DETECTION ACCURACY OF ACTIVE FAULT
BASEDON SEISMIC REFLECTION METHOD DATA
Katsumi Ebisawa, Toshio Kogure, Hideaki Tsutsumi, Shohei Motohashi and Masaharu Sakagami
Japan Nuclear Energy Safety Organization (JNES), Japan
Abstract
The authors examined the accuracy of detection of active fault detected by seismic reflection method
using the active fault data at Headquarters for Earthquake Research Promotion and National Institute of
Advanced Industrial Science and Technology. It was demonstrated that the detection possibility of active
faults was quite large and the detection percentage was about 90 %, when the seismic reflection method was
applied suitably against them under sedimentary layer. The percentage of dip-slip faults to total faults was
detected 86 % and that for strike-slip faults is 14 %. The detection percentage of the former fault was about
six times as large as that of the latter. It was predicted that active fault of fault length over 6 km may be
detected by the seismic reflection method.
.
1
1. Introduction
As a part of the advanced seismic design guideline, Japan Nuclear Energy Safety Organization has
been conducting the development of methodology for evaluating the exceedance probability spectra of
seismic motion occurred by the buried fault. On the other hand, Headquarters for Earthquake Research
Promotion (HERP) and National Institute of Advanced Industrial Science and Technology (AIST) have
been conducting the investigation of active faults by using the various investigation methods (geophysical
prospecting included the seismic reflection method, trenching, boring and outcrop investigation etc.).
In order to estimate the accuracy of detection of active faults by the seismic reflection method, the
authors examined the accuracy using 55 active faults data in HERP’s and AIST’s database 1),2). Furthermore
we also examined the relationship between the fault displacement obtained by the seismic reflection method
and the fault characteristics (fault length, average slip rate etc.) obtained by trenching, boring and outcrop
investigation, etc.
This paper describes the above detection accuracy and the relationship between the fault displacement
and the fault characteristics.
2. Overview of active fault data
(1) Data used the examination
The active fault data of Headquarters for Earthquake Research Promotion (HERP) 1) and National
Institute of Advanced Industrial Science and Technology (AIST) 2) were used to examine.
The active faults in HERP’s and AIST’s data were classified into the following two types.
- Buried fault: Fault with no surface appearance due to overburden
- Surface fault: Fault with any surface appearance such as a fault outcrop
Both HERP’s and AIST’s data were obtained by the seismic reflection method and other investigation
methods that are the trench, boring and outcrop, etc. The parameters of the data obtained by the seismic
reflection method are as follows.
- Displacement at the sedimentary layer
- Displacement at the basement
The sedimentary layer is defined as after the neogene period. The basement is defined as before the
paleogene period.
On the other hand, the parameters of the data obtained by the trench and outcrop etc. are as follows.
- Types of displacement; dip-slip or strike-slip or dip & strike-slip
- Average slip rate
- Fault length
The confidence for the above three parameters is described in the HERP’s data but that is not described in
the AIST’s data.
(2) Arrangement way of data
Both the displacement data obtained by seismic reflection method, and the fault characteristics data
obtained by the trenching, boring and outcrop investigation, etc. in the HERP’s data were arranged for the
buried and surface faults as shown in Table 1 (A). Those in the AIST’s data were also arranged as shown in
Table 1 (B). The total number of faults in Table 1 (A) and (B) is 55.
2
The profile location and the displacement obtained by the seismic reflection method for the Kyoto-Nara
north basin fault zone at the sedimentary layer and basement are shown in Fig.1. The prospecting depth is
about 800 m. The former displacement is from 50 to 100 m. The latter is 150 to 200 m.
The fault of Tonami west plain was investigated by the seismic reflection method, Gravity method and
the boring. The prospecting depth is about 1000 m. The location and the displacement of this fault are
shown with the columnar by the boring in Fig.2. The displacement at sedimentary layer is about 380 m.
