Publications

Export 4 results:
Sort by: Author Title Type [ Year  (Desc)]
2004
Sandwell, D, Fialko Y.  2004.  Warping and cracking of the Pacific plate by thermal contraction. Journal of Geophysical Research-Solid Earth. 109   10.1029/2004jb003091   AbstractWebsite

Lineaments in the gravity field and associated chains of volcanic ridges are widespread on the Pacific plate but are not yet explained by plate tectonics. Recent studies have proposed that they are warps and cracks in the plate caused by uneven thermal contraction of the cooling lithosphere. We show that the large thermoelastic stress produced by top-down cooling is optimally released by lithospheric flexure between regularly spaced parallel cracks. Both the crack spacing and approximate gravity amplitude are predicted by elastic plate theory and variational principle. Cracks along the troughs of the gravity lineaments provide conduits for the generation of volcanic ridges in agreement with new observations from satellite-derived gravity. Our model suggests that gravity lineaments are a natural consequence of lithospheric cooling so that convective rolls or mantle plumes are not required.

1996
Levitt, DA, Sandwell DT.  1996.  Modal depth anomalies from multibeam bathymetry: Is there a south Pacific superswell? Earth and Planetary Science Letters. 139:1-16.   10.1016/0012-821x(95)00247-a   AbstractWebsite

A region west of the southern East Pacific Rise (SEPR), between the Marquesas and Austral Fracture Zones has previously been found to exhibit anomalous depth-age behavior, based on gridded bathymetry and single-beam soundings. Since gridded bathymetry has been shown to be unsuitable for some geophysical analysis and since the area is characterized by unusually robust volcanism, the magnitude and regional extent of depth anomalies over the young eastern flank of the so called 'South Pacific Superswell' are re-examined using a mode-seeking estimation procedure on data obtained from several recent multibeam surveys. The modal technique estimates a representative seafloor depth, based on the assumption that bathymetry from non-edifice and edifice-populated seafloor has a low and a high standard deviation, respectively. Flat seafloor depth values are concentrated in a few bins which correspond to the mode. This method estimates a representative seafloor value even on seafloor for which more than 90% of coverage is dominated by ridge and seamount clusters, where the mean and median estimates may be shallow by hundreds of meters. Where volcanism-related bias is moderate, the mode, mean and median estimates are close. Depth-age results indicate that there is only a small anomaly (< 200 m) over 15-35 Ma Pacific Plate seafloor with little age-dependent shallowing, suggesting that the lithosphere east of the main hot-spot locations on the 'superswell' is normal. An important implication is that, in sparsely surveyed areas, depths from ETOPO-5 are significantly different from true depths even at large scales (similar to 1000 km) and thus are unsuitable for investigations of anomalies associated with depth-age regressions. We find that seafloor slopes on conjugate profiles of the Pacific and Nazca Plates from 15 to 35 Ma are both slightly lower than normal, but are within the global range. Proximate to the SEPR, seafloor slopes are very low (218 m Myr(-1/2)) on the Pacific Plate (0-22 Ma) and slightly high (similar to 410 m Myr(-1/2)) on the Nazca Plate (0-8 Ma); slopes for older Pacific seafloor (22-37 Ma) are near normal (399 m Myr(-1/2)). Seafloor slopes are even lower north of the Marquesas Fracture Zone but are highly influenced by the Marquesas Swell. We find that the low subsidence rate on young Pacific seafloor cannot be explained by a local hot-spot or a small-scale convective model exclusively and a stretching/thickening model requires implausible crustal thickness variation (similar to 30%).

