Academia.eduAcademia.edu

Figure 1 – uploaded by Richard Walker

See full PDFdownloadDownload figure
Table 1. Source parameters of the main Gowk valley earthquakes. Epicentres are from Engdahl er al. (1998). Magnitudes (m, and M,) are from the USGS. An m after the M,, value indicates that two subevents were required to model the body waves. Seismic moment (Mp) is in units of 10'8 N m. TF is the duration of the time function in seconds (the time for 95 per cent of the seismic moment to be released), and sv is the slip vector azimuth, assuming that the west-dipping nodal plane is the fault plane. The last column gives the origin of the earthquake source parameters on each line: from body wave modelling in this paper (B, shown in bold type, with B1 and B2 signifying the first and second subevents in the 1981 June 11 earthquake), or from the CMT solutions by Harvard (H) or the USGS (U). The * after the Harvard CMT depth indicates that it was fixed at 15 km in the inversion. The number in brackets after the H or U in the last column indicates the extent to which the CMT solution can be represented by a double-couple source, expressed as a percentage (y) according to the formula y = 100{1 — [(2]A2| x 1.5)/(\A;| + |A3|)]}, where 2;, 2 and 23 are the maximum, intermediate and minimum eigenvalues of moment tensor. In this (arbitrary) definition, y= 100% for a pure double-couple source (e.g. with eigenvalues —1, 0, +1) and 0 per cent for a linear vector dipole (e.g. with eigenvalues —0.5, —0.5, + 1.0). +second subevent 2.2 s after the first, offset 8.1 km in direction 256°.

Table 1 Source parameters of the main Gowk valley earthquakes. Epicentres are from Engdahl er al. (1998). Magnitudes (m, and M,) are from the USGS. An m after the M,, value indicates that two subevents were required to model the body waves. Seismic moment (Mp) is in units of 10'8 N m. TF is the duration of the time function in seconds (the time for 95 per cent of the seismic moment to be released), and sv is the slip vector azimuth, assuming that the west-dipping nodal plane is the fault plane. The last column gives the origin of the earthquake source parameters on each line: from body wave modelling in this paper (B, shown in bold type, with B1 and B2 signifying the first and second subevents in the 1981 June 11 earthquake), or from the CMT solutions by Harvard (H) or the USGS (U). The * after the Harvard CMT depth indicates that it was fixed at 15 km in the inversion. The number in brackets after the H or U in the last column indicates the extent to which the CMT solution can be represented by a double-couple source, expressed as a percentage (y) according to the formula y = 100{1 — [(2]A2| x 1.5)/(\A;| + |A3|)]}, where 2;, 2 and 23 are the maximum, intermediate and minimum eigenvalues of moment tensor. In this (arbitrary) definition, y= 100% for a pure double-couple source (e.g. with eigenvalues —1, 0, +1) and 0 per cent for a linear vector dipole (e.g. with eigenvalues —0.5, —0.5, + 1.0). +second subevent 2.2 s after the first, offset 8.1 km in direction 256°.

Related Figures (29)

Article (7 Geophysical Journal International - August 2001 The 1998 March 14 Fandoqa earthquake (Mw 6.6.) in Kerman province, Southeast Iran: Re- rupture of the 1981 Sirch...
Article (7 Geophysical Journal International - August 2001 The 1998 March 14 Fandoqa earthquake (Mw 6.6.) in Kerman province, Southeast Iran: Re- rupture of the 1981 Sirch...
Figure 1. (a) Seismicity of Iran 1964-1998, with epicentres from the catalogue of Engdahl er al. (1998). Note how the cut-off in seismicity follows the NE and E borders of Iran. The Zagros is marked by Z, the Alborz by A, the Kopeh Dagh by K, the relatively aseismic central Iran block by C, and the Lut block by L. The 1998 Fandoga epicentral region is marked with a white circle on both maps. (b) A velocity field for Iran showing how the NNE motion of Arabia relative to Asia is absorbed in Iran. The distribution of velocities within Iran is estimated from the spatial variation in the style of strain rates indicated by earthquakes (from Jackson et a/. 1995). Note the expected right-lateral shear and shortening along the eastern border of Iran. The boxed region is the area shown in Fig. 2.
Figure 1. (a) Seismicity of Iran 1964-1998, with epicentres from the catalogue of Engdahl er al. (1998). Note how the cut-off in seismicity follows the NE and E borders of Iran. The Zagros is marked by Z, the Alborz by A, the Kopeh Dagh by K, the relatively aseismic central Iran block by C, and the Lut block by L. The 1998 Fandoga epicentral region is marked with a white circle on both maps. (b) A velocity field for Iran showing how the NNE motion of Arabia relative to Asia is absorbed in Iran. The distribution of velocities within Iran is estimated from the spatial variation in the style of strain rates indicated by earthquakes (from Jackson et a/. 1995). Note the expected right-lateral shear and shortening along the eastern border of Iran. The boxed region is the area shown in Fig. 2.
