Interpretation of Seismic Risk

Assessment Data

United Nations Development Program

July, 12-14, 2010
Aqaba
The Dead Sea Transform (DST) fault system, also sometimes
referred to as the Dead Sea Rift, is a geologic fault which extends
through the Jordan River Valley in the Middle East. It runs along the
boundary of two tectonic plates, the African Plate on the west and
the Arabian Plate on the east. It is a left lateral transform fault,
signifying the relative motions of the two plates. Both plates are
moving in a general north-northeast direction, but the Arabian Plate
is moving faster, resulting in the observed left lateral motions along
the fault of approximately 107 km. A component of extension is also
present, which has contributed to the depression, or pull apart basin
.in which the Dead Sea is situated

The Dead Sea Transform runs from the northern end of the
Red Sea Rift just offshore of the southern tip of the Sinai Peninsula,
.Turkey to a junction with the East Anatolian Fault in southeastern
A transform fault or transform boundary, also known as conservative plate
boundary, is a fault which runs along the boundary of a tectonic plate. The
relative motion of such plates is horizontal in either sinistral or dextral direction.
Typically, some vertical motion may also exist, but the principal vectors in a
transform fault are oriented horizontally. Not all faults are transform faults, and
not all plate boundaries are transform faults. Most transform faults are found on
the ocean floor, where they often offset active spreading ridges to form a zigzag
.plate boundary. However, the best-known transform faults are found on land
Transform faults are one of the three types of plate boundaries in plate tectonics.
Transform faults occur where plates slide past each other, and crust is neither
destroyed nor created. Divergent faults occur where magma seeps up through the
earth's crust, and new crust is formed as the plates are pushed away from each
other. Convergent faults occur where two plates collide and one plate is forced
under the other plate (in a process known as subduction or obduction), and as a
consequence, the plate being forced under is melted and destroyed. When two
continental plates converge, they may push up against each other (in a process
known as continental collision) forming mountain ranges, however, subduction
.may still occur
SESAME seismic source zones (351 sources)

Jimenez, Giardini and Grünthal, 2001

PGA map of the Israeli
seismic code SI 413 (1995,
the 2004 amendment)
 REDUCTION OF EARTHQUAKE LOSSES IN
THE EXTENDED MEDITERRANEAN REGION
(RELEMR)

Cooperation in Seismic Hazard Mapping

of the Dead Sea Rift Region

Barcelona, Spain
December, 2006
Seismic Source model for the Levant region

Seismic Source
Model for the
Levant region
Two categories of soil condition have been
considered:

•Rock (Vs > 750m/s) in the Ambraseys et al. (1996)

•Generic Rock (Vs = 620m/s) in the Boore et al.
(1997) predictive models
Seismic hazard map for the
Levant region using
Ambraseys et al. (1996)
peak ground acceleration
attenuation relationship.
PGA is assessed for a 10%
probability of exceedance
in an exposure time of 50
years for rock.
Seismic hazard map for the
Levant region using Boore
et al. (1997) peak ground
acceleration attenuation
relationship. PGA is
assessed for a 10%
probability of exceedance in
an exposure time of 50 years
for generic rock.
Assessment of Seismic Hazard in Jordan

Maria Jimenez
Institute of Earth Sciences “Jaume Almera” (CSIC)
Barcelona, Spain

Hanan Al-Nimry, Adnan Khasawneh, Tareq Al-Hadid, Khaled Kahhaleh

Royal Scientific Society (RSS)
Amman, Jordan
 Definition and Characterization of discrete
homogeneous seismic sources: 18 sources
 Characterization of Seismic Sources: Magnitude-
frequency relation

 Attenuation Relationship: relate expected ground

motion at a site with the magnitude and distance of
the earthquake. The uncertainty of the ground
motion values about the predicted mean is
considered. Ambraseys et al. (1996) for both PGA
and spectral acceleration ordinates
 Determine the probabilities that any acceleration
value is exceeded within the design life of a
structure by integration over all seismic sources
(EQ occurrences follow a Poisson distribution).
SEISRISK III (Bender and Perkins, 1987)
Subjective process
 Geologic and tectonic studies (e.g. location of
the fault is clearly identified and earthquake
activity can be associated with it)
 Seismic events:
- Historical records (1-1899 A.D.)
- Instrumental records (1900-1999 A.D.)

Within a specific source, it is assumed that

earthquakes can occur at any point with
equal probability.
 Yamoune-Roum
Cyprus

Palmera

SE-
ME 3

SE-
ME 1
Farah
Haifa DS-RJ (20)
SE- (20)
ME 2 Sirhan Fault (30)
NE Gaza
(30) Wadi Karak Faiha (20)
Central Sinai Wadi
Araba (20)

SE Maan (30)
Gulf of
Gulf of Aqaba
Suez-North

East of Gulf of Aqaba (40)

Gulf of
Suez-South
b-constant of the Gutenberg-Richter relationship

Log10 N = a- bML
where
N: number of earthquakes per year in the zone with
magnitude larger or equal to ML
a, b: constants that characterize the area.

Mmax: maximum EQ which could physically be expected to occur

(rupture length).
It is the upper limit of validity of the magnitude-frequency relation.

