Hawaiian Swell Research Articles

On the Origin of the Hawaiian Swell: Lithosphere and Asthenosphere Seismic Structure From Rayleigh Wave Dispersion

AbstractIn this study, we revisit the shear‐wave velocity structure of the lithosphere and asthenosphere surrounding the Hawaiian hotspot and Hawaiian swell using Rayleigh wave data spanning periods of 20–125 s from the PLUME project. A primary goal of this investigation is to probe the origin of the Hawaiian swell and the mechanism that elevates the topography, providing insights into mantle dynamics beneath hotspot swells. In the shear velocity model, the 30–70 km depth range is largely featureless with weak and local anomalies, indicating that the elevation of the Hawaiian swell cannot be attributed to upper lithospheric reheating or replacement. In contrast, at 80–150 km depth, a pronounced region of anomalously low velocities is well‐resolved, with the lowest velocities found beneath the Hawaii‐Maui‐Molokai part of the island chain. Minimum shear velocities are approximately 4.0 km/s at 100–120 km depth, which is an ∼8%‐10% velocity decrease relative to the surrounding velocities away from the swell. This pattern suggests that hot, buoyant mantle from deeper plume sources laterally spread out near the top of the normal oceanic asthenosphere. We find that the low‐velocity pattern in the asthenosphere exhibits a strong correlation with the overall shape of the Hawaiian swell topography. Assuming that density anomalies are proportional to shear velocity anomalies, we demonstrate that the anomalous elevation of the swell can be explained by the uplift of a 30‐km‐thick elastic plate loaded from below by this buoyant, low‐seismic‐velocity layer in the asthenosphere.

Reprocessing of Legacy Seismic Reflection Profile Data and Its Implications for Plate Flexure in the Vicinity of the Hawaiian Islands

AbstractDuring 1975–1988, an academic research ship, R/V Robert D. Conrad, acquired more than 150,000‐line‐km of multichannel seismic reflection profile data from each of the world's main ocean basins and their margins. This extensive legacy seismic data set, which involved both single ship and two‐ship data acquisition, has been widely used by the marine geoscience community. We report on our experience in reprocessing seismic reflection profile data acquired during Conrad cruise RC2308 to the Hawaiian Islands region in August/September 1982. We show that the application of modern, industry standard processing techniques, including filtering, de‐bubble, deconvolution, and migration, can significantly enhance 40+ year old legacy seismic reflection profile data. The reprocessed data reveals more precisely, and with much less scatter, the flexure of Cretaceous Pacific oceanic crust caused by the Pliocene‐Recent volcanic loads that comprise the Hawaiian Islands. A comparison of observed picks of top oceanic crust which has been corrected for the Hawaiian swell and the Molokai Fracture Zone with the calculations of a simple 3‐dimensional elastic plate (flexure) model reveals a best fit elastic plate thickness of the lithosphere, Te, of 26.7 km, an average infill density of 2,701 kg m−3, and a Root Mean Square difference between observations and calculations of 305 m. Tests show these results depend weakly on the load density assumed and that the average infill density is close to what would be predicted from an arithmetic average of the flanking moat infill density and the infill density that immediately underlies the volcanic edifice.

Seismic Constraints on Crustal and Uppermost Mantle Structure Beneath the Hawaiian Swell: Implications for Plume‐Lithosphere Interactions

AbstractWe infer the lithospheric structure beneath the Hawaiian Swell based on a joint inversion of ambient noise and teleseismic Rayleigh waves collected during the PLUME experiment. These combined datasets let us use Rayleigh waves with a period range of 8–50 s that provides imaging resolution for the shallow lithosphere and constrains interactions between the upwelling plume and migrating plate. We find an elongated low‐velocity anomaly beneath the lithosphere along the island chain that connects to the plume conduit, consistent with the melting region associated with a restite hotspot swell root that has mechanically eroded the base of its overlying lithosphere. It could also be consistent with a plume refracted by the overriding plate motion if the plume could manage to deeply erode the lithospheric base, as only a more viscous restite root is seen to do in thermomechanical experiments. There is also a low‐velocity body beneath the North Arch that coincides with the location of recent off‐chain volcanic fields discovered there. Its location relative to the Molokai Fracture Zone (MFZ) supports the concept that swell‐root material has preferentially spread beneath the younger side of the MFZ. We also find a clear low‐velocity anomaly associated with the uppermost ∼50 km of the fracture zone lithosphere. The relatively well‐resolved shallow lithospheric structure determined here allows us to estimate the volumes of underplating and the total flux of the rising plume material that is added beneath the Hawaiian lithosphere, which we constrain to be in the range of 0.5–0.6 km3/yr.

