Publication Date:
2019
Description:
Abstract
Seismic spectral ratios between the 3 September 2017 North Korean nuclear test (NKT2017) and four earlier Korean tests (25 May 2009, NKT2009; 12 February 2013, NKT2013; 6 January 20016, NKT2016J; and 9 September 2016, NKT2016S) are investigated. All the observed teleseismic P wave spectral ratios exhibit a unique notch at approximately 2.5 Hz, but it is not observed for spectral ratios of regional P and Lg waves. Meanwhile, for a given event pair, the network‐averaged Lg wave spectral ratio is similar to that of regional P wave, but with the explosions' source corner frequencies significantly reduced, supporting the conjecture by Fisk (2006, https://doi.org/10.1785/0120060023). We demonstrated that the observed notch of teleseismic P wave spectral ratios may be well modeled with interference between pP and P waves with an apparent pP‐P time delay ΔtpP of about 0.4 s for NKT2017, while regional P wave spectral ratios may be well fitted with source spectral ratios predicted by classical explosion source models including Mueller and Murphy (1971, https://doi.org/10.2172/4107998) (MM71), Denny and Johnson (1991, https://doi.org/10.1029/gm065p0001) (DJ91), and their two hybrids. Via the above modeling, yield sizes of the explosions and the depth of burials of NKT2009–2016S are determined as functions of NKT2017's buried depth h0. Our result indicate that, for h0 in the range of 600–1,100 m, the MM71‐related models give yield estimations about 100–300 kt for NKT2017, 3–7 kt for NKT2009, 6–15 kt for NKT2013 and NKT2016J, and 10–25 kt for NKT2016S, while yield sizes obtained by model DJ91 are much smaller. Nevertheless, for all the above source models, the obtained depths of burial estimations of NKT2009–2016S are always significantly smaller than NKT2017's depth of burial h0.
Print ISSN:
2169-9313
Electronic ISSN:
2169-9356
Topics:
Geosciences
,
Physics
Permalink
