9th International Workshop on Statistical Seismology

Does the Spatial Analysis of 2012 Ahar-Varzeghan Seismic

Sequence Corroborate the Tectonic Interpretations?

Introduction To The Sequence

P. Yazdi, Universidad Politécnica de Madrid

M.A Santoyo García-Galiano, Universidad Nacional Autónoma de Mexico

J.M. Gaspar Escribano, Universidad Politécnica de Madrid

On 11 August 2012, two destructive shallow earthquakes Mw=6.4 andMw=6.2 occurred within distance of about 6 km and time lag of 11minutes in Northwestern Iran between the small cities of Ahar andVarzeghan with no previously well known active fault or significantseismicity. The following aftershock sequence remained strongly activeduring four months with more than 3500 events registered by theIranian Seismological Center in an area within a radius of 50 km. Themagnitude distribution of the sequence gives a b-value of 0.97 forcutoff magnitude Mw=2.0.

Re-Localization Of The Sequence

1st Shock Mw=6.4 2nd Shock Mw=6.2

Strike, Dip, Rake(source)

267, 81, -175 (IRSC)

7, 51, 21 (IRSC)

Depth to Top 0 km 11 km

Epi (lat, long) 38.39 N, 46.84 E 38.42 N , 46.77 E

Hypo Depth 6 km 13 km

Fault Size 22 x 12 km 15 x 9 km

Doublet’s Fault Geometry

An observed surface rupturewhich starts at 4 km west ofthe first mainshock continuesabout 8 km westward.

1

2

Figure 1) Surface Rupture location

Figure 2) Location of the sequence inNorthwestern Iran

Figure 3) Relative location of the fault rupturegeometries considered in this study, including asmoothed elliptical source model for the slip onthe fault.

Conclusions

Regional stress: looking at the regional stressmap If we consider a mean angle of 135 formaximum horizontal stress in the area on the andassume an approximately zero plunge then theoptimum strike slip faulting will be on verticalplanes that have angle of δ =(2β)’ with themaximum horizontal stress. It means an angle of169 for left lateral and 101 for a right lateralplanes for 𝜇=0.4.

CFS Changes Due To The First Mainshock

The seismicity since August 2012 to the end of November 2012has been re-localized using double difference algorithm ofWaldhauser and Ellswarth (2000). The maximum distance of 200km was considered between a pair and station and maximumdistance of 15 km for each pair of events. The principal clusterhas over 600 events with centroid at ∼ 38.42N, ∼46.74E and∼13.40 Km depth. For the velocity model the result of study byDonner et al.(2015) was used . (7 layers and Vp/Vs ratio 1.7)

Figure 4) Events with Mw>2.0 fromAugust to November 2012 within aradius of 50 Km

Figure 5) Distribution of epicenters after relocationFigure 7) Depth distribution of the re-located events in time

Figure 6) Gap between the re-located sequence and the locationof the faults

Figure 10) Stress map obtained with the online softwareCASMO using IRSC focal mechanism solution for theearthquakes on the Ahar-Varzeghan area during 4 months(August - November 2012)

References

Coulomb Failure Stress Changes: For strike, dip and rake of 101,90 and 180 the change of CFS due to both mainshocks and for frictional coefficient 0.4and 0.6 for the strike slip and strike slip with inverse components respectively were calculated. The main focus was on depths between 11 to 17 km and theresult shows high density of the events on the positive stress change areas. The CFS changes for optimum rives and normal planes was also calculated.

Z= 12 km Z= 14 km Z= 16 km

Figure 11) CFS changes at a depth of 12 km and events with depth

between 11-13 km : 75 % in the positive stress changes area

Figure 13) CFS changes at a depth of 16 km and events with depth

between 15-17 km : 55 % in the positive stress changes area

Figure 12) CFS changes at a depth of 14 km and events with depth

between 13-15 km : 72 % in the positive stress changes area

-5 bar +5 bar

Z= 12 km Z= 14 km Z= 16 km

Figure 15) CFS changes in different depths for reverse optimum faultingFigure 14) CFS changes in different depths for optimum normal faulting

Based on our knowledge from previous works investigating the mechanism of the faulting for thedoublet of Ahar-Varzeghan, we tried to find the best explanation using the changes in the coulombfailure stress and the aftershock sequence spatial distribution in different depths. The results for theassumption of North-South orientation plane for the second mainshock at lower depth in respect tothe first shock’s fault, was presented in this study. This assumption made us to be able to justify thespatial distribution of the sequence coming during 4 months and improve our understanding about thesource placement. The agreement between the spatial distribution of more than 500 relocated data atthe mean depth of 14 km with the CFS changes for the optimum strike slip faulting (as the mostprobable regional mechanism) leads us to accept that the doublet most likely occurred on aperpendicular structure and corroborate with the result of previous studies.

A. Copley et al. 2014, “The 2012 August 11 Ahar earthquakes: consequences for tectonics andearthquake hazard in the Turkish–Iranian Plateau”, Geophysics Journal International 196, 15–21

S. Donner et al. 2015, “The Ahar-Varzeghan Earthquake Doublet (Mw 6.4 and 6.2) of 11 August 2012:Regional Seismic Moment Tensors and a Seismotectonic Interpretation”, BSSA, Vol. 105, No. 2a

M. Faridi et al. 2012, “The Ahar-Varzeghan Earthquake Doublet (Mw 6.4, 6.2, Iran) ”, Geological Surveyof Iran

M. Mahood et al. 2014 “Simulation of the first earthquake August 11, 2012 Ahar-Varzaghan usingstochastic finite fault method”, Iranian Journal of Earth and Space Physics, Vol 40, No.2

Figure 8) CFS (bar) due to the 1st shock with 267-81-(-175) with smoothed elliptical weighting to the slip amount and for the 2ndshock with 7-57-21 on a horizontal plane with a) 12 Km depth and b) 13 Km depth. Slip was calculated using Wells andCoppersmith (1994) empirical relation for average slip and Hanks and Kanamori (1979) relation for released energy.

The coulomb stress change because of the first mainshock was studied in order to see if it has triggeredthe second mainshock. Considering the epicenter given by Donner et al. 2015 and with the mentionedgeometry for the fist mainshock, it is more likely that the second shock had happened at a hypocentraldepth lower than the edge of the first fault. using a smoothed elliptical source model the slip wasweighted over the fault plane considering the released energy. A fiction coefficient of 0.4 was consideredsince the faulting is strike slip.

a b

Figure 9) Simple Mohr Circle which explains that Coulomb Failure Stress 𝜎𝑓 = 𝜏 − 𝜇 𝜎𝑛depends on the friction coefficient 𝜇. The angle of between normal to the plane and themaximum stress is β.

𝜏

𝜎𝑛𝜎𝑛

𝜇

𝜑𝜎𝑓 2𝛽

𝜏 = 𝜇 𝜎𝑛 − 𝜎𝑓

07 Nov

14 Aug

15 Aug

East West parallel planes ?If yes, then both events rupturedmore or less the same area.(Copley et al. 2014 , Donner et al. 2015)

…. BUT

Most probably the rupture was due to the first mainshock

• Not in good agreement with rock mechanical perspectivesince they have almost the same size.

• The rupture orientation and focal mechanism of the secondshock do not agree

CFS Changes Due To The Doublet

14-19 June 2015

p.yazdi@alumnos.upm.essantoyo@geofísica.unam.mxJorge.Gaspar@upm.es