Volume 1 , Issue 1 , January 2016 , Pages 445-458
Ghassan I. Aleqabi 1 ; Hafidh A. A. Ghalib 2 ; Michael E. Wysession 3
1 Washington University in Saint Louis, Campus Box 1169, One Brookings Drive, Saint Louis, MO 63130, USA,
2 Array Information Technology, 5130 Commercial Drive, Suite B, Melbourne, FL 32940, USA,
3 Washington University in Saint Louis, Campus Box 1169, One Brookings Drive, Saint Louis, MO 63130, USA
The shear wave (S-wave) velocity structure of the crust and its effect on seismic-wave
propagation are of fundamental interest in many geophysical studies. S-wave velocities
and layer thicknesses have a dominating influence on the waveforms of Rayleigh waves.
Rayleigh wave waveforms from the Northern Iraq Seismographic Network (NISN) are
inverted to constrain the crustal and upper mantle velocity structure in Iraq and
surrounding regions. The inversion uses a niching genetic algorithm (NGA), which
optimizes four parameters of earth properties at different frequencies and in different
sub-populations: P-wave velocity, S-wave velocity, density, and Earth layer thicknesses.
Each subpopulation separately performs the genetic algorithm processes of selection,
crossover, and mutation of the velocity models, with the niching function removing
models that are too similar to maintain distinct subpopulations. Using both observed
earthquake surface waves waveforms and synthetic forward-modeled waveforms; we
obtain regional S-wave profiles for the crust (including top-most sedimentary layers),
and the upper mantle. We demonstrate that the NGA method is a robust means of
interpreting observed surface-wave waveform data.