Table 1 (A) Active fault data of Headquarter for Earthquake Research Promotion
Surface fault
Buried fault
Active fault name
Ishikari-teichi-toen
fault zone
Kitakami-teichi-seien
fault zone
Nagamachi-Rifusen
fault zone
Motoarakawa fault zone
Isehara fault
Kurehayama fault zone
Morimoto & Togashi
fault zone
Isewan fault zone
Yanagase & Sekigahara
fault zone
Kyoto-bonchi – Narabonchi fault zone
(Southern areas)
[Nara-bonchi-toen fault
zone]
Uemachi fault zone
Ikoma fault zone
Chuo Tectonic Line fault
zone
(Kongo-sanchi-toen –
Iyo-nada)
Mashike-sanchi-toen
fault zone
Tobetu fault
Hakodate-heiya-seien
fault zone
Yamagata-bonchi fault
zone
Tachikawa fault zone
Kannawa & Kozu –
Matuda fault zone
Miura-hanto faults
(Main areas)
Tukioka fault zone
Tonami-heiya fault zone
(Western aeras)
Tonami-heiya fault zone
(Eestern aeras)
Inadani fault zone
Fujikawa-kako
fault
zone
Yoro-KuwanaYokkaichi fault zone
Suzuka-toen fault zone
Nosaka & Syufukuji
fault zone
Biwako-seigan
fault
zone
Kohoku-sanchi
fault
zone
Mikata fault zone
Hanaore fault zone
(Southern areas)
Arima-Takatsuki fault
zone
Yamasaki fault zone
Kikukawa fault
Nagao fault zone
Seismic reflection method
Displacement
Displacement
at basement
at sedimentary
(m)
layer (m)
(5)
500
150-200
150
500-1500
Slip direction
Confidence
Dip
***
Fault characteristics
Average velocity rate
(m/103years)
Confidence
0.8-1.5
*
Fault length
(km)
Confidence
66
**
200
Dip
***
0.2-0.4
**
62
**
150-200
Dip
***
0.5-0.7
**
21-40
*
200
50
500
50-100
!
Dip
Dip
Dip
***
***
***
(0.05)
0.3-0.4
0.4-0.6
1
!
**
*
*
25
21
22
26
!
**
**
**
(10-20)
30-150
Dip
Dip
***
***
0.1-0.2
0.6-0.8
*
**
42
48
**
**
50-100
Dip
***
0.6
**
35
**
Dip
Dip
Strike
***
***
***
0.4
0.5-1
1.8-3.5
**
*
**
42
38
66-74
**
**
**
400-500
50
?
?
Dip
***
1
*
60
**
?
?
?
?
Dip
Dip
***
***
0.1-0.2
0.2-0.4
*
**
20
24
*
**
350-400
Dip
***
1-2
**
60
**
?
?
?
?
Strike
Dip
***
!
0.2-0.3
1-3
**
!
33
25
**
!
?
?
Strike
***
0.5-2.8
*
14
**
380
Dip
Dip
***
***
0.4
0.3-0.4
**
**
30
26
*
**
?
Dip
***
0.3-0.4
**
30
**
200
(30)
Dip
Dip
***
!
0.5-1.3
2-7
*
!
78
20
**
!
1500
Dip
***
3-4
**
60
***
Dip
Dip &
strike
Dip
***
***
0.1-0.4
0.2-0.8
**
**
34-47
31
**
**
***
1.1-1.6
*
59
***
Dip &
strike
Dip
Dip
***
0.5
**
25
***
***
***
0.8
0.3
*
*
26
15
***
**
50
Strike
***
1.5
**
55
**
??
?
Dip
Strike
Dip
**
***
***
0.06-0.09
!
0.05-0.1
*
32
44
24
**
**
***
120-130
?
?
100-200
?
?
?
??
??
(10-30)
(10-30)
??
?
(30)
3
*
? : :In the case that fault could be detected but its displacement could not be estimated by seismic reflection method
??: In the case that fault could not be detected by seismic reflection method but that could be detected by other methods, trench
investigation and outcrop etc.
