The real EQ prediction is possible and must be based on three suppositions:

1) The behaviour of the focal area of the future EQ is non-linear. Therefore, the so called "stress waves" originate in the focal area (Li, J.Z., Z. Q. Bai, W. S. Chen, Y. Q. Xia, Y. R. Liu, & Z. Q. Ren (2003): Strong earthquakes can be predicted: a multidisciplinary method for strong earthquake prediction. Natural Hazards and Earth System Sciences (2003) 3: 703–712.)

2) These "stress waves" are detectable everywhere on the litosphere plate (due to long-periods, which result to their small attenuation). The apparatus for EQ prediction must measure directly or indirectly the stress field and not the seismic waves. The variations of the stress field precede the mainshock by days up to months (http://meetingorganizer.copernicus.org/EGU2013/EGU2013-7592.pdf).

3) The recognition and understanding of the triggering mechanisms can help to the localisation of the focal area.

I am really thankful for this answer but the question still remains if there is a fault patch and a nucleation takes place will the entire nucleation zone be affected or part of it may not be affected asthe Malle Earthquake of Chile.How can pre seismic locking be explained.It s something like this if we consider earthquake to be like a heart attack how can stress waves be identified.

Secondly if there is a stage where there is a uniform stress applied to a region across every direction how will residual strain behave....

The locking of the fault can be explained by the asperity (the hardest part of the fault plain or more rough part than the other). The asperity becomes the stress concentrator. If the stress reaches the strength limit of the asperity, some parts of focal zone start to deform in non-linear way. At that time, the stress waves are created and they spread to the surrounding and can travel around the whole lithosphere plate. They can be detectable everywhere in the case of the biggest asperities, which can accumulate a lot of energy proportional to M >7 (Stuart Crampin and Yuan Gao - Earthquakes can be predicted).

Secondly - it depends on their values.

Sir I beleive asperity is the endogenetic factor of rocks and heterogeneity as the exogenetic stabling factor that characterizes the nature of earthquake behavior of rock material.How can we establish the time of fracture of rock material...

If we detect the stress waves, we can recognise that some part of the rock mass is approaching to its destruction. The stress waves are the consequence of non-linear behaviour of the material close to the strength limit. The wavelength of the stress wave (and the amplitude) is proportional to the nucleation period and to the magnitude of the future mainshock. For example, if the stress wave has the length of one week, we can suppose that the mainshock will be of magnitude 7 during next 30 days. With the help of stress-based methods we can estimate both parameters - magnitude and time of the mainshock according this empirical function: log t (days) = 0.46 M - 2.2 (Kalenda et al. 2010, 2012). On the other hand, stress-based methods have problem with the localisation of the focal area. We must use other methods.

Sir I am yet to understand how will stress waves be detected can we define any wireless monitoring unit or a remote monitoring system for detection.How will the threshold upper and lower limit for stress waves be defined .

Secondly we have found already that if the nucleation mechanism is considered as a response mechanism with precursory behavior as a feedback mechanism earthquake occurence is an evident output function.

Utkin et al,2013 Journal of Geodetic Science. Volume 3, Issue 3, Pages 203–208, ISSN (Online) 2081-9943, ISSN (Print) 2081-9919, DOI: 10.2478/jogs-2013-0024, October 2013

we have already defined that if radon is a precursory function then it is found there is either an exponential increase or a exponential decrease in radon count before earthquake ,In Dutta et al,2012 we have defined that earthquake occurence follows a mathematical function and stress strain is related to pore pressure where a bayesian function can explain Volumetric analysis of radon.But we cannot give estimate of the exact state whereby the rupture may take place in microcrack evolution mechanism

I would like to know is there any specific parameter or set of parameters which can be regularly monitored by which an earthquake warning based wireless remote montioring unit can be defined and developed .......

The problem of the radon measurement and prediction of the earthquake occurrence is in the fact that radon reacts on any stress changes (and movement of the fault). Then, we can not estimate the absolute level of the stress field in the focal area. The EQ prediction must be based on another parameter - stress waves, which demonstrate themselves the closeness of the stress field to the strength limit.

I put the PPT presentation from IUGG meeting in Melbourne 2011: "The prediction of earthquakes can be solved by stress measurement"

https://www.researchgate.net/publication/259148832_The_prediction_of_earthquakes_can_be_solved_by_stress_measurement?ev=prf_pub

On the page 12 and 13 there are examples of "stress waves", which were detected before sthe strongest world-wide earthquakes in 2009 and 2010.

The non-linear asperity model is on the page 14: The stress waves are generated between points b and c of stress-strain diagram, when the material shows non-linear behaviour.

