Geodetic and Seismic Constraints on Strain Accumulation on the Hellenic Subduction Zone off Crete

Michael Floyd (1), Demitris Paradissis (2), Athanassis Ganas (3), Hayrullah Karabulut (4), Robert King (1), and Robert Reilinger (1)

(1) Massachusetts Institute of Technology, Cambridge, Massachusetts, United States [mfloyd@mit.edu]; (2) National Technical University of Athens, Athens, Greece;(3) National Observatory of Athens, Athens, Greece; (4) Kandilli Observatory and Earthquake Research Institute, Bogaziçi University, Istanbul, Turkey

References

20˚ 22˚ 24˚ 26˚ 28˚ 30˚34˚

36˚

38˚

40˚

42˚B l a c k

S e a

A e g e a n

S e a

M e d i t e r r a n e a n S e a

H e l l e n i c

S u b d u c t i o n

Z o

n e

K. T

. F

.

A n a t o l i a

20 mm/yr (95% conf.)

SurveyContinuous

1. Introduction

20˚ 22˚ 24˚ 26˚ 28˚ 30˚34˚

36˚

38˚

40˚

42˚B l a c k

S e a

A e g e a n

S e a

M e d i t e r r a n e a n S e a

H e l l e n i c

S u b d u c t i o n

Z o

n e

K. T

. F.

A n a t o l i a

10 mm/yr (95% conf.)

SurveyContinuous

Floyd, M. A. (2008), Continuous and survey Global Positioning System observations of the deformation of the Aegean, D.Phil. thesis, University of Oxford, 202 pp.

Pirazzoli, P. A., J. Thommeret, Y. Thommeret, J. Laborel, L. F. Montaggioni (1982), Crustal Block Movements from Holocene Shorelines: Crete and Antikythira (Greece), Tectonophys., 86, 27–43, doi:10.1016/0040-1951(82)90060-9.

Shaw, B., and J. Jackson (2010), Earthquake mechanisms and active tectonics of the Hellenic subduction zone, Geophys. J. Int., 181, 966–984, doi:10.1111/j.1365-246X.2010.04551.x.

Vernant, P., R. Reilinger, and S. McClusky (2014), Geodetic evidence for low coupling on the Hellenic subduction plate interface, Earth Planet. Sci. Lett., 385, 122–129, doi:10.1016/j.epsl.2013.10.018.

4. Updated and Expanded GPS Velocity Solution

20˚ 22˚ 24˚ 26˚ 28˚

34˚

36˚

38˚A

a

B

b

C

c

D

d

E

e

5 mm/yr (95% conf.)

Volcanoes

SurveyContinuous

−8

−6

−4

−2

0

2

4

v⊥ / m

m/y

r

0 200 400 600

Distance along profile / km

−8

−6

−4

−2

0

2

4

6

v|| / m

m/y

r

A a

−3

−2

−1

0

1

2

v⊥ / m

m/y

r

0 50 100 150 200 250

Distance along profile / km

−4

−3

−2

−1

0

1

2

3

4

v|| / m

m/y

r

C c

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

v⊥ / m

m/y

r

0 50 100 150 200 250

Distance along profile / km

−4

−3

−2

−1

0

1

2

v|| / m

m/y

r

B b

Figure 6 Block model of contraction (negative numbers in parentheses) across the Hellenic Subduction Zone and extension of the Aegean. Nubian subduction interface (black line), Hellenic Trench (blue line) and Kephalonia Transform Fault (red line) have their locking (coupling) coefficient systematically varied in the inversion to assess the effect of locking on fit to GPS velocities. Note that east-west extension is accommodated by two small blocks (circled).