! : The type of active fault could not be decided.
Blank: Not description
Dip: Dip-slip fault,
Strike: Strike-slip fault,
Dip & strike: Dip & strike-slip fault
*** :Confidence of decision for dip and strike- fault- High
** : the above confidence- Middle
* : the above confidence- Low
Table 1 (B) Active fault data of National Institute of Advanced Industrial Science and Technology
Active fault name
Seismic reflection method
Displacement
Displacement
at basement
at sedimentary
(m)
layer (m)
Slip
direction
Confidence
V
-
Fault characteristics
Average velocity rate
(m/103years)
Confidence
-
Fault length
(km)
Surface fault
Buried fault
Nara-bonchi-toen-kishin
150-200
50-100
0.2
12-20
fault
Uemachi-kishin fault
400-500 V
0.4
44
Ikoma-kishin fault
150
50 V
0.2-0.4
34
Rokko-kishin fault
400-500
100-150 H
1
35
(Rokko-san segment)
Rokko-kishin fault
300 0.5-1.9
20
(Hokutan segment)
Rokko-kishin fault
200 V
0.1-0.2
10
(Senzan segment)
Shizuki-kishin fault
300 V
0.1
12
Minato－Honjo
?
? V
0.1
19
-kishin
Fault
Chuo-kozosen Izumi100-150
V
0.1-0.6
18
Kongo-kishin fault
(Kongo segment)
Chuo-kozosen Izumi(4-5) V
0.8-1
30
Kongo-kishin fault
(Kitan-kaikyo segment)
Chuo-kozosen Shikoku(5) 25
kishin fault
(Naruto-kaikyo
segment)
Mikata-kishin fault
(10-30) V
0.2-1
24
Hanaore-kishin fault
(30-50)
57
(Southern areas)
Nosaka-kishin fault
?
V
0.1
32
Biwako-seigan-kishin
??
?? V
2
65
fault(Aibano segment)
Biwako-seigan-kishin
?
V
2
65
fault(Hira segment)
Turuga-kishin fault
??
?? 0.5-0.6
16
Yoro-kishin fault
1500
V
2
55
Rokko-kishin fault
(50) H
1.5
38
(Arima-Takatuki
Tectonic Line)
? : :In the case that fault could be detected but its displacement could not be estimated by seismic reflection method
??: In the case that fault could not be detected by seismic reflection method but that could be detected by other methods, trench
investigation and outcrop etc.
! : The type of active fault could not be decided.
Blank: Not description
Dip: Dip-slip fault,
Strike: Strike-slip fault,
Dip & strike: Dip & strike-slip fault
*** :Confidence of decision for dip and strike- fault- High
** : the above confidence- Middle
* : the above confidence- Low
4
Confidence
-
-
Sanbyaku fault
Obitoke fault
Takai fault
Tenri frexure
0
5
Level (m)
-200
-400
basement
-600
Profile location
-800
0
1000
2000
3000
1000
Profile location
Result of seismic reflection method
Fig.1 Results of seismic reflection method for Kyoto-Nara basin fault zone 3)
4000
Tonami plain fault zone
Gravel
Sandy mudstone
6
Depth (m)
Altitude (m)
Mudstone
Sandy mudstone
Mudstone
Faults location
Result of seismic reflection method with columnar by boring
Fig.2 Results of seismic reflection method for Tonami west plain fault zone 4)
3. Examination results
(1) Methods of geophysical prospecting
The following four methods in the geophysical prospecting were used to each fault of 55.
- Seismic reflection method (95 %)
- Gravity method
(21 %)
- Electric method
(19 %)
- Electromagnetic method
( 2 %)
The value in ( ) is the percentage of the use frequency. The percentage of the seismic reflection method is
95 % and that is the highest frequency.
(2) Detection accuracy of active fault by reflection shooting method
In order to examine the accuracy of the seismic reflection method for detecting active faults, the fault data
in Table 1 were used. These faults were divided into the following three cases.
