hi
Basic guide line/thumb rules for major earthquake prediction
There are two aspects of predicting dates of major quakes,as per my findings,
1) potent month/weeks
2) Trigger dates
1) potent months have indicators like
a)Jupiter closer/est----about 4.0 AU or less
b)Venus closer/sest----about 0.3 to 0.5 AU
c)Sun closer/est-------0.98 AU or so
For the time being ,let us keep aside ,direct .retro gate motions
First find out such month where all three/ any two exists
2) date and time are decided by trigger Moon
a)Moon at max declinations (both)
b) Moon closer---57.0 ER or so
c) Full Moon/New Moon
d) Moon Joining /opposing planets/Sun mentioned at sr no 1) above
So instead of taking all above as indicators take sr 1) as first requirement and sr2) as trigger to fix date and time
It would be an interesting idea to find dates when these indicators and triggers are present and no 6.5+ quake
Amit
Basic guide line/thumb rules for major earthquake prediction
There are two aspects of predicting dates of major quakes,as per my findings,
1) potent month/weeks
2) Trigger dates
1) potent months have indicators like
a)Jupiter closer/est----about 4.0 AU or less
b)Venus closer/sest----about 0.3 to 0.5 AU
c)Sun closer/est-------0.98 AU or so
For the time being ,let us keep aside ,direct .retro gate motions
First find out such month where all three/ any two exists
2) date and time are decided by trigger Moon
a)Moon at max declinations (both)
b) Moon closer---57.0 ER or so
c) Full Moon/New Moon
d) Moon Joining /opposing planets/Sun mentioned at sr no 1) above
So instead of taking all above as indicators take sr 1) as first requirement and sr2) as trigger to fix date and time
It would be an interesting idea to find dates when these indicators and triggers are present and no 6.5+ quake
Amit
62 comments:
Hi
For more accurate results ,also take in to account maximum declinations of Jupiter ,Venus and Sun in sr no 1) above
AMIT
Hi Amit;
That's a testable hypothesis so I tested it.
I created an array containing Right Ascension, declination and distance for sun, moon, venus and jupiter every day from 2003 to 2012.
Then I found days containing the smallest values for each of the declinations and distances..
Then I added the largest quake for each day, 6.5+
Then I looked for a day containing 2 or more minimums, calling it a prediction.
Finally I checked for prediction days containing a quake, calling that a hit.
There were 14 predictions and only one hit out of 415 quakes.
Now I'm sure you're going to want some leeway in the size of a prediction window so let me know how much and I'll rerun it.
I'll also add the maximums you wanted.
Roger
Adding those brought you up to 22 predictions and 2 hits.
Roger
Hi again Amit;
Disregard my last posts, I'm still debugging and polishing.
It's looking better for you. I'm trying a 5 day window, unless you have other ideas?
Roger
Roger
No.I do not want any change in my window period.
Yess.I am conservative. in giving magnitudes.
Hence , when I say 6.5 ( + or - 0.3) ,it often go beyond 6.8.
comments on what you have stated above
1) No need to take Right Ascension
2) Distance minimum but. Declinations should be maximum ie + or - 22 etc
3)Can it be done in two parts ,Roger
Part A.. Minimun distance and Maximum Declination dates of Sun,Venus,Jupiter to be fond out first
PartB.... In these selected dates check for Moon minimum distance ,maximum declination, FullMoon and Moon opposing ,joiñing closer/Maximum declination planets
PartA is first requirement then only teigger Moon will act
Amit
Amit;
Ok, back to a 3 day window.
I mark minimum distances and max and min declination for sun, moon, venus and jupiter.
Then I start a window if any one of those is marked and search the next 2 days for others marked. If 2 or more things are marked I accept that as a prediction.
Then I check to see if there are any quakes in the window and if so, count the prediction as a hit.
Finally I compute the odds on a 6.5 quake in a 3 day window.
The results are below chance.
Is this close enough to what you are doing or should I restrict a prediction to windows involving the moon?
Roger
Amit;
Ok, I defined a prediction window as a 3 day period with 2 or more indicators, one of which had to be the moon distance or declination.
This reduced the count to 54 predictions in 10 years with only 15 hits (27.8%)
Since the odds of a 6.5+ quake in a 3 day window are 28.7% you still lose.
Roger
Roger
Thanks.
Few things
1) For Sun , Venus and Jupiter , Minimum distance and Maximum declinations only. to be considered
2) Three day window with 2 or more indictors one of the indicator should be
Moon closer/closest....57 to 58 ER
Moon at maximum declination both North and South ( 22 degree to 28 degree...whichever the case.
or. Full Moon, New Moon ,Moon in Opposition
or joning the closer/est Planets ie Venus or Jupiter.