Figure 2. Regional summary map of faults and shallow earthquake focal mechanisms. Mechanisms constrained by body wave inversion (various sources) are in black, dark grey spheres are Harvard CMT solutions, and light grey spheres are first motion fault plane solutions from Jackson & McKenzie (1984). Gray circles are epicentres of 1900-1963 earthquakes with M,>5.7. Sub-crustal earthquakes associated with the Makran subduction zone have been removed. Faults are from Berberian & Yeats (1999). The boxed region is the area of Fig. 3. The Nayband, Gowk and Sabzevaran faults are marked N, G and S.
Figure 2. Regional summary map of faults and shallow earthquake focal mechanisms. Mechanisms constrained by body wave inversion (various sources) are in black, dark grey spheres are Harvard CMT solutions, and light grey spheres are first motion fault plane solutions from Jackson & McKenzie (1984). Gray circles are epicentres of 1900-1963 earthquakes with M,>5.7. Sub-crustal earthquakes associated with the Makran subduction zone have been removed. Faults are from Berberian & Yeats (1999). The boxed region is the area of Fig. 3. The Nayband, Gowk and Sabzevaran faults are marked N, G and S.
Figure 3. Summary map of the faulting, topography and focal mechanisms in the Gowk Fault zone. Thick white lines are the positions of the Nayband, Gowk and Sabzevaran faults (see Fig. 2). Large white circles are the approximate epicentres of the five main earthquakes whose source parameters are discussed in this paper (Table 1). Small black circles are other earthquakes from the catalogue of by Engdahl et al. (1998) for the period 1964-1998. Lines P1—P4 are the positions of the topographic profiles in Fig. 4, and the box outlines the area of the SAR images in Fig. 12. The line A-A’ shows the location of the cross-section in Fig. 13.
Figure 3. Summary map of the faulting, topography and focal mechanisms in the Gowk Fault zone. Thick white lines are the positions of the Nayband, Gowk and Sabzevaran faults (see Fig. 2). Large white circles are the approximate epicentres of the five main earthquakes whose source parameters are discussed in this paper (Table 1). Small black circles are other earthquakes from the catalogue of by Engdahl et al. (1998) for the period 1964-1998. Lines P1—P4 are the positions of the topographic profiles in Fig. 4, and the box outlines the area of the SAR images in Fig. 12. The line A-A’ shows the location of the cross-section in Fig. 13.
Figure 4. Topographic swath profiles (500 m wide) crossing the Gowk fault zone along the lines in Fig. 3, showing the elevation contrast between the Kerman plateau in the SW, the Dasht-e-Lut desert in the NE, and the mountains either side of the Gowk fault zone. The position of the Gowk Fault is marked GF. The frontal ridge of the Shahdad thrust system is marked ST. Each profile P1—P4 is shown in black, while the other profiles are in grey in each box to show the similarities and variations between their shapes. The profiles are aligned roughly on the position of the Shahdad thrust or the eastern flank of the Abbarik mountains.
Figure 4. Topographic swath profiles (500 m wide) crossing the Gowk fault zone along the lines in Fig. 3, showing the elevation contrast between the Kerman plateau in the SW, the Dasht-e-Lut desert in the NE, and the mountains either side of the Gowk fault zone. The position of the Gowk Fault is marked GF. The frontal ridge of the Shahdad thrust system is marked ST. Each profile P1—P4 is shown in black, while the other profiles are in grey in each box to show the similarities and variations between their shapes. The profiles are aligned roughly on the position of the Shahdad thrust or the eastern flank of the Abbarik mountains.
Figure 5. Meisoseismal areas (shaded, stippled) of significant earthquakes (top) and space-time diagram (bottom) of nearly 145 years’ seismicity along the Gowk fault system. Open white teeth are on the hanging walls of thrusts. Thick lines are faults that were activated at the surface in the five main earthquakes of 1981-1998 that we discuss here. Z is Zamanabad, A is Ab-e-Garm, CF is Chahar Farsakh, F is Fandoga and J is Jowshan.
Figure 5. Meisoseismal areas (shaded, stippled) of significant earthquakes (top) and space-time diagram (bottom) of nearly 145 years’ seismicity along the Gowk fault system. Open white teeth are on the hanging walls of thrusts. Thick lines are faults that were activated at the surface in the five main earthquakes of 1981-1998 that we discuss here. Z is Zamanabad, A is Ab-e-Garm, CF is Chahar Farsakh, F is Fandoga and J is Jowshan.
