EARTHQUAKE  FORECASTING  REPORTS

Latest  Update:   June 22, 2017

Earthquake Forecasting Breakthroughs     Earthquake Forecasting Data

Volume 4, Number 3, September 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

Volume 4, Number 2, June 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

Volume 3, Number 3, September 2015. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

       The actual headings for the various reports are:

Volume 4, Number 3, September 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

Volume 4, Number 3, September 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

From the Editor Central Italy earthquake in August 2016 and its precursors…….........352

Valentino Straser, Gabriele Cataldi and Daniele Cataldi………......473

Jetstream anomalies appeared prior to the M6.2 Italy earthquake on 24 August 2016, Hong-Chun Wu…….…478

Latent heat anomalies prior to the Amatrice, Italy M6.2 Earthquake, Ariel R. Césped………..…..480

Relative humidity and OLR as pre-earthquake signals – A study of Central Italy earthquakes (August 24, 2016),

Natarajan Venkatanathan and Rubidha Devi Duraisamy…………………...482

Time-dependent neo-deterministic seismic hazard scenarios: Preliminary report on the M6.2 Central Italy earthquake,

24^th August 2016, Antonella Peresan, Vladimir Kossobokov, Leontina Romashkova, Andrea Magrin, Alexander Soloviev

and Giuliano F. Panza……………….....487

Volume 4, Number 2, June 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

Subionospheric VLF propagation anomaly prior to the Kumamoto Earthquake in April, 2016, Masashi Hayakawa

and Tomokazu Asano………….…273

Anomalies in jet-streams prior to the M6.6 Taiwan Earthquake on 5 February 2016 and the M7.0 Kumamoto

Earthquake on 15 April 2016, Hong-Chun Wu…………….276

Cataldi and Valentino Straser……………...279

The April 2016 M7.0 Kumamoto Earthquake swarm: Geology, thermal energy transmigration, and precursors,

Fumio Tsunoda and Dong R. Choi………………286

Volume 3, Number 3, September 2015. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

Special papers: The 16 September 2015 M8.3 Chile Earthquake

Analysis of psychrometric parameters associated with seismic precursors in Central Chile. Ariel R. CÉSPED….383

Blot’s energy transmigration law and the September 2015 M8.3 Coquimbo Earthquake, Chile. Dong R. CHOI and

John CASEY…………....387

A surge and short-term peak in northern solar polar field magnetism prior to the M8.3 earthquake near Chile on

September 16, 2015. Ben DAVIDSON……….....391

Solar wind ionic and geomagnetic variations preceding the M8.3 Chile Earthquake. Valentino STRASER, Gabriele

CATALDI and Daniele CATALDI……….394

Outgoing longwave radiation anomaly prior to big earthquakes: A study on the September 2015 Chile Earthquake.

N. VENKATANATHAN, Philip PHILIPOFF and S. MADHUMITHA………400

Space weather conditions prior to the M8.3 Chile Earthquake, Kongpop U-YEN….…...405

Anomalies in jet streams that appeared prior to the 16 September 2015 M8.3 Chile Earthquake. Hong-Chun WU…..407

Volume 4, Number 3, September 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

From the Editor Central Italy earthquake in August 2016 and its precursors….….........352

SELF and VLF electromagnetic signal variations that preceded the Central Italy earthquake on August 24, 2016,

Valentino Straser, Gabriele Cataldi and Daniele Cataldi…………......473

Jetstream anomalies appeared prior to the M6.2 Italy earthquake on 24 August 2016, Hong-Chun Wu..…478

Latent heat anomalies prior to the Amatrice, Italy M6.2 Earthquake, Ariel R. Césped…...…..480

Relative humidity and OLR as pre-earthquake signals – A study of Central Italy earthquakes (August 24, 2016),

Natarajan Venkatanathan and Rubidha Devi Duraisamy…………...482

Time-dependent neo-deterministic seismic hazard scenarios: Preliminary report on the M6.2 Central Italy earthquake,

24^th August 2016, Antonella Peresan, Vladimir Kossobokov, Leontina Romashkova, Andrea Magrin, Alexander Soloviev

and Giuliano F. Panza……….....487

FROM THE EDITOR

Central Italy earthquake in August 2016 and its precursors

We are glad to launch here another NCGT issue jam-packed with leading-edge articles, information and lively discussions. It also contains articles on a topical event, the latest Amatrice earthquake in Central Italy, which claimed several hundreds of fatalities.We welcome several contributions by Italian teams on this disaster.

The Amatrice earthquake is significant from various points of view:1) it was a repeat of the M6.3 L’Aquila earthquake (2009) with a similar magnitude, number of casualties, and geology of the epicenter, and 2)although its magnitude was moderate (M6.2), retrospective analysis by a team of international seismologists disclosed plenty of precursors, which must be documented and referred to in forecasting future devastating earthquakes in the region and in other parts of the world.

Though little known, this quake had been warned well in advance by Wu (p. 478-479), who reported a jetstream anomaly near the epicenter seven months prior to the Amatrice earthquake. The Editor himself who has intensively watched the region since last year found indisputable evidence of the seismo-tectonic link between the southwestern Aegean Sea (Sea of Crete) and the entire Italian peninsula. Furthermore, though supporting studies such as tomographic images are needed, he could link the three devastating Italian quakes (1997, 2009 and 2016) with deeper precursory quakes in the Sea of Crete,based on the shallow Earth thermal energy transmigration concept. The following list summarizes the time of appearance of various earthquake precursory signals, which are documented in this NCGT issue or in unpublished internal notes.

* 1) 26 Jan. 2014, M6.0, Lat. 38.21, Long 20.43, H=19km, 2) 3 Feb. 2014. M6.1, Lat. 38.29, Long. 20.34, H= 14 km. Further forerunners of these quakes are sought in deeper quakes (36-60 km) in the northern coast of the Crete Island in 2011 and 2013.

The above facts demonstrate that if an integrated monitoring program using multi-parameter precursors had been in place, the chances of issuing a warning for a quake in the general area of Amatrice would have increased significantly.

Because of the aftermath of the 2009 L’Aquila earthquake in which seven public officials were convicted, Italian earthquakes are still a touchy issue for world seismological communities,possibly discouraging companies with a proven track record for earthquake prediction from going anywhere near Italy. However, according to Alexander (Natural Hazards, 2014. DOI 10.1007/s1 1069-014-1062-2), the court decision was made on the basis of “incomplete, imprecise and contradictory information on the nature, causes, dangers and future development of seismic activity in the area in question”. Clearly some of the responsibilities lie on our side; we could not correctly understand the significance of medium and short-term precursory phenomena and earthquake generation mechanisms in general. On the other hand, as Gregori rightly puts (p. 470 of this issue), because “earthquakes will never be predicted with absolute certainty, it is very important, however, that legislators enact a law that considers the concrete possibility of an eventually false short-term alert with no criminal charge for people who released it.” I believe the entangled web of political and societal issues will be sorted out soon in Italy in a constructive direction.

However, even with a 100% accurate earthquake prediction system (none exist at this time), there are practical matters that must be considered. For example: 1) Would those in the affected area take the warning seriously? Remember, geologists warned of an impending catastrophic earthquake, yet 100,000 still died in the 10 January, 2010, Haiti quake; 2) Most quake deaths are attributed to poorly constructed homes and buildings that are unlikely to be strengthened even with a warning; 3) Most countries, including the USA, lack the infrastructure and/or societal discipline to practice, much less conduct large scale evacuations.

SELF and VLF electromagnetic signal variations that preceded the Central Italy earthquake on August 24, 2016

Valentino Straser¹, Gabriele Cataldi² and Daniele Cataldi³

¹valentino.straser@alice.it, Independent researcher

²ltpaobserverproject@gmail.com,Radio Emissions Project (Rome)

³ltpaobserverproject@gmail.com,Radio Emissions Project (Rome)

Abstract: The strong earthquake hit Central Italy on August 24, 2016 with the loss of human lives and extensive damages to buildings and the historic and architectural heritage. It reopens the delicate theme of seismic precursors candidates and their future application. The main shock of magnitude (Mw) 6.2 occurred at 01:36:32 (UTC), with geographical coordinates (lat. and long.) 42.7, 13.23 at a depth of 8 km, and in the next ten days was followed by more than 5,500 events, in which 159 earthquakes of magnitude between 3.0 and 4.0, and 15 with magnitude in the range between 4.0 and 5.0. The observation of signals and radio anomalies detected by the LTPA Observer Project, Observatory of Rome, showed the appearance of unusual signals in the VLF band, appeared on August 18, 2016 and a radio anomaly recorded at 9:00 UTC on August 23, 2016. The natural change (variation) of electromagnetic emissions, in particular the anomaly observed in VLF band in 5 days before the main earthquake and the clear interference by SELF registered roughly 18 hours before the quake, bring us to suppose that these two elements could be pre-seismic signals of the strong earthquake and the following seismic swarm.

Keywords:seismic precursors, SELF-VLF frequency band, radio anomalies.

(Received on 8 September, 2016. Accepted on 14 September 2016)

Introduction

The relative proximity of the survey station LTPA Observer Project di Roma to the epicentral zone of the earthquake with magnitude Mw6.2 (see: USGS) of Central Italy on August 24, 2016 (Fig. 1), gives us an opportunity to verify the potential variation of the electromagnetic emissions occurred on a global scale before a quake of more than magnitude 6.0 (Molchanov and Hayakawa, 2008; Straser, 2011; Cataldi et al., 2014; Hayakawa, 2015). At the beginning of 1988 some hypotheses on possible correlation among the earthquake, variation of electromagnetic background and vertical and horizontal atmospheric flows were born. These studies were possible for the historic data of Black Sea. The link between the local variation of geomagnetic background and terrestrial electric flows was ascertained in the 2001 during a seminar organized by Institute for Nuclear Researcher and Nuclear Energy – INRNE in Sofia, Bulgaria (Mavrodiev et al., 2001). Anomalies of electromagnetic background on VLF band were observed 16 days before the earthquake with magnitude M6.8 occurred in Chamoli (India) on March 29 1999 (Singh et al., 2001). The satellite “Intercosmos-24” was launched into the Space on September 29 1989, in order to study the pre-seismic radio emission. The satellite during its journey through the 180 orbits from November 16 1989 to December 31 1989, analyzed 28 seismic event with medium-high intensity, between M5.2 and 6.1. This process of survey was possible thanks to a series of radio receptacle that are able to monitor ELF, ULF and VLF bands. The results of electromagnetic monitoring highlighted the presence of pre-seismic radio emission on ELF-ULF bands (f< 1000 Hz) and VLF band. The pre-seismic radio emissions achieve the highest intensity in the 12-24 hours before the seismic event. The pre-seismic radio emissions were observed on the ELF-ULF band roughly over the epicenter zone, while the VLF emission was observed far from the epicenter point (Molchanov et al., 1993). Anomalies of VLF propagation are revealed in the time between 22 and 31 in the month of December. In this period of time a train of strong earthquakes (M6+) was registered; they predated the earthquake M9.1 that hit the northern Sumatra on December 26, 2004 (Chakrabarti et al., 2005).

Radio anomalies are interferences that usually appear from 16 hours to a few minutes before an earthquake of magnitude equal or greater than 6.0. Similarly, other authors (Hayakawa et al., 2010; Molchanov and Hayakawa, 1998; Pulinets, 1998) believe that changes in the ELF-VLF frequency band may be associated to the seismic precursors, as reported by studies directly conducted on the field. Data presented in this report show the analysis of electromagnetic waves generated by tectonic stress in the preparatory stages of an earthquake (Freund et al., 2006).

Fig. 1. Index Map.

Anomalies detected in VLF band

The VLF electromagnetic (EM) signal monitoring is available 24/7. The spectrogram (Fig. 2) has been realized through a radio receiver prototype developed by Gabriele Cataldi, designed to work efficiently in the VLF band. The core of the receiver is represented by the LM386 chip, a very common operational amplifier in electronics, capable of working with a bandwidth of 300kHz and provide an amplification included between 20 and 74 dB. This is the same chip found in the famous receiver "INSPIRE VLF-3", but unlike this, the prototype developed by Gabriele Cataldi has a single amplification stage (always represented by the LM386 chip) that provides a gain of 44.95dB (177x). This receiver is connected to a loop antenna of square shape of dimensions of 60 x 60cm, containing 50 turns of enameled copper wire of 0.18mm diameter. The antenna is aligned in the direction of 310°NW and maintains a high directivity in this direction and in the opposite direction, ie. at 130°SE. Instead, it is "blind" to the electromagnetic radiation in the direction of 40°NE and 220°SW corresponding to the two null points (Fig. 2).

Fig. 2.VLF Monitor: The image represents the dynamic spectrogram of the Earth's electromagnetic field registered on August 18, 2016 between the 24:00 UTC and 07:30 UTC from the electromagnetic environment monitoring station of Radio Emissions Project, located at Albano Laziale (RM), Italy. In the middle of the picture, inside the red dotted line, a radio anomaly was recorded which predates the M6.2 Italian earthquake. The strange radio emission appeared at 02:47 UTC and disappeared at 06:21 UTC. The labels at the top of the spectrogram (in light blue) indicate known radio stations, mainly anthropogenic. The spectrogram recording speed is 1 line/minute.

Fig. 3. Alignment of the loop antenna in respect to the “F” geomagnetic component: The image shows the alignment of the loop antenna which is equipped with the Radio Emissions Project’s VLF monitoring station compared to the “F” geomagnetic component. The red line provided with arrows at the end represents the direction in which the loop antenna retains its maximum sensitivity (210°NW-130°SE) to the magnetic component. The blue line provided with arrows at the ends represents the direction in which the loop completely loses its sensitivity (40°NE-220°SW) and is "blind" to the magnetic radiation.

Anomalies detected in the SELF band

In general the pre-seismic radio emissions reach their maximum intensity 12-24 hours before the earthquake. Pre-seismic radio emissions that have been observed in the ELF-ULF band are observed almost above the epicenter, while VLF emissions are observed far away from the epicenter (Molchanov et al., 1993). Anomalies in VLF radiowave propagation were observed from 22 to 31 December 2004, the period in which a seismic train of strong earthquakes (M6+) was registered preceding the M9.1 Sumatra Island quake that occurred on December 26, 2004 (Chakrabarti et al., 2005).

The spectrogram, Fig. 4, shows an increase of the natural electromagnetic background mainly between 0 and 0.7Hz that predated the Italian M6.2 earthquake of about 2 hours (magnitude of the main earthquake is indicated by the red square and the vertical yellow line represents the temporal marker). This increase (indicated by the large red arrow) predated the M6.2 earthquake occurred on August 24, 2016 at 01:36 UTC, by approximately 70 minutes. It was characterized by strongest impulsive changes that predate the three big earthquakes after the main one (M6.2), whose magnitude is indicated in yellow. The emission peak identified by the acronym “SGP” is an intense emission which lasted about 40 minutes and that preceded the M6.2 earthquake for 17 hours. The similar ELF impulsive emissions (f=1~10Hz) have been observed by Ohta et al. (2013) prior the 2011 Tohoku EQ and to other seismic events (Shekotov et al., 2013).

Fig. 4. SELF-ELF Monitor. The image represents the dynamic spectrogram of the Earth's electromagnetic field registered between 02:00 UTC, August 23, 2016 and 06:30 UTC, August 24, 2016 by environmental electromagnetic monitoring station of Radio Emissions Project, located in Lariano (RM), Italy; it monitors the band SELF and ELF with a resolution of 10.1mHz. The upper portion of the spectrogram is centered in the SELF band between 0 and 1.5Hz, while the lower portion is centered in the SELF band between 0 and 0.31Hz. The spectrogram is acquired through a radio receiver prototype developed by Gabriele Cataldi designed to work efficiently between the SELF band (<3 Hz) and the ELF band (3-30Hz). The used antenna is a coil antenna aligned vertically. The word "SGP" is an acronym coined by the authors that identifies the radio emission of electromagnetic nature that was observed to precede large earthquakes (seismic geomagnetic precursor).