Annual rate of seismicity 4: seismic activity within a certain

period for earthquake magnitude greater than or equal to
4.0.
4 Mmax b-value Model Source
0.33±0.04 7.5±0.6 0.75±0.06 narrow area Dead Sea-
source, 20 km Jordan River
width
0.11±0.05 6.6±0.5 0.82±0.01 narrow area Wadi Araba
source, 20 km
width
1.47 8.0 0.92 Area Source Yamoune-
Roum
0.12 6.0 0.96 Area Source Palmera

1.51±0.3 6.5±0.5 0.85±0.01 Area Source Gulf of

Aqaba
4 Mmax b-value Model Source
0.54 7.0 1.07 Area Source Gulf of
Suez-South
0.19 7.0 0.8 Area Source Gulf of
Suez-North
0.05±0.01 7.0±0.8 0.71±0.05 narrow area source, Sirhan Faults
30 km width
0.09±0.04 5.8±0.4 0.86±0.07 narrow area source, Farah Haifa
20 km width Fault
2.74 8.0 0.98 Area Source Cyprus Fault

0.023±0.01 4.7±0.34 0.44±0.12 narrow area source, Wadi Karak-

20 km width Faiha Fault
4 Mmax b-value Model Source
1.75 5.8 0.8 Area Source SE-MT 1
0.49 5.8 1.05 Area Source SE-MT 2
0.09 7.5 0.92 Area Source SE-MT 3
0.029±0.01 4.6±0.28 0.29±0.12 narrow area source, SE Maan
30 km width
0.054±0.01 5.9±0.5 0.40±0.08 narrow area source, East of Gulf
40 km width of Aqaba
0.010±0.007 4.0±0.40 0.30±0.10 narrow area source, Central
40 km width Sinai
0.022±0.01 4.5±0.24 0.34±0.13 narrow area source, North East
30 km width Gaza
PGA Peak ground acceleration in terms of g.
SA Spectral acceleration in terms of g.
MS Surface-wave magnitude

d Shortest distance from the site to the surface projection of the fault rupture in km.
h Focal depth of earthquake.
log Standard deviation.

and r2 = d2+h2
PGA (peak acceleration) is what is experienced by a particle
on the ground. SA (spectral acceleration) is approximately
what is experienced by a building, as modeled by a particle on
a massless vertical rod having the same natural period of
vibration as the building.
PGA is a good index to hazard for short buildings, up to about 7 stories. To be a
good index, means that if you plot some measure of demand placed on a
building, like interstory displacement or base shear, against PGA, for a number of
different buildings for a number of different earthquakes, you will get a strong
correlation.

PGA is a natural simple design parameter since it can be related to a force and for
simple design one can design a building to resist a certain horizontal force.

PGV, peak ground velocity, is a good index to hazard to taller buildings.

However, it is not clear how to relate velocity to force in order to design a taller
building.

SA would also be a good index to hazard to buildings, but ought to be more

closely related to the building behavior than peak ground motion parameters. I
presume design might also be easier, but the relation to design force is likely to
be more complicated than with PGA, because the value of the period comes into
the picture.
 Ambraseys attenuation relationships are
given for MS.
 Regression has been performed on the
magnitude values from the Jordan
catalogue for all events in both the
historical and instrumental parts.
 Source Model:
* Latitude: 27oN and 36oN
* Longitude: 30.5oE and 38.0oE

 Area Covered:
* Latitude: 28oN and 34oN
* Longitude: 33.75oE and 40.0oE

 Grid Interval 0.1o, approx. 9 km

 No. of computation nodes: 3355

PGA (g)
Rock
CHANCE of EXCEEDENCE (Y) %

The percent (%) chance that a certain amount of mapped

shaking distribution will occur over the time period being
.considered

.Typically the values of 2%, 5% and 10% are used

Keep in mind that a 5% chance of exceedence means there

is a 95% chance that the shaking will NOT exceed the
.value
0.2s SA (g)
Rock
0.3s SA (g)
Rock
0.5s SA (g)
Rock
1.0s SA (g)
Rock
2.0s SA (g)
Rock
 34

33.5

33

32.5

32

31.5
Lat

31

30.5

30

29.5

29

28.5
34.5 35 35.5 36 36.5 37 37.5 38 38.5 39 39.5 40

Long
 34

33.5

33

32.5

32

31.5
Lat

31

30.5

30

29.5

29

28.5
34.5 35 35.5 36 36.5 37 37.5 38 38.5 39 39.5 40

Long
 0.1
Z= 0.093/2.5 = 0.0372
0.09

0.08

0.07
Avg Spectra of all points
UBC Spectra
0.06
Avg Spectra of 28 selected points
Avg plus sigma of selected points
SA

0.05
Avg plus sigma of all points
0.04

0.03

0.02

0.01

0
0 0.5 1 1.5 2 2.5
Period (sec)
 0.15

Z = 0.138/2.5 = 0.0552

0.1
Avg Spectra of all points
UBC Spectra
Avg plus sigma of all points
SA

0.05

0
0 0.5 1 1.5 2 2.5
Period (sec)
 0.25

Z = 0.21/2.5 = 0.084

0.2

Avg Spectra of all points

0.15
UBC Spectra

Avg plus sigma of all points

SA

0.1

0.05

0
0 0.5 1 1.5 2 2.5
Period (sec)
 0.3

Z = 0.25/2.5 = 0.1

0.25

Avg Spectra of all points

0.2
UBC Spectra
Avg plus sigma of all points
SA

0.15

0.1

0.05

0
0 0.5 1 1.5 2 2.5
Period (sec)
 0.45

Z= 0.37/2.5 = 0.148
0.4

0.35

Avg Spectra of all points

0.3 UBC Spectra

Avg plus sigma of all points

0.25
SA

0.2

0.15

0.1

0.05

0
0 0.5 1 1.5 2 2.5
Period (sec)
 34

33.5

33

32.5
Z=0.15
32

31.5
Lat

31

30.5 Z=0.10 Z=0.06

30

29.5

29

28.5
34.5 35 35.5 36 36.5 37 37.5 38 38.5 39 39.5 40
Long
 Each of the zones is
characterized with a
seismic zone factor, Z
that describes the level
of expected seismicity
and that is directly
associated with the
effective peak ground
acceleration.