Can earthquakes be predicted?

Total Pageviews

Search This Blog

Monday, May 31, 2010

sample study-planets changing direction and quakes

hi
A sample study for the year 2006 was done by me ,to check concentartion of quakes near dtaes when planets changes the direction. For study purpose Mercury was not considere as it is a small planet and have less tidal pull
In year 2006 there are 5 dates when planets changes the direction ie
3rd Feb, 5th March, 5th April, 6th July,and 7th December.
The quakes 6 -6.9, 7.0-8.0 and 8+ were checked before and after 15 days of the date on which planet changes the direction
Results are
------ ---- ---Actual----- Average (expected during the period ie 5 months)
6-6.9 -- ---------66 -----------56
7.0-8.0 ----------6 ------------6
8+ ---------------1 -----------0.5(?)
A study was also done for Jupiter when changing direction as well as closer to Earth
7th April 2007 and 7th August 2007 Jupiter changes the direction as well as closer to earth (<5.0 AU)
For 15 days before and after these dates the quakes are as follows
-- For 7th April 07 -----For Date 7h August 07
---------Actual -----------Actual --------Average (for a Month)
6-6.9 -----31 ---------------20 -------------11
7-7.9 ------1-----------------2--------------1.25
8+---------1 ----------------1 -------------0.1 (?)
I would like you to comment as to weather this is above average or not?
regards
Amit

11 comments:

AMIT said...

hi
when Jupiter and saturn both cahnges the direction say saturn on 27th January 2001 and Jupiter on 26th January 2001,as well as when both closer,what happens
Jupiter--4.5 AU from Earth
Saturn----8.7 AU
Venus---- 0.62 AU
Earth to Sun --0.985 AU
Results are
6-6.9------25
7-7.9------7
8+---------0
please see quakes of 7-7.9 are several times the average (average is 1.2 per month)
Amit

Roger Hunter said...

Amit;

It means nothing.

Quakes happen in clusters due to aftershocks. This confuses the statistics if averages are used.

The Jones method I use minimizes the effect of clusters by counting hits in a yes-or-no manner, giving odds for a "one or more" situation.

Roger

AMIT said...

ROGER
I have some different opinion.
1) All quakes are not aftershocks
2)They occur at different places all over the globe and hence they need to be counted
3) Even aftershocks(major) have defined pattern
4) Those termed as clusters and which are not counted actually shows that the date is potent,as far as tidal pull is concerned
regards
Amit

Roger Hunter said...

Amit;

1) True

2) Why? You aren't predicting location anyway.

3) True but not relevant.

4) Maybe but you have to show it happens consistently and you can't do it with a small sample. You're just cherry picking.

Roger

AMIT said...

Roger
As per USGS the average quakes of 6-6.9 is 134 ,and that for 7-7.9 is 15.
I wonder how these averages are calculated.Wheather aftershocks and clusters are considered while ariving at these average figures?
Any idea ?
Regards
Amit

Roger Hunter said...

Amit;

I don't know for certain but I suspect it's a simple count from the catalog.

The question is "How many mag 6 quakes happen on average?" which is quite different from our question which is "What are the chances that a 30 day window will contain a mag 6+ quake?"

Roger

AMIT said...

Roger
I would like to know more about Jones method.Could you suggest some link for that?
regards
Amit

AMIT said...

Roger
NEIC records says from 1.1.1980 to
1.1.2010 the quakes are
6-6.9------3749 ie av per year 125
7-7.9------375 ie av per year 13
8+---------24 ie av per year say 1
It imply that all quakes (clusters and aftershocks) are included in total.How than we can compare both,one inclusive of clusters and aftershocks and another without it?.
What could be these average figures excluding both?
regards
Amit

Roger Hunter said...

Amit;

I don't know of a link but the method is very simple.

Suppose you want probabilities for 2009 using a 3 day window for mag 6+ quakes.

Divide the time into consecutive 3 day windows. There are 122 of them.

Get a list of all 6+ quakes for 2008.

Place each quake in the window corresponding to it's date.

Count how many windows have one or more quakes in them. Say there were 91 of them.

The odds are then 91/122 = 0.75

The situation gets much more complicated if you have a group of predictions with different probabilities. Dr Jones had to invent a new system for that.

Roger

Divide the

AMIT said...

Roger
1)you mean previous ears list of 6+ quakes is to be taken?
2)If we take NEIC list,it will give more value of odds as all quakes are considered in list. Why then multiple quakes and clusters are not counted for calculating accuracy of prediction?
If we include we should include both time,if we exclude we should exclude both time
regards
Amit

Roger Hunter said...

Amit;

The catalog includes all quakes so that one can do his own selections.

Some users may want to see all the aftershocks. Others may not.

My program excludes them. Not just aftershocks, but any multiple hits in a window because I want the probability of a hit in a time window.

If it's a prediction for a certain size or certain area, only those quakes are searched.

Roger