Conclusions

The seismic events potentially destructive are rare. For this reason a statistical analysis of the data is currently unreliable. This paper is preliminary, so that further detailed analysis is highly required. However, elements of coincidence between the appearance of radio anomalies in the SELF-VLF band before strong earthquakes, suggests that the search address can be developed waiting for a data modeling, finalized to the seismic prediction. The data collected before the Italian M6.2 earthquake occurred on August 24, 2016, is a new element to complement the complex mosaic of the seismic prediction.

Acknowledgments: The author wish express special thanks to Prof. Masashi Hayakawa for his interesting suggestion and for the improvement brought to our paper. The last but not the least the thanks to Dr. Dong Choi for his critical reading of the manuscript and his encouragement in order to continue our research.

References cited

Cataldi, G., Cataldi, D. and Straser, V., 2014. Earth’s magnetic field anomalies that precede the M6+ global seismic activity. Geophysical Research Abstracts, v. 16, EGU2014-1068, 2014.

Chakrabarti, S.K., Saha, M., Khan, R. Mandal, S, Acharyya, K. and Saha, R., 2005. Unusual sunset terminator behaviour of VLF signals at 17KHz during the Earthquake episode of Dec., 2004. http://www.ursi.org/Proceedings/ProcGA05/pdf/EP.18(01596).pdf

Freund, F.T., Takeuchi, A. and Lau, B.W.S., 2006. Electric currents streaming out of stressed igneous rocks – A step towards understanding pre-earthquake low frequency EM emissions. Physics and Chemistry of the Earth, v. 31, p. 389-396.

Hayakawa, M., Horie, T., Muto, F., Kasahara, Y., Ohta, K., Liu, J.-Y. and Hobara, Y., 2010. Subionospheric VLF/LF probing of ionospheric perturbations associated with earthquakes: A possibility of earthquake prediction. SICE Journal of Control, Measurement, and System Integration, v. 3, no. 1, p. 010–014.

Hayakawa, M, 2015. Earthquake Prediction with Radio Techniques, John Wiley & Sons, Singapore, 294p.

Mavrodiev, S. Cht. and Thanassoulas, C., 2001. Possible correlation between electromagnetic Earth fields and future earthqukes. Seminar Proceedings. Institute for Nuclear Research and Nuclear Energy, Bulgarian Academy of Sciences. 23-27 July, 2001. Sofia.

Molchanov, O.A. and Hayakawa, M., 1998. Subionospheric VLF signal perturbations possibly related to earthquakes. Jour. Geophys. Res., v. 103, p. 17489–17504.

Molchanov, O.A., Mazhaeva, O.A., Golyavin, A.N. and Hayakawa, M., 1993. Observation by the Intercosmos-24 satellite of ELF-VLF electromagnetic emissions associated with earthquakes. Annales Geophysicae (ISSN 0992-7689), v. 11, no. 5, p. 431-440.

Molchanov, O. A. and Hayakawa, M., 2008. Seismo Electromagnetics and Related Phenomena: History and latest results. TERRAPUB, Tokyo, 189p.

Ohta, K., Izutsu, J., Schekotov, A., and Hayakawa, M., 2013. The ULF/ELF electromagnetic radiation before the 11 March 2011 Japanese earthquake, Radio Sci., v. 48, 589–596, doi:10.1002/ rds.20064.

Pulinets, S.A., 1998. Seismic activity as a source of the ionospheric variability. Adv. Space Res., v. 22, p. 903-906.

Schekotov, A., Fedorov, E., Molchanov, O. A. and Hayakawa, M., 2013. Low frequency electromagnetic precursors as a prospect for earthquake prediction. In “Earthquake Prediction Studies: Seismo Electromagnetics”, Ed. by M. Hayakawa, TERRAPUB, Tokyo, 81-99.

Singh, R.P., Mishra, P.K., and Singh, B., 2001. Anomalous VLF electric field perturbations associated with Chamoli earthquakes of March/April 1999. Current Science, v. 80, no. 11, p. 1416-1421.

Straser, V., 2011. Radio anomalies and variations in the Interplanetary Magnetic Field (IMF) used as seismic precursors on a global scale. New Concepts in Global Tectonics Newsletter, no. 61, p. 52-65.

Jetstream anomalies appeared prior to

the M6.2 Italy Earthquake on 24 August 2016

Hong-Chun Wu^{1, 2}

¹ Institute of Labor, Occupational Safety and Health, Taiwan

² Formosa Science Center

wuhongchun094@gmail.com

Abstract: The M6.2 Italy earthquake on 24 August. 2016, killed 292 people with extensive property damage and social disruptions. The 60 knots (108 km/h) jetstream contour crossed near the epicenter on 24 Jan. 2016 at 06:00 (UTC), seven months prior to the major M = 6.2 earthquake near Amatrice, Central Italy. The distance between epicenter and the jetstream contour cross point was about 100 km. The Amatrice quake is characterized by the longest time window from the jetstream anomaly appearance to the actual quake occurrence (214 days): It was somewhat comparable withthe M8.3 Chile earthquake in 2015,which was 96 days, and with the M7.0 Kumamoto Japan earthquake in 2016, 140 days. The Amatrice earthquake again proved that the jetstream anomaly is one of the most reliable short to medium-term precursory signals in forecasting impending earthquakes.

Keywords: jet stream, earthquake, precursor

(Received on 13 September 2016. Accepted 18 September)

1. Introduction

As reported earlier (Wu, 2015 and 2016), the jetstream anomaly signals allowed successfully forecasting recent two catastrophic earthquakes; the M8.3 Chile earthquake in 2015 and the M7.0 Japan earthquake in 2016. In both cases, the geographic deviation between epicenter and jetstream anomaly point was less than 100 km, but the time delay to mainshock occurrence of these earthquakes (4 to 5 months) was larger than earlier earthquakes, which had usually been one to two months. This study reports the jetstream anomaly appeared before the M6.2 Italy earthquake.

2. The M6.2 Italy earthquake on 24August 2016

The Central Italy earthquake on 24 August 2016 brought 292 human casualties (Wiki web, 2016). Satellite data observation by the author found possible atmospheric disturbances in jetstream velocity before the M = 6.2Italy Earthquake on 24August 2016. The 60 knots (108 km/h) jetstream speed contour crossed at approximately 43.7N in longitude and 13.2E in latitude on 24January 2016 at 06:00 UTC (Figure 1), 214 days prior to the major M = 6.2 shock. The anomalous cross point was near the epicenter less than 100 km. Because of measurement deviation and satellite map resolution, the predicted point was placed at 44.3N and 12.0E(Fig. 1), not at the crosspoint (43.7N, 13.2E). The predicted map was posted on the author’s website:

https://www.facebook.com/photo.php?fbid=1042124835839565&set=a.657516484300404.1073741826.100001261760990&type=3&theater

Forecast parameters posted on 25 January 2016:

2016/01/24~2016/02/24 Northern Italy (44.3N12.0E), M>6.0

Actual event:

M6.2 2016-08-24 01:36:33 (UTC)42.714N 13.172E10.0 km

Figure 1. The anomalous behavior of jet stream: (a) The original jet stream map (San Francisco State University in USA), (b) the jet stream at a speed of 60 knots (108 km/h) crossed near the epicenter on 24Jan. 2016 at 06:00 (UTC). The crossed point of jet stream (about43.7N 13.2E) is shown by . The epicenter (42.714N 13.172E) is shown by , and the predicted point (44.3N 12.0E) .

3. Discussion and conclusion

The distance between the epicenter and the jet stream contour cross point was about 100 km. If the frequency of satellite image is increased from 1 map/6 hours to 1 map/3 hours, the location deviation can be reduced.

This earthquake has shown a much longer time window (216 days) from the jetstream anomaly appearance to the mainshock occurrence than previous earthquakes; Other long time window was observed in the M8.3 Chile earthquake in 2015, which was 96 days, and in the M7.0 Kumamoto Japan earthquake in 2016, 140 days. The Amatrice Earthquake extended these time windows much longer.

The method of jetstream precursor needs further improvement by enhancing resolution and frequency of the satellite images. Further improvement can be achieved by launching a new satellite for air velocity measurement in the tropical region. However, despite some constraints, the jet stream precursor is still a powerful, short to medium-term tool, and is capable of forecasting future strong earthquakes.

References cited

Wiki web, 2016:https://en.wikipedia.org/wiki/2016_Central_Italy_earthquake

Wu, H.C., 2015. Anomalies in jet streams that appeared prior to the 16 September 2015 M8.3 Chile earthquake. New Concepts in Global Tectonics Journal, v. 3, no. 3, p407-408

Wu, H.C., 2016.Anomalies in jet-streams prior to the M6.6 Taiwan Earthquake on5 February 2016 and

the M7.0 Kumamoto Earthquake on 15 April 2016. New Concepts in Global Tectonics Journal, v. 4,

no. 2, p.276-278.

Latent heat anomalies prior to the Amatrice, Italy M6.2 Earthquake

Ariel R. Césped

Chilesismos

chilesismos@gmail.com

Abstract: The following short note presents the analysis of the air energy at Central Italy, during the period of July-August, 2016. Anomalies were detected on August 12, twelve days prior to the deadly M6.2 earthquake at Amatrice (42.71°N, 13.17°E). This anomalous behavior suggests potential evidence of the Lithosphere-Atmosphere-Ionosphere Coupling (LAIC) and the ion-induced nucleation mechanism, a well-known phenomenon triggered by radon daughters once they are released from the stressed fault lines as terminal process for impeding earthquakes.

Keywods: earthquake prediction, latent heat, LAIC model, Gibbs free energy, statistic analysis

(Received on 30 August 2016. Accepted on 14 September 2016)

PROCEDURE AND ANALYSIS

Contemporary studies about the airborne latent heat have demonstrated it is a trustworthy parameter for earthquake forecast (Dey et. al., 2003, Cervone et. al., 2005, Li et. al., 2009, Habibeh et. al., 2013, and Césped, 2015). Radon and its daughters that leaks from the ground and fault lines prior to earthquakes (Pulinets, 2011) are neutralized by water molecules suspended in the air, while thermal energy is released in accordance with the Gibbs Free-Energy Theory. Depending on the size of the signal (which is proportional to the earthquake that is due to occur), the heat might affect strongly the meteorology of the future epicenter, so one can detect anomalies linked to the process described previously by using specific statistical tools.

Based on the Gibbs principle, an analysis of the latent heat of condensation (LHC) has been done. Meteorological data from the Alberto Blanchetti airport at Rieti (located 42 kilometers from the mainshock epicenter) were collected.

Corrected LHC is calculated with equation (1) (Rogers and Yau, 1989):

(1)

where T is the air temperature, in Celsius degrees. In order to get a more confident result, the dew-point temperature () is replaced for T. is calculated by integrating the well-known Clausius-Clapeyron equation (2):

(2)

where T is the air temperature and is the relative humidity. The anomaly index added to determine the threshold of anomaly, in accordance with equation (3):

(3)

Where is the daily value of LHC, is the average of the LHC dataset, and σ is its standard deviation. The threshold is set as 2.5 times the interquartile range. Figure 1 shows the anomaly index of the LHC during period July 24 – August 24, 2016.

Figure 1. LHC index anomaly at Rieti in period July 24 – August 24, 2016.

As one can see in the results, an anomaly in LHC has been detected on August 12, 2016 in the vicinity of the earthquake epicenter. Since weather conditions and anthropogenic aerosols (contamination levels) at Rieti are very stable and there is no records of storms and hurricanes in the test period, the peak might be related strongly to radionuclide activity in the air as earthquake precursors. Even though the potential success of this technique, further parameters should be included, to verify (or discard) the observations and create a more robust multi-data system to avoid false positives.

REFERENCES

Alvan, H.V., Azad, F.H. and Mansor,S.,2013. Latent heat flux and air temperature anomalies along an active

fault zone associated with recent Iran earthquakes. Advances in Space Research, v. 52, issue 9,

p. 1678–1687.

Cervone,G., Singh, R. P., Kafatos,M. and Yu1,C., 2005. Wavelet maxima curves of surface latent heat flux anomalies associated with Indian earthquakes.Natural Hazards and Earth System Sciences, v. 5, p. 87–99 SRef-ID: 1684-9981/nhess/2005-5-87.

Césped, A.R., 2015. Analysis of psychrometric parameters associated with seismic precursors in central Chile:

a new earthquake or the great 2010 Maule M8.8 aftershock?New Concepts in Global Tectonics Journal, v. 3, p. 383-386.

Dey, S. and Singh, R.P., 2003.Surface latent heat flux as an earthquake precursor. Nat. Hazards Earth Syst. Sci.,v. 3, p. 749-755, doi:10.5194/nhess-3-749-2003.

Li, J., Wu, L., Wu, H., Liu, S. and Yu, J., 2009. Surface Latent Heat Flux (SLHF) Prior to Major Coastal and Terrestrial Earthquakes in China.Progress in Electromagnetics Research Symposium, Beijing, China, March 23–27, 2009.

Pulinets, S. and Ouzounov, D., 2011. Lithosphere–Atmosphere–Ionosphere Coupling (LAIC) model – An unified concept for earthquake precursors validation.Journal of Asian Earth Sciences, v. 41, issues 4–5,p. 371–382.

Relative humidity and OLR as pre-earthquake signals – A study of Central Italy earthquake (August 24, 2016)

Natarajan Venkatanathan^1* and Rubidha Devi Duraisamy²

¹Department of Physics, SASTRA University, Thirumalaisamudram, Thanjavur, Tamil Nadu, India. venkatanathan@eee.sastra.edu

²School of Computer Science and Engineering, SRC Campus, SASTRA University, Kumbakonam, Tamil Nadu, India. rubidhadevi@src.sastra.edu

Abstract: Prior to the occurrence of several devastating earthquakes like Sumatra earthquake 2004, China earthquake 2008, Haiti earthquake 2010 and Japan earthquake 2011 anomalous behaviour of atmospheric parameters was detected with the help of satellite technology. The results obtained by the precursory studies have encouraged the scientists to correlate multi-parameter precursors to forecast earthquakes on a short-term basis. In this paper, the authors’ approach is based on monitoring the relative humidity (RHUM) and outgoing longwave radiation (OLR). The data sets are obtained from NOAA and processed further to identify the anomalous deviations in the atmospheric parameters. The authors have analyzed the recent central Italy earthquake occurred on August 24, 2016 with the magnitude of 6.2. It is observed that precursors like relative humidity and outgoing long wave radiation begin to appear 3-30 days prior to the earthquakes, near the epicenter. This enables the authors to conclude that it is possible to forecast earthquakes of greater magnitude on a short term basis with reasonable accuracy by using relative humidity and anomalous outgoing longwave radiation.

Keywords: relative humidity, outgoing longwave radiation, anomaly, precursors and earthquake forecasting

(Received on 23 September 2016. Accepted on 26 September 2016)

1.Introduction

Forecasting the earthquake requires a good understanding of complex physical phenomena, but the devastation produced by earthquakeshas forced the scientists to explore the possibilities of short-term earthquake prediction. Scientists from various countries like India, Italy, China, Turkey, and Japan are exploring the possibility of using various ground-based and space-based monitoring techniques to forecast big earthquakes. From the retrospective analysis of Relative Humidity (RHUM) and Outgoing Longwave Radiation (OLR) scenarios along the seismically active region, it was found by many scientists across the globe that there were anomalous deviations prior to the various devastating large earthquakes.

