Content uploaded by Martin Vallée
Author content
All content in this area was uploaded by Martin Vallée on May 11, 2019
Content may be subject to copyright.
Active deformation in Ecuador enlightened by a new
1
waveform-based catalog of earthquake focal
2
mechanisms
3
Sandro Vacaa,b*, Martin Valléea, Jean-Mathieu Nocquetc,a and Alexandra Alvaradob
4
(a) Université de Paris, Institut de physique du globe de Paris, CNRS, F-75005 Paris,
5
France
6
(b) Instituto Geofísico-Escuela Politécnica Nacional, Quito, Ecuador
7
(c) Université Côte d’Azur, IRD, CNRS, Observatoire de la Côte d’Azur, Géoazur,
8
Sophia Antipolis, France
9
*Corresponding author: svaca@igepn.edu.ec
10
Abstract
11
The recent development of a national seismic broadband network in Ecuador
12
enables us to determine a comprehensive catalog of earthquake focal mechanisms at
13
the country-scale. Using a waveform inversion technique accounting for the spatially
14
variable seismic velocity structure across the country, we provide location, depth,
15
focal mechanism and seismic moment for 282 earthquakes during the 2009-2015
16
period. Our results are consistent with source parameter determinations at the
17
global scale for the largest events, and increase the number of waveform-based focal
18
mechanism solutions by a factor of two. Our new catalog provides additional
19
constraints on the active deformation processes in Ecuador. Along the Ecuador
20
margin, we find a correlation between the focal mechanisms and the strength of
21
interseismic locking at the subduction interface derived from GPS measurements:
22
thrust earthquakes predominate in Northern Ecuador where interseismic locking is
23
high, while the low-to-moderate locking in Central and Southern Ecuador results in
24
variable fault plane orientations. Focal mechanisms for crustal earthquakes are
25
consistent with the principal axis of strain rate field derived from GPS data and with
26
the location of the main active faults. Our catalog helps to determine the earthquake
27
type to be expected in each of the seismic zones that have recently been proposed for
28
probabilistic seismic hazard assessment.
29
1. Introduction
30
The northern Andes is an area of complex tectonics due to the interaction of the
31
Nazca, South America and Caribbean plates (Pennington, 1981; Kellogg and Bonini, 1982;
32
Ego et al., 1996). The oblique convergence of the oceanic Nazca plate below the South
33
America continent (Fig. 1) is partitioned between westward slip at the subduction interface
34
and a northeastward escape of the North Andean Sliver (NAS) relative to South America
35
(e.g., Pennington, 1981, Kellogg et al., 1985, Freymueller et al., 1993, Audemard &
36
Audemard, 2002, Trenkamp et al., 2002, Nocquet et al., 2014, Mora-Paez et al., 2018, Fig. 1).
37
The NAS motion is predominantly accommodated by a large-scale regional dextral fault-
38
system (Soulas et al., 1991), starting at the southern boundary of the Caribbean plate in
39
Venezuela, running across Colombia along the foothills of the Eastern Cordillera (e.g.,
40
Taboada et al., 1998), entering into Ecuador where it crosses the Andean cordillera, before
41
finally reaching the gulf of Guayaquil (e.g., Audemard, 1993; Nocquet et al., 2014; Alvarado
42
et al., 2016; Yepes et al., 2016; Fig. 1). In Ecuador, this major fault system has been named
43
the Chingual-Cosanga-Pallatanga-Puná (CCPP) fault system, in reference to its individual
44
segments (Alvarado et al., 2016). Secondary fault systems, with significant seismic hazard
45
shown by large historical earthquakes (Beauval et al., 2010; Beauval et al., 2013), are also
46
found west of this major fault system in the inter-andean valley, and east of it in the sub-
47
andean domain (Alvarado et al., 2014; Alvarado et al., 2016; Yepes et al., 2016) (Fig. 1).
48
Along the Ecuadorian margin, elastic strain accumulation along the subduction is
49
heterogeneous. In northern Ecuador, the high interseismic locking imaged by GPS (Fig. 1) is
50
consistent with the large megathrust earthquakes observed during the XXth century [1906,
51
Mw 8.4-8.8 (Kelleher, 1972; Kanamori and McNally, 1982; Ye et al., 2016; Yoshimoto et al.,
52
2017); 1942, Mw 7.8 (Mendoza and Dewey, 1984); 1958, Mw 7.7 (Swenson and Beck, 1996);
53
1979, Mw 8.1 (Beck and Ruff, 1984); and the recent Mw 7.8 2016 Pedernales earthquake
54
(Ye et al., 2016; Nocquet et al., 2017; Yoshimoto et al., 2017)].
55
56
Fig. 1. Tectonic map of Ecuador. The Nazca plate converges obliquely with respect to the stable South
57
American plate (SOAM) at 58 mm/yr (Kendrick et al., 2003), and relatively to the North Andean Sliver
58
(NAS) at 47mm/yr (Nocquet et al., 2014). The interseismic coupling (ISC) model from Chlieh et al.
59
(2014) is shown by the colored contours. With respect to Stable South America, the NAS moves NNE-
60
ward at ~9 mm/yr along the Chingual-Cosanga-Pallatanga-Puná (CCPP) fault system (Nocquet et al.,
61
2014; Alvarado et al., 2016). The Inca Sliver is moving toward the SSE at ~5 mm/yr (Nocquet el al.,
62
2014; Villegas-Lanza et al., 2016), inducing shortening in the eastern sub-Andean belt. The Grijalva
63
fracture separates two domains of the Nazca plate with different ages and densities (Lonsdale, 2005).
64
Faults: SLL: San Lorenzo lineament; EF: Esmeraldas Fault; EAF: El Angel Fault; JiF: Jipijapa Fault; Py:
65
Pisayambo zone; QFS: Quito active Fault System. Cities: E: Esmeraldas; B: Bahía; M: Manta; G:
66
Guayaquil; Q: Quito; L: Latacunga; C: Cuenca; R: Riobamba.
67
Between latitudes 0.8S and 1.5S, the average coupling is low and only a small, shallow area
68
close to La Plata island is found to be locked (Vallée et al., 2013; Chlieh et al., 2014; Collot et
69
al., 2017). South of this area, the GPS data do not detect any significant coupling (Nocquet
70
et al., 2014; Villegas-Lanza et al., 2016).
71
Aside from the large subduction earthquakes, destructive events mostly occurred
72
along or close to the CCPP (Beauval et al., 2010; Beauval et al., 2013; Alvarado et al., 2016;
73
Yepes et al., 2016). Large crustal events are expected to have long recurrence intervals
74
(Baize et al., 2015), and as a consequence, historical events cannot fully characterize the
75
type and locations of potential future earthquakes. An approach complementary to the
76
historical earthquake catalog is to determine the rupture mechanism of small and moderate
77
earthquakes. A preliminary attempt to characterize the seismogenic zones in Ecuador was
78
made by Bonilla et al. (1992) who determined the spatial distribution of the active fault
79
systems, using the earthquake depths and faulting styles provided by the focal mechanisms
80
solutions. More recently, Yepes et al. (2016) proposed a new classification for the seismic
81
source zones (SSZs) for subduction interface, intraslab and crustal events. Their
82
classification takes into account focal mechanisms from the Global centroid moment tensor
83
(GCMT) catalog (Dziewonski et al., 1981; Ekström et al., 2012), geological and geophysical
84
information (tectonic and structural features of major faults, geodesy and paleoseismology).
85
In total 19 SSZs have been characterized corresponding to the shallow subduction interface
86
(3), intraslab (6), crustal (9), and outer rise (1) zones. Each of the SSZ is assumed to have a
87
homogeneous seismogenic potential (Yepes et al., 2016). In this study, we use the zonation
88
proposed by Yepes et al. (2016) and discuss its relations with our newly determined focal
89
mechanism catalog.
90
2. Development of the broadband seismic network and new
91
potential for source parameters determination
92
For the main part of the XXth century, Ecuador has been seismically instrumented
93
only by sensors in the vicinity of its capital Quito. In 1904, the Astronomical Observatory
94
installed there the first seismic instrument (Bosh-Omori), which was then replaced by a
95
Mainka instrument in 1928. Later, a set of Sprengnether seismometers (two horizontal and
96
one vertical components) was deployed in 1954. In 1963, the QUI station (composed of 3
97
high-gain and 3 long period instruments, both with horizontal and vertical components
98
(López, 2005)) was installed in the western part of Quito in the framework of the World-
99
Wide Standardized Seismographic Network (WWSSN). This station was moved in 1975 to
100
South-west of Quito and was maintained until the 1980’s by the “Instituto Geofísico de la
101
Escuela Politécnica Nacional” (IG-EPN).
102
The local Ecuadorian seismic network started in the 1970’s, with short period
103
seismic stations mostly deployed temporarily in order to monitor volcanic activity and
104
specific areas of the Inter-Andean-Valley (Yepes, 1982; Durand et al., 1987). The density of
105
the stations in the Andes improved after the creation of the IG-EPN in 1983, and the seismic
106
network was eventually extended to the coastal areas after 1991 (Vaca, 2006). In 2002, the
107
IRIS GSN station OTAV (close to Otavalo city) was the first permanent broadband station
108
with real time transmission installed in the country. Since 2006, the seismic network has
109
been regularly improved thanks to the efforts of IG-EPN together with the support of
110
national government agencies (SENESCYT and SENPLADES), national and international
111
partners (local governments, IRD, JICA, USAID, see Alvarado et al., 2018). The densification
112
of the broad band network in the north-western zone of Ecuador started at the end of 2008
113
with the ADN-project (Nocquet et al., 2010). Since 2011, a country-scale broadband
114
network is progressively being installed, with the final objective to cover the most
115
seismically active regions, from the coastal zone to the eastern foothills with an average ~50
116
km inter-station distance (Fig. 2).
117
Among other applications, the development of a country-wide broadband seismic
118
network now allows us to determine the earthquake source parameters by waveform
119
modeling. Since 2009, most of the events with moment magnitudes Mw > 3.5 could be
120
analyzed with the method described in the next section. Even with only a few years of data,
121
a significant information increase is expected compared to the GCMT catalog, which has a
122
magnitude threshold of about Mw 5.0. We also expect to improve the focal mechanism
123
information previously provided by IG-EPN, which was based on first arrival polarities. Our
124
final objective is to contribute to the “Ecuadorian focal mechanism catalog”, in which we will
125
also provide more reliable information about source depths and moment magnitudes.
126
Fig. 2. Seismic (broadband, green triangles) and GPS networks (orange hexagons, Mothes et al., 2013,
127
2018; Alvarado et al., 2018) in Ecuador as of December 2015. The dense arrays of seismic stations in
128
the central-northern part of the country are used for volcano monitoring. Additional GPS stations in
129
northern Peru and southern Colombia helping to define the regional kinematics are also shown.
130
3. Focal mechanism, depth and magnitude determination
131
Several similar methods exist to analyze the broadband seismic waveforms in order
132
to retrieve the source parameters [e.g. FMNEAR (Delouis, 2014); ISOLA (Zahradník et al.,
133
2008)]. Here, we use the MECAVEL method, already used in several studies of moderate
134
magnitude earthquakes (Mercier de Lepinay et al, 2011; Grandin et al., 2017). A specificity
135
of the MECAVEL method is its ability to solve for the velocity model simultaneously with the
136
searched source parameters (strike, dip, and rake of the focal mechanism, centroid location,
137
source origin time and duration, and moment magnitude). The method starts from an initial
138
solution (for origin time, hypocenter, and magnitude), here determined by IG-EPN. The
139
velocity model is parametrized by a superficial low-velocity layer above a crustal structure
140
with variable Moho depth. Crustal velocities are searched over a wide range, between
141
5.5km/s and 6.7km/s, and Moho depth can reach up to 67km. This approach is particularly
142
useful when analyzing earthquakes occurring in different tectonic environments, as it is the
143
case in Ecuador. Modeled waveforms in the tabular velocity model are computed using the
144
discrete wave number method from Bouchon (1981), and the inverse problem is solved
145
through the Neighborhood Algorithm (Sambridge, 1999). Within the MECAVEL method, the
146
three-component displacement waveforms are bandpassed between a low frequency (Fc1)
147
and a high frequency (Fc2) threshold. Fc1 is typically chosen above the low-frequency noise
148
that may affect the waveforms for a moderate earthquake and Fc2 is mostly controlled by
149
the limited accuracy of the simplified one-dimensional structure model. Fc2 must also not be
150
chosen above the earthquake corner frequency, because the earthquake time history is
151
simply modeled by a triangular source time function whose only inverted parameter is the
152
global duration. In most of the cases analyzed here, Fc1 on the order of 0.02-0.04Hz and Fc2
153
on the order of 0.05-0.07Hz are found to be suitable values. As a consequence, the source
154
duration has a real role in the inversion procedure only for large events (Mw>6.5), for which
155
it affects frequencies close to Fc2.
156
We extract all events reported with local magnitude larger than 3.8 from the
157
Ecuadorian national earthquake catalog (IG-EPN) for the period 2009-2015, and collect all
158
the available broadband seismic data in Ecuador. We then manually select the most suitable
159
waveforms, taking into account distance and azimuthal coverage and eliminating
160
components with a poor signal-to-noise ratio. For the 544 events with magnitude above 3.8,
161
326 were recorded with a quality sufficient for the waveform analysis.
162
Fig. 3. Example of a solution determined by the MECAVEL method. The map in (a) shows the inverted
163
source parameters (focal mechanism, moment magnitude Mw and depth Z) and the location of the
164
broadband seismic stations used. The red line represents the trench. The left-bottom inset provides
165
the origin time (T0), the epicentral location and the angles (strike, dip, and rake) of the conjugate
166
nodal planes and rakes. The agreement between observed (blue) and synthetic (red) waveforms is
167
shown in b) for each station and component. The stations are sorted by increasing epicentral distance
168
from top to bottom. Here, data and synthetics are filtered between 0.04Hz and 0.06Hz. We excluded
169
some components because of their poor signal-to-noise ratio in the selected frequency range.
170
We use a criterion based on the misfit between data and synthetics, azimuthal
171
coverage, and number of available stations and components, in order to ensure that only
172
reliable solutions are kept. 44 events not meeting these criteria were rejected, resulting in
173
a final catalog of 282 events. This catalog is provided as a public dataset linked to the present
174
study (Vaca et al., 2019). Rejections are mostly related to earthquakes with low magnitudes
175
located far away from the seismic network, and/or to earthquakes with an erroneous initial
176
location preventing the MECAVEL method to converge. An example of focal mechanism
177
determination for a Mw 3.8 earthquake is shown in Fig. 3.
178
As a first validation of our method, we compared our results to the Global CMT
179
solutions for the 34 events found in common during the 2009-2015 period. These
180
earthquakes have a magnitude between Mw 4.8 and Mw 7.1 (Figs. 4 and 5). The focal
181
mechanisms are very similar in almost all cases, even when compared to the full GCMT
182
solution which includes the non-double-couple components. Only one event (2014/10/20,
183
marked with a black asterisk in Fig. 4) located in the Andes close to the Ecuador-Colombia
184
border, is significantly different. This event occurred during a seismic crisis related to a
185
magmatic intrusion in the Chiles-Cerro Negro volcanic complex (IG-EPN, OSVP internal
186
reports, Ebmeier et al., 2016). In such a context, a complex mechanism reflecting the
187
superposition of volumetric changes and shear faulting (McNutt, 2005; Minson et al., 2007;
188
Shuler and Ekström, 2009) would explain the strong non-double-couple component of the
189
GCMT-solution and the difficulty to resolve this event with the double-couple MECAVEL
190
method.
191
Compared with GCMT results, no general bias is observed for the full depth range,
192
down to 200 km depth, and the average difference is 8 km (Fig. 5). This difference is due to
193
the numerous events with depths shallower than 50km, for which GCMT determines deeper
194
values than MECAVEL. This trend is likely due both to the minimum allowed depth in the
195
inversion (12km for GCMT and 3km for MECAVEL) and to slower velocity structures found
196
by MECAVEL. The comparison of magnitudes shows that those determined with the
197
MECAVEL method are slightly lower than the GCMT ones (average difference of 0.13, Fig.
198
6). On the contrary, magnitudes from IG-EPN catalog are systematically larger (average
199
difference of 0.38). Such observation should help to homogenize the magnitudes of the local
200
IG-EPN catalog, a step required to use a magnitude catalog for seismic hazard estimation
201
(e.g. Beauval et al., 2013).
202
Fig. 4. Comparison between MECAVEL (double couple) with GCMT solutions (full solution) for the
203
common events of the 2009-2015 period. The date of the earthquake occurrence is shown to the left
204
of the focal mechanisms. Depths (Z) and magnitudes (Mw) are shown to the right of the focal
205
mechanisms.
206
Fig. 5. Depth comparison for the events common to GCMT and MECAVEL (this study). Dashed lines
207
show the line along which the considered depths are equal.
208
Fig. 6. Magnitude comparison for the events common to GCMT, MECAVEL (this study) and IG-EPN
209
local catalog. a) Comparison between GCMT and MECAVEL b) Comparison between MECAVEL and
210
IG-EPN. The equation of the linear regression between the two magnitude catalogs (and the
211
associated correlation coefficient R2) is shown in the figure. This equation can be used to convert the
212
local magnitudes to moment magnitudes in order to homogenize the local catalog. In a) and b) the
213
dashed lines show the line along which the considered magnitudes are equal. “mg_IGEPN” refers to
214
the preferred magnitude reported by IG-EPN.
215
As another validation of the MECAVEL method, we show in Fig. 7 that the optimized
216
1D model is consistent with the large-scale features of the crustal thickness in Ecuador. In
217
particular, the Moho depth approaches 50 km beneath the ~150 km-wide Andes mountain
218
range (Robalino, 1977; Chambat, 1996); and as expected, the crustal thickness is thinner
219
when entering into the subandean area or into the coastal domain. Crustal thicknesses
220
obtained from Receiver Functions show Moho depths of ~53 km under the Cotopaxi volcano
221
in the central Andes (Bishop et al., 2017), and of ~50 km below OTAV station (Poveda et al.,
222
2015). We show in Fig. 7 that the latter values are consistent with the neighboring Moho
223
depths inferred from MECAVEL. Crustal depths determined in the subduction area (not
224
shown in Fig. 7) are less consistent from one earthquake to the other, which can be simply
225
understood by the fact that the one-dimensional parametrization is too simplistic in a
226
subduction context. This generally illustrates that in a structurally complex area, the
227
velocity structure determined by the MECAVEL method has to be considered as an
228
equivalent model, possibly not directly related to the actual structure.
229
230
Fig. 7. Moho depths inferred from the MECAVEL inversion results. The depth contours are
231
interpolated from the Moho depths individually determined for each earthquake (colored points).
232
The subduction area is not considered here, as the tabular model is not expected to provide
233
meaningful information in a context of 2D/3D structure complexities. Colored triangles show Moho
234
depths obtained from Receiver Functions at the two following locations: CTPXI (Cotopaxi Volcano)
235
and OTAV (IRIS GSN station close to Otavalo).
236
In Fig. 8, we finally show the 210 solutions reported by GCMT (Ekström et al., 2012)
237
for the 1976-2015 period together with the 282 solutions determined here in the 2009-
238
2015 period. We observe a general consistency of the focal mechanisms between the two
239
catalogs, in all of the seismically active areas of the country. The two catalogs complement
240
each other, with areas where information about the earthquake mechanism type is richer
241
in the MECAVEL or, on the contrary, in the GCMT catalog. In the next section, we discuss the
242
combined catalog in the light of the active deformation processes in Ecuador. This combined
243
catalog uses the MECAVEL solution for the events common with GCMT, but as shown by the
244
similarities of the 34 common solutions in Figure 4, this choice does not influence any
245
further interpretation of the focal mechanisms.
246
Fig. 8. Focal mechanisms provided (a) by the MECAVEL method (this study, 2009-2015) and (b) by
247
the GCMT double-couple solutions (1976-2015). In a) and b) the earthquake depths and the iso-
248
depths (in km) contours of the slab (Hayes et al., 2012) are color-coded with the same color scale
249
(shown at the bottom right). The thick red line with triangles represents the trench.
