![]() Figure 5a shows a cross section of the scattering kernel K for the SPE‐2 to L4‐05 source–receiver geometry where the source depth is 45.7 m and the epicentral distance is 500 m. The diffusion constant D = 65 × 10 6 m 2 / s used to calculate the sensitivity kernel is from a study by Wesley (1965) at the Nevada Test Site, as reported by Dainty and Toksöz (1981). We model the perturbed region as a sphere, which is expected for such a deeply buried shot, with slowness perturbation that decreases to zero at the boundary. In order to make a statement on medium change, we forward model possible scenarios and compare their predictions with observations. The shift time measurement becomes stable, approximately, 3 s after the direct arrival, when the multiple scattering assumption is more valid. Figure 3 shows the measurement of t s at station L4‐05. Douma and Snieder (2006) show how to correct R max ( t, t w ) to remove the bias due to noise. In addition, noise can contaminate the late‐time coda wave signal, which can reduce the number of usable windows for analysis. Each of these windows of 2 t w length offers an independent assessment of t s, so a stable observation can be found from repeated measurement windows. Snieder (2004) shows that an appropriately large t w of about five times the dominant period of the waveform is needed to stabilize the measurement. The cross correlation attains its maximum value R max ( t, t w ) when t s = ⟨ τ ⟩ ( t, t w ), which is the mean travel‐time perturbation of the arrivals in the time window. (1) R ( t, t w ) ( t s ) = ∫ t − t w t + t w u ( t ′ ) u ^ ( t ′ + t s ) d t ′ ∫ t − t w t + t w u 2 ( t ′ ) d t ′ ∫ t − t w t + t w u ^ 2 ( t ′ ) d t ′ ,in which t s is the shift time, and R ( t, t w ) ( t s ) measures the change between the reference u and perturbed u ^ scattered velocity field over a time window of length 2 t w. This observational study will focus on applying the CWI method to SPE data, and future work will focus on detailed numerical and physical modeling of how such damage can occur. The use of CWI with the well‐studied SPE explosions offers a unique opportunity to assess its use in other experiments and at other nuclear test sites. ![]() (2011) report a velocity change of 20% in the near‐source region due to damage in granite for an experiment in Vermont that could affect shear‐wave generation ( Stroujkova et al., 2012). Modeling of the 1993 “tuna‐can” shaped Non‐Proliferation Experiment explosion by Stevens and O’Brien (2012) shows a somewhat spherical damage region, and Martin et al. Patton and Taylor (2011) attempt to quantify the contribution of late‐time damage to seismic moment and find that it is significant for certain emplacement conditions. Johnson and Sammis (2001) found that the damage radius is up to 10 times the post‐shot cavity radius and is a separate source of seismic radiation that must be accounted for. It is important to quantify the damage component of the seismic wavefield to better understand the effects of source, path, and site. We will use CWI to infer small changes in the near‐source environment of the SPE explosions due to damage and test the ability to retrieve known source location offsets. (2006) measured medium velocity perturbation due to an imposed stress change using CWI. Snieder and Hagerty (2004) measured changes in volcanic tremor location, and Grêt et al. One of the methods-coda wave interferometry (CWI)-has shown promise in finding very small changes in source and medium properties. (2008) gives a good review of differential measurement techniques. However, differential measurements of sources recorded at the same site do not suffer from these complications and can get at near‐source properties more directly ( Poupinet et al., 1996). The great difference in the waveforms is primarily due to differences in path heterogeneity, which is difficult to model. ![]() Figure 2 shows recordings of the SPE Phase I chemical explosions from this network. 1), and we employ these recordings in the analysis. All shots were recorded on a local network of short‐period seismometers (Fig.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |