Earthquake Forecasting Program Jan. 20, 2005

From: edgrsprj (edgrsprj_at_ix.netcom.com)
Date: 01/20/05 

Date: Thu, 20 Jan 2005 08:38:02 GMT

EARTHQUAKE FORECASTING PROGRAM

Newsgroup Readers - Please forward copies of this report to any government
agency and disaster mitigation officials you believe might be interested in
earthquake forecasts. The report will also explain why copies have been
posted to weather related Newsgroups.

SUMMARY - This report is primarily intended to let government and disaster
mitigation officials around the world know about the existence of the
following earthquake forecasting data Web page:

http://www.freewebz.com/eq-forecasting/Data.html

       The report also contains information regarding a much larger
earthquake forecasting data table that might be of interest to statisticians
and weather forecasters.

Posted on January 20, 2005 edgrsprj@ix.netcom.com
Web sites: http://www.freewebz.com/eq-forecasting/index.html
http://home.netcom.com/~edgrsprj/index.html

       For responses to this report which are going to be posted to just one
Newsgroup, sci.geo.earthquakes might be the most appropriate choice. Names
have not been used here in part to make it easier for me to deal with the
incoming spam e-mail which always seems to follow the posting of a note to a
Newsgroup. The information in this report represents expressions of
personal opinion.

Earthquake Forecasting Program

       That previously mentioned Data.html Web page contains a table which
lists earthquake forecasting data generated by the latest version of my Perl
computer language Earthquake and Tornado Data Evaluation Computer Program
(ETDPROG.pl). The table data are not true earthquake predictions. I am
working with what I believe are earthquake fault zone activity related
electromagnetic energy field fluctuations (EM signals). And the table data
are probability type numbers which show how well the ETDPROG.pl program
indicated those EM signals matched the earthquakes in my database of about
28,000 earthquakes. Information at the top of the Web page explains how to
interpret the data.

       At any time about 100 earthquakes are listed in the data table on
that Web page. And just making a guess I believe that a significant
earthquake will then probably occur within a few months at the location of
perhaps 1 in 25 on the list. In part the goal of those data is to recommend
that government officials around the world check the locations of the
earthquakes listed in the table to see if there might be any obvious signs
of approaching seismic activity in those areas.

       This latest ETDPROG.pl program can quickly compare any or all of the
28,000 earthquakes in my database file with one or more of the 2000 EM
signal times in my database going back to late in 1994. The computer number
crunching and display operations for doing that which used to take about 8
hours are now possible in as little as 2 minutes. As the data table on that
Data.html Web page shows, the most effective data display approach appears
to be to create warning signal time slices or windows, each 3 months long
which are offset from one another by about 10 days. I believe that by
comparing those time windows with one another it can at times be seen that
seismic activity is approaching in some area.

       At this time the most serious weakness in the program is probably the
fact that it is completely reliant on just one person. And if I am busy
with other things I cannot run any data analyses or even collect many
earthquake precursor data. The program was not running between November 6
of 2004 and January 13, 2005 while a major software rewrite was being done.
No warnings were circulated and no data were being stored on my Data.html
Web page during the weeks before that catastrophic December 26, 2004
earthquake in the Indian Ocean area.

       Another program limitation is probably associated with the length of
those time windows. During any given 3 month period of time there will
undoubtedly be some significant seismic activity in certain locations such
as in the Japan area. And so Japan area earthquakes will almost always be
listed in the data table. The program is presently probably most helpful
for spotting the approach of significant earthquakes which are likely to
occur in locations where they are less frequent such as in most South
America countries.

STATISTICAL ANALYSES AND WEATHER DATA DISPLAYS

       As stated earlier, the primary goal in posting this report to a
number of Newsgroups is to let government and disaster mitigation officials
around the world know about the existence of that Data.html Web page. A
secondary goal is to provide statisticians and weather forecasters with the
following information:

       Only about 100 of the highest probability earthquakes are presently
being listed on my Data.html Web page. That is because of bandwidth
limitations. If I store a file there which is too large and enough people
download it then the Web site will stop running until the next month when
the bandwidth counter resets itself. However in addition to the table on
that Data.html Web page I can generate ones which have similar data for all
of the 28,000 earthquakes in my database file and as many time slices or
windows as desired. The time windows can also be of any length desired. 3
month duration windows simply appear to work the best at this time.

       The data in those larger tables are quite impressive. You can
visually examine them and easily see the patterns surrounding the occurrence
of these catastrophic earthquakes. I don't have the time or resources to
work on this. But if there are any statisticians who would like to try to
develop some computer program routines which can automatically identify
those patterns in the data then I believe that that could greatly enhance
the power of this approach to forecasting earthquakes. And since the
ETDPROG.pl program can itself be easily used by researchers around the world
and since the necessary precursor data are probably widely available for
free I believe that with improvements like that the program might enable us
to accurately spot the approach of a good percentage of our highly
destructive earthquakes.

       Over the years weather forecasters around the world have developed
powerful, high quality data display programs for their meteorological data.
And my data files should be compatible with them. As the table on that
Data.html Web page shows, the data are organized into columns which can be
easily processed by both spread*** and conventional data processing
computer programs. A color encoded "movie" of the time window data showing
where around the world they are indicating an earthquake could be about to
occur as time progresses would undoubtedly be impressive.

       If there is anyone interested in considering such a project then they
can try contacting me by e-mail to see what might be arranged. There are
e-mail addresses at my Web sites. A few of the larger data files that I
have generated were about 6.5 Meg in size each. In a zipped form they were
around 2.5 Meg. If you do send me an e-mail note and do not hear from me
within a few days then try contacting me again using a different heading on
your e-mail note. My spam filter occasionally deletes a legitimate e-mail
letter because of some word or phrase that was in its heading or body.

Quantcast