Figure 6. LANDSAT TM images of the Gowk fault zone. (a) The whole central part of the fault zone, from its junction with the southern tip of the Nayband Fault in the north to the south Golbaf depression in the south. This covers the entire rupture zones of the five 1981-1998 earthquakes whose focal mechanisms are shown in Fig. 3. Note the subparallel ridges formed by anticlines above the blind faults of the Shahdad thrust system in the east. (b) Detailed image of the rupture zone of the 1998 March 14 Fandoga earthquake. Thin black lines in the Gowk valley indicate the principal coseismic surface faults (see also Fig. 7).
Figure 6. LANDSAT TM images of the Gowk fault zone. (a) The whole central part of the fault zone, from its junction with the southern tip of the Nayband Fault in the north to the south Golbaf depression in the south. This covers the entire rupture zones of the five 1981-1998 earthquakes whose focal mechanisms are shown in Fig. 3. Note the subparallel ridges formed by anticlines above the blind faults of the Shahdad thrust system in the east. (b) Detailed image of the rupture zone of the 1998 March 14 Fandoga earthquake. Thin black lines in the Gowk valley indicate the principal coseismic surface faults (see also Fig. 7).
Figure 7. (a) Detailed map of the coseismic ruptures (thick lines) in the 1998 March 14 Fandoqa earthquake (see also Fig. 6b). Thin lines are other faults that showed no surface reactivation in 1998. Relative uplift and subsidence across the 1998 ruptures are marked by + or —. Groups of arrows on the east side of the valley are earthquake-triggered landslides. Numbers in the two boxes between Fandoqa and Hashtadan show the locations of maximum horizontal/vertical slip (in cm) in 1998. The Hashtadan and Fandogqa playas are marked by horizontal dashed lines. Elevations are in metres. Note the right-lateral offset and diversion of the Hashtadan river where it crosses the Gowk fault. (b) Detail of the ruptures near Fandoqa playa, covering the area of the box marked ‘inset’ in Fig. 7(a).
Figure 7. (a) Detailed map of the coseismic ruptures (thick lines) in the 1998 March 14 Fandoqa earthquake (see also Fig. 6b). Thin lines are other faults that showed no surface reactivation in 1998. Relative uplift and subsidence across the 1998 ruptures are marked by + or —. Groups of arrows on the east side of the valley are earthquake-triggered landslides. Numbers in the two boxes between Fandoqa and Hashtadan show the locations of maximum horizontal/vertical slip (in cm) in 1998. The Hashtadan and Fandogqa playas are marked by horizontal dashed lines. Elevations are in metres. Note the right-lateral offset and diversion of the Hashtadan river where it crosses the Gowk fault. (b) Detail of the ruptures near Fandoqa playa, covering the area of the box marked ‘inset’ in Fig. 7(a).
Figure 8. Observed amplitudes of coseismic right-lateral (middle) and vertical (bottom) offsets on the Gowk fault system following the 1981 June 11 Golbaf and 1981 July 28 Sirch earthquakes (thin lines, shaded) and the 1998 March 14 Fandoga earthquake (thick lines). Offsets are in centimetres, plotted against distance along the fault measured on the map in the top panel. Where coseismic fault segments from the same earthquake overlapped, their cumulative value has been plotted. In the bottom panel, dashed lines, marked E+, indicate that the vertical sense of slip was east-side-up. Observed points at which offsets were measured have been joined by lines to distinguish the 1981 and 1998 displacements, and not to imply that the offsets varied between those points in the jagged manner indicated by the profiles. Note the large offsets from the M,, 6.6 Fandogqa (1998) earthquake compared with the much bigger (~ M,, 7.1) 1981 Sirch earthquake. Z is Zamanabad, A is Ab-e-Garm, CF is Chahar Farsakh, H in the north is Hasanabad, H in the south is Hashtadan, S is Sirch, F is Fandoqa, G is Golbaf and J is Jowshan.
Figure 8. Observed amplitudes of coseismic right-lateral (middle) and vertical (bottom) offsets on the Gowk fault system following the 1981 June 11 Golbaf and 1981 July 28 Sirch earthquakes (thin lines, shaded) and the 1998 March 14 Fandoga earthquake (thick lines). Offsets are in centimetres, plotted against distance along the fault measured on the map in the top panel. Where coseismic fault segments from the same earthquake overlapped, their cumulative value has been plotted. In the bottom panel, dashed lines, marked E+, indicate that the vertical sense of slip was east-side-up. Observed points at which offsets were measured have been joined by lines to distinguish the 1981 and 1998 displacements, and not to imply that the offsets varied between those points in the jagged manner indicated by the profiles. Note the large offsets from the M,, 6.6 Fandogqa (1998) earthquake compared with the much bigger (~ M,, 7.1) 1981 Sirch earthquake. Z is Zamanabad, A is Ab-e-Garm, CF is Chahar Farsakh, H in the north is Hasanabad, H in the south is Hashtadan, S is Sirch, F is Fandoqa, G is Golbaf and J is Jowshan.