Data The prediction of earthquakes can be solved by stress measurement

What are the limitations of your mechanism in analyzing the shear strain of rock which can be analyzed in the future work

The limitation of the detection of "stress waves" is due to tensor of the stress. No all apparatus can detect such stress waves everywhere on the globe. It depends on the tectonic settings of the specific place, where the apparatus is installed. For example, the apparatus in Central Europe, which would be installed close to NW-SE active fault, will be sensitive to any movement from Pakistan, Indonesia, Tonga and Chile, because their movement is in this direction.

So can wireless sensor network come into use for detecting a certain parameter of stress waves,I saw in one of your ppts that stress waves has a very high wavelength the bandwidth to support such a system may be infeasible

But if we have a mechanism to detect source of a seismic event associated with a weaker geological feature or with a softer patch rockmass yields more slowly under lower stress, and radiates less seismic energy per unit of inelastic coseismic deformation, than an equivalent source within strong and

highly stressed rock. Therefore by comparing radiated seismic energies, or stress

drops, of seismic events with similar moments one can gain insight into the stresses acting within the part of the rockmass affected by these events

Our main aim is to identify Location of Potential Rockbursts: To alert management by indicating the locations of potential rockbursts associated with intermediate or large seismic events and to assist in possible rescue operations - it is important then to

monitor the locations of associated aftershocks using wireless sensor networks.

Is stress wave a stationary wave

I have followed this discussion on and off. Interesting points. However, the whole issue as presented in the past comments strongly leans towards a conventional seismological approach. The conventional seismological approach, however, has shown to be miserably inadequate to "predict" earthquakes except within 5-20 years of statistical uncertainties.

Seismology has its roots in the latter part of the 19th century, in the late 1800's, when mechanical physics ruled supreme. For well over 100 years seismologists have stuck to mechanical physics as the best and (in their minds) only way to study earthquakes. This is unfortunate. Seismologist have spent an inordinate amount of time, intellectual energy (and funding) on the pursuit of a fundamentally unreachable goal: how to "predict" future earthquakes on the basis of knowledge derived from studying the mechanics of past earthquakes. Even if seismologists build up huge libraries showing them how thousands of past earthquakes have happened, any attempt to apply this knowledge to the "prediction" of a future seismic event is subject to the law of statistics. Statistics of a single future event are inevitably subject to great uncertainties. That's why seismologists have not be able in the past and will never be able in the future to "predict" major earthquake within any probability window narrower than years to decades, sometimes even centuries.

The reason for this sorry state of affairs is that the mechanical approach to forecasting earthquakes is fundamentally flawed.

(I prefer the word "forecast" over "prediction")

My own work over the past 10-15 years has shown, I think conclusively, that we can derive far better and more meaningful information about impending earthquake activity from the study of non-seismic processes that accompany the build-up of high levels of stress in the Earth's crust. This, however, requires some familiarity with solid state physics (in particular its non-mechanical aspects) with electrochemistry and semiconductor physics.

Friedemann Freund

Dear Friedmann, I fully agree with your attempt to this problem (EQ prediction). We must measure the parameters, which can tell us how far (how close) to the failure limit the rock mass is. You measure the parameters of ElMag field, which is activated (changed) right before the mainshock. I measure the deformation of the rock mass in the distant quiet area to detect so called "stress waves", which are generated in the non-linear part of the deformation process. Both methods can indirectly measure the stress field and not the seismicity, which occures at the end of the deformation process.

To Pushan Dutta: The stress waves are not stationary waves, but long-period waves, which are spreading to the surrounding with the velocity of seismic P-waves. Their periods are days, weeks up to months (in the case of the largest asperities, which produce the EQs of magnitude 8.5 and higher).

Dear Pavel, what you write is VERY intriguing. I cheerfully admit, it's new to me.

You mention long-period stress waves, which spread at the velocity of P waves, but you also state unambiguously that their periods are VERY: long, days, weeks up to months.

I'm wondering whether there is a contradiction between the speed of propagation and the long periods? Is the high speed a theoretical postulate (which I expect to be the case) that these stress waves propagate as elastic waves and not as inelastic waves, which travel by plastically deforming their surrounding?

If they propagate as elastic waves, I have question #1: How do stress waves decay as a function of distance? is it between 1/r(square) and 1/r(cube) or is it faster than 1/r(cube)?

A related question #2: How far do your long-period stress waves typically travel in the case of an asperity that leads to a magnitude 8.5 or 9 event?

Thanks for this stimulating input

Friedemann

I would like to add a point that there is a difference in the measurement to the accurate limit why earthquake predetection is very difficult,We always use a correlated length of scale not actual scale of rupture.In any non linear seismic process we are estimating a pre detection threshold based on stiffness change for radon count for instant.or finding quasi static rupture mechanism by measuring zero crossing rates ,if we measure the change in variance between two points can we measure earthquake occurence.

