Lateral Variations in Compressional/Shear Velocities at the Base of the Mantle
10.1126/science.284.5411.120
Crossref journal-article
American Association for the Advancement of Science (AAAS)
Science (221)
Abstract

Observations of core-diffracted P ( P diff ) and SH ( SH diff ) waves recorded by the Missouri-to-Massachusetts (MOMA) seismic array show that the ratio of compressional ( P ) seismic velocities to horizontal shear ( SH ) velocities at the base of the mantle changes abruptly from beneath the mid-Pacific ( V P / V S = 1.88, also the value predicted by reference Earth models) to beneath Alaska ( V P / V S = 1.83). This change signifies a sudden lateral variation in material properties that may have a mineralogical or textural origin. A textural change could be a result of shear stresses induced during the arrival at the core of ancient lithosphere from the northern Pacific paleotrench.

Bibliography

Wysession, M. E., Langenhorst, A., Fouch, M. J., Fischer, K. M., Al-Eqabi, G. I., Shore, P. J., & Clarke, T. J. (1999). Lateral Variations in Compressional/Shear Velocities at the Base of the Mantle. Science, 284(5411), 120–125.

Authors 7
  1. Michael E. Wysession (first)
  2. Amy Langenhorst (additional)
  3. Matthew J. Fouch (additional)
  4. Karen M. Fischer (additional)
  5. Ghassan I. Al-Eqabi (additional)
  6. Patrick J. Shore (additional)
  7. Timothy J. Clarke (additional)
References 68 Referenced 59
  1. Jordan T. H., Proc. Natl. Acad. Sci. U.S.A. 76, 4192 (1979); (10.1073/pnas.76.9.4192) / Proc. Natl. Acad. Sci. U.S.A. by Jordan T. H. (1979)
  2. 10.1029/GL013i013p01517
  3. 10.1038/33083
  4. 10.1038/386578a0
  5. Grand S. P., van der Hilst R. D., Widiyantoro S., GSA Today 7, 1 (1997). / GSA Today by Grand S. P. (1997)
  6. M. E. Wysession et al. in The Core-Mantle Boundary Region M. Gurnis M. E. Wysession E. Knittle B. A. Buffett Eds. (American Geophysical Union Washington DC 1998) pp. 273–297.
  7. E. J. Garnero J. Revenaugh Q. Williams T. Lay L. H. Kellogg ibid. pp. 319–334.
  8. T. Lay E. J. Garnero Q. Williams L. Kellogg M. E. Wysession ibid. pp. 299–318; E. J. Garnero and T. Lay J. Geophys. Res. 102 8121 (1997). (10.1029/GD028p0299)
  9. Matzel E., Sen M. K., Grand S. P., Geophys. Res. Lett. 23, 2417 (1996). (10.1029/96GL02186) / Geophys. Res. Lett. by Matzel E. (1996)
  10. J.-M. Kendall and P. G. Silver in (5) pp. 97–118.
  11. Relative to SV observed SH velocities have been either fast (24) [
  12. Vinnik L. P., Farra V., Romanowicz B., Bull. Seismol. Soc. Am. 79, 1542 (1989); / Bull. Seismol. Soc. Am. by Vinnik L. P. (1989)
  13. Vinnik L., Romanowicz B., Le Stunff Y., Makeyeva L., Geophys. Res. Lett. 22, 1657 (1995); (10.1029/95GL01327) / Geophys. Res. Lett. by Vinnik L. (1995)
  14. ] slow [
  15. Pulliam J., Sen M. K., Geophys. J. Int. 135, 113 (1998); (10.1046/j.1365-246X.1998.00612.x) / Geophys. J. Int. by Pulliam J. (1998)
