Observations of PKKP precursors used to estimate small-scale topography on the core-mantle boundary

Science

Volumn 277, Issue 5326, 1997, Pages 667-670

  Earle, Paul S ^a   Shearer, Peter M ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

CORE/MANTLE BOUNDARY; PKKP PHASE; TOPOGRAPHY;

ARTICLE; DYNAMICS; EARTHQUAKE; ENERGY TRANSFER; GEOLOGY; PRIORITY JOURNAL;

EID: 0030881835     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.277.5326.667     Document Type: Article

References (47)

1. A. M. Dziewonski and D. L. Anderson, Phys, Earth Planet. Inter. 25, 297 (1981).
2. D. E. Loper and T. Lay, J. Geophys. Res. 100, 6397 (1995).
3. A. Morelli and A. M. Dziewonski, Nature 325, 678 (1987).
4. J. Neuberg and J. Wahr, Phys. Earth Planet. Inter. 68, 132 (1991); A. Rodgers and J. Wahr, Geophys. J. Int. 115, 991 (1993); R. J. Pulliam and P. B. Stark, J. Geophys. Res. 98, 1943 (1993); P. B. Stark and N. W. Hengartner, ibid., p. 1957.
5. J. Neuberg and J. Wahr, Phys. Earth Planet. Inter. 68, 132 (1991); A. Rodgers and J. Wahr, Geophys. J. Int. 115, 991 (1993); R. J. Pulliam and P. B. Stark, J. Geophys. Res. 98, 1943 (1993); P. B. Stark and N. W. Hengartner, ibid., p. 1957.
6. J. Neuberg and J. Wahr, Phys. Earth Planet. Inter. 68, 132 (1991); A. Rodgers and J. Wahr, Geophys. J. Int. 115, 991 (1993); R. J. Pulliam and P. B. Stark, J. Geophys. Res. 98, 1943 (1993); P. B. Stark and N. W. Hengartner, ibid., p. 1957.
7. J. Neuberg and J. Wahr, Phys. Earth Planet. Inter. 68, 132 (1991); A. Rodgers and J. Wahr, Geophys. J. Int. 115, 991 (1993); R. J. Pulliam and P. B. Stark, J. Geophys. Res. 98, 1943 (1993); P. B. Stark and N. W. Hengartner, ibid., p. 1957.
8. W. Menke, Geophys. Res. Lett. 13, 1501 (1986); J. Vidale and H. M. Benz, Nature 359, 627 (1992); K. Bataille and F. Lund, Geophys. Res. Lett. 23, 2413 (1996).
9. W. Menke, Geophys. Res. Lett. 13, 1501 (1986); J. Vidale and H. M. Benz, Nature 359, 627 (1992); K. Bataille and F. Lund, Geophys. Res. Lett. 23, 2413 (1996).
10. W. Menke, Geophys. Res. Lett. 13, 1501 (1986); J. Vidale and H. M. Benz, Nature 359, 627 (1992); K. Bataille and F. Lund, Geophys. Res. Lett. 23, 2413 (1996).
11. D. J. Doornbos, Phys. Earth Planet. Inter. 21, 351 (1980).
12. R. A. Haddon and J. R. Cleary, ibid. 8, 211 (1974); D. J. Doornbos, Geophys. J. R. Astron. Soc. 53, 643 (1978); K. Bataille et al., Pure Appl. Geophys. 132, 151 (1990).
13. R. A. Haddon and J. R. Cleary, ibid. 8, 211 (1974); D. J. Doornbos, Geophys. J. R. Astron. Soc. 53, 643 (1978); K. Bataille et al., Pure Appl. Geophys. 132, 151 (1990).
14. R. A. Haddon and J. R. Cleary, ibid. 8, 211 (1974); D. J. Doornbos, Geophys. J. R. Astron. Soc. 53, 643 (1978); K. Bataille et al., Pure Appl. Geophys. 132, 151 (1990).
15. K. Bataille and S. M. Flatte, J. Geophys. Res. 93, 15057 (1988).
16. A. C. Chang and J.R. Cleary, Bull. Seismol. Soc. Am. 68, 1059 (1978).
17. w ≥5.5.
18. P. S. Earle and P. M. Shearer, Bull. Seismol. Soc. Am. 84, 366 (1994); L. Astiz et al., Seismol. Res. Lett. 67, 8 (1996).
19. P. S. Earle and P. M. Shearer, Bull. Seismol. Soc. Am. 84, 366 (1994); L. Astiz et al., Seismol. Res. Lett. 67, 8 (1996).
20. Including seismograms with impulsive noise bursts resulting from unrelated earthquakes or instrument error would contaminate the stacked image. Seismograms containing such artifacts were removed by visual inspection. In addition, possible contamination from double events was further avoided by removing traces with impulsive energy arriving before the theoretical P arrival. Other seismic phases, including SS, SP, and SPP, overlap a portion of the time-distance window in which the precursors were imaged. However, these phases have a different move-out and are not consistently observed at high frequencies (11) because of the high attenuation of the S phase.
21. Eighty-nine percent of the CMB is within 10° of a PKKP underside CMB reflection point. However, 83% of the reflection points are in the Southern Hemisphere, due to a concentration of seismic stations in the Northern Hemisphere.
22. E. R. Kanasewich, Time Sequence Analysis in Geophysics (Univ. of Alberta Press, Alberta, Canada, 1981).
23. We subtracted the average noise from a 60-s window starting 140 s before the PKKPbc arrival. Shifting the noise window to a time that overlaps with the precursory wave train changes the precursors' zero offset but not their shape.
24. The maximum amplitude was taken from the processed trace in a 15-s window starting at the theoretical PKKPbc arrival time.