- Case 1: In the case that fault can be detected and its displacement is also estimated by seismic
reflection method,
- Case 2: In the case that fault can be detected but its displacement can not be estimated by seismic
reflection method and
- Case 3: In the case that fault can not be detected by seismic reflection method but that can be
detected by other methods, trench investigation and outcrop etc.
The numbers of faults for each case are shown in the Table 2.
Case 1
Table 2 Breakdown of Active Fault Data
Type 1:
Type 2:
Type 3:
Dip-slip fault Dip & Strike-slip fault Strike-slip fault
30
0
4
Case 2
9
1
3
Case 3
2
1
0
Total
41
2
( ): The type of active fault is not described.
< >: The type of active fault can not be decided.
7
Total
34 (3)<1>
13
3 (1)
50 (4)<1>
In this table 2, the total number of faults is 55. The number of case 1 & 2 that are detected by seismic
reflection method is 51. The percentage of the number of Case 1 & 2 to the total number is about 90 %. The
active faults that could not be detected by seismic reflection method are 10 %. The reasons that could not be
detected depend on the following characteristics of the seismic reflection method.
- The application of seismic reflection method to the strike-slip faults is not suitable.
- Since the decision of the fault segmentation is difficult, the profile location predicted by seismic
reflection method is not always suitable.
7
(3) Tendency of dip and strike-slip faults detected by seismic reflection method
The tendency of the dip and strike-slip faults detected by reflection shooting method is shown in Table 2.
The faults were divided into the following three types;
- Type 1: Fault of dip-slip component
- Type 2: Fault of both dip and strike-slip components
- Type 3: Fault of strike-slip component
The numbers of Type 1, 2 and 3 are 41, 2 and 7, respectively. The numbers of Type 1 and 2 including
dip-slip component are 43 and the percentage of these fault to total fault is 86 %. The percentage of Type 3;
strike-slip fault, is 14 %. The detection percentage of fault of dip-slip component was larger than that of
strike-slip component by about six times.
(4) Histograms of displacement at sedimentary layer or basement detected by seismic reflection method
The histograms of displacement at the sedimentary layer or basement detected by seismic reflection
method are shown in Fig.3. The histogram at the sedimentary layer is the uniform distribution from 100 to
500 m except below the 100 m. On the other hand, the histogram at the basement is the concentrated
distribution from 100 to 200 m. and before and behind 500 m.
(5) Relationship between displacement at sedimentary layer or basement detected by seismic reflection
method and average slip rate estimated by other geological investigation methods
The relationship between the displacement at sedimentary layer or basement detected by seismic
reflection method, and the average slip rate estimated by other geological investigation methods are shown
in Fig.4. From Fig.4, there can be seen a clear correlation between the displacement at the basement and the
average slip rate, but no correlation between the displacement at the sedimentary layer and that.
(6) Relationship between displacement at sedimentary layer or basement detected by seismic reflection
method, and fault length estimated by other geological investigation methods
In the same way as (5), the relationship between the displacement at sedimentary layer or basement
detected by seismic reflection method, and the fault length estimated by other geological investigation
methods are shown in Fig.5. From Fig.5, the fault length at both sedimentary layer and basement is 10 to
100 km. There can be seen no correlation between the displacement at the sedimentary layer and the fault
length. However we can see a tendency of correlation between that of basement and that. The dotted line in
Fig.5 represents the correlation. The magnitude was calculated by the following Matsuda equation;
logＬ=0.6Ｍ-2.9 (M:6.2-8.4),
where L is fault length and M is magnitude.
The chain line in Fig.5 represents the resolving power limit of seismic reflection method and its value is
about 25 m. The magnitude of intersecting point of dotted and chain lines is about 6.1. The magnitude 6
corresponds about the fault length of 6 km. It may well be predicted that the active fault of fault length over
6 km will be detected by seismic reflection method.