Melt-affected ocean crust and uppermost mantle near Hawaii—clues from ambient-noise phase velocity and seafloor compliance

SUMMARYWe present models of crustal and uppermost mantle structure beneath the Hawaiian Swell and surrounding region. The models were derived from ambient-noise intermediate-period Rayleigh-wave phase velocities and from seafloor compliance that were estimated from continuous seismic and pressure recordings collected during the Hawaiian Plume-Lithosphere Undersea Mantle Experiment (PLUME). We jointly inverted these data at the locations of over 50 ocean-bottom instruments, after accounting for variations in local bathymetry and sediment properties. Our results suggest that the crystalline crust is up to 15 km thick beneath the swell and up to 23 km thick closer to the islands. Anomalously thick crust extends towards the older seamounts, downstream of Hawaii. In a second region, anomalies immediately to the south of Hawaii may be associated with the leading edge of the shallow Hawaiian magma conduit. In a third region, thickened crust to the immediate west of Hawaii may be related to Cretaceous seamounts. Low seismic velocities identified in the uppermost mantle to the northeast of Hawaii may be linked to the Molokai fracture zone and may be manifest of complex non-vertical pathways of melt through the upper lithosphere. Velocity anomalies decrease in amplitude towards the surface, suggesting that melt becomes focused into conduits at depths between 20 and 40 km that escape the resolution capabilities of our data set.

Why Is Crustal Underplating Beneath Many Hot Spot Islands Anisotropic?

AbstractHigh‐frequency harmonic regression (2.0–4.0‐Hz cutoff) of receiver functions at two long‐running seismic observatories at mid‐Pacific hot spot islands confirms earlier detections of underplated material with seismic velocities intermediate to crust and mantle, and reveals it to be multilayered and anisotropic within ~30 km of the surface. Magmatic underplating beneath the oceanic Moho has been proposed to accompany basaltic melt that erupts at the seafloor and (eventually) atop a subaerial volcano. An alternate hypothesis is “metasomatic underplating” whereby crustal fractures developed during magma ascent allow seawater to infiltrate and to serpentinize the sub‐Moho mantle partially. Metasomatic underplating would lower seismic wave speeds, promote the buoyancy of the hot spot swell, and induce textural anisotropy as metamorphic expansion of olivine‐rich peridotite promotes a crack network along which serpentinization spreads. Differential expansion of mantle peridotite and crustal gabbro promotes cracks in the crust that offer new pathways for seawater to descend to the Moho, allowing metasomatic underplating to expand laterally and to contribute anisotropy to the underplated layer. Rare serpentinized mantle xenoliths confirm that crack textures can develop during serpentinization at depth. The discovery of iron‐oxidizing microbial mats on the seafloor flank of the Loihi volcano, and many locations of diffuse low‐temperature venting worldwide, is consistent with the circulation of metasomatic fluids with reducing chemistry, sourced from serpentinization at depth. Posteruptive uplift of Santa Maria Island (Azores), and asymmetry of the Hawaiian swell, suggests that underplating requires 2–4 Myr to complete, suggesting that fluid infiltration is slow, subject to cycles of blockage and fresh fracturing.

