Embed Size (px)
Citation preview
REFERENCES
Bandy, W., C. Mortera-Gutierrez, J. Urrutia-Fucugauchi, and T. W. C. Hilde (1995), The subducted Rivera-Cocos plate boundary: Where is it, what is it, and what is its relationship to the Colima rift?, Geophys. Res. Lett., 22(22), 3075-3078.
DeMets, C., and D. S. Wilson (1997), Relative motions of the Pacific, Rivera, North American, and Cocos Plates since 0.78 Ma, J. Geo-phys. Res., 102(B2), 2789-2806.
Lin, J. and R. S. Stein (2004), Stress triggering in thrust and subduction earthquakes and stress interaction between the southern San Andreas and nearby thrust and strike-slip faults, J. Geophys. Res., 109, B02303.
Masterlark, T. (2003), Finite element model predictions of static deformation from dislocation sources in a a subduction zone: Sensi-tivities to homogeneous, isotropic, Poisson-solid, and half-space assumptions, J. Geophys. Res., 108(B11), 2540.
Schmitt, S. V., C. DeMets, J. Stock, O. Sánchez, B. Márquez-Azúa, and G. Reyes (2005), A geodetic study of the 22 January 2003 Tecomán, Colima, Mexico earthquake, in preparation for submission to Geophys J. Int.
Serpa, L., S. Smith, C. Katz, C. Skidmore, R. Sloan, and T. Pavlis (1992), A geophysical investigation of the southern Jalisco Block in the State of Colima, Mexico, Geofís. Int., 31(4), 475-492.
Singh, S. K., L. Ponce, and S. P. Nishenko (1985) The great Jalisco, Mexico, earthquakes of 1932: Subduction of the Rivera Plate, Bull. Seism. Soc. Am., 75(5), 1301-1313.
Yagi, Y., T. Mikumo, J. Pacheco, and G. Reyes (2005), Source rupture process of the Tecomán, Colima, Mexico earthquake of January 22, 2003, determined by joint inversion of teleseismic body wave and near-source data, Bull. Seism. Soc. Am., 94, 1795-1807 (2004)
5.3 Comparison to 1932 earthquakesThe 1932 earthquakes share attributes with the 1995 and 2003 earthquakes that suggest possible triggering relationship for earth-
quakes along this part of the trench. Singh et al. (1985) approximate the rupture areas of the 3 June (M
w ≈ 7.9) and 18 June 1932
(Mw ≈ 7.8) earthquakes from
isoseismal lines defined from damage reported by regional newspapers (Fig. 7). However, not all large earthquakes in the region trigger large adjacent ruptures—the 30 January 1973 M
w = 7.5 Colima earth-
quake did not trigger a sub-sequent rupture in the Manzanillo Trough.
Both the up-dip and down-dip limits of the 2003 rup-ture are deeper than those of the 1995 earthquake. This may suggest:
(A) A fundamental change along the strike of the trench in the depth of seis-mogenesis, possibly caused by differences in the rates and angles of Rivera and Cocos plate subduction, or
(B) The presence of low-strengh rock to greater depths, consistent with Bandy et al. (1995).
5.2 Comparison to 1995 earthquakeThe 1995 and 2003 earthquakes ruptured distinctly different parts of the subduction interface (Fig. 6) and only slightly overlap at the northwest edge of the Manzanillo Trough. Modeling of generic thrust faults indicates that the highest Coulomb stress changes occur along strike immediately at the edge of a rupture (Lin and Stein, 2004). This suggests that the 2003 Tecomán earthquake re-sulted from stress transfer from the adjacent 1995 event and its postseismic effects.
5. DISCUSSION5.1 Comparison to seismological solutionOnly one seismological slip distribution has been published (Yagi et al., 2004), based on a joint inversion of local and teleseismic data. The chief difference between the seismological and geodetic slip distributions is that the former extends to great depths. We tested the seismological slip distribution in our FEM (Fig. 5) and found that it produced large misfits of the vertical displacements, particularly at sites sensitive to deeper slip. Consequently, we contend that rup-ture did not extend as deeply as the seismological model suggests.
4. RESULTSThe GPS-derived slip distribution is located offshore from the Colima Graben (Fig. 4). The rupture is focused on the earthquake epicenter found by the local seismic network (Schmitt et al.).
Moment release: 6.3 × 1019 N•m (Mw=7.1)
Slip direction: N26°W (11° CCW of dip direction) Depth range: 12 km – 40 km Width of rupture: 80 km Maximum slip: 3.85 m at 22 km depth
3. METHODSWe use the commercially-available finite element package ABAQUS to model the region. Our 3D FEM mesh (Fig. 3) is designed to approxi-mate the geometrical characteristics of the northern Middle America subduction zone. The model configuration and material properties share many similarities to those employed by Masterlark (2003).
To perform inversions with the FEM, we construct a linear system of the 3D surface deformations that result from slip on each pair of nodes (one oceanic and one continental) on the subduction inter-face. We allow for oblique slip, though all nodes must slip in the same direction. Finally, we impose a down-dip slip constraint.