Scientists have been using satellite based technologies to identify the earthquake precursory parameters since 1988 (Gorny et al., 1988). The notable anomalous OLR observations were observed prior to the big earthquakes of this century – Sumatra 2004 earthquake and Japan 2011 earthquake (Ouzounov et al., 2007; Ouzounov et al., 2011). Similarly, prior to the earthquake occurred in Awaran, Pakistan on September 24, 2013, short-lived anomalies were observed thrice (Venkatanathan and Natyaganov, 2014). Prior to the earthquakes of Italy and Indian regions abnormal OLR flux values were observed (Vineeta Rawat et al., 2011). In recent years, anomalous changes in the ‘total electron concentration’ (TEC) of F2 layer of the ionosphere were observed prior to the occurrence of big earthquakes (Pulinets, 2011; Liu et al., 2004). Before the occurrence of China earthquake prominent OLR singularities were found from the analysis of wavelet maxima curves (Xiong et al., 2009). Prior to the earthquakes of Iran region occurred during 2003 and 2005, temperature rise of 5˚C to 10˚C were observed near the epicenters of the earthquakes (Choudhury et al., 2006). In the recent past similar studies were done by the scientists prior to the earthquakes of recent past (Gorny and Shilin, 1992; Qiang et al., 1999; Tronin, 2000; Choudhury et al., 2006; Saraf et al., 2009; Panda et al., 2007; Ouzounov and Freund, 2004; Tramutoli et al., 2005; Filizzola et al., 2004). The outgoing longwave radiation is reflected from the Earth's surface to the atmosphere and it is measured at the top of the atmosphere. It is often influenced by cloud and surface temperature of the Earth (Pan Xiong et al., 2009).

The anomalous thermal rise in the atmosphere may be due to the emission of radon gas from the interior of the Earth (Pulinets and Ouzounov, 2011). The emanation of radioactive gases like radon gas makes the air in the atmosphere to get ionized. These ionized airs are the collection centers of the water vapours, which condense and release latent heat to a larger extent into the atmosphere. These ion-induced nucleation process plays important role in the coupling between thermal and electric phenomena of the atmosphere. Thus the Lithosphere – Atmosphere – Ionosphere – Magnetosphere Coupling develops prior to the occurrence of devastating earthquakes.

In this manuscript the authors have chosen the recent Italy earthquake that occurred on August 24, 2016 with the magnitude 6.2, in order to validate the appearance of OLR and relative humidity in an anomalous manner prior to the earthquake.

2. Methodology

Prior to the occurrence of the earthquake the radon gas emission gets increased due to the increased tectonic activity. These radioactive gases are moving upward from the interior of the Earth and ionize the air molecules present in the surface of the Earth. The ions produced are acting as condensation centre of the water vapour, which collects more water vapour around it. During the condensation process enormous amount of latent heat energy is released, which triggers a drop in humidity near the epicentral region. Due to convective process, the hot air moves up and heats the atmosphere, thus anomalous increase in OLR flux value occurs.

2.1. Drop in relative humidity

The anomalous drop below the -2 level in relative humidity (RHUM) is calculated as follows,

Where “n” is the number of predefined preceding years for which mean OLR flux is calculated for the selected spatial region and time.

Where,

2.2. OLR flux index method

Outgoing Long wave Radiations are low frequency electromagnetic radiations reflected from the Earth’s surface and clouds, which can be measured above the cloud level. The basic data obtained from the satellite is processed using an algorithm to obtain the OLR data ranging between 10 and 13 µm, since atmosphere is most transparent in this wavelength range, and not influenced by other factors like water vapour, carbon dioxide and Ozone. Normally OLR anomaly was observed 3 to 30 days prior to the devastating earthquakes.

The OLR anomaly can be identified by finding variation in the anomalous signal level index (AS_index), which indicates the maximum change in the level of OLR flux (Ouzounov et al., 2011):

Where “n” is the number of predefined preceding years for which mean OLR flux is calculated for the selected spatial region and time.

Where,

3. Results and Discussion

Earthquake in Amatrice, Central Italy (Latitude: 42.71N and Longitude: 13.22E) struck on August 24, 2016 with magnitude of 6.2. The depth of the earthquake is 4 km, which is very shallow in nature and hence wide spread devastation was reported.

Prior to the earthquake both drop in relative humidity and anomalous spike in OLR was observed. First the drop in relative humidity below the -2 level was observed approximately a month before the occurrence of earthquake in Italy and the drop was lasted for three days starting from July 23, 2016 to July 25, 2016. On July 23, 2016 the relative humidity recorded was 48%, which is 29.4% less than 10 years average; hence the change index of -3.1769 was observed (Fig. 1).

Fig. 1. Graph showing variations in relative humidity observed prior to the earthquake at the location 42.5N latitude and 12.5E longitude.

Again the drop in relative humidity was observed on August 15, 2016, just 9 days before the earthquake. This time the drop was observed for a day only.

The anomalous OLR again spiked on July 30, 2016 – seven days after the relative humidity drop was observed for the first time (Fig. 2). The anomalous index of 2.6214 was observed at the nearby location 42.5N latitude and 13.5E longitude.

The drop in relative humidity and the positive deviation in OLR flux were observed due to the emanation of radon gas. The increased tectonic activity in the Central Italy region, reduces the volume of voids present in the rocks, thus the radon gas present in the voids get released and moving up to reach the Earth’s surface. At the surface, due to the radioactive nature of the radon gas, the air molecules get ionized. These ions are acting as condensation centers of water vapour and form aerosol sized particles. This process leads to drop in relative humidity, which leads to the release of latent heat into the atmosphere. Due to the convection process the released heat energy moves in the atmospheric column and contributed to the increase in OLR flux value in an anomalous manner (Pulinets and Ouzounov, 2011). This possible mechanism can be inferred from the sequence of events occurred prior to the Central Italy earthquake. At first the drop in relative humidity was observed on July 23, 2016 and it was followed by positive deviation in OLR flux on July 30, 2016.

Fig. 2. Anomalous OLR scenario was observed at the nearby location of Central Italy earthquake occurred on July 30, 2016, 15 days before the Amatrice earthquake.

4. Conclusion

Radon from the vicinity of the tectonically active region is the basis for the drop in relative humidity and anomalous OLR variations. Normally these anomalies are observed prior to the devastating earthquakes. In this paper we have analyzed, the relative humidity and OLR flux prior to the Central Italy earthquake occurred with the magnitude of 6.2. Even though anomalous variations in RHUM and OLR can provide vital clue to impending earthquakes, a lot of work need to be done in order to fine tune the earthquake forecasting research. Thus the authors propose interdisciplinary study of earthquake precursors to resolve the complex earthquake phenomena. In order to improve the accuracy of earthquake forecasting studies, the authors proposed the multi-parameter and interdisciplinary precursory techniques by including the parameters like total electron content, infra sound studies, radon gas emanation and EM emissions; then earthquake forecasting will no longer be considered to be ‘ELUSIVE’.

Acknowledgements: The authors acknowledge National Centers for Environmental Information, NOAA for providing Relative humidity and OLR data. We sincerely acknowledge the anonymous reviewers for giving valuable inputs, which helped us to the fine tune this paper.

References

Filizzola, C., Pergola, N., Pietrapersota, C. and Tramutoli, A., 2004. Robust satellite techniques for seismically active areas monitoring: a sensitivity analysis on September 7, 1999 Athen’s earthquake. Phys Chem Earth, v. 29, p. 517–527.

Choudhury, S., Dasgupta, S., Saraf, A.K., Panda, S. K., 2006. Remote sensing observations of pre-earthquake thermal anomalies in Iran. Int J. Remote Sens., v. 27, no. 20, p. 4381–4396.

Gorny, V.I., Salman, A.G., Tronin, A.A. and Shilin, B. B., 1988. “The earth’s outgoing IR radiation as an indicator of seismic activity”. Proc Acad Sci USSR., v. 301, p. 67–69.

Gorny, V.I. and Shilin, B.V., 1992. “The thermal methods of remote sensing for study of natural resources”. In the Proceedings of the 18th annual conference of the remote sensing society, University of Dundee. 15–17 Sep 1992, UK, p. 244–263.

Liu, JY, Chuo, Y. J., and Shan, S.J., 2004. “Pre-earth- quake ionospheric anomalies monitored by GPS TEC”. Ann Geophys, v. 22, p. 1585-93.

Ouzounov, D. and Freund, F., 2004. “Mid‐infrared emission prior to strong earthquakes analyzed by remote sensing data”. Adv. Space Res., v. 33, p. 268–273.

Ouzounov, D., Liu, D., Kang, C., Cervone, G., Kafatos, M. and Taylor, P., 2007. “Outgoing long wave radiation variability from IR satellite data prior to major earthquakes”.Tectonophysics, v. 431, p. 211-220.

Ouzounov, D., Pulinets, S., Romanov, A., Romanov, A., Tsybulya, K., Davidenko, D., Kafatos, M. and Taylor, P., 2011. “Atmosphere-ionosphere response to the M9 Tohoku earthquake revealed by multi-instrument space-borne and ground observations: Preliminary results”.Earthquake Science, v 24, no. 6, p. 557-564.

Panda, S.K., Choudhury, S., Saraf, A.K. and Das, J.D., 2007. “MODIS land surface temperature data detects thermal anomaly proceeding 8 October 2005 Kashmir earthquake”. Int. J. Remote Sens., v. 28, no. 20, p. 4587–4596.

Pulinets, S., 2012. “Low-Latitude Atmosphere-Ionosphere Effects Initiated by Strong Earthquakes Preparation Process”. Int. J. Geophy, v. 2012, p. 1-14.

Pulinets, S. and Ouzounov, D., 2011. “Lithosphere–Atmosphere–Ionosphere Coupling [LAIC] model – an unified concept for earthquake precursors validation”. Journal of Asian Earth Sciences, v. 41, p. 371–382.

Qiang, Z., Chang-gong, D., Lingzhi, L., Min, X., Fengsha, G., Tao, L., Yong, Z. and Manhong, G., 1999. “Satellite thermal infrared brightness temperature anomaly image—short-term and impending earthquake precursors”. Sci. China., v. 42, no. 3, p. 313–324.

Saraf, A.K., Vineeta, R., Choudhury, S., Dasgupta, S. and Das, J., 2009. Advances in understanding of the mechanism for generation of earthquake thermal precursors detected by satellites. International Journal of Applied Earth Observation and Geoinformation, v. 11, p. 373-379.

Tramutoli, V., Cuomo, V., Filizzola, C., Pergola, N. and Pietrapertosa, C., 2005. “Assessing the potential of thermal infrared satellite surveys for monitoring seismically active areas: the case of Kocaeli [Izmit] earthquake, August 17, 1999”, Remote Sens. Environ., v. 96, p. 409–426.

Tronin, A.A., 2000. Thermal satellite data for earthquake research., IGARSS 2000, IEEE 2000 International Geosciences Symposium, Taking the Pulse of the planet: the role of remote sensing in managing the environment, Honolulu, HI, IEEE.

Venkatanathan, N. and Natyaganov, V., 2014. “Outgoing Longwave Radiations as Pre-Earthquake Signals: Preliminary Results of September 24, 2013 [M7.7] Earthquake”. Current Science, v. 106, no. 9, p. 1291 – 1297.

Vineeta, R., Saraf, A.K., Das, J., Sharma, K. and Shujat, Y., 2011. “Anomalous land surface temperature and outgoing long-wave radiation observations prior to earthquakes in India and Romania”. Nat Hazards, v. 59, p. 33–46.

US Geological Survey.www.neic.usgs

Xiong, P., Bi, Y.Z. and Shen, X.H., 2009. “Study of Outgoing Longwave Radiation Anomalies Associated with Two Earthquakes in China Using Wavelet Maxima”, Hybrid Artificial Intelligent Systems, Berlin; New York: Springer, p. 78.

Time-dependent neo-deterministicseismic hazard scenarios:

Preliminary report on the M6.2 Central Italy earthquake,

24th August 2016

Antonella Peresan¹, Vladimir Kossobokov², Leontina Romashkova³, Andrea Magrin⁴, Alexander Soloviev⁵ and Giuliano F. Panza⁶

¹CRS-OGS, National Institute of Oceanography and Experimental Geophysics. Udine, Italy.

Department of Mathematics and Geosciences, University of Trieste, Italy. International Seismic Safety Organization (ISSO). aperesan@units.it

²Institute of Earthquake Prediction Theory and Mathematical Geophysics, Russian Academy of Sciences, Moscow, Russian Federation. International Seismic Safety Organization (ISSO). volodya@mitp.ru

³ Institute of Earthquake Prediction Theory and Mathematical Geophysics, Russian Academy of Sciences, Moscow, Russian Federation. International Seismic Safety Organization (ISSO). lina@mitp.ru

⁴Department of Mathematics and Geosciences, University of Trieste, Italy. amagrin@units.it

⁵ Institute of Earthquake Prediction Theory and Mathematical Geophysics, Russian Academy of Sciences, Moscow, Russian Federation. soloviev@mitp.ru

⁶International Seismic Safety Organization (ISSO). Institute of Geophysics, China Earthquake Administration, Beijing. giulianofpanza@fastwebnet.it

Abstract: A scenario-based Neo-Deterministic approach to Seismic Hazard Assessment (NDSHA) is available nowadays, which permits considering a wide range of possible seismic sources as the starting point for deriving scenarios by means of full waveforms modeling. The method does not make use of attenuation relations and naturally supplies realistic time series of ground shaking, including reliable estimates of ground displacement, readily applicable to complete engineering analysis. Based on the neo-deterministic approach, an operational integrated procedure for seismic hazard assessment has been developed that allows for the definition of time dependent scenarios of ground shaking, through the routine updating of earthquake predictions, performed by means of the algorithms CN and M8S. The integrated NDSHA procedure for seismic input definition, which is currently applied to the Italian territory, combines different pattern recognition techniques, designed for the space-time identification of strong earthquakes, with algorithms for the realistic modeling of ground motion. Accordingly, a set of deterministic scenarios of ground motion at bedrock, which refers to the time interval when a strong event is likely to occur within the alerted area, is defined both at regional and local scale. The CN and M8S predictions, as well as the related time-dependent ground motion scenarios associated with the alarmed areas, are routinely updated since 2006. The prospective application of the time-dependent NDSHA approach provides information that can be useful in assigning priorities for timely mitigation actions and, at the same time, allows for a rigorous validation of the proposed methodology. The results from real-time testing of the time-dependent NDSHA scenarios are illustrated with specific reference to the August 24th, 2016 Central Italy earthquake.

Keywords: intermediate-term earthquake prediction, seismic hazard scenario, time-dependent hazard, Italy

(Received on 26 September 2016. Accepted on 29 September 2016)

Introduction

A reliable and comprehensive characterization of expected seismic ground shaking, eventually including the related time information, is essential in order to develop effective mitigation strategies and increase earthquake preparedness. Forecasting earthquakes and related ground shaking, however is not an easy task and it implies a careful application of statistics to data sets of limited size and different accuracy. Nowadays it is well recognized by the engineering community that standard hazard indicator estimates (e.g. seismic PGA, peak ground acceleration) alone are not sufficient for the adequate design, mainly for special buildings and infrastructures (Panza et al., 2011). Moreover, any effective tool for seismic hazard assessment (SHA) must demonstrate its capability in anticipating the ground shaking related with large earthquake occurrences, a result that can be attained only through rigorous verification and validation process.

A scenario-based Neo-Deterministic approach to Seismic Hazard Assessment (NDSHA) is available nowadays, which considers a wide range of possible seismic sources (including the largest deterministically or historically defined credible earthquake, MCE) as the starting point for deriving scenarios by means of full waveforms modeling, either at national and local scale. The method does not make use of attenuation relations and naturally supplies realistic time series of ground shaking, including reliable estimates of ground displacement readily applicable to seismic isolation techniques. The NDSHA procedure permits to incorporate, as they become available, new geophysical and geological data, leading to the natural definition of a set of scenarios of expected ground shaking at the bedrock. At the local scale, further investigations can be performed taking into account the local soil conditions, in order to compute the seismic input (realistic synthetic seismograms) for engineering analysis of relevant structures, such as historical and strategic buildings. The standard NDSHA has been already applied in several regions worldwide, including a number of local scale studies accounting for two-dimensional and three-dimensional lateral heterogeneities in inelastic media.

Based on the neo-deterministic approach, an operational integrated procedure for seismic hazard assessment has been developed that allows for the definition of time dependent scenarios of ground shaking, through the routine updating of earthquake predictions, performed by means of the algorithms CN and M8S (Peresan et al., 2005 and references therein). The CN and M8S algorithms belong to a family of fully formalized procedures for intermediate-term medium-range earthquake prediction, which are based on a quantitative analysis of the seismic flow. These procedures are tested over decades of the on-going real-time prediction experiments in several regions worldwide, including Italy. The integrated NDSHA procedure for seismic input definition, which is currently applied to the Italian territory, combines different pattern recognition techniques, designed for the space-time identification of strong earthquakes, with algorithms for the realistic modeling of ground motion. Accordingly, a set of deterministic scenarios of ground motion at bedrock, which refers to the time interval when a strong event is likely to occur within the alerted area, can be defined by means of full waveform modeling, both at regional and local scale. The CN and M8S predictions, as well as the related time-dependent ground motion scenarios associated with the alarmed areas, are routinely updated every two months since 2006 (Panza et al., 2012; Peresan et al., 2011).

