Processing innovations to improve PP precursor analysis and increase frequency content of studies in the Mid-Pacific



Journal Title

Journal ISSN

Volume Title



PP wave data is gathered for bouncepoints primarily in the Pacific Ocean using earthquakes of magnitude 6.2 or greater occurring along the Mariana/Izu-Bonin subduction region, as well as Tonga and the Indonesia plate. The data is composed of midpoints that fall in the Mid-Pacific with a particular concentration under Hawaii. Data is recorded in seven different seismic arrays in the United States but is mainly from the Transportable Array (TA) and USArray (US) stations. The data underwent rigorous cleaning before final analysis. On top of the normal aspects, such as rotating and cutting of data, we utilized a beaming technique on both sides. On the receiver end, it’s simply called beamforming; on the source end it is called simultaneous iterative deconvolution (SID). This cleaning technique is utilized to see how well PP waves can be used in long offset mantle discontinuity studies. Of particular interest to our study are the 410, 520, and 660 discontinuities, which react differently to thermal anomalies such as subducting slabs or hotspots. The dataset had very good ray coverage around Hawaii but dispersed a bit on the outskirts of the dataset, particularly in the southern region. The double-beaming technique is applied to the PP data. Frequency content is increased to as high as 1 Hz while still getting legitimate results. SID was able to get frequencies as high as 4 Hz. 1 Hz is significantly higher than frequencies typically used in these types of studies, which is around .1 Hz. Though frequency was attempted as high as 8 Hz, it was found that PP waves had too little of these upper frequencies. The high frequencies were wiped out during waterlevel deconvolution, a method used to help stabilize the data by filling spectral nulls in frequency with white noise. The improved PP method allows interpretation of both cubes around Hawaii as well as a long line that intersects the majority of the data. Overall, the Tahiti and Hawaii hotspots are analyzed, as well as the subduction zone at the Southern Explorer Ridge (SEXP). It is found that in the Tahiti region, the 410 and 520 both deepen significantly due to the high thermal anomalies associated with hotspots. Hawaii, however, does not display these deepened horizons. This implies thermal anomalies associated with the Tahiti hotspot could be thermally stronger than the Hawaii hotspot. A double 520 can be seen in regions around Hawaii in datasets that have not been ocean corrected. Discontinuities seem stronger in non-ocean-corrected datasets overall, probably due to receiver function familiarity. Farther north, we anticipate a deepening of the 410 associated with the data line moving to continental crust, but it is difficult to decipher. We also look for a rise in this region for the subducting slab, but it is difficult to see due to resolution on the extremities of the dataset. The 410 may rise at this point, but it could possibly be an inaccurate interpretation.



Transition zone, Hawaii, Pacific, Simultaneous iterative deconvolution, Tahiti, PP precursor data