PROPOSED EARTHQUAKE FORECASTING
COMPUTER PROGRAM DEVELOPMENT EFFORT
Posted July 11, 2005
My main earthquake forecasting Web page is:
http://www.freewebz.com/eq-forecasting/Data.html
Newsgroup Readers: If you circulate copies of this report to groups of
computer programmers at different universities etc. around the world then
they might find the subject matter to be interesting.
The information in this report represents expressions of personal opinion.
THE GOAL OF THIS REPORT
This is part of an effort to get some idea regarding how many
computer programmers and other researchers around the world might be
interested in participating in a project aimed at developing life saving
earthquake forecasting computer programs.
That effort is not presently underway. And I don't know when or if
it will get started. I am simply attempting to determine if other people
believe that large numbers of volunteers would be interested in working on
such a project or if there would instead be little interest in it. That
information would be helpful for developing a plan for establishing a Web
site where the project would be centered. Personnel running the following
Web site have volunteered to make their site available for such an effort.
But nothing has gotten underway so far.
http://www.ictwhoiswho.net/comprend/index.cfm
If quite a few people were interested in such a Web site based
computer program development effort then after it got started work would
undoubtedly progress quite rapidly. If only a few were interested then it
might never get started. If you would like to express an opinion on the
likelihood of people being interested in the idea then you can try posting a
note in response to this one. The sci.geo.earthquakes or comp.lang.misc
newsgroups might be appropriate if you wish to post to just one newsgroup.
Of you can try contacting me by e-mail.
THE CORE OF THE PRESENT EARTHQUAKE FORECASTING PROGRAM
In connection with an earthquake forecasting effort which has been
underway for the past 15 years I believe I have been able to crack the
"Earthquake Code." That means making crucially important discoveries
regarding how earthquakes are being triggered. Two of them which are
discussed on the following Web page are called the "Gravity Point" and
"Earthquake Triggering Symmetry."
http://www.freewebz.com/eq-forecasting/90-05.html
The earthquake triggering and forecasting theories and data on that
Web page were discussed on my behalf by one of my research colleagues in the
People's Republic of China at a disaster mitigation conference in that
country in December of 2003. Governments and disaster mitigation groups
around the world were told about the Web site earlier this year. And my Web
site visitor counter indicates that some 100 to 200 people around the world
are presently downloading information from the site each day.
My earthquake forecasting computer programs use those Gravity Point
and Earthquake Triggering Symmetry discoveries and others to compare
electromagnetic energy field fluctuation type signals (EM signals) with more
than 30,000 earthquakes which occurred since the beginning of 1990. Some
100 to 200 signals detected during a 3 month period of time are involved.
For some as yet unknown reason they are often highly selective for
earthquakes which are likely to occur near populated areas, making them
unusually valuable. The earthquake which is the best match with all of
those signals is rated # 1. The worst match would have a rating number
greater than 30,000. A listing of more than 100 of the best matches is then
posted perhaps once a week to the following Web page:
http://www.freewebz.com/eq-forecasting/Data.html
For a recent example of how well that approach to forecasting
earthquakes can work, data displayed on that Web page on June 27, 2005 gave
the following earthquake a # 3 rating (possible rating range: 1 to 30,000+):
2005/01/11 19:19:48 11.40N 86.51W 40.7 5.0 Near the Coast of Nicaragua
(U.S. National Earthquake Information Service data)
And less than a week later on July 2, 2005 the following powerful and
strongly felt earthquake occurred:
2005/07/02 02:16:46 11.18N 86.40W 45.5 6.7 Near the Coast of Nicaragua
11.40N and 86.51W versus
11.18N and 86.40W
Pretty good accuracy for a forecasting program!
That earthquake was reportedly strongly felt in Managua. Had it
occurred directly beneath the city and near the surface then I expect that
it would have been devastating.