Intermediate-term middle-range earthquake predictions by CN and M8S algorithms

Two formally defined and globally tested algorithms for intermediate-term middle-range earthquake prediction, namely CN and M8S (Keilis-Borok and Soloviev, 2003; Peresan et al., 2005), are routinely applied to the Italian territory with the aim to provide space and time constraints about the impending strong earthquakes. The algorithms CN and M8S have been designed following the general concepts of pattern recognition, which automatically imply strict definitions and reproducible prediction results, in agreement with the guidelines provided by the United States National Research Council, Panel on Earthquake Prediction of the Committee on Seismology (Allen et al., 1976). Based on a multiple set of premonitory patterns, CN and M8S allow for a diagnosis of the Times of Increased Probability (TIPs) for the occurrence, inside a given region and time window, of earthquakes within a specified magnitude range. Quantification of the seismicity patterns is obtained through a set of empirical functions of time, which are evaluated on the sequence of earthquakes occurred within a set of localized regions. The functions account for significant variability of space-time clustering of moderate size earthquakes, as well as for specific changes in seismic activity, including anomalous activation and quiescence. Several experiments have been dedicated to assess a high level of robustness of the methodology against unavoidable uncertainties and possible changes in the input data. Italy represents nowadays the only region of moderate seismic activity where the two algorithms are applied simultaneously for the prospective monitoring of seismic activity aimed at forecasting strong earthquakes.

The CN and M8S predictions, as well as the related time-dependent ground motion scenarios associated with the alarmed areas, are routinely updated every two months since 2006. The rules for the real-time application of the CN and M8S algorithms to the Italian territory are described in detail in Peresan et al. (2005), whereas the procedure for the definition of the related ground shaking scenarios is illustrated in Peresan et al. (2011).

The intermediate-term middle-range earthquake prediction experiment, aimed at a real-time testing of M8S and CN predictions for earthquakes with magnitude larger than a given threshold (namely 5.4 and 5.6 for CN algorithm, and 5.5 for M8S algorithm) in the Italian region and its surroundings, is ongoing since 2003. Predictions are regularly updated every two months and a complete archive of predictions is made available on-line (http://www.geoscienze.units.it/esperimento-di-previsione-dei-terremoti-mt.html), thus allowing for a rigorous validation of the applied algorithms. The results obtained during more than nine years of real-time monitoring already permitted a preliminary assessment of the significance of the issued predictions (Peresan et al., 2011). So far, 14 out of the 16 strong earthquakes, occurred within the monitored territory since 1954, have been correctly preceded by an alarm (TIP, Time of Increased Probability) declared by CN algorithm, with about 30% of the overall space–time volume occupied by alarm; the confidence level for such predictions is above 99%. The prediction experiment by CN algorithm has been recently expanded to the Adria Region, where predictions are routinely updated since 2005 (Peresan, 2016). When including the Adria Region the score increases to 21 out of 25 strong earthquakes correctly diagnosed by CN algorithm, with about 31% of TIPs, and a confidence level well above 99%.Similarly, the algorithm M8S correctly identified 14 of the 23 earthquakes with magnitude M5.5+ (i.e., between 5.5 and 6.0), occurred since 1972 within the monitored territory, with a space–time volume of alarm of about 31%; the confidence level of M5.5+ predictions has been estimated to be above 98% (no estimation is yet possible for higher magnitude levels).

A strong earthquake (M=6.2) hit the town of Amatrice in Rieti Province, Central Italy, on 24^th August 2014 (data from ANSS, USGS catalog). The epicenter was located inside the Central Region (Fig. 1), alerted by CN algorithm for an earthquake with magnitude M5.6, starting on 1 November 2012, whereas it occurred outside the areas alerted by M8S algorithm for the corresponding magnitude interval (Fig. 2). Therefore the earthquake scores as a successful real-time prediction, for CN algorithm only. The epicenter of the event falls outside the area alerted by M8S for an earthquake with magnitude 6.0≤M<6.5, as on August 2016 (Fig. 2B); however, the epicentral area was in state of alarm less than one year before, i.e. up to December 2015 (Fig. 2A). During 2016 the alarm area shrunk down to the south, thus resulting in a failure to predict.

The uncertainties associated with intermediate-term middle-range earthquake predictions are intrinsically quite large. CN and M8S algorithms, however, already proved effective in predicting strong earthquakes, by rigorous prospective testing over the Italian territory. Specifically, 6 out of 8 target earthquakes have been predicted in real-time by CN, and 5 out of 9 by M8S algorithm. For both the algorithms the confidence level achieved in real-time testing is above 97%. From the diagram of alarms (TIPs) reported in Fig. 1, it is possible to observe that 5 out of 8 alarms where followed by an earthquake, with a false alarm rate around 35% for CN Central region.

Neo-deterministic time-dependent seismic hazard scenarios for the Italian territory

Based on NDSHA approach, an operational integrated procedure for seismic hazard assessment has been developed (Peresan et al., 2011) that allows for the definition of time dependent scenarios of ground shaking, through the routine updating of earthquake predictions, performed by means of the algorithms CN and M8S (Peresan et al., 2005). The integrated NDSHA procedure for seismic input definition, which is currently applied to the Italian territory, combines different pattern recognition techniques, designed for the space-time identification of strong earthquakes, with algorithms for the realistic modelling of ground motion. Accordingly, a set of deterministic scenarios of ground motion at bedrock, which refers to the time interval when a strong event is likely to occur within the alerted areas can be defined by means of full waveform modelling, both at regional and local scale.

Fig. 1. Central Region (in yellow) alerted by CN algorithm for the time interval 1.11.2012-1.5.2017. In the diagram of TIPs, the black boxes represent periods of alarm, while each triangle surmounted by a number indicates the occurrence of a strong event (MM₀=5.6), together with its magnitude. Full red triangles indicate failures to predict. The blue star in the map and the blue arrow indicate the August 24^th 2016 Central Italy earthquake. The complete predictions archive is available at: www.geoscienze.units.it/esperimento-di-previsione-dei-terremoti-mt.html.

A) B)

Fig. 2. Areas alerted by M8S algorithm (in yellow) for a possible earthquake with magnitude 6.0≤M<6.5 determined for two time windows shifted by one year, namely: A) 1.7.2015 – 31.12.2015 and B) 1.7.2016 – 31.12.2016. The blue stars mark the epicenter of the August 24^th 2016 earthquake.

In Italy and surrounding regions the areas prone to strong earthquakes have been systematically identified based on the morphostructural zonation and pattern recognition analysis, considering two magnitude thresholds (M≥6.0 and M≥6.5) as described by Gorshkov et al. (2002 and 2004). The identified seismogenic nodes are used, along with the seismogenic zones ZS9 (Meletti and Valensise, 2004), to characterize the earthquake sources used in the seismic ground motion modelling, as described by Peresan et al.(2011). The earthquake epicenters reported in the catalogue are grouped into 0.2°x0.2° cells, assigning to each cell the maximum magnitude recorded within it. A smoothing procedure is then applied, to account for spatial uncertainty and for source dimensions. Only the sources located within the alarmed areas are considered to define the time-dependent scenarios (Panza et al., 2012; Peresan et al., 2011). From the set of complete synthetic seismograms, different maps that describe the maximum ground shaking at the bedrock can be produced, including peak ground displacement (PGD), velocity (PGV) or acceleration (PGA).

The August 24^th 2016 Central Italy earthquake took place within one of the seismogenic nodes previously identified as prone to possible earthquakes with magnitude M≥6.0 by Morphostructural Zonation and pattern recognition analysis (Gorshkov et al., 2002). This earthquake prone area experienced other destructive earthquakes in the past, including a M6.2 earthquake in 1639 and the large M6.9 Valnerina earthquake in 1703.

The seismic hazard maps at the bedrock defined by the Neo-Deterministic approach (NDSHA), correctly anticipated the recorded ground shaking. Specifically, in the NDSHA map published in 2001 (Panza et al., 2001) the predicted value is 0.15-0.3g, and the revision published in 2012 (Panza et al., 2012) gives 0.3-0.6g, which well contain therecorded values as high as 0.45g (RAN data: http://ran.protezionecivile.it/IT/dettaglio_evid.php?evid=340867).

This is quite natural since the August 24 earthquake did not necessarily generate the largest possible shaking in the area, as evidenced by the M6.9 Valnerina earthquake that struck the area in 1703, and should be factored in reconstruction considering that source and local effects may lead to values >0.6g. Notably, although the earthquake occurred in a well known seismic region, the ground shaking for this event exceeded (about 50% higher) at several sites the values expected at the bedrock according to current Italian seismic regulation (i.e. PGA<0.275g), which is based on a classical PSHA map (Gruppo di Lavoro, 2004) and overlooks source and site specific conditions. In conclusion, even when grossly adjusted for soil type corrections, following the rules given in the current Italian building code (NTC08, 2008), PSHA still underestimates the maximum recorded ground shaking by about 30%. NDSHA estimates, instead, when accounting for soil conditions, following the same rules given by NTC08 (2008), are well compatible with available recordings, and represent a serious warning to be considered in any future action.

The time-dependent ground shaking scenario associated to CN Central region (Fig. 3A) defined for the period 1 November 2012 – 1 September 2016, appears also well consistent with the ground shaking recorded for this earthquake. Since the time NDSHA time-dependent scenarios are regularly computed, namely starting on 2006, this is the third large earthquake that struck the Italian territory, along with L’Aquila earthquake. In all cases the method correctly predicted the observed ground motion, although L’Aquila earthquake scores as a failure in the earthquake prediction experiment, because the epicenter was located about ten km outside the alarmed territory.

A)B)

Fig. 3. Time-dependent scenarios of ground shaking associated to the alarm in CN Central Region (Fig. 1). On the left maps of peak ground velocity (PGV) are shown, computed using simultaneously all of the possible sources within the alarmed area and for frequencies up to 10 Hz. On the right, the same maps are provided, but for PGV>15 cm/s (modified after Peresan et al., 2015). The circle on maps A) and B) evidences the area within 30 km distance from the epicenter of the Central Italy earthquake (M6.2, 2016).

Conclusions

Prudent cost-effective actions can be taken if the prediction certainty is known, but not necessarily high (Davis et al., 2012; Peresan et al., 2012). The time-dependent ground motion scenarios associated with the areas alarmed by the CN and M8S algorithms on the territory of Italy, are routinely updated every two months since 2006. The prospective application of the time-dependent NDSHA approach provides information that can be useful in assigning priorities for timely mitigation actions and, at the same time, allows for a rigorous prospective testing and validation of the proposed methodology. A broad spectrum of interrelated actions can be undertaken for mitigation of earthquake impact on cultural heritage, including temporary safety measures and long term (tens of years) planning of interventions and retrofitting (Vaccari et al., 2009). A remarkable recent result of a practical application of NDSHA is the performed retrofitting of schools and strategic buildings of Provincia di Trieste (http://www.provincia.trieste.it/opencms/opencms/it/attivita-servizi/cantieri-della-provincia/immobili/Programma_verifiche_sismiche/).

The possible role of intermediate-term middle-range earthquake predictions in forecasting and planning safeguard interventions for cultural heritage, was discussed at the Conference “Resilienza delle città d’arte ai terremoti” (AccademiaNazionaledeiLincei,Rome, 3-4 November 2015).The above mentioned methodologies and results were recently presented both at the International Conference on “Data Intensive System Analysis for Geohazard Studies” (Sochi, 18–21 July 2016), as well as at the Moscow School for Young Scientists on "System Analysis and Seismic Hazard Assessment" (12-15 July 2016).

References

Allen, C.R., Edwards, W., Hall, W.J., Knopoff, L., Raleigh, C.B., Savit, C.H., Toksoz, M.N. and Turner R.H., 1976. Predicting earthquakes: A scientific and technical evaluation—with implications for society. Panel on Earthquake Prediction of the Committee on Seismology, Assembly of Mathematical and Physical Sciences, National Research Council, US National Academy of Sciences, Washington, DC.

Davis, C., Keilis-Borok, V., Kossobokov, V. and Soloviev, A.,2012. Advance prediction of the March 11, 2011 Great East Japan Earthquake: A Missed Opportunity for Disaster Preparedness. International Journal of Disaster Risk Reduction, 1: 17-32; DOI: 10.1016/j.ijdrr.2012.03.001

Gorshkov, A., Panza, G.F., Soloviev, A.A. and Aoudia, A.,2002. Morphostructural zonation and preliminary recognition of seismogenic nodes around the Adria margin in peninsular Italy and Sicily.JSEE. J. of Seismology and Earthquake Engeneering. Spring 2002, v. 4, no. 1, p. 1-24.

Gorshkov, A.I., Panza, G.F., Soloviev, A.A. and Aoudia, A., 2004. Identification of seismogenic nodes in the Alps and Dinarides, Boll. Soc. Geol. It., v. 123, p. 3-18.

Gruppo di Lavoro, 2004.Redazione della mappa di pericolosità sismica prevista dall’Ordinanza PCM 3274 del 20 marzo 2003. Rapporto conclusivo per il Dipartimento della Protezione Civile, INGV, Milano-Roma, april 2004, 65p. + 5 appendix.

Keilis-Borok, V.I., and A.A. Soloviev Editors, 2003. Non-linear Dynamics of the Lithosphere and Earthquake Prediction. Springer, Heidelberg, 337p.

Meletti, C. and Valensise, G., 2004.Zonazione sismogenetica ZS9 – App.2 al Rapporto Conclusivo. In: Gruppo di Lavoro MPS, 2004. Redazione della mappa di pericolosità sismica prevista dall’Ordinanza PCM 3274 del 20 marzo 2003. Rapporto Conclusivo per il Dipartimento della Protezione Civile, INGV, Milano-Roma, aprile 2004, 65p. + 5 allegati

NTC08, 2008, D.M. 14 gennaio 2008 – Norme tecniche per le costruzioni, Ministero delle Infrastrutture, in Italian, [online] http://www.cslp.it.

Panza, G.F., Romanelli, F. and Vaccari, F., 2001.Seismic wave propagation in laterally heterogeneous anelastic media: theory and applications to the seismic zonation.Advances in Geophysics, Academic press 43, p. 1-95.

Panza, G.F., Irikura, K., Kouteva, M., Peresan, A., Wang, Z. and Saragoni, R.(eds.), 2011. Advanced seismic hazard assessment.Pure Appl. Geophys., Topical Volume 168, 752p.

Panza, G.F., La Mura, C., Peresan, A., Romanelli, F. and Vaccari, F., 2012. Seismic Hazard Scenarios as Preventive Tools for a Disaster Resilient Society, Advances in Geophysics, Elsevier, London, 93–165. DOI http://dx.doi.org/10.1016/B978-0-12-380938-4.00003-3.

Peresan, A., Kossobokov, V., Romashkova, L. and Panza, G.F., 2005.Intermediate-term middle-range earthquake predictions in Italy: a review. Earth Science Reviews, v. 69, nos. 1-2, p. 97-132.

Peresan, A., Zuccolo, E., Vaccari, F., Gorshkov, A. and Panza, G.F., 2011.Neo-deterministic seismic hazard and pattern recognition techniques: time dependent scenarios for North-Eastern Italy. Pure and Applied Geophysics, v. 168, nos. 3-4, p. 583-607. DOI 10.1007/s00024-010-0166-1.

Peresan, A., Kossobokov, V. and Panza, G.F., 2012.Operational earthquake forecast/prediction.Rend. Fis. Acc. Lincei (2012) 23:131-138. DOI 10.1007/s12210-012-0171-7.

