بِسْمِ اللَّهِ الرَّحْمَنِ الرَّحِيمِ

Earthquake Clouds – Awan Gempa

Posted by agorsiloku pada Juli 24, 2006

Zhonghao Shou
Published in Science and Utopya 64, page 53~57, October 1999 (in Turkish)

History Background
Ancient Chinese and Italians studied special clouds which were indicative of impending earthquakes. The Chronicle of Lon-De County (35.7 E, 106.1 N) in Ningxia province, China, 300 years ago (recompiled in1935) recorded, “It was sunny and warm; the sky was blue and clear. Suddenly, there appeared threads of a black cloud spanning the sky like a long snake. The cloud stayed for a long time, so there would be an earthquake ” [1].

Following the message, I found its corresponding earthquake, the 7.0 Guyuan (36.5 N, 106.3 E), Ningxia earthquake on October 25, 1622. It was the only big one in the western China (Empirical Evidence
In order for an earthquake prediction to be useful to government planners and the general population, it must have two important features. First, a prediction must be informative. It must include three windows – a time, a location, and a magnitude. Furthermore, these must all be specific. For example, announcing that “There will be a big earthquake” does not sufficiently motivate people to plan ahead. To demonstrate that earthquake clouds offer a means to make informative and specific earthquake predictions, I have analyzed my set of 39 predictions which have been officially certified by the U.S. Geological Survey.

Here I present an example – the 4.1 ML San Fernando Earthquake Prediction. On June 3, 1994, I predicted that from June 8 to 25 (an 18-day time window), there would be an earthquake of 4 ~ 5 ML in San Fernando, California (an area of 5,500 km2), 10~80o northwest of and 30~100 km away from Pasadena (34.138N, 118.143 W) . A 4.1 ML earthquake indeed occurred in this area (34.310 N, 118.398 W) on June 14. The following is a comparison:

Date Degrees (northwest) Distance(km) Size (ML)
The prediction 6/8~6/25 10~80 o 30~100 >4~5
The earthquake 6/14 50.6 o 30.3 4.1
The Southern California Earthquake Database shows that this earthquake was the only one greater than 3.9 ML in the 5,500 km2 predicted area from May 26, 1994 to June 25, 1995 (396 days).

The San Fernando prediction described above is informative, as it contains the necessary windows. Now, to quantitatively evaluate its specificity, I propose a statistics calculation. First, select those earthquakes whose epicenters are within the predicted area and whose sizes are within the predicted magnitude range from a certain earthquake database; second, divide the period covered by the database into time intervals of the same length as the predicted time span, and the sum of all intervals is counted as A, whereas the sum of intervals that do not contain any selected earthquake as B; finally, calculate the probability P =1-B/A, where P represents the probability of a random guess being true under the limits of the prediction.

Now, let’s use the method to evaluate the San Fernando Earthquake Prediction, which has an 18-day time window, a 4~5 ML size window, and a 5,500 km2 area window (10~ 80 degree northwest of and 30~100 km away from Pasadena). To calculate P in this case, we applied four computer filters (one to select for size 4~5, one for distance 30~100 km away from Pasadena, one for 10~ 80 degree northwest of Pasadena, and the last for the 18-day time window) to the Southern California Earthquake Database from April 1, 1981 to May 20, 1998, which spans a period of 6097 days (excluding February 1 to July 12, 1983, due to the lack of data); 5844 out of 6063 intervals are devoid of specified earthquakes, and Table 1 shows the distribution of the intervals. Therefore, P = 1 – B/A = 1- 5844/6063 = 3.6%.

Based on satellite images and direct visual observation of earthquake clouds, I have made 39 predictions to the US Geological Survey. The most specific prediction had a probability of P=2.5%, while most have probability less than P = 26%. Thus, random guessing would have yielded a success rate of less than 26%. To contrast, 25 of my 39 predictions were correct, yielding a success rate of more than 60%. The likelihood of lucky guesses consistently achieving this rate of success is extremely small. I believe this unequivocal evidence that there exists a correlation between clouds and earthquakes. I will propose a model describing how a cloud can be a direct precursor to an earthquake.

Model for Earthquake Clouds
I predicted the 6.1 Afghanistan earthquake on February 4, 1998 to both the USGS and the L. A. Times on January 5, 1998 with a P = 13% of being correct*.

