News
 
Banner place
 

Report of Pakistan board of GNFE

Earthquake Forecasting by Gravity Variations Precursors Recorded at ATROPATENA-PK System
M.A Mubarak, M.Qaisar, Nabeel Ahmad, M.Awais, M.Shahid, M.Irfan

ABSTRACT

The gravity anomalies, prior to major earthquakes, is a very well established fact and have been noticed in cases of many major earthquakes. The ATROPATENA system which is based on principle of Cavendish balance is developed by scientists, working at Scientific-Research Institute on prognosis, International Academy of Sciences, Baku, Azerbaijan to measure such gravity variations. One station of the ATROPATENA system network was installed at the end of January, 2009 at Center for Earthquake Studies (CES), Islamabad Pakistan to study gravity anomalies for short-term forecasting of strong earthquakes. Six months data recorded at ATROPATENA-PK from March to August 2009 was analyzed. The forecast of earthquakes were made on the basis of pattern recorded on three components of the ATROPATENA system. Different patterns are recognized from the recorded signals originated from different regions and defined the region of future earthquake. The over all success ratio was more than 65% which is a good achievement for successful forecasting earthquakes.

INTRODUCTION

The study of gravity anomalies in the perspective direction of short-term forecasting of earthquakes is the most important aspect of researches in the field of geodynamics. Recently, significant anomalies were observed not only near the epicenter, but also in the region very far from the epicenter before the occurrence of strong earthquake. Such anomalies in the epicenter region are known as “source precursor” and those far away from the epicenter as “field precursors” (Ma Zongjin, 1980). Such precursors have been noticed by different seismologists, ranging from behavioral changes in animals to changes in Lithosphere, Atmosphere and Ionosphere. One of these types of earthquake precursor is due to changes in gravitational field and this is noticed as “field precursor” (Walsh and Rice, 1979). The variations in gravitational field may be stipulated by a number of geophysical and tectonic reasons (Khain and Khalilov, 2006 and Khalilov, 2007). These are i) Stress conditions of the earthquake preparation zone when approached to the critical level, it causes either squeezing/compaction of the rocks or stretching that result in mass cracking of rocks in earthquake preparation zone, which eventually causes breaking of rocks with increase/decrease of densities. ii) High and low density area appears due to deformational processes arising in the central earthquake preparation zone before the occurrence of strong earthquake. iii) The critical stresses in earthquake preparation zone causes active movements of fluids in the layers of the Earth and as a result either it increases or decreases the level of subsoil waters that has been observed in the shafts and bores before the occurrence of earthquakes. Probably, there are also other factors involved in changes of gravity, but all of them don’t have large radius of range near earthquake preparation zone of strong earthquake. It is due to fact that this effect of changes of gravity, connected directly with the geodynamical processes in earthquake preparation zone, is observed in the radius from tens to thousands of kilometers from station of registration. The system ATROPATENA, designed to monitor such type of variations of gravity, is installed at CES Islamabad, Pakistan. Current study is based on the analysis of six months data (March 2009 to August 2009) registered at ATROPATENA-PK.

ATROPATENA SYSTEM

The ATROPATENA system (Fig. 1) consists of two Torsion detectors to register variations in gravitational field in two horizontal directions – NS and EW and a gravimeter to detect the variation in vertical component of gravitational field. Two sets of small masses are attached at the end of two bars of low density material. These small masses and one vertical gravimeter are placed in a jar. The system of detectors is completely isolated, due to it’s highly sensitiveness, from the environment by means of vacuum and registered very weak displacements of sensitive elements of the system. The system is set in equilibrium by two heavy masses placed outside the jar, to eliminate the effect of any small changes in gravitational field, not related to any geodynamic phenomenon. As a whole the system ATROPATENA registers the variations of gravitation field in three perpendicular directions – X, Y, Z.

Figure 1: Schematic diagram of ATROPATENA

Recording these variations is done by using laser beams and optical matrix. There are small mirrors attached to each bar and gravimeter. The displacement of the bars is noted by using laser beams directed on these small mirrors. The laser beams reflected from mirrors to the optical matrixes. Then these changes in positions are recorded by the sensitive cameras attached on back of optical matrixes. The analogue signals are then converted to digital form by software Power Graph is transferred to the computer for recording.

DATA RECORDING AND UPLOADING TO SERVER

Gravity variations signal generated before occurrence of strong earthquake is recorded after every one second at CES station. The recorded signal is then stored in Microsoft SQL server locally. After every ten minutes, the recorded data from the network as shown in Fig.2 is being uploaded to database server, based in Canada, through TCP after automatic formatting. The data is now available to download by web interface and ready to analyze.

