EXPECTED SEISMIC ACTION IN ALMERIA AND GRANADA CITIES (SOUTHERN SPAIN) COMBINING REGIONAL- AND

LOCAL-SCALE INFORMATIONJ. M.Gaspar-Escribano1, B. Benito1, M. Navarro2, F. Vidal3

1 ETSI Topografía, Geodesia y Cartografía, Universidad Politécnica de Madrid, Spain2 Facultad de Ciencias, Universidad de Almería, Spain

3 Andalusian Institute of Geophysics, Granada University, Spain

A hybrid approach that relates results from a regional seismic hazard assessment study (SISMOSAN Project) with local-scale site-effect characterizations is presented (Figure 1). Results of a regional-scale probabilistic seismic hazard analysis ofSouthern Spain on rock conditions are disaggregated to infer hazard controlling earthquakes for different target motions. Acollection of controlling magnitude-distance pairs and the corresponding site-specific response spectra at main capital citiesof the region is obtained. These spectra are first-order approximations to expected seismic actions required in localearthquake risk assessments and earthquake resistant design.In addition, results of independent, local-scale studies mapping predominant soil periods of Almeria and Granada (SouthernSpain) are used to show areas where period-dependent resonant effects are likely to occur. Similarly, measurements ofbuilding vibration periods available in Granada city may be also used to anticipate buildings that are more prone to undergoresonant effects. In these cases, if a local seismic risk assessment study or an earthquake-resistant structural design is tobe developed, it is recommended to use different seismic actions on sites characterized by distinct predominant periods.

SUMMARY

Figure 1. Scheme of the proposed approach

1. REGIONAL - SCALE APPROACH (SISMOSAN PROJECT)

2. REGIONAL + LOCAL - SCALE APPROACH

ALMERIA

0.000.050.100.150.200.250.300.350.400.45

0.0 0.5 1.0 1.5 2.0Period (s)

SA (g

)

RockSoil II

CADIZ

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.0 0.5 1.0 1.5 2.0Period (s)

SA (g

)

SA RockSA Soil III

CORDOBA

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.0 0.5 1.0 1.5 2.0Period (s)

SA (g

)

SA RocaSA Suelo III

GRANADA

0.000.100.200.300.400.500.600.700.800.901.001.10

0.0 0.5 1.0 1.5 2.0Period (s)

SA

(g)

SA Rock

SA Soil IV-A

HUELVA

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.0 0.5 1.0 1.5 2.0Period (s)

SA (g

) SA RockSA Soil III

JAEN

0.000.050.100.150.200.250.300.350.400.45

0.0 0.5 1.0 1.5 2.0Period (s)

SA

(g)

SA RockSA Soil III

MALAGA

0.000.050.100.150.200.250.300.350.400.450.50

0.0 0.5 1.0 1.5 2.0Period (s)

SA

(g)

SA RockSA Soil III

SEVILLE

0.000.050.100.150.200.250.300.350.400.450.50

0.0 0.5 1.0 1.5 2.0Period (s)

SA (g

)

SA RockSA Soil IV-A

AlmeríaCádiz

Córdoba

GranadaHuelva

Jaén

Málaga

Sevilla

CITY TARGET MOTION (cm/s2 PR=475yrs)

Magnitude (Mw)

Distance (km)

ALMERIA PGA= 127 cm/s2 4.5-5.0 5-10 ALMERIA SA(0.2s)= 323 cm/s2 4.5-5.0 5-10 ALMERIA SA(0.5s)= 160 cm/s2 5.0-5.5 5-10 ALMERIA SA(1.0s)= 72 cm/s2 5.5-6.0 5-10 CADIZ PGA= 91 cm/s2 4.5-5.0 5-10 CADIZ SA(0.2s)= 231 cm/s2 4.5-5.0 5-10 CADIZ SA(0.5s)= 112 cm/s2 5.0-5.5 5-10 CADIZ SA(1.0s)= 60 cm/s2 5.5-6.0 5-10 CORDOBA PGA= 69 cm/s2 4.5-5.0 5-10 CORDOBA SA(0.2s)= 177 cm/s2 4.5-5.0 5-10 CORDOBA SA(0.5s)= 90 cm/s2 5.0-5.5 5-10 CORDOBA SA(1.0s)= 47 cm/s2 5.5-6.0 15-20 GRANADA PGA= 201 cm/s2 5.0-5.5 0-5 GRANADA SA(0.2s)= 506 cm/s2 4.0-4.5 0-5 GRANADA SA(0.5s)= 260 cm/s2 5.5-6.0 5-10 GRANADA SA(1.0s)= 112 cm/s2 5.0-5.5 5-10 HUELVA PGA= 70 cm/s2 4.0-4.5 5-10 HUELVA SA(0.2s)= 176 cm/s2 4.0-4.5 5-10 HUELVA SA(0.5s)= 87 cm/s2 5.0-5.5 5-10 HUELVA SA(1.0s)= 54 cm/s2 7.5-8.0 310-315 JAEN PGA= 97 cm/s2 4.5-5.0 5-10 JAEN SA(0.2s)= 247 cm/s2 4.5-5.0 5-10 JAEN SA(0.5s)= 127 cm/s2 4.5-5.0 5-10 JAEN SA(1.0s)= 60 cm/s2 5.5-6.0 5-10 MALAGA PGA= 113 cm/s2 4.5-5.0 5-10 MALAGA SA(0.2s)= 287 cm/s2 4.5-5.0 5-10 MALAGA SA(0.5s)= 146 cm/s2 5.0-5.5 5-10 MALAGA SA(1.0s)= 68 cm/s2 6.0-6.5 30-35 SEVILLE PGA= 71 cm/s2 4.5-5.0 5-10 SEVILLE SA(0.2s)= 180 cm/s2 4.5-5.0 5-10 SEVILLE SA(0.5s)= 69 cm/s2 5.0-5.5 5-10 SEVILLE SA(1.0s)= 50 cm/s2 6.0-6.5 25-30