250
251
4. Focal mechanisms and deformation processes in Ecuador
252
For the sake of clarity, we separate the focal mechanism (FMs) according to their
253
depths and their locations along the margin or in the continental domain. Figs. 9 and 10
254
show events shallower than 35 km (used to analyze the partitioning features in Fig. 11), and
255
Fig. 12 shows the events deeper than 35 km. Although this division is somehow arbitrary, it
256
is convenient to first discuss the state of stress at the subduction interface and within the
257
overriding plate. Within the continental domain, it allows to separate the events related to
258
crustal tectonics from deep slab-related events.
259
4.1. Subduction
260
261
Overall, the location of the earthquakes studied here is in agreement with the study
262
from Font et al. (2013), who found that earthquakes during the interseismic period are
263
spatially organized into several stripes of seismicity, most of them being perpendicular to
264
the trench (Fig. 9).
265
4.1.1. Northern Ecuador
This zone hosted a large megathrust earthquakes sequence during the XXth century
266
with magnitudes Mw 7.7-8.8 (Kelleher, 1972; Beck and Ruff, 1984; Mendoza and Dewey,
267
1984; Swenson and Beck, 1996). In our catalog, this area is characterized by thrust events
268
at or close to the subduction interface (Fig. 9). An interesting spatial correlation shows up
269
with the interseismic locking models. Indeed, from latitude 0.8°N and further north, high
270
locking is found from the trench to a depth of ~30 km. Location of the focal mechanisms
271
determined here appears to outline the area of high locking, with only very few events
272
located within areas of locking higher than 60% (Fig. 9). Focal mechanisms rather correlate
273
with areas of the largest interseismic locking gradients, either downdip or laterally. This
274
observation is for instance similar to the Himalaya where the small seismicity appears to
275
delimit the downdip limit of locking, where shear stress at the interface is the largest during
276
the interseismic period (Avouac et al., 2015). This seismicity also appears to occur during
277
seismic swarms related to slow slip events as found by Vaca et al. (2018) for the Punta-
278
Galera Mompiche zone area located around lat. 0.8°N. The thrust mechanisms found in this
279
study are compatible with this interpretation, although a few shallow strike-slip
280
mechanisms reflect additional deformation within the overriding plate along the San
281
Lorenzo lineament and the Esmeraldas fault (Fig. 1).
282
Fig. 9. Combined GCMT and MECAVEL (1976-2015) shallow FMs solutions (depth shallower than 35
283
km) for the central and northern Ecuador margin. The interseismic locking model from Chlieh et al.
284
(2014) is shown by the colored contours. The thick red line with triangles represents the trench.
285
4.1.2. The Pedernales segment
The Pedernales segment, between lat. 0.7°N and 0.5°S, possibly ruptured during the
286
1906 earthquake, and hosted the 1942 Mw 7.8-7.9 and the recent 2016 Pedernales Mw 7.8
287
earthquakes (Swenson and Beck, 1996; Ye et al., 2016; Nocquet et al., 2017). Along this
288
segment, interseismic locking is confined between 10 and 30 km depth, in agreement with
289
the location of the 2016 Pedernales earthquake, whose main rupture propagated below the
290
coast between latitudes 0.4°N and 0.4°S (Nocquet et al., 2017). Our catalog, which ends in
291
2015, exhibits interesting spatial relationships with the rupture areas of the forthcoming
292
Pedernales earthquake. First, our catalog shows that very few earthquakes occurred within
293
the area of large (>1 m) co-seismic slip of the Pedernales earthquake (Fig. 9 and Nocquet et
294
al., 2017) during the years before the event. Secondly, at lat. ~0.2°S, our catalog highlights
295
a larger density of events. That area did not rupture during the 2016 earthquake but
296
experienced large and rapid localized afterslip immediately after (Rolandone et al., 2018).
297
It also hosted regular seismic swarms (Segovia, 2016) and repeating earthquakes during
298
the years before the Pedernales earthquakes, although no associated slow slip event here
299
has been geodetically found yet (Rolandone et al., 2018). The focal mechanisms found in
300
this study are also predominantly thrust, consistent with slip at the interface (Figs. 9, 10).
301
In this area, deformation therefore does not appear to be accommodated by infrequent large
302
earthquakes, but rather by numerous moderate earthquakes, seismic swarms (possibly
303
associated with aseismic slip) and afterslip.
304
305
4.1.3. The Bahía de Caráquez and La Plata Island segments
The Bahía area (Figs. 9, 10), between latitudes 0.5°S and 1°S, experienced three M
306
~7 earthquakes in 1896, 1956 and 1998 (Mw 7.1) (Segovia et al., 1999; Yepes et al., 2016).
307
In that area, our study shows mostly thrust mechanisms, compatible with interface
308
subduction earthquakes. Although Yepes et al. (2016) consider the Pedernales and Bahía
309
asperities to behave independently one from each other, the seismicity distribution does
310
not show clear patterns to support this view. The Bahía and Pedernales segments are now
311
considered as the same seismic zone (Beauval et al., 2018).
312
Between latitudes 1°S and 1.5°S, the central margin in Ecuador includes a 50 x 50
313
km2 area of high ISC (Figs. 1 and 9), around the “La Plata Island”, found to correlate with the
314
presence of a subducted oceanic relief (Collot et al., 2017). This zone marks a transition
315
between the mostly locked areas to the north and the southern Talara zone (Fig. 10) which
316
shows weak to negligible interplate locking. Episodic slow slip events, associated with
317
seismic swarms seem to release part of the slip deficit there (Vaca et al., 2009; Font et al.,
318
2013; Vallée et al., 2013; Chlieh et al, 2014; Jarrin, 2015; Segovia et al., 2018). In the central
319
margin, the mechanisms of the abundant seismicity are more diverse than in Northern
320
Ecuador (Figs. 9, 10), varying from reverse to strike-slip. The presence of Carnegie ridge
321
may be an element explaining this variability, perhaps through the influence of various
322
seamounts locally perturbing the stress field (Collot et al., 2017). Alternatively, strike-slip
323
events might be located within the slab, indicating internal deformation of the subducting
324
Carnegie ridge.
325
This area also shows outer-rise seismicity occurring within the Nazca plate, west of
326
the trench, in the Carnegie ridge domain (Figs. 9, 10). Part of the seismicity might be related
327
to the slab flexure (Collot et al., 2009), which is evidenced here by the presence of a few
328
normal mechanisms. Nevertheless, most of the earthquakes show strike-slip mechanisms
329
with planes azimuths ranging from N-S to NE-SW, like the one of 2011/11/17 (Mw 5.9
330
MECAVEL, Mw 6.0 GCMT; Figs. 4, 8, 9 and 10). Such kind of seismicity could be related to
331
two aligned ridges of ~W-E direction (with a 30km separation) and to some structures of
332
the Nazca Plate aligned N55°E, observed in the bathymetry (Michaud et al., 2006; Collot et
333
al., 2009). Because of the recurrent seismicity and the reported magnitudes, we suggest that
334
the outer-rise Carnegie ridge could be added as an additional seismic zone for future PSHA
335
models.
336
Further inland, an aligned N-S cluster with mostly reverse FMs is observed between
337
lat. 1.8°S and 1.2°S. Béthoux et al. (2011), observing a similar pattern of focal mechanisms,
338
suggest that some of them are not related to the interface but to the N-S oriented Jipijapa
339
fault (Fig. 1) (Egüez et al., 2003), especially those showing steep dips (~30°) and shallow
340
hypocenters (less than 20 km depth).
341
342
4.1.4. Southern Ecuador and northern Peru
South of the Grijalva fracture (Fig. 10), very few thrust events are observed. This can
343
be related to the very low subduction interface locking in this area (Nocquet et al., 2014;
344
Villegas-Lanza et al., 2016). The faulting mechanisms are dominantly strike-slip with a few
345
normal events (Fig. 10), in agreement with the NNE-SSW opening of the Gulf of Guayaquil
346
(Deniaud et al., 1999; Calahorrano, 2005; Witt et al., 2006) and the relative motion between
347
the NAS and the Inca Sliver (Nocquet et al., 2014).
348
Interestingly, we observe a general correlation between the level of locking and the
349
diversity of focal mechanisms. For the locked Northern segments (see Fig. 1), thrust
350
mechanisms consistent with the Nazca-NAS convergence dominate, suggesting that locking
351
at the plate interface controls the stress field both at the plate interface and within the
352
overriding margin. Oppositely, south of La Plata Island (~1.5°S) where locking is weak or
353
confined to the shallowest part of the subduction interface, thrust mechanisms show
354
variable orientations of shortening. Additionally, normal and strike-slip mechanisms,
355
consistent with known crustal faults (e.g. Bethoux et al., 2011) are frequent. Thus, the
356
heterogeneous stress field along that segment therefore appears to result from a
357
combination of crustal stress associated with slow straining of the overriding plate and
358
reduced compressional stress in the plate convergence direction.
359
Fig. 10. Joint CGMT and MECAVEL shallow FM solutions (depth shallower than 35 km). We keep the
360
conventions chosen by Yepes et al. (2016) for interface (non-colored polygons) and upper-crustal
361
(colored polygons) seismic source zones (SSZs). Faults distribution is modified from Alvarado et al.
362
(2016). The strain rate axes are calculated from GPS velocities measured within or close from each
363
SSZ (see Fig. S1 and Table S1). We exclude the strain tensor of El Angel SSZ, because too few velocities
364
are available there. The red line represents the trench.
365
4.1.5 Strain partitioning in the Ecuadorian subduction regime
In subduction contexts with oblique convergence, the motion obliquity is generally
366
not fully accommodated by slip at the plate interface (e.g. McCaffrey, 1992). In this case, a
367
forearc sliver is expected to move in a direction parallel to the trench, resulting in strike-
368
slip faulting along one or several faults within the overriding plate (e.g., Chemenda et al.,
369
2000). In addition, the trench perpendicular component of plate convergence may also be
370
partitioned between slip at the subduction interface and thrust in the back-arc domain.
