Author manuscript,
in "Geophysical Journal International 189 (2012) 6-18"
Geophysical
Journal published
International
DOI : 10.1111/j.1365-246X.2012.05365.x
Fo
rP
ev
rR
ee
w
ie
hal-00681707, version 1 - 14 May 2012
Page 1 of 28
New evidence for large earthquakes on the Central Iran plateau: Paleoseismology
2
of the Anar fault
3
Foroutan1,2,3 M., M. Sébrier2,3, H. Nazari4, B. Meyer2,3, M. Fattahi5,6,7, A. Rashidi8, K. Le
4
Dortz2,3,* and M. D. Bateman6
5
6
1: Geological Survey of Iran, Azadi square, Meraj avenue, PO Box: 13185-1494, Tehran, Iran
email: foroutan@gsi.ir
7
8
2: UPMC Univ Paris 06, ISTEP UMR 7193, Université Pierre et Marie Curie, F-75005, Paris, France
email: mohammad.foroutan@upmc.fr, bertrand.meyer@upmc.fr, michel.sebrier@upmc.fr,
9
3: CNRS, ISTEP, UMR 7193, F-75005, Paris, France
10
11
4: Research Institute for Earth Sciences, Geological Survey of Iran, Po Box: 13185-1494,Tehran, Iran
12
13
5: The Institute of Geophysics, University of Tehran, Tehran, Iran
14
15
16
6: Sheffield Centre for International Dryland Research, Department of Geography, University of
17
18
7: The School of Geography, University of Oxford, OX1 3QY, UK
19
20
21
22
8: Geological Survey of Iran, Postal Code 7615736841, Kerman, Iran
email: h.nazari@gsi.ir
email: mfattahi@ut.ac.ir
email: morteza.fattahi@ouce.ox.ac.uk
ev
email: kermangeo@gsi-iran.org
rR
ee
Sheffield, Winter Street, Sheffield S10 2TN, UK
email: M.D.Bateman@Sheffield.ac.uk
* Now at: Laboratoire de Géologie, ENS, UMR 8538, 75231 Paris, ledortz@geologie.ens.fr
Abstract
iew
23
24
rP
hal-00681707, version 1 - 14 May 2012
1
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
25
The Central Iran plateau appears aseismic during the last few millenniums based on
26
instrumental and historical seismic records. Nevertheless, it is sliced by several strike-slip
27
faults that are hundreds kilometres-long. These faults display along-strike, horizontal offsets
28
of intermittent gullies that suggest the occurrence of earthquakes in the Holocene.
29
Establishing this is crucial for accurately assessing the regional seismic hazard. The first
30
paleoseismic study performed on the 200-km-long, NS striking Anar fault shows that this
31
right-lateral fault hosted three large (Mw§ earthquakes during the Holocene or possibly
32
Uppermost Pleistocene for the older one. These three seismic events are recorded within a
33
sedimentary succession, which is not older than 15 ka, suggesting an average recurrence of at
Geophysical Journal International
most 5 ka. The six OSL ages available provide additional constraints and allow estimating
35
that the three earthquakes have occurred within the following time intervals: 4.4±0.8 ka,
36
6.8±1 ka, and 9.8±2 ka. The preferred age of the more recent event, ranging between 3600 yrs
37
and 5200 yrs, suggests that the fault is approaching the end of its seismic cycle and the city of
38
Anar could be under the threat of a destructive earthquake in the near future. Additionally, our
39
results confirm a previous minimum slip rate estimate of 0.8±0.1 mm/yr for the Anar
40
fault indicating the westernmost prominent right‐lateral faults of the Central Iran
41
plateau are characterized by slip rates close to 1 mm/yr. These faults, which have
42
repeatedly produced destructive earthquakes with large magnitudes and long
43
recurrence interval of several thousands of years during the Holocene, show that the
44
Central Iran plateau does not behave totally as a rigid block and that its moderate
45
internal deformation is nonetheless responsible for a significant seismic hazard.
46
Introduction
ee
47
rP
48
The Central Iran Plateau is a wide region experiencing low GPS deformation rates
49
and is commonly described as a rigid block (e.g., Jackson & McKenzie, 1984; Vernant et
50
al., 2004). The region (Figure 1) is nonetheless sliced by several strike‐slip faults with
51
clear morphological traces (Walker & Jackson, 2004; Meyer et al., 2006; Meyer & Le