  16. ] the same (9) or variable at small to intermediate spatial scales (24) [
  17. Vinnik L., Bregér L., Romanowicz B., Nature 393, 564 (1998); (10.1038/31208) / Nature by Vinnik L. (1998)
  18. Russell S. A., Lay T., Garnero E. J., ibid. 396, 255 (1999) ]. / ibid. by Russell S. A. (1999)
  19. R. W. Valenzuela and M. E. Wysession in (5) pp. 57–71.
  20. The MOMA Broadband Deployment involved the temporary installation of 18 broadband-frequency seismic sensors in a linear array between permanent stations CCM (Cathedral Caves MO) and HRV (Harvard MA). Instruments were borrowed from the PASSCAL program of the Incorporated Research Institutions for Seismology (IRIS) and data were recorded between January 1995 and April 1996. The location was chosen because it records on average more than half (53%) of large global earthquakes in the range 100° to 140° and as such it is ideal for the detection of core-diffracted waves [
  21. Wysession M. E., et al., Eos 77, 477 (1996); (10.1029/96EO00317) / Eos by Wysession M. E. (1996)
  22. ]. MOMA was unusual for seismic arrays in that it was largely designed for the purpose of examining structure at the base of the mantle.
  23. Wysession M. E., Okal E. A., Bina C. R., J. Geophys. Res. 97, 8749 (1992); (10.1029/92JB00511) / J. Geophys. Res. by Wysession M. E. (1992)
  24. Robertson G. S., Woodhouse J. H., ibid. 101, 20041 (1996); / ibid. by Robertson G. S. (1996)
  25. Bolton H., Masters G., Eos 77, F697 (1996). / Eos by Bolton H. (1996)
  26. The V P / V S ratios across the bottom 250 km of the mantle are 1.888 to 1.878 for the Preliminary Reference Earth Model (PREM) (25) and 1.875 to 1.876 for the Iasp91 model [
  27. Kennett B. L. N., Engdahl E. R., Geophys. J. Int. 105, 429 (1991)]. (10.1111/j.1365-246X.1991.tb06724.x) / Geophys. J. Int. by Kennett B. L. N. (1991)
  28. For P -velocity models with slower than average D " velocities beneath the northernmost Pacific and Alaska see (3);
  29. 10.1016/0031-9201(90)90236-Q
  30. Pulliam R. J., Vasco D. W., Johnson L. R., J. Geophys. Res. 98, 699 (1993); (10.1029/92JB01053) / J. Geophys. Res. by Pulliam R. J. (1993)
  31. Wysession M. E., Nature 382, 244 (1996); (10.1038/382244a0) / Nature by Wysession M. E. (1996)
  32. . For S -velocity models with faster than average D " velocities beneath the northernmost Pacific and Alaska see (4);
  33. Tanimoto T., Geophys. J. Int. 100, 327 (1990); (10.1111/j.1365-246X.1990.tb00688.x) / Geophys. J. Int. by Tanimoto T. (1990)
  34. 10.1029/93JB03408
  35. 10.1098/rsta.1996.0054
  36. 10.1029/96JB01306
  37. Sylvander M., Souriau A., Phys. Earth Planet. Inter. 94, 1 (1996); (10.1016/0031-9201(95)03093-X) / Phys. Earth Planet. Inter. by Sylvander M. (1996)
  38. ; A. M. Dziewonski G. Ekström X.-F. Liu in Monitoring a Comprehensive Test Ban Treaty (Kluwer Academic Dordrecht Netherlands 1996) pp. 521–550; (10.1007/978-94-011-0419-7_28)
  39. Kuo B. Y., Wu K. Y., J. Geophys. Res. 102, 11775 (1997). (10.1029/97JB00305) / J. Geophys. Res. by Kuo B. Y. (1997)
  40. 10.1029/93JB02216
  41. 10.1029/97RG02282
  42. Melt may accumulate at the base of the mantle because it may have a greater density than surrounding solids [
  43. Rigden S. M., Ahrens T. J., Stolper E. M., J. Geophys. Res. 94, 9508 (1989); (10.1029/JB094iB07p09508) / J. Geophys. Res. by Rigden S. M. (1989)
  44. ] especially if iron partitions preferentially into the liquid [
  45. McFarlane E., Drake M. J., Rubie D. C., Geochim. Cosmochim. Acta 58, 5161 (1994); (10.1016/0016-7037(94)90301-8) / Geochim. Cosmochim. Acta by McFarlane E. (1994)
  46. Ohtani E., Kato T., Ito E., Geophys. Res. Lett. 18, 85 (1991)]. (10.1029/90GL02717) / Geophys. Res. Lett. by Ohtani E. (1991)
  47. M. E. Wysession C. R. Bina E. A. Okal in Dynamics of the Earth's Deep Interior and Earth Rotation J.-L. LeMoüel et al. Eds. (American Geophysical Union Washington DC 1993) pp. 181–190.