Peresan, A., Gorshkov, A., Soloviev, A. and Panza, G.F., 2015. The contribution of pattern recognition of seismic and morphostructural data to seismic hazard assessment. Bollettino di Geofisica Teorica ed Applicata, v. 56, n. 2, p. 295-328. DOI 10.4430/bgta0141.

Peresan, A., 2016. Recent developments in the monitoring of seismicity patterns for the Italian region. Chapter submitted to AGU Book on “Pre-Earthquake Processes: a multi-disciplinary approach to earthquake prediction studies”, Eds: Ouzounov, D., Pulinets, S., Hattori, K. and Taylor, P.

Vaccari, F.,Peresan,A., Zuccolo, E., Romanelli, F., Marson, C., Fiorotto, V. and Panza, G.F., 2009.Neo-deterministic seismic hazard scenarios: application to the engineering analysis of historical buildings. Atti del convegno PROHITECH 2009 - Protection of Historical Buildings Mazzolani (ed). Taylor & Francis Group, London. ISBN 978-0-415-55803-7, p. 1559-1564.

Volume 4, Number 2, June 2016. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

From the Editor……148

Special papers: The April 2016 M7.0 Kumamoto Earthquake, Japan

Subionospheric VLF propagation anomaly prior to the Kumamoto Earthquake in April, 2016, Masashi Hayakawa

and Tomokazu Asano…………….…273

Anomalies in jet-streams prior to the M6.6 Taiwan Earthquake on 5 February 2016 and the M7.0 Kumamoto

Earthquake on 15 April 2016, Hong-Chun Wu………….276

Solar activity correlated to the m7.0 Japan earthquake occurred on April 15, 2016.Gabriele Cataldi, Daniele

Cataldi and Valentino Straser………...279

The April 2016 M7.0 Kumamoto Earthquake swarm: Geology, thermal energy transmigration, and precursors,

Fumio Tsunoda and Dong R. Choi………286

FROM THE EDITOR

The special papers section discusses precursory signals that appeared prior to the devastating Kumamoto Earthquake. Although the mass media have shown little interest, this quake had been accurately predicted by two tools: jet stream (Wu, p. 276-278, five months prior) and VLF propagation anomaly (Hayakawa and Asano, p. 273-275, one week prior). They have again demonstrated that we are capable of predicting major earthquakes accurately and well in advance. These predictions are corroborated by Cataldi et al. (p. 279-285), who noteda solar activity anomaly (ionic density change in the solar wind) that appeared several days before the Earthquake.

SPECIAL PAPERS: THE APRIL 2016 M7.0 KUMAMOTO EARTHQUAKE, JAPAN

Editor’s note: The devastating Kumamoto Earthquake in April this year had been successfully predicted indepently by two methods; 1) jet stream anomaly appeared about five months prior by Wu, and 2) VLF electromagnetic wave propagation anomaly one week prior to the event by Hayakawa. This NCGT issue documents these successful predictions together with solar activity anomaly observed several days prior and geological analysis of this quake.

Subionospheric VLF propagation anomaly prior to the Kumamoto earthquake in April, 2016

Masashi Hayakawa¹ and Tomokazu Asano¹

¹Hayakawa Institute of Seismo Electromagnetics, Co. Ltd. (Hi-SEM), University of Electro-Communications (UEC) Incubation Center, 1-5-1 Chofugaoka, Chofu, Tokyo, 182-8585, Japan; E-mail: hayakawa@hi-seismo-em.jp, Phone/FAX: +81-42-444-6349.

Abstract: Strong earthquakes (EQs) occurred in the Kumamoto area of Kyushu Island on April 14 and 16, 2016

are definitely of the fault type. The occurrence probability of EQs with magnitude (M) greater than 7 (in the coming 30 years) was estimated to be less than 1% according to the national medium-term EQ prediction, but a series of EQs (with magnitude of 6.5 and 7.0) have taken place with a lot of casualties. Our short-term EQ prediction system based on the VLF/LF observation network of pre-EQ ionospheric perturbation has predicted an EQ in Kyushu Island, but our predicted M was 5.0-5.5 though the time and place were acceptable. This paper presents some preliminary analysis results, and we will indicate the reason why there was observed such a discrepancy in M estimation.

Keywords: subionospheric VLF propagation anomaly, 2016 Kumamoto EQ, fault-type EQ

(Received on 16 May 2016. Accepted on 25 May 2016)

1. Short-term EQ prediction system

It is a recent consensus that electromagnetic effects are the most powerful tool for the short-term EQ prediction (e.g., Hayakawa (ed.), 2012 and 2013; Hayakawa, 2015). Among the different kinds of possible electromagnetic precursors, we consider that the ionospheric perturbations as most promising because a clear statistical correlation was established between EQs and magnitude (M) greater than 6 and with shallow depth (Hayakawa et al., 2010). This is the reason why we have established a venture company to release such EQ prediction information to the public (but only in a closed society or only to the people and companies signed up to be a registration member of our company).

Fig. 1 illustrates our VLF/LF network in Japan, and there are eight VLF/LF observing stations all over Japan. We can list those stations from the north: NSB (Nakashibetsu), AKT (Akita), KTU (Katsuura), KMK (Kamakura), IMZ (Imizu), TYH (Toyohashi), and ANA (Anan). The VLF/LF receiver at each station is designed to receive simultaneously signals from several transmitters, in our case, two Japanese transmitters with call signs of JJY (Fukushima, 40kHz) and JJI (Miyazaki, 22.4kHz) and three foreign transmitters like NWC (Australia), NPM (Hawaii) and NLK (Seattle). In Fig. 1 we plot only the propagation paths associated with the JJI transmitter and the corresponding great-circle paths are indicated, because we are interested in the Kumamoto EQ in April, 2016.

Fig. 1. VLF/LF network. JJI is a Japanese VLF transmitter in south Kyushu, and eight receiving stations. A cross in Kyushu Island indicates the EQ epicenter.

2.Observational result

Fig. 2 illustrates the propagation characteristics for the propagation path of JJI-IMZ during about two months from 17 February to 16 April, because this propagation showed the most noticeable effect. The upper plot refers to the trend (TR) as characterized by the average nighttime amplitude normalized with its previous value (blue, during the previous 21 days and pink, 17days). The value is expressed in terms of standard deviation (σ). The second plot refers to the fluctuation in amplitude (dispersion, DP), with the same format as the upper one. Normally, an anomaly is known to be characterized by a decrease in TR and an enhancement in DP (Hayakawa et al., 2010 and 2016). It is clearly seen from this figure (upper plot) that the most important parameter of TR shows a significant decrease during the period of 4 to 7 April. Similar characteristics are seen for other propagation paths, so that we might consider that an EQ will happen in the Kyushu area about one week later (such as around 14 or so). Our forecast was acceptable in the two parameters of EQs (when and where), but the last parameter M was not so good because our prediction based on Fig. 2 and similar plots for other propagation paths, led us to conclude that the predicted M was 5.0-5.5 based on the degree of VLF propagation anomaly in Fig. 2. However, the actual M was M=7.0. As is clearly seen from Fig. 1 and the corresponding Fresnel zone for the JJI-IMZ, the discrepancy in estimating M is apparently due to the absence of any VLF/LF receiver in the north of Kyushu Island. If there were a receiver in Nagasaki in the north Kyushu, the anomaly in Fig. 2 would be expected to be much larger than -3σ or so because the EQ epicenter was just on the propagation path of JJI-Nagasaki.

Fig. 2. Observational results for the propagation path of JJI-IMZ during 17 February to 16 April, 2016. The upper panel refers to TR (trend) of average night time amplitude, and the second panel, its dispersion (DP).

3. Concluding remark

We could identify the precursory signatures on the subionospheric VLF/LF propagation data, because one of the two Japanese transmitters (JJI) is located in the south of Kyushu. The time and place we predicted were acceptable, but the M was considerably smaller than the actual value probably because of no VLF/LF receiver in the north Kyushu.

The results for the 2016 Kumamoto EQs may indicate the importance of our subionospheric VLF/LF network, but there have arisen a few difficulties. It seems impossible to predict any EQs all over Japan with our VLF/LF network consisting of only eight VLF/LF receivers. We can suggest a few possible ways to resolve this problem. As an example of this direction, we (or Hi-SEM) are now establishing a region-selected EQ prediction network. Namely, we pay particular attention to the Kanto (Tokyo) district with the expectation of huge EQs there, with the use of multi-parameter observations including VLF/LF subionospheric propagation, upper ionospheric sounding (vertical sounding of VHF transmitter signals, and oblique reception at remote stations of those VHF transmitter signals), ULF/ELF radiation and depression, line-of-sight VHF signals, etc.

References

Hayakawa, M. (ed.), 2012. The Frontier of Earthquake Prediction Studies, Nihon-Senmontosho-Shuppan, Tokyo, 794p.

Hayakawa, M. (ed.), 2013. Earthquake Prediction Studies: Seismo Electromagnetics, TERRAPUB, Tokyo, 168p.

Hayakawa, M., 2015. Earthquake Prediction with Radio Techniques. John Wiley & Sons, Singapore, 294p.

Hayakawa, M., Kasahara, Y., Nakamura, T., Muto, F., Horie, T., Maekawa, S., Hobara, Y., Rozhnoi, A.A., Solovieva, M. and Molchanov, O.A., 2010. A statistical study on the correlation between lower ionospheric perturbations as seen by subionospheric VLF/LF propagation and earthquakes. Jour. Geophys. Res., v. 115, A09305, doi:10. 1029/2009JA015143, 2010.

Hayakawa, M., Asano, T., Rozhnoi, A.A. and Solovieva, M., 2016. VLF/LF sounding of ionospheric perturbations in possible association with earthquakes. AGU volume, Pre-Earthquake Processes: A Multi-disciplinary Approach to Earthquake Prediction Studies, Ouzounov, D. et al. (eds.) (in press).

Anomalies in jet-streams prior to the M6.6 Taiwan Earthquake on

5 February 2016 and the M7.0 Kumamoto Earthquake

on 15 April 2016

Hong-Chun Wu^{1, 2}

¹ Institute of Labor, Occupational Safety and Health, Taiwan

² Formosa Science Center

wuhongchun094@gmail.com

Abstract: A jet stream is a rapidly flowing narrow air stream with almost horizontal axis in the upper troposphere or the low stratosphere. When the front of a jet stream remains at the same place for 6 hours or more, or at the intersection of wind speed contour, it implies the occurrence of abnormal precursors. The jet-stream was interrupted at the epicenter on 28 Dec 2015 at 12:00 UTC, 40 days prior to the major M = 6.6 Taiwan earthquake. It was also interrupted at the epicenter on 25 Nov. 2015 at 18:00 UTC, 140 days prior to the major M= 6.4 and M7.0 Japan earthquake. This broad time window between the jet-stream precursor appearance and the actual earthquake occurrence must be taken into consideration in future earthquake forecast using jet-stream precursors. It is possible to combine other precursors to make more precise forecasts.

Keywords: jet stream, earthquake, precursor

(Received on 6 May 2016. Accepted on 25 May 2016)

1.Introduction

The M6.6 Taiwan earthquake on 5 Feb. 2016 killed 117, and the M6.4 on 14 April and the M7.0 earthquakeon 15 April 2016 in Kyushu Japan killed 59 (Wiki web, 2016). To predict major earthquakes and warn the public in advance is one of the most important missions of earthquake scientists. This study introduces the jet stream precursors that had appeared prior to these two devastating earthquakes.

A jet stream is a rapidly flowing narrow air stream with almost horizontal axis in the upper troposphere or the low stratosphere. Wind speed is maximal on a jet axis. Linear size of a certain jet stream (length, width and depth) are determined according to wind speed contour (isotach) of 108 km/h (30 m/s). Usually, the length of a jet stream is several thousand kilometers, and its width could be hundreds of kilometers and the thickness could be 4-5 km. When a front of jet stream remains at the same place during 6 hours or more, or at the intersection of wind speed contour, imply the occurrence of abnormal precursors (Wu, 1999, 2004 and 2014).

According to the Lithosphere-Atmosphere-Ionosphere-Magnetosphere (LAIM) system, the crustal regions release radioactive elements such as radon (222Rn), and then reaction with air and water, producing the reaction heat. It results in temperature rise, and pressure drop, and finally, changes in the velocity line of jet streams (Pulinets et al., 2015).

2. The Taiwan earthquake on 5 February 2016

Satellite data observation found possible atmospheric disturbances in jet stream velocity (Wu, 2014) before the M = 6.6 Taiwan Earthquake on 05 Feb. 2016. The jet-stream was interrupted at the epicenter on 28 Dec 2015 at 12:00 UTC (Figure 1), 40 days prior to the major M = 6.6 shock, near the epicenter less than 60 km. It was posted on the author’s website:

https://www.facebook.com/photo.php?fbid=1032014433517272&set=pb.100001261760990.-2207520000.1462264459.&type=3&theater

Forecast parameters posted on 5 January 2016:

2015/12/28~2016/01/28 western Taiwan (23.5N120.7E) M>6.0

Actual event:

M6.6 2016-02-05 19:57:26 UTC 22.830°N 120.625°E 10.0 km

Figure 1. The anomalous behavior of jet stream: The original jet stream map (San Francisco State University), left figure, and the jet stream at a speed of 90 knots (162 km/hour), left figure. The jet stream was interrupted near the future epicenter on 28 Dec. 2015 at 12:00 (UTC). The epicenter was located near the interrupted region.

3. The 2016 Kumamoto Earthquakes, Japan

The jet-stream was interrupted at the epicenter on 25 Nov. 2015 at 18:00 UTC (Figure 2), 140 days prior to the major M= 6.4, M7.0 Kumamoto Earthquake, and the epicenter deviation was less than 40 km. It was posted on the web:

https://www.facebook.com/photo.php?fbid=1009833709068678&set=pb.100001261760990.-2207520000.1462264459.&type=3&theater

Forecast posted on 2015/11/26:

2015/11/25~2015/12/25 Southern Japan (33.1N131.0E) M>6.0

Actual events:

M6.4 2016-04-14 12:26:36 UTC 32.849°N 130.635°E 10.0 km

M7.3 2016-04-15 16:25:06 UTC 32.782°N 130.726°E 10.0 km

Figure 2. The anomalous behavior of jet stream: The original jet stream map (S.F. State University), left figure, and the jet stream (right figure) at a speed of 130 knots (234 km/hour), which was interrupted at the epicenter on 25 Nov. 2018 at 12:00 (UTC). The epicenter was located at the interrupted region.

4. Discussion and conclusion

As demonstrated in this study, the time lag from the precursor appearance to the earthquake occurrence shows a wide range; 40 days for the M6.4 Taiwan earthquake on 5 February 2016. It was 140 days for the M6.4 and M7.3 Japan earthquakes on 14 and 15 April 2016, respectively, and 96 days for the M8.3 Chile earthquake on 16 September 2015 (Wu, 2015). The advantage of jet stream precursor was precise for location, therefore it only confirmed M>6.0 earthquake, but not for magnitude, M>7.0 or M>8.0. Due to the longer time window, it only indicates impending an earthquake event.

This broad time window between the jet-stream precursor appearance and the actual earthquake occurrence must be taken into consideration in future earthquake prediction using jet-stream precursors. It is possible to combine the other precursors to make more precise forecasts.

References cited

Pulinets, S.E., Ouzounov, D., Arelin, A.K. and Davidenko, D., 2015. Physical bases of the generation of short-term earthquake precursors: A complex model of ionization-induced geophysical processes in the lithosphere–atmosphere–ionosphere–magnetosphere system. Gemagnetism and Aeronomy, v. 55, no. 4, p. 540-558.

Wiki web, 2016: https://en.wikipedia.org/wiki/2016_Taiwan_earthquake, https://en.wikipedia.org/wiki/2016_Kumamoto_earthquakes

Wu, H.C., 1999. Preliminary finding on perturbation of jet stream by earthquake. Chinese Taipei Geophysical Society Meeting, Taiwan, p. 429-434.

Wu, H.C., 2004. Preliminary findings on perturbation of jet stream prior to earthquakes. Eos Trans AGU;85:T51B-0455.

