. (1976, 1977) assume that this plate was always the Kula plate. Woods & Davies (1982) suggested that the Kula plate only came into existence at Chron 34 time (84 Ma) and that the plate along the northern boundary of the Pacific plate during M-sequence time was another plate, the Izanagi. I suggest, however, that the earlier interpretations are correct and that there was always only one plate bordering the northern Pacific plate. For convenience I follow the convention developed over the past 24 years and refer to the Izanagi as the older plate and Kula as the younger, but finally use the name Kula/Izanagi for the single plate.

Tectonic evolution is presented here as a series of figures showing progressive changes in the Pacific, Izanagi, Kula and Farallon plate boundaries, using the time scale of Gradstein et al. (2004). I use the gravity and tectonic data map, Figure 2, as a base. Figures step though time showing areas of sea floor appearing as they are created. Figure 3 is a copy of Figure 2 with the area north and west of the trench blanked out. This is done to emphasize that the sea floor we are dealing with was a long way from any continental margin when it formed, as shown in Figure 4. This is a Pacific-wide reconstruction for 80 Ma, using the plate circuit approach for calculating relative plate positions (rotation poles from Norton, 1995). It is not possible to calculate the position of the Pacific plate relative to the continents during M-sequence time, as the Pacific was totally surrounded by subduction zones and the fixed hotspot reference frame can no longer be regarded as valid (Tarduno et al., 2003). The plate circuit can only be used for times later than 90 Ma, when the Pacific became attached to New Zealand and rifted from West Antarctica.

Figure 3. Same as Figure 2 but with area north and west of the subduction zone blanked out. This is done to emphasize that the area of the Pacific that we are dealing with was tectonically active while it was a long way from the margin. Click here or on Figure for enlargement.

Figure 4. Plate reconstruction for 80 Ma. Arrows indicate relative motion across plate boundaries (red lines) which are dashed where inferred. Green lines are magnetic lineations, magenta lines are fracture zones. OJP = Ontong Java plateau.

Figures 5 through 10 step through the tectonic evolution from 125 to 71 Ma.

125 Ma (M0, Figure 5)

  • Pacific plate boundaries are constrained by magnetic lineations.
  • Izanagi – Farallon boundary is drawn assuming a RRR triple junction.
  • Shatsky Rise formed along the track of the migrating triple junction (Sager et al., 1999).

Figure 5. Tectonic setting at M0 time, 125 Ma. Heavy red lines are plate boundaries, arrows show relative motion directions. Click here or on Figure for enlargement.

110 Ma (Figure 6)

  • Pacific plate boundaries extrapolated into the quiet zone from the M sequence using spreading rates from Nakanishi et al. (1989).
  • Izanagi – Farallon boundary drawn assuming a RRR triple junction.
  • Hess Rise straddles the presumed Pacific-Farallon spreading axis, suggesting that the rise formed as a result of extra volcanism at a spreading center, possibly associated with a spreading direction change.

Figure 6. Tectonic setting at approximately 110 Ma. Click here or on Figure for enlargement.

90 Ma (Figure 7)

  • Pacific plate boundaries extrapolated into the quiet zone back in time from Chron 34.
  • Izanagi – Farallon boundary drawn assuming a RRR triple junction.
  • Izanagi–Pacific spreading assumed to continue in a NW direction as implied by the Kruzenstern and parallel fracture zones in the far north.

Figure 7. Tectonic setting late in the quiet zone at 90 Ma. Click here or on Figure for enlargement.

90 Ma (Figure 8)

  • The time maybe as little as 1 m.y. after Figure 7.
  • Triple junction jumps to southeast from A to B; drawn assuming a RRR configuration.
  • Emperor Trough formed as a sinistral transform fault.

Figure 8. Tectonic setting at 90 Ma after the triple junction jumped from A to B, creating the Emperor Trough as a sinistral transform fault. Click here or on Figure for enlargement.

84 Ma (Chron 34, Figure 9)

  • Pacific – Izanagi/Kula spreading starts to change direction.
  • The Chinook Trough formed at a spreading ridge as a result of a spreading direction change.

Figure 9. Tectonic setting at Chron 34, 84 Ma. Click here or on Figure for enlargement.

71 Ma (Chron 32, Figure 10)

  • Pacific – Izanagi/Kula spreading is now north-south.
  • Meiji Seamount forms at a spreading ridge abandoned as a result of the spreading direction change.