Subduction zones produce the largest recorded earthquakes. During the inter-seismic period they tend to exhibit contraction on the upper plate, consistent with elastic strain accumulation due to locking on the thrust interface, which produces GPS velocities directed away from the trench relative to the overriding plate. These motions are commonly found to reverse episodically in slow slip events. However, the Hellenic Subduction Zone is observed to have inter-seismic GPS velocities towards the trench and no recorded slow slip events. We consider here the nature of deformation and elastic strain accumulation along the Hellenic Subduction Zone from recent GPS velocity observations.

2. Previous Thrust Earthquakes

← Figure 1 Typical GPS velocities above subduction zones around the world: Japan (left), Cascadia (top), Chile (bottom) and our study area, the Hellenic Subduction Zone (right).

USGS Slab1.0 (https://earthquake.usgs.gov/data/slab/)

UNAVCO GPS Velocity Viewer

Hashimoto et al. (2009, Figure 2) Métois et al. (2013, Figure 3)

← Figure 2 Pirazzoli et al. (1982) reported significant uplift (up to 9 m) in south-west Crete. This uplift, when dated, coincides with an earthquake in 365 CE that was widely reported in historical records throughout the Mediterranean basin.

→ Figure 3 Shaw et al. (2008) revisited and confirmed the result of Pirazzoli et al. (1982). If the uplift occurred in a single event, it was likely a Mw 8.3–8.5 earthquake.

3. Inter-seismic Strain Accumulation from GPS Velocities

Pirazzoli et al. (1982, Figure 10)

Shaw et al. (2008, Figure 3a)

Figure 5 Inter-seismic strain accumulation model fitting the upper plate strain rates shown in Figure 4, left. Here, two patches near southwestern Crete require approximately 3 mm/yr slip rate deficit.

Figure 4 Inter-seismic strain rates derived from a network of continuous on baselines approximately perpendicular and parallel to the convergence azimuth (NNE) across the plate boundary (offshore to the southeast).

Figure 9 Latest iteration of the GPS velocity solution in the Aegean and western Anatolia relative to Nubia. The network consists of survey (blue) and continuous (red) GPS sites that have been observed by the institutions of the authors or is publicly available.

Figure 7 Schematic SW-NE cross-section through the interface geometry shown in Figure 6, left (letter B for reference).

Modified from Vernant et al. (2014, Figure 3B)

Figure 8 Misfit to GPS velocities as a function of locking (coupling) coeffic i en t on t he Nub ian subduction interface (black), Hel lenic Trench (blue) and Kephalonia Transform Fault (red). This model shows a general preference for low (< 40%) coupling, implying a slip rate deficit of < 15 mm/yr given the convergence rate of ~ 35 mm/yr.

Vernant et al. (2014, Figure 2B)

Vernant et al. (2014, Figure 4)

• Contraction between southwest Crete and the central Aegean Sea is now a persistent and reproducible geodetic result• Coupling is at least 10–25% (3–9 mm/yr slip rate deficit) but may be greater,

assuming extensional deformation of the Aegean is not confined to southwestern and southeastern Aegean

• Geometric constraints also suggest potential bounds of approximately:• 200 km along-strike• Wells and Coppersmith (1994) empirical relationships suggest maximum

potential Mw7.8–7.9• Up to 40 km down-dip, but no resolution of offshore up-dip extent• 3 mm/yr of slip rate deficit, without accounting for possible long-term extension• Lower bound of moment accumulation rate equal to:• Mw7.0 every 50–100 years• Mw7.5 every 275–550 years• Mw8.0 every 1550–3110 years

• It has been 1650 years since last major interplate earthquake (M > 7, several m of slip) off southwest Crete• Most recent smaller event was Mw6.8 interplate thrust on 14 February 2008

Conclusions

6. End-member Models of Upper Plate Deformation and Bounds of Inter-seismic Strain Accumulation

The Vernant et al. (2014) model, shown bottom-left of this poster, and previous block models such as Reilinger et al. (2006), provides a lower bound on elastic strain accumulation along the subduction interface by assuming that the upper plate is rigid, with elastic strain due to relative plate motions at the boundaries and no internal deformation. Viewing the velocities in an Aegean reference frame (Figure 12, right) shows broad extension both radially (perpendicular to the arc) and tangentially (parallel to the arc) throughout the southern Aegean. An upper bound on elastic strain accumulation along the interface may be found by a model or assumption of radial extension along the entire arc, including southwest Crete, which would then require a larger slip rate deficit on the subduction interface to produce the same observations.

