1212KB - 東京大学地震研究所

ΐ ῒ ῐ ῏ ῑ ῕ ῔
Bull. Earthq. Res. Inst.
Univ. Tokyo
Vol. 2- ῍,**2῎ pp. ,/+ῌ,0-
The ,**, Seismic Refraction/Reflection Surveys in the
Kozu-Matsuda Fault Zone and Ashigara Valley
Yasuhisa Tanaka+῎*, Kazuki Koketsu+῎, Hiroaki Yamanaka,῎, Tomiichi Uetake-῎, Toshikazu
Tanada.῎, Shinji Kawasaki/῎ and Hiroe Miyake+῎
+῎
Earthquake Research Institute, University of Tokyo, Tokyo, Japan
Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology,
Yokohama, Japan
-῎
R & D Center, Tokyo Electric Power Company, Yokohama, Japan
.῎
Hot Springs Research Institute of Kanagawa Prefecture, Odawara, Japan
/῎
JGI, Inc., Tokyo, Japan
,῎
Abstract
We conducted seismic surveys in western Kanagawa prefecture including the Kozu-Matsuda
fault zone and Ashigara valley, in February and March ,**,. The reflection surveys were carried
out using vibrator sources along , receiver lines. Line A is +/-km long and extends east to west
across the Kozu-Matsuda fault zone and Ashigara valley. Line B is +*-km long and extends north to
south in the valley. ++/+ receivers were deployed at intervals of ,/ m for ,+* vibration points. --D
refraction surveys were also conducted with .-dynamite shots at both sides of the lines A and B and
repeated vibroseis sweeps were made at - points. In addition to the ++/+ reflection receivers, +*2
receivers were deployed along the / lines between the shot points, and +,2 independent receivers
were located over a wide area in the Ashigara valley. We overview the surveys and present datasets
of seismograms and travel times.
Key words : Kozu-Matsuda fault zone, Ashigara valley, refraction survey, reflection survey, seismogram
+.
Introduction
fault are called Oiso-type earthquakes.
The Kozu-Matsuda fault is an active fault +,-km
type earthquake is one of - regions of earthquakes in
long with a westward declination. The fault is con-
the Sagami trough zone, and has relatively short
nected to the Matsuda-Kita fault at its northwestern
interval between earthquakes (Matsuda, +32/). These
The Oiso-
end. The average slip rate of the fault is about ,ῌ-
earthquakes cause uplifts in the Oiso hills and subsi-
mm/y, and the average recurrence interval is esti-
dence in the Ashigara valley. Earthquake Research
mated to be 2**ῌ+-** years (Earthquake Research
Committee (,**0) evaluates the magnitudes of the
Committee, ,**0). The most recent event on the fault
earthquakes to be about 1./ from vertical displace-
occurred 0/*ῌ3** years ago (Kanagawa Pref., ,**-).
ment and length of the fault.
The Kozu-Matsuda fault is considered to be a splay
The Ashigara valley is a sedimentary basin .-km
fault from the Sagami trough and constructs part of
wide east to west and +,-km long north to south in
subduction boundary of the Philippine Sea plate.
the western part of Kanagawa prefecture, which is
Several historical events in and around the Sagami
located about 2* km from southwest Tokyo. Figure +
trough have been identified (Ishibashi, +32/ ; Tsuji,
shows this location and the topography around the
+32/), and the Kozu-Matsuda fault zone is also located
Ashigara valley. The valley is bounded by the Hak-
in the source region of the +3,- Kanto earthquake
one volcanos to the west and the Tanzawa moun-
(Matsuda, +33-). The events on the Kozu-Matsuda
tains to the north. The valley is open to the Sagami
* e-mail : [email protected] (+ῌ+ῌ+ Yayoi, Bunkyo-ku, Tokyo ++-ῌ**-,, Japan)
ῌ 251 ῌ
Y. Tanaka, K. Koketsu, H. Yamanaka, T. Uetake, T. Tanada, S. Kawasaki and H. Miyake
bay to the south, so it appears to extend from north-
sive velocity structure model in the Kozu-Matsuda
west to southeast. The Kozu-Matsuda fault zone is
fault zone and Ashigara valley.
located along the boundary between the Ashigara
,.
valley and the Oiso hills (Figure ,).
Overview
Several studies have been carried out to obtain
The ,**, surveys were performed from February
an understanding of the ground motion characteris-
,* to March +., by a joint team of the Earthquake
tics and velocity structure in the Ashigara valley.