New constraints on the origin of the Hawaiian swell from wavelet analysis of the geoid to topography ratio

Abstract Analyzing the formation mechanism of hotspot swells enhances our understanding of intraplate volcanism and the underlying geodynamical processes. The two main hypotheses for the origin of the archetypal Hawaiian swell are thermal lithospheric thinning, and dynamic support by an ascending plume. Any successful model would have to be able to simultaneously explain the swell topography and the corresponding geoid anomaly. In simple models of isostatic compensation, the geoid-to-topography ratio (GTR) is linearly related to the depth of the compensating mass; therefore it is often considered a fundamental parameter to assess swell support mechanisms. Previous estimates for the geoid-to-topography ratio (GTR) of the Hawaiian swell however are biased towards low values by incomplete removal of the effects of volcanic loading and lithospheric flexure. In order to resolve these issues, we here apply a continuous wavelet transform, which allows resolution of lateral variations of the GTR at various spatial scales. In a series of synthetic tests, the robustness of this approach and its power to identify the origin of hotspot swells are established. With 8 m/km on the youngest part of the chain, the recovered GTR agrees well with the predictions for dynamic support, therefore ruling out thermal rejuvenation as an important mechanism. We also find that the depth of the compensating mass decays by 20 km over a distance of 500 km from Hawaii to Kauai, and identify sublithospheric erosion by small-scale convection in the ponded plume material as a viable mechanism to support this decay.

Shear wave splitting at the Hawaiian hot spot from the PLUME land and ocean bottom seismometer deployments

We examine upper mantle anisotropy across the Hawaiian Swell by analyzing shear wave splitting of teleseismic SKSwaves recorded by the PLUME broadband land and ocean bottom seismometer deployments. Mantle anisotropy beneath the oceans is often attributed to flow‐induced lattice‐preferred orientation of olivine. Splitting observations may reflect a combination of both fossil lithospheric anisotropy and anisotropy due to present‐day asthenospheric flow, and here we address the question whether splitting provides diagnostic information on possible asthenospheric plume flow at Hawaii. We find that the splitting fast directions are coherent and predominantly parallel to the fossil spreading direction, suggesting that shear wave splitting dominantly reflects fossil lithospheric anisotropy. The signature of anisotropy from asthenospheric flow is more subtle, although it could add some perturbation to lithospheric splitting. The measured delay times are typically 1 s or less, although a few stations display larger splitting delays of 1–2 s. The variability in the delay times across the different stations indicates differences in the degree of anisotropy or in the thickness of the anisotropic layer or in the effect of multilayer anisotropy. Regions with smaller splitting times may have experienced processes that modified the lithosphere and partially erased the fossil anisotropy; alternatively, asthenospheric splitting may either constructively add to or destructively subtract from lithospheric splitting to produce the observed variability in delay times.

Mantle heterogeneity and off axis volcanism on young Pacific lithosphere

Plate tectonics and mantle plumes explain most volcanism on earth, but there are numerous actively forming linear volcanic chains in the middle of tectonic plates that are not explained by these theories. Using the multidisciplinary geophysical dataset of the MELT and GLIMPSE experiments, we show that associated with 3 volcanic chains west of the East Pacific Rise there are low seismic velocities and densities in the asthenosphere that extend to the East Pacific Rise spreading center. Analogous to the Hawaiian swell, the low-density anomalies produce swells beneath the volcanoes on young seafloor. The associated gravity anomalies are part of a set of gravity lineaments that have been previously interpreted as being due to thermo-elastic cracking of the lithosphere or small-scale convection. The correlation between the surface volcanism and subsurface density and velocity anomalies and their extension to the spreading center suggest that pre-existing, buoyant or fertile asthenospheric mantle heterogeneities are stretched in the direction of plate motion by shear between the plate and the underlying mantle. These heterogeneities seed small-scale convection, producing upwelling and pressure release melting, forming volcanic chains that extend nearly to the ridge axis.

Early growth of Kohala volcano and formation of long Hawaiian rift zones

AbstractTransitional-composition pillow basalts from the toe of the Hilo Ridge, collected from outcrop by submersible, have yielded the oldest ages known from the Island of Hawaii: 1138 ± 34 to 1159 ± 33 ka. Hilo Ridge has long been interpreted as a submarine rift zone of Mauna Kea, but the new ages validate proposals that it is the distal east rift zone of Kohala, the oldest subaerial volcano on the island. These ages constrain the inception of tholeiitic volcanism at Kohala, provide the first measured duration of tholeiitic shield building (≥870 k.y.) for any Hawaiian volcano, and show that this 125-km-long rift zone developed to near-total length during early growth of Kohala. Long eastern-trending rift zones of Hawaiian volcanoes may follow fractures in oceanic crust activated by arching of the Hawaiian Swell in front of the propagating hotspot.