  48. Kendall J.-M., Silver P. G., Nature 381, 409 (1996). (10.1038/381409a0) / Nature by Kendall J.-M. (1996)
  49. Using the formulation of G. P. Tandon and G. J. Weng [ Polym. Composites 5 327 (1984)] in an analysis similar to that of (9) and (19) we found that two different classes of heterogeneous inclusions could provide 0.5% anisotropy for V SH versus V SV and 2.7% for V SH versus V P without altering the matrix composition: (i) horizontal sheets of anomalously fast material (transverse isotropy) and (ii) horizontal cigar-shaped inclusions oriented perpendicular to the direction of propagation (north-south perpendicular to the paleotrench). For an example of the horizontal sheets if the inclusions have S velocities that are increased by 11% over a pyrolite composition at CMB temperature and pressure (which has K S = 601 and μ = 337) and P velocities that are increased by 5% (most of the increase is in the shear modulus) then the sheets would need to occupy 30% of the rock volume. All of the velocities including V P would increase relative to ambient conditions so an additional factor would have to reduce all velocities for V P to be slower than the global average. The inclusion of seismically slow sheets cannot explain the V SH versus V P difference without overestimating the V SH versus V SH anisotropy unless the D " matrix is chemically different from that beneath the mid-Pacific. The cigar-shaped inclusions taken here to have a 20-to-1 aspect ratio may be either fast or slow to explain the anisotropy. Elongated tubules could have either Δ V S = +50% and Δ V P = +32% occupying 13% of the rock volume or Δ V S = –25% and Δ V P = –30% with a volume of 24%. Elongated anomalies might result from the lateral shearing of D " beneath the more viscous lower mantle though the inclusion volumes and velocity differences are extremely large. With the fast-velocity tubules all velocities are increased and a mechanism (most likely chemical) is required to lower them all so that the P velocities are slower than the global average. With the slow-velocity tubules all velocities are decreased so a mechanism (most likely thermal) is required to make the S velocities faster than the global average.