When a huge rock is stressed by external forces, its weak parts first break and small earthquakes occur. Table 2 shows that all of the eleven big earthquakes in Southern California between January 1, 1975 and May 31, 1999 have many foreshocks within 10km. Since a strong earthquake produces a huge gap, it is likely that the foreshocks generate small cracks, which reduce the cohesion of the rock. Next, underground water fills in the cracks. Its expansion, contraction, and chemistry further reduces the cohesion.

The external forces cause(Refer to Fig. 15 and 16[4] ) between neighboring particles of rock to move against each other, and the resulting friction generates heat. The amount of heat can be surprisingly large. Scientific analysis of frictional melt and recrystallization of fault-rock indicates that temperatures from 300 – 1500o C can be generated along fault lines[5-8]. Anecdotal evidence of extreme heat is also plentiful. For example, very hot erupting matter was reported to have burned a man during the 7.8 Tangshan, China earthquake in 1976 [9]. Before the 7.3 Haicheng, China earthquake in1975, part of the ice in a shade of a frozen reservoir melted during a very cold winter [10]. Temperatures of 250~ 350o C were directly measured in steam and groundwater before three big earthquakes in Mexico [11].

High temperature makes groundwater become vapor. Haas showed that water at 300oC boils at a pressure of 86 atmospheres[12]. It is plausible that underground water at temperatures of 300~1500o C boils at great depths, where friction acts prior to an earthquake. In fact, the vapor has actually been observed before[11] and after[13] earthquakes. The tremendous pressure of the vapor forces it to the surface through pores and cracks in the ground. The effect of this superheated steam has been seen at the surface on numerous occasions. “Water spouts erupted from as high as 115 feet above the valley floor at an estimated 400 cubic feet per second” during the 7.3 Borah Peak, Idaho earthquake on October 28, 1983 [14], and “Petroleum erupted about 20 meters high” from a well eleven days before the Tangshan earthquake [15]. Furthermore, the pressure in certain oil wells sharply rose 20~ 50 atm. about a month before the Tangshan earthquake[16].

The high temperature of the vapor makes it difficult to observe as it escapes from the ground. However, Giang et al. reported, “Before medium and strong earthquakes, due to local force effect, a lot of gas emitted, which has already been evidenced by many monitored results.” [17]. The vapor is subsequently transported by surface winds. As it rises and meets colder air, it condenses to form a cloud. The entire process is similar to making an artificial silk. The fault, the vapor, the holes, the cold air, and the cloud are like the spinning pump, the viscous liquid, the jet, the spinning bath, and the artificial silk respectively. So, the most common form of an earthquake cloud is line-shaped. Due to a variety of factors, such as the distribution of vapor sources and surface winds, the shape may look like a line, a snake, a few parallel lines, a bind of parallel waves, a feather, a radiation or a lantern pattern. They are clearly distinct from weather clouds.

The image in Fig.1 contrasts the giant, white, mass-shaped weather cloud and the dwarf, linear earthquake cloud alongside. I asked a meteorologist from UCLA, whose field is special clouds, to interpret this subtle phenomenon, and he agreed that the dwarf is a cloud, but not a weather cloud.

What is it? It is an earthquake cloud. Meteorology theory could explain neither how the hollow formed in the giant weather cloud, nor how the dwarf formed in the hollow. Secondly, it was well-known that the darker the region, the hotter its temperature, so the hollow was much hotter than the giant. This phenomenon reflected that the ground temperature under the hollow was much hotter than that under the giant. Since both were over land in Asia, the difference of the temperatures implied that there was a big geothermal area that heated the air over it, and the warmed air convected and pushed a part of the weather cloud away, so the hollow formed. Counting on the Y-axis at 140 E of the original image, I calculated the location of the earthquake. This example demonstrates objectively that earthquake clouds exist.

These special clouds are useful in predicting earthquakes for three reasons. First, as the cloud’s tail points toward the position of the fault, it is possible to locate the epicenter. Second, the size of the cloud reflects the pressure around the fault, and can serve as an indicator of the magnitude of the impending earthquake. Finally, since an earthquake generally occurs within 49 days of the first appearance of the cloud, the time of the earthquake can be estimated. These clouds are therefore “Earthquake Clouds”.

