Suchergebnisse

Abstract. The problem of detecting the occurrence of an earthquake precursor is faced in the general framework of the statistical learning theory. The aim of this work is both to build models able to detect seismic precursors from time series of different geochemical signals and to provide an estimate of number of false positives. The model we used is k-Nearest-Neighbor classifier for discriminating "no-disturbed signal", "seismic precursor" and "co-post seismic precursor" in time series relative to thirteen different hydrogeochemical parameters collected in water samples from a natural spring in Kamchachta (Russia) peninsula. The measurements collected are ion content (Na, Cl, Ca, HCO3, H3BO3), parameters (pH, Q, T) and gases (N2, CO2, CH4, O2, Ag). The classification error is measured by Leave-K-Out-Cross-Validation procedure. Our study shows that the most discriminative ions for detecting seismic precursors are Cl and Na having an error rates of 15%. Moreover, the most discriminative parameters and gases are Q and CH4 respectively, with error rate of 21%. The ions result the most informative hydrogeochemicals for detecting seismic precursors due to the peculiarities of the mechanisms involved in earthquake preparation. Finally we show that the information collected some month before the event under analysis are necessary to improve the classification accuracy.

Abstract. We report the information that in the days of the radio anomaly presented in the paper Biagi et al. (2009) an interruption of the broadcasting from the transmitter (RMC, France) happened. It remains unclear if the action resulted in a complete power off of the system, or in a reduction in the radiated power, and if this has affected France only, or every direction. Should a complete power off have occurred, the proposed pre-seismic defocusing is inexistent. Our doubts on this action are reported.

Abstract. On 6 April 2009 a strong (Mw=6.3) earthquake occurred in the Abruzzo region (central Italy). Since 1996, the intensity of CLT (f=189 kHz, Sicily, Italy), MCO (f=216 kHz, France) and CZE (f=270 kHz, Czech Republic) broadcast signals has been collected with a ten minutes sampling rate by a receiver operating in a place located about 13 km far from the epicenter. During March 2009, the old receiver was substituted with a new one able to measure, with one minute sampling rate, the intensity of five VLF signals and five LF signals radiated by transmitters located in different zones of Europe. The MCO and CZE transmitters mentioned above are included among them. From 31 March to 1 April the intensity of the MCO radio signal dropped and this drop was observed only in this signal. The possibility that the drop was connected to problems in the transmitter or in the receiver was investigated and excluded. So, the drop indicates a defocusing of the radiated signal. Since no particular meteorological situation along the path transmitter-receiver happened, the defocusing must be related to other causes, and a possibility is presented that it is a precursor of the Abruzzo earthquake.

Abstract. The VLF/LF radio signals method for studying preseimic activity is applied to the Abruzzo earthquake (M=6.3, 6 April 2009). The data collected by three receivers located in Moscow (Russia), Graz (Austria) and Bari (Italy) at about 3000 km, 1000 km and 500 km from the epicenter were used. The signals received from the Sardinia (20.27 kHz) and the Sicily (45.9 kHz) transmitters, both located in Italy, were compared with those received from the Iceland (37.5 kHz), the Great Britain (19.58 kHz) and the Germany (23.4 kHz) transmitters. The propagation paths of the two Italian transmitters cross the epicentral area (seismic paths) unlike the paths of the other three signals (control paths). Using two different analyses, that are the study of the night-time signal and the research of shifts in the evening terminator times, clear anomalies were revealed 2–8 days before the occurrence of the Abruzzo earthquake in the seismic paths, while no anomalies have been found in the control paths.

Abstract. The energy released by the earthquakes occurred in the seismogenetic area of the southern Kamchatka (Russia) from January 1977 to December 2004, reveals an increase in the peak energy in the period 1992–1999. This increase is related to the occurrence of seven earthquakes with magnitude ranging from 6.9 to 7.7; the first of these earthquakes happened on 2 March 1992 with M=7.1. For many years, hydro-geochemical data have been collected with a mean sampling rate of three days, in the form of the most common ions and gases in the water of deep wells and natural springs of a network operating in the south area of the Kamchatka, where the capital city Petropavlovsk is located. The collected data were analysed and differences in the trend and in the spectral content of some hydro-geochemical parameters at the springs were pointed out before and after the occurrence of the March 1992 earthquake, indicating clear post-seismic effects. Then an evident increase in the Carbon Dioxide and in the Hydrogen content appeared practically at each measurement site during the two-four years preceding the earthquake, so that long term precursors can be claimed. Finally, an anomalous decrease in the Carbon Dioxide content at the springs was revealed two/three months prior the earthquake, as a middle term precursor. The earthquake on March 1992 was the event nearest (90–120 km) to the measurement sites that happened in last thirty years.