WHAT NEEDS TO BE DONE
Basically, more sophisticated data processing and data display
computer program subroutines need to be developed. They could be built on
my already existing computer programs and data. People would develop new
subroutines, give them a try, and see if they did a better job of
determining or displaying where an earthquake might be about to occur. The
subroutines could be stored at the proposed Web site. Ones that were
especially helpful could be merged into the main program one version of
which might run as a CGI program at the Web site. Other versions of the
programs would be downloadable for free for use on personal computers.
The basic form of the data processing routine for my existing program
and probably many other earthquake forecasting programs might be expressed
in the following manner:
Prob = aA + bB + cC + dD + eE + ...
"Prob" is the probability that an earthquake of a given magnitude will occur
at a specific latitude, longitude, depth, and time.
A, B, C, D, and E . are things such as:
A - the gradual buildup of strain in a fault zone due to the movement of
the Earth's tectonic plates relative to one another
B - temporary strain added to the fault zone by bending, stretching, and
compression forces related to the Solid Earth Tide. (The ground shifts a
little in response to the sun and moon gravities just as ocean water does -
hence the Solid Earth Tide)
C - temporary strain added to the fault zone by the weight of ocean water
shifting from one location to another in connection with ocean tides
a, b, c, d, and e .. are "weight" factors which specify how important A, B,
and C etc. are at different points in time.
"A," the gradual buildup of strain in a fault zone related to
tectonic plate movement etc. is undoubtedly the most important factor and
perhaps the only one which scientists around the world are in agreement on.
It can probably be determined with a certain amount of accuracy for some
fault zones at the present time, but not too many. With my present computer
programs I do not use actual values for "A" as they would be impossible to
determine. Instead when a strong EM signal is detected I simply assume that
a fault zone somewhere has stored enough strain energy that it is about
ready to fracture. And an effort is made to determine where it is located.
With each probability calculation my present computer programs do
about 30 separate comparisons between each of the 100 to 200 EM signals and
the more than 30,000 earthquakes in my database file. The comparisons
involve things such as the positions of the sun and the moon in the sky and
the locations of ocean and Solid Earth Tide crests and troughs around the
world when the EM signal was detected and when the earthquake occurred.
For an example of one possible and relatively easy computer program
improvement, an effort could be made to see if factoring in earthquake fault
zone orientation - north and south versus east and west - improved the
probability calculations. Another improvement would involve determining the
importance of the latitudes of the Gravity Points and the sublunar points.
At the present time only their longitudes are used in my calculations. (The
sublunar point is the location on the surface of the Earth which a line
drawn between the center of the Earth and the center of the moon would pass
through.)
An earthquake forecasting group at Madras University in India has
already developed some advanced earthquake location determination routines
which it appears might be helpful to this effort. Routines developed for
use at the following Web sites might also be helpful.
http://pasadena.wr.usgs.gov/step/
http://www-aig.jpl.nasa.gov/public/dus/quakesim/
THE PROGRAMMING LANGUAGE IN PRESENT USE
The original programs were written in a number of languages including
Basic. The main program is presently written in Perl. That language was
chosen because it is fast and powerful, the compiler can be downloaded for
free by anyone, and because it looks like it is getting sufficient support
that calculations can be trusted and it will be around for a while. There
is another feature of that language which I am guessing many other
programming languages probably presently offer though that was not the case
in the past. That is the ability to make changes to the program code itself
at the same time that the program is running.
I myself do that in the following manner:
Ordinarily the main program "P1" starts running and performs a group
of calculations. That takes about 5 minutes. It then waits for a keyboard
instruction telling it how the output data should be displayed. As with
probably any program, once it is running in the normal mode no changes can
be made to the program itself. However when it is run in the following
experimental mode that rule does not apply.
A short program I will call "P2" starts running and immediately uses
a "do" statement to get the regular program P1 to compile and start running.