Figure 2: Network of System ATROPATENA with their server based at USA

DATA ANALYSIS

The data acquired during March 2009 to August 2009 from ATROPATENA system installed at CES, Islamabad, Pakistan was analyzed. The unusual Low-frequency changes in the gravitational field registered at the system were observed before the occurrence of strong earthquakes. This phenomenon was observed when the epicenters are at large distances (in the radius from one thousand to tens of thousands km) from the registering station. Some peculiarities were also observed during registration of signals, which allow the increase of accuracy of the forecast. The detail of six months data is given in Table 1-6.

The statistical analysis of recorded data shows that the gravitational signals were registered on the average 7-15 days (90%) before the occurrence of strong earthquakes. Some of the recorded variations of gravity before strong earthquakes during March 2009 to August 2009 are shown in Figs. 3-9. These graphs shows, in most of the cases before the distant strong earthquakes, there is at first decrease of gravity and then increase of gravity. It is also observed “vibration of the record” – relatively high-frequency oscillations of gravimeter readings which is stopped right away after the earthquake. However, in some cases before the occurrence of distant strong earthquakes, the changes of anomalies of gravity have more complicated character.

Starting from March 2009 to August 2009, total 57 events were forecasted numbers of earthquakes and out of these 57 the successful forecasts were 36 on the basis of strong gravity variations precursors. Six forecasts were not possible either due to weak or no signals recorded at ATROPATENA-PK and 12 forecasts were also impossible due to unknown pattern (pattern which has not yet been standardized or recognized). Only three earthquake forecast were unsuccessful. The success ratio during the months of March, April, May, June, July and August was 50%, 89%, 78%, 43%, 60% and 80% respectively.

The summary of analyzed data is shown in Table-7 and bar graph is shown in Fig.10. The result shows over all success ratios as 66% which is good success ratio for successful forecasting earthquakes.

SOME RECORDS OF FORECASTED EARTHQUAKES

Earthquake of March 10, 2009 of Indonesia

The Fig. 3 shows the earthquake forecast for Indonesia in the month of March 2009. The first gravity variation (precursor no. 3 March) was recorded on 5 and 6 March 2009. It showed continued gravity variation in north- south and east-west component. The earthquake was forecasted and the forecast was it may strike in the region of Indonesia from 7-15 March 2009. This forecast was successful because an earthquake of magnitude 5.6 took place in Indonesia on 10 March 2009 at 23:42:14 PST.

A similar Pattern was observed soon after the first earthquake forecast and gravity variations (precursor no.4 March) were recorded on 7 & 8 March 2009. It also showed continued gravity variations in north- south and east-west component. We forecasted that Earthquake that it may strike in the region of Indonesia from 9-18 March 2009. This forecast was successful because an earthquake of magnitude 6.3 took place in Indonesia on 16 March 2009 at 19:15:55 PST. We can conclude that first earthquake of Indonesia with magnitude 5.6 may be the foreshock for second earthquake (Main shock) of Indonesia having magnitude 6.3.

Figure 3: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred in Indonesia

Earthquakes of April 6 & 7, 2009 of Central Italy

The Fig. 4 shows the earthquake forecast for Central Italy in the month of April 2009. The first gravity variation (precursor no.1 April) was recorded on 31 March & 1 April 2009. It showed continued gravity variation in north- south, east-west and as well as in vertical component. The forecast was that Earthquake may strike in the region of Italy from 2-12 April 2009. This Forecast was successful because an earthquake of magnitude 6.3 took place in Central Italy on April 6, 2009 at 21:32:39 PST.

A similar Pattern was observed soon after the first earthquake forecast and gravity variation (precursor no. 4 March) was recorded on 2 April 2009. It also showed continued gravity variation in north- south, east-west and as well as in vertical component. The forecast was that Earthquake may strike in the region of Italy from 3-15 April 2009. This Forecast was successful because an earthquake of magnitude 5.5 took place in Central Italy on 7 April 2009 at 13:47:37 PST.

Figure 4: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred Central Italy

Earthquake of June 5, 2009 of Japan

The Fig. 5 shows the earthquake forecast for Japan in the month of June 2009. The first gravity variation (precursor no.1 June) was recorded on 1 & 2 June 2009. It showed dominant gravity variation in east-west and in vertical component and minor gravity variation in north-south component. The forecast was that earthquake may strike in the region of Japan from 3-12 June 2009 based on strong gravity precursor. This forecast was successful because an earthquake of magnitude 6.4 took place in Japan on 5 June 2009 at 08:30:33 PST.

