Here are a few comments. I haven’t said anything about this particular type of forecasting work in a while.

Just take my word for the following for the moment since this is just comment material.

My data and data from other researchers clearly show that powerful earthquakes are often being triggered by forces related to the gravitational pulls of the sun and the moon on the crust of the Earth. And that means that the triggering times are related to the positions of the sun and the moon in the sky.

Another researcher’s data on this were very crude. But he processed so many earthquakes that clear trends could be seen.

My own earthquake triggering model is quite a bit more advanced. And my results are a lot clearer. My own forecasting data which are based on that triggering model show that.

From what I understand, Amit’s forecast model is generally the same. However, he includes planet location data in addition to sun and moon position data. And I would make a guess, and it is only a guess, that those additional planet data degrade his results.

The point of this is that there are specific higher probability earthquake triggering (and forecasting) time windows related to the positions of at least the sun and the moon in the sky. Those time windows can be easily identified. And charts showing the results could be displayed on my Roger Web pages or my general forecasting research Web pages.

The very basic calculations that show some good results are simple. And once they are done, additional layers of complexity can be added to take into account the type of fault zone, its north and south or east and west orientation, the distance between the sun and Earth and the moon and Earth as opposed to just the subsolar and sublunar locations, and on and on.

With the increasing levels of complexity the time windows get more accurate. They probably really need to be optimized for each individual fault zone. And it would be worthwhile for governments to do that. I wouldn’t attempt that myself as it could require months or even years of work.

A high probability time window does not mean that an earthquake will occur during the time window. It simply means that if an earthquake is getting ready to occur anyway, it is more likely to occur during that time window than at some other time.

People can say that this or that researcher investigated this and got no good results. That doesn’t mean anything. This process is like opening a combination lock. If you have the right combination in the first place the lock opens with a few turns of the dial. If you don’t you can spin the dial for years and the lock won’t open.

People who failed to find links between earthquake occurrence times and the positions of the sun and the moon in the sky in the past were likely just using the wrong combination (equations). They used ones linking the occurrence times and the sun and moon positions that were predestined to fail. So no matter how many calculations they did they got no useful results.

One researcher sent me some very interesting data not too long ago showing links between earthquakes that have occurred in recent years and fault zone depth. The data looked rather startling. Unfortunately, most of the people doing this type of research are not storing their data on Web pages where other people can see them. And I usually don’t have time to process all of them and store them on my own Web pages.

So, getting back to the original subject matter,

I expect that Amit Dave’s forecast accuracy is strongly affected by the nature of the equations he is using. I intend to add those types of high probability time window data to my own forecasting computer program. But that will take a while. First the actual correlations need to be established. My present forecast programs don’t make use of time window data. My older WaveChart data did. Roger and I developed a procedure for displaying Amit’s and Shan’s forecast data on my charts so they could be easily compared with my data, with sun and moon position data etc., and with earthquake occurrence times. But that involved a FreeBasic program. And that program is not presently running.