Inferring nonlinear mantle rheology from the shape of the Hawaiian swell

The convective circulation generated within the Earth's mantle by buoyancy forces of thermal and compositional origin is intimately controlled by the rheology of the rocks that compose it. These can deform either by the diffusion of point defects (diffusion creep, with a linear relationship between strain rate and stress) or by the movement of intracrystalline dislocations (nonlinear dislocation creep). However, there is still no reliable map showing where in the mantle each of these mechanisms is dominant, and so it is important to identify regions where the operative mechanism can be inferred directly from surface geophysical observations. Here we identify a new observable quantity--the rate of downstream decay of the anomalous seafloor topography (swell) produced by a mantle plume--which depends only on the value of the exponent in the strain rate versus stress relationship that defines the difference between diffusion and dislocation creep. Comparison of the Hawaiian swell topography with the predictions of a simple fluid mechanical model shows that the swell shape is poorly explained by diffusion creep, and requires a dislocation creep rheology. The rheology predicted by the model is reasonably consistent with laboratory deformation data for both olivine and clinopyroxene, suggesting that the source of Hawaiian lavas could contain either or both of these components.

Underplating of the Hawaiian Swell: evidence from teleseismic receiver functions

SUMMARY The Hawaiian Islands are the canonical example of an age-progressive island chain, formed by volcanism long thought to be fed from a hotspot source that is more or less fixed in the mantle. Geophysical data, however, have so far yielded contradictory evidence on subsurface structure. The substantial bathymetric swell is supportive of an anomalously hot upper mantle, yet seafloor heat flow in the region does not appear to be enhanced. The accumulation of magma beneath pre-existing crust (magmatic underplating) has been suggested to add chemical buoyancy to the swell, but to date the presence of underplating has been constrained only by local active-source experiments. In this study, teleseismic receiver functions derived from seismic events recorded during the PLUME project were analysed to obtain a regional map of crustal structure for the Hawaiian Swell. This method yields results that compare favourably with those from previous studies, but permits a much broader view than possible with activesource seismic experiments. Our results indicate that the crustal structure of the Hawaiian Islands is quite complicated and does not conform to the standard model of sills fed from a central source. We find that a shallowP-to-s conversion, previously hypothesized to result from the volcano-sediment interface, corresponds more closely to the boundary between subaerial and subaqueous extrusive material. Correlation between uplifted bathymetry at ocean-bottomseismometer locations and presence of underplating suggests that much of the Hawaiian Swell is underplated, whereas a lack of underplating beneath the moat surrounding the island of Hawaii suggests that underplated crust outward of the moat has been fed from below by dykes through the lithosphere rather than by sills spreading from the island centre. Local differences in underplating may reflect focusing of magma-filled dykes in response to stress from volcanic loading. Finally, widespread underplating adds chemical buoyancy to the swell, reducing the amplitude of a mantle thermal anomaly needed to match bathymetry and supporting observations of normal heat flow.

Bathymetry of the Pacific plate and its implications for thermal evolution of lithosphere and mantle dynamics