Model for Earthquakes
The presence of groundwater is critical to the formation of earthquake clouds. I will propose a model for earthquakes which accounts for the importance of water. The USGS performed a very important experiment at the Rangely Oil Field in Western Colorado in 1969. They regularly injected water into the oil wells or pumped out of them and assisted with other important works. “The result showed an excellent correlation between the quantity of fluid injected and the local earthquake activity”. Bolt further proposed, “If there were no water in the rocks, there would be no tectonic earthquakes” [18].

Other researchers[19] have observed the severe weakening effect of groundwater. They contribute two excellent charts for yield strength of rock vs. temperature (Fig.10, [19]). They show that after dehydration, the yield strength drops sharply.

Dehydration is the most important phase for understanding earthquakes. During a dehydration period, carbon dioxide and other gases may erupt in some cases [15] because limestone and other rocks dissolve; earthquake sound may release due to underground gas disturbance [20]; hibernating snakes or lizards may commit suicide on frozen land because their holes are too hot to sleep in [20].

I propose that earthquake vapor escapes in the beginning of the dehydration, i.e. when the yield strength begins to drop sharply. Once the yield strength has dropped sufficiently, the rock yields and an earthquake occurs. Thus, the creation of an earthquake cloud by evaporating groundwater is directly linked to the generation of the earthquake itself. The whole process is similar to using the color change of phenolphthalein indicator from pink to no color to detect a rapid pH fall from 8 to 6 while putting HCl solution drops into NaOH solution.

Given the devastating impact of earthquakes, it is essential that predictions are based on a reliable precursor. I will attempt to justify the reliability of earthquake clouds as a precursor. Of my 39 predictions based on earthquake clouds, 14 were incorrect. Five of them were incorrect because their time windows were not big enough. For example, the 7.0 Mexico earthquake on January 11, 1997 * had a 6-day delay out of a 30 day-time window, but according to the National Earthquake Information Center, the earthquake was “The largest earthquake in this general area since a magnitude 7.0 event on April 30, 1986.” Compared with the 10-year period, the 6-day error is quite small. In fact, this kind of mistake can be prevented since by using a more conservative time window. According to more than 100 cases of my reliable records and predictions, earthquakes always follow within 49 days of the appearance of the cloud. The other 9 predictions were incorrect due to my inexperience and inability to determine the precise origin of an earthquake cloud. For example, the 6.2 Mexico earthquake on January 30, 1995 * had been predicted in Southern California just by my estimate, but the probability including the missed place is only 5.6 %.

Since a cloud is always moved and re-shaped by wind, it is critical to observe the initial position, track and size of an earthquake cloud, and to measure the wind velocity during the course of its motion. In principle, one should be able to predict a distant earthquake, if the cloud retains its shape after traveling a distance toward the observer, and the track can be estimated. Without adequate wind data, this is impossible. Thus, the mistakes have been due to deficiencies in my understanding of the details, and lack of resources, not due to any unreliability of the earthquake cloud precursor itself.

Earthquake clouds are a reliable precursor, but to be used properly, they require significant effort in the acquisition and analysis of cloud and wind data. For example, I found a big earthquake cloud near Sri Lanka (5N, 80E) from satellite images on July 16 [Fig. 2], and told three witnesses that there would be a very big earthquake in the northwest area from Sri Lanka.

On July 30, I made a prediction that there would be an earthquake of size more than 6.9 in the area 25~45 N and 15~65 E within 34 days, but I had no idea to reduce the location because I did not have surface current distributions to trace where the cloud came from.

On August 17, the 7.4 Turkey earthquake occurred. It is the only one larger than 6.9 in the west area from Sri Lanka since May 11, 1997. Its low probability suggests that the hometown of that cloud should be in Turkey. This example tells us that to detect the origin of the cloud, we have a tough job to do.

To prevent a large earthquake, I suggest Turkish people to make an hourly surface wind velocity distribution, and a greatly magnified hourly satellite image, or to detect the vapor directly. I believe that earthquake clouds are reliable for short term prediction, and hope that Turkish people would like this paper.

I thank the USGS, Two “jpg” web pages of uk, Caltech libraries, Dr. Moore, G. and three Caltech Ph.D. students Shou, W.Y., Harrington, D. and Wang, A. S.

References and Notes
[1] Li, D.J. Earthquake Clouds, 148-150 (Xue Lin Public Store, Shanghai, China, 1982).