8
10
: National Institute of Advanced Industrial Science
and Technology (AIST)
Numbers 断層数
of faults
8
: Headquarters for Earthquake Research Promotion
(HERP)
6
産総研
推本
4
2
0
0-49
50-99
100-149
150-199
250-299
200-249
300-349
350-399
400-449
450-499
500～
堆積層変位量(m )
Displacement at sedimentary layer (m)
5
産総研
: AIST
Numbers断層数
of faults
4
推本
: HERP
3
2
1
0
0-49
50-99
100-149
150-199
250-299
200-249
300-349
350-399
400-449
450-499
500～
基盤変位量(m )
Displacement at basement (m)
Fig.3 Histogram of displacement at sedimentary layer or basement detected
by seismic reflection method
9
Displacement
at sedimentary layer (m)
堆積層変位量(m)
10000
: National Institute of Advanced Industrial Science
and Technology (AIST)
: Headquarters for Earthquake Research Promotion
(HERP)
1000
推本
産総研
100
A class : 1-10 (m/103 years)
B class : 0.1-1 (m/103 years)
C class : 0.01-0.1 (m/103 years)
25
25
B class
C class
10
0.01
A class
0.1
1
平均変位速度（m/千年）
Average
slip rate (m/103 years)
10
10000
Displacement
at basement (m)
基盤変位量(m)
: AIST
: HERP
1000
A class : 1-10 (m/103 years)
B class : 0.1-1 (m/103 years)
C class : 0.01-0.1 (m/103 years)
推本
産総研
100
25
10
0.01
C class
B class
0.1
1
平均変位速度(m/千年)
Average
slip rate (m/103 years)
25
A class
10
Fig.4 Relationship between displacement at sedimentary layer or basement detected by seismic
reflection method, and average slip rate estimated by other geological investigation methods
10
Displacement 堆積変位量(m
at sedimentary
) layer (m)
10000
:
National Institute of Advanced Industrial
Science and Technology (AIST)
：Headquarters for Earthquake Research
Promotion
(HERP)
1000
推本
産総研
100
25
25
10
1
10
100
1000
断層の長さ
)
Fault
length (m
(m)
基盤変位量(m)
Displacement
at basement (m)
10000
: AIST
: HERP
1000
推本
産総研
100
25
25
10
1
10
M6.1
100
1000
Fault
length (m)
断層の長さ(m)
M7.0
M8.0
M : Magnitude estimated by using Matsuda Equation
Fig.5 Relationship between displacement at sedimentary layer or basement detected by seismic
reflection method, and fault length estimated by other geological investigation methods
11
4. Summarization and future plan
(1) Summarization
The main examination results are summarized as follows.
- When the seismic reflection method was applied suitably against the active faults under sedimentary
layer, it was demonstrated that the possibility of their detection was quite large and the detection
percentage was about 90 %.
- The percentage of the dip-slip fault to the total fault is 86 % and that for the strike-slip fault is 14 %.
The detection percentage of the former fault was larger than that of the latter by about six times.
- It may well be that the active fault over about fault length of 6 km will be detected by seismic
reflection method.
(2) Future plan
In order to obtain high quality data concerning geophysical prospecting, the main issues are as follows:
- To accumulate the related geophysical prospecting data with the deep boring data,
- To examine the detection accuracy of buried faults using the various geophysical prospecting data.
In order to realize these plans, it is important that the deep boring data are accumulated.
References
1) Headquarters for Earthquake Research Promotion: Investigation report of active faults based on
earthquake investigation research grant from 1995 to 1999 fiscal year. *
2) National Institute of Advanced Industrial Science and Technology: Overview report of active fault
research survey from 1996 to 2000. *
3) Koji Okumura, et al: Total investigation of Nara basin east fault zone, Overview report of active fault
research survey on Heisei 8 fiscal year, Geological Survey of Japan, 1997. *
4) Toyama prefecture: Survey report of Tonami west plain fault zone based on a basic survey grant related
earthquake on Heisei 10 fiscal year, 1999. *
*: In Japanese
12