25. The weight assigned to a processed trace was defined as the ratio of the maximum value found in a signal window to that taken in a noise window. An incorrect choice for the position of the noise window can bias the amplitude of the precursors, highlighting the importance of the P-wave test (Fig. 2C). A gap between the end of the noise window and the start of the reference phase will falsely amplify the observed precursors by ∼5%. This was verified in stacks using P as a reference phase. A false precursory wave train was imaged, for distance ranges with poor STN, when using a noise window ending 50 s before the P rival time. A noise window that overlaps the precursory time window may reduce the amplitude of the observed precursors, down-weighting seismograms with large-amplitude precursors. This effect is negligible, given that the amplitude of the noise on a single trace is up to 10 times greater than that of the precursors. Thus, we use a 140-s noise window ending 5 s before PKKPbc and a 15-s signal window starting at the PKKPbc onset time. It is also possible for the stack to be biased by an anomalously large reference phase (possibly resulting from focusing effects); this was avoided by assigning a maximum weight of 7 to each individual trace. Although the choice of noise window and weighting scheme can influence the absolute level of the observed precursors, the relative amplitude of the PKKPbc precursors with range was independent of the choice of windowing.
26. The source-receiver range for deep events was corrected to its zero depth equivalent by ray tracing from the hypocenter to the surface with the use of the PKKPbc ray parameter.
27. B. Efron and R. Tibshirani, Science 253, 390 (1991).
28. J. Tolstoy and C. S. Clay, Ocean Acoustics: Theory and Experiment in Underwater Sound (McGraw-Hill, New York, 1966); P. Scott and D. Helmberger, Geophys. J. R. Astron. Soc. 72, 237 (1983); J. A. Ogilvy, Theory of Wave Scattering from Random Rough Surfaces (Institute of Physics, London, 1991).
29. J. Tolstoy and C. S. Clay, Ocean Acoustics: Theory and Experiment in Underwater Sound (McGraw-Hill, New York, 1966); P. Scott and D. Helmberger, Geophys. J. R. Astron. Soc. 72, 237 (1983); J. A. Ogilvy, Theory of Wave Scattering from Random Rough Surfaces (Institute of Physics, London, 1991).
30. J. Tolstoy and C. S. Clay, Ocean Acoustics: Theory and Experiment in Underwater Sound (McGraw-Hill, New York, 1966); P. Scott and D. Helmberger, Geophys. J. R. Astron. Soc. 72, 237 (1983); J. A. Ogilvy, Theory of Wave Scattering from Random Rough Surfaces (Institute of Physics, London, 1991).
31. W. Kampfmann and G. Müller, Geophys. Res. Lett. 16, 653 (1989).
32. The extreme values for the grid search were λ = 5 to 30 km and σ = 100 to 1000 m. In general, increasing σ scaled the amplitude of the predicted precursory and increasing λ concentrated the energy near the PKKPbc arrival.
33. . A model of CMB topography could not be found that fit our observations within their error bounds over the entire distance range, giving rise to the possibility of additional sources of scattering. Thus, limits were placed on the model parameters by consideration of models that matched the general character of the observations. Synthetics for λ < 7 km predict a precursory wave train that does not monotonically increase with time at shorter distance ranges, which is a result of the increased contribution of ab-to-ab scattering. Synthetics calculated for λ > 10 m concentrate the observed energy near the PKKPbc arrival.
34. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
35. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
36. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
37. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
38. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
39. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
40. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
41. G. Poupinet et al., Nature 305, 204 (1983); A. Morelli et al., Geophys. Res. Lett. 13, 1545 (1986); J. H. Woodhouse et al., ibid., p. 1549; P. M. Shearer and K. M. Toy, J. Geophys. Res. 96, 2233 (1991); K. C. Creager, Nature 356, 309 (1992); P. M. Shearer, J. Geophys. Res. 99, 19647 (1994); W. J. Su and A. M. Dziewonski, ibid. 100, 9831 (1995); X.-D. Song and D. V. Helmberger, ibid., p. 9805.
42. E. R. Engdahl, thesis, Saint Louis University, St. Louis, MO (1968).
43. R. A. W. Haddon, Eos 53, 600 (1972); M. Hedlin et al., Nature 387, 145 (1997).
44. R. A. W. Haddon, Eos 53, 600 (1972); M. Hedlin et al., Nature 387, 145 (1997).
45. E. R. Engdahl, Science 161, 263 (1968).
46. P. G. Richards, Geophys. J. R. Astron. Soc. 35, 243 (1973).
47. We thank the personnel from the IRIS-GSN stations for providing the data and specifically R. Benson for maintaining the FARM archive. This work benefited from the suggestions of B. Engdahl, two anonymous reviewers, and many co-workers and was supported by NSF grant EAR93-15060.