371
We examine here how focal mechanisms observed in Ecuador help to constrain the
372
degree of partitioning. To do so, we first compute the angle difference between the surface-
373
projected slip vector of interplate earthquakes and the Nazca/South America convergence
374
direction. This approach neglects the small 3D component of the slip vector, and relies on
375
the assumption that motion at the subduction interface is fully characterized by
376
earthquakes (even if it can also be accommodated by aseismic processes). We select the FMs
377
(GCMT and MECAVEL) of events located at the margin with hypocentral depths less than 35
378
km, nodal planes with strikes between 0° to 45°, dip shallower than 25°, and rakes between
379
90° and 150°, as these events are expected to have sources along the interface (Fig. 11a).
380
After averaging over the earthquakes (Fig. 11 b), the angle difference between the slip
381
vector and the Nazca/South America convergence direction is found equal to 5.4° (+/-6.2°),
382
clockwise with respect to the Nazca/SOAM convergence (Fig. 11c and 11d). Although
383
marginally significant, the average direction of subduction slip vector suggests that the
384
subduction obliquity is not fully accommodated by slip at the subduction interface. Its value
385
is further consistent with the escape of the NAS with respect to SOAM.
386
Fig. 11. Partitioning evidenced from FMs (GCMT and MECAVEL) subduction interface events. (a) The
387
surface-projected slip vector of each focal mechanism is shown by blue lines, together with the
388
Nazca/SOAM convergence direction (green lines). (b) The histogram shows the angle (in degrees)
389
between Nazca/SOAM convergence direction and slip vector direction, by bins of 5 degrees. Values
390
range between -5° and 15°, resulting in an average and standard error of 5.4° and 6.2°, respectively.
391
(c) Construction of the kinematic triangle. Average azimuth of the trench and its normal (red lines),
392
Nazca/SOAM convergence direction and amplitude (green vector), and the mean slip direction
393
deduced from FMs (blue) are first reported. The additional information on the NAS/SOAM relative
394
direction (black arrow, ~50° azimuth), constrained by the purely strike-slip motion observed in the
395
Chingual area (zone 1 in Figure 10), allows us to determine the kinematic triangle. This kinematic
396
triangle is shown in (d) with the assumed velocity of Nazca/SOAM (green), and the computed
397
velocities of Nazca/NAS (blue) and NAS/SOAM (black).
398
To further quantify the amount of partitioning, we use the seismicity observed in
399
the Chingual area (zone 1 in Figure 10), where most of the FMs are purely strike-slip with
400
one of their nodal planes directed along a ~50° azimuth. Using this additional information
401
together with the amplitude of the Nazca/SOAM convergence (55.7 mm/yr as reported at
402
0°N by Kendrik et al. (2003)) allows us to determine the kinematic triangle (Fig. 11c and
403
11d). The relative motion Nazca/NAS is found equal to 49.2 mm/yr, to be compared with
404
the value of 47.5mm found by Nocquet et al. (2014). The relative motion NAS/SOAM is
405
found equal to 8.2mm/y, in agreement with the geological slip rates of 7.3±2.7 mm/yr (Ego
406
et al., 1996; Tibaldi et al., 2007) and the values between 7.5 and 9.5 mm/yr derived from
407
GPS data (Nocquet et al., 2014). If using an average azimuth of 27° for the trench in Northern
408
Ecuador, the ratio of partitioning is about 25% for the along-trench component of the
409
Nazca/South America convergence (Fig. 11c). Normal trench convergence is also
410
partitioned with 6% of the convergence being transferred to the motion of the NAS.
411
4.2. Crustal deformation
412
Using the selection of focal mechanisms shown in Fig. 10, we compare the style of
413
faulting with recent kinematic models for inland Ecuador (Alvarado et al., 2016) and the
414
seismic zonation proposed by Yepes et al. (2016). We further compare the principal axes of
415
the horizontal strain rate tensor against the focal mechanisms. The strain rate tensors
416
provided in Table S1 (Supplemental Information) are derived from the GPS velocities shown
417
in Fig. S1, using least-squares and estimating a constant velocity gradient (Aktug et al.,
418
2009) within the individual areas shown in Fig. 10.
419
4.2.1. The Chingual-Cosanga-Pallatanga-Puná Fault System (CCPP)
420
The CCPP is the main fault system accommodating the 7.5 – 9.5 mm/yr motion of
421
the NAS with respect to the stable part of the South America plate (Nocquet et al., 2014,
422
Alvarado et al., 2016). Variation in strike, slip rate and faulting styles have been used to
423
define separated segments for the seismic zonation presented in Yepes et al. (2016).
424
The Chingual seismic zone is the northern segment of CCPP (Fig. 1, marked as zone
425
1 in Fig. 10), crossing the border with Colombia. It delimits the boundary between the NAS
426
and the Amazon basin, which is assumed to be part of the stable part of the South America
427
plate, as indicated by small GPS velocity residuals (<2 mm/yr) east of the Andes. As a
428
consequence, the motion is expected to be mostly right-lateral strike-slip (Ego et al., 1995;
429
Tibaldi et al., 2007). Little seismicity is observed along this segment. The two focal
430
mechanism solutions (see zone 1 in Fig. 10) are consistent with dextral strike-slip on the
431
NE-oriented planes. Nonetheless, the faults located at the feet of the eastern Andes indicate
432
shallow dipping thrust. The strain rate tensor derived from GPS is in good agreement with
433
right-lateral shear along N30° trending faults.
434
South of the Chingual segment, the Cosanga fault system (zone 2, Fig. 10) delimits
435
the boundary between the NAS and the Sub-andean domain. Focal mechanisms show
436
reverse slip with a slight right-lateral strike-slip component along the NS nodal plane. This
437
seismic source is described as a transpressive zone (Ego et al., 1996; Alvarado et al., 2016;
438
Yepes et al., 2016). Two destructive earthquakes (Mw ~7.0), in the last 60 years (1955 and
439
1987) (Hall, 2000; Yepes et al., 2016), occurred along the northern portion of this segment.
440
Focal mechanism for the 1987, Mw 7.0 (mainshock) and Mw 5.8 (aftershock) show thrust
441
and strike-slip respectively (Kawakatsu and Proaño, 1991). The strain rate tensor is also in
442
agreement with a right-lateral transpressive regime for this segment.
443
The Pallatanga seismic source (zone 3) includes the Pallatanga fault itself and the
444
continuously active Pisayambo seismic nest (Aguilar et al., 1996; Troncoso, 2008). The fault
445
cuts diagonally the Inter-Andean-Valley across the Riobamba basin where it seems to divide
446
into several segments (Baize et al., 2015). In its southwestern part, the Pallatanga fault is a
447
right-lateral strike-slip fault (Winter et al., 1993), for which a 1300-3000 year-long
448
recurrence time of Mw ~7.5 earthquakes has been reported from a paleo-seismology study
449
(Baize et al., 2015). The last earthquake occurred in 1797 and generated the highest
450
intensities [magnitude 7.6 derived from intensities, XI MKS] reported in Ecuador (Egred,
451
2000; Beauval et al., 2010). Focal mechanisms of small magnitude earthquakes show a
452
combination of right-lateral and thrust motions. The northern part of the Pallatanga fault
453
system (Pisayambo) shows highly recurrent seismicity (Segovia and Alvarado, 2009). In
454
this area, the analysis of an Mw 5.0 earthquake in 2010, combining InSAR, seismic and field
455
observations, evidences a steeply dipping fault plane (> 50°) with right lateral displacement
456
(Champenois et al., 2017). Compressional behavior with right-lateral component is
457
indicated by the GPS derived strain rate tensor.
458
The southernmost segment of CCPP, the Puná seismic source (zone 4) is described
459
as a strike-slip structure, based on geomorphic observations in the Puná Island (Dumont et
460
al., 2005). Dumont et al. (2005) calculated a Holocene slip rate of 5.8 to 8 mm/yr which is
461
consistent with the relative motion between the NAS and Inca Sliver (Nocquet et al., 2014).
462
No large historical earthquakes have been reported for this segment. The FMs show dextral
463
mechanisms on NE oriented planes, which are consistent with the expected fault direction
464
and the predominant dextral components derived from the strain rate (Fig. 10). A small
465
group of events including a Mw 5.0 earthquake at the foot of the western Andes shows
466
reverse motion with ~EW shortening. This area behaves like a restraining bend linked to
467
non-coplanar segments of the CCPP fault system, similarly to the New Madrid seismic zone
468
(Marshak et al., 2003). In the Gulf of Guayaquil, the diversity of FMs solutions are the result
469
of the complex tectonic environment. Strike-slip motion can be interpreted as a result of
470
activity in a transpressional structures like those observed in Puná Island (Deniaud et al.,
471
1999; Fig. 10). The normal mechanism solutions are related to the drift of NAS, which
472
induces a N-S tensional regime (Witt et al., 2006).
473
4.2.2. The subandean domain
The Quaternary tectonics of the northern sub-Andean is not well known. The scarce
474
seismicity and the lack of instrumentation, before 2009 (Alvarado et al., 2018), are
475
responsible for the incomplete knowledge of the active tectonics in that zone. Toward the
476
South, thanks to specific works carried out after the 1995 Macas earthquake (Mw 7.1), the
477
tectonics of the Cutucú uplift is better understood.
478
The subandean domain is dominated by reverse faulting (Rivadeneira and Baby,
479
2004). In northern Ecuador, the Napo uplift (zone 5) is considered to result from a sub-
480
horizontal crustal decollement steepening close to the surface (Rivadeneira et al, 2004).
481
Reverse FMs show variable fault plane azimuths (Fig. 10) which can be the expression of
482
such type of structures. The strain rates are the lowest of the described zones, probably
483
because most of the deformation is absorbed by the CCPP system and faults in the NAS.
484
From the strain tensor, a dominant E-W shortening is expected for this area (Fig. 10).
485
The southeastern Cutucú seismic zone (zone 6) is the source of the 1995 Macas
486
earthquake (Mw 7.0, GCMT); it is a complex system with almost parallel thrusts and
487
decollements with a NNE trend (Bes de Berc, 2003). The complexity of the fault system
488
could be an element explaining the diversity of the observed mechanisms (reverse and
489
strike-slip, shown by both MECAVEL and GCMT solutions; see Fig. 8). However, we cannot
490
exclude the possibility that some solutions (e.g. around latitude ~2°S and longitude
491
~77.6°W) are not accurately determined, due to the absence of stations east of the
492
earthquakes. The strain rate is relatively low and shows shortening in a NW-SE direction
493
(Fig. 10), which is consistent with the existence of the Cutucú Range and its NNE strike.
494
4.2.3. Western cordillera
495
The El Angel fault (zone 8) is the southernmost expression of NNE trending
496
structures that are clearly recognizable along the western slopes of the Cordillera Central
497
in Colombia, defined as the Romeral fault system (París et al., 2000). Geomorphic
498
lineaments have right-lateral strike-slip motion (Ego et al., 1995). In 1868, a segment
499
attributed to this system ruptured twice with Mic (magnitude based on intensity
500
observations) of 6.6 and 7.2, respectively (Beauval et al., 2010). In the recent years analyzed
501
here, seismicity concentrates close to the Cerro Negro-Chiles volcanic complex, where a
502
magmatic intrusion likely started in the second semester of 2013 (Ebmeier et al., 2016). The
503
main Mw 5.6 earthquake, studied using satellite radar data (Ebmeier et al., 2016), shows a
504
predominantly right-lateral slip with a slight reverse component, in agreement with the
505
MECAVEL solution. The other FMs also show right-lateral strike-slip motion and E-W
506
shortening, but no comparison can be done here with a GPS-derived strain tensor, due to
507
the insufficient GPS coverage of the area (Fig. S1).
508
The Quito and Latacunga seismic sources (zone 9) are composed by blind reverse
509
faults, folds and flexures at the surface, delimiting a possible block separated from the NAS
510
(Alvarado et al., 2016). The Quito portion (N-S direction and ~60 km long) is a five sub-
511
segments structure, which can rupture individually or simultaneously with magnitudes
512
from 5.7 to 7.1 (Alvarado et al., 2014). The FMs in this section show ~N-S reverse planes
513
which are consistent with the shortening predicted by the strain rates derived from GPS
514
velocities. Along the Latacunga segment, we only have one solution with a N-S nodal plane
515
indicating EW shortening, in agreement with the proposed kinematic model from Lavenu et
516
al. (1995) and Alvarado et al. (2016).
517
4.3. Deep sources
518
The intermediate and deep seismicity is related to the subduction of Nazca and
519
Farallon slabs (Fig. 12). The Nazca slab, however, only hosts a weak and low magnitude
520
seismicity in Ecuador. We could solve for two intermediate-depth FMs (~100 km), in the
521
area of La Mana (Fig. 12), which both show a combination of normal and strike-slip
522
mechanisms. The small number of events prevents us to properly describe the rupture
523
characteristics of this seismic source.
524
The Farallon seismic sources located south of the extension of the Grijalva margin
525
(Loja, Morona, Loreto and Puyo in Fig. 12) exhibit recurrent seismicity along the slab, from
526
shallow (~35 km) to intermediate depths (~250 km). We generally observe that at least
527
one of the nodal planes has a strike following the slab contour. At relatively shallow depths
528
(between 40 and 90 km), from the Peru border to the Guayaquil area, the seismicity is
529
mostly strike-slip (Fig. 12). The deeper seismicity is dominated by normal events, in
530
agreement with the old deep part of the Farallon plate generating strong slab-pull forces
531
(e.g. Chen et al., 2004). A highly active seismicity cluster is related to the el Puyo seismic
532
nest, which spans over a wide range of depths from 130 to 250 km. Higher magnitude
533
earthquakes appear to occur in the deeper portion of the nest, at depths around 200 km, as
534
illustrated by the Mw 7.1 earthquake of August 2010. Another less dense and active normal-
535
faulting cluster is located in the southeastern part of our study zone (Fig. 12), and shallower
536
events (~120-160 km deep) are observed there.
537
Fig. 12. Joint CGMT and MECAVEL deep focal mechanisms (depth larger than 35 km). We follow the
538
denomination chosen by Yepes et al. (2016) for the name of the seismic sources zones and related
539
features. The Grijalva rifted margin (black dashed line) is believed to have a relevant role because it
540
separates two different slab domains with very different seismic activities (Yepes et al., 2016). The
541
focal mechanism depths and the iso-depths contours of the slab (Hayes et al., 2012) are color-coded
542
with the same scale, shown in the legend. The red thick line represents the trench.
543
5. Conclusions
544
We provide here a new catalog of earthquake focal mechanisms in Ecuador,
545
obtained by waveform modeling. Our catalog includes 282 reliable solutions of source
546
parameters for the period 2009-2015. This information includes the nodal plane angles as
547
well as depth and moment magnitude determinations. Together with the GCMT solutions,
548
our results provide new constraints on the interpretation of the tectonic processes at work
549
in Ecuador. Combined with GPS-derived strain rates, these solutions put a better control on
550
the deformation to be expected along and around the CCPP (Cosanga-Chingual-Pallatanga-
551
Puná) fault system, which delimits the eastern boundary of the North Andean Sliver. In
552
particular, the strike-slip character of the Puná fault, predicted by GPS strain rates and
553
which was not fully recognized by the large magnitude GCMT mechanisms, now appears
554
more clearly. At the Ecuador subduction zone, the focal mechanisms reflect the interseismic
555
coupling derived from GPS: thrust interface mechanisms characterize the coupled interface
556
in Northern Ecuador, while the low-to-moderate coupling in Central and Southern Ecuador
557
results in variable fault plane orientations. This suggests that in case of low locking at the
558
subduction interface, the stress field within the surrounding medium is poorly controlled
559
by the plate motion and rather reflects heterogeneous deformation within the slab or the
560
overriding crust.
561
Acknowledgments
562
We are grateful to the Secretaría Nacional de Educación Superior, Ciencia y
563
Tecnología (SENESCYT, Ecuador) for funding the PhD scholarship of the first author of this
564
study. This work has been supported by the Institut de Recherche pour le Développement
565
of France (IRD) and the Instituto Geofísico, Escuela Politécnica Nacional (IG-EPN), Quito,
566
Ecuador in the frame of the Joint International Laboratory ‘Earthquakes and Volcanoes in
567
the Northern Andes’ (grant IRD 303759/00). Fundings from the Agence Nationale de la
568
Recherche of France (grant ANR-07-BLAN-0143-01), SENESCYT (grant Fortalecimiento del
569
Instituto Geofísico), SENPLADES (grant Generación de Capacidades para la Difusión de
570
Alertas Tempranas), and collaboration with Instituto Geográfico Militar (IGM) are
571
acknowledged. We thank all these partners for their strong support to the projects of
572
geophysical instrumentation in Ecuador (national seismic and geodetic networks, and local
573
ADN and JUAN projects). The installation and maintenance of the seismic and geodetic
574
arrays would not have been possible without the help of numerous colleagues from the IRD
575
and IG-EPN. The data from the OTAV station (IU network, https://doi.org/10.7914/SN/IU),
576
retrieved at the IRIS Data Management Center, were used in this study. Numerical
577
computations were partly performed on the S-CAPAD platform, IPGP, France. Comments
578
and encouraging exchange of ideas with M. Segovia have contributed to enrich this study.
579
Important remarks made during the review process, in particular by one of the two
580
anonymous reviewers, were very helpful for this study. We finally thank the Editor (F.
581
Audemard) for his careful rereading of the manuscript.
582
Bibliography
583
Aguilar, J., Chatelain, J.-L., Guillier, B., Yepes, H., 1996. The Pisayambo, Ecuador,
584
Seismicity Nest: Towards the birth of a volcano? Geodinámica Andina, Troisième
585
symposium international sur la Géodynamique Andine, Collection Colloques et Séminaires,
586
ORSTOM édition, Paris, 126-129.
587
Aktug, B., Nocquet, J.-M., Cingöz, A. Parsons, B., Erkan, Y., England, P., 2009.
588
Deformation of western Turkey from a combination of permanent and campaign GPS data:
589
Limits to block-like behavior, J. Geophys. Res., 114, B10404.
590
https://doi.org/10.1029/2008JB006000
591
Alvarado, A., Audin, L., Nocquet, J.-M., Lagreulet, S., Segovia, M., Font, Y., Lamarque,
592
G., Yepes, H., Mothes, P., Rolandone, F., Jarrin, P., Quidelleur, X., 2014. Active tectonics in
593
Quito, Ecuador, assessed by geomorphological studies, GPS data, and crustal seismicity,
594
Tectonics, 33, 67–83. https://doi.org/10.1002/2012TC003224
595
Alvarado, A., Audin, L., Nocquet, J.-M., Jaillard, E., Mothes, P., Jarrin, P., Segovia, M.,
596
Rolandone, F., Cisneros, D., 2016. Partitioning of oblique convergence in the northern Andes
597
subduction zone: Migration history and present-day boundary of the North Andean sliver
598
in Ecuador, Tectonics, 35, 1048–1065. https://doi.org/10.1002/2016TC004117
599
Alvarado, A., Ruiz, M., Mothes, P., Yepes, H., Segovia, M., Vaca, M., Ramos, C., Enríquez,
600
W., Ponce, G., Jarrín, P., Aguilar, J., Acero, W., Vaca, S., Singaucho, J.C., Pacheco, D., Córdova,
601
A., 2018. Seismic, Volcanic, and Geodetic Networks in Ecuador: Building Capacity for
602
Monitoring and Research, Seismol. Res. Lett., 89, 432–439.
603
https://doi.org/10.1785/0220170229
604
Audemard, F. A., 1993. Néotectonique, sismotectonique et aléa sismique du Nord-
605
ouest du Vénézuéla (système de failles d'Oca-Ancón), PhD thesis, Université Montpellier II,
606
Montpellier, France.
607
Audemard, F. E., & Audemard, F. A. (2002). Structure of the Mérida Andes,
608
Venezuela: relations with the South America–Caribbean geodynamic interaction.
609
Tectonophysics, 345(1-4), 1-26.
610
Avouac, J.-P., Meng, L., Wei, S., Wang, T., Ampuero, J.-P., 2015. Lower edge of locked
611
Main Himalayan Thrust unzipped by the 2015 Gorkha earthquake, Nat. Geosci., 8, 708-711,
612
https://doi.org/10.1038/ngeo2518
613
Baize, S., Audin, L., Winter, T., Alvarado, A., Pilatasig, L., Taipe, M., Reyes, P.,
614
Kauffman, P., Yepes, H., 2015. Paleoseismology and tectonic geomorphology of the
615
Pallatanga fault (central Ecuador), a major structure of the South American crust,
616
Geomorphology, 237, 14–28. https://doi.org/10.1016/j.geomorph.2014.02.030
617
Beauval, C., Yepes, H., Bakun, W., Egred, J., Alvarado, A., Singaucho, J.-C., 2010.
618
Locations and magnitudes of historical earthquakes in the Sierra of Ecuador (1587-1996),
619
Geophys. J. Int., 181, 1613-1633. https://doi.org/10.1111/j.1365-246X.2010.04569.x
620
Beauval, C., Yepes, H., Palacios, P., Segovia, M., Alvarado, A., Font, Y., Aguilar, J.,
621
Troncoso, L., Vaca, S., 2013. An earthquake catalog for seismic hazard assessment in
622
Ecuador, Bull. Seismol. Soc. Am., 103(2A), 773–786. https://doi.org/10.1785/0120120270
623
Beauval, C., Marinière, J., Yepes, H., Audin, L., Nocquet, J.-M., Alvarado, A., Baize, S.,
624
Aguilar, J., Singaucho, J.-C., Jomard, H., 2018. A New Seismic Hazard Model for Ecuador, Bull.
625
Seismol. Soc. Am., 108(3A), 1443-1464. https://doi.org/10.1785/0120170259
626
Beck, S., Ruff, L., 1984. The rupture process of the great 1979 Colombia earthquake:
627
Evidence for the asperity model, J. Geophys. Res., 89, 9281–9291.
628
https://doi.org/10.1029/JB089iB11p09281
629
Bes de Berc, S., 2003. Tectonique de chevauchement, surrection et incision fluviatile
630
(Exemple de la zone Subandine Equatorienne, haut bassin Amazonien), Thèse de doctorat,
631
Université Toulouse III – Paul Sabatier.
632
Béthoux, N., Segovia, M., Alvarez, V., Collot, J.-Y., Charvis, P., Gailler, A., Monfret, T.,
633
2011. Seismological study of the central Ecuadorian margin: Evidence of upper plate
634
deformation, J. South Am. Earth Sci., 31, 139–152.
635
https://doi.org/10.1016/j.jsames.2010.08.001
636
Bishop, J.W., Lees, J.M., Ruiz, M.C., 2017. Receiver function stacks: initial steps for
637
seismic imaging of Cotopaxi volcano, Ecuador. AGU Fall Meet. Abstr. 31.
638
Bonilla, L., Ruiz, M., Yepes, H., 1992. Evaluation of seismic hazard in Ecuador,
639
Simposio Internacional sobre Prevención de Desastres Sísmicos, Mem. UNAM, Mexico, 118–
640
125.
641
Bouchon, M., 1981. A simple method to calculate Green’s functions for elastic layered
642
media, Bull. Seismol. Soc. Am., 71(4), 959–971.
643
Calahorrano, A., 2005. Structure de la marge du Golfe de Guayaquil (Equateur) et
644
propriétés physiques du chenal de subduction à partir de données de sismique marine
645
réflexion et réfraction, Thèse de Doctorat, Université Pierre et Marie Curie (Paris VI).
646
Chambat, F, 1996. Figure de la Terre : Gravimétrie, régime de contraintes et
647
vibrations, Thèse de Doctorat. Université Paris 7.
648
Champenois, J., Baize, S., Vallée, M., Jomard, H., Alvarado, A., Espin, P., Ekström, G.,
649
Audin, L., 2017. Evidences of surface rupture associated with a low-magnitude (Mw 5.0)
650
shallow earthquake in the Ecuadorian Andes, J. Geophys. Res., 122, 8446–8458.
651
https://doi.org/10.1002/2017JB013928
652
Chemenda, A., Lallemand, S., Bokun, A., 2000. Strain partitioning and interplate
653
friction in oblique subduction zones: Constraints provided by experimental modeling, J.
654
Geophys. Res., 105(B3), 5567–5581. https://doi.org/10.1029/1999JB900332
655
Chen, P.-F., Bina, C. R., Okal, E. A., 2004. A global survey of stress orientations in
656
subducting slabs as revealed by intermediate-depth earthquakes, Geophys. J. Int., 159(2),
657
721-733. https://doi.org/10.1111/j.1365-246X.2004.02450.x
658
Chlieh M., Mothes, P., Nocquet, J.-M., Jarrin, P., Charvis, P., Cisneros, D., Font, Y., Collot,
659
J.-Y., Villegas, J.-C., Rolandone, F., Vallée, M., Regnier, M., Segovia, M., Martin, X., Yepes, H.,
660
2014. Distribution of discrete seismic asperities and aseismic slip along the Ecuadorian
661
megathrust, Earth Planet. Sci Lett., 400, 292-301.
662
https://doi.org/10.1016/j.epsl.2014.05.027
663
Collot J-Y., Michaud, F., Alvarado, A., Marcaillou, B., Sosson, M., Ratzov, G., Migeon, S.,
664
Calahorrano, A., Pazmiño, A., 2009. Visión general de la morfología submarina del margen
665
convergente de Ecuador-Sur de Colombia: implicaciones sobre la transferencia de masa y la
666
edad de la subducción de la Cordillera de Carnegie, Geología y Geofísica Marina y Terrestre.
667
Ecuador. Spec. Pub. INOCAR-IRD, 47-74.
668
Collot, J-Y., Sanclemente, E., Nocquet, J-M., Leprêtre, A., Ribodetti, A., Jarrin, P., Chlieh,
669
M., Graindorge, D., Charvis, P., 2017. Subducted oceanic relief locks the shallow megathrust
670
in central Ecuador, J. Geophys. Res., 122, 3286–3305.
671
https://doi.org/10.1002/2016JB013849
672
Delouis, B., 2014. FMNEAR: Determination of focal mechanism and first estimate of
673
rupture directivity using near-source records and a linear distribution of point sources, Bull.
674
Seismol. Soc. Am., 104, 1479–1500. https://doi.org/10.1785/0120130151
675
Deniaud, Y., Baby, P., Basile, C., Ordoñez, M., Montenegro, G., Mascle, G., 1999.
676
Opening and tectonic and sedimentary evolution of the Gulf of Guayaquil: Neogene and
677
Quaternary fore-arc basin of the south Ecuadorian Andes, C. R. Acad. Sci. Paris, 328, 181–
678
187.
679
Dumont, J.-F., Santana, E., Vilema, W., Pedoja, K., Ordóñez, M., Cruz, M., Jiménez, N.
680
Zambrano, I., 2005. Morphological and microtectonicanalysis of Quaternary deformation
681
from Puná and Santa Clara Islands, Gulf of Guayaquil, Ecuador (South America),
682
Tectonophysics, 399(1-4), 331–350. https://doi.org/10.1016/j.tecto.2004.12.029
683
Durand, P., Massinon, B., Menechal, Y., Troiville, C., 1987. Etude sismique du site de
684
Mica-Tambo (Equateur), Commissariat à l'Energie Atomique, Laboratoire de Géophysique,
685
Rapport No.1.
686
Dziewonski, A., Chou, T., Woodhouse, J., 1981. Determination of earthquake source
687
parameters from waveform data for studies of global and regional seismicity, J. Geophys.
688
Res., 86, 2825–2852. https://doi.org/10.1029/JB086iB04p02825
689
Ebmeier, S., Elliott, J., Nocquet, J.-M., Biggs, J., Mothes, P., Jarrín, P., Yépez, M., Aguaiza,
690
S., Lundgren, P., Samsonov, S., 2016. Shallow earthquake inhibits unrest near Chiles–Cerro
691
Negro volcanoes, Ecuador–Colombian border, Earth Planet. Sci Lett., 450, 283-291.
692
https://doi.org/10.1016/j.epsl.2016.06.046
693
Ego, F., Sébrier, M., Yepes, H., 1995. Is the Cauca-Patia and Romeral fault system left
694
or right lateral? Geophys. Res. Lett., 22, 33–36. https://doi.org/10.1029/94GL02837
695
Ego, F., Sébrier, M., Lavenu, A., Yepes, H., Egüez, A., 1996. Quaternary state of stress
696
in the northern Andes and the restraining bend model for the Ecuadorian Andes,
697
Tectonophysics, 259(1), 101–116. https://doi.org/10.1016/0040-1951(95)00075-5
698
Egred, J., 2000. El Terremoto de Riobamba, Ed. Abya Yala, Quito.
699
Egüez, A., Alvarado, A., Yepes, H., Machette, M., Costa, C., Dart, R., 2003. Database and
700
Map of Quaternary faults and folds of Ecuador and its offshore regions. U.S. Geol. Survey
701
Open File Report, 03-289, 1–77.
702
Ekström, G., Nettles, M., Dziewonski, A., 2012. The global CMT project 2004–2010:
703
centroid-moment tensors for 13,017 earthquakes, Phys. Earth. Planet. Inter., 200(201), 1–
704
9. https://doi.org/10.1016/j.pepi.2012.04.002
705
Font, Y., Segovia, M., Vaca, S., Theunissen, T., 2013. Seismicity pattern along the
706
Ecuadorian subduction zone: new constraints from earthquake location in a 3D a priori
707
velocity model, Geophys. J. Int, 193, 263–286. https://doi.org/10.1093/gji/ggs083
708
Grandin, R., Vallée, M., Lacassin, R., 2017. Rupture process of the Oklahoma Mw 5.7
709
Pawnee earthquake from Sentinel-1 InSAR and seismological data, Seismol. Res. Lett., 88,
710
994-1004. https://doi.org/10.1785/0220160226
711
Freymueller, J., Kellogg, J., Vega, V., 1993. Plate Motions in the north Andean region,
712
J. Geophys. Res., 98, 21853–21863. https://doi.org/10.1029/93jb00520
713
Hall, M., 2000. Los terremotos del Ecuador del 5 de marzo de 1987 deslizamientos
714
y sus efectos socioeconómicos, Colección de Estudios de Geografía, Volumen 9, Corporación
715
Editora Nacional, Quito, Ecuador.
716
Hayes, G., Wald, D., Johnson, R., 2012. Slab1.0: a three-dimensional model of global
717
subduction zone geometries, J. Geophys. Res., 117, B01302.
718
https://doi.org/10.1029/2011JB008524
719
Jarrin, P., 2015. Modelamiento de datos GPS aplicado al estudio de la subducción en
720
el Ecuador. Reporte de Master. Escuela Politécnica Nacional, Quito-Ecuador.
721
Kanamori, H., McNally, K., 1982. Variable rupture mode of the subduction zone along
722
the Ecuador-Colombia coast, Bull. Seismol. Soc. Am., 72, 1241–1253.
723
Kawakatsu, H., Proaño, G., 1991. Focal Mechanisms of the March 6, 1987 Ecuador
724
Earthquakes, J. Phys. Earth. 39, 589-597. https://doi.org/10.4294/jpe1952.39.589
725
Kelleher, J., 1972. Rupture zones of large South American earthquakes and some
726
predictions, J. Geophys. Res., 77, 2087–2103. https://doi.org/10.1029/JB077i011p02087
727
Kellogg, J., Bonini, W., 1982. Subduction of the Caribbean Plate and Basement, uplifts
728
in the overriding South American Plate, Tectonics, 1(3), 251-276.
729
https://doi.org/10.1029/TC001i003p00251
730
Kellogg, J., Ogujiofor, I., Kansaka, D., 1985. Cenozoic tectonics of the Panama and
731
North Andes blocks. Memorias Congreso Latinoamericano de Geologia, 6, 34-49.
732
Kendrick, E., Bevis, M., Smalley, R., Brooks, B., Vargas, R., Lauría, E., Fortes, L., 2003.
733
The Nazca-South America Euler vector and its rate of change, J. S. Am. Earth Sci., 16, 125–
734
131. https://doi.org/10.1016/S0895-9811(03)00028-2
735
Lavenu, A., Winter, T., Dávila, F., 1995. A Pliocene-Quaternary compressional basin
736
in the Interandean Depression Central Ecuador, Geophys. J. Int., 121, 279–300.
737
https://doi.org/10.1111/j.1365-246X.1995.tb03527.x
738
Lonsdale, P., 2005. Creation of the Cocos and Nazca plates by fission of the Farallon
739
plate, Tectonophysics, 404, 237-264. https://doi.org/10.1016/j.tecto.2005.05.011
740
López, E., 2005. 132 años de historia del Observatorio Astronómico de Quito,
741
Observatorio Astronómico Nacional, Escuela Politécnica Nacional.
742
Marshak, S., Nelson, W., McBride, J., 2003. Phanerozoic strike-slip faulting in the
743
continental interior platform of the United States: Examples from the Laramide Orogen,
744
midcontinent, and Ancestral Rocky Mountains, Geological Society Special Publication, 210,
745
159-184. https://doi.org/10.1144/GSL.SP.2003.210.01.10
746
McCaffrey, R., 1992. Oblique plate convergence, slip vectors, and forearc
747
deformation, J. Geophys. Res., 97(B6), 8905–8915. https://doi.org/10.1029/92JB00483
748
McNutt, S., 2005. Volcano seismology, Annu. Rev. Earth Planet. Sci., 33, 461 – 491.
749
Mendoza, C., Dewey, J., 1984. Seismicity associated with the great Colombia-Ecuador
750
earthquakes of 1942, 1958, and 1979: Implications for barrier models of earthquake
751
rupture, Bull. Seismol. Soc. Am., 74(2), 577–593.
752
Mercier de Lépinay, B., Deschamps, A., Klingelhoefer, F., Mazabraud, Y., Delouis, B.,
753
Clouard, V., Hello, Y., Crozon, J., Marcaillou, B., Graindorge, D., Vallée, M., Perrot, J., Bouin, M.-
754
P., Saurel, J.‐M., Charvis, P., St‐Louis, M., 2011. The 2010 Haiti earthquake: A complex fault
755
pattern constrained by seismologic and tectonic observations, Geophys. Res. Lett., 38,
756
L22305. https://doi.org/10.1029/2011GL049799
757
Michaud, F., Collot, J.‐Y., Alvarado, A., López, E., and the scientific and technical team
758
of INOCAR, 2006. Batimetría y Relieve Continental, publication IOA‐CVM‐02‐Post, Inst.
759
Oceanogr. de la Armada del Ecuador, Guayaquil.
760
Minson, S., Dreger, D., Burgmann, R., Kanamori, H., Larson, K., 2007. Seismically and
761
geodetically determined nondouble-couple source mechanisms from the 2000 Miyakejima
762
volcanic earthquake swarm, J. Geophys. Res., 112, 10308.
763
https://doi.org/10.1029/2006JB004847
764
Mora-Páez, H., Kellogg, J., Freymueller, J., Mencin, D., Fernandes, R., Diederix, H.,
765
LaFemina, P., Cardona-Piedrahita, L., Lizarazo, S., Peláez-Gaviria, J.-R., Díaz-Mila, F.,
766
Bohórquez-Orozco, O., Giraldo-Londoño, L., Corchuelo-Cuervo, Y., 2018. Crustal
767
deformation in the northern Andes – A new GPS velocity field, J. South Am. Earth Sci., 89,
768
76–91. https://doi.org/10.1016/j.jsames.2018.11.002
769
Mothes, P., Nocquet, J.-M., Jarrin, P., 2013. Continuous GPS network operating
770
throughout Ecuador. Eos. Transactions AGU, 94 (26), 229-231.
771
https://doi.org/10.1002/2013EO260002
772
Mothes, P., Rolandone, F., Nocquet, J.-M., Jarrin, P., Alvarado, A., Ruiz, M., Cisneros, D.,
773
Mora-Páez, H., Segovia, M., 2018. Monitoring the earthquake cycle in the northern Andes
774
from the Ecuadorian cGPS network. Seismol. Res. Lett., 89, 534-541.
775
https://doi.org/10.1785/0220170243
776
Nocquet, J.-M., Yepes, H., Vallée, M., Mothes, P., Regnier, M., Segovia, S., Font, Y., Vaca,
777
S., Béthoux, N., Ramos, C., 2010. The ADN project: an integrated seismic monitoring of the
778
northern Ecuadorian subduction, EGU General Assembly 2010, Vienna, Austria, 9913.
779
Nocquet, J.-M., Villegas, J-C., Chlieh, M., Mothes, P., Rolandone, F., Jarrin, P., Cisneros,
780
D., Alvarado, A., Audin, L., Bondoux, F., Martin, X., Font, Y., Régnier, M., Vallée, M., Tran, T.,
781
Beauval, C., Maguiña, J., Martinez, W., Tavera, H., Yepes, H., 2014. Motion of continental
782
slivers and creeping subduction in the northern Andes, Nat. Geosci., 7, 287–291.
783
https://doi.org/10.1038/ngeo2099
784
Nocquet, J-M., Jarrin, P., Vallée, M., Mothes, P., Grandin, R., Rolandone, F., Delouis, B.,
785
Yepes, H., Font, Y., Fuentes, D., Régnier, M., Laurendeau, A., Cisneros, D., Hernandez, S.,
786
Sladen, A., Singaucho, J.-C., Mora, H., Gomez, J., Montes, L., Charvis, P., 2017. Supercycle at the
787
Ecuadorian subduction zone revealed after the 2016 Pedernales earthquake, Nat. Geosci.,
788
10, 145-149. https://doi.org/10.1038/NGEO2864
789
París, G., Machette, M., Dart, R., Haller, K., 2000. Map and database of Quaternary
790
faults and folds in Colombia and its offshore regions, U.S. Geol. Survey Open File Report, 00-
791
0284.
792
Pennington, W., 1981. Subduction of the Eastern Panama Basin and seismotectonics
793
of northwestern South America, J. Geophys. Res., 86.
794
https://doi.org/10.1029/JB086iB11p10753
795
Poveda, E., Monsalve, G., Vargas, C., 2015. Receiver functions and crustal structure
796
of the northwestern Andean region, Colombia, J. Geophys. Res. Solid Earth, 120, 2408–2425.
797
https://doi.org/10.1002/2014JB011304
798
Rivadeneira, M., Baby, P., 2004. Características geológicas de los principales campos
799
petroleros de Petroproducción, Travaux de l’Institut Français d’Etudes Andines, 144.
800
Rivadeneira, M., Baby, P., Barragán, R., 2004. La Cuenca Oriente: geología y petróleo,
801
Travaux de l’Institut Français d’Etudes Andines, 144.
802
Robalino, F., 1977. Espesor de la corteza en Quito mediante el análisis del espectro
803
de las ondas longitudinales P de período largo, Instituto Panamericano de Geografía e
804
Historia, XI Asamblea General, Sección Nacional del Ecuador. Quito-Ecuador.
805
Rolandone, F., Nocquet, J.-M., Mothes, P.A., Jarrin, P., Vallée, M., Cubas, N., Hernandez,
806
S., Plain, M., Vaca, S., Font, Y., 2018. Areas prone to slow slip events impede earthquake
807
rupture propagation and promote afterslip, Sci. Adv., 4, eaao6596.
808
https://doi.org/10.1126/sciadv.aao6596
809
Sambridge, M., 1999. Geophysical inversion with a neighbourhood algorithm- I.
810
Searching a parameter space, Geophys. J. Int., 138, 479-494.
811
https://doi.org/10.1046/j.1365-246X.1999.00876.x
812
Segovia, M., Pacheco, J., Shapiro, N., Yepes, H., Guillier, B., Ruiz, M., Calahorrano, A.,
813
Andrade, D., Egred. J., 1999. The Agust 4, 1998, Bahía Earthquake (Mw=7.1): Rupture
814
mechanism and comments on the potencial seismic activity, Fourth ISAG, Germany, 673-
815
677.
816
Segovia, M., Alvarado, A., 2009. Breve análisis de la sismicidad y del campo de
817
esfuerzos en el Ecuador, Geología y Geofísica Marina y Terrestre, Ecuador. Spec. Pub.