→ Figure 12 Velocities in an Aegean reference frame.

Figure 10 Same as Figure 9, left, but relative to the central Aegean, defined by minimising the velocity of the GPS sites circled in yellow.

    2 mm/yr observed contraction      =     1–5 mm/yr long-term extension      +     3–7 mm/yr potential contraction(lower bound of potential moment)            (e.g. Figure 12; realistic physical model?)     (upper bound of potential moment)

25˚E 26˚E 27˚E 28˚E 29˚E 30˚E

33˚N

34˚N

35˚N

36˚N

37˚N

A

A'

B

B'

C

C'

0203550

100

150

km Mw

7

6

5

4

020406080

100120140160180

De

pth

(km

)

0 50 100 150 200 250 300 350

Distance along profile (km)

Profile A (E. Crete)Profile B (Rhodes)

A A'

B B'Trenchsystems

Rhodes E. CreteAegean Sea

Kos

20˚E 22˚E 24˚E 26˚E 28˚E 30˚E

34˚N

36˚N

38˚N

40˚N

42˚N

0203550

100

150

km Mw

7

6

5

4

Profiles through the new GPS velocity solution allow us to assess the relative contribution of extension in the upper plate and potential inter-seismic strain accumulation on the subduction interface. Profile A (Figure 11, below) shows the general pattern of tangential (arc-parallel) extension along the entire arc, as well as contraction around southwest Crete and radial (arc-perpendicular) extension elsewhere (Peloponnese in the west and Dodecanese Islands in the east). Profile C (Figure 11, below) shows this same contraction as well as a best-fit model assuming dip-slip on a fault dipping 20° that aligns with the Hellenic Subduction Zone. Profile B shows the extension (top) and a velocity gradient (bottom) consistent with block rotation and/or right-lateral shear of the Peloponnese.

←↓ Figure 11 Profiles through the latest GPS velocity solution. Note the difference in scale from Figure 10, above, and that only velocities less than 10 mm/yr are plotted.

3.0 mm/yr

3.1 mm/yr

4 mm/yr extension

Radial extension

200 km

Radial contraction

Radial extension

Peloponnese Crete

Dip-slip model: 9.2 ± 2.2 mm/yr

Perturbed by 2011 Santorini

inflation episode

−8

−6

−4

−2

0

2

v⊥ / m

m/y

r

0 50 100 150 200 250

Distance along profile / km

−4

−2

0

2

4

6

8

10

12

14

v|| / m

m/y

r

E e

1.5 mm/yr extension

20 mm/yr

7. Why the Contraction Around Southwest Crete?

↓ Figure 13 A schematic example of the summation of competing deformation signals—inter-seismic strain accumulation on the subduction interface and long-term extension of the Aegean—which may result in the overall pattern of observed GPS velocities.

Gavdos

Sedimentary wedge Crete Sea of Crete

←↑ Figure 14Shallow seismicity (from ISC) associated with subduction is closest to land around southwest Crete, implying that any contraction due to elastic strain accumulation is most likely to be observed there. The Gavdos block may also be impinging, causing an asperity.

←↑ Figure 15Foundering of the downgoing slab may be more extreme further east along the trench system, as seen in cross-sections of seismicity (from ISC). Profile C is not shown, above.

40 km

Abstract EGU2018-11522

European Geosciences UnionGeneral Assembly 2018

Vienna | Austria | 8–13 April 2018

Floyd (2009), DPhil thesis, Figure 3.1 Floyd (2009), DPhil thesis, Figure 3.4a