Roger
One more way is to find days when Jupiter ,venus and Sun are closer and and one of them have Maximm declinatio
And
Moon as trigger ie closer, Max declination ,Full Moon , New Moon etc and check whether this gives 7+ quakes
More the indicator combine more the magnitude
Amit
Amit;
You're welcome.
Now you're clutching at straws.
I already know moon phase has nothing to do with quakes and aspects to other planets?
Be serious.
Roger
Roger
pl do not check Moon phases in isolation.
First requirement is Part A , then only trigger will work.
Otherwise , we should have quakes on all Full Moon and New Moon.
However , yes more potent trigger is Moon Closest and at Maximum declination
Amit
Amit;
I've tested that above. It doesn't work.
Roger
Amit;
I sent you the output file from my 3 day window program.
Please let me know if/when you get it.
Roger
Roger
I may not be able to study it for few more days, as my wife is again operated on wednesday for rectosigmoid obstruction.
Whatever I am posting is from my mobile phone and from hospital. I am stationed at hospital since last Tuesday.
I will revert back as soon as I get time after discharge fro hospital.
Amit
Roger
please note 6.5 NZ quake on 21st.
Note date ,magnitude and hours accuracy
May be I am not properly explaining the merhod to you.Hence your results just equals to
averages.
Further averages are calculatrd by including cluster of quakes
Pin pointed prediction is the proof.Several occatios ( mentioned on this blog) date accuracy of 6+ quakes achieved
Amit
Amit;
Sorry about your wife's situation. My wife just had a double mastectomy so I know what you're going thru.
Latest test involving maximum declinations and minimum distances for mag 7+ quakes produced 7 predictions and no hits.
Roger
Hi Amit;
Good news.
Your criteria for mag 7+ quakes works (almost).
Sun, Venus or Jupiter close or with max. declination with moon new or full or max. declination gives 32 predictions and 5 hits while the odds on a hit are 7.4%
That's better than chance but not enough better to pass my test.
Still it's the best you've done so far.
Roger
Roger
Thanks. I am sure , if I could explain you better , it should be much more than average
Now it is more than double the chance
Amit
Roger
Once you said , the prediction test results should be equal to calculated averages, where as
these results are more than double the averages
Amit
Amit;
Don't get too excited, I'm still working on the program.
The latest revision was to require that something other than the moon was involved and that killed most of the predictions. Apparently the correspondence between new moon and perigee was creating the windows.
You have to be very specific as to what conditions are required and that can be hard to set up.
Roger
Roger
I am neigher exicted nor disappointed. I have crossed that phase. Down the 18 years, I am laughed at, rediclued, threatened and even labelled as" An Astrologer Engineer"
After part A is done iepotent dates of Sun,Jupiter and Venus closer/closest ,main task is to fix trigger for date
Moon at maximum declination and closer/closest is. most potent trigger
Full Moon and New Moon are next potent trigger
Moon joining or opposing plannets (Jupiter ,venus) to one which is closer or at maximum declination is ranked as third
Amit
Roger
It would be interesting to check
1) Maximum indictors from part A
and
2) Maximum triggers from part B
and still no 7+ quakes
Amit
hi
Hereby I give yet another hypothesis for fixing the. location of quakes..Part C
Part A..Fixing potent dates of major quake
Part B..Fixing date and time as per Moon trigger explained earlier
Part C..fixing location
i) After the date and time fixed as per part B
place on Earth. is the intersection of Earth surface and line joining center of Earth and Moon as well as opposite side of the Earth ie 180 degrees apart
or
Intersection of Earth surface and Planet (potent as per part A) and 180 degree apart
Alreenatevly , to check major quakes (7.5+) occured in last few years with this location hypothesis, since we have date,time and locations
Amit
Amit;
Since you can't predict the date there's no point in worrying about the location.
I've made a condensed version of the days listing only the indicators for each day; declination and distance for each of the 4 bodies, plus the 6.5+ quakes.
This makes it possible to see everything without side scrolling.
Your task is to find the rules for predicting the 3 day window containing a quake.
I don't see any. There are some matches but not enough to matter.
Roger
Roger
Rules are already mentioned above.
Amit
Heres a couple of programs i gave before
Add time/date of earthquake in question to the below program
http://douglasallen.github.io/planets/
= the exact point that the planet would strike the earth if a line was drawn from the center of that plant/moon sun etc to the center of the earth.