Or in other words how is the differential stress threshold determined or fixed for a region

If we are talking about stress waves, only Pavel will be able to provide an answer

Dear Friedmann, your question is essential - the stress waves are deformation types of waves, which means that their velocity inside blocks is almost the velocity of P-waves, but at the edges of the blocks they must wait for the closing of the free spaces and therefore the average velocities are smaller than those of elastic P-waves.

We could not measure their amplitude decay because their wavelenghts were longer than the circumference of the Earth and they were visible on the whole lithosphere plate. It depends only on the local tectonic settings of the apparatus, which stress waves would be detectable much more easily.

Download the presentation https://www.researchgate.net/publication/259390335_Pribram2013_predikce?ev=prf_pub

and on slides 12 and 13 there are some examples of stress waves, which were detected before Okhotsk Sea EQ (M7.7) and Indonesian and Chile EQ (M8.8).

You question #2: The stress waves are detectable everywhere on the Earth before M7.5+ events, but it depends mostly on tectonic settings. We can "see" stress waves from Indonesia and Kurile Islands in Central Europe, but Japan area and Central America is a problem for us.

The period of stress waves is proportional to the length of asperity and therefore to the magnitude of future EQ (see slide 17).

Data Pribram2013 predikce

Pavel Kalenda and Friedman Freund: how do you understand that a certain stress wave implies a certain earthquake will occur in a certain region without measuring the conditions locally...

Secondly you have said that stress waves are long period wave if so then is it affected by the medium through which propogates or else how can you consider the wave to be a stress wave.

thirdly if it has a time duration of days how can stress wave for a certain condition be called as a stress wave due to the breakage of say x rupture and not y rupture.

Do they have any specific signature which will differentiate them from the rest say P PKP or so and so forth

Dear Pushan, the localisation of the asperity, is a bit a problem from the point of view of stress waves. We use at that time 2 methods, how to recognise the focal area:

1) use the independent method, which can localise the focal area, for example IR satelite measurement (see slides 14 and 15) or

2) to recognise the triggering mechanism. For example the tides have two maxima per day, where the Coulomb criterion would be fulfilled, which imply the local meridian. Another example is triggering by LOD, which give us another two local meridians. The best examples are on slides 11 and 12, where are peaks of deformation noise around midnight of local Central Europe time. If we suppose that the triggering mechsanism is solar irradiance, which has its maximum around noon, we could recognise that the local meridian is opposite to Central Europe, i.e. of Kurile - Kamchatka region. Both anomalies and stress waves were detected in Moravian Karst on N-S active fault, which has its direction to Kurily - Kamchatka region. Therefore, we successfuly predicted both EQs in 2008 and in 2012, as we refered at EGU meeting:

https://www.researchgate.net/publication/258959868_The_Multi-parameter_observation_of_pre-earthquake_signals_and_theirpotential_for_short_term_earthquake_forecasting?ev=prf_pub

Data The Multi-parameter observation of pre-earthquake signals an...

Dear Pavel, thanks for sending this ppt presentation. I'm still trying to wrap my mind around these ideas. In principle they look very reasonable and logical (though daring). In practice, the question arises how elastic stresses can be transmitted across large distances taking into account that the brittle (elastic) portion of the Earth crust is as thin as the skin of an apple, if viewed in the context of the size of the Earth. Most likely the transmission of an elastic response across a large continental plate like Eurasia might be better than the transmission across wide stretches of oceanic plate with even the possibility of a mid-ocean spreading ridge, where the thickness of the elastic crust goes to near-zero.

Fascinating. Surely more stations around the globe would be necessary to improve the resolution in time and space. Wonderful prospect.

Friedemann

I think if a triggering mechanism exist a function earthquake warning system based on location error analysis can be designed for the near future .If a location sensitive sensor network can be set up we can say when earthquake will take place.

Please read my idea:

I, the developer of a new measuring tools - magneto elastic power meter based on which developed original geophones to measure and record the wave fields in solids. Development carried out in order to study and forecast of earthquakes:

http://earthsciences.typepad.com/blog/2012/05/listening-to-seismicity.html 

However, during the research it became apparent that the unique characteristics of the geophones allow them to be used not only in seismic exploration seismology, but also for many other solutions and scientific applications.

Firstly, they are perfectly suited to the frequency and dynamic range, reliability and long eternity to the role of acoustic detectors detect neutrinos (especially in the "ice" antennas).

Second, they can be an alternative when underwater hydrophones trials, having unlimited depth of immersion.

Currently, I am constantly live in San Francisco, California. I have tested samples of geophones (one and three-component) and digital recording system - visualization for their presentation.