Figure 9. Field photographs of coseismic surface ruptures and geomorphology following the 1998 March 14 Fandogqa earthquake. (e) was taker in April 1999. All the others were taken in March 1998, a few days after the 1998 Fandoga earthquake. (a) View E of west-facing, left-steppins en echelon scarps about 500 m south of the Hashtadan river. (b) View west of the east-facing scarp south of Hashtadan at 30°03.28’N 57°39.09’E near the location of maximum offset shown by the boxes in Fig. 7(a). (c) View east in the same area as 9(b). The person is standing on the faul trace, which has offset gullies ~ 2.5 m in a right-lateral sense. (d) View south of the east-facing scarp in the same region as 9(b) and 9(c). (e) View W o: a 2 m right-lateral offset in a field boundary (between the two people) in the northern part of the Fandoqa depression at 30°02.15’N 57°39.78'E (f) Overview of the Fandoga depression, looking SW from its northern end. The scarp crossing the road is up on its western side but loses this vertica component before following the western (right) side of the bluff in the distance. The location of 9e is between the bluff and the road. (g) A fielc boundary offset 1.5 m (between the white markers) in the Fandoqa playa. View W to Fandoga village in the background. (h) A shattered zone of er echelon cracks and fissures in the Fandogqa playa, on the west side of the bluff in 9(f). View south to the pass containing an earth dam (see Fig. 7a) (i) View north of a gorge cutting through the frontal anticline of the Shahdad thrust system, south of Shahdad at 30°19.16’N 57°46.01’E. Note the folding of the surface gravels and the incision of the river. (j) Detail of the west side of the gorge at 9(i), near its exit onto the Shahdad plain Small, west-dipping thrusts (which were checked, and had not moved, after the 1981 and 1998 earthquakes) are seen cutting the uplifted and foldec gravels.
Figure 9. Field photographs of coseismic surface ruptures and geomorphology following the 1998 March 14 Fandogqa earthquake. (e) was taker in April 1999. All the others were taken in March 1998, a few days after the 1998 Fandoga earthquake. (a) View E of west-facing, left-steppins en echelon scarps about 500 m south of the Hashtadan river. (b) View west of the east-facing scarp south of Hashtadan at 30°03.28’N 57°39.09’E near the location of maximum offset shown by the boxes in Fig. 7(a). (c) View east in the same area as 9(b). The person is standing on the faul trace, which has offset gullies ~ 2.5 m in a right-lateral sense. (d) View south of the east-facing scarp in the same region as 9(b) and 9(c). (e) View W o: a 2 m right-lateral offset in a field boundary (between the two people) in the northern part of the Fandoqa depression at 30°02.15’N 57°39.78'E (f) Overview of the Fandoga depression, looking SW from its northern end. The scarp crossing the road is up on its western side but loses this vertica component before following the western (right) side of the bluff in the distance. The location of 9e is between the bluff and the road. (g) A fielc boundary offset 1.5 m (between the white markers) in the Fandoqa playa. View W to Fandoga village in the background. (h) A shattered zone of er echelon cracks and fissures in the Fandogqa playa, on the west side of the bluff in 9(f). View south to the pass containing an earth dam (see Fig. 7a) (i) View north of a gorge cutting through the frontal anticline of the Shahdad thrust system, south of Shahdad at 30°19.16’N 57°46.01’E. Note the folding of the surface gravels and the incision of the river. (j) Detail of the west side of the gorge at 9(i), near its exit onto the Shahdad plain Small, west-dipping thrusts (which were checked, and had not moved, after the 1981 and 1998 earthquakes) are seen cutting the uplifted and foldec gravels.
Table 2. Details of ERS-2 data used in this study. All SAR data copyright ESA. (*) At the middle of the Fandoga fault break within this frame, the inclination angle is ~25°, and the unit vector fi pointing towards the spacecraft is i =(0.419, —0.087, 0.904) in the coordinate system (east, north, up), i.e. the spacecraft moves almost from north to south with the SAR looking to the right or west. (+) The altitude of ambiguity, the magnitude of topographic error required to cause a single interference fringe, again estimated at the centre of the fault.
Table 2. Details of ERS-2 data used in this study. All SAR data copyright ESA. (*) At the middle of the Fandoga fault break within this frame, the inclination angle is ~25°, and the unit vector fi pointing towards the spacecraft is i =(0.419, —0.087, 0.904) in the coordinate system (east, north, up), i.e. the spacecraft moves almost from north to south with the SAR looking to the right or west. (+) The altitude of ambiguity, the magnitude of topographic error required to cause a single interference fringe, again estimated at the centre of the fault.
Figure 9. (Continued.)
Figure 9. (Continued.)
