RESEARCH LETTER Oblique Blind Faulting Underneath the Luzon Volcanic Arc

10.1029/2023GL103659
During the 2022 Mw 7.0 Abra Earthquake, the Philippines
Key Points:
Chi-Hsien Tang1 , Ya-Ju Hsu1 , Teresito Bacolcol2, Yunung Nina Lin1 , Horng-Yue Chen1 ,
• G lobal Navigation Satellite System
Yu-Ting Kuo3 , Hsuan-Han Su1, Hsin-Ming Lee1, Alfie Pelicano2 , Genesis Sapla2, and
and Sentinel-1 Aperture Radar reveal
the Abra rupture on an oblique blind Shui-Beih Yu1
fault beneath northern Luzon
• The coseismic rupture brought the Institute of Earth Sciences, Academia Sinica, Taipei, Taiwan, 2Philippine Institute of Volcanology and Seismology, Quezon
1

shallow parts (5 km depth) of adjacent City, Philippines, 3Department of Earth and Environmental Sciences, National Chung Cheng University, Chiayi, Taiwan
faults up to 2 bars closer to failure
• Inferred slip and depth distribution
of seismicity reveal the thick Abstract Unknown seismogenic structures lurking beneath convergent margins introduce substantial
seismogenic layer of 30 km on the
northern Philippine Fault Zone
uncertainty in seismic hazard assessments. In northwestern Luzon, the Mw 7.0 Abra earthquake on 27 July 2022
highlights the seismic activity along an unmapped blind fault underneath the Cordillera Central. By integrating
coseismic displacements constrained by radar satellite imagery and Global Navigation Satellite System, we
Supporting Information:
image oblique coseismic slip at 11–22 km depth with peak slip of ∼1 m beneath the Philippine Fault Zone in
Supporting Information may be found in
the online version of this article. northern Luzon. The southward propagation of coseismic slip and aftershocks terminated at a distance of 50 km
from the northern end of the 1990 Luzon earthquake rupture, leaving a seismicity gap in between. Coulomb
Correspondence to: stress changes of reaching 2 bars are imparted at the shallow portions of the Vigan-Aggao and Abra River
Y.-J. Hsu, faults, where the updated 100-year seismic potential is increased to Mw 7.0–7.7, given the thick seismogenic
[email protected] layer of ∼30 km in northern Luzon.

Citation: Plain Language Summary On 27 July 2022, the Mw 7.0 Abra earthquake hit northwestern Luzon,
Tang, C.-H., Hsu, Y.-J., Bacolcol, T., the Philippines, killing 11 people, damaging more than 30,000 buildings, and causing US$34 million worth
Lin, Y. N., Chen, H.-Y., Kuo, Y.-T., of damage. The earthquake occurred on a previously unrecognized deep fault with no observable surface
et al. (2023). Oblique blind faulting
underneath the Luzon volcanic arc
ruptures. In this study, we collected coseismic displacements from Global Navigation Satellite System and
during the 2022 Mw 7.0 Abra earthquake, Interferometric Synthetic Aperture Radar to investigate the source fault and coseismic slip distribution. Our
the Philippines. Geophysical Research model shows that the coseismic slip occurred beneath the Philippine Fault Zone in northern Luzon at a depth
Letters, 50, e2023GL103659. https://doi.
org/10.1029/2023GL103659
of 11–22 km. The coseismic slip and aftershocks propagated southward and terminated at a distance of 50 km
from the northern tip of the 1990 Luzon coseismic surface rupture. This leaves a seismicity gap between
Received 13 MAR 2023 latitudes 16.5°N–17°N along the Philippine Fault Zone. The mainshock-induced stress perturbation on the
Accepted 26 APR 2023 nearby active faults, together with the unusually thick seismogenic layer on the northern Philippine Fault Zone,
implies a high seismic hazard in Luzon. Successive crustal deformation and seismicity monitoring will provide
Author Contributions: valuable constraints for the energy budget of earthquakes across northern Luzon, where local communities are
Conceptualization: Chi-Hsien Tang, threatened by inland crustal earthquakes and megathrust events in the future.
Ya-Ju Hsu, Yunung Nina Lin, Yu-Ting Kuo
Data curation: Ya-Ju Hsu, Horng-Yue
Chen, Hsuan-Han Su, Hsin-Ming Lee,
Alfie Pelicano, Genesis Sapla, Shui- 1. Introduction
Beih Yu
Funding acquisition: Ya-Ju Hsu, On 27 July 2022 (UTC), the Mw 7.0 Abra earthquake struck the northwestern region of Luzon, the Philippines.
Teresito Bacolcol This earthquake is the largest crustal seismic event in northern Luzon since the 1990 Mw 7.7 earthquake (Silcock
Investigation: Teresito Bacolcol, Yunung
& Beavan, 2001; Yoshida & Abe, 1992). The earthquake occurred ∼20 km east of Vigan city, beneath the
Nina Lin, Horng-Yue Chen, Yu-Ting Kuo,
Hsuan-Han Su, Hsin-Ming Lee, Alfie western flank of the Cordillera Central, with a hypocenter depth of 15 km (PHIVOLCS, 2022) (Figure 1). The
Pelicano, Genesis Sapla earthquake is characterized by an oblique faulting and the focal mechanism of the mainshock exhibits two nodal
planes dipping ∼70° to the north-west and ∼30° to the east, respectively (Bonita et al., 2015; Ekström et al., 2012;
Punongbayan et al., 2015; USGS, 2022a) (Table S1 in Supporting Information S1). Strong ground shaking gener-
© 2023. The Authors. ated by the mainshock was perceived throughout Luzon island, including Manila, one of the most populous
This is an open access article under metropolises in the world, ∼300 km away from the epicenter (PHIVOLCS, 2022). The emergency field investi-
the terms of the Creative Commons
Attribution-NonCommercial-NoDerivs gation after the mainshock reported massive damages of local infrastructures, induced landslides in mountainous
License, which permits use and areas, and liquefaction in western Luzon (Perez et al., 2023). The distribution of aftershocks shows a north-south
distribution in any medium, provided the trend, roughly parallel to the strike of geological units in this region (Figure 1). However, no coseismic surface
original work is properly cited, the use is
non-commercial and no modifications or ruptures were found, leaving the seismogenic fault of the mainshock concealed. According to the focal mecha-
adaptations are made. nism (Figure 1), this event is likely associated with an oblique blind fault.

