Characterization of active and passive sources for frequency content & image quality

Date

2014-05

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Selection of source type and its parameters are very crucial for a successful seismic survey. The purpose of this study is to select optimal thumper source parameters for active seismic surveys. Many authors working on receiver function (RFs) believe that there is a magnitude of earthquakes above which the source function can be complex enough to cause interference and reduce receiver function quality. In this study, we test this by producing RFs at cratonic stations from events that are grouped by magnitude and determine if the quality of source functions is dependent on magnitude. For active seismic survey of this study, the seismic data were acquired with eight different pressure settings of the thumper source in September, 2012 in the western part of Lubbock County. 46 (vertical component) geophones with 55 foot receiver interval were deployed to record the seismic traces. There were eight jugs per array with a 55 foot array aperture. The 55 foot array was used even though the operator recommended 110 as the best of the area. In the processing of the CMP data, we combined neighboring traces that essentially doubles the effective array length. Total length of the seismic line was 2475 foot – approximately. Objectives were to choose an optimum hammer pressure and an optimum thumps per stack (TPS) for the thumper truck to have optimal seismic imaging. Series of these tests revealed that 700 PSI source provides the largest amplitudes as compared to other hammer pressures when 1 TPS is used. However, with two or higher TPS, 900 PSI source produces the largest amplitudes especially at frequencies below 20 Hz. If frequencies above 20 Hz, the 700 and 800 PSI sources are almost as strong as the 900 PSI source and for certain frequency outperform the 900PSI source. There was almost never improvement using 8 TPS as compared to 4 TPS. So, we recommended no more than 4 TPS. For the best result, we suggested combining several different PSI source in one stacked thump; this may not be possible with the current system. For the passive seismic sources study, earthquake data were acquired from IRIS for data rich cratonic stations; ARU, FFC, SAML, WRAB and QSPA for events occurring from 30 to 90 degrees between January 1st, 1995 and January 1st, 2011. In this part, the main objective was to determine if there are magnitude ranges that produce the best RFs and if there is a range that should be excluded. Previous experiences report that magnitudes below Mw6.0 do not tend to have a signal far enough above the noise level to be optimal. To analyze the events, two different receiver function production techniques were used; the frequency domain water level deconvolution (FWLD) and curved spectral deconvolution (CSD). All tests and interpretations showed that the events with magnitude between Mw≥6.3 and Mw≤6.6 provide the largest amplitudes for phases of interest from the Moho (P, Ps and PPs). But there was not a magnitude above which the quality of RFs reduced significantly to warrant their exclusion. The results also showed that the standard deviations become smaller with a larger number of samples per deconvolution. Therefore, to produce the best RFs, we suggested including all events with magnitudes between Mw6.0 and Mw7.7 (we did not have larger events at the stations tested) where including 8 to 16 (or more) events simultaneous deconvolution.

Description

Keywords

Thumper truck, Curved spectral deconvolution

Citation