The polarization of volcanic seismic signals: medium or source?

With the proliferation of three-component seismometers on volcanoes, the temptation is great to use polarization analysis, as we do in earthquake seismology, to determine propagation direction and/or wavetype in order to locate and characterize the wave’s source. In volcano seismology, there are two impediments to such a procedure: the complexity of the volcano’s structure and the often long-lasting volcanic seismic signals. I develop a simple model of acoustic scattering and apply it to three simple, theoretical source signals, which represent the classes of signals encountered at volcanoes. When the medium is strongly scattering, the polarization at a receiver location for impulsive, sinusoidal or square-wave sources mimics the characteristics for volcanic shocks, tornillos with a single frequency and multichromatic tremor or tornillos. The polarization observed for the square-wave source resembles that observed for multichromatic tremor at Lascar Volcano, Chile, and multichromatic tornillos at Galeras Volcano, Colombia. At low frequencies, the particle motion is fairly linear and nearly constant, but cannot necessarily be used directly to indicate the direction to the source. At all frequencies, the particle motion exhibits some characteristics of Rayleigh waves, namely ellipticity of varying degrees and an ‘RZ’ product that oscillates with twice the frequency of the mode. If this model is correct, the lack of change of fundamental wavefield parameters, such as the polarization, for low frequencies during any given tornillo and from one tornillo to the next implies that the location of the source is stable to within a resolution of about 200 m.