--
Edward Cranswick                Tel: 303-273-8609
US Geological Survey, MS 966    Fax: 303-273-8600
PO Box 25046, Federal Center    cranswick@........
Denver, CO 80225-0046  USA      E.M. Forster said, "Only connect".


------------- Begin Forwarded Message -------------

Mime-Version: 1.0
X-Sender: mblanpied@..................
Date: Tue, 21 Sep 1999 13:04:41 -0700
To: wehz@........................
From: William Lee  (by way of Michael Blanpied)
Subject: Taiwan earthquake information from Willie Lee

Dear Mike:

I was probably the first person outside Taiwan to know about this
earthquake.  The e-mail containing the shock arrived several
minutes after the earthquake, and a dozen more later on its
aftershocks.  The automatic system produced very reasonable
estimates for epicenter, focal depth, and magnitude of the main
shock in about 100 seconds after the earthquake's origin time.

I faxed all the info to Dave Oppenheimer in the early afternoon
of 9/20 after he called me.  I came to the office later and left
a big map with the "dots" in the press room of Building 3.

Last night, I wrote a short note about the Taiwan EQ and
sent it primarily to John Filson as he left me a phone message
for more info.  Please feel free to circulate this e-mail.

One sad comment is that I was too optimistic.  It appears that
the epicentral area was hard hit and the dead toll is mounting.
The peak PGA recorded at Chiayi (about 50 km southwest of the
quake) was about 16% g acceleration.  There are closer strong-
motion stations, but they are not telemetered, and it appears
that most phone communications are out as of now.  Internet to
Taiwan is still out as of now (12 noon; 9/21).

After an intensive programs to instrument urban areas during
1991-1996, Taiwan has the highest concentrations of modern
digital accelerographs of the world (about 5 times closer station
spacing than Japan):  about 1,000 digital accelerograhs in free
sites, and an equivalence of over 500 accelerographs in structures
(about 50 structures have been instrumented with an array of
accelerometers and are monitored by a realtime system).  The results
reported below derived from about 70 continuously telemetered
digital accelerographs, many at the local weather stations in
urban areas.

A few simple facts:
(1) Taiwan has an area of about 8% of California or Japan.
     More than 1/2 of the area are mountainous (up to 4,000 meters).
(2) It is very seismic active, because Philippines plate/Pacific
     Plate/Eurasia Plate converge there.  Seismicity is about
     3-5 times higher than California.
(3) I am fortunate to have the opportunity to participate
     in their earthquake program while I was at the USGS before
     and after my retirement.
(4) The Taiwan instrumentation program costed about $40 millions
     in 6 years (about 90% were spent on strong-motion).  It was
     completed on time, under the budget (the program was budgeted
     at $72 million), and with only a few additional staff from
     their original staff of about 30 people.
(5) Finally, I believe the "KISS" principal works, i.e., keeping
     it simple.

Willie

====== Prepared by about 11 p.m, 9/20/99, Calif. time =============
 
              Taiwan Quake: Results and Performance of
      Taiwan's Rapid Earthquake Information Release System

    At 1:47 a.m. Taiwan local time, a major earthquake occurred in
Taiwan, about 140 km SSW of Taipei, its capital.  Fortunately,
the epicenter was located in the mountainous area, about 50 km
southeast and northeast from two major cities: Taichung and Chiayi,
respectively.  Hundreds of people were killed and thousands were
injured.  The disaster would be similar to the recent Turkey quake
if the epicenter were 50 km west in the coastal plain, where the
population is very dense.

    I received the main shock information in an e-mail, several
minutes after the earthquake occurred.  The Taiwan's realtime
system is based on about 70 continuously digital-telemetered
accelerographs all over the island, and has been in operation
since March 5, 1996.  Data were automatically processed in
Taipei using hardware and software designed and implemented
in the early 1990's (Lee et al., 1996; Shin et al., 1996;).
Early results had been published by Teng et al. (1997) and
Wu et al. (1997). [see References after the table].

    In the following table, I listed the results received as of
10 p.m., Sept. 20, California time (or 1 p.m., Taiwan local time),
i.e., for a period of about 11 hours after the main shock.