Wu, H.C. and Tikhonov, I.N., 2014. Jet streams anomalies as possible short-term precursors of earthquakes with M> 6.0. Research in Geophysics, Special Issue on Earthquake Precursors, v. 4, no. 1, p. 12–18. doi:10.4081/rg.2014.4939.

Wu, H.C., 2015. Anomalies in jet streams that appeared prior to the 16 September 2015 M8.3 Chile earthquake. New Concepts in Global Tectonics Journal, v. 3, no. 3, p. 407-408.

Solar activity correlated to the M7.0 Japan earthquake occurred

on April 15, 2016

Gabriele Cataldi

ltpaobserverproject@gmail.com

Radio Emission Project, Rome

Daniele Cataldi

ltpaobserverproject@gmail.com

Radio Emission Project, Rome

Valentino Straser

valentino.straser@alice.it

Independent Researcher

Abstract: The authors of this study wanted to verify a possible relationship between the M7.0 earthquake, which occurred near Kumamoto, Kyushu, Japan on April 15, 2016 at 16:25:06 UTC and the solar activity. The authors have scientifically demonstrated that there is a relationship between M6+ global seismic activity and solar activity and have had occasions to explain that large earthquakes, that can also generate tsunami, are always preceded by a variation of the interplanetary medium ion density; this variation has been defined by the authors as a "Interplanetary Seismic Precursor” (ISP) because, as already mentioned, it always precedes the strong earthquakes. The Interplanetary seismic precursors are phenomena that originate from the Sun through what the authors defined as "Solar Seismic Precursors" (SSP) and which are represented by: Coronal mass ejection (CME), coronal holes, solar flares and magnetic loops placed above the sunspots. The authors were able to confirm that from 2012 to today, all M6+ earthquakes that occur on our planet are always preceded by this type of event and, therefore, the authors had expected that the Japanese M7.0 earthquake was preceded by an ionic change in the solar wind.

Keywords: Coronal mass ejection, solar seismic precursors, earthquake prediction, interplanetary seismic precursor, solar wind.

(Received on 10 May 2016. Accepted on 22 May 2016)

INTRODUCTION

There are many data suggesting a potential relationship between the potentially destructive earthquakes on a global scale and solar activity (Straser et al., 2014-2015). Therefore, it is imperative to understand how this happens. The explanations that the scientific community has proposed from the 1970 to today have been numerous but what has been definitively understood is that the variations of the ion density of the interplanetary medium are the only physical phenomenon to have the highest specificity in the context of seismic prediction (Anagnostopoulos et al., 2010; Odintsov et al., 2006; Makarova and Shirochkov, 1999; Simpson, 1968; Nikouravan et al., 2012; Afraimovich and Astafyeva, 2008; Rabeh et al., 2014). Since 2011 the authors have started tracking the chemical and physical parameters of the solar wind near Earth (interplanetary magnetic field modulation; ion energy and concentration; polarity, speed, temperature and direction of the solar wind; dynamic pressure of the solar wind; magnetopause standoff distance) after noting, between 2009 and 2010, that strong earthquakes were preceded by an increase of natural radio emissions (SELF and ELF bands) and to check if these emissions were connected to geomagnetic and solar activities. The analysis of solar activity related to the analysis of natural radio emissions and M6+ global seismic activity allowed the authors to understand that the seismic activity is always preceded by an increase of solar activity, namely: that the strong earthquakes that occur on a global scale are always preceded by an increase of the solar wind ion density near Earth and by associated geomagnetic field disturbances. These conclusions indicate that the physical phenomenon responsible for the coupling between solar activity and seismic activity is a form of electromagnetic interaction that must be investigated in the coming years. According to the authors this is the new scientific approach that must be followed in the future to predict the potentially destructive earthquakes.

In this context, in 2013 the Italian Space Agency (ASI) and the China National Space Administration (CNSA) have signed an agreement for the construction of the CSES (China Seismo-Electromagnetic Satellite). The first satellite dedicated to monitoring electromagnetic field and waves, plasma and particles perturbations of the atmosphere, ionosphere and magnetosphere induced by natural sources and anthropocentric emitters; and to study their correlations with the occurrence of seismic events. This space mission underlines the importance of the studies that the authors have realized since 2010 and what is the future of scientific research on earthquake predictability. The study that was presented in this paper shows that the strong M7.0 Japanese earthquake occurred on April 15, 2016 was preceded by a solar wind proton density increase, began on April 12, 2016 at 14:15 hours UTC (74 hours before the earthquake). The authors constantly monitor the interplanetary medium ionic variation, and knew that the increase began on April 12, 2016, at 14:15 UTC would be followed by an M6+ seismic event and actually this ion increment preceded eight M6+ earthquakes occurred on a global scale, including the Japanese one. The authors chose to present in this work the M7.0 earthquake occurred in Japan on April 15, 2016 (Fig. 1) because this earthquake, unlike other earthquakes, occurred precisely during a strong variation of the interplanetary magnetic field (IMF).

Fig. 1. Index Map. M7.0 earthquake’s epicenter occurred on April 15, 2016 at 16:25:06 UTC (credits: USGS, http://earthquake.usgs.gov/earthquakes/eventpage/us20005iis#general).Concentric colored lines represent a measure (colorimetric scales) of the energy dispersion of the earthquake measured in magnitude (Mw).

METHODS AND DATA

To realize this study, the authors analyzed the space weather conditions (near Earth) and the characteristics of the geomagnetic field in the days that preceded the strong earthquake. In particular, the data taken into consideration were: data on the solar activity concern variation in the ionic density of the solar wind detected by the ACE (Advanced Composition Explorer) satellite orbiting the L1 point (Lagrange point) at 1.5 million kilometers from Earth; Solar Wind Density (ENLIL Heliosphere Ecliptic Plane), variations in interplanetary magnetic field or IMF (GOES); X-ray flux (GOES), temporal monitoring of CMEs events or Solar Coronal Mass Ejections (ISWA); monitoring of the coronal holes position on the Sun's surface (NSO/SOLIS-VSM Coronal Hole); Solar Wind Velocity (ENLIL Heliosphere Ecliptic Plane); Electron flux (NOAA/SWPC); Magnetopause Standoff Distance (CCMC/RT). The data on geomagnetic activity used for the study are represented by Kp-Index and were provided by Space Weather Prediction Center (SWPC).

RESULTS

The results of the study have confirmed the hypothesis of the authors: that the Japanese M7.0 earthquake was preceded by an increase of the interplanetary medium ion density (near Earth), and also confirming also what the authors ascertained from the 2012 data. Specifically, the M7.0 earthquake occurred after an increase of the solar wind proton density (Fig. 2) started on April 12, 2016 at 14:15 UTC.

251704320

Fig. 2. Solar wind proton density variation. Graph contains the data on the variation of solar wind proton density recorded between 8 and 17 April 2016 at the L1 Lagrange point by Advanced Composition Explorer Satellite; the variation of Kp-Index and the temporal markers (black vertical arrows) of M6+ earthquakes recorded in the same period. The vertical purple arrow represents the beginning of the “gradual” proton density increase (beginning of Interplanetary Seismic Precursor). The yellow areas surrounded by the red dashed line indicates increases of Kp-Index that preceded the M6+ earthquakes (Geomagnetic Seismic Precursor). The data on the proton density variation and the Kp-Index were provided by iSWA. iSWA is a flexible, turn-key, Web-based dissemination system for NASA-relevant space weather information that combines forecasts based on the most advanced space weather models with concurrent space environment information. The data on seismic activity were provided by United States Geological Survey (USGS).

DISCUSSION

By analyzing the proton density variation curve and the data on seismic activity provided by the USGS, it has emerged that indeed the M7.0 Japanese earthquake was one of eight seismic events occurred as a result of this proton increase. The energy proton fraction analyzed in this study is comprised between 310 and 1900 keV. Additional confirmation of the close connection between M6+ global seismic activity and solar activity comes from the analysis of the interplanetary magnetic field (IMF) variation (Fig. 3). Another important fact connected to the M7.0 Japan earthquake and, in general, connected to all M6+ global seismic activity is the increase of the Earth's geomagnetic field (Cataldi et al., 2013-2015). The increase of geomagnetic activity is thus a direct effect of increased solar activity. If the M7.0 Japan earthquake is truly related to the solar activity, it is clear that this earthquake had to be preceded by an increase of the Earth's geomagnetic field strength. In fact, the instrumental data confirm this thesis and also observations made by the authors from 2012 to today. If one analyzes the trend of the Kp-Index recorded between 9 and 13 April 2016, it becomes clear that the M7.0 Japanese earthquake (and more generally all eight earthquakes registered between 13 and 16 April 2016) was preceded by an increase in the Earth's geomagnetic field that began on April 11, 2016 at 09:00 UTC (Fig. 4). This phenomenon has been described by the authors as "Geomagnetic Seismic Precursor” (SGP) and is represented by a geomagnetic disturbance or variation. Analysis of DST and AL Index leaves no doubt in this sense (Fig. 5): the M7.0 Japan earthquake was preceded by a weak geomagnetic storm and by an increase of polar magnetic emission. The sequentiality of these seismic precursors (Solar Seismic Precursor, Interplanetary Seismic Precursor and Geomagnetic Seismic Precursor) is a direct evidence of a connection that exists between solar activity and M6+ global seismic activity; evidence that the authors have already had the opportunity to discuss at the international level for some years (Straser et al., 2015; Straser and Cataldi, 2014-2015).

Fig. 3. Interplanetary magnetic field (IMF) variation. The magnetogram contains the data on the interplanetary magnetic field (IMF) variation recorded between 13 and 15 April 2016. The recording was made by two satellites placed in geostationary orbit: GOES 13 and GOES-15. From the graph it is clear that the Japanese M7.0 quake was preceded by an increase in the interplanetary magnetic field (IMF) that started on April 15, 2016 at 06:00 UTC. The main feature of this increase is that has superimposed itself at the time at which it was recorded the M7.0 Japanese earthquake.

Fig. 4. Kp-Index variation. The graph contains the K-Index (recorded at College, Fredericksburg and Boulder) and Kp-Index data recorded between 9 and 16 April 2016. The Estimated 3-hour Planetary Kp-index is derived at the NOAA Space Weather Prediction Center using data from the following ground-based magnetometers: Sitka, Alaska; Meanook, Canada; Ottawa, Canada; Fredericksburg, Virginia; Hartland, UK; Wingst, Germany; Niemegk, Germany; and Canberra, Australia. The K-index, and by extension the Planetary K-index, are used to characterize the magnitude of geomagnetic storms. Kp is an excellent indicator of disturbances in the Earth's magnetic field and is used by SWPC to decide whether geomagnetic alerts and warnings need to be issued for users who are affected by these disturbances.

Fig. 5. Low-dimensional model of the energy transfer from the solar wind through the magnetosphere and into the ionosphere (WINDMI). The picture shows the variation of the AL-Index (at top) and the DST-Index (at bottom) in the hours that preceded the Japanese M7.0 earthquake occurred on April 15, 2016. The DST-Index is a direct measure of the Earth’s geomagnetic horizontal (H) component variation due to the equatorial ring current, while the AL-Index (Auroral Lower) is at all times, the minimum value of the variation of the geomagnetic H component of the geomagnetic field recorded by observers of reference and provides a quantitative measure of global Westward Auroral Electroject (WEJ) produced by increased of ionospheric currents therein present. The WINDMI data analysis showed that the Japanese earthquake occurred on April 15, 2016 was preceded by an increase of solar and geomagnetic activity. Model developed by the Institute for Fusion Studies, Department of Physics, University of Texas at Austin.

CONCLUSION

In conclusion we can confirm that the seismic events of strong intensity (M6+) that occur on a global scale are always preceded by an increase of the proton density of the solar wind. On average, considering the time intervals recorded from January 1, 2012 to April 30, 2016 (604 seismic event, USGS data), 140.9 hours elapse between the beginning of the increase of solar wind proton density increase and the earthquake. The M7.0 Japan earthquake occurred at a time of 74 hours from the start of the proton density increase and at a distance greater than 96 hours from the start of the Kp-Index increase that preceded it. We are convinced that the future of seismic prediction should take space weather, the solar phenomena and the heliophysics as well as the phenomena of geomagnetic nature, into serious consideration.

Acknowledgements: We wish to thank the reviewers for the constructive and useful comments on the early draft of this paper.

References

Afraimovich, E.L. and Astafyeva, E.I., 2008. TEC anomalies-Local TEC changes prior to earthquakes or TEC response to solar and geomagnetic activity changes? Earth Planets Space, no. 60, p. 961–966.

Anagnostopoulos, G., Papandreou, A. and Antoniou, P., 2010. Solar wind triggering of geomagnetic disturbances and strong (M>6.8) earthquakes during the November – December 2004 period. Demokritos University of Thrace, Space Research Laboratory, 67100 Xanthi, Greece. Cornell University Library.

Cataldi, G., Cataldi, D. and Straser, V., 2013. Variations Of Terrestrial Geomagnetic Activity Correlated To M6+ Global Seismic Activity. EGU (European Geosciences Union) 2013, General Assembly, Geophysical Research Abstracts, v. 15. Vienna, Austria. Harvard-Smithsonian Center for Astrophysics, High Energy Astrophysics Division, SAO/NASA Astrophysics Data System.

Cataldi, G., Cataldi, D. and Straser, V., 2015. Solar wind proton density variations that preceded the M6.1 earthquake occurred in New Caledonia on November 10, 2014. European Geosciences Union (EGU) General Assembly 2015, Natural Hazard Section (NH5.1), Sea & Ocean Hazard - Tsunami, Geophysical Research Abstract, v. 17, Vienna, Austria. Harvard-Smithsonian Center for Astrophysics, High Energy Astrophysics Division, SAO/NASA Astrophysics Data System.

Makarova, L.N. and Shirochkov, A.V., 1999. “On the connection between the Earth‟s magnetosphere magnetopause position and the earthquakes occurrence”, In: Abstracts of XXVI General Assembly LJRSI, Toronto, Canada, August 13-21, p. 755.

Nikouravan, B., Rawal, J.J., Sharifi, R. and Nikkhah, M., 2012. Probing relation between solar activities and seismicity. International Journal of the Physical Sciences, v. 7, no. 24, p. 3082-3088.

Odintsov, S., Boyarchuk, K., Georgieva, K., Kirov, B. and Atanasov, D., 2006. Long-period trends in global seismic and geomagnetic activity and their relation to solar activity. Physics and Chemistry of the Earth, no. 31, p. 88–93.

Rabeh, T., Cataldi, G. and Straser, V., 2014.Possibility of coupling the magnetosphere–ionosphere during the time of earthquakes. European Geosciences Union (EGU) General Assembly 2014, Geophysical Research Abstract, v. 16, Natural Hazard Section (NH4.3), Electro-magnetic phenomena and connections with seismo-tectonic activity, Vienna, Austria. Harvard-Smithsonian Center for Astrophysics, High Energy Astrophysics Division, SAO/NASA Astrophysics Data System.

Simpson, I.F., 1968. Solar activity as a triggering mechanism for earthquakes. Earth and Planet, Sci. Letter, v. 3, no. 5, p. 417-425.

Straser, V. and Cataldi, G., 2014. Solar wind proton density increase and geomagnetic background anomalies before strong M6+ earthquakes. Space Research Institute of Moscow, Russian Academy of Sciences, MSS-14. Moscow, Russia.

Straser, V. and Cataldi, G., 2015. Solar wind ionic variation associated with earthquakes greater than magnitude M6.0. New Concepts in Global Tectonics Journal, v. 3, no. 2, p. 140-154.

Straser, V., Cataldi, G. and Cataldi, D., 2015. Solar wind ionic and geomagnetic variations preceding the Md8.3 Chile Earthquake. New Concepts in Global Tectonics Journal, v. 3, no. 3, p. 394-399.