Figure 10. Tectonic setting at Chron 32, 71 Ma. Kula spreading direction has reoriented to north relative to the Pacific plate. Meiji seamount, which was close to the ridge axis in the previous figure, is now several hundred kilometers from the ridge. Click here or on Figure for enlargement.

Implications for the origin of the Hawaii hotspot

As presented here, Jurassic through end Cretaceous tectonic evolution of the North Pacific involved just three plates. These were the Pacific, Farallon and Kula/Izanagi plates. There is no reason to invoke extra plates such as the Chinook (Rea & Dixon, 1983). Changes in spreading direction implied by changes in fracture-zone- and magnetic-lineation strike can all be accounted for with a simple three-plate system. Anomalous topographic features in the area were created in several ways: Shatsky Rise by excess ridge axis-type volcanism at a triple junction (Sager et al., 1999; Mahoney et al., 2005); Hess Rise by excess volcanism at a ridge-transform intersection perhaps associated with a spreading direction change (Figure 6); Chinook Trough by ridge reorganization also associated with a spreading direction change (Figure 9); Emperor Trough as a fracture zone (Figure 8) and Meiji Seamount at a spreading ridge associated with both a spreading direction change and perhaps at the ridge left behind by a ridge jump (Figures 9 and 10).

The three-plate tectonic model presented above, which is similar to models proposed by Larson & Chase (1972), Hilde et al. (1977) and Smith (2003), leads to a scenario for initiation of the Hawaii – Emperor seamount chain. The oldest dated seamount in the chain, Detroit seamount (Figure 10) is 75.8 million years old (40Ar/39Ar from whole rock basalt samples and feldspar separates; Tarduno et al., 2003, Doubrovine & Tarduno, 2004; Clouard & Bonneville, 2005), although 40Ar/39Ar ages as old as 81.2 Ma have been obtained from basalts cored on the seamount (Keller et al., 1995). Meiji seamount at the northern end of the chain is not reliably dated. Based on extrapolation from Detroit seamount, its age is thought to be about 85 Ma (Tarduno & Cottrell, 1997; Regelous et al., 2003). Direct evidence yields younger ages. Dalrymple et al. (1980) reported a minumun K-Ar age of 61.9 ± 5 Ma. Fossil assemblages in overlying sediments suggest an age of 68-70 Ma (Worsley, 1973). I assume here that these ages are indeed minimum ages and that Meiji is as old as 85 Ma.

Geochemical evidence presented by Keller et al. (2000), particularly strontium, lead and neodymium ratios, are also consistent with formation of Meiji and Detroit seamounts close to a ridge axis. I suggest here that Meiji formed at the Pacific – Kula/Izanagi ridge axis as a result of excess volcanism associated with the 30° change in spreading direction from 84 to 71 Ma shown in Figures 9 and 10. As inferred in Figure 10, Meiji may actually have formed at an abandoned spreading center left behind as north-directed spreading became established.

Meiji is the oldest preserved seamount in the Hawaii-Emperor seamount chain. Whether there were older seamounts in the chain that have been subducted is, of course, not known. However, if the inference from gravity data, isotopic data and the plate boundary scenario presented above that Meiji formed at a ridge axis is correct, it is likely that Meiji is the oldest seamount in the chain. If this is so, and it formed by processes related to sea floor spreading, an important implication is that the Hawaii hotspot also formed by processes initiated at the (abandoned?) spreading ridge. This means that the hotspot formed as a result of spreading processes and plate tectonic motion. It would, of necessity, have a shallow (upper mantle) origin. This is different to the usual model for a hotspot as being derived from the mantle or core remote from the influences of plate tectonic processes and thus superimposed on plate processes, with little interaction between the two. The shallow model is along the lines of the "plate model" for the Earth proposed by Anderson (2005) and the "alternative Earth" of Hamilton (2003).

