By studying changes in the Earth’s ionosphere, an Israeli research team discovered a method to predict earthquakes 48 hours ahead with 80% accuracy.

Israeli researchers have discovered a method to predict earthquakes 48 hours ahead with 80% accuracy, as detailed in a peer-reviewed study published in the scientific journal Remote Sensing in May.

How did they work it out?
By studying changes in the Earth’s ionosphere, the sliver of atmosphere which meets the vacuum of space, the Ariel University and Center for Research & Development Eastern Branch research team was able to evaluate potential precursors to several major earthquakes that occurred in the past 20 years.

www.jpost.com/science/article-712972

¿ so those rumours of primitive animals detecting earthquakes in advance may be true ?

SCIENCE said:

¿ so those rumours of primitive animals detecting earthquakes in advance may be true ?

We do know that it doesn’t work for Italian scientists.

https://www.researchgate.net/publication/361256521_Using_Support_Vector_Machine_SVM_with_GPS_Ionospheric_TEC_Estimations_to_Potentially_Predict_Earthquake_Events

That’s the weirdest idea I’ve seen in a proper geophys journal for many years.

dv said:

https://www.researchgate.net/publication/361256521_Using_Support_Vector_Machine_SVM_with_GPS_Ionospheric_TEC_Estimations_to_Potentially_Predict_Earthquake_Events

That’s the weirdest idea I’ve seen in a proper geophys journal for many years.

Looking up article. Yeah, thought so. The basic idea is piezoelectricity, rocks produce an electric charge when squeezed.

The absolute best that you could get out of this, if anything, is to find regions with high geologic stress. Shearing fracture along fault zones is not in itself directly related to high stress because it depends also on the thickness of rock that has to be sheared through in order to produce the earthquake. And magnitude is not strongly related because it depends on the length of the fracture. For example, rock stress along the San Andreas / Howard fault system is increasing almost linearly but that doesn’t tell us when the earthquake will be.

What the article is is a correlation between high ionospheric activity in times of solar quietness (after subtracting the solar component) … there is no check against any other measurements of stresses in rocks, such as measurements of bulging, or cessation of micro-tremors.

No, look, what we have here is a correlation generated by machine learning on random datasets.

The following is a figure from the paper. They are claiming that the ionospheric strength (four lines) predicts the earthquake time. I don’t see any significant correlation.

mollwollfumble said:

dv said:

https://www.researchgate.net/publication/361256521_Using_Support_Vector_Machine_SVM_with_GPS_Ionospheric_TEC_Estimations_to_Potentially_Predict_Earthquake_Events

That’s the weirdest idea I’ve seen in a proper geophys journal for many years.

Looking up article. Yeah, thought so. The basic idea is piezoelectricity, rocks produce an electric charge when squeezed.

The absolute best that you could get out of this, if anything, is to find regions with high geologic stress. Shearing fracture along fault zones is not in itself directly related to high stress because it depends also on the thickness of rock that has to be sheared through in order to produce the earthquake. And magnitude is not strongly related because it depends on the length of the fracture. For example, rock stress along the San Andreas / Howard fault system is increasing almost linearly but that doesn’t tell us when the earthquake will be.

What the article is is a correlation between high ionospheric activity in times of solar quietness (after subtracting the solar component) … there is no check against any other measurements of stresses in rocks, such as measurements of bulging, or cessation of micro-tremors.

No, look, what we have here is a correlation generated by machine learning on random datasets.

The following is a figure from the paper. They are claiming that the ionospheric strength (four lines) predicts the earthquake time. I don’t see any significant correlation.

Are you questioning the excellence of their hindsight here?

But I await their 80% accuracy in predicting future earthquakes with interest.

we’re 100% sure there will be a future earthquake

SCIENCE said:

we’re 100% sure there will be a future earthquake

You don’t say.

The Rev Dodgson said:

mollwollfumble said:

dv said:

https://www.researchgate.net/publication/361256521_Using_Support_Vector_Machine_SVM_with_GPS_Ionospheric_TEC_Estimations_to_Potentially_Predict_Earthquake_Events

That’s the weirdest idea I’ve seen in a proper geophys journal for many years.

Looking up article. Yeah, thought so. The basic idea is piezoelectricity, rocks produce an electric charge when squeezed.

The absolute best that you could get out of this, if anything, is to find regions with high geologic stress. Shearing fracture along fault zones is not in itself directly related to high stress because it depends also on the thickness of rock that has to be sheared through in order to produce the earthquake. And magnitude is not strongly related because it depends on the length of the fracture. For example, rock stress along the San Andreas / Howard fault system is increasing almost linearly but that doesn’t tell us when the earthquake will be.

What the article is is a correlation between high ionospheric activity in times of solar quietness (after subtracting the solar component) … there is no check against any other measurements of stresses in rocks, such as measurements of bulging, or cessation of micro-tremors.