The 15 April 2016 Kumamoto Earthquake swarm: Geology, thermal energy transmigration, and precursors

Fumio Tsunoda

agatsuma.terao@gmail.com

Dong R. Choi

Raax Australia Pty Ltd. dong.choi@raax.com.au

International Earthquake and Volcano Prediction Center. dchoi@ievpc.org

Abstract: The devastating April 2016 M7.0 (UTC; M7.3<JMA, 2016>)Kumamoto Earthquake was an extraordinarily strong quake for inland quakes in Japan. It was generated in the southwestern slope of the Aso Volcano.The centrums of the earthquake swarms movedinside the Beppu-Shimabara Graben. The original energy of this quake has come from the 2010 deep Celebes Sea quakes with magnitudes 6.6 to 7.6. It transmigrated through a major thermal energy transmigration route or surge channel developed in the upper mantle from the Philippines, via Taiwan, to Kyushu, Japan (PJ route). The Mayon Volcano eruption in the Philippines in 2013 is attributed to this deep-Earth sourced energy. The Kumamoto quake is strongly related to volcanic activity in the graben represented by Aso Volcano – both are generated by the same energy. Therefore,seismicity of the Kumamoto earthquake swarm will weaken when the volcanic activity of the Aso Volcano declines. This study highlights the importance of energy flow concepts in generating earthquakes and volcanic activities (surge channel and the VE process).

Keywords: The 2016 Kumamoto Earthquake, surge channel, volcanic eruption, VE process, the 1965 Matsushiro Earthquake Swarm, Beppu-Shimabara Graben

(Received on 25 May 2016. Accepted on 21 June 2016)

Introduction

The April 2016 Kumamoto Earthquake killed about 50 people and injured about 3,000 people with an extensive property damages (https://en.wikipedia.org/wiki/2016_Kumamoto_earthquakes). Two strong foreshocks (IRIS) and a mainshock have been registered as follows:

Foreshocks:

14 April 2016 12:26:36 (UTC) M6.2 (JMA; M6.3) 10 km

14 April 2016 15:03:46 (UTC) M6.0(JMA; M6.5) 6 km

Mainshock:

15 April 2016 16:25:06 (UTC) M7.0 (JMA; M7.3)10 km

The Kumamoto Earthquake swarm that began on April 14, 2016 has been continuing today (24 May 2016; JMA HP, 2016). The number of felt earthquakes has exceeded 1,000 already. Moreover, the active region moved and spread: The epicenters of these earthquakes are distributed within the Beppu-Shimabara Graben (Matsumoto, 1979 and 2000; Matsumoto and Yamasaki, 1984) over the dotted area in Fig. 1. The lateral eastward spread was most rapid from 14 to 16 April, as seen in the M4.5+ quake distribution map (Fig. 2).

Strong thrust shock pushed up because the focus was shallow. As a result, faults that were near the surface of the Earth vibrated intensely. The number of the completely destroyed houses is 7,996 and the partial destroyed houses 19,100. Moreover, the number of the damaged dwelling houses is 73,035.

This quake was one of the largest inland quakes in Japan since 1970, Fig. 3. It implies a seismic energy accumulation and convergence in the epicentral area. It is also characterized by a close association with volcanic activities, particularly Aso Volcano, indicating the same energy source.

Fig. 1. The 2016 Kumamoto Earthquake, graben and volcanoes. Dotted area = fore-, main- and aftershocks. B. S. G. = Beppu-Shimabara Graben. Red star: Kumamoto Earthquake mainshock.

Figure 2. The Kumamoto Earthquake (represented by M4.5+ shocks), foreshocks on 14 April, mainshock on 15 April, and aftershocks spreading eastward on 15 to 16 April. Generated by IRIS website program.

Fig. 3. M7.0+ earthquakes in and around Japan. 1970 to 2016 (April). Kumamoto quake in 2016 is one of the largest inland quakes in Japan since 1970.

Activity process of the Kumamoto Earthquake

The volcano-earthquake (VE) activity in Beppu-Shimabara Graben (Fig. 1) changed as follows; volcanic earthquakes increased rapidly in 2013. Smoke of the Aso Volcano came to spout from 2014 to the height of 1,000m. From about June, 2015, volcanic earthquakes further increased. In addition, the southern front of Aso Volcano has begun to expandalong the fault of the Beppu-Shimabara Graben. An M6.2 earthquakeon April 14, 2016 was a harbinger, which was followed by series of major shocks including the mainshock on the following day. The seismic activity expanded throughout the Beppu-Shimabara Graben rapidly, Fig. 2 (Japan Meteorological Agency, 2016).

Precursory signals

Two distinctive precursory signals had been detected and publicly announced prior to the Kumamoto Earthquake: 1) the jet stream anomaly appeared 140 days before the mainshock near the epicenter, and 2) the VLF electromagnetic wave propagation anomaly appeared about one week prior. They are described in this NCGT issue, by Wu (p.276-278), and by Hayakawa and Asano (p. 273-275),respectively. It is noteworthy in the former that the time delay from the appearance of the jet stream anomaly to the mainshock occurrence was 140 days, which is the longest ever recorded (Wu, 2016). The prediction based on the VLF anomaly was also accurate except for magnitude (prediction, 5.0-5.5, and actual one 7.0): Hayakawa and Asano attribute this magnitude discrepancy to the missing receiving stations in Kyushu.

In addition, the junior author of this paper (DRC) noticed on the Japanese Himawari 8 satellite images that the electromagnetic energy discharges occurred actively mainly from 3 and 5 March (and intermittently after that) about 40-43 days prior to the mainshock along the southern graben boundary fault, including the Aso Volcano and its western area. The active discharge occurred again from 8 to 12 April, or 2 to 6 days prior in the same area. However, the strength and scale of energy leaks observed from the area before the quake is on much smaller scale compared to other M7.0 class quakes.

Apart from the above precursors, other signals such as the outgoing longwave radiation (OLR) had not shown noticeable anomalous indications. Sea surface temperature (SST) near Kumamoto City also showed no particular anomalies.

Geological background of the volcanic Kumamoto earthquake swarm

The geological development of this graben area is summarizedin Fig. 4; from A, B, C, D and E in ascending order.

(A) The mass infusion of andesite magma into the crust

The Conrad discontinuity which divides the Earth crust into the upper and the lower parts has weak coherency. Therefore, massive andesite magma invaded the Conraddiscontinuity in the early Miocene (Fig. 4-A; Tsunoda, 2001). The middle crust (MC in Fig. 4-A; Tsunoda, 2016 - this NCGT issue p. 174-193) was formed and gradually thickened. The upper crust (UC in Fig. 4-A)was domed and fractured.

(B) Massive volcanic eruption stage

The massive volcanic effusion which was named the Hohi volcanic rocks (Matsumoto, 1979; Matsumoto and Yamasaki, 1984) which are composed of a large quantity of pyroclastic rocks and lava erupted in the middle Miocene periods.

(C) The Beppu-Shimabara Graben (BSG) stage

As a consequence of the bulky magma extrusion, the middle crust became thin and its surface caved in. As a result of these events, the highly fractured upper crustsubsided in the hollow of the middle crust and the Beppu-Shimabara Graben (Matsumoto, 2000; BSG in Fig.4-C) was formed in the late Miocene. At this time, the both ends of BSG were cut by deep faults.

(D) Active volcano stage (late Pliocene)

Magma invaded into the crust again. The middle crust thickened and pushed up the upper crust. As a result, Daisen and Kirishima volcanic zones were formed (BSG; buried, VS; volcanic sediments, Pyr; pyroclastic rocks).

(E) The Kumamoto volcanic earthquake swarm stage

The most of centrums of Kumamoto earthquake swarm are distributed inside the BSG areawith depths ranging from 5 to 13 km (JMA, HP, 2016). On the other hand, the Curie point ofthe BSG area (Okubo et al., 1989), which expresses a high temperature state as some rocks melt, is roughly same with the place where the epicenters of the Kumamoto earthquake swarm are distributed. In other words, this earthquake swarm moved inside the BSG area both to the east and west. The characteristic of such Kumamoto earthquake closely resembles the Matsushiro volcanic earthquake swarm (Matsuzawa, 1976). The BSG in Fig.4-E shows the buried BSG. Why was the basement of the BSG area in a high temperature condition? It was because there was a supply of thermal energy.

Fig. 4. Developmental stage of the Beppu-Shimabara Graben. BSG = Beppu- Shimabara Graben. LC=Lower crust, MC=Middle crust, UC=Upper crust, V1=Unzen Volcano, V2=Aso Volcano, V3=Kuju Volcano, BSG=Beppu-Shimabara Graben (Fig.4-C), BSG (Fig.4-D and Fig.4-E)= graben-filling sediments, F=fault.

Transmigration of thermal seismic energy

We consider that the original energy of the Kumamoto quake has come from a swarm of the powerful July 2010 deep earthquakes in the Celebes Sea (see Choi, 2010 for details). This massive energy is responsible for the Mayon Volcano eruption in May 2013 (L2 volcanic explosivity index; Smithsonian HP, 2016), Fig. 5.Part of the same energy appeared again in Taiwan, which then moved to Kyushu.Please note an increased seismic and volcanic activity in 2014 to 2015 in the southern Kyushu,Fig. 6.Also note a clear shallowing trend of the 150 to 300 km earthquakes between Taiwan and Kyushu, Fig. 7.

Figure 5. Migration of VE activity along the PJ thermal transportation route or surge channel. A massive energy was released by deep (600 km) earthquakes in the Celebes Sea in July 2010 (five M6.6-7.6 quakes between 23 and 29 July 2010. Choi, 2010). The Mayon Volcano eruption three years after the Celebes Sea deep quakes was caused by the same energy.

Fig. 6. The M4.5+ earthquakes from 2007 to 2015. Note almost total lack in quakes in the Beppu-Shimabara Graben before the Kumamoto Earthquake in 2016 (compare with Fig. 2). Also note the 150-300 km deep earthquakes in the southernmost Kyushu – yellow circle, and the 70-150 km deep earthquakes – green circle.Overall increase in relatively deep quakes since 2013 in South Kyushu. We consider this area is part of the energy source for the Kumamoto quake.

Figure 7.The northward shallowing trend of relatively deep quakes, 150 km or deeper, and strong (M6.0+) quakes in the South China Sea from 1970 to 2016; near Taiwan – 254 km and Kyushu Island 160-170 km. The same trend is also confirmed in M5.5 to 6.0 quakes with the same depth range. While writing this paper (31 May 2016), another M6.4 quake rocked just north of Taiwan, which was 244 km deep (IRIS, ds.iris.edu/seismon/).We interpret this overall trend to indicate the northward shallowing surge channel from Taiwan to Kyushu. The earthquakes will represent the depth of the upper wall of the surge channel.<

The shallowing quake zone suggests the presence of a thermal energy transmigration route or surgechannel developed in the upper mantle. Note three active volcanoes in the Kagoshima Graben; Kaimondake, Sakurajima and Kirishima Volcanoes – so called the Kirishima Volcanic Zone. The earthquake depths (250 to 150 km, northward shallowing) would indicate the depth of upper wall of the surge channel. The surge channel is considered 20 to 40 km wide judging from the width of Kagoshima Graben.

The surge channel in the NE-SW direction, after passing the axis of perpendicular Precambrian geanticline (Fig. 8, Korea-Kyushu-Palau Ridge system, Choi, 1993), heads north. At this junction, the world renowned Hishikari Gold Mine (Kubota, 2016) has developed. It then enters the Shimabara-Beppu Graben, where major active volcanoes are nested, notably Aso Volcano (Fig. 8).Unzen, Aso and Kuju Volcanoes in the Beppu-Shimabara Graben belong to the Daisen Volcanic Zone (fig.5-2 of Minato, 1977). Kirishima Volcanic Zone and Daisen Volcanic Zone, diagonally intersecting each other, are separate volcanic zones.Judging from the earthquake depth trend and seismic tomographic profiles (Matsubara and Obara, 2011;see also fig. 8 of Kubota, 2016, this NCGT issue, page 199),there are two energy transmigration channels; one at around 150 to 170 km depth (fig.4 of Tsunoda et al., 2013) and anotherat20-30 km in the southern Kagoshima and theBeppu-Shimabara Grabens.

>

Fig. 8. Energy transmigration route from the south, Kagoshima Graben (dotted by Quaternary volcanoes), via eastern boundary of the Korea-Kyushu-Palau Ridge Geanticline (where gold mines are swarmed), and entered theBeppu-Shimabara Graben. Base map from Choi (1993).

In reference to the opinions such as Wever and Davis (1990), Tsunoda (2009) and Tsunoda et al. (2013),there are three low velocity layers in the shallower depth above 400km. Tsunoda (2010) assumed the existence of high temperature transportation route. Moreover, as stated earlier, there was a powerful heat supplyin 2010 from Celebes Sea throughthe Philippine-Japan (PJ) route (Tsunoda et al., 2013 and 2015). In other words, thermal energy that produced the Kumamoto Earthquake is the same energy that let the MayonVolcano erupt in 2013,which was the biggest in the 20th century along the thermal PJ route (Tsunoda, 2010; Tsunoda et al., 2013).

In Kumamoto, the variationin the GPS data at the southern foot of the Aso Volcano occurred alongadeepfault bounding the southern margin of the Beppu-Shimabara Graben. The direction of expansion of GPS variation accords with that of the boundary fault (Fig.9). This can be explained by thatground temperature rose by the arrival of thermal energy, which resulted in heating of the fault plane, and reactivated thefault zone. This happened about half year prior to the Kumamoto Earthquake, judging from the ground movement record.

We will present more detailed accounts on this topicby comparing the Kumamoto Earthquake, 2016, andthe Matsushiro Earthquake swarmin the central Japan (Fig. 1) in 1965 to 1966 in the near future.

Fig. 9. Reactivation of the deep-fault of the Beppu-Shimabara Graben.

Conclusions

1. The Kumamoto Earthquakes in April 2016 is primarily a volcanic earthquake closely related to the Aso Volcano.

2. Its energy was supplied originally from the deep Earth in Celebes Sea which wasmanifested in July 2010 as a series of M6.7 to 7.6 earthquakes.

3. The thermal energy transmigrated through a surge channel developed in the upper mantle along the Philippines-Taiwan-Kyushu (Japan) route (PJ route). This energy, originated in 2010 in deep Celebes Sea, is responsible for the Mayon Volcano eruption in 2013, strong quakes in Taiwan and southern Kyushu in 2013-2014, and reawakened volcanoes in Kagoshima Graben and Aso Volcano in Beppu-Shimabara Graben in 2015-16.

4. The Kumamoto Earthquake had been predicted by two precursory signals – jet stream anomaly (150 days prior) and VLF wave propagation anomaly (one week prior). Other signals, however, were not strongly manifested as compared to the magnitude 7.0 class earthquakes in the past.

5. This study proved the control of deep geological structures on earthquake/volcanic activities, and emphasized the important concept of thermal energy flow through surge channels and the VE process in studying earthquake and volcanic activities, hence their prediction.

Acknowledgements:This paper was built up onmany laborious field works made by Japanese geologists; studies on grabens and mineral deposits in Kyushu, Japan, especially by M. Matsumoto and Y. Kubota. Our special thanks are offered to Y. Kubota for providing us with precious Hishikari Gold Mine information and geology of the region.

References

Blot,C.,1976.Volcanismeetsismicitedanslesarcsinsulaires.Previsiondecesphenomenes.Geophysique13,

ORSTOM,Paris,206p.

Choi, D.R., 1993. On the Korea-Kyushu-Palau Ridge System. Hokuriku Geology Institute Report, no. 3, p. 123-132.

Choi, D.R., 2010. Blot’s energy transmigration concept applied for forecasting shallow earthquakes: a swarm of strong

deep earthquakes in the northern Celebes Sea in July 2010. NCGT Newsletter, no. 56, p. 75-85.

Hayakawa, M. and Asano, T.,2016. Subionospheric VLF propagation propagation anomaly prior to the Kumamoto

Earthquake in April, 2016. NCGT Journal, v. 4, no. 2, p. 273-275.

Japan Meteorological Agency HP, 2016. Earthquake and volcanism (in Japanese; http:

//www. jma.go.jp/jma/index.html). http: //www. jma.go.jp/jma/indexe.html (in English).

Kubota, Y., 2016. Spatial distribution of high grade epithermal Quaternary gold deposits at Japanese island arc

junctions and their global implications. NCGT Journal, v. 4, no. 2, p. 194-203.