P1 does the original calculations like normal. But instead of waiting for
the operator to enter a data display command it exits and P2 becomes active
again. A display instruction is entered and P1 is told to recompile and
start running again. Instead of doing all of the calculations from scratch
it jumps strait to the display routine and uses the entered command to begin
printing the output data. The data generated when it originally did the
calculations are still in memory.
The advantage here is that at any time, program P1 can be called into
a text editor, modified as desired, and then saved. A new display
subroutine can be added to it or an existing routine can be modified while
the previously generated data are still active in the computer. Then when
P1 is directed by P2 to recompile (that takes about a second) and run, the
new subroutine is included just as if it were in the original program. If
the compiler encounters a programming error it terminates and returns
control to program P2. A correction can then be made to the new program
code and the sequence repeated. No data are lost because of the error. And
you don't have to wait for 5 minutes while the program recalculates
everything.
Data processing and display routines could be written with other
programming languages besides Perl. If done with sufficient care more than
one language could be used at the same time. The different language
routines would simply be linked with one another.
A LIFE SAVING EFFORT
It appears to me (sadly) that few governments have a very good sense
of direction with regard to the development of life saving earthquake
forecasting programs. The work is usually undertaken by independent
research groups at various government agencies or universities. They go
their own ways, uninterested in or unable to work with one another to
effectively forecast deadly earthquakes. They often claim that if they
share information then they could lose their patent rights and potential
profits etc. And that could actually happen. But isn't saving tens of
thousands of lives more important than simply making a meager profit (which
no one that I am aware of is presently able to do with their forecasting
programs anyway)?
I myself presently own 3 U.S. copyrights related to my forecasting
technology. But I have been letting interested parties around the world use
it for free. See:
http://www.freewebz.com/eq-forecasting/301.html
One country where forecasting program information is shared to some
extent is the People's Republic of China were some 10,000 people reportedly
work full-time in a state sponsored earthquake forecasting program. They
are also supported by a small army of volunteer workers. But even there the
forecasting efforts of different groups are not effectively coordinated very
often from what I can see. And advanced forecasting technology being
developed in other countries is frequently ignored.
A listing of some other earthquake forecasting programs around the
world can be found on the following Web page:
http://www.freewebz.com/eq-forecasting/141.html
A book discussing one of the forecasting programs listed on that Web
page is scheduled for release some time in late 2005. See:
http://www.sentientpublications.com/catalog/earthquakes.php
I myself assisted with that effort by providing the author with some
free technical information and book content advice.
The proposed Internet Web site effort to develop an effective
earthquake forecasting program discussed in this present report could not be
ignored by governments around the world. Once they saw how well the
programs worked then they would be forced to begin using them to predict
their own earthquakes.
United Nations personnel appeared to like this concept when I
formally proposed it to them in July of 2004.
http://www.unisdr.org/wcdr-dialogue/t3-dialogue.htm#34
And they discussed it repeatedly in their summary reports of the
ideas proposed during that Internet Web site based conference. But no
governments or disaster mitigation groups expressed any interest in
developing the concept.
Since the starting point for the proposed forecasting program
development efforts would be the computer programs that I already have
running, success would be guaranteed.
So, the important question would then be,
"How many computer programmers and other researchers would be interested in
helping with such an effort?"
This type of work is quite interesting and exciting when you become
actively involved with it. You can generate a forecast. And then when the
earthquake occurs where you expected you can get a pretty good shock. If
the earthquake is also destructive the experience can be rather frightening.
The importance of the work should be obvious. More than a quarter of
million people reportedly perished in connection with the December 26, 2004
earthquake generated tsunami (tidal wave) in the Indian Ocean. My present
forecasting computer programs did not become operational until several weeks
after it occurred. But when I ran my EM signal data from around that time
through the programs the results indicated to me that the earthquake could
have been predicted.
Why devote valuable free time to developing computer programs which
do ordinary things when that time could be devoted to developing programs
which might eventually help save tens or even hundreds of thousands of
lives?
--
http://mail.python.org/mailman/listinfo/python-list