No, look, what we have here is a correlation generated by machine learning on random datasets.

The following is a figure from the paper. They are claiming that the ionospheric strength (four lines) predicts the earthquake time. I don’t see any significant correlation.

Are you questioning the excellence of their hindsight here?

But I await their 80% accuracy in predicting future earthquakes with interest.

How would it be used I wonder.
Would anyone actually listen if data points towards an forthcoming earthquake.
I imagine it would have margins of error no matter what even if that’s just days
Evacuation isn’t a minor undertaking and if nothing happens then they are discredit even if the theory holds merit and perhaps they didn’t have all the data required.

The Rev Dodgson said:

mollwollfumble said:

dv said:

https://www.researchgate.net/publication/361256521_Using_Support_Vector_Machine_SVM_with_GPS_Ionospheric_TEC_Estimations_to_Potentially_Predict_Earthquake_Events

That’s the weirdest idea I’ve seen in a proper geophys journal for many years.

Looking up article. Yeah, thought so. The basic idea is piezoelectricity, rocks produce an electric charge when squeezed.

The absolute best that you could get out of this, if anything, is to find regions with high geologic stress. Shearing fracture along fault zones is not in itself directly related to high stress because it depends also on the thickness of rock that has to be sheared through in order to produce the earthquake. And magnitude is not strongly related because it depends on the length of the fracture. For example, rock stress along the San Andreas / Howard fault system is increasing almost linearly but that doesn’t tell us when the earthquake will be.

What the article is is a correlation between high ionospheric activity in times of solar quietness (after subtracting the solar component) … there is no check against any other measurements of stresses in rocks, such as measurements of bulging, or cessation of micro-tremors.

No, look, what we have here is a correlation generated by machine learning on random datasets.

The following is a figure from the paper. They are claiming that the ionospheric strength (four lines) predicts the earthquake time. I don’t see any significant correlation.

Are you questioning the excellence of their hindsight here?

But I await their 80% accuracy in predicting future earthquakes with interest.

Aye. I hope they tested this as a forecaster before publishing.

Cymek said:

The Rev Dodgson said:

mollwollfumble said:

Looking up article. Yeah, thought so. The basic idea is piezoelectricity, rocks produce an electric charge when squeezed.

The absolute best that you could get out of this, if anything, is to find regions with high geologic stress. Shearing fracture along fault zones is not in itself directly related to high stress because it depends also on the thickness of rock that has to be sheared through in order to produce the earthquake. And magnitude is not strongly related because it depends on the length of the fracture. For example, rock stress along the San Andreas / Howard fault system is increasing almost linearly but that doesn’t tell us when the earthquake will be.

What the article is is a correlation between high ionospheric activity in times of solar quietness (after subtracting the solar component) … there is no check against any other measurements of stresses in rocks, such as measurements of bulging, or cessation of micro-tremors.

No, look, what we have here is a correlation generated by machine learning on random datasets.

The following is a figure from the paper. They are claiming that the ionospheric strength (four lines) predicts the earthquake time. I don’t see any significant correlation.

Are you questioning the excellence of their hindsight here?

But I await their 80% accuracy in predicting future earthquakes with interest.

How would it be used I wonder.
Would anyone actually listen if data points towards an forthcoming earthquake.
I imagine it would have margins of error no matter what even if that’s just days
Evacuation isn’t a minor undertaking and if nothing happens then they are discredit even if the theory holds merit and perhaps they didn’t have all the data required.

Actually, their statistical method is pretty good. I was expecting more false positives.

As a prediction method for earthquakes – no.

But let’s look at it another way. From the purely scientific perspective. Are ionospheric fluctuations due to stresses in rocks of the Earth’s crust actually measurable? It’s an interesting question. The way to check it would be to compare stress measurements of rocks on the Earth’s surface with ionospheric fluctuations directly overhead.

mollwollfumble said:

Cymek said:

The Rev Dodgson said:

Are you questioning the excellence of their hindsight here?

But I await their 80% accuracy in predicting future earthquakes with interest.

How would it be used I wonder.
Would anyone actually listen if data points towards an forthcoming earthquake.
I imagine it would have margins of error no matter what even if that’s just days
Evacuation isn’t a minor undertaking and if nothing happens then they are discredit even if the theory holds merit and perhaps they didn’t have all the data required.

Actually, their statistical method is pretty good. I was expecting more false positives.

As a prediction method for earthquakes – no.

But let’s look at it another way. From the purely scientific perspective. Are ionospheric fluctuations due to stresses in rocks of the Earth’s crust actually measurable? It’s an interesting question. The way to check it would be to compare stress measurements of rocks on the Earth’s surface with ionospheric fluctuations directly overhead.

Or could ionospheric fluctuations cause earthquakes?

Or could they both be triggered by a third event?