TANG ET AL. 1 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Methodology: Chi-Hsien Tang, Ya-Ju Luzon is bounded by the eastward subduction of the Sunda Plate to the west and the westward subduction of
Hsu, Yunung Nina Lin
the Philippine Sea Plate to the east with a convergent rate of ∼90 mm/yr (Hsu et al., 2016; Kreemer et al., 2014;
Project Administration: Ya-Ju Hsu,
Teresito Bacolcol, Shui-Beih Yu Sella et al., 2002). In northern Luzon, the Philippine Fault Zone (PFZ) is characterized by a series of subparallel
Resources: Ya-Ju Hsu, Teresito Bacolcol, north-south trending sinistral strike-slip branches (Pinet & Stephan, 1990; Ringenbach et al., 1990, 1993). These
Yunung Nina Lin, Horng-Yue Chen
horsetail splays accommodate an oblique shortening rate of 20–25 mm/yr with strain partitioning (Aurelio, 2000;
Software: Chi-Hsien Tang, Yunung
Nina Lin Barrier et al., 1991; Galgana et al., 2007; Hsu et al., 2016; Yu et al., 2013). Although the faults are well-mapped
Supervision: Ya-Ju Hsu on the surface, their geometry and slip behaviors are largely unknown, posing considerable uncertainties in the
Validation: Chi-Hsien Tang
seismic hazard assessment (Aurelio et al., 2017; Hsu et al., 2011; Yeats & Huftile, 1995). The 2022 Abra earth-
Visualization: Chi-Hsien Tang
Writing – original draft: Chi-Hsien quake therefore presents a rare opportunity to investigate the elusive seismogenic fault beneath northern Luzon
Tang and may help improve seismic potential estimates. While previous studies have attempted to interpret the source
Writing – review & editing: Chi-Hsien
of the Abra earthquake (Perez et al., 2023; J. Rimando et al., 2022), none has objectively explored, which is
Tang, Ya-Ju Hsu, Yunung Nina Lin
crucial for characterizing a blind fault.
In this study, we use Interferometric Synthetic Aperture Radar (InSAR) and Global Navigation Satellite System
(GNSS) measurements to characterize the coseismic deformation of the 2022 Abra earthquake, objectively
explore the fault geometry, and image the coseismic slip distribution. We show that the earthquake occurred on
an oblique blind fault beneath the Central Cordillera. We then examine the slip behavior of the Abra mainshock,
the earthquake-induced stress changes on nearby active faults, and the seismogenic thickness of northern Luzon
to characterize the seismic hazards there.

2. Geodetic Data
We generated the coseismic interferogram using the synthetic aperture radar images from Copernicus Sentinel-1A
descending track 32 on 21 July and 2 August 2022 (six days before and after the mainshock). The flight direction
is ∼N190° with a westward look angle ranging from 36° to 45°. We processed the interferogram using InSAR
Scientific Computing Environment (ISCE) software (Rosen et al., 2012) and corrected for the tropospheric effects
(Yu et al., 2018) (Figure S1 in Supporting Information S1). To improve the computational efficiency of coseismic
slip modeling, we downsampled the interferogram from ∼300,000 to ∼4,700 data points based on the interfero-
metric coherence and the spatial gradient of the line-of-sight (LOS) displacement field (Text S1 and Figure S2 in
Supporting Information S1). The sampled step is ∼2 km near the epicenter and ∼5 km in mountainous areas or
the far field (i.e., ∼80 km east of the epicenter).
The interferogram shows no clear discontinuities across any active faults (Figure 2a). Instead, the deformation
exhibits a long-wavelength pattern, suggesting that the coseismic slip is distributed deeply in the crust and did
not propagate to the surface. This aligns with the fact that no surface rupture was observed after the earthquake
(Perez et al., 2023). The deformation pattern contains two quadrants of opposite displacement signs—a shorten-
ing quadrant near the epicenter and a lengthening quadrant to the southeast—with a peak-to-trough displacement
of ∼26 cm. Because only the LOS displacement field along the descending track is available, it is challenging
to determine the fractions of horizontal and vertical coseismic motions. However, the dense spatial sampling of
the InSAR data compensates for the sparse GNSS measurements and allows a better determination of the seis-
mogenic fault at play.
The GNSS surveys in Luzon began in 1995 under the collaboration between the Institute of Earth Sciences,
Academia Sinica, Taiwan, and the Philippine Institute of Volcanology and Seismology (Hsu et al., 2012, 2016;
Yu et al., 1999, 2013). The GNSS network consists of both continuous GNSS (cGNSS) sites and survey-mode
GNSS (sGNSS). We collected data from five cGNSS sites (BRGC, CLAV, PAGP, TGDN, and VIGN) in north-
western Luzon (Figure 2a). The time series include five days after the mainshock. We used the final orbit products
from International GNSS Service (IGS) and processed cGNSS data with GAMIT/GLOBK software (Herring
et al., 2002). We incorporated 34 sites from IGS and 12 sites from the Taiwan network to process the daily solu-
tions in Luzon. Final solutions were transformed into the iGb08 reference frame (Rebischung et al., 2012). To
extract the coseismic displacements, we removed the interseismic linear trend from the whole cGNSS time series,
and then estimated the difference between the positions before and after the 2022 mainshock (Figures S3 and S4
in Supporting Information S1).
For sGNSS sites, we have conducted campaign surveys repeatedly since 1998. Each campaign includes two
to three days of observations. To examine the coseismic displacements of the 2022 mainshock, we reoccupied
two sites (BR14 and LUZD) between 30 July and 1 August 2022. To strengthen the data coverage east of the