Then use the GHA (Greenwich hour angle ) and DEC (Declination) of object requiring about and put it
into the great circle calculator along with the coordinates of the Celestial body being inquired about.
http://williams.best.vwh.net/gccalc.htm
(Geocentric to geodetic angle conversion)
Answers in feet....
putting this to future predictions would be life draining without the computer power needed.
Also look for kelpers Harmonics of the planets and the 3 laws
Jazzman :)
Amit;
Following those rules, results are not significantly better than chance.
Furthermore, there is no modification to the rules which will improve the results.
You're way up a blind alley.
Roger
Jazzman
Thanks. I will check it,seems useful
Amit
Roger
I am unable to comprehend.
Once ,you said ,it appears better than chance( more than twice)
Rules are clear
Part A)
Select dates /weeks ,where any two indicators are available
a)Sun closer(0.98 AU approx) ,Sun at Maximum Declination
b)Jupiter closer (4.0 AU or so), Jupiter at Maximum declination
c) Venus closer (0.4 AU or so),Venus at maximum declination
Out of these three a),b) and c) of part A indicators any one is required.
or
out of six indicators{ two of each a),b) and c)} of part A) any four are required for 6.5+ quake to occur
Part B)
once probable dates from part A are fixed exact date is decided by trigger Moon.
without part A,Moon may not be able to trigger the quake, hence we do not have quakes on all full Moon and new Moons
Potential triggers, in descending order are
a) Moon closer/closest (57 ER), and Moon at Maximum declination (+ or - 22 or whichever is highest)
b) Full Moon OR New Moon
c)moon joining or opposing ,planets closer (in part A)
Amit
Amit;
Please clarify; by "maximum" do you mean largest, smallest or both?
Roger
Roger
Maximum declination means.....
Maximum declination both north and south
say for Moon +22 and -22. both
if it is +27 and -27 in perticular year that should be taken and not +22/-22
Amit
Amit;
Ok, then you mean maximum absolute value.
Not a problem as I compute the values, then look for peaks and troughs
Roger
Roger
window period of + pr- 1 day is decided by the trigger Moon. The selected day (middle of window) is the day when Moon enters trigger(PartB) ie Moon closest,at max declination ,fullMoon,New New Moon etc
Amit
Amit;
Now with the planets indicators starting a possible window and the moon triggering it within 3 days, causing the window to include 1 day either side of the moon trigger day
we find 18 windows and 8 hits for 44.4% hit ratio.
The odds are only 28.7% for a 3 day window so this is significant at the 88.6% level.
This is suggestive but not sufficient to pass the 99% success level I require.
So see if you can adjust the rules to reach that level.
Roger
Amit;
You'd need 11 out of 18 to pass my test.
Roger
Roger
8 out of 18 ie 44.4% is well above average(28.7%)
What exactly you mean by 99% ?
Is it 99% of average ie 28.7 ?
11 out of 18 is 61.1 % which is more than 200 % of average
pl clarify
As far as I know, it was said" no one is better than average odds". and
" Any body can guess the same hits I am getting"
Now atleast , I am well above anybodys guess prediction
Amit
Amit;
No, not averages. Statistical significance levels.
You take the number of predictions, the number of hits and the chances of success and calculate the binomial probability of getting that many hits or more by chance.
The 99% level I require tell us that there is only 1 chance in 100 of doing better by chance.
Roger
Roger
Ok. Got it.
Though it is very stringent ,it is fine
As main trigger is fast moving Moon, we can also check ,how narrowing the window works.It would also be interesting to check following
First fix the trigger Moon ,in terms of hour ie exactly the hour Moon enters the trigger position. say for example For full Moon trigger ,the hour in which Moon exactly opposes the Sun, and then,
1) take window as + or - 24 hours ,to the trigger hour ( ie total 2 day window period)
2) take window as + or -12 hours ,to the trigger hour ( ie only one day window)
I am not sure , what results ,such window period will give.
Roger, I request you to check ,how it works.
Idea is ,the trigger is fast moving, it should work in narrow window period
Amit
Amit
You filed the easy test so you want to see how it does with a herder test??
Your logic escapes me.
In any case, I can't do it. I'm working with tables of daily planetary positions. An hourly table would be 24 times as large.
Sorry.
Roger
Roger
I have other version of it.
I have improved a lot. It is. double the average
I have not heard of any one reaching this.You are the correct person to throw light on this
Any way.My prediction trigger is based on Moon we can atleast check the one day window ie the day on which Moon enters trigger position
Amit
Amit;
I can test nearly anything you can think up, within reason.
As for better, Petra Challus got a significance level in the mid-nineties so she's the best so far.
Roger
Amit;
Good news.