Abstract. The analysis of the VLF signals radiated by ground transmitters and received on board of the French DEMETER satellite, reveals a drop of the signals (scattering spot) connected with the occurrence of large earthquakes. The extension of the "scattering spots" zone is large enough (1000–5000 km) and, probably, it increases with the magnitude of the "relative" earthquake. A possible model to explain the phenomenology, based on the acoustic gravity waves and the ionosphere turbulence, is proposed. The method of diagnostics applied to this study has the advantage to be a global one due to the world wide location of the powerful VLF transmitters and of the satellite reception. However, a specific disadvantage exists because the method requires rather a long time period of analysis due to the large longitudinal displacements among the successive satellite orbits. At the moment, at least, one month seems to be necessary.

Abstract. Starting from 1992 geochemical data are being collected with a mean sampling frequency of three days in the form of the pH value and of the most common ions and gases in the groundwater in one deep well located in Petropavlovsk, the capital city of Kamchatka (Russia). On 1 January 1996 a strong eruption started from the Karymsky volcano, that is located about 100km far from the well, in the north-northeastern direction. At the same time, a large earthquake (M=6.9) occurred in the Karymsky area. On 5 December 1997 a very large earthquake (M=7.7) occurred offshore, at a distance of 350km from the well and towards the same direction. The analysis of the geochemical data shows clear variations in the raw temporal trends on both cases. For the first event, a clear premonitory phase appeared; for the second one, some pre-seismic variations could be revealed but permanent modifications of the chemistry of the water subsequent to the earthquake are very clear. In both cases the feature of the geochemical variations is consistent with an afflux of new water in the aquifer connected with the well and with an escape of the Carbon dioxide gas from the ground in different directions. A schematic model able to justify such a phenomenology and the connections of the geochemical variations with the previous tectonic activities is proposed.

Abstract. In February 2002, a receiver was put into operation at the Department of Physics of Bari University (Southern Italy) to record VLF-LF radio signals. The intensity and the phase of the signals transmitted by GB (f=16 kHz, United Kingdom), FR (f=20.9 kHz, France), GE (f=23.4 kHz, Germany), IC (f=37.5 kHz, Island) and IT (f=54 kHz, Sicily, Italy) has been monitored with a 5 s sampling rate. The intensity raw data averaged over 10 min, from February 2002 to April 2006, have been analysed. Several decreases of the electric field intensity of the radio signals with a duration of some days were revealed, generally occurring in not concomitant periods. The GE signal decreases systematically in winter and summer each year and so, it could be supposed that such decreases are related to the transmitter. On the contrary, all the other decreases pointed out are sporadic and the previous justification does not seem realistic. On the time occurrence of these decreases, the geomagnetic activity, the meteorological conditions in the receiver area and the regional seismic activity were investigated. The main result is that, generally, a pre or post seismic effect seems to give the most convincing justification. The effect seems to appear for earthquakes with magnitude M≥4.3, when the epicentres are within the third Fresnel zone of the radio signals or near enough to some radio path.

Abstract. The results of the monitoring of three VLF/LF signals collected in Petropavlovsk station (Kamchatka, Russia) and one VLF signal collected on board of the DEMETER French satellite are presented. Two periods of the seismic activity occurred in the Japan-Kamchatka area during November–December 2004 and July–September 2005 were investigated and the earthquakes with M≥6.0 in the Japan-Kamchatka area, located inside one or more of the third Fresnel zones of the three radio paths were considered. The ground data were analysed using residual signal of phase dP or of amplitude dA, defined as the difference between the signal and the average of few quiet days (±5 days) immediately preceding or following the current day. Also the satellite data were processed by a method based on the difference between the real signal and the reference one, but in order to obtain this last signal it was necessary to construct previously a model of the signal distribution over the selected area. The method consists: (a) in averaging all the data available in the considered region over a period characterized by low level seismicity, regardless of the global disturbances, in particular, of the magnetic activity; (b) in computing a polynomial expression for the surface as a function of the longitude and the latitude. The model well describes the real data in condition of their completeness and in absence of magnetic storms or seismic forcing. In the quoted periods of seismic activity clear anomalies both in the ground and in satellite data were revealed. The influence of the geomagnetic activity cannot to be excluded, but the seismic forcing seems more probable.