Figure 10. P (top) and SH (bottom) observed (solid) and synthetic (dashed) waveforms for the 1998 March 14 Fandoga earthquake (Table 1). Station positions on focal spheres are identified by capital letters and arranged clockwise starting from north. STF is the source time function. Vertical ticks on the seismograms indicate the inversion window. Numbers beneath the header line are strike, dip, rake, centroid depth (km) and moment (N m). Stations were weighted according to azimuthal density and then the S seismogram weights were halved, to compensate for their larger amplitudes.
Figure 10. P (top) and SH (bottom) observed (solid) and synthetic (dashed) waveforms for the 1998 March 14 Fandoga earthquake (Table 1). Station positions on focal spheres are identified by capital letters and arranged clockwise starting from north. STF is the source time function. Vertical ticks on the seismograms indicate the inversion window. Numbers beneath the header line are strike, dip, rake, centroid depth (km) and moment (N m). Stations were weighted according to azimuthal density and then the S seismogram weights were halved, to compensate for their larger amplitudes.
Table 3. Source parameters of the 1998 Fandogqa earthquake from SAR and seismology. Source parameters estimated from SAR interferometry were calculated using Lamé elastic constants 4=3.22x 10! Pa and =3.43 x 10'° Pa. Column 1 shows the results from an inversion in which all parameters were free to change. In the inversion shown in column 2 the rake was fixed at 180° (pure right-lateral strike-slip). The source parameters from the seismological inversion (Fig. 10, Table 1) are shown for comparison in column 3: the slip was calculated from the moment and fault area, assuming a length from the surface ruptures and a depth equal to twice the centroid depth. The n or t after the dip-slip component indicates normal or thrust slip. The last column shows the result of a free inversion for the fringes on the Shahdad thrust. The dip of the Shahdad fault determined by SAR is given as 8°W(*) and is actually the angle between the fault and the ground surface, which slopes 2° to the east, so that the fault dip is 6°W relative to the horizontal.
Table 3. Source parameters of the 1998 Fandogqa earthquake from SAR and seismology. Source parameters estimated from SAR interferometry were calculated using Lamé elastic constants 4=3.22x 10! Pa and =3.43 x 10'° Pa. Column 1 shows the results from an inversion in which all parameters were free to change. In the inversion shown in column 2 the rake was fixed at 180° (pure right-lateral strike-slip). The source parameters from the seismological inversion (Fig. 10, Table 1) are shown for comparison in column 3: the slip was calculated from the moment and fault area, assuming a length from the surface ruptures and a depth equal to twice the centroid depth. The n or t after the dip-slip component indicates normal or thrust slip. The last column shows the result of a free inversion for the fringes on the Shahdad thrust. The dip of the Shahdad fault determined by SAR is given as 8°W(*) and is actually the angle between the fault and the ground surface, which slopes 2° to the east, so that the fault dip is 6°W relative to the horizontal.
Figure 11. Tests to check the inversion for the 1998 March 14 earthquake (Fig. 10) for sensitivity to source time function duration, rake and possible later ruptures on the Shahdad thrust system (see text). Synthetic seismograms are dashed, observed are solid lines. The first line contains seismograms at selected stations from Fig. 10, the final inversion result. P and SH focal spheres are shown, with the time function and numbers showing the strike, dip, rake, depth and moment. In the second line the time function duration was restricted to 6 s, with all other parameters left free in the inversion. In the third line the rake was fixed to 180° (pure strike-slip) with other parameters free. The resulting solution violates the clear downward first motion at LZH. If the rake is fixed to 160° (to require a thrust component of slip) many first motions are violated in the NE (see Fig. 10). The fourth line is a forward model, showing the effect of a rupture 10 s after the first motion, with the orientation of the thrust found from the SAR data beneath the Shahdad anticlines (Table 3) and a short time function of duration 4 s. The fifth line shows a forward model with the same parameters as in line 4, but with the dip steepened to 40°W.
Figure 11. Tests to check the inversion for the 1998 March 14 earthquake (Fig. 10) for sensitivity to source time function duration, rake and possible later ruptures on the Shahdad thrust system (see text). Synthetic seismograms are dashed, observed are solid lines. The first line contains seismograms at selected stations from Fig. 10, the final inversion result. P and SH focal spheres are shown, with the time function and numbers showing the strike, dip, rake, depth and moment. In the second line the time function duration was restricted to 6 s, with all other parameters left free in the inversion. In the third line the rake was fixed to 180° (pure strike-slip) with other parameters free. The resulting solution violates the clear downward first motion at LZH. If the rake is fixed to 160° (to require a thrust component of slip) many first motions are violated in the NE (see Fig. 10). The fourth line is a forward model, showing the effect of a rupture 10 s after the first motion, with the orientation of the thrust found from the SAR data beneath the Shahdad anticlines (Table 3) and a short time function of duration 4 s. The fifth line shows a forward model with the same parameters as in line 4, but with the dip steepened to 40°W.