Research Institute, University of Tokyo and other
Uetake and Kudo (+332, ,**/) found large variations
institutions. We first conducted reflection surveys
of seismic amplification, depending on the locations
along , receiver lines with large vibrator sources.
of stations in the Ashigara valley, using records from
We then conducted refraction surveys with . explo-
eastern Yamanashi and western Kanagawa prefec-
sions and - vibrator sources. The sources and receiv-
ture earthquakes in +330. They also showed varia-
ers are shown in Figure ,. The reflection surveys aim
tions of coda waveforms for the Japan Meteorologi-
to establish the structure of reflection boundaries
cal Agency (JMA) magnitude ../ earthquake in west-
and the fault along the receiver lines, while the re-
ern Kanagawa prefecture, and confirmed them to be
fraction surveys aim to reveal the velocity structure
due to the complex --D velocity structure beneath
of layers along the lines and the --D velocity struc-
the Ashigara valley. These studies suggest that we
ture beneath the Ashigara valley.
need accurate velocity structure models to precisely
The reflection sources were set every +** m
predict strong ground motions from an earthquake.
along the receiver lines with a large vibrator. The
In addition, the detailed velocity structure model
sweep frequency was changed from 0 to /* Hz, and
around the Kozu-Matsuda fault zone can help us
the sweep length was +0 s. We used vertical-com-
explore the geometry of this fault. Some reflection
ponent geophones with natural frequencies of +* Hz
surveys (Hasegawa et al., +33+ ; Sato et al., ,**/) and
as receivers located every ,/ m along the lines. The
refraction surveys (Kasahara et al., ,**, ; Sato et al.,
+/-km long receiver line A consists of +,, reflection
,**/) have already been conducted in parts of the
sources and 1/+ receivers. The +*-km long line B has
target region of this experiment including the Kozu-
32 sources and .** receivers.
Matsuda fault zone, Ashigara valley, and Oiso hills.
reflection surveys were logged by a seismic record-
However, the ,**, seismic refraction/reflection sur-
ing system called GDAPS-. with a sampling rate of .
veys are the first attempt to construct a comprehen-
ms and a recording length of 0 s.
The records of the
The refraction
sources were explosions of ,** kg of dynamite at TD
+ and TD,, /* kg of dynamite at TD-, and +** kg of
dynamite at TD. (Table +).
1/+ and .** receivers
were deployed along lines A and B ; +*2 independent
stations were installed around the sources along the
lines connecting the C, D, E, F, and H source points ;
and, +,2 independent stations were located over a
wide area in the Ashigara valley. The independent
stations cover the sparse zone of stations in the
valley and increase the numbers of pairs between
sources and stations. The receivers along lines C, D,
E, F, and H consisted of vertical-component seismographs of L-,,D, and loggers of Datamark LS2***SH/
WD with a sampling frequency of +** Hz and a recording length of +* minutes. We also used - vibraFig. +. Location map of the survey area in western
Kanagawa prefecture. The Ashigara valley is located in the southwestern Kanto region of Japan.
Broken squares are the source region of the +3,Kanto earthquake (Matsuda, +33-).
tor sources, which repeatedly swept +** times at
VP,, VP-, and VP. on lines A and B. Stacking the
recorded waveforms and cross-correlations with the
sweep signals clarified the distribution of reflection
ῌ 252 ῌ
The ,**, Surveys in the Kozu-Matsuda Fault Zone and Ashigara Valley
boundaries along the receiver lines. The refracted
shows a summary of the recordings along the re-
waves from the vibrator sources were recorded at the
ceiver lines. The TD+ and TD- shots are recorded in
receivers along lines A and B.
almost all of the survey lines, and the TD, and TD.
shots are recorded in lines close to them.
-.
Results
We show the corrected seismograms in Figures
The reflection seismograms were processed us-
-, ., /, and 0, respectively, and obtained travel times
ing the Common Depth Point (CDP) method. The
along lines A, B, D, and E. The record sections of line
refraction records were corrected for errors due to
A show clear arrivals from the shots of TD+, TD,,
the altitudes of the sources and receivers assuming
VP,, and VP.. Plotted travel times show a gradual
P-wave velocity῍+.2 km/s for the top of the layer.
increase of apparent P-wave velocity at some dis-
We read travel times for each pair of source and
tance from each shot, and a sudden change of veloc-
receiver from these corrected seismograms. Table ,
ity around the Kozu-Matsuda fault zone. The record
Fig. ,. Distribution of receivers and survey lines. Stars indicate the shot points of TD+, TD,,
TD-, and TD.. Triangles denote the vibroseis points of VP,, VP-, and VP.. VP- and VP.
coincide with the refraction receivers No.,-33 and No..33, respectively. Crosses and lines
indicate receivers. Lines A and B consist of reflection and refraction receivers. Lines D and
E consist of refraction receivers. Solid line illustrates the Kozu-Matsuda fault (KMF) and
broken line illustrates the Matsuda-Kita fault (NKMF).