1. INTRODUCTION & TECTONIC SETTINGThe 22 January 2003 Tecomán subduction earthquake ruptured the Middle America trench subduction interface near Tecomán, Colima, Mexico (Fig. 1). With an estimated magnitude between 7.1 and 7.6, the Tecomán earthquake caused the deaths or injuries of more than 1,000 people and damaged or destroyed more than 40,000 homes and businesses, mostly in and near the inland city of Colima. This area has a history of large earthquakes (1995, 1973, and two in 1932), and the 2003 event is the second to be observed geodetically. The latest GPS results give insight into the seismic behavior of this region.
The seismotectonics of western Mexico are dominated by subduc-tion of the Rivera and Cocos plates northeastward beneath the west-ern edge of North America (Fig. 1). Subduction of the Rivera plate di-minishes rapidly to the northwest along the Mexican coastline (Fig. 1), from a maximum rate of 38 ± 4 mm/yr at the Manzanillo Trough (DeMets and Wilson, 1997). Significantly faster subduction of the Cocos plate (52 ± 3 mm/yr) at the Manzanillo Trough implies differ-ential Cocos-Rivera plate motion offshore. The transition between Rivera and Cocos plate lithosphere is poorly defined by kinematic and seafloor morphologic data and most likely consists of a broad zone of distributed earthquakes and deformation (Bandy et al., 1995) that intersect the trench near the Manzanillo Trough. The Manza-nillo Trough and adjacent southern Colima graben are slow or inac-tive (Serpa et al., 1992) extensional features with deep sediment fill (Bandy et al., 1995).
2. DATAOur GPS network consists of 31 geodetic markers that have been oc-cupied annually since 1995. The Tecomán earthquake occurred one day after we initiated our 2003 occupation. Eight GPS receivers were operating during the earthquake, half at sites located relatively close to the rupture zone. We occupied an additional 19 sites within one week of the earthquake and one site five months after the earth-quake. For sites operating during the earthquake, we determined coseismic offsets by differencing the locations from the day before the earthquake and the 22 h following the earthquake. For the other sites (hereafter called “quasi-coseismic”), we extrapolate the pre-earthquake location and difference it with the first post-earthquake position observed (Fig. 2).
Stuart Schmitt Department of Geology and Geophysics, University of Wisconsin-Madison, e-mail: [email protected]
Charles DeMets Department of Geology and Geophysics, University of Wisconsin-Madison
Joann Stock Seismological Laboratory, California Institute of Technology
Osvaldo Sánchez Instituto de Geofísica, Universidad Autonoma Nacional de México
Bertha Márquez-Azúa Departamento de Geografía y Ordenación Territorial, Universidad de Guadalajara
A GEODETIC STUDY OF THE 22 JANUARY 2003TECOMÁN, COLIMA, MEXICO EARTHQUAKE
EGG
Rivera
PacificPlate
Plate
CocosPlate
JB
NorthAmerican
Plate
CG
CTFZ
TZFZ
TME
M=7.1, 01.22.03 Tecomán EQ
-6-5-4-3-2-101234
MT
Rivera transform
110°W 108°W 106°W 104°W 102°W 100°W
16°N
18°N
20°N
22°N
24°N
GPS sites
'95 rupture
EGG
Rivera
PacificPlate
Plate
CocosPlate
JB
NorthAmerican
Plate
CG
CTFZ
TZFZ
TME
M=7.1, 01.22.03 Tecomán EQ
-6-5-4-3-2-101234
MT
Rivera transform
'95 rupture
250 km
250
km
250 km
Elastic continental uppercrust, E = 37.5 GPa
Viscoelastic mantle,E = 150 GPa
Elastic oceanic crust,E = 54.5 GPa
Viscoelastic continentallower crust, E = 37.5 GPa
earthquake
2003
2003
100
0
200
300
400
500
600
700
800
900
100
0
200
300
400
500
600
700
800
900
1996
1997
1998
1999
2000
2001
2002
2003
23 J
an.
25
Jan.
27 J
an.
29 J
an.
31 J
an.
2 Fe
b.
4 Fe
b.
East
-wes
t p
osi
tio
n (m
m)
East
-wes
t p
osi
tio
n (m
m)
CRIP
MIRASJDL
NOVIVICT
MILN
TENA
AUTA
TAPA
CGUZ
LIM2CEBOCEBOCEBO
COSACOSACOSA
SEBA
AYUT
CHAC
GUAC
PORT
UCOL
earthquake
2003
2003
100
0
200
300
400
500
600
700
800
900
100
0
200
300
400
500
600
700
800
900
1996
1997
1998
1999
2000
2001
2002
2003
23 J
an.
25
Jan.
27 J
an.
29 J
an.
31 J
an.
2 Fe
b.
4 Fe
b.