    Several general conclusions are:

(1) The Taiwan system performed as designed and has shown to be
     capable of handling probably the largest earthquake of this
     century in Taiwan.  The first located aftershock occurred
     just 10 minutes after the main shock.  There were many
     aftershocks and some were quite large.

(2) The preliminary result of the main shock is reasonably
     accurate.  The local magnitude was estimated to be 7.3
     in about 100 seconds after the earthquake's origin time
     and agreed well with the Mw = 7.6 determined by NEIC in about
     3 hours after the earthquake.  Focal depth was 10 km vs 5 km
     given by the moment tensor solution of NEIC.

(3) Hypocenter and magnitude were determined in about 100 seconds
     after the origin time, or about 90 seconds after sufficient
     number of stations recorded the first P-arrivals.  Some
     extra time was spent in making sure that the largest
     amplitudes had been recorded.  For some aftershocks, the
     processing time (Proc. Time in the following table) was less
     than 1 minute.  The Proc. Time is defined to be the elapsed
     time from the earthquake's origin time to the time the results
     sent to the e-mail server.

(4) E-mail transit time is quick in the beginning (middle of the
     night), taking only several minutes to get from Taipei to my
     home.  However, by 5 a.m. local time, e-mails were backed up
     and for several hours, no e-mails arrived from Taiwan.  Phone
     calls to Taiwan are not possible at this moment (12 hours after
     the quake and e-mails from Taiwan started to dribble in.


    The following lists the results from the Taiwan realtime system
as received in e-mails after the first 11 hours of the quake.
Results after the first two hours are probably not complete as
the e-mails could not get out of Taiwan rapidly.

-------------------------------------------------------------------
Date    Origin Time  Latitude  Longitude  Depth(Km)  ML  Proc. Time
-------------------------------------------------------------------

9/20/99 17:47:15.89  23.87 N   120.75 E   10.0      7.3    102 sec.

9/20/99 17:57:17.16  23.95 N   121.03 E    6.9      6.1     57 sec.

9/20/99 18:03:42.64  23.80 N   120.85 E    5.0      6.5     75 sec.

9/20/99 18:11:27.71  24.03 N   120.98 E    5.2      5.2     56 sec.

9/20/99 18:16:21.01  23.86 N   121.03 E   13.5      6.9     85 sec.

9/20/99 18:21:31.57  23.99 N   121.07 E    8.9      5.1     50 sec.

9/20/99 18:32:55.60  23.82 N   121.01 E    2.9      5.2     62 sec.

9/20/99 19:28:43.48  23.88 N   120.97 E    2.4      4.9     53 sec.

9/20/99 19:40:33.27  23.57 N   120.88 E    2.5      5.4     55 sec.

9/20/99 19:57:52.19  24.17 N   120.75 E    6.9      5.9     76 sec.

9/20/99 20:02:15.36  24.01 N   120.67 E    2.5      5.7     69 sec.

9/21/99 01:37:09.01  23.73 N   120.75 E    2.5      5.2     55 sec.

9/21/99 02:24:46.48  23.93 N   121.00 E    8.4      4.9     56 sec.

9/21/99 03:31:49.58  23.99 N   121.02 E    3.1      5.1     58 sec.
-------------------------------------------------------------------

REFERENCES:

Lee WHK, Shin TC and Teng TL (1996). Design and implementation of
   earthquake early warning systems in Taiwan. Proc. 11th World Conf.
   Earthq. Eng., Paper No. 2133.

Shin TC, Tsai YB, and Wu YM (1996). Rapid response of large earthquake
   in Taiwan using a realtime telemetered network of digital
   accelerographs. Proc. 11th World Conf. Earthq. Eng., Paper No. 2137.

Teng TL, Wu L, Shin TC, Tsai YB, and Lee WHK (1997).  One minute after:
   strong motion map, effective epicenter, and effective magnitude.
   Bull. Seism. Soc. Am., vol 87, p 1209-1219.

Wu YM, Shin TC, Chen CC, Tsai YB, Lee WHK, and Teng TL (1997).
   Taiwan rapid earthquake information release system. Seism. Res.
   Letters, vol 68, p 931-943.




 

------------- End Forwarded Message -------------