A long‐standing question in geodynamics is the cause of deviations of ocean depth or seafloor topography from the prediction of a cooling half‐space model (HSC). Are the deviations caused entirely by mantle plumes or lithospheric reheating associated with sublithospheric small‐scale convection or some other mechanisms? In this study we analyzed the age and geographical dependences of ocean depth for the Pacific plate, and we removed the effects of sediments, seamounts, and large igneous provinces (LIPs), using recently available data sets of high‐resolution bathymetry, sediments, seamounts, and LIPs. We found that the removal of seamounts and LIPs results in nearly uniform standard deviations in ocean depth of ∼300 m for all ages. The ocean depth for the Pacific plate with seamounts, LIPs, the Hawaiian swell, and South Pacific super‐swell excluded can be fit well with a HSC model till ∼80–85 Ma and a plate model for older seafloor, particularly, with the HSC‐Plate depth‐age relation recently developed by Hillier and Watts (2005) with an entirely different approach for the North Pacific Ocean. A similar ocean depth‐age relation is also observed for the northern region of our study area with no major known mantle plumes. Residual topography with respect to Hillier and Watts' HSC‐Plate model shows two distinct topographic highs: the Hawaiian swell and South Pacific super‐swell. However, in this residual topography map, the Darwin Rise does not display anomalously high topography except the area with seamounts and LIPs. We also found that the topography estimated from the seismic model of the Pacific lithosphere of Ritzwoller et al. (2004) generally agrees with the observed topography, including the reduced topography at relatively old seafloor. Our analyses show that while mantle plumes may be important in producing the Hawaiian swell and South Pacific super‐swell, they cannot be the only cause for the topographic deviations. Other mechanisms, particularly lithospheric reheating associated with “trapped” heat below old lithosphere (Huang and Zhong, 2005), play an essential role in causing the deviations in topography from the HSC model prediction.

The relationship between depth, age and gravity in the oceans

We reassess the applicability of the thermal plate cooling model to the subsidence of the North Pacific, Atlantic and North Indian Ocean Basins. We use a new numerical plate model in which the thermophysical parameters of the lithosphere vary with temperature according to the results of laboratory experiments, and the ridge temperature structure is consistent with the thickness of the oceanic crust. We first attempt to exclude thickened crust from our data set, and then to exclude swells and downwellings by masking regions of the data that remains that have significant gravity anomalies when there exists a clear regional correlation between intermediate-wavelength gravity and topography. We find that the average variation of depth with age is consistent with conventional half-space models until about 90 Myr. Thereafter, the departure from the half-space cooling curve is more rapid than predicted using simple conductive plate cooling models. The depth–age curves in the Pacific and Atlantic show ∼250 m of temporary shallowing between the ages of 90–130 Myr, a result consistent with the outcome of experiments on the initiation of small-scale boundary layer convection. The results do not change significantly if the estimated component of the gravity arising from plate cooling is subtracted prior to calculation of the correlation between gravity and topography. A 90-km-thick conductive plate is nevertheless a reasonable model for the average temperature structure of the oldest part of the Pacific ocean lithosphere. In the Pacific, the broad topographic undulations associated with the Line Island Swell, the Hawaiian Swell and surrounding basins have correlated gravity anomalies and an admittance of approximately 30 mGal km−1 and are likely to result from convective circulation in the upper mantle. In the Northeast Atlantic, the intermediate-wavelength admittance over the Cape Verde swell is similar; in the Northwest Atlantic over the Bermuda Swell it is slightly larger but not as well constrained.

Hawaiian mantle xenoliths and magmas: Composition and thermal character of the lithosphere

