Cataldo Godano

Wave Polarization and Location of the Source of the Explosion Quakes at Stromboli Volcano

Portable 1 Hz, three component, digital seismic stations were used to investigate the spectral content and the wave composition of the volcanic earthquakes at Stromboli, Aeolian Islands, — Southern Italy. Two successive surveys were carried out in spring 1986 and summer 1988 and more than 1000 seismograms were recorded. Two distinct phases are visible on the seismograms. The first phase is that of the first arrival and is peaked at 1 Hz. The second phase begins two seconds later and shows two peaks at 2 and 5 Hz. Two different techniques were used to investigate the properties of the wave polarization. The first is based on the pattern of the wave vector along the seismogram, while the second one analyzes the properties of the covariance matrix among the three components of the signal. Both methods lead to the same results. Linear polarization with a mainly horizontal ground particle motion is found for the first phase of the seismograms. This phase is presumed to be composed mainly of P waves coming from the source. The pattern of polarization shows a possible source area located some hundreds of meters north of the crater with linear dimensions of about several hundreds of meters. The existence of two phases on the seismograms cannot be explained using simple step pressure models as, for example, that recently developed by Crosson and Bame. The present experimental evidence has to be used as a new constraint for future development of theoretical models.

Very-Long-Period Volcanic Tremor at Stromboli, Italy

We analyze a long time–space series of Stromboli volcanic tremors. A very-low-frequency content in the range of 0.02–0.5 Hz has been found by using spectral analysis and independent component analysis. Independent component analysis is an entropy-based technique. We observe the occurrence of a component having a period of 30 sec. Polarization analysis shows that the wave field comes mainly from the crater area, well evidenced by seismometers located around the summit ring, whereas the radiation becomes increasingly scattered at stations located around the base of this volcano. Based on its apparent velocity, the 30-sec component appears to be a slow wave, related to inhomogeneities of the source and/or gas-pressure fluctuations inside the shallow plumbing system.

Strombolian volcanic activity as an intermittent phenomenon

Considering the characteristics of the Strombolian behaviour and the classical intermittency, we suggest that the well-known superposition of tremor and explosion-quakes of the volcanic activity can be described as an intermittent phenomenon. Indeed, a slightly modified non-linear map of Venkataramani et al. (Phys. Rev. Lett., 77 (1996) 5361), characterized by an intermittent behaviour, reproduces the statistical features of the experimental time series associated with the Strombolian activity, matching the distributions of the amplitude (size) and of the inter-times between two successive explosions.

Self-organized criticality and earthquake predictability

Abstract We analyse a seismic catalogue of South California to investigate the possibility of earthquake prediction using the hypothesis that the seismic events are self-organized critical phenomena. The relation found previously is valid only in a mean field approximation, but cannot be used for earthquake prediction because the time clustering of seismic events makes the definition of a standard deviation of waiting times of earthquakes impossible.

Dependence of the apparent seismic quality factor on epicentral distance: an interpretation in terms of fractal structure of the seismic medium

Abstract Following Sato (1988a, b, Pageoph, 128:43–47) we evaluated an explicit formulation of the scattered energy in the single scattering approximation for a fractal distribution of scatterers in the lithosphere. We find that by using this hypothesis it is possible to explain the widely observed lapse time dependence of Q-coda and the distance dependence of the measured direct-wave Q. On the basis of this approach we can estimate the fractal dimension of the scatterers distribution by means of a simple fit of the experimental curves Q-coda versus lapse time (or distance). We estimated the fractal dimension of the scatterer distribution for a number of regions of Italy and New Mexico, finding that it is in the range 2–3 for all of the regions.