Figure 12. (a) Observed SAR interferogram for the Fandoqa earthquake derived as described in the text. Each colour cycle, from dark blue to red, represents an increase in line-of-sight distance away from the satellite of 28 mm. The white lines are traces of the Gowk fault (west) and Shahdad fault (east). The blotches in the bottom left quadrant of the figure probably represent atmospheric disturbances. (b) Synthetic fringes predicted by an elastic dislocation model with two faults described by the parameters in columns | and 4 of Table 3.
Figure 12. (a) Observed SAR interferogram for the Fandoqa earthquake derived as described in the text. Each colour cycle, from dark blue to red, represents an increase in line-of-sight distance away from the satellite of 28 mm. The white lines are traces of the Gowk fault (west) and Shahdad fault (east). The blotches in the bottom left quadrant of the figure probably represent atmospheric disturbances. (b) Synthetic fringes predicted by an elastic dislocation model with two faults described by the parameters in columns | and 4 of Table 3.
Figure 13. Cross-sections (vertically exaggerated by 4.1 above, true scale below) along the line A—A’ in Fig. 3, showing the results of the SAR interferogram inversion (Table 3). The inversion program calcu- lates the position of the dislocation relative to sea level assuming a flat surface, whereas the SAR interferogram measures deformation at the topographic surface. The solid lines show the adjusted positions of the faults (from their original dashed positions) to allow for surface elevation. In summary, in spite of the trade-off between rake and displacement, the unconstrained SAR inversion gives a result that is consistent with the seismological inversion (Table 3). These inversions are in agreement with most of the field obser- vations also. The only problematic result is that both the SAR
Figure 13. Cross-sections (vertically exaggerated by 4.1 above, true scale below) along the line A—A’ in Fig. 3, showing the results of the SAR interferogram inversion (Table 3). The inversion program calcu- lates the position of the dislocation relative to sea level assuming a flat surface, whereas the SAR interferogram measures deformation at the topographic surface. The solid lines show the adjusted positions of the faults (from their original dashed positions) to allow for surface elevation. In summary, in spite of the trade-off between rake and displacement, the unconstrained SAR inversion gives a result that is consistent with the seismological inversion (Table 3). These inversions are in agreement with most of the field obser- vations also. The only problematic result is that both the SAR
Figure 14. Block perspective views looking west across the anticline ridges above the Shahdad thrust system towards the Abbarik mountains, made by combining digital topography with the LANDSAT TM image. In the lower picture, the SAR interferogram has been draped over the upper image, to show how the edge of the fringe pattern coincides with the frontal ridge. The salt flats of the Dasht-e-Lut are in the foreground.
Figure 14. Block perspective views looking west across the anticline ridges above the Shahdad thrust system towards the Abbarik mountains, made by combining digital topography with the LANDSAT TM image. In the lower picture, the SAR interferogram has been draped over the upper image, to show how the edge of the fringe pattern coincides with the frontal ridge. The salt flats of the Dasht-e-Lut are in the foreground.
Figure 15. Detailed map of the northern end of the Gowk fault zone, including the area affected by the 1998 November 18 earthquake near Chahar Farsakh (M,, 5.3) and the location of surface cracks observed after that earthquake (thick line). The stippled regions with arrows are old fans and their flow directions, now abandoned by the deep incision of the Shahdad river through the uplifted (west) side of the Gowk fault. Except for the Nayband fault, the transverse fault W of Deh Malshahi, the Shahdad fault and the fault SW of the Tabagqsar anticline, all other faults were reactivated at the surface during the 1981 July 28 Sirch earthquake (©, 7.1).
Figure 15. Detailed map of the northern end of the Gowk fault zone, including the area affected by the 1998 November 18 earthquake near Chahar Farsakh (M,, 5.3) and the location of surface cracks observed after that earthquake (thick line). The stippled regions with arrows are old fans and their flow directions, now abandoned by the deep incision of the Shahdad river through the uplifted (west) side of the Gowk fault. Except for the Nayband fault, the transverse fault W of Deh Malshahi, the Shahdad fault and the fault SW of the Tabagqsar anticline, all other faults were reactivated at the surface during the 1981 July 28 Sirch earthquake (©, 7.1).
Figure 16. E—W cross-section through Chahar Farsakh, showing two uplifted terraces of the Shahdad river west of the Gowk fault, and the recen travertine deposit that drapes them. The attitude of the Gowk fault at depth, and whether it has a reverse or normal component at this location, are no known, so it has been drawn as vertical (see text).
Figure 16. E—W cross-section through Chahar Farsakh, showing two uplifted terraces of the Shahdad river west of the Gowk fault, and the recen travertine deposit that drapes them. The attitude of the Gowk fault at depth, and whether it has a reverse or normal component at this location, are no known, so it has been drawn as vertical (see text).