Matsumoto, Y., 1979. Some problems on volcanic activities and depression structures in Kyushu, Japan. Mem. Geol. Soc. Japan, no. 16, p. 127-139. (in Japanese with English abstract)

Matsumoto, Y., 2000. Geotectonic development of the Beppu-Shimabara Graben in central Kyushu, Japan. NCGT Newsletter, no. 16, p. 16-20.

Matsubara M. and Obara K., 2011. The 2011 Off the Pacific Coast of Tohoku earthquake related to a strong velocity gradient with the Pacific plate. Earth Planets Space, v. 63, p. 663-667.

Matsumoto, Y. and Yamasaki, T., 1984. Volcanic stratigraphy and structure of geothermal areas in Kyushu. Res. Nat. Energy, no. 8, p. 465-472.

Matuzawa, T., 1964. Study of earthquakes. Uno Shoten, Tokyo, 213p. (in Japanese)

Matsuzawa, T., 1976. Matsushiro Earthquake swarm. The theory of the earthquake and the application (in Japanese), p. 39-58, University of Tokyo Press, 223p.

Meyerhoff, A.A., Taner, I., Morris, A.E.L., Agocs, W.B., Kamen-Kaye, M., Bhat, M.I., Smoot, N.C., Choi, D.R. and Meyerhoff-Hull, D. (ed.), 1996. Surge tectonics: a new hypothesis of global geodynamics. Kluwer Academic Publishers, 323p.

Minato, M. (ed.), 1977. Japan and its nature. Heibonsha Ltd., Tokyo, 220p.

Okubo,Y., Tsu H. and Ogawa, K., 1989. Estimation of Curie point temperature and geothermal structure of island arcs of Japan. Tectonophysics, v.159, p. 279-290.

Smithsonian HP, 2016. volcano.si.edu/search_eruption.cfm

Takeda,T., Sato,H., Iwasaki,T., Matsuda,N., Sakai,S., Iidaka,T. and Kato,A., 2004. Crustal structure in northern Fossa Magna region, central Japan, modelled from refraction/wide-angle reflection data. Earth Planets Space, v.56, 1293-1299.

Tsunoda,F., 2001. Middle Pleistocene uplift of the South Fossa Magna Region. Himalayan Geology, v.22, p. 17-25.

Tsunoda, F., 2009. Habit of earthquake Part 1. Mechanism of earthquakes and lateral seismic energy transmigration. NCGT Newsletter, no. 53, p. 38-46.

Tsunoda, F., 2010. Habits of earthquakes. Part 2. Earthquake corridors in East Asia. NCGT Newsletter, no.54, p. 45-56.

Tsunoda, F., 2016. Origin of the central Honshu Arc and the Izu Ridge, Japan. NCGT Journal, v. 4, no. 2, p. 174-193.

Tsunoda, F., Choi, D.R. and Kawabe, T., 2013. Thermal energy transmigration and fluctuation. NCGT Journal, v. 1, no. 2, p. 65-80.

Tsunoda, F., Choi, D.R. and Kawabe, T., 2013. Thermal energy transmigration and fluctuation. NCGT Journal, v. 1, no. 2, p. 65-79.

Tsunoda, F., T. Kawabe, M. Hayakawa and Dong, R. Choi, 2013. Tendency of volcano-seismic activity developed in the central part of the Honshu Arc, Japan. NCGT Journal, v. 3, no. 1, p. 34-42.

USGS HP, 2016. earthquake.usgs.gov/earthquakes/search

Wever, M. and Davis,J.P., 1990. Evidence of a laterally variable lower mantle structure from P– and S- waves. Geophys. Jour. Inst., v.102, 231-255.

Wu, H.-C., 2016.Anomalies in jet-streams prior to the M6.6 Taiwan Earthquake on 5 February 2016 and the M7.0 Kumamoto Earthquake on 15 April 2016. NCGT Journal, v. 4, no. 2,p. 276-278.

Volume 3, Number 3, September 2015. ISSN 2202-0039. Editor: Dong R. CHOI (editor@ncgt.org). www.ncgt.org

Special papers: The 16 September 2015 M8.3 Chile Earthquake

Analysis of psychrometric parameters associated with seismic precursors in Central Chile. Ariel R. CÉSPED… 383

Blot’s energy transmigration law and the September 2015 M8.3 Coquimbo Earthquake, Chile. Dong R. CHOI and

John CASEY………....387

A surge and short-term peak in northern solar polar field magnetism prior to the M8.3 earthquake near Chile on

September 16, 2015. Ben DAVIDSON…….....391

Solar wind ionic and geomagnetic variations preceding the M8.3 Chile Earthquake. Valentino STRASER, Gabriele

CATALDI and Daniele CATALDI………….394

Outgoing longwave radiation anomaly prior to big earthquakes: A study on the September 2015 Chile Earthquake.

N. VENKATANATHAN, Philip PHILIPOFF and S. MADHUMITHA………400

Space weather conditions prior to the M8.3 Chile Earthquake, Kongpop U-YEN…...405

Anomalies in jet streams that appeared prior to the 16 September 2015 M8.3 Chile Earthquake. Hong-Chun WU…..407

THE 16 SEPTEMBER 2015 M8.3 CHILE EARTHQUAKE

Editor’s note: The latest great earthquake in Chile, South America was successfully predicted by a concerted effort of world scientists. It was initially warned more than one year prior, and a short-term warning was given 21 days prior to the mainshock. It was made possible by well-established, but relatively little-appreciated, concepts and a multiparameter approach. The entire procedures taken by the group will set a new direction for the future earthquake prediction practice. Another significance of this success is that it has broken the myth that ‘No one can predicted earthquakes’, which has been claimed by world seismological authorities. Due to time constraint, this NCGT issue includes only some short preliminary reports and introduces precursory signals which had appeared prior to the mainshock.

ANALYSIS OF PSYCHROMETRIC PARAMETERS ASSOCIATED WITH SEISMIC PRECURSORS IN CENTRAL CHILE:

A NEW EARTHQUAKE OR THE GREAT 2010 MAULE M8.8 AFTERSHOCK?

Ariel R. CÉSPED

Principal Researcher, Chilesismos

chilesismos@gmail.com

(Author’s note: The original of this paper was issued on August 26, 2015 – 21 days prior to the mainshock of the Chilean earthquake on September 16, 2015)

Abstract: This paper shows the results of the analysis on psychrometric parameters associated with earthquake precursors in the central area of Chile, during the period of July - August 2015. A comparative study found equivalences between the behavior of the current readings and that occurred in preceding days of the Maule M8.8 mega-earthquake in 2010, which allowed to infer the possibility of a significant earthquake for the period of September - October, 2015.

Keywords: earthquake prediction, psychrometrics, atmospheric research

LITHOSPHERE-ATMOSPHERE ELECTRO-CHEMICAL CONNECTION

State-of-the-Art studies in the field of atmospheric physics suggest that areas near active geological faults show a seismic preparation process in days or weeks before earthquakes, emitting chemical and electromagnetic signals such as radioactive gases (mostly Radon Rn222), ions, "p-holes", and electromagnetic pulses (Rozhnoi et. al., 2009; Freund, 2010; Pulinets, 2011). Ions and free electrons released by the Earth's crust, possibly associated with piezoelectric phenomena in the area of greater stress, or product of the induced radiolysis by the radioactive decay of Radon, saturate the sub-soil and migrate to the surface. The gradient of potential between the ground (cathode) and the ionosphere (anode) (which in standard conditions is up to 100 Volts/m) activates the relative motion of these particles through the air, and eventually collide with the airborne water molecules, breaking electrical links and triggering an "avalanche" of corpuscles (Townsend Discharge) in a large scale. Thermodynamical instability of these radicals is quickly balanced (the average reaction time is approximately 11 nanoseconds) by a mechanism of neutralization of ion induced nucleation (IIN), bringing together the atoms in suspension and developing aerosols of several nanometers in diameter.

This process occurs with a large associated energy transfer process consisting of latent heat, which is released into the atmosphere. Inherent phase change alters the amount of water vapour in the air, thus, the relative humidity in areas close to the future epicenter spontaneously. If the phenomenon described above is sufficiently extensive, the variations could finally reach the upper levels of the atmosphere, re-setting properties of the cloud layer (linear precursor clouds) (Zhong, 2014) and the high-speed winds in the high troposphere (jet-streams) (Wu, 2014) until they can be visible from space by satellites.

EXPERIMENTAL PROCEDURE

To identify the possible action of the mechanism described above, an analysis of some psychrometric parameters associated with seismic precursors is performed. The meteorological database is provided by the Dirección Meteorológica de Chile at the Carriel Sur, Talcahuano (36.7S, 73.0W) station. Taken Periods of studies are (1) January-February 2010 and (2) July-August, 2015.

In order to detect a possible participation of radicals in the composition of electro-chemistry of the air and the presence of induced nucleation processes, an analysis of the corrected chemical potential (ΔU) (Boyarchuk et. al., 2006) is suggested, in accordance with the equation (1):

Where TS and HR are the temperature in °C and relative humidity of the air respectively. Then it is necessary to check possible alterations associated with the process of stabilization of ionized particles and its interaction energy with water vapour in the air through an analysis of the condensation latent heat L(T) (Roger and Yau, 1989) according to the equation (2):

Where T_D is the condensation temperature of water, considering, in this case, the dew point in °C. If this process reaches a level of considerable influence, the traces of latent heat around the area over the future epicenter should therefore be traceable from space via satellite. To verify this, an analysis of behavior of the surface latent heat flux (SHLF) is performed. The data is obtained from the database of Surface Flux Re-analysis webserver, managed by the National Center of Environment Prediction (NCEP). Finally the three parameters ΔU, L(T), and SHLF are evaluated through the anomaly index, according to the following algorithm (3):

Where x is the daily value of the sample, and x and  average and standard deviation of the data series. A threshold was established to discriminate the anomalous values equal to x 2.5. The results are tabulated and graphed for further analysis.

RESULTS

The following are the comparative results of the ΔU, L(T), and SHLF analysis respectively, for the period January-February 2010 and July-August 2015 in Bío-Bío and surrounding regions.

Chart No. 1 (left) shows a significant variation of ΔU occurred on January 30, 2010, twenty-eight days before the Maule M8.8 earthquake. Chart No. 2 (right) shows an analogue anomaly occurred on August 16, 2015.

Chart No. 3 (left) shows an important variation in L(T) occurred on January 30, 2010, twenty-eight days before the Maule M8.8 earthquake. Chart No. 4 (right) shows an analogue anomaly occurred on August 16, 2015.

Chart No. 5 (left), shows a SHLF anomaly on January 02, 2010, three days after the increase of the ΔU and L(T) and twenty-six days before the Maule M8.8 earthquake. Chart No. 6 (right) shows a SHLF anomaly on August 16, 2015, the same day as the detection of ΔU and L(T) anomalies observed in Chart No. 2 and Chart No. 4

BLOT’S ENERGY TRANSMIGRATION LAW AND THE SEPTEMBER 2015 M8.3 CHILE EARTHQUAKE

Dong R. CHOI¹ and John CASEY²

¹ International Earthquake and Volcano Prediction Center, Australia. dchoi@ievpc.org, www.ievpc.org

² International Earthquake and Volcano Prediction Center, USA. jcasey@ievpc.org;

Veritence Corporation. mail@veritence.net

Abstract: The powerful earthquake which rocked south of Coquimbo, Chile on 16 September 2015 with a tsunami was successfully predicted by a concerted effort by International Earthquake and Volcano Prediction Center (IEVPC) and local seismologists. The initial prediction more than one year prior by IEVPC was based on the energy transmigration concept which allows to predict the delay time of earthquake energy from deep to shallow Earth. A team of local seismologists, Chilesismos, who continued monitoring short-term signals found psychrometric anomalies about one month prior and publicly warned 21 days prior to the mainshock. The Coquimbo quake prediction exercise has proven that catastrophic earthquakes are predictable if we take a right approach armed with right earthquake model and tools supported by a synergetic international cooperation. The formation of a well-funded international research organization is in order to save human lives and help mitigate damages.

Keywords: September 2015 Coquimbo Earthquake, earthquake prediction, Blot’s ET law, seismic energy convergence

Introduction

The International Earthquake and Volcanic Prediction Center (IEVPC) has been issuing long to medium-term earthquake alerts in the last few years as part of its test program. Their major tool is the Claude Blot’s (1976) energy transmigration concept or the ET law (Grover, 1998). It links the deep and shallow earthquakes, and allows to predict the time of shallow appearance 3 to 5 years before the major earthquakes hit.

Since 1973 USGS recorded a total of 44 deep (300km+) very strong (7.0+) shocks worldwide. After a few years, almost all of them had appeared at shallow depths as catastrophic earthquakes with similar or stronger magnitude (IEVPC internal data); their genetic link between the deep and shallow shocks were verified by the ET law. Today we are seeing the shallow appearance of the 2011 deep energies worldwide. The latest Chile quake is one of those examples.

All of the deep quakes, 300 km or deeper, have occurred along major deep rooted fault zones, hence they appear in the Circum Pacific region; South and Western Pacific, and South America. By analysing geological structure, the loci of shallow appearance can be roughly predicted, because we know that the deep energy transmigrates along deep fractures and is trapped in culminated structures (Choi, 2011; Choi and Kubota, 2015), although many areas especially oceanic areas are geologically poorly controlled which makes it difficult to judge in which way the energy will flow. In the Pacific coast of South America, we have clarified southward seismic energy propagation in the upper mantle and lower crust. In the Central American coast, the direction of energy movement was found to be controlled by solar cycle (Choi, 2014). Many other studies mainly by our NCGT colleagues proved the energy transmigration phenomena under the mobile belts (Tsunoda et al., 2013, and others). Today energy flow in the mantle is a well-established fact.

In the above backdrop, IEVPC analysed a strong (M7.0) deep shock that occurred near Santiago del Estero, Argentina in January 2011 (Fig. 1), and issued a warning in early September 2014, which was lifted at the end of December. In 2015 our study was followed up by a team of local seismologists who is properly equipped with earthquake analysis tools and concepts. The present paper briefly introduces the scientific ground which led to the announcement of the Coquimbo warning in September 2014 on the IEVPC website as one of its test programs, #004-09-01-14, http://www.ievp.org/.

The ET law applied to a deep precursory shock

Fig. 1 illustrates the ET analysis with deep faults and energy flow direction. This figure was circulated among the IEVPC and other colleagues on 10 September 2014. For calculation purpose, based on geological data, a hypothetical shallow epicenter was placed at the 30-35 km depth in the coast between Coquimbo and Valparaiso where NE-SW block fault crosses. The main reason of this site selection was that a large structural culmination is present in the north or the Coquimbo area, therefore the seismic energy would move into this structure and be trapped, which is a prerequisite for major earthquakes to be generated. Based on geological analysis of numerous powerful earthquakes, we have learnt that the mainshocks tend to occur at a deep fault zone bounding major structural culminations (Choi, 2011 for example). Interestingly, the actual mainshock of the September 2015 quake occurred right near this point.

Figure 1. Deep fracture zones and energy flow of the study region. Prepared and circulated as an internal confidential document among the IEVPC colleagues and some selected researchers on 10 September 2014, almost one year prior to the actual occurrence. Base map from the IRIS site http://ds.iris.edu/seismon/.

The ET formula resulted in the following shallow appearance time:

Appearance at 20 km, 15 June, 2015

Appearance at 25 km, 27 February, 2015

Appearance at 30 km, 1 December, 2014

Appearance at 35 km, 17 September, 2014

Based on these results mainly at the depths from 30 to 35 km, IEVPC issued a first warning on 1 September, 2014 with the following parameters.

Epicenter; Coastal area between Coquimbo and Valparaiso

Magnitude: M7.0 to 8.3

Depth: 10 to 60 km

Time of occurrence: Before the end of December 2014

Southward energy flow

In analysing the Coquimbo quake, it is essential to take another energy flow into account – along the Pacific coast of the South America from north to south. This energy flow was identified by the latitude-time diagram of strong earthquakes (M7.0 or greater), Fig. 2 (Choi, 2014).