  50. Lay T., Garnero E. J., Maupin V., Eos 79, F607 (1998). / Eos by Lay T. (1998)
  51. L. Stixrude in (5) pp. 83–96;
  52. 10.1126/science.270.5235.458
  53. Karato S., et al., Eos 79, F607 (1998). / Eos by Karato S. (1998)
  54. M. J. Fouch K. M. Fischer M. E. Wysession ibid. p. F617.
  55. 10.1016/0031-9201(81)90046-7
  56. M. J. Fouch K. M. Fischer M. E. Wysession T. J. Clarke in preparation.
  57. Mula A. H., Müller G., Pure Appl. Geophys. 118, 1270 (1980). / Pure Appl. Geophys. by Mula A. H. (1980)
  58. Okal E. A., Geller R. J., Bull. Seismol. Soc. Am. 69, 1039 (1979). (10.1785/BSSA0690041039) / Bull. Seismol. Soc. Am. by Okal E. A. (1979)
  59. Ellipticity corrections were made according to H. Jeffreys and K. E. Bullen [ Seismological Tables (British Association for the Advancement of Science London 1970)]. Travel-time corrections were made for mantle paths outside of the region of D " sampled by tracing the ray paths through whole-mantle tomographic models separately for P velocities (3) and S velocities (4). Corrections for receiver-side mantle anisotropy above D " were made using shear-wave splitting parameters determined from SKS and SKKS waves recorded at MOMA (26). To estimate S diff polarizations exiting the CMB we measured amplitudes of SH diff and SV diff at the nearest stations for each earthquake; these were predicted at more distant stations with the use of theoretical amplitude decay constants [
  60. Doornbos D. J., Mondt J. C., Geophys. J. R. Astron. Soc. 57, 353 (1979); (10.1111/j.1365-246X.1979.tb04783.x) / Geophys. J. R. Astron. Soc. by Doornbos D. J. (1979)
  61. ] as SV diff energy decays rapidly around the CMB and waveforms are often lost in the codae of SKKS arrivals. Synthetic time offsets for SH diff were then obtained by computing how an S diff phase with the appropriate polarization would be split by shallower mantle anisotropy with the orientation and magnitude inferred from the SKS and SKKS phases. Although the estimated S diff polarizations can only approximate actual S diff polarizations as they exit the core the SH diff time corrections are relatively insensitive to reasonable S diff polarization errors. In addition although the assumption that splitting in SKS / SKKS phases reflects anisotropy above D " may not be entirely accurate if azimuthal anisotropy exists in D " beneath the eastern United States D " contributions to SKS / SKKS splitting appear to be small if they exist at all. Predicted time shifts in P diff phases due to receiver-side mantle anisotropy above D " are small and time corrections for this effect are not applied to the data.
  62. Selecting onset arrival times common for most other seismic phases is not possible for core-diffracted waves because of the rapid loss of high frequencies during diffraction. A linear regression through the times of the wave peak maxima is the common means of determining the apparent slowness (18 27) [
  63. Mondt J. C., Phys. Earth Planet. Inter. 15, 46 (1977); (10.1016/0031-9201(77)90009-7) / Phys. Earth Planet. Inter. by Mondt J. C. (1977)
  64. Souriau A., Poupinet G., ibid. 84, 227 (1994); / ibid. by Souriau A. (1994)
  65. ] and this has been shown to be as reliable as using a multiwaveform cross-correlation [
  66. Wysession M. E., Okal E. A., Geophys. Res. Lett. 16, 1417 (1989)]. (10.1029/GL016i012p01417) / Geophys. Res. Lett. by Wysession M. E. (1989)
  67. Wysession M. E., Bartkó L., Wilson J., J. Geophys. Res. 99, 13667 (1994). (10.1029/94JB00691) / J. Geophys. Res. by Wysession M. E. (1994)
  68. We thank the many people with the IRIS PASSCAL and DMC programs who helped with the MOMA experiment and K. Koper A. Li E. Roth L. Salvati R. Valenzuela and J. Zaslow for additional help with the MOMA deployment. Supported by NSF grants EAR-9319324 EAR-9315971 and EAR-9315925 and by the David and Lucile Packard Foundation.
Dates
Type When
Created 23 years, 1 month ago (July 27, 2002, 5:40 a.m.)
Deposited 1 year, 7 months ago (Jan. 13, 2024, 12:25 a.m.)
Indexed 1 year, 2 months ago (July 4, 2024, 3:50 a.m.)
Issued 26 years, 5 months ago (April 2, 1999)
Published 26 years, 5 months ago (April 2, 1999)
Published Print 26 years, 5 months ago (April 2, 1999)
Funders 0

None

@article{Wysession_1999, title={Lateral Variations in Compressional/Shear Velocities at the Base of the Mantle}, volume={284}, ISSN={1095-9203}, url={http://dx.doi.org/10.1126/science.284.5411.120}, DOI={10.1126/science.284.5411.120}, number={5411}, journal={Science}, publisher={American Association for the Advancement of Science (AAAS)}, author={Wysession, Michael E. and Langenhorst, Amy and Fouch, Matthew J. and Fischer, Karen M. and Al-Eqabi, Ghassan I. and Shore, Patrick J. and Clarke, Timothy J.}, year={1999}, month=apr, pages={120–125} }