Recent seismological investigations into the Hawaiian mantle and geophysical studies of the Hawaiian Swell have led to divergent models regarding the thermal structure and role of the lithosphere in Hawaiian magmatism. Whereas some models require that the Hawaiian plume erodes the lower lithosphere and replaces it with plume-derived residues and cumulates, others do not require any thinning of the lithosphere. In this paper, we develop a model for the physical and thermal characteristics of the oceanic lithosphere beneath Oahu, Hawaii, from the mantle xenoliths included in the rejuvenated stage lavas (Honolulu Volcanics, HV) that erupted through the Salt Lake Crater (SLC) and Kaau-Pali-Kalihi (KPK) vents. Three main xenolith suites are found at Oahu: spinel peridotite, garnet pyroxenite, and dunite. Dunites are generally shallow cumulates, whereas the peridotite and pyroxenite xenoliths are from the mantle. Plagioclase peridotite (harzburgite) is rare, and represents a ~15 km thick lithospheric layer below the ~15 km thick crust (MOR-generated oceanic crust + thickened crust formed by Koolau shield volcano activity). The harzburgite layer is underlain by a spinel peridotite layer. Spinel peridotites from SLC and KPK differ in chemical and isotopic composition. SLC spinel peridotites have unusually enriched 176 Hf/ 177 Hf but highly nonradiogenic 187 Os/ 188 Os isotopic compositions, likely indicating that they are metasomatized, recycled (previously subducted) oceanic lithospheric fragments. On the other hand, KPK spinel peridotites exhibit compositional characteristics of the 90 m.y. old lithosphere beneath Oahu that has been variably metasomatized by passing HV magmas. We suggest that the lower lithosphere beneath SLC and KPK are very different in their petrologic composition. The lower lithosphere (60–90 km) beneath SLC is extensively veined and fragmented by garnet clinopyroxenite intrusives and mixed with spinel peridotite blobs that are recycled subducted lithospheric fragments. For the KPK lithosphere, we favor a model in which the lower part is peridotitic. The thickness of our model lithosphere beneath Oahu is ~90 km. The seismic low-velocity zone beneath Oahu is likely due to the presence of small amounts of hydrous-alkalic and carbonatitic (+kimberlitic) melts. The isotopic data suggest little or no interaction between the basal lithosphere and Koolau shield magmas (mostly plume-derived, particularly those with near bulk-earth Nd- and Sr-isotope ratios), suggesting that during shield-stage volcanism, magmas used well-insulated (i.e., a reaction zone of 60 km) xenoliths are isotopically depleted and lack the typical Koolau-like signature. Therefore, the lower lithosphere was not significantly eroded by Koolau magmatism. KPK spinel peridotites give temperatures of 900–1040 °C for the intermediate lithosphere, which is about 200–400 °C hotter than what would be expected of a normal 90 Ma lithosphere. We suggest that such an anomalously high temperature is due to heating of the wall rocks (lithosphere) by ascending HV magmas and not due to simple basal heating of the lithosphere by the hot spot/plume followed by gradual (conductive) upward transport of heat. Thermobarometry of garnet-bearing xenoliths suggests that these rocks equilibrated at ~2–3 GPa and 1200–1350 °C. This temperature range is 50–200° less than that predicted by a recent model (Ribe and Christensen 1999). The plume that generated the Koolau shield magmas was a complex mixture composed predominantly of Koolau source materials at its center and blocks of old recycled lithosphere in its outer fringes that were later rafted into the lithosphere beneath SLC.

Time variation in igneous volume flux of the Hawaii‐Emperor hot spot seamount chain

Satellite gravity, ship track bathymetry, sediment thickness, and crustal magnetic age data were combined to calculate the residual bathymetry and residual mantle Bouguer gravity anomaly (RMBA) for the northwest Pacific Ocean. The Hawaii‐Emperor hot spot track appears on the RMBA map as a chain of negative anomalies, implying thickened crust or less dense mantle. The hot spot swell is clearly visible in a broad band of half‐width ∼500 km for about 2000 km downstream from the current hot spot location, corresponding to hot spot ages of 0–25 Ma. A much narrower expression of the hot spot is visible for the rest of the chain at hot spot ages of 25–80 Ma. Comparison of the observed RMBA with various compensation models reveals that the relatively narrow features of the Hawaii‐Emperor seamounts are best explained as being supported by plate flexure, while the Shatsky Rise, Hess Rise, and Mid‐Pacific Mountains oceanic plateaus are best fit by Airy isostasy with a thickened crustal root. Amplitude comparisons between the RMBA predictions of various compensation models and the observed RMBA for the Hawaiian swell are ambiguous. However, on the basis of the shape of the predicted anomalies, we favor a model of flexure in response to a buried load at 120 km depth. We further calculate igneous (i.e., crustal) volume flux along the axis of the Hawaii‐Emperor hot spot by integrating cross‐sectional areas of gravity‐derived excess crustal thickness and seafloor elevation, respectively, with respect to the normal oceanic crust. The highest values of the calculated igneous volume flux along the Hawaiian and Emperor ridges (∼8 m3/s) occur at present and at about 20 Ma. The flux was reduced to only 50% of this maximum (∼4 m3/s) at 10 Ma. The calculated igneous volume flux is systematically smaller (maximum values of ∼4 m3/s) along the Emperor ridge. Overall, the Hawaiian and Emperor ridges appear to have experienced quasi‐periodic variations in fluxes on timescales of 6–30 Ma. Furthermore, during the low‐flux periods at 25–48, 57, and 75 Ma the height and size of individual hot spot seamounts appear to be noticeably less than those of the high‐flux periods. We hypothesize that the variations in the fluxes of the Hawaiian ridge might be controlled by the thickness of the overlying lithosphere at the time of hot spot emplacement, while the variations along the Emperor ridge may be influenced by the dynamics of the slow absolute motion of the hot spot at the time.