Abstract. On February 1996, a receiver able to measure the electric field strength of LF radio broadcastings, with a sampling frequency of ten minutes, was put into operation in a site (AS) located in central Italy. One of the broadcasting stations selected is MCO (f=216 kHz), located in southeast France, 518 km far from the receiver. The MCO data collected since February 1996 up to September 2004 were examined and, at first, the night time data and the day time data (in winter and summer) were separated. Then, the wavelet analysis on the night and day time data was applied. The main result of the analysis was the appearance of a very clear anomaly during summer (July–August) 1998, at day time and at night time. The anomaly is a strong exaltation of the signal components with period in the 25–40 days range. Theoretical calculations of electric field strength were made and the only way to justify this anomaly seems to be the occurrence of an increase of the ground wave propagation mode of the radio signal. Such an increase could have been produced by an increase of the ground conductivity and by modifications of some parameter of the troposphere, mainly the refractive index. On 15 August 1998 a seismic sequence started with 17 earthquakes (M=2.2–4.6) on the Reatini mountains, a seismogenic zone located 30 km far from the AS receiver along the path MCO-AS. In this paper, the possibility that the previous radio anomaly can be a precursor of this seismic sequence is proposed.

Abstract. We examined the groundwater Argon content data sampled from 1988 to 2001 at two wells in Kamchatka (Russia) and anomalous increases appeared clearly during June-July 1996. On 21 June, a shallow (1km) earthquake with M=7.1 occurred at a distance less than 250km from the wells and so the previous increases could be related to this earthquake and, in particular, could be considered premonitory anomalies. In order to support this raw interpretation, we analysed the data collected in details. At first we smoothed out the high frequency fluctuations arising from the errors in a single measurement. Next we considered the known external effects on the water of a well that are the slow tectonic re-adjustment processes, the meteorology and the gravity tides and we separated these effects applying band-pass filters to the Argon content raw trends. Then we identified the largest fluctuations in these trends applying the 3 σ criterion and we found three anomalies in a case and two anomalies in other case. Comparing the time occurrence of the anomalies at the two wells we found out that a coincidence exists only in the case of the premonitory anomalies we are studying. The simultaneous appearance of well definite anomalies in the residual trends of the same parameter at two different sites supports their meaning and the possibility that they are related to some large scale effect, as the occurrence of a strong earthquake. But, other earthquakes similar to the June 1996 event took place during the Argon content measurements time and no anomaly appeared in this content. In the past, some of the authors of this paper studied the Helium content data collected in three natural springs of the Caucasus during seven years. A very similar result, that is the simultaneous appearance of clear premonitory anomalies only on the occasion of a strong (M=7.0) but shallow (2–4km) earthquake, was obtained. The correspondence with the case of the Caucasus validates the interpretation of the Kamchatkian anomalies as precursors.

Abstract. On 22 August 2005 an earthquake with magnitude M=4.7 occurred in the Anzio (central Italy) offshore area. From 2002, a VLF-LF radio receiver is into operation in Bari (southern Italy). The intensity and the phase of the signals transmitted by GB (f=16 kHz, United Kingdom), FR (f=20.9 kHz, France), GE (f=23.4 kHz, Germany), IC (f=37.5 kHz, Island) and IT (f=54 kHz, Sicily, Italy) has been monitored with a 5 s sampling rate. The previous epicenter is near enough to some of the radio paths and the data collected were studied in order to reveal possible seismic effects. The raw analysis revealed a clear drop in the intensity of the FR radio signal on 19 August. Then the wavelet analysis was applied to the intensity and the phase data of the different radio signals. In the mentioned day an increase in the band 60–120 min was revealed in the spectra of the FR signal. Then the principal component analysis was applied and again the 19 August stood up as an anomalous day for the FR radio signal. The path of this signal, among the paths of the five radio signals collected by the Bari receiver, is the nearest to the mentioned epicentre and the anomaly revealed on 19 August appears as a precursor of the earthquake. This result confirms the possible precursor revealed by other researchers in the air Rn content in a site located 5 miles far from the epicenter.

Abstract. At the beginning of 2002 an OmniPAL receiver was put into operation at the Department of Physics of Bari University (Southern Italy). The electric field strength of five VLF-LF signals transmitted from United Kingdom (f=16kHz), France (f=20.9kHz), Germany (f=23.4kHz), Iceland (f=37.5kHz) and Italy (f=54kHz) has been monitoring with a 5s sampling frequency. In a first step we reduced the amount of the data taking one datum each 10min (mean of the ±5min raw data) and then we smoothed these data by a running adjacent averaging over 7 days. Analysing the trends we obtained, we revealed at first in the signal from the Italian transmitter two clear intensity decreases in April 2002 and in August-September 2002. At these times we observed earthquakes with M=4.3 and M=5.6 respectively near the transmitter-receiver path and a precursory effect in the previous decreases appeared. Then, we noted that all of the five radio trends in the time interval March 2002-February 2003 are more disturbed than in other periods; in particular an evident simultaneous decrease appears in January-February 2003. We propose that these disturbances are related to general excitation of the margin between the African and European plates. In a second step we examined the terminator time (evening) changes for the Italian transmitter in July-September 2002, and we found significant deviations from the mean value at the end of August, which is supportive for some precursory ionospheric signature of earthquakes.