The odds on a 6.5+ quake on a given day are 11%
Selecting a 1 day window based on the moon trigger date gave 26 predictions with 6 hits.
That's a significance level of 94.1% so you're doing good.
However, that's a small sample so I'll need to extend the testing time range.
It's also rather useless, getting only 6 hits in 10 years with no idea of location.
Roger
Amit;
You would need 8 hits to pass the test.
Roger
Roger
Thanks.
I wonder why there are only 26 predictions.
This is because , once ,Say Jupiter is closer/closest (3.8 to 4.0 AU), there could be two or three Moon triggers with in say 15/20 days.
Amit
Amit;
it's because I only check for moon trigger within 3 days.
I'll try 20 days.
Roger
Roger
Once Jupiter is closer to closest,it will be so for some days say 15/20. By that much day Moon may have one full Moon and /or New Moon,closest to earth (57 ER or so),and /or Maximum declination (either north or south) .Thus ,there will be at least two trigger ,may be three.
As Venus is faster,it will hardly give two trigger of Moon or may not give
Thus,there should be more windows, as per my observations
Amit
Amit;
In the 20 day search, do you want to make a prediction for each moon trigger or only the first one?
Roger
Roger
In 20 days, 10 days on either side of the most potent position (closest) of Jupiter.( similarly Venus ,Sun)
yes, Each Moon trigger in these 20 days will give quakes ( ie why we have clusters of quakes in some weeks)
ie Moon coming close in these 20 days
Moon reaching Maximum declination ( either North or South)in these 20 days
Full Moon or New Moon in these 20 days
Amit
Amit;
That makes no sense. you're pulling the trigger before the gun is loaded.
I looked for a moon trigger(s) within the 20 days starting with a planetary indicator, finding 248 predictions but only 33 quakes.
That's an 85.6% significance level.
Roger
Roger
It is not single shot country gun. Rather an automatic pre loaded AK 56.
As long as pre loaded cartige are available( potent position as per PartA)
the trigger(Part B) will fire.
After that even if you pull the trigger, it will not fire.
Amit
Amit;
That's my point. Until the potent position (planet) happens, pulling the (moon) trigger will do nothing.
And the loaded gun will not fire until the (moon) trigger is pulled.
Roger
Roger
Potent position (PartA) does not come and go suddenly.It slowly increases ,reach the peak and slowly decreases.
In the process,it gives quakes.It is not one day peak only
Amit
Amit;
Ok, the planet is not a one day thing but the moon is.
Now if you're saying that there is a critical period 20 days long centered on a planetary location, triggered by the moon indicator falling within that period then we have an entirely different situation.
I'm not sure how to figure the odds except as a 20 day window containing a quake and I would guess that's very high. If a one day window is 11% a 20 day window must be 100%, making your prediction worthless.
Roger
Roger
There can nt be and should not be 20 day window period.If there are two seperate Moon triggers in these 20 days,there will be two windows ,each of one day
Amit
Amit;
You're right. The possible window is 20 days but the predicted window is one day.
Using those rules we have 258 predictions but only 27 hits.
That's 10.5% hit ratio while chance is 11%
Next?
Roger
Roger
it appears Something wrong in number of windows
As you said in last 10 years ,there are 26 occasions,when Jupiter, Venus OR Sun were closest. ie 10 for Jupiter ( every one year and one month) and 10 for Sun ( every year). where as Venus was 6 times closest
Thus in all 26.
Now ,each planet can have maximum Moon triggers either 3 or 4 ( some times 2) in 20 days.
ie closest, full Moon/New Moon and maxi declination
Let us take each of the window of 20 days ,have 4 Moon triggers
Thus it comes out to 26*4=84 say maximum 90 odd
How come there are 258 predictions?
Amit
sorry
104 , and not 84
so ,100 odd is the correct window figure
Amit
Amit;
You are correct for distances, total 26 but for declination we have 20 sun 268 moon, 32 venus and 18 jupiter.
Then there are 248 moon phases to consider.
258 predictions is correct but only 27 hits
Roger
Roger
Part A consists of only distances ie 26
Part B ie Moon trigger includes all ie Moon closest,Moon at max declination(+ and-) and full moon new moon
Thus 90-100. 0dd windows will be there
Amit
Roger
Part B all to be taken inside those 20 days only
Thus each one of 26 may have 3 or 4 moon triggers
Amit
Amit;
What happened to declination?
Roger
Amit;
No good.
83 predictions, 7 hits, 71.6% significance level.
I have one that reaches 94.1% level but only makes 26 predictions and 6 hits, too few for reliable testing.
Roger
Post a Comment