Hawaiian hot-spot swell structure from seafloor MT sounding

Seafloor magnetotelluric (MT) data were collected at seven sites across the Hawaiian hot spot swell, spread approximately evenly between 120 and 800 km southwest of the Hawaiian-Emperor island chain. All data are consistent with an electrical strike direction of 300°, aligned along the seamount chain, and are well fit using two-dimensional (2D) inversion. The major features of the 2D electrical model are a resistive lithosphere underlain by a conductive lower mantle, and a narrow, conductive, ‘plume’ connecting the surface of the islands to the lower mantle. This plume is required; without it the swell bathymetry produces a large divergence of the along-strike and across-strike components of the MT fields, which is not seen in the data. The plume radius appears to be less than 100 km, and its resistivity of around 10 Ωm, extending to a depth of 150 km, is consistent with a bulk melt fraction of 5–10%. A seismic low velocity region (LVR) observed by Laske et al. [Laske, G., Phipp Morgan, J., Orcutt, J.A., 1999. First results from the Hawaiian SWELL experiment, Geophys. Res. Lett. 26, 3397–3400] at depths centered around 60 km and extending 300 km from the islands is not reflected in our inverse model, which extends high lithospheric resistivities to the edge of the conductive plume. Forward modeling shows that resistivities in the seismic LVR can be lowered at most to 30 Ωm, suggesting a maximum of 1% connected melt and probably less. However, a model of hot subsolidus lithosphere of 10 2 Ωm (1450–1500 °C) within the seismic LVR increasing to an off-swell resistivity of >10 3 Ωm (<1300 °C) fits the MT data adequately and is also consistent with the 5% drop in seismic velocities within the LVR. This suggests a ‘hot, dry lithosphere’ model of thermal rejuvination, or possibly underplated lithosphere depleted in volatiles due to melt extraction, either of which is derived from a relatively narrow mantle plume source of about 100 km radius. A simple thermal buoyancy calculation shows that the temperature structure implied by the electrical and seismic measurements is in quantitative agreement with the swell bathymetry.

Constraints on the dynamics of mantle plumes from uplift of the Hawaiian Islands

The ∼0.2 mm/yr uplift of Hawaiian islands Lanai and Molokai and Hawaiian swell topography pose important constraints on the structure and dynamics of mantle plumes. We have formulated 3-D models of mantle convection to investigate the effects of plume–plate interactions on surface vertical motions and swell topography. In our models, the controlling parameters are plume radius, excess plume temperature, and upper mantle viscosity. We have found that swell height and swell width constraints limit the radius of the Hawaiian plume to be smaller than 70 km. The additional constraint from the uplift at Lanai requires excess plume temperature to be greater than 400 K. If excess plume temperature is 400 K, models with plume radius between 50 and 70 km and upper mantle viscosity between 1020 and 3×1020 Pa s satisfy all the constraints. Our results indicate that mantle plume in the upper mantle may be significantly hotter than previously suggested. This has important implications for mantle convection and mantle melting. In addition to constraining plume dynamics, our models also provide a mechanism to produce the observed uplift at Lanai and Molokai that has never been satisfactorily explained before.