TANG ET AL. 2 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Figure 1. Tectonics of northern Luzon. Brown lines are mapped active faults (Styron & Pagani, 2020). Major earthquakes
(M > 6) since 1900 with depths less than 35 km are marked with white circles. The white open circle shows the 2022
epicenter from PHIVOLCS. The centroid moment tensors of the 1990 and 2022 events are indicated with black focal
mechanisms from GCMT (Ekström et al., 2012). The green focal mechanism corresponds to our preferred coseismic slip
model. Six days of aftershocks following the 2022 mainshock (PHIVOLCS, 2022) are shown as red dots. The orange zone
illustrates the 1990 surface rupture (Silcock & Beavan, 2001). Yellow triangles mark volcanoes. The inset illustrates the
regional geography and plate tectonics. ARF: Abra River Fault; DF: Dalton Fault; DDF: Digdig Fault; NCF: North Cordillera
Fault; PHF: Philippine Fault; SMF: San Manuel Fault; TF: Tubao Fault; VAF: Vigan-Aggao Fault; CB: Cagayan Valley
Basin; CC: Cordillera Central; CVB: Central Valley Basin; NSM/SSM: Northern/Southern Sierra Madre; ZR: Zambales
Range; PH: Philippine Sea Plate; SU: Sunda Plate; YZ: Yangtze Plate.

mainshock, we acquired observations from IFG1 and KA08 on 29 November 2022 (Figure S5 in Supporting
Information S1). We employed the same data processing for cGNSS and estimated coseismic displacements by
fitting the sGNSS time series using a Heaviside function at the time of the mainshock and a linear trend.
At BR14, the site closest to the mainshock and located within the shortening quadrant of the InSAR data, the
GNSS coseismic displacements reach ∼22 cm of northwestward motion and ∼22 cm of uplift (Figure 2b). The
coseismic displacements gradually change to westward motion in the west coast area around Vigan. GNSS sites
more than 80 km away from the mainshock record coseismic displacements of less than ∼1 cm. All the GNSS
coseismic displacements are listed in Table S2 in Supporting Information S1 (Figure S1f in Supporting Informa-
tion S1 shows the comparison between GNSS and InSAR data).

3. Coseismic Slip Modeling

We relied on a kinematic inversion combining InSAR and GNSS data (Text S2 in Supporting Information S1) to
explore the fault geometry and the slip distribution during the Abra earthquake in a two-step strategy. In the first
step, we attribute surface deformation to a uniform slip on a planar fault in an elastic half-space (Okada, 1985). We

TANG ET AL. 3 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Figure 2. Coseismic displacements and model fit. (a) Unwrapped interferogram from descending Sentinel-1A images acquired on 21 July and 2 August 2022. The
flight and line-of-sight (LOS) directions are shown with black vectors. White vectors indicate horizontal GNSS coseismic displacements. Error ellipses show 95%
confidence intervals. The white star marks the epicenter of the 2022 mainshock. (b) Survey-mode GNSS position time series at BR14 with respect to the Philippine Sea
Plate since 2009. Error bars show 95% confidence intervals. (c) Modeled coseismic displacements and (d) residuals produced by the coseismic slip model.

performed a Markov Chain Monte Carlo (MCMC) algorithm (Hastings, 1970) to approximate the posterior distri-
bution of fault geometry parameters, including the location, length, width, strike, and dip of the fault (Figures S6
and S7 in Supporting Information S1). Despite the lack of surface rupture, the higher earthquake intensity on the
hangingwall compared to the footwall (Perez et al., 2023) and the north-south-trending distribution of aftershocks
(Figure 1) suggest that the Abra earthquake likely occurred on the north-south-striking, east-dipping fault plane.
Therefore, we searched for the fault strike and dip in the ranges of −30°–30° and 0°–60°, respectively. We gener-
ated 100,000 MCMC samples with an acceptance rate of ∼25%, indicating a good exploration throughout the
model space. To mitigate the effects of initial values and the autocorrelation in the MCMC process, we truncated
the first 25% of samples (burn-in) and downsampled the rest by a factor of 0.2 (thinning). From the remaining

TANG ET AL. 4 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

15,000 samples, we estimated the fault geometry with the maximum posterior density. We then used it as an
initial value in the nonlinear inversion to find the fault geometry that minimizes the residual norm. To remove
long-wavelength orbital errors in the InSAR data, we simultaneously fit a bilinear ramp of the LOS displacement
field in the inversion. As a result, we note strong constraints on fault location and length, but relatively limited
constraints on width, strike, and dip (Figure S6 and Table S3 in Supporting Information S1).
In the second step, we explored the coseismic slip distribution using the same data set and the fault geometry
of the best-fit uniform slip model. We fixed the location, strike, and dip of the fault but extended the length and
width to 100 and 60 km, respectively, according to the aftershock distribution. The top of the new fault geom-
etry is at a depth of 5 km and the bottom at a depth of ∼34 km. We meshed the fault into 5- by 5-km subfaults,
resulting in 240 subfaults in total. We employed the non-negative least squares approach to penalize right-lateral
slip and normal faulting on the fault. Because the data variance of InSAR is unknown, we tuned the weight
between GNSS and InSAR data to balance their misfit (Text S2 and Figure S8 in Supporting Information S1).
The inversion is stabilized by imposing second-order Tikhonov regularization (smoothing) to the strike- and
dip-slip components. The weight of the smoothing was assigned to balance the trade-off between data misfit and
model roughness (Figure S9 in Supporting Information S1). A checkerboard test was performed to ensure that
the geodetic data provide sufficient resolution to track the mainshock rupture (Figure S10 in Supporting Infor-
mation S1). Moreover, we show that the additional 4-month postseismic transient at GNSS sites IFG1 and KA08
does not significantly change the inverted coseismic slip distribution (Figure S11 in Supporting Information S1).

4. Discussion and Conclusions

The optimal fault geometry of the uniform slip model exhibits a strike of N353° and a dip of 29° to the east
(Figure S7 and Table S3 in Supporting Information S1). A uniform slip of 1.2 m with a rake of 33° occurred on
a 41-km by 22-km fault plane at depths ranging from ∼11 to ∼22 km. The inferred fault is located in the south
of the hypocenter at a slightly deeper depth, indicating a southward and downdip propagation of the coseismic
rupture (Figure S7 in Supporting Information S1). This inference is consistent with the earthquake centroid
located to the south of the hypocenter at a deeper depth (Figure 1 and Figure S7 in Supporting Information S1),
and the finite fault model constrained by teleseismic body waves and long-period surface waves (USGS, 2022b).
Figures 2c and 2d show the data fitting of our preferred coseismic slip model with multiple subfaults (see GNSS
vertical fit in Figure S12 in Supporting Information S1 and residuals in a greater scale in Figure S13 in Support-
ing Information S1). The model reduces the variance of InSAR and GNSS data by 94% and 93%, respectively,
with a root-mean-square error of ∼1.3 cm for both. The coseismic slip distribution along the fault shows peak slip
of ∼1 m, slightly smaller than the uniform slip model, at the depth of 16–18 km (Figures 3a–3c). The coseismic
slip is dominated by left-lateral slip at shallow depths and changes to predominantly reverse slip at depths greater
than 17 km (Figure 3c). The strike- and dip-slip motions peak at depths of ∼14 and ∼18 km, respectively, with
comparable amounts of slip. Our model shows that the mainshock rupture was likely initiated by the left-lateral
slip in the proximity of the hypocenter and subsequently drove reverse slip at deep depths. An average rake angle
of ∼41° is estimated using strike-slip and dip-slip occurred above the depth of ∼24 km, where the model resolu-
tion is sufficient (Figure S10 in Supporting Information S1). The corresponding focal mechanism is illustrated in
Figure 1 as a green-shaded beach ball, which agrees with the solution from the GCMT (Ekström et al., 2012). The
azimuth of the P-axis estimated from our geodetic slip model is ∼N300°, roughly consistent with the compres-
sional stress axes derived from the focal mechanisms before the mainshock (Wu et al., 2017). This shows that the
occurrence of the Abra earthquake is anticipated, although the ruptured fault is elusive.
We estimate the geodetic moment magnitude given 𝐴𝐴 by log10 𝑀𝑀0 − 9.095 ∕1.5, where M0 is the moment release
( )

that can be expressed in a discrete summation of ∑μAisi, where μ is the rigidity (30 GPa), Ai is the area of each
subfault with a slip si. The moment releases of our uniform slip model and distributed slip model are 3.12 × 10 19
and 4.96 × 10 19 N·m −1, respectively, corresponding to geodetic moment magnitudes of 6.93 and 7.07. The
moment magnitude of the distributed slip model is slightly greater than that of the uniform slip model, but they
both fall within a range of estimates derived from seismological observations based on either centroid moment
tensor or W-phase (Table S1 in Supporting Information S1). Our geodetically-derived distributed slip model is
comparable to the seismologically-constrained slip model with peak slip of ∼1 m and an equivalent moment
magnitude of 7.09 (USGS, 2022b). Our approach, however, objectively explores the coseismic fault geometry
through the MCMC algorithm and thus is more informative in terms of characterizing the regional seismogenic
structures. This gives us confidence that the resulting north-south trending fault is creditable.

TANG ET AL. 5 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Figure 3. Coseismic slip model and seismicity. (a) Coseismic slip distribution inferred from InSAR and GNSS data. The red dashed line marks the fault top projected
on the surface. (b) Spatial correlation of the 2022 coseismic slip, aftershocks, seismicity since 1990 (Hsu et al., 2016; M > 3 and depths <35 km), and the 1990 surface
rupture. Black circles are seismicity within three months after the 1990 earthquake. (c) Maximum coseismic slip at each depth. Gray bars show the normalized number
of 3-month aftershocks with an interval of 3 km. (d) The normalized number of earthquakes between 1900 and the 2022 Abra mainshock. The sampling region of
seismicity is enclosed by the dashed box in (b). Magenta dashed lines indicate the shallowest depth of the subducting Sunda Plate interface in the region. D95 marks the
depth of the 95% seismicity using earthquakes above 40 km depth.
The inferred fault geometry apparently deviates from the Abra River Fault (ARF), a vertically dipping sinistral
strike-slip fault near the epicenter (Figure 3b). Instead, the updip surface projection of the fault roughly coin-
cides with the southern segment of the Vigan-Aggao Fault (VAF), which represents the active deformation front
onshore along northwestern Luzon (Pinet & Stephan, 1990; Ringenbach et al., 1993). The morphology of the

TANG ET AL. 6 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

VAF suggests that it is an oblique sinistral strike-slip fault (R. E. Rimando & Rimando, 2020), consistent with
the predominant left-lateral slip above ∼15 km depth during the mainshock (Figure 3c). These findings suggest
that the Abra coseismic rupture may occur on the downdip extension of the VAF. Alternatively, the fault could
be another branch of the PFZ between the VAF and ARF and does not reach the surface. Because the coseismic
slip is distributed on both sides of the ARF, our inferred blind fault is likely merged with the ARF at depths.
Despite limited constraints on strike angle provided by geodetic data (Figure S6 in Supporting Information S1),
the inferred blind fault generally aligns with the PFZ.

Interestingly, there appears to be a “seismic gap” between the 2022 Abra rupture and the surface rupture of the
1990 Mw 7.7 Luzon earthquake along the Digdig Fault (Figure 3b). The southward propagating rupture during
the Abra earthquake and the triggered aftershocks roughly terminated at 17°N, whereas the aftershocks following
the 1990 earthquake (M > 3) mainly concentrate between 16°N and 16.5°N. The seismic activity is relatively
sparse between 16.5°N and 17°N since 1990. The seismic gaps may be viewed as an indicator of where future great
earthquakes are likely to occur in a fault system (Bilek & Lay, 2018; McCann et al., 1979; Nishenko, 1991). It has
been shown to match the occurrence of some large earthquakes along the San Andreas fault (Bakun et al., 2005;
U.S. Geological Survey Staff, 1990) and the subduction zone megathrusts in Chile (Moreno et al., 2010, 2012)
and Aleutian (Elliott et al., 2022; Herman & Furlong, 2021). If no other mechanism (e.g., aseismic slip) accom-
modates the accumulated elastic strain along the PFZ between 16.5°N and 17°N, the elastic strain may, at least
in principle, eventually be released as earthquakes to close the seismic gap. The concept of seismic gap is,
however, often challenged due to the lack of statistical significance led by insufficient observations (Kagan &
Jackson, 1991, 1995). Given this limitation, we also recommend caution with these interpretations.

To elucidate future seismic hazards in northern Luzon, we examine the static Coulomb stress changes imparted
by the Abra earthquake on the shallow portions of nearby active faults, particularly on the three major branches
of the PFZ from west to east: the VAF, the ARF, and the Dalton Fault. We adopt the fault model proposed by
Hsu et al. (2016), which captures the first-order geometry of the complex braided system in northern Luzon, as
the receiver faults. All faults are assigned a vertical dip angle with left-lateral strike-slip motions except for the
VAF, which is dipping 30° to the east with a rake assumed to be 30° based on our slip model. We relate the Abra
coseismic slip to stress tensor changes at any given location in an elastic half-space (Okada, 1992). A friction
coefficient of 0.4 is employed to weigh normal stress (King et al., 1994). The coseismic Coulomb stress changes
on each fault at a depth of 5 km show stress increases of >0.5 bar on the three aforementioned active faults
(Figure 4a). The stress increase on the VAF is especially significant (∼2 bars), which is not surprising because it
is right at the updip of the source fault. The M > 3 aftershocks mostly scatter around the hypocenter of the main-
shock within a depth of 20 km, where the amplitude of stress changes is significant (Figure 4b). Some aftershocks
occurred in the quadrants of negative Coulomb stress changes, likely indicating a substantial prestress in play or
the uncertainties in aftershock locations. Nevertheless, we estimated a Coulomb stress increase of ∼0.2 bars at
the depth of 16 km where a Mw 6.4 aftershock occurred three months after the mainshock (Figure S14 in Support-
ing Information S1). The mainshock also alters stress changes in the deep crust around the subducting Sunda
Plate (Figure 4b), which may promote aseismic creep and viscoelastic flow there (e.g., Tang et al., 2019, 2020).
Continued geodetic monitoring would reveal the seismic-cycle deformation following the Abra earthquake and
its possible interaction with the subducting plate.

The deep Abra coseismic slip also urged us to examine the seismogenic thickness in northern Luzon. To do so,
we calculate the 95% seismicity depth (D95) using both the 3-month aftershocks and the seismicity since 1900.
We use seismicity above 40 km depth to preclude the influence of the subducted plate. The resulting D95 is 31
and 34 km for aftershocks and background seismicity, respectively (Figures 3c and 3d). The estimated seismo-
genic layer is thicker than most other continental strike-slip fault systems (Donzé et al., 2021). For instance,
the seismogenic thickness is ∼15 km surrounding the San Andreas fault (Hauksson & Meier, 2019; Zuza &
Cao, 2020), ∼15 km for north-central Anatolia (Yolsal-Çevikbilen et al., 2012), and ∼8 km for the Alpine fault
(Warren-Smith et al., 2022). Hsu et al. (2016) used GNSS velocities and a fault locking depth of 20 km to assess
the seismic potential of inland faults in Luzon ranging from Mw 6.9 to 7.6 over a centennial recurrence inter-
val. Because the seismic moment is proportional to the size of the fault plane, the 50% increase in seismogenic
thickness (from 20 to 30 km) results in the same proportion of the increase in the seismic potential. The updated
seismic moment of large inland faults in Luzon is equivalent to Mw 7.0–7.7. The cause of the thick seismogenic
layer remains perplexing but may be related to cracks and conduits left beneath the volcanic arc and/or water

TANG ET AL. 7 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Figure 4. Coulomb stress change induced by the 2022 earthquake. (a) Stress change on the faults in northern Luzon.
The VAF is dipping 30° to the east with a rake of 30°, while the other faults are dipping vertically with left-lateral
slip. Background color shows the stress change at a depth of 5 km along the slip direction (Strike = N353°, Dip = 29°,
Rake = 41°). Rigidity and friction coefficient are 30 GPa and 0.4, respectively. The white star and circles mark the mainshock
and M > 3 aftershocks, respectively. The dashed box outlines the source fault. (b) Stress change along the AA’ cross section.
The gray line shows the side projection of the source fault. The black dashed curve illustrates the subducting Sunda Plate
interface from Hsu et al. (2016). White circles indicate M > 3 aftershocks within ±15 km width.

dehydrated from the subducted sediments (Armada et al., 2020). These conditions weaken the lower crust of
northern Luzon, leading to lower yield stress for brittle deformation than for ductile deformation, and eventually
thickening the seismogenic layer (e.g., Hauksson & Meier, 2019). This concept was also invoked to explain the
strain localization in the lower crust beneath the Japan arc (Iio et al., 2002). Yet, we are aware of the uncertainty in
hypocentral depth introduced by velocity models and limited seismic network coverage. Future studies on earth-
quake relocation, Vp/Vs tomography, and magnetotelluric imaging in Luzon will provide additional constraints
to delineate real causes.