Figure 17. Observed and synthetic seismograms for the 1981 June 11 Golbaf earthquake (Table 1). Layout and conventions are the same as i Fig. 10. Note that in this case there are two subevents used to generate the synthetic seismograms. A ‘w’ beneath the station code denotes that th waveform was hand-digitized from a WWSSN record. A ‘d’ denotes a digital station.
Figure 17. Observed and synthetic seismograms for the 1981 June 11 Golbaf earthquake (Table 1). Layout and conventions are the same as i Fig. 10. Note that in this case there are two subevents used to generate the synthetic seismograms. A ‘w’ beneath the station code denotes that th waveform was hand-digitized from a WWSSN record. A ‘d’ denotes a digital station.
Figure 18. Tests to investigate the sensitivity of the 1981 Golbaf earthquake seismograms to various source parameters (see text). The layout is the same as in Fig. 11. The first line shows selected seismograms from the solution in Fig. 17. Lines 2 and 3 are forward models showing the contribution of the first (line 2) and second (line 3) subevents to the synthetic seismograms. Line 4 shows the result of an inversion in which the strike, dip and rake are fixed to the values of the Harvard CMT ‘best-double-couple’ solution, while other parameters are free to change. Line 5 shows the result of a single-source inversion in which the strike, dip and rake fixed to the values of the first subevent in line 1, but with the depth fixed at 10 km (see text).
Figure 18. Tests to investigate the sensitivity of the 1981 Golbaf earthquake seismograms to various source parameters (see text). The layout is the same as in Fig. 11. The first line shows selected seismograms from the solution in Fig. 17. Lines 2 and 3 are forward models showing the contribution of the first (line 2) and second (line 3) subevents to the synthetic seismograms. Line 4 shows the result of an inversion in which the strike, dip and rake are fixed to the values of the Harvard CMT ‘best-double-couple’ solution, while other parameters are free to change. Line 5 shows the result of a single-source inversion in which the strike, dip and rake fixed to the values of the first subevent in line 1, but with the depth fixed at 10 km (see text).
Figure 19. Two views of the co-seismic faulting in the 1981 Sirch earthquake. (a) Two parallel ruptures, each with right-lateral offsets of ~ 10-15 cm. 10 km north of Chahar Farsakh, looking east. These were typical of the largest offsets seen in 1981, and are far smaller than the maximum coseismic offsets of ~ 300 cm seen after the Fandoga earthquake in 1998. (b) View south near the Shahdad river about 1 km east of Chahar Farsakh. The 1981 scarp, up on the eastern side, is by the vehicle. on the Gowk fault system. Ruptures in the last three earth- quakes of this sequence, in November 1989 (M,, 5.8), March 1998 (M,, 6.6) and November 1998 (M,, 5.4), precisely followed shorter portions (~ 10, 20 and 4 km respectively) of the much longer surface ruptures that formed in June and July 1981 (14 km and 65 km respectively). There is little doubt that the surface ruptures coincided in these earthquakes, as they were mapped in the field by the same people (MB and MQ) each time. It can be argued that the two smallest events of the sequence in November 1989 (M,, 5.8) and November 1998 (M,, 5.4) were too small to be unequivocally associated with rupture at depth on the faults whose surface reactivation showed only minor cracking, and that those surface cracks may Between June 1981 and December 1998 five earthquakes occurred that were associated with observed surface ruptures
Figure 19. Two views of the co-seismic faulting in the 1981 Sirch earthquake. (a) Two parallel ruptures, each with right-lateral offsets of ~ 10-15 cm. 10 km north of Chahar Farsakh, looking east. These were typical of the largest offsets seen in 1981, and are far smaller than the maximum coseismic offsets of ~ 300 cm seen after the Fandoga earthquake in 1998. (b) View south near the Shahdad river about 1 km east of Chahar Farsakh. The 1981 scarp, up on the eastern side, is by the vehicle. on the Gowk fault system. Ruptures in the last three earth- quakes of this sequence, in November 1989 (M,, 5.8), March 1998 (M,, 6.6) and November 1998 (M,, 5.4), precisely followed shorter portions (~ 10, 20 and 4 km respectively) of the much longer surface ruptures that formed in June and July 1981 (14 km and 65 km respectively). There is little doubt that the surface ruptures coincided in these earthquakes, as they were mapped in the field by the same people (MB and MQ) each time. It can be argued that the two smallest events of the sequence in November 1989 (M,, 5.8) and November 1998 (M,, 5.4) were too small to be unequivocally associated with rupture at depth on the faults whose surface reactivation showed only minor cracking, and that those surface cracks may Between June 1981 and December 1998 five earthquakes occurred that were associated with observed surface ruptures
Figure 20. Observed and synthetic seismograms for the 1981 July 28 Sirch earthquake (Table 1). Layout and conventions are the same as in Figs 10 and 17. (a) An inversion with the strike, dip and rake fixed to values from the Harvard CMT ‘best double-couple’ solution. (b) An inversion with all parameters free, using a single source.