Upper mantle structure beneath the Hawaiian swell: Constraints from the ocean seismic network pilot experiment

Data from two broadband, ocean‐bottom seismographic stations deployed ∼225 km southwest of Oahu, Hawaii during the Ocean Seismic Network Pilot Experiment provide constraints on upper mantle structure beneath the Hawaiian swell. Receiver functions show that the mantle transition zone is thinned by &gt;50 km relative to reference model PA5, which, in the absence of compositional changes, implies excess temperatures of &gt;350 K in the transition zone. The combination of the measurements reported here and the thickness variations reported by Li et al. [2000] imply that the transition zone is thinned by 30 ± 15 km over an along‐swell dimension of at least 700 km. At ∼80 km depth, P‐to‐S converted phases are identified from the Gutenberg discontinuity marking the lid of the oceanic low‐velocity zone and the base of the lithosphere. Shear‐wave splitting measurements imply that fast‐polarization azimuths are intermediate between the absolute plate‐motion vector and the fossil spreading direction; multi‐event stacked values of ø and δt are −80° and 1.5 s, respectively.

Mapping the Hawaiian plume conduit with converted seismic waves

The volcanic edifice of the Hawaiian islands and seamounts, as well as the surrounding area of shallow sea floor known as the Hawaiian swell, are believed to result from the passage of the oceanic lithosphere over a mantle hotspot. Although geochemical and gravity observations indicate the existence of a mantle thermal plume beneath Hawaii, no direct seismic evidence for such a plume in the upper mantle has yet been found. Here we present an analysis of compressional-to-shear (P-to-S) converted seismic phases, recorded on seismograph stations on the Hawaiian islands, that indicate a zone of very low shear-wave velocity (< 4 km s(-1)) starting at 130-140 km depth beneath the central part of the island of Hawaii and extending deeper into the upper mantle. We also find that the upper-mantle transition zone (410-660 km depth) appears to be thinned by up to 40-50 km to the south-southwest of the island of Hawaii. We interpret these observations as localized effects of the Hawaiian plume conduit in the asthenosphere and mantle transition zone with excess temperature of approximately 300 degrees C. Large variations in the transition-zone thickness suggest a lower-mantle origin of the Hawaiian plume similar to the Iceland plume, but our results indicate a 100 degrees C higher temperature for the Hawaiian plume.

Geoid height versus topography for a plume model of the Hawaiian swell

We use a three-dimensional variable-viscosity convection model of a stationary plume beneath a drifting lithosphere to study the factors that control the geoid-to-topography ratio (GTR) of the Hawaiian swell. The rate of melting in the plume is predicted using a batch melting parameterization, and the melt is assumed to migrate to the surface where it builds a volcanic edifice, equivalent to the Hawaiian island chain. Viscous stresses, elastic deformation of the lithosphere and (optionally) the volcanic material deposited on the ocean floor are included in the calculation of surface topography and the corresponding geoid. The derivation of the GTR from the model imitates methods that have previously been used to estimate the ‘observed’ GTR for the Hawaiian swell. The first method we use here is that of Marks and Sandwell [J. Geophys. Res. 96 (1991) 8045–8055], which applies bandpass filters to retain only wavelengths from 400 to 4000 km as most characteristic of the swell topography and geoid, and the GTR is estimated from the slope of the regression line of geoid versus topography. Another group of methods analyzes the data along profiles drawn across the hotspot swell and eliminates the unwanted signals, e.g. the volcanic islands and the flexural moat around them, by cutting out parts of the sections. The GTR is then calculated from curves which best fit the topography and geoid profiles on the swell flanks only. In our plume model, when the effects of the volcanic surface loading are included, Marks and Sandwell’s method yields 4.4 m/km for the GTR, while profile-fitting on the swell flanks gives 7–8.5 m/km. Ignoring the volcanic load leads to 7–8 m/km in all processing methods. The observed GTR for the Hawaiian swell has been reported to lie between 4 and 5 m/km. Analysis of the data processing methods shows that the applied bandpass filters cannot completely remove the signal due to the volcanic edifice and lithospheric flexure and this causes apparent GTRs around 4 m/km. The ‘pure’ swell model containing no volcanic load on the surface demonstrates that the Hawaiian swell may have a proper GTR near 7 m/km.