TANG ET AL. 8 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Data Availability Statement

Figures and maps were prepared using Generic Mapping Tools (Wessel et al., 2013). The topography was down-
loaded from Global Multi-Resolution Topography Synthesis (https://www.gmrt.org/GMRTMapTool/). The
Sentinel-1 images were available through ASF Data Search Vertex (https://search.asf.alaska.edu/). The original
and downsampled unwrapped InSAR LOS displacements used in this study can be downloaded from a Zenodo
repository (https://doi.org/10.5281/zenodo.7699489). The GNSS coseismic displacements are listed in Table S2
in Supporting Information S1. The seismicity in the Philippines was acquired from PHIVOLCS Earthquake
Bulletins (https://www.phivolcs.dost.gov.ph/index.php/earthquake/earthquake-information3).

Acknowledgments References
We thank Zachary Albert Ragadio,
Cassandra Joy Cabigan, Abegail Armada, L. T., Hsu, S.-K., Dimalanta, C. B., Yumul, G. P., Jr., Doo, W.-B., & Yeh, Y.-C. (2020). Forearc structures and deformation along the
Abrenica, Bon Jovi Morales, Alyssa Manila Trench. Journal of Asian Earth Sciences X, 4, 100036. https://doi.org/10.1016/j.jaesx.2020.100036
Dane Pariñas, Justine Hadap, Alberto Aurelio, M. A. (2000). Shear partitioning in the Philippines: Constraints from Philippine Fault and global positioning system data. Island Arc,
Baloto, and Patricia Belardo for their 9(4), 584–597. https://doi.org/10.1111/j.1440-1738.2000.00304.x
contributions to the Luzon GNSS Aurelio, M. A., Dianala, J. D. B., Taguibao, K. J. L., Pastoriza, L. R., Reyes, K., Sarande, R., & Lucero, A. (2017). Seismotectonics of the 6
network. We thank John Dale Dianala, February 2012 Mw 6.7 Negros earthquake, central Philippines. Journal of Asian Earth Sciences, 142, 93–108. https://doi.org/10.1016/j.
Jeremy Rimando, and John Emmanuel jseaes.2016.12.018
Fungo for their useful discussions. We Bakun, W. H., Aagaard, B., Dost, B., Ellsworth, W. L., Hardebeck, J. L., Harris, R. A., et al. (2005). Implications for prediction and hazard assess-
appreciate the editor Lucy Flesch and ment from the 2004 Parkfield earthquake. Nature, 437(7061), 969–974. https://doi.org/10.1038/nature04067
two anonymous reviewers who provided Barrier, E., Huchon, P., & Aurelio, M. (1991). Philippine fault: A key for Philippine kinematics. Geology, 19(1), 32–35. https://doi.
constructive comments on the manuscript. org/10.1130/0091-7613(1991)019<0032:PFAKFP>2.3.CO;2
We thank European Space Agency Bilek, S. L., & Lay, T. (2018). Subduction zone megathrust earthquakes. Geosphere, 14(4), 1468–1500. https://doi.org/10.1130/GES01608.1
(ESA) and Alaska Satellite Facility Bonita, J. D., Kumagai, H., & Nakano, M. (2015). Regional moment tensor analysis in the Philippines: CMT solutions in 2012–2013. Journal of
(ASF) for providing Sentinel-1 images. Disaster Research, 10(1), 18–24. https://doi.org/10.20965/jdr.2015.p0018
This study was supported by Academia Donzé, F.-V., Klinger, Y., Bonilla-Sierra, V., Duriez, J., Jiao, L., & Scholtès, L. (2021). Assessing the brittle crust thickness from strike-slip fault
Sinica (AS-GC-110-03) and the National segments on Earth, Mars and Icy moons. Tectonophysics, 805, 228779. https://doi.org/10.1016/j.tecto.2021.228779
Science and Technology Council of Ekström, G., Nettles, M., & Dziewoński, A. M. (2012). The global CMT project 2004–2010: Centroid-moment tensors for 13,017 earthquakes.
Taiwan (MOST 110-2923-M-006-001). Physics of the Earth and Planetary Interiors, 200–201, 1–9. https://doi.org/10.1016/j.pepi.2012.04.002
This is IESAS contribution 2413. Elliott, J. L., Grapenthin, R., Parameswaran, R. M., Xiao, Z., Freymueller, J. T., & Fusso, L. (2022). Cascading rupture of a megathrust. Science