Figure 20. Observed and synthetic seismograms for the 1981 July 28 Sirch earthquake (Table 1). Layout and conventions are the same as in Figs 10 and 17. (a) An inversion with the strike, dip and rake fixed to values from the Harvard CMT ‘best double-couple’ solution. (b) An inversion with all parameters free, using a single source.
Figure 21. Observed and synthetic seismograms for the 1989 November 20 South Golbaf earthquake (Table 1). Layout and conventions are the same as in Figs 10 and 17.
Figure 21. Observed and synthetic seismograms for the 1989 November 20 South Golbaf earthquake (Table 1). Layout and conventions are the same as in Figs 10 and 17.
Figure 22. E-W cartoon cross-sections to illustrate possible fault geometries in and beneath the Gowk valley and Shahdad thrust system (see text) Velocity triangles to the right of cartoons (b) to (e) illustrate the relative motions of fault-bounded blocks A-C. In the text, faults are referred to by the blocks they bound; thus ‘fault AB’ is the fault bounding blocks A and B.
Figure 22. E-W cartoon cross-sections to illustrate possible fault geometries in and beneath the Gowk valley and Shahdad thrust system (see text) Velocity triangles to the right of cartoons (b) to (e) illustrate the relative motions of fault-bounded blocks A-C. In the text, faults are referred to by the blocks they bound; thus ‘fault AB’ is the fault bounding blocks A and B.
Figure 23. Cartoon cross-section from Berberian et al. (1984) to illustrate their suggestion that the depressions in the Gowk valley were formed by subsidence in the footwalls of two opposing thrusts. This suggestion is not compatible with the normal faulting component inferred from seismological and SAR data after the 1998 Fandoqa earthquake. More problematic than the sense of throw 1s that some ruptures observed in 1998 and 1981 had an apparent reverse sense of displacement, on steeply east- or west-dipping faults. The significance of this apparent shortening is not clear and unambiguous exposures of the faults in cross-section were rare. In 1998 only two were visible, exposed for a depth of about a metre at the base of a steep scarp, dipping west at 65° and 78° between Hashtadan and Fandoqa. At these two places the local strike of the ruptures was 10°—20° anticlockwise from the overall strike of 156°, and the shortening may be a localized restraining-bend effect. In 1981 the vertical offsets were generally smaller (Fig. 8), but clearly involved shortening in some places (see photos in Figs 11 and 12 of Berberian et al. 1984). It is difficult to know how to interpret these observations. The 1981 fractures in particular were often very complex, involving distri- buted fissuring and cracking over a relatively wide zone, and normal faults were also seen, sometimes associated with thrusts (Fig. 14 in Berberian et al. 1984). Some of the apparent shorten- ing may have been related to minor irregularities in the trend of the fault zone or to local topographic effects steepening faults at shallow depth adjacent to older scarps. These surface obser- vations in 1981 greatly influenced our tectonic interpretation in Berberian et al. (1984), causing us to suggest that the Gowk valley was bounded by outward-dipping reverse faults (Fig. 23),
Figure 23. Cartoon cross-section from Berberian et al. (1984) to illustrate their suggestion that the depressions in the Gowk valley were formed by subsidence in the footwalls of two opposing thrusts. This suggestion is not compatible with the normal faulting component inferred from seismological and SAR data after the 1998 Fandoqa earthquake. More problematic than the sense of throw 1s that some ruptures observed in 1998 and 1981 had an apparent reverse sense of displacement, on steeply east- or west-dipping faults. The significance of this apparent shortening is not clear and unambiguous exposures of the faults in cross-section were rare. In 1998 only two were visible, exposed for a depth of about a metre at the base of a steep scarp, dipping west at 65° and 78° between Hashtadan and Fandoqa. At these two places the local strike of the ruptures was 10°—20° anticlockwise from the overall strike of 156°, and the shortening may be a localized restraining-bend effect. In 1981 the vertical offsets were generally smaller (Fig. 8), but clearly involved shortening in some places (see photos in Figs 11 and 12 of Berberian et al. 1984). It is difficult to know how to interpret these observations. The 1981 fractures in particular were often very complex, involving distri- buted fissuring and cracking over a relatively wide zone, and normal faults were also seen, sometimes associated with thrusts (Fig. 14 in Berberian et al. 1984). Some of the apparent shorten- ing may have been related to minor irregularities in the trend of the fault zone or to local topographic effects steepening faults at shallow depth adjacent to older scarps. These surface obser- vations in 1981 greatly influenced our tectonic interpretation in Berberian et al. (1984), causing us to suggest that the Gowk valley was bounded by outward-dipping reverse faults (Fig. 23),
Related topics:
GeologyGeophysicsActive TectonicsGeomatic EngineeringContinental Tectonics
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved