Hi again Roger,

As the following will show, I knew that it was worthwhile to keep comparing notes with you regarding earthquake forecasting even though up until now you have not trusted my forecasting method.

As I said in an earlier note, my Perl data generation program is presently extremely versatile and powerful. And instead of waiting to see if you could generate that chart I mentioned in the previous note I tried it myself.

Holy Cow! Another major breakthrough.

I had the program go through each of the 27 windows on my Data.html Web page, each one 3 months in duration. And I had it create a 180E to 180W array like on my charts containing the earthquakes which occurred in each time window. For each earthquake that occurred at a specific longitude the program added 1 to that array longitude value. Then I plotted the results to create one of those charts.

The results were shocking!

The chart created from recent earthquakes looked quite similar to my synthetic charts created from that EM signal – earthquake matching system. As I said in the previous note, I thought that the recent strong chart peaks at 125E might be longitude offset errors. But they aren't! They appeared in the true earthquake chart just like in the synthetic one. So there must have been a large number of low magnitude earthquakes that occurred at that longitude recently and I did not notice that.

There were three differences between the chart:

First, the true earthquake chart does not have those +90, +180, +270 longitude degree earthquakes. And that is not surprising. In the synthetic chart they are a necessary artifact resulting from how my programs do the calculations.

Second, the onset date for the peak groups is different. With the November 15, 2006 Kuril Islands earthquake for example, peaks begin appearing on my synthetic chart in early August of 2006. But in the true earthquake chart they don’t start to appear until a few weeks or so before the actual earthquake. They must be from foreshock earthquakes.

Third, the peaks appear to continue further out in time after the earthquakes for the true earthquake chart versus the synthetic one. I would imagine that those are aftershock earthquakes appearing. And my forecasting method does not detect them unless they are especially powerful.

Now I am really anxious to do some more work on this and will probably get started on that in a few weeks. For example, I simply added one earthquake to each longitude value for each time window if there was an earthquake in the time window. What I might try next is factoring in the earthquake magnitude. Also, there is a programming procedure that I can use to have the true earthquake chart display those +90, +180, +270 earthquakes as well. You wouldn’t be able to do that yourself without knowing about a “weight value” assignment procedure for the earthquakes that my program uses.

But, before waiting several weeks I am going to see if I can generate one of those “true earthquake” charts and store it on my Data.html Web page right underneath the synthetic chart. And I will try to post a note here telling people when it is stored there. What a stunning development!

Thanks for the nice Easter present Roger.

And a happy Easter to everyone.