Specific response spectra on soil conditions (M,D,ε)

0

100

200

300

400

500

600

700

800

900

0.0 0.5 1.0 1.5 2.0

Period (s)

SA

(cm

/s2 )

Almería (4.75, 7.5, 1.01)Cádiz (4.25, 7.5, 1.26)Córdoba (4.25, 7.5, 0.8)Granada (5.25, 2.5, -0.47)Huelva (4.25, 7.5, 0.82)Jaén (4.75, 7.5, 0.55)Málaga (4.75, 7.5, 0.81)Sevilla (4.25, 7.5, 0.85)

Specific response spectra on rock conditions (M,D,ε)

0

100

200

300

400

500

0.0 0.5 1.0 1.5 2.0

Period (s)

SA

(cm

/s2 )

Almería (4.75, 7.5, 1.01)Cádiz (4.25, 7.5, 1.26)Córdoba (4.25, 7.5, 0.8)Granada (5.25, 2.5, -0.47)Huelva (4.25, 7.5, 0.82)Jaén (4.75, 7.5, 0.55)Málaga (4.75, 7.5, 0.81)Sevilla (4.25, 7.5, 0.85)

GRANADA

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 0.5 1 1.5 2Period (s)

SA (g

)

UHS Soil (g)

SRS_PGA Soil (g)

SRS_SA0.2s Soil (g)

SRS_SA0.5s Soil (g)

SRS_SA1s Soil (g)

ALMERIA

0.0

0.1

0.2

0.3

0.4

0.5

0 0.5 1 1.5 2Period (s)

SA

(g)

UHS Soil (g)

SRS_PGA Soil (g)

SRS_SA0.2s Soil (g)

SRS_SA0.5s Soil (g)

SRS_SA1s Soil (g)

Regional-scale seismic hazard analysis on rockcondition (Benito et al., 2008), following the standardzonified method with a logic tree that accounts fordifferent zoning and attenuation models (Figure 2).

Figure 2. Seismic hazard map on rock conditions (SISMOSAN Project): Expected PGA for the 475-year return period

Regional-scale geotechnical classification map(Figure 3). Soils are sorted depending on their expectedamplification: I(A) stiffest material, lowest amplificationand IV(B) softest material, highest amplification .

Figure 3. Geotechnical amplification map (SISMOSAN Project)

Regional-scale seismic hazard assessment includingsoil amplification effects (Figure 4), resulting bymultiplying maps of Figures 2 and 3.

Figure 4. Seismic hazard map including soil amplification (SISMOSAN Project): Expected PGA for the 475-year return period

The uniform hazard spectra (UHS, 475-year return period)are derived from the regional-scale hazard study at capitalcities (Figure 5). These are approximate representations of theseismic action for risk studies or engineering applications.

Figure 5. UHS at capital cities on rock and soil conditions.

Hazard-controlling earthquakes for specific target motions are derived byhazard deaggregation (Table 1). These provide probabilistic earthquakescenarios that can be adapted to each site and expected target motion.Specific response spectra consistent with these controlling events may beobtained using the weighted combination of ground-motion model utilizedfor the direct hazard calculations (Figure 6). These SRS are hazard-consistent representations of seismic action particularised for a given siteand ground-motion level. Hence, SRS constitute a more refinedrepresentation of the seismic action than UHS for a given site and targetmotion (Gaspar-Escribano et al., 2008).

Figure 6. SRS at capital cities on rock and soil conditions for target motions equalling the expected PGA for the 475-year return period at each site.

Table 1. Controlling earthquakes for target motions expected for a 475-year return period.

ACKN

OW

LED

GEM

ENTS

: Th

e SI

SMO

SAN

Pro

ject

was

fina

nced

by

the

Gov

enm

ent o

f And

alus

ia (C

onse

jería

de

Gob

erna

ción

, Jun

ta d

e An

dalu

cía)

3. ALMERIA AND GRANADA SCENARIOSSome work on urban scale is already advanced in Almeria and Granada, where microtremor measurements, geotechnicalsoundings, Vs(30) measurements and (sub-) surface structural models are available and provide predominant period maps ofurban soils (Figure 7, Navarro et al., 2001, 2004). Moreover, measurements of fundamental vibration periods of buildings areavailable in Granada (Figure 9, Navarro et al., 2004).

Figure 9. Fundamental vibration period map of Granada.

Figure 8. SRS in Almeria (top) and Granada (bottom) for different hazard-consistent target motions.

Different SRS response spectra for different sitesof the city may be adopted depending on thehazard-consistent target motion considered(Figure 8).

Potential damaging scenarios, where soilpredominant periods and building vibration periodscoincide, may be anticipated. In such cases, it isrecommended to represent the seismic action bythe SRS corresponding to the hazard-consistenttarget motion of the same vibration period.

CONCLUSIONS

REF

EREN

CES

-Ben

ito, N

avar

ro, G

aspa

r-Esc

riban

o, V

idal

, Gón

gora

, Jim

énez

Peñ

a, G

arcí

a R

odríg

uez,

Pas

tor (

2008

). S

eism

ic h

azar

d in

And

alus

ia re

gion

(Sou

th S

pain

).

SIS

MO

SA

N P

roje

ct. 1

4WC

EE,

Bei

jing

-Gas

par-E

scrib

ano,

Ben

ito, G

arcí

a-M

ayor

dom

o (2

008)

Haz

ard-

Con

sist

ent R

espo

nse

Spe

ctra

in th

e R

egio

n of

Mur

cia

(SE

Spa

in).

Bul

l. E

arth

quak

e E

ng.,

6, 1

79-1

96.

-Nav

arro

, Eno

mot

o, S

anch

ez, M

atsu

da, I

wat

ate,

Pos

adas

, Luz

ón, V

idal

, Seo

(200

1). S

urfa

ce s

oil e

ffect

s st

udy

usin

g sh

ort-p

erio

d m

icro

trem

or o

bser

vatio

ns in

A

lmer

ia C

ity, S

outh

ern

Spa

in. P

ure

App

l. G

eoph

ys.,

158,

2481

-249

7.-N

avar

ro, V

idal

, Fer

iche

, Eno

mot

o, S

anch

ez, M

atsu

da (2

004)

. Exp

ecte

d gr

ound

–RC

bui

ldin

g st

ruct

ures

reso

nanc

e ph

enom

ena

in G

rana

da C

ity (S

outh

ern

Spa

in).

13

WC

EE, V

anco

uver

, Pap

er N

o. 3

308

CO

NTA

CT:

ETSI

Top

ogra

fía, G

eode

sia

y C

arto

graf

ía,

Uni

vers

idad

Pol

itécn

ica

de M

adrid

, SPA

IN

jgas

par@

topo

graf

ia.u

pm.e

s+

34

9133

6644

1

- The transition from regional- to local-scale seismic hazard assessment isattained by designing probabilisticearthquake scenarios by means of hazarddeaggregation.

- Specific response spectra correspondingto the controlling events may be used torepresent the seismic action in local-scalestudies.

- At municipal scale, the superposition ofpredominant soil period maps andbuilding vibration period maps helpsidentifying areas where resonant effectsmay be expected. Ad hoc SRS for thesescenarios may be adopted for planning ordesign purposes.

-This approach is specially suitable forlow-to-moderate seismic areas, where thedata quality required to perform more-detailed seismic hazard assessmentanalyses is very difficult to reach.Figure 7. Predominant period maps of

Almeria (top) and Granada (bottom).

1

2

3

+ =

Regional-scale seismichazard mapping of

Andalusia

Hazard-consistent, period-dependent response

spectra

• Hazard dissagregation

• Controlling earthquakes

• Specific Response Spectra

• Local-scale weak-motion measurements

• Vs(30) measurements

• (Sub-)surfacestructural model

Urban zonation in termsof predominant soil

periods

WORST-CASE SCENARIOS IN ALMERIA AND GRANADA

Building fundamental vibration periods