1995
Sandwell, DT, Winterer EL, Mammerickx J, Duncan RA, Lynch MA, Levitt DA, Johnson CL.  1995.  Evidence for Diffuse Extension of the Pacific Plate from Pukapuka Ridges and Cross-Grain Gravity Lineations. Journal of Geophysical Research-Solid Earth. 100:15087-15099.   10.1029/95jb00156   AbstractWebsite

Satellite altimeter measurements of marine gravity reveal 100 to 200-km wavelength lineations over a wide area of the Pacific plate oriented roughly in the direction of absolute plate motion. At least three mechanisms have been proposed for their origin: small-scale convective rolls aligned in the direction of absolute plate motion by shear in the asthenosphere; diffuse N-S extension of the lithosphere resulting in lineated zones of extension (boudins); and minihotspots that move slowly with respect to major hotspots and produce intermittent volcanism. Recently, several chains of linear volcanic ridges have been found to be associated with the gravity lineations. Following ridgelike gravity signatures apparent in high-resolution Geosat gravity measurements, we surveyed a series of volcanic ridges that extend northwest from the East Pacific Rise flank for 2600 km onto 40 Ma seafloor. Our survey data, as well as radiometric dates on samples we collected from the ridges, provide tight constraints on their origin: (1) Individual ridge segments and sets of ridges are highly elongate in the direction of present absolute plate motion. (2) The ridges formed along a band 50 to 70-km-wide in the trough of one of the more prominent gravity lineations. (3) Radiometric dates of the largest ridges show no hotspot age progression. Moreover, the directions predicted for minihotspot traces older than 24 Ma do not match observed directions of either the gravity lineations or the ridges. Based on this last observation, we reject the minihotspot model. The occurrence of the ridges in the trough of the gravity lineation is incompatible with the small-scale convection model which would predict increased volcanism above the convective upwelling. We favor the diffuse extension model because it is consistent with the occurrence of ridges in the trough above the more highly extended lithosphere. However, the multibeam data show no evidence for widespread normal faulting of the crust as predicted by the model. Perhaps the fault scarps are buried under more than 30 m of sediments and/or covered by the elongated ridges. Finally, we note that if ridge-push force is much smaller than trench-pull force, then near the ridge axis the direction of maximum tensile stress must be perpendicular to the direction of absolute plate motion.

1991
Marks, KM, Sandwell DT, Vogt PR, Hall SA.  1991.  Mantle Downwelling beneath the Australian-Antarctic Discordance Zone - Evidence from Geoid Height Versus Topography. Earth and Planetary Science Letters. 103:325-338.   10.1016/0012-821x(91)90170-m   AbstractWebsite

The Australian-Antarctic discordance zone (AAD) is an anomalously deep and rough segment of the Southeast Indian Ridge between 120-degrees and 128-degrees-E. A large, negative (deeper than predicted) depth anomaly is centered on the discordance, and a geoid low is evident upon removal of a low-order geoid model and the geoid height-age relation. We investigate two models that may explain these anomalies: a deficiency in ridge-axis magma supply that produces thin oceanic crust (i.e. shallow Airy compensation), and a downwelling and/or cooler mantle beneath the AAD that results in deeper convective-type compensation. To distinguish between these models, we have calculated the ratio of geoid height to topography from the slope of a best line fit by functional analysis (i.e. non-biased linear regression), a method that minimizes both geoid height and topography residuals. Geoid/topography ratios of 2.1 +/- 0.9 m/km for the entire study area (38-degrees-60-degrees-S, 105-degrees-140-degrees-E), 2.3 +/- 1.8 m/km for a subset comprising crust less-than-or-equal-to 25 Ma, and 2.7 +/- 2.0 m/km for a smaller area centered on the AAD were obtained. These ratios are significantly larger than predicted for thin oceanic crust (0.4 m/km), and 2.7 m/km is consistent with downwelling convection beneath young lithosphere. Average compensation depths of 27, 29, and 34 km, respectively, estimated from these ratios suggest a mantle structure that deepens towards the AAD. The deepest compensation (34 km) of the AAD is below the average depth of the base of the young lithosphere (approximately 30 km), and a downwelling of asthenospheric material is implied. The observed geoid height-age slope over the discordance is unusually gradual at -0.133 m/m.y. We calculate that an upper mantle 170-degrees-C cooler and 0.02 g/cm3 denser than normal can explain the shallow slope. Unusually fast shear velocities in the upper 200 km of mantle beneath the discordance, and major-element geochemical trends consistent with small amounts of melting at shallow depths, provide strong evidence for cooler temperatures beneath the AAD.