[2] Dunbar, P.K., Lockridge, P.A. & Whiteside, L.S. World Data Center A for Solid Earth Geophysics, 146 (National Geophysical Data Center, Colorado, 1992).

[3] Zhou, H.L. Moment magnitudes of historical earthquakes in China. Earthquake Research in China 1, No. 3, 347-360 (1987).

[4] Haicheng Earthquake Study Delegation. Prediction of the Haicheng earthquake. Eos 58, 236-272 (1977).

[5] Spray, J.G. A physical basis for the frictional melting of some rock-forming minerals. Tectonophysics 204, 205-221 (1992).

[6] Swanson, M.T. Fault structure, wear mechanisms and rupture processes in pseudotachylyte generation. Tectonophysics 204, 223-242 (1992).

[7] Koch, N. & Masch, L. Formation of Alpine mylonites and pseudotachylytes at the base of the Silvretta nappe, Eastern Alps. Tectonophysics 204, 289-306 (1992).

[8] Techmer, K.S., Ahrendt, H. & Weber, K. The development of pseudotachylyte in the Ivrea-Verbano zone of the Italian Alps. Tectonophysics 204, 307-322 (1992).

[9] Shi, H. X., Cai, Z.H. & Gao, M.X. Anomalous migration of shallow groundwater and gases in the Beijing region and the 1976 Tangshan earthquake. Acta Seismologica Sinica 2, No.1, 55-64 (1980).

[10] Yang, C.S. Temporal and spatial distribution of anomalous ground water changes before the 1975 Haicheng earthquake. Acta Seismologica Sinica 4, No.1, 84-89 (1982).

[11] Glowacka, E. & Nava, F. A. Major earthquakes in Mexicali Valley, Mexico, and fluid extraction at Cerro Prieto geothermal field. Bulletin of the Seismological Society of America 86, No.1A , 93-105 (1996).

[12] Haas, J.L.Jr. The effect of salinity on the maximum thermal gradient of a hydrothermal system at hydrostatic pressure. Eco. Geol. 66, 940-946 (1971).

[13] Chandrasekharam, D. Ateam emanation due to seismic pumping, Killari, Maharashtra. Geol. Surv. Ind. Spl. Pub. No.27, 229-233 (1995).

[14] Lane, T. & Waag, C. Ground-water eruptions in the Chilly Buttes area, Central Idaho. Special Publications 91, 19 (1985).

[15] Shi, H.X. & Cai, Z.H. Case examples of peculiar phenomena of subsurface fluid behavior observed in China preceding earthquakes. Acta Seismologica Sinica 2, No.4, 425-429 (1980).

[16] Zhang, D.Y. & Zhao, G.M. Anomalous variations in oil wells distributed in the Bohai bay oil field before and after the Tangshan earthquake of 1976. Acta Seismologica Sinica 5, No.3, 360-369 (1983).

[17] Giang, Z. J. et cl. An experimental study of temperature increasing mechanism of satellitic thermo-infrared. Acta Seismologica Sinica 19, No. 2, 197-201 (1997).

[18] Bolt, B.A. Stimulation of earthquakes by water. Earthquakes, 135-139 (W.H. Freeman and Company, New York, 1988).

[19] Kirby, S.H & McCormick, J.W. Inelastic properties of rocks and minerals: strength and rheology. Practical Handbook of Physical Properties of Rocks and Minerals, 179-185 (ed. Carmichael, R.S., CRC Press, Boca Raton, Florida, 1990).

[20] Tang, X. Anomalous meteorology. A General History of Earthquake Study in China, 49-84 (Science Press, Beijing, 1988, in English).

Table 1 The interval distribution of the 4.1 San Fernando Earthquake Prediction.

Table 2 Small cracks had occurred around and before all big earthquakes in Southern California between 1/1/1975 and 9/14/1999.

* All predictions are listed on Web Site: (Before) and (Now)


Tinggalkan Balasan

Isikan data di bawah atau klik salah satu ikon untuk log in:


You are commenting using your account. Logout /  Ubah )

Foto Google+

You are commenting using your Google+ account. Logout /  Ubah )

Gambar Twitter

You are commenting using your Twitter account. Logout /  Ubah )

Foto Facebook

You are commenting using your Facebook account. Logout /  Ubah )


Connecting to %s

%d blogger menyukai ini: