The three longest-lived volcanic hotspot chains on the Pacific plate - Hawaii, Louisville, and Rurutu - record mantle plumes that moved relative to one another over the last ~80 million years. Tomography-based plume-motion models reproduce the first-order inter-hotspot distances only when each plume is rooted in the lowermost mantle: they miss the rapid southward motion of the Hawaiian plume at 60-50 Ma, fail to reproduce shorter-timescale changes in inter-hotspot distance (that is, inter-hotspot distance variations on time scales shorter than the first-order ~80-0 Ma trends, most specifically the sharp southward drift of the Hawaiian plume between about 60 and 50 Ma), and generally fit the observations better when plume roots are allowed to move with lower-mantle flow than when the roots are held fixed. The Rurutu chain also lacks age control between 48 and 10 Ma, and its apparent rapid 66-72 Ma motion along the Gilbert Ridge may be an artifact of preferentially sampling late-stage volcanism. The problem remains open. Work from 2019-2026 confirmed the ~47 mm/yr southward Hawaiian drift, partially bridged the Rurutu age gap, and produced competing quantitative explanations (ridge-plume interaction, slab-driven flow, plume-lithosphere interaction, longitudinal hotspot drift, plate-motion change, and a revived true-polar-wander episode), but no consensus mechanism exists, the mobility history of plume roots and large low shear velocity provinces remains unquantified over the relevant timescales, no model reproduces the full short-timescale inter-hotspot record, and the Gilbert Ridge sampling-bias hypothesis has not been tested.

Background

Volcanic hotspots are long-lived sites of intraplate volcanism fed by mantle plumes - columns of hot, buoyant rock rising from deep in Earth's mantle. As an oceanic plate moves over a plume, the plume leaves a chain of volcanoes (seamounts and islands) that youngs toward the currently active volcano; this age progression is the primary record of both plate motion above and plume motion below. Because hotspots were long assumed to be nearly stationary in the mantle, seamount chains have been used to define a 'hotspot reference frame' for absolute plate motion (APM) models, which describe plate motion relative to the deep mantle.

The northwestern Pacific holds the archetypal test case. The Hawaiian-Emperor chain shows a prominent ~60-degree change in trend, the Hawaiian-Emperor Bend (HEB), formed about 47-50 million years ago (Ma). One school attributes the bend to an abrupt change in Pacific plate motion (Sharp and Clague, 2006); paleomagnetic work on Emperor Seamounts drill cores instead indicates that the Hawaiian plume itself drifted southward by roughly 15 degrees between 80 and 50 Ma relative to Earth's spin axis (Tarduno et al., 2009), while the Louisville hotspot in the South Pacific shows little latitudinal motion over the same interval. Geodynamic modeling (Hassan et al., 2016) suggested that focused subduction along the Aleutian arc drove a rapid 'burst' of Hawaiian hotspot motion through interaction of plate tectonics and deep mantle flow. Deciding how much of the bend and of the observed paleolatitude changes reflects plate motion change, plume drift, or true polar wander (TPW - a reorientation of the whole solid Earth relative to the rotation axis) remains unresolved.

Konrad et al. (2018) added a third, independent constraint: the newly mapped Rurutu hotspot chain, identifiable across the mid-Pacific by a distinctive HIMU isotopic fingerprint (HIMU denotes mantle sources with high U/Pb ratios, expressed geochemically as 206Pb/204Pb at or above about 20 and low 87Sr/86Sr at or below about 0.704), traced via new 40Ar/39Ar incremental-heating ages (a potassium-argon-based radioisotopic dating method applied to dredged seamount basalts) from the Gilbert Ridge and Tuvalu chains (about 72-47 Ma) into the young Cook-Austral islands (10-0 Ma). Comparing the three chains' inter-hotspot distances through time, and testing thousands of plume-motion simulations, they found: (i) the Hawaiian plume is unique in its rapid motion at 60-50 Ma, while Rurutu and Louisville moved roughly in tandem; (ii) models reproduce first-order inter-hotspot distances only if plumes are rooted near the core-mantle boundary (CMB, at about 2900 km depth) rather than at 660 km depth, and represent inter-hotspot motions generally better when the roots are advected by mantle flow than when the roots are kept at fixed locations - implying that the margins of the large low shear velocity provinces (LLSVPs, the two continent-sized, seismically slow regions at the base of the mantle beneath Africa and the Pacific, from which many plumes appear to rise) may not be stable through time; and (iii) the models cannot reproduce shorter-timescale changes in inter-hotspot motion.

The components of this problem that remain unresolved are as follows. The mechanism of the sharp 60-50 Ma southern drift of the Hawaiian plume is not captured by current tomography-based models and likely involves regional processes such as plume-ridge capture or slab-induced flow. The Rurutu chain has a gap in age coverage between about 48 and 10 Ma, limiting characterization of that plume's motion. The apparent rapid motion of the Rurutu plume along the Gilbert Ridge between 66 and 72 Ma (originally documented by Koppers et al., 2007, as strongly nonlinear 40Ar/39Ar age systematics) may be an artifact of preferentially sampling late-stage or rejuvenated volcanism rather than true plume motion. The mobility of deep mantle structure (LLSVPs and plume roots) over tens of millions of years is left poorly constrained by the paper's inference that LLSVP margins may not be stable through time. The fast 60-50 Ma motion may involve regional processes that are "not fully resolved in the tomographic models used in our calculations", and most of the tomography and viscosity models tested cannot reproduce the shorter-timescale changes in inter-hotspot motion.

Problem Statement

Determine, dynamically and observationally, what controls the relative motions of the three longest-lived Pacific mantle plumes - the Hawaiian, Louisville, and Rurutu plumes - as recorded by their seamount chains from about 80 Ma to the present, and produce a quantitative account of those motions tested against the available observational record. Konrad et al. (2018) established that tomography-based plume-motion models reproduce only the first-order inter-hotspot distance histories; the unresolved components of this question are:

(1) Mechanism of the rapid Hawaiian plume motion: identify and quantify the dynamical mechanism or combination of mechanisms - plume-ridge capture, slab-induced mantle flow, plume-lithosphere interaction, migration of the plume root within or along the Pacific large low shear velocity province, true polar wander, and/or changes in Pacific plate motion - that produced the rapid southward (and any concurrent longitudinal) motion of the Hawaiian plume between approximately 60 and 50 Ma, which the current tomography-based models miss, and reconcile it with the observed age progressions, chain geometries, and paleolatitudes.

(2) Mobility of plume roots and LLSVPs: establish whether plume roots and the large low shear velocity provinces are fixed, passively advected by lower-mantle flow, or actively migrating over timescales of tens of millions of years, and at what rates, given that models with mobile roots generally represent the observed inter-hotspot distances better than fixed-root models.

(3) Observational limits and discrimination: the Rurutu chain cannot at present be mapped across its 48-10 Ma gap between the Tuvalu segment (~64-47 Ma) and the Cook-Austral segment (10-0 Ma), for lack of sampling and datable material, and it remains uncertain whether the apparent rapid motion of the Rurutu plume along the Gilbert Ridge between 66 and 72 Ma reflects true plume migration or preferential sampling of late-stage and rejuvenated volcanism. On the data available today both alternatives remain open.

An account of the three chains' relative motions over ~80 Ma must identify and quantify the dynamical mechanism or combination of mechanisms responsible for the rapid southward (and any concurrent longitudinal) motion of the Hawaiian plume at 60-50 Ma, establish whether the plume roots and large low shear velocity provinces are fixed, passively advected, or actively migrating and quantify that regime and its rates over the relevant timescales, reproduce the observed inter-hotspot distance histories among the three chains, including the shorter-timescale changes that current models cannot reproduce, within the uncertainties of the observed histories, and be consistent with the independent paleomagnetic, geochronological, and geochemical constraints, taking account of the observational limitations and unresolved sampling questions noted in (3).

