HR: 16:15h
AN: S22G-02    [PDF]
TI: The Physical Basis of Lg Generation by Explosion Sources
AU: * Stevens, J L
EM: Jeffry.L.Stevens@saic.com
AF: Science Applications International Corporation, 10260 Campus Point Drive, San Diego, CA 92121 United States
AU: Baker, G E
EM: glenn.e.baker@saic.com
AF: Science Applications International Corporation, 10260 Campus Point Drive, San Diego, CA 92121 United States
AU: Xu, H
EM: heming.xu@saic.com
AF: Science Applications International Corporation, 10260 Campus Point Drive, San Diego, CA 92121 United States
AU: Bennett, T J
EM: theron.j.bennett@saic.com
AF: Science Applications International Corporation, 10260 Campus Point Drive, San Diego, CA 92121 United States
AU: Rimer, N
EM: norton.rimer@saic.com
AF: Science Applications International Corporation, 10260 Campus Point Drive, San Diego, CA 92121 United States
AU: Day, S M
EM: day@moho.sdsu.edu
AF: San Diego State University, Dept. Geological Sciences MC-1020 5500 Campanile Dr., San Diego, CA 92182 United States
AB: We use observations of explosion-generated Lg together with numerical modeling to determine how underground nuclear explosions generate shear wave phases. This question is fundamental to how Lg phases are interpreted for use in explosion yield estimation and earthquake/explosion discrimination. We analyze several explosion data sets: 1) Degelen Mountain explosions recorded at distances less than 100 km and corresponding recordings at Borovoye (BOR) at a distance of approximately 650 km; 2) recordings from Russian deep seismic sounding experiments; 3) Nevada Test Site (NTS) explosion sources including the Nonproliferation Experiment (NPE) and nuclear tests covering a range of source depths and media properties. Observations that constrain possible models include: 1) Small Sg phases are observed at distances from less than a km to 10s of km; 2) a strong Rg phase can persist to 100s of km; 3) at 10s to 100s of km, Sg or Lg remain distinct from Rg, with no indication of scattered energy preceding Rg; 4) Sn is impulsive and large for Degelen explosions recorded at BOR; 5) Shear wave phases are larger on horizontal than vertical seismograms. We model the overburied NPE, and an underburied Degelen explosion, using point sources and two-dimensional nonlinear finite difference calculations. A small Sg distinct from large Rg near the Degelen explosion and large Lg and Sn at regional distance are consistent with the signals from a shallow, axisymmetric CLVD source. This source has an S node in the horizontal direction, and therefore makes only a small direct S wave at the near regional stations, however it is a strong generator of S at takeoff angles corresponding to the crustal phases Lg and Sn. Large Lg from the NPE can be generated from just a point explosion due to the low velocity of the source medium. We are considering three candidate mechanisms for explosion-generated Lg: 1) Direct generation by the explosion source, where the explosion is modeled as a point compressional source; 2) Secondary generation by the explosion source, where Lg is generated primarily by the nonspherical parts of the explosion source, with strong influence from the free surface; and 3) Rg scattering. The observations discussed above favor explanation number 2 in a high velocity source medium and a combination of 1 and 2 in a low velocity medium. The spherically symmetric part of an explosion source in a high velocity medium generates very little Lg, and therefore does not explain the observations. However, the observed vertical and radial shear waves can be explained by adding a CLVD, which is the lowest order non-spherical correction to the spherical source. That Rg persists to such large distances argues against explanation 3. More difficult to explain is the observed presence of Lg on the tangential components. This cannot be modeled with an axisymmetric source, and must be due to either a non-axisymmetric source effect, or to polarization changes along the source to receiver path.
DE: 7200 SEISMOLOGY
DE: 7219 Nuclear explosion seismology
SC: Seismology [S]
MN: 2003 Fall Meeting