2003
2003
2003
2003
2003
2003
2003
No
rth
-so
uth
po
siti
on
(mm
)
No
r th
-so
uth
po
siti
on
(mm
)
UCOL
CRIP
MIRA
SJDL
NOVIVICT
MILN
TENA
TAPA
CGUZ
CEBOCEBOCEBOAUTA
COSACOSACOSALIM2
SEBA
AYUTGUAC
PORTCHAC
200
0
400
600
800
1000
1200
1400
1600
1800
0
400
600
800
1000
1200
1400
1600
1800
300
100
500
700
900
1100
1300
1500
1700
1900
300
200
100
500
700
900
1100
1300
1500
1700
1900
23 J
an.
25
Jan.
27 J
an.
29 J
an.
31 J
an.
2 Fe
b.
4 Fe
b.
earthquake
2003
2003
Vert
ical
po
siti
on
(mm
)
Vert
ical
po
siti
on
(mm
)
1996
1997
1998
1999
2000
2001
2003
2002
CRIP
MIRA
SJDL
NOVI
VICT
MILN
TENA
AUTA
TAPA
CGUZ
LIM2
CEBOCEBOCEBO
COSACOSACOSA
SEBA
AYUT
CHAC
GUAC
PORT
UCOL
Slip (cm)
100
50
150
200
250
300
350
400
0
106°W 105°W 104°W 103°W
18°N
19°N
20°N
21°N
CRIP
MIRA
SJDL
NOVI
COLIMELA
GUFICHAM
VICT
COOBAVAL
PURIMILN
TENA
AUTA TAPACGUZ
LIM2
CEBO
COSA
UGEOSEBA
AYUT
GUAC
CHAC
PORT
PredictedObserved
100 mm
UCOL
Quasi-coseismic
25 mm
Observed Predicted
105°W
104°W
103°W
18°N
19°N
20°N
21°N
106°W
MIRA
SJDL
VICT
NOVI
CRIP
COLICOOB
AVALCGUZ
TAPA
CEBOCOSA
0 50 100 150 200 250 300
-60
-40
-20
0
20
40
300
600
-60
-40
-20
0
20
40
300
600
Distance from trench (km)
Vert
ical
dis
pla
cem
ent (
mm
)
prediction
change in scale
UCOL
0 50 100 150 200 250 300
prediction
-60
-40
-20
0
20
40
300
600
-60
-40
-20
0
20
40
300
600
MIRA
SJDL
VICT
NOVI
CRIP
COLICOOB
AVALCGUZ
TAPA
CEBOCOSA
Distance from trench (km)
Vert
ical
dis
pla
cem
ent (
mm
) change in scale
UCOL
CRIP
MIRASJDL
NOVI
COLIMELA
GUFICHAM
VICT
COOBAVAL
PURIMILN
TENA
AUTA TAPACGUZ
LIM2
CEBO
COSA
UGEOSEBA
AYUT
GUAC
CHAC
PORT
Slip (cm)
PredictedObserved
106°W 105°W 104°W 103°W
18°N
19°N
20°N
21°N
100
50
150
200
250
300
350
400
0
UCOL
100 mm
Quasi-coseismic
0
100
200
300
400
500
600
700
Cos
eism
ic s
lip (c
m)
18°N
19°N
20°N
106°W 105°W 104°W
18°N
19°N
20°N
18°N
19°N
106°W 105°W 104°W 103°W
18°N
19°N
20°N
21°N
18 June 1932
3 June 1932 MMI > IX
MMI >
VIII1995rupture
2003rupture
1973rupture
Fig. 7. Historical seismicity.
(c)
(b)
(a)
Fig. 6. (a) 1995 Mw = 8.0 earthquake
slip distribution found with new FEM. (b) 2003 slip distribution. (c) Cumulative 1995 and 2003 slip.
(a) (b)
Fig. 5. (a) Yagi et al. (2004) slip distribution; observed and pre-dicted horizontal displacements. (b) Predicted and observed verti-cal displacements for sites inland from the rupture.
(c)
(b)
(a)
Fig. 4. (a) Best-fitting slip distribution; observed (gray and black arrows) and predicted (red arrows) horizontal displacements. (b) Ver-tical displacements; colors as in (a). (c) Predicted and observed verti-cal displacements for sites inland from rupture.
Fig. 3. Exploded view of finite element mesh employed for modeling. Values of Young’s modulus E are shown. Coseismic deformation is modeled as a static step in the FEM, so the entire model behaves elasti-cally.
Fig. 2. GPS coordinate time series used to find the 19 quasi-coseismic offsets. The linear regression used to determine preearthquake site motion is shown by the medium-thickness lines. Thick vertical lines represent the quasi-coseismic offsets. Horizontal medium-thickness lines after the earthquake illustrate how post-earthquake site position is obtained from the first available GPS solution. Time scale differs before and after the earthquake.
Fig. 1. Tectonic setting of western Mexico.
JB: Jalisco Block;CG: Colima Graben;MT: Manzanillo Trough.
