Do Earthquakes cause Propagation dropouts? Can the RF-Seismograph see earthqakes?

I looked through the graphs of the 2016 events. A first glance there does seem to be some relationship of the (inserted) quake spike with the 80m noise. However, it seems to me that just as often there is an increase in 80m noise without any quake. Have you performed any statistical analysis to see if the results are any better than mere chance?

 

We have over a list over 180 M6+ events for the 4 year time period and they all fit the same pattern as we have detected with the M5 discussed at this forum. I believe that has statistical significance!
What makes this hard is that we only have looked at M6+ events and there is so much more going on.
For your assistance we have created a little snooper document that explains and discuses further details!
Check this:
http://www3.telus.net/public/bc237/MDSR/IntroductionRF-SeismographandEarthqakes.pdf

All the best;

Alex

 

Let's not get too picky. I think the intention was to say the Sun is not the only element in considering propagation. The real silliness is to be found in the kind of blind, unqualified assertion seen earlier, viz: "No" as to any possibility of seismicity being a factor. I know it's Trump America, but...

 

I certainly strongly support examining this potential link. But I tend to agree with VE7DXW that, there is currently an absence of a robust statistical analysis, which is much more than simply presenting a graph of HF noise in relation to a quake, the timing of which doesn't seem evident to me on the majority of the graphs. Quakes are not single-moment events. There are stress build-up periods of various duration, stages of energy release, and then aftershocks (not that I pretend to be a quake expert, but you get the idea.)

I suppose it is is straightforward in seismically-active areas for hams to monitor the bands regularly as part of their normal activities, with no special effort. The West Coast US, New Zealand, Japan, Latin America all have a reasonable number of operators. One thing we learned in Wales in relation to the persistence of radiation from Chernobyl was that farmers held a lot of local knowledge, such as how upland plants grow, but that scientists were apt to dismiss this 'yocal stuff' as irrelevant. I note that one person has already noticed noise increases ahead of a past quake, so it is probably an idea to try some form of quizzing of those who have operated over lengthy periods in seismic zones.

 

I take that to mean there has been no statistical analysis. Statistical significance is a mathematical term with a precise meaning. To determine it involves a lot more than just graphing some data and eyeballing it.

 

When you have list of the times the earthquakes released and you look at the effect on propagation and in 80% of the cases it creates the same pattern... I call that statistically significant!

Alex

 

There

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

There is an interesting article in Scientific American that explains it really well.

Earthquakes in the sky (page 44). I is in the October 2018 edition - a purplish white cover stating the "The Unsolvable Problem"


This explores the interesting measurements that were done in Japan on how earthquakes can create electric field lines that extend into the atmosphere. Could they be used to detect earthquakes before they cause damage on the planet?

You can get the issue in the Library or on line.

Alex

 

I really do appreciate your enthusiasm. But you also need to look at the all the times when there was no earthquake and yet the propagation was affected (by something else, obviously).

The hypothesis of a relationship between the ionosphere and earthquakes must be tested with with appropriate mathematical statistical tools on the entire dataset, not just near the times when an earthquake occurred.

 

We did that too. A small portion of the earthquakes do not have a RF-signatures. For instance the earthquake that occurred in Alaska on Nov 30 was a M5.2 and it did not have a measurable effect on the RF-Seismograph. And since we are only looking for quakes bigger than 5.9 it did not make our list.
In the 4 years that we conduct this study we experienced a total of 180 6+ earthquakes. And we looked at all of them.
So there is lot of data and this has nothing to do with eyeballs!
The RF-seismograph takes a 6 measurements every 52s, one per band. This amounts to 1660 individual measurements per day each, for 6 bands. Each day RF-Seismograph logs a 110k data file containing the individual data points. As of it the measurement has been running with an up-time of 99.999% since we started to record data in Aug 2016. As of today we have collected 107MB of data of band activity and noise level. We are definitely not eyeballing this!

Maybe you should set up an RF-Seismograph yourself. You might actually record an earthquake too!

Alex

 

Alex:
-You have failed to define the hypothetical propagation event [commencement and ending]. What are the specific parameters? Without a hard definition, you ARE just 'eyeballing" it.

-You have failed to statistically test all recorded observations for events matching the definition. You have no statistical validity. Without this test, you are just cherry picking propagation events.

-You have not statistically correlated defined propagation events in time to earthquake events.

-Bottom line: You have an interesting curiosity, but nothing more, at this time. There is no proof of propagation events occurring, and nothing more than visual interpretation associating them with earthquakes. shrug.

p.s. You have not specified what characteristic you are using to identify an earthquake, e.g. S wave, P wave etc. nor have you specified what parameter(s) identifies the commencement (and ending) of the earthquake event e.g. amplitude, displacement distance, duration, intensity, moment, etc.

[Heavily edited]

 

How can you cherry pick observation events when the time is defined by the time the quake released.

We are trying to learn something from this and obviously you did not even look at the data, otherwise you had a bit of a more respect of the work we went though to get to this conclusion. I did not wake up one day and said let make up a story about earthquakes and propagation. It took us 10 years to get this stage! We released this info so that others can have a go at it and maybe for you to learn something too!

We are still in the early stages of this research and if you want to go ahead and help us define each earthquake on our list so that we can make more specific conclusion please do so! We will release a list of all observations soon and I welcome you to further analyze the data and make more precise definitions.

Thank you;
Alex - VE7DXW

 

Two events are occurring.
1. Earthquake events (most can agree these exist and generally when they occur).
2. Propagation events (though you have failed to define them or prove that they exist as statistically valid events).

Your hypothesis is that these are correlated.
1. You have not proven that earthquakes or the hypothetical propagation events are correlated, and to what degree.
2. You have not statistically tested for (false) or negative correlation of hypothetical propagation events.

Your observations are definitely interesting. I hope that you continue observations and explore your hypothesis!!!

-Time to get mathematicians/statisticians involved in fully analyzing your data.

 

https://earthdata.nasa.gov/user-resources/sensing-our-planet/warnings-from-the-ionosphere

 

One more tool in the box. Before you start to evacuate a whole district you want multiple sources to tell that it actually is going to shake!
Alex

 

Have you also looked at all similar prop events in your data and then matched against non-earthquake events?
How do you know these prop events don’t occur many more times without earthquakes than with?