References

  1. Konrad, K., Koppers, A. A. P., Steinberger, B., Finlayson, V. A., Konter, J. G., Jackson, M. G. (2018). On the relative motions of long-lived Pacific mantle plumes. Nature Communications 9, 854. doi:10.1038/s41467-018-03277-x. https://doi.org/10.1038/s41467-018-03277-x
  2. Koppers, A. A. P., Staudigel, H., Phipps Morgan, J., Duncan, R. A. (2007). Nonlinear 40Ar/39Ar age systematics along the Gilbert Ridge and Tokelau Seamount Trail and the timing of the Hawaii-Emperor Bend. Geochemistry, Geophysics, Geosystems 8, Q06L13. doi:10.1029/2006GC001489. https://doi.org/10.1029/2006GC001489
  3. Sharp, W. D., Clague, D. A. (2006). 50-Ma Initiation of Hawaiian-Emperor Bend Records Major Change in Pacific Plate Motion. Science 313, 1281-1284. doi:10.1126/science.1128489. https://doi.org/10.1126/science.1128489
  4. Tarduno, J., Bunge, H.-P., Sleep, N., Hansen, U. (2009). The Bent Hawaiian-Emperor Hotspot Track: Inheriting the Mantle Wind. Science 324, 50-53. doi:10.1126/science.1161256. https://doi.org/10.1126/science.1161256
  5. Hassan, R., Müller, R. D., Gurnis, M., Williams, S. E., Flament, N. (2016). A rapid burst in hotspot motion through the interaction of tectonics and deep mantle flow. Nature 533, 239-242. doi:10.1038/nature17422. https://doi.org/10.1038/nature17422
  6. Finlayson, V. A., Konter, J. G., Konrad, K., Koppers, A. A. P., Jackson, M. G., Rooney, T. O. (2018). Sr-Pb-Nd-Hf isotopes and 40Ar/39Ar ages reveal a Hawaii-Emperor-style bend in the Rurutu hotspot. Earth and Planetary Science Letters 500, 168-179. doi:10.1016/j.epsl.2018.08.020. https://doi.org/10.1016/j.epsl.2018.08.020
  7. Bono, R. K., Tarduno, J. A., Bunge, H.-P. (2019). Hotspot motion caused the Hawaiian-Emperor Bend and LLSVPs are not fixed. Nature Communications 10, 3370. doi:10.1038/s41467-019-11314-6. https://doi.org/10.1038/s41467-019-11314-6
  8. Wessel, P., Conrad, C. P. (2019). Assessing Models for Pacific Absolute Plate and Plume Motions. Geochemistry, Geophysics, Geosystems 20, 6016-6032. doi:10.1029/2019GC008647. https://doi.org/10.1029/2019GC008647
  9. Rose, J., Koppers, A. A. P. (2019). Simplifying Age Progressions within the Cook-Austral Islands using ARGUS-VI High-Resolution 40Ar/39Ar Incremental Heating Ages. Geochemistry, Geophysics, Geosystems 20, 4756-4778. doi:10.1029/2019GC008302. https://doi.org/10.1029/2019GC008302
  10. Arnould, M., Coltice, N., Flament, N., Mallard, C. (2020). Plate tectonics and mantle controls on plume dynamics. Earth and Planetary Science Letters 547, 116439. doi:10.1016/j.epsl.2020.116439. https://doi.org/10.1016/j.epsl.2020.116439
  11. Jiang, Z., Li, S., Liu, Q., Zhang, J., Zhou, Z., Zhang, Y. (2021). The trials and tribulations of the Hawaii hot spot model. Earth-Science Reviews 215, 103544. doi:10.1016/j.earscirev.2021.103544. https://doi.org/10.1016/j.earscirev.2021.103544
  12. Hu, J., Gurnis, M., Rudi, J., Stadler, G., Müller, R. D. (2021). Dynamics of the abrupt change in Pacific Plate motion around 50 million years ago. Nature Geoscience 15, 74-78. doi:10.1038/s41561-021-00862-6. https://doi.org/10.1038/s41561-021-00862-6
  13. Buff, L., Jackson, M. G., Konrad, K., Konter, J. G., Bizimis, M., Price, A., Rose-Koga, E. F., Blusztajn, J., Koppers, A. A. P., Herrera, S. (2021). Missing links for the long-lived Macdonald and Arago hotspots, South Pacific Ocean. Geology 49, 541-544. doi:10.1130/G48276.1. https://doi.org/10.1130/G48276.1
  14. Chase, A., Wessel, P. (2022). Analysis of Pacific Hotspot Chains. Geochemistry, Geophysics, Geosystems 23, e2021GC010225. doi:10.1029/2021GC010225. https://doi.org/10.1029/2021GC010225
  15. Xie, S., Cao, Z., Liu, L., Yang, D., Liu, M., Li, Y., Qi, R. (2024). The role of plume-lithosphere interaction in Hawaii-Emperor chain formation. Nature Communications 15, 6571. doi:10.1038/s41467-024-51055-9. https://doi.org/10.1038/s41467-024-51055-9
  16. Zhang, J., Hu, J. (2025). Dynamics of longitudinal Hawaiian hotspot motion and the formation of the Hawaiian-Emperor Bend. Proceedings of the National Academy of Sciences 122, e2510972122. doi:10.1073/pnas.2510972122. https://doi.org/10.1073/pnas.2510972122
  17. Müller, R. D. (2025). The moving mantle beneath Hawaii: A new look at an old bend. Proceedings of the National Academy of Sciences 122, e2524801122. doi:10.1073/pnas.2524801122. https://doi.org/10.1073/pnas.2524801122
  18. Konter, J. G., Finlayson, V. A., Konrad, K., Jackson, M. G., Koppers, A. A. P., Wessel, P., Beethe, S., Bizimis, M., Alverson, A., Kelley, C. (2025). Pacific hotspots reveal a Louisville-Ontong Java Nui tectonic link. Nature 641, 388-394. doi:10.1038/s41586-025-08889-0. https://doi.org/10.1038/s41586-025-08889-0
  19. Woodworth, D., Gordon, R. G., Gaastra, K. M. (2026). Changes in Latitude of Pacific Hotspots From ~56 Ma to ~48 Ma Inferred From Pacific Plate Apparent Polar Wander Before and After Chron C24r: Implications for True Polar Wander. Journal of Geophysical Research: Solid Earth 131, e2026JB033790. doi:10.1029/2026JB033790. https://doi.org/10.1029/2026JB033790
  20. Gaastra, K. M., Gordon, R. G., Mifflin, G. (2026). Geologically Current Rates of Hotspot Motion. Geophysical Research Letters 53, e2026GL122416. doi:10.1029/2026GL122416. https://doi.org/10.1029/2026GL122416