Some philosophical points regarding plate tectonic processes that are pertinent to further discussion are:

  1. Plate motion is primarily driven by slab pull. Spreading ridges, once established, can provide some driving force, but they are essentially passive features that react to spreading-direction changes by quickly re-orienting themselves to be orthogonal to the spreading direction.
  2. Volcanism can be generated almost anywhere. There is always a supply of magma available for eruption, provided a suitable channel to the surface such as a rift is created by plate forces (Favela & Anderson, 1999). This is why almost all oceanic spreading ridges spread symmetrically; there is no preferred supply direction that, if it existed, would make ridges spread asymmetrically. An impressive example of ongoing volcanism in an extensional environment is the Cenozoic volcanism of western North America. See the animations by Glazner et al. (2005) and Walker et al. (2004); over 4,000 radiometric ages in this compilation demonstrate how volcanism occurs all over the area, with little evidence for geometric patterns.
  3. Volcanism has a supply and demand balance. Most spreading ridges move along with supply (asthenospheric melt) matching demand (the volume of material required by spreading). As slab pull provides most of the plate driving force, spreading rates vary widely but for most of the world’s ridges, supply matches demand. Hess Rise could be an example of a case when supply briefly exceeded demand. An example of an extensional environment where volcanism is lacking may be the Australia-Antarctica Discordance. In this spreading ridge system south of Australia there seems to be a lack of volcanism (i.e. insufficient supply) in the spreading process (West et al., 1997).
  4. There must be flow of asthenospheric material towards ridge axes to provide new oceanic crust. Although most of this flow would be perpendicular to the ridge, i.e. bringing material in from the ridge flanks, there could be some flow along the ridge. This scenario was elegantly investigated for several triple junctions by Georgen & Lin (2002). These authors showed how triple junction geometry and spreading rates can affect axial flow in the vicinity of the triple junction. I suggest that at Shatsky Rise there was a net flow towards the triple junction that resulted in excess volcanism that created the rise.
In our supply-demand scenario, if demand changes (spreading slows or increases), supply simply changes as well. I hypothesize, though, that there can be cases where supply becomes so organized that it continues even if the spreading ridge moves away. This could be what happened at Meiji: the spreading ridge moved away in response to changing spreading geometries, but volcanism was so well-established that it continued lava generation, forming the seamount. To form the hotspot, i.e. a long-lived oversupply situation, requires that a melting center be formed that remains self-sustaining. An intriguing thought is that, if a melting center can be initially generated as a result of spreading, i.e. shallow, processes, it could eventually tap deeper mantle processes (the classic plume, for instance). This could be what happened at the bend in the Hawaii-Emperor chain – the hotspot intersected a mantle plume and changed character. It still remains to explain, though, what could generate a long-lived melting anomaly (hotspot) like Hawaii, starting with tectonically-induced excess volcanism at Meiji seamount.


I thank Gillian Foulger for the encouragement to write this note. It arose from conversations with Will Sager and John Tarduno at the AGU Chapman Conference “The Great Plume Debate: The Origin and Impact of LIPs and Hot Spots” and their insight is appreciated. I also thank Brian Bell, Ian Campbell, Gillian Foulger, Warren Hamilton, Dean Presnall, Dave Sandwell and Tony Watts for stimulating discussions. I thank Lisa Gahagan and Larry Lawver of the Plates Project, Univ. Texas Institute for Geophysics, for permission to use their magnetic lineation and fracture zone compilations. Permission from Dave Sandwell to use his gravity data in the figures, and Doug Robertson and Bob Brovey for their help in generating the plots, is gratefully acknowledged.