Table +. Locations, shot times, and charge amounts of dynamite shots (TD+ to TD.) and vibroseis
shots (VP, to VP.) for refraction surveys. The shots are in Japan Standard Time (JST).
ῌ 253 ῌ
Y. Tanaka, K. Koketsu, H. Yamanaka, T. Uetake, T. Tanada, S. Kawasaki and H. Miyake
Table ,. Quality of records for the pairs of shots and receiver lines. ῎ : Travel
times were picked from almost all records. ῏ : Travel times were picked from
some records. ῍ : Travel times were picked from only a few records.
Fig. -. Record sections (top : shot TD+, second : vibroseis VP,, third : vibroseis VP., fourth : shot TD,) and the
travel time curve (bottom) for line A. Amplitude is recovered by Automatic Gain Control (AGC) with .*** ms
gate length. Travel time curve is reduced with a reduction velocity of 0.* km/s.
ῌ 254 ῌ
The ,**, Surveys in the Kozu-Matsuda Fault Zone and Ashigara Valley
Fig. .. Record sections (top : shot TD-, second : vibroseis VP., third : vibroseis VP-, fourth : shot TD.) and travel
time curve (bottom) for line B.
ῌ 255 ῌ
Y. Tanaka, K. Koketsu, H. Yamanaka, T. Uetake, T. Tanada, S. Kawasaki and H. Miyake
Fig. /.
Fig. 0.
Record section of shot TD+ (top) and travel time curve (bottom) for line D.
Record sections (top : shot TD+, middle : shot TD-) and travel time curve (bottom) for line E.
ῌ 256 ῌ
The ,**, Surveys in the Kozu-Matsuda Fault Zone and Ashigara Valley
Table -.
Locations of stations and travel times.
ῌ 257 ῌ
Y. Tanaka, K. Koketsu, H. Yamanaka, T. Uetake, T. Tanada, S. Kawasaki and H. Miyake
Table -.
(continued)
ῌ 258 ῌ
The ,**, Surveys in the Kozu-Matsuda Fault Zone and Ashigara Valley
Table -.
(continued)
ῌ 259 ῌ
Y. Tanaka, K. Koketsu, H. Yamanaka, T. Uetake, T. Tanada, S. Kawasaki and H. Miyake
Table -.
(continued)
ῌ 260 ῌ
The ,**, Surveys in the Kozu-Matsuda Fault Zone and Ashigara Valley
Table -.
section and travel time curve for line B also show
(continued)
Acknowledgments
clear arrivals and an increase of apparent velocity.
We thank Profs. Atsushi Yasuda and Shin’ichi
The record section of line D also shows clear arrivals
Sakai for their valuable comments and constructive
from the TD+ shot. The TD- shot was also observed
suggestions. We all thank the survey participants
well on line D, although TD- is located away from
from Hokkaido University, Kyoto University, Kyu-
line D. The record sections of line E show arrivals
shu University, Kyushu Tokai University, Nagoya
from TD+ and TD-. The travel time curve for TD+
University, Tokai University, Disaster Management
also shows a sudden change of velocity around the
Bureau of Kanagawa Prefecture, Tokyo Institute of
Kozu-Matsuda fault zone as a local low apparent
Technology, University of Tokyo, Yokohama City
velocity zone around the Kozu-Matsuda fault zone
University, Hot Spring Research Institute of Kana-
and the shot point of TD-.
gawa Prefecture, and National Research Institute of
Table - lists the locations of stations and travel
Fire and Disaster, for their e#orts during the survey
times obtained for the explosions. All of the stations
and its preparation.
We also thank the Cities of
tabled are also plotted in Figure ,.
The error in
Odawara, Minamiashigara, Hiratsuka, and Hadano,
reading travel times should be mostly within +* ms,
and the Towns of Yamakita, Matsuda, Kaisei, Oi,
but the records for shots away from the receiver lines
Nakai, Ninomiya, and Oiso, and JGI, Inc. This survey
provide rather large errors in their travel times of up
was sponsored by the Coordinating Committee of
to -* ms.
Earthquake Prediction Research, and Disaster Management Bureau of Kanagawa Prefecture.
..