818
INOCAR-IRD, 131-149.
819
Segovia Reyes, M., 2016. Imagerie microsismique d’une asperité sismologique dans
820
la zone de subduction Équatorienne. Thèse de doctorat, Université Nice-Sophia Antipolis.
821
Segovia, M., Font, Y., Régnier, M., Charvis, P., Galve, A., Nocquet, J.‐M., et al.,
822
2018. Seismicity distribution near a subducting seamount in the Central Ecuadorian
823
subduction zone, space‐time relation to a slow‐slip event. Tectonics, 37, 2106–
824
2123. https://doi.org/10.1029/2017TC004771
825
Shuler, A., Ekström, G. 2009. Anomalous earthquakes associated with Nyiragongo
826
Volcano: Observations and potential mechanisms, J. Volc. Geotherm. Res., 181, 219-230.
827
https://doi.org/10.1016/j.jvolgeores.2009.01.011
828
Soulas, J.-P., Egüez, A., Yepes, H., Pérez, V-H., 1991. Tectónica activa y riesgo sísmico
829
en los Andes Ecuatorianos y en el extremo Sur de Colombia, Bol. Geol. Ecuatoriano, 2(1), 3–
830
11.
831
Swenson, J., Beck, S., 1996. Historical 1942 Ecuador and 1942 Peru subduction
832
earthquakes and earthquake cycles along Colombia, Ecuador and Peru subduction
833
segments, Pure Appl. Geophys., 146(1), 67–101.
834
Taboada, C., Dimaté, C., Fuenzalida, A., 1998. Sismotectónica de Colombia:
835
deformación continental activa y subducción, Física de la Tierra, 10, 111-147.
836
Tibaldi, A., Rovida, A., Corazzato, C., 2007. Late Quaternary kinematics, slip-rate and
837
segmentation of a major Cordillera-parallel transcurrent fault: The Cayambe-Afiladores-
838
Sibundoy system, NW South America, J. Struct. Geol., 29(4), 664–680.
839
https://doi.org/10.1016/j.jsg.2006.11.008
840
Trenkamp, R., Kellogg, J., Freymueller, J., Mora, H., 2002. Wide plate margin
841
deformation, southern Central America and northwestern South America, CASA GPS
842
observations, J. South Am. Earth Sci., 15, 157-171.
843
Troncoso, L., 2008. Estudio sismológico del nido de Pisayambo, Reporte de Master.
844
Escuela Politécnica Nacional, Quito-Ecuador.
845
Vaca, S., 2006. Elaboración de un modelo 3D de velocidades sísmicas de ondas para
846
el Ecuador y relocalización de eventos sísmicos entre 1995 y 2005, Tesis de ingeniería,
847
Facultad de Geología, Escuela Politécnica Nacional, Quito-Ecuador, 63-66.
848
Vaca, S., Régnier, M., Bethoux, N., Alvarez, V., Pontoise, B., 2009. Sismicidad de la
849
región de Manta: Enjambre sísmico de Manta-2005, Geología y Geofísica Marina y Terrestre,
850
Ecuador. Spec. Pub. INOCAR-IRD, 155-166.
851
Vaca, S., Vallée, M., Nocquet, J.-M., Battaglia J., Regnier, M., 2018. Recurrent Slow Slip
852
Events as a barrier to the northward rupture propagation of the 2016 Pedernales
853
earthquake (Central Ecuador), Tectonophysics, 724-725, 80-92.
854
https://doi.org/10.1016/j.tecto.2017.12.012
855
[dataset] Vaca, S., Vallée, M., Nocquet, J.-M, Alvarado, A., Seismic source parameters
856
of the 282 Ecuador earthquakes (2009-2015) determined by the MECAVEL method,
857
Mendeley data, 2019.
858
Vallée M, Nocquet, J.-M., Battaglia, J., Font, Y., Segovia, M., Régnier, M., Mothes, P.,
859
Jarrin, P., Cisneros, D., Vaca, S., Yepes, H., Martin, X., Béthoux, N., Chlieh, M., 2013. Intense
860
interfece seismiscity triggered by a shallow slow slip event in the Central Ecuador
861
subduction zone, J. Geophys. Res., 118, 1-17. https://doi.org/10.1002/jgrb.50216
862
Villegas-Lanza, J.-C., Chlieh, M., Cavalié, O., Tavera, H., Baby, P., Chire-Chira, J.,
863
Nocquet, J.-M., 2016. Active tectonics of Peru: Heterogeneous interseismic coupling along
864
the Nazca megathrust, rigid motion of the Peruvian Sliver, and Subandean shortening
865
accommodation, J. Geophys. Res., 121, 7371–7394.
866
https://doi.org/10.1002/2016JB013080
867
Winter, T., Avouac, J.-P., Lavenu, A., 1993. Late Quaternary kinematics of the
868
Pallatanga strike-slip fault (Central Ecuador) from topographic measurements of displaced
869
morphological features, Geophys. J. Int., 115, 905–920. https://doi.org/10.1111/j.1365-
870
246X.1993.tb01500.x
871
Witt, C., Bourgois, J., Michaud, F., Ordoñez, M., Jiménez, N., Sosson, M., 2006.
872
Development of the Gulf of Guayaquil (Ecuador) during the Quaternary as an effect of the
873
North Andean Block tectonic escape, Tectonics, 25, TC3017.
874
https://doi.org/10.1029/2004TC001723
875
Ye, L., Kanamori, H., Avouac, J.-P., Li, L., Cheung, K., Lay, T., 2016. The 16 April 2016,
876
Mw 7.8 (MS 7.5) Ecuador earthquake: A quasi-repeat of the 1942 MS 7.5 earthquake and
877
partial re-rupture of the 1906 MS 8.6 Colombia–Ecuador earthquake, Earth Planet. Sci. Lett.,
878
454, 248-258. https://doi.org/10.1016/j.epsl.2016.09.006
879
Yepes, H., 1982. Estudio de la actividad microsísmica en el Valle Interandino entre
880
las latitudes 0.00° y 1.00° S, Tesis Ing. Geólogo, Escuela Politécnica Nacional. Quito, 1-98.
881
Yepes, H., Audin, L., Alvarado, A., Beauval, C., Aguilar, J., Font, Y., Cotton, F., 2016. A
882
new view for the geodynamics of Ecuador: Implication in seismogenic source definition and
883
seismic hazard assessment, Tectonics, 35, 1249–1279.
884
https://doi.org/10.1002/2015TC003941
885
Yoshimoto, M., Kumagai, H., Acero, W., Ponce, G., Vásconez, F., Arrais, S., Ruiz, M.,
886
Alvarado, A., Pedraza-García, P., Dionicio, V., Chamorro, O., Maeda, Y., Nakano, M., 2017.
887
Depth-dependent rupture mode along the Ecuador-Colombia subduction zone, Geophys.
888
Res. Lett., 44, 2203–2210. https://doi.org/10.1002/2016GL071929
889
Zahradník, J., Janský, J., Plicka, V., 2008. Detailed waveform inversion for moment
890
tensors of M~4 events; examples from the Corinth Gulf, Greece, Bull. Seismol. Soc. Am., 98,
891
2756-2771. http://doi.org/10.1785/0120080124
892
Supplementary Information for ”Active deformation
893
in Ecuador enlightened by a new waveform-based
894
catalog of earthquake focal mechanisms”
895
Sandro Vacaa,b*, Martin Valléea, Jean-Mathieu Nocquetc,a and Alexandra Alvaradob
896
(a) Institut de Physique du Globe de Paris, Sorbonne Paris Cité, Université Paris
897
Diderot, UMR 7154 CNRS, Paris, France
898
(b) Instituto Geofísico-Escuela Politécnica Nacional, Quito, Ecuador
899
(c) Geoazur, IRD, Université Côte d’Azur, CNRS, Observatoire de la Côte d’Azur,
900
Valbonne, 06103, Nice Cedex 2, France
901
*Corresponding author: svaca@igepn.edu.ec
902
903
904
Contents of this file:
905
- Table S1
906
- Figure S1
907
Description of this file :
908
Table S1 provides the details of the determination of the horizontal strain rate
909
tensor. Fig. S1 shows the GPS velocity field with respect to stable South America plate.
910
Seismic
source
zone
Chingual
(1)
Cosanga
(2)
Pallatanga
(3)
Puná
(4)
Napo
(5)
Cutucu
(6)
UIO_Lat
(9)
ε1
25.50
41.7±2.9
35.0±3.2
35.3±3.0
14.1±4.0
3.1±3.1
3.5±4.8
ε2
-20.98
-60.8±4.4
-20.8±7.1
-24.3±2.8
-20.3±4.4
-30.3±5.1
-51.1±7.3
θ
107.81
95.8±1.3
71.4±3.7
88.1±2.1
84.1±5.7
123.8±4.5
113.5±4.2
GPS
Stations
CNJO
PSTO
TULC
AHUA
ELCH
HUAC
LATA
PAPA
PUYX
RIOP
CUER
LATA
RIOP
TOTO
ZHUD
CUEC
CUER
GPH1
GYEC
MACH
NARI
PROG
TU01
ZHUD
AHUA
AUCA
CNJO
ELCH
HUAC
LIMO
PUYX
CUEC
HONA
MONT
SNTI
TOTO
CULA
HSPR
LATA
MOCA
PAPA
Table S1. Results of the horizontal strain rate tensor determination for the seismic
911
source zones (we refer to the corresponding indexes in the main text and in Fig. 10). The
912
strain axes ε1 (most extensional eigenvalue of strain tensor) and ε2 (most compressional
913
eigenvalue of strain tensor) are both given with their uncertainties in nanostrain/year (10-
914
9/year). Extension is taken positive. Azimuth θ is the angle from the North of ε2 (mean value
915
and its uncertainty is given in degrees). The GPS stations used for each zone (located around
916
or inside the seismic zone) can be seen in Fig. S1.
917
Fig. S1. GPS velocity field with respect to stable South America from Nocquet et al. (2014). Error
918
ellipses are 95% confidence level. Faults are modified from Alvarado et al. (2016). The red line
919
represents the trench.
920
921