  • Atwater, T, 1989: Plate tectonic history of the northeast Pacific and western North America, in E.L. Winterer, D.M. Hussong and R.W. Decker, eds., The Geology of North America, Vol. N, The Eastern Pacific Ocean and Hawaii, Geol. Soc. Amer., pp. 21-70.
  • Dalrymple, G.B., M.A. Lanphere and J.H. Natland, 1980: K-Ar minimum age for Meiji Guyot, Emperor seamount chain, in E.D. Jackson et al., eds., Initial reports of the Deep Sea Drilling Project, 55, pp. 677-683.
  • Doubrovine, P.V. and J.A. Tarduno, 2004: Late Cretaceous paleolatitude of the Hawaiian hot spot: New paleomagnetic data from Detroit Seamount (ODP Site 883), Geochem. Geophys. Geosys., Q11L04, doi:10.1029/2004GC000745.
  • Georgen, J.E. and J. Lin, 2002: Three-dimensional passive flow and temperature structure beneath oceanic ridge-ridge-ridge triple junctions, Earth Planet. Sci. Lett., 204, 115-132.
  • Glazner, A.F., 2005: Animations of the NAVDAT database.
  • Gradstein, F.M., J.G. Ogg and A.G. Smith, 2004: A Geologic Time Scale 2004, Cambridge Univ. Press.
  • Hilde. T.W.C., N. Isezaki and J.M. Wageman, 1976: Mesozoic sea-floor spreading in the North Pacific, Geophysical Monograph 19, The geophysics of the Pacific Ocean basin and its margin, pp. 205-226.
  • Hilde, T.W.C., S. Uyeda and L. Kroenke, 1977: Evolution of the western Pacific and its margin, Tectonophysics, 38, 145-165.
  • Keller, R.A., M.R. Fisk and W.M. White, 2000: Isotopic evidence for Late Cretaceous plume-ridge interaction at the Hawaiian hotspot, Nature, 405, 673-676.
  • Larson, R.L. and C.G. Chase, 1972: Late Mesozoic evolution of the western Pacific Ocean, Geol. Soc. Amer. Bull., 83, 3627-3644.
  • Mahoney, J.J., R.A. Duncan, M.L.G. Tejada, W.W. Sager and T.J. Bralower, 2005: Jurassic-Cretaceous boundary age and mid-ocean-ridge-type mantle source for Shatsky Rise, Geology, 33, 185-188.
  • Mammerickx J. and G.F. Sharman, 1988: Tectonic evolution of the North Pacific during the Cretaceous quiet period, J. Geophys. Res., 93, 3009-3024.
  • Norton, I.O., 1995: Plate motions in the Norh Pacific: the 43 Ma non-event. Tectonics, 14, 1080-1094.
  • Rea, D.K. and J.M. Dixon, 1983: Late Cretaceous and Paleogene tectonic evolution of the North Pacific Ocean, Earth Planet. Sci. Lett., 65, 145-166.
  • Norton, I.O., 2006: Speculations on tectonic origin of the Hawaii hotspot, paper in preparation for G.R. Foulger and D.M. Jurdy, eds., The Origins of Melting Anomalies: Plumes, Plates, and Planetary Processes, Geol. Soc. Am. Special paper.
  • Regelous, M., A.W. Hofmann, W. Abouchami and S.J.G. Galer, 2003: Geochemistry of lavas from the Emperor Seamounts, and the geochemical evolution of Hawaiin magmatism from 85 to 42 Ma, J. Petrol., 44, 113-140.
  • Sager, W.W., K. Jinho, A. Klaus, M. Nakanishi and L.M. Khankishiyeva, 1999: Bathymetry of Shatsky Rise, northwest Pacific Ocean; implications for ocean plateau development at a triple junction, J. Geophys. Res., 104, 7557-7576.
  • Sandwell, D.T., 2005: Global gravity version 11.1.
  • Smith, A.D., 2003: A re-appraisal of stress field and convective roll models for the origin and distribution of Cretaceous to Recent intraplate volcanism in the Pacific basin, Int. Geol. Rev., 45, 287-302.
  • Tarduno, J.A. and R.D. Cottrell, 1997: Paleomagnetic evidence for motion of the Hawaiian hotspot during formation of the Emperor seamounts, Earth Planet. Sci. Lett., 153, 171-180.
  • Tarduno, J.A., R.A. Duncan, D.W. Scholl, R.D. Cottrell, B. Steinberger, T. Thordarson, B.C. Kerr, C.R. Neal, F.A. Frey, M. Torii, C. Carvallo, 2003: The Emperor seamounts: Southward motion of the Hawaiian hotspot plume in Earth’s mantle, Science, 301, 1064-1069.
  • Walker, J.D., T.D. Bowers, A.F. Glazner, G.L. Farmer, and R. Carlson, 2004: Creation of a North American volcanic and plutonic rock database (NAVDAT), Rocky Mountain (56th Annual) and Cordilleran (100th Annual) Joint Meeting (May 3–5, 2004) Paper No. 2-1
  • West, B.P., W.S.D. Wilcock, J.C. Sempéré and L. Géli, 1997: Three-dimensional structure of asthenospheric flow beneath the Southeast Indian Ridge, J. Geophys. Res., 102, 7783-7802.
  • Woods, M.T. and G.F. Davies, 1982: Late Cretaceous genesis of the Kula plate. Earth Planet. Sci. Lett., 58, 161-166.
  • Worsley, J.R. 1973: Calcareous nannofossils, Leg 19 of the Deep Sea Drilling Project, Init. Reports. DSDP 19, 741-650.

last updated 24th January, 2006