Conclusions
We conducted seismic refraction/reflection sur-
veys in the Kozu-Matsuda fault zone and Ashigara
valley in order to construct a --D velocity structure
model of the Kozu-Matsuda fault. From the reflection surveys, 2 reflection profiles were obtained.
From the refraction surveys, we picked /2. travel
times from the corrected seismograms.
References
Earthquake Research Committee, ,**0, A partial revision of
a long-term evaluation of the Kannawa/Kozu-Matsuda
fault zone, Publications of Earthquake Research Committee ῌJanuary-December ,**/ῌ, II, -1/ῌ.*3 (in Japanese).
Hasegawa I., K. Ito, T. Satoh, S. Watanabe, M. Komazawa, Y.
Ninomiya, H. Ito, T. Tosha and M. Sugihara, +33+, Underground structure in the Ashigara plain by refraction
and reflection method, +33+ SSJ Fall Meeting, C.* (in
Japanese).
Ishibashi K., +32/, Possibility of huge earthquakes around
ῌ 261 ῌ
Y. Tanaka, K. Koketsu, H. Yamanaka, T. Uetake, T. Tanada, S. Kawasaki and H. Miyake
Odawara, Chikyu Monthly, 1., .,*ῌ.,0 (in Japanese).
Kanagawa Pref., ,**-, Report of the result from Heisei +.
Kozu-Matsuda fault zone survey in Kanagawa Pref., pp /0
(in Japanese).
Kasahara K., T. Tanaka, T. Ikawa, Y. Ohta, S. Kawasaki and
T. Ito, ,**,, New analysis of seismic reflection data 3*AS and 3+-TAN (NIED), Ashigara and Tanzawa areas,
central Japan, Bull. Earthq. Res. Inst. Univ. Tokyo, 11,
,01ῌ,1/ (in Japanese with English abstract).
Matsuda T., +32/, Oiso-type earthquakes, Chikyu Monthly,
1., .1,ῌ.11 (in Japanese).
Matsuda T., +33-, Seismotectonics of the Northwestern Sagami Bay Area, J. Geogr., +*,, -/.ῌ-0. (in Japanese).
Sato H., N. Hirata, K. Koketsu, D. Okaya, S. Abe, R. Kobay-
ashi, M. Matsubara, T. Iwasaki, T. Ito, T. Ikawa, T.
Kawanaka, K. Kasahara and S. Harder, ,**/, Earthquake source fault beneath Tokyo, Science, -*3, .0,ῌ.0..
Tsuji Y., +32/, Historical earthquakes striking Odawara,
Chikyu Monthly, 1., .-+ῌ.-3 (in Japanese).
Uetake T. and K. Kudo, +332, Evaluation of site e#ects in a
wide frequency band at Ashigara valley, Japan, using
strong motion records from remote and large events,
Zisin, /*, -31ῌ.+. (in Japanese with English abstract).
Uetake T. and K. Kudo, ,**/, Assessment of site e#ects on
seismic motion in Ashigara valley, Japan, Bull. Seismol.
Soc. Am., 3/, ,,31ῌ,-+1.
῍Received August ., ,**2῎
῍Accepted November ,1, ,**2῎
ῌ 262 ῌ
The ,**, Surveys in the Kozu-Matsuda Fault Zone and Ashigara Valley
ῌ ,**, D¦§¨+ῌ©ª«¬+ῌ­¦®¯,ῌ°±²«-ῌ³D´µ.ῌ#¶·¸/ῌi¹º»+
-
.
!"#$%&
/
'
+
,
( +*2 )*+,-./0.123456789) +,2 )
,**, : , ;<= - ;.<>? @ABῌCDEFGH
IJK L4=9(MNOPQRSTU0VWXYO
Z[,-\]^!"#$_`\ab.c? defῌg
Xh. /0.123456789)OPQRiTU0
jf.kKWXYlm\noc5 lmpfOqHr
VWXYsOt.uv34K wxyIRlmOz
R{|}~€.kKgjflmI @ABῌCDEF
 ‚=45ƒV„…PQ *+†‡ˆ‰5Š‹Œ
G*+Z[,-\_.ŽK +/ km O9( A H,-
€\‘’
\“E’K +* km O9( B I”‰5 •5 –. A,
B —9(˜™O . ).š>K›|œ|ž H - )O
: @ABῌCDEFG Z[,- defl
{|}~€ Ÿ.kK/0Vdeflm\noc5
Ÿ)R A B 9(O ++/+ ).¡¢
m gjflm „…
£)\¤¥ / 9
263