Advances, 8(18), eabm4131. https://doi.org/10.1126/sciadv.abm4131
Galgana, G., Hamburger, M., McCaffrey, R., Corpuz, E., & Chen, Q. (2007). Analysis of crustal deformation in Luzon, Philippines using geodetic
observations and earthquake focal mechanisms. Tectonophysics, 432(1–4), 63–87. https://doi.org/10.1016/j.tecto.2006.12.001
Hastings, W. K. (1970). Monte Carlo sampling methods using Markov chains and their applications. Biometrika, 57(1), 97–109. https://doi.
org/10.2307/2334940
Hauksson, E., & Meier, M.-A. (2019). Applying depth distribution of seismicity to determine thermo-mechanical properties of the seismogenic
crust in Southern California: Comparing lithotectonic blocks. Pure and Applied Geophysics, 176(3), 1061–1081. https://doi.org/10.1007/
s00024-018-1981-z
Herman, M. W., & Furlong, K. P. (2021). Triggering an unexpected earthquake in an uncoupled subduction zone. Science Advances, 7(13),
eabf7590. https://doi.org/10.1126/sciadv.abf7590
Herring, T. A., King, R. W., & McClusky, S. C. (2002). Documentation for the GAMIT analysis software, release (Vol. 10.0). Massachusetts
Institute of Technology.
Hsu, Y.-J., Yu, S.-B., Kuo, L.-C., Tsai, Y.-C., & Chen, H.-Y. (2011). Coseismic deformation of the 2010 Jiashian, Taiwan earthquake and implica-
tions for fault activities in southwestern Taiwan. Tectonophysics, 502(3–4), 328–335. https://doi.org/10.1016/j.tecto.2011.02.005
Hsu, Y.-J., Yu, S.-B., Loveless, J. P., Bacolcol, T., Solidum, R., Luis, A., et al. (2016). Interseismic deformation and moment deficit along the
Manila subduction zone and the Philippine Fault system: Slip-Deficit rates in Luzon. Journal of Geophysical Research: Solid Earth, 121(10),
7639–7665. https://doi.org/10.1002/2016JB013082
Hsu, Y.-J., Yu, S.-B., Song, T.-R. A., & Bacolcol, T. (2012). Plate coupling along the Manila subduction zone between Taiwan and northern
Luzon. Journal of Asian Earth Sciences, 51, 98–108. https://doi.org/10.1016/j.jseaes.2012.01.005
Iio, Y., Sagiya, T., Kobayashi, Y., & Shiozaki, I. (2002). Water-weakened lower crust and its role in the concentrated deformation in the Japanese
Islands. Earth and Planetary Science Letters, 203(1), 245–253. https://doi.org/10.1016/S0012-821X(02)00879-8
Kagan, Y. Y., & Jackson, D. D. (1991). Seismic gap hypothesis: Ten years after. Journal of Geophysical Research, 96(B13), 21419–21431. https://
doi.org/10.1029/91JB02210
Kagan, Y. Y., & Jackson, D. D. (1995). New seismic gap hypothesis: Five years after. Journal of Geophysical Research, 100(B3), 3943–3959.
https://doi.org/10.1029/94JB03014
King, G. C. P., Stein, R. S., & Lin, J. (1994). Static stress changes and the triggering of earthquakes. Bulletin of the Seismological Society of
America, 84, 935–953. https://doi.org/10.1785/BSSA0840030935
Kreemer, C., Blewitt, G., & Klein, E. C. (2014). A geodetic plate motion and global strain rate model. Geochemistry, Geophysics, Geosystems,
15(10), 3849–3889. https://doi.org/10.1002/2014GC005407
McCann, W. R., Nishenko, S. P., Sykes, L. R., & Krause, J. (1979). Seismic gaps and plate tectonics: Seismic potential for major boundaries. Pure
and Applied Geophysics, 117(6), 1082–1147. https://doi.org/10.1007/BF00876211
Moreno, M., Melnick, D., Rosenau, M., Baez, J., Klotz, J., Oncken, O., et al. (2012). Toward understanding tectonic control on the Mw 8.8 2010
Maule Chile earthquake. Earth and Planetary Science Letters, 321–322, 152–165. https://doi.org/10.1016/j.epsl.2012.01.006
Moreno, M., Rosenau, M., & Oncken, O. (2010). 2010 Maule earthquake slip correlates with pre-seismic locking of Andean subduction zone.
Nature, 467(7312), 198–202. https://doi.org/10.1038/nature09349
Nishenko, S. P. (1991). Circum-Pacific seismic potential: 1989–1999. Pure and Applied Geophysics, 135(2), 169–259. https://doi.org/10.1007/
BF00880240