Figure 5: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred in Japan

Earthquake of June 6, 2009 of Japan

The Fig. 6 shows a similar Pattern was observed soon after the first earthquake forecast gravity variation precursor 2 (June) was recorded on 4 June 2009. It also showed continued gravity variation in north- south, east-west and as well as more dominant gravity variation in vertical component. The forecast was that earthquake may strike in the region of Japan from 5-13 June 2009. This forecast was successful because an earthquake of magnitude 5.8 took place in the region of Japan on 6 June 2009 at 10:52:43 PST. This earthquake of Japan with magnitude 5.8 was most probably the after shock of the major earthquake with magnitude 6.4 in Japan.

Figure 6: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred in Japan

Earthquake of July 13, 2009 of Taiwan

The Fig. 7 shows a comparatively different pattern from the rest of standardized pattern. This gravity variation precursor shows the earthquake forecast for Taiwan in the month of July 2009. The first gravity variation precursor 3 (July) was recorded on 8 July 2009. It showed dominant Gravity variation in north-south and east-west component with minor gravity variation in vertical component. The forecast was that earthquake may strike in the region of Taiwan from 9-16 July 2009 based on strong gravity precursor. This forecast was successful because an earthquake of magnitude 6.3 took place in the region of Taiwan on 13 July 2009 at 23:05:03 PST.

Figure 7: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred in Taiwan

Earthquake of July 13, 2009 of Taiwan

The Fig. 8 shows a similar pattern as described in the previous figure. This gravity variation precursor shows the earthquake forecast for Taiwan in the month of July 2009. The first gravity variation precursor 4 (July) was recorded on 10 July 2009. It showed dominant gravity variation in north-south and east-west component with minor gravity variation in vertical component. The forecast was that earthquake may strike in the region of Taiwan from 11-20 July 2009. This Forecast was successful because an earthquake of magnitude 5.4 took place in the region of Taiwan on 16 July 2009 at 15:48:15 PST. This earthquake of Taiwan with magnitude 5.4 was most probably the after shock of the major earthquake of Taiwan with magnitude 6.3.

Figure 8: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred in Taiwan

Earthquake of May 17, 2009 of Hind Kush (Afghanistan)

The Fig. 9 shows the earthquake forecast for Hindu Kush (Afghanistan) in the month of May 2009. The gravity variation precursor 3 (May) was recorded on 8th May 2009. It showed continued gravity variation in north- south and east-west component. The forecast was that earthquake may strike in the region of Hindu Kush from 9-18 May 2009. This forecast was successful because an earthquake of magnitude 4.6 took place in Hindu Kush on May 17, 2009 at 17:57:58 PST.

Figure 9: Gravity variations recorded at CES and details of
corresponding Earthquake that occurred in Hindu Kush


Figure 10: Graph showing Summary of six months earthquake forecasts.

Conclusions:

The result of six months data shows that there are definite variations in gravity data recorded at ATROPATENA system installed at CES, Islamabad, before the occurrence of major earthquake. The specific patterns of gravity variations were observed during analysis of data that helped in standardization of the patterns. Most of the earthquakes were successfully forecasted those occurred in different part of the globe on the basis of these standardized patterns. The forecasted earthquakes occurred in Indonesia, Japan, Taiwan, Italy and Hindu-Kush region of Afghanistan on the basis of their standardized pattern. The result indicates over all success ratios as more than 65% which is a good achievement for successful forecasting earthquakes. As the gravity variation data was analyzed on the basis of changes in gravitational field in three components of ATROPATENA system but there is still need to supplement other geophysical parameters that help in the accuracy of forecasting major earthquakes which requires doing a lot of work. However, at present there are three stations of ATROPATENA working and with experience by recognizing the pattern we may forecast the future major earthquakes with more accuracy, higher probability and reliability.

Acknowledgement:

The authors are thankful to Dr. Ishfaq Ahmad N.I. H.I. S.I, for his constant encouragement and patronization of the project.

REFERENCES

1. Ma Zong-jin. 1980. Relation between earthquakes and field of stress locus in crust in North China [J]. Seismology and Geology, 2(1 )
2. Khain V. Y. and Khalilov E. N. (2006). Tideless variations of gravity before strong distant earthquakes. Science without Borders Volume 2 2006/2006 ICSD/IAS H&E, Innsbruck, pp. 319-339
3. Khalilov E.N. (2007). About the possibility of creation of International Global System of forecasting of earthquakes “ATROPATENA” (Baku-Yogyakarta-Islamabad). Natural cataclysms and global problems of the modern civilization. Special edition of Transaction of the International Academy of Sciences. H&E ICSD/IAS, Innsbruck, pp51-69.
4. Walsh J B, and Rice J B, 1979. Local changes in gravity resulting from deformation. J Geophys Research, 16:445~44