MANTLE LAYERING FROM SCS REVERBERATIONS .3. THE UPPER MANTLE

被引:227
作者
REVENAUGH, J
JORDAN, TH
机构
[1] MIT, DEPT EARTH ATMOSPHER & PLANETARY SCI, CAMBRIDGE, MA 02139 USA
[2] UNIV CALIF SANTA CRUZ, EARTH SCI BOARD, SANTA CRUZ, CA 95064 USA
关键词
D O I
10.1029/91JB01487
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
This is the third paper in a four-part sequence that examines the nature of mantle layering using the multiple-ScS phases and internal reflections observed within the reverberative interval of SH-polarized seismograms. In this paper, migration techniques am applied to ScS reverberations to image discontinuities in shear impedance beneath a tectonically diverse study area that includes the western Pacific, Japan, the Philippine Sea, and Australasia. Between the M and 410-km discontinuities (Bullen's region B), the analysis reveals four reflectors, designated H, G, L, and X. The H (Hales) reflector is a positive impedance increase occurring at an average depth of about 60 km with a mean reflection coefficient of about 3.5%. It can be seen on all profiles except those where its signature is overwhelmed by the G discontinuity and is best explained as the seismic expression of the spinel --> garnet facies boundary. The G (Gutenberg) reflector, marking the lid-to-low-velocity zone (LVZ) interface, is the only negative impedance contrast identified from ScS reverberations anywhere in the mantle. It is observed along most of the oceanic paths sampled in our study, where its depth generally correlates with Q(ScS), but it is not found on any of the profiles from predominantly continental corridors, consistent with the notion that the LVZ is absent or only weakly expressed beneath the cratons. G occurs at a depth of approximately 60 km beneath the western Pacific, shallower than most estimates of thermal boundary layer (plate) thickness; the data are consistent with a sharp drop in shear velocity owing to the breakdown of hydrous phases such as amphibole along a steeply rising portion of the geotherm within the thermal boundary layer. The G contrasts with the largest magnitude (up to 10%) occur on profiles for back are regions, where upper mantle melting may be accentuated by volatiles fluxing from subducted oceanic lithosphere. None of the eight profiles having a G discontinuity show an L reflector, or Lehmann discontinuity. whereas most of the ones that lack G require an L of substantial magnitude (R0(z(L)) almost-equal-to 0.02). This positive impedance contrast is not a global feature but has a continental affinity within the study area, increasing from an average depth of about 210 km beneath the continental margin north of Australia to nearly 300 km near the center of the western Australian craton. As other investigators have noted, this discontinuity is not easily explained by plausible phase transitions or chemical changes. We propose that the L reflector represents a transition from an anisotropic mechanical boundary layer to an annealed, nearly isotropic asthenosphere within the continental tectosphere. The substantial differences between the reflectivity structures of continents and oceans documented in this study thus provide new evidence for the thick plate model of the continents. The X reflector is an enigmatic feature observed at a depth near 300 km on profiles sampling regions of active subduction. Like L, this impedance increase has been observed in refraction studies but is not readily explained by standard mantle models. Based on its limited geographic occurrence, we speculatively ascribe the X discontinuity to hydration reactions occurring in parts of the mantle that have been volatile charged by subducting lithosphere. Overall, our results indicate an upper mantle where impedance layering varies substantially over a broad spectrum of scale lengths, but correlates with surface geology and plate structure down to at least 250 km (the L discontinuity), and perhaps to depths as great as 350 km (the X discontinuity). The lateral variability observed in region B reflectivity structure appears to be governed primarily by the thermal and chemical gradients related to continental deep structure and subduction. The large magnitude of this variability explains why no consistent view of upper mantle structure has developed from global data sets, which tend to average out the region B features observed by ScS reverberations.
引用
收藏
页码:19781 / 19810
页数:30
相关论文
共 117 条
[61]  
JORDAN TH, 1988, J PETROL, P11, DOI DOI 10.1093/PETROLOGY/SPECIAL_VOLUME.1.11
[62]   HOW THICK IS LITHOSPHERE [J].
KANAMORI, H ;
PRESS, F .
NATURE, 1970, 226 (5243) :330-&
[63]   THE VELOCITY STRUCTURE AND HETEROGENEITY OF THE UPPER MANTLE [J].
KENNETT, BLN ;
BOWMAN, JR .
PHYSICS OF THE EARTH AND PLANETARY INTERIORS, 1990, 59 (03) :134-144
[64]   OBSERVATION AND INVERSION OF SURFACE-WAVE DISPERSION [J].
KNOPOFF, L .
TECTONOPHYSICS, 1972, 13 (1-4) :497-&
[65]   THE THICKNESS OF THE LITHOSPHERE FROM THE DISPERSION OF SURFACE-WAVES [J].
KNOPOFF, L .
GEOPHYSICAL JOURNAL OF THE ROYAL ASTRONOMICAL SOCIETY, 1983, 74 (01) :55-81
[66]   A LOWER MANTLE S-WAVE TRIPLICATION AND THE SHEAR VELOCITY STRUCTURE OF D'' [J].
LAY, T ;
HELMBERGER, DV .
GEOPHYSICAL JOURNAL OF THE ROYAL ASTRONOMICAL SOCIETY, 1983, 75 (03) :799-837
[67]   LITHOSPHERIC THICKNESS IN WESTERN PACIFIC [J].
LEEDS, AR .
PHYSICS OF THE EARTH AND PLANETARY INTERIORS, 1975, 11 (01) :61-64
[68]   S AND THE STRUCTURE OF THE UPPER MANTLE [J].
LEHMANN, I .
GEOPHYSICAL JOURNAL OF THE ROYAL ASTRONOMICAL SOCIETY, 1961, 4 :124-138
[69]  
Lehmann I., 1959, ANN GEOPHYS, V15, P93
[70]  
LEMERLAM AL, 1987, J GEOPHYS RES, V92, P14007