TANG ET AL. 9 of 10
 19448007, 2023, 9, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GL103659 by Cochrane Philippines, Wiley Online Library on [06/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Geophysical Research Letters 10.1029/2023GL103659

Okada, Y. (1985). Surface deformation due to shear and tensile faults in a half-space. Bulletin of the Seismological Society of America, 75(4),
1135–1154. https://doi.org/10.1785/bssa0750041135
Okada, Y. (1992). Internal deformation due to shear and tensile faults in a half-space. Bulletin of the Seismological Society of America, 82(2),
1018–1040. https://doi.org/10.1785/bssa0820021018
Perez, J. S., Llamas, D. C. E., Dizon, M. P., Buhay, D. J. L., Legaspi, C. J. M., Lagunsad, K. D. B., et al. (2023). Impacts and causative fault of
the 2022 magnitude (Mw) 7.0 Northwestern Luzon earthquake, Philippines. Frontiers of Earth Science, 11, 1091595. https://doi.org/10.3389/
feart.2023.1091595
PHIVOLCS. (2022). Retrieved from https://www.phivolcs.dost.gov.ph/index.php/earthquake/earthquake-information3
Pinet, N., & Stephan, J. F. (1990). The Philippine wrench fault system in the Ilocos Foothills, northwestern Luzon, Philippines. Tectonophysics,
183(1–4), 207–224. https://doi.org/10.1016/0040-1951(90)90417-7
Punongbayan, B. J. T., Kumagai, H., Pulido, N., Bonita, J. D., Nakano, M., Yamashina, T., et al. (2015). Development and operation of a regional
moment tensor analysis system in the Philippines: Contributions to the understanding of recent damaging earthquakes. Journal of Disaster
Research, 10(1), 25–34. https://doi.org/10.20965/jdr.2015.p0025
Rebischung, P., Griffiths, J., Ray, J., Schmid, R., Collilieux, X., & Garayt, B. (2012). IGS08: The IGS realization of ITRF2008. GPS Solutions,
16(4), 483–494. https://doi.org/10.1007/s10291-011-0248-2
Rimando, J., Williamson, A., Mendoza, R. B., & Hobbs, T. (2022). Source model and characteristics of the 27 July 2022 MW 7.0 northwestern
Luzon earthquake, Philippines. Seismica, 1(1). https://doi.org/10.26443/seismica.v1i1.217
Rimando, R. E., & Rimando, J. M. (2020). Morphotectonic kinematic indicators along the Vigan-Aggao Fault: The western deformation front of
the Philippine Fault Zone in Northern Luzon, the Philippines. Geosciences, 10(2), 83. https://doi.org/10.3390/geosciences10020083
Ringenbach, J. C., Pinet, N., Stéphan, J. F., & Delteil, J. (1993). Structural variety and tectonic evolution of strike-slip basins related to the Phil-
ippine Fault System, northern Luzon, Philippines. Tectonics, 12(1), 187–203. https://doi.org/10.1029/92TC01968
Ringenbach, J. C., Stephan, J. F., Maleterre, P., & Bellon, H. (1990). Structure and geological history of the Lepanto-Cervantes releasing bend on
the Abra river fault, Luzon Central Cordillera, Philippines. Tectonophysics, 183(1–4), 225–241. https://doi.org/10.1016/0040-1951(90)90418-8
Rosen, P. A., Gurrola, E., Sacco, G. F., & Zebker, H. (2012). The InSAR scientific computing environment. In Proceedings of the 9th European
conference on synthetic aperture radar (EUSAR) (pp. 730–733). Nuremberg.
Sella, G. F., Dixon, T. H., & Mao, A. (2002). REVEL: A model for recent plate velocities from space geodesy. Journal of Geophysical Research,
107(B4), ETG11-1–ETG11-30. https://doi.org/10.1029/2000JB000033
Silcock, D. M., & Beavan, J. (2001). Geodetic constraints on coseismic rupture during the 1990 Ms 7.8 Luzon, Philippines, earthquake. Geochem-
istry, Geophysics, Geosystems, 2(7), 2000GC000101. https://doi.org/10.1029/2000GC000101
Styron, R., & Pagani, M. (2020). The GEM global active faults database. Earthquake Spectra, 36(1), 160–180. https://doi.
org/10.1177/8755293020944182
Tang, C.-H., Barbot, S., Hsu, Y.-J., & Wu, Y.-M. (2020). Heterogeneous power-law flow with transient creep in Southern California follow-
ing the 2010 El Mayor-Cucapah earthquake. Journal of Geophysical Research: Solid Earth, 125(9), e2020JB019740. https://doi.
org/10.1029/2020JB019740
Tang, C.-H., Hsu, Y.-J., Barbot, S., Moore, J. D. P., & Chang, W.-L. (2019). Lower-crustal rheology and thermal gradient in the Taiwan orogenic
belt illuminated by the 1999 Chi-Chi earthquake. Science Advances, 5(2), eaav3287. https://doi.org/10.1126/sciadv.aav3287
U.S. Geological Survey Staff. (1990). The Loma Prieta, California, earthquake: An anticipated event. Science, 247(4940), 286–293. https://doi.
org/10.1126/science.247.4940.286
USGS.(2022a).Retrievedfrom https://earthquake.usgs.gov/earthquakes/eventpage/us6000i5rd/moment-tensor?source=us&code=us_6000i5rd_mww
USGS. (2022b). Retrieved from https://earthquake.usgs.gov/earthquakes/eventpage/us6000i5rd/finite-fault
Warren-Smith, E., Townend, J., Chamberlain, C. J., Boulton, C., & Michailos, K. (2022). Heterogeneity in microseismicity and stress near
rupture-limiting section boundaries along the late-interseismic alpine fault. Journal of Geophysical Research: Solid Earth, 127(10),
e2022JB025219. https://doi.org/10.1029/2022JB025219
Wessel, P., Smith, W. H. F., Scharroo, R., Luis, J., & Wobbe, F. (2013). Generic mapping tools: Improved version released. Eos, Transactions
American Geophysical Union, 94(45), 409–410. https://doi.org/10.1002/2013EO450001
Wu, W.-N., Lo, C.-L., & Lin, J.-Y. (2017). Spatial variations of the crustal stress field in the Philippine region from inversion of earthquake focal
mechanisms and their tectonic implications. Journal of Asian Earth Sciences, 142, 109–118. https://doi.org/10.1016/j.jseaes.2017.01.036
Yeats, R. S., & Huftile, G. J. (1995). The Oak Ridge fault system and the 1994 Northridge earthquake. Nature, 373(6513), 418–420. https://doi.
org/10.1038/373418a0
Yolsal-Çevikbilen, S., Biryol, C. B., Beck, S., Zandt, G., Taymaz, T., Adıyaman, H. E., & Özacar, A. A. (2012). 3-D crustal structure along the
North Anatolian Fault Zone in north-central Anatolia revealed by local earthquake tomography: Structure revealed by earthquake tomography.
Geophysical Journal International, 188(3), 819–849. https://doi.org/10.1111/j.1365-246X.2011.05313.x
Yoshida, Y., & Abe, K. (1992). Source mechanism of the Luzon, Philippines earthquake of July 16, 1990. Geophysical Research Letters, 19(6),
545–548. https://doi.org/10.1029/91GL02467
Yu, C., Li, Z., Penna, N. T., & Crippa, P. (2018). Generic atmospheric correction model for interferometric synthetic aperture radar observations.
Journal of Geophysical Research: Solid Earth, 123(10), 9202–9222. https://doi.org/10.1029/2017JB015305
Yu, S.-B., Hsu, Y.-J., Bacolcol, T., Yang, C.-C., Tsai, Y.-C., & Solidum, R. (2013). Present-day crustal deformation along the Philippine Fault in
Luzon, Philippines. Journal of Asian Earth Sciences, 65, 64–74. https://doi.org/10.1016/j.jseaes.2010.12.007
Yu, S.-B., Kuo, L.-C., Punongbayan, R. S., & Ramos, E. G. (1999). GPS observation of crustal deformation in the Taiwan-Luzon Region.
Geophysical Research Letters, 26(7), 923–926. https://doi.org/10.1029/1999GL900148
Zuza, A. V., & Cao, W. (2020). Seismogenic thickness of California: Implications for thermal structure and seismic hazard. Tectonophysics,
782–783, 228426. https://doi.org/10.1016/j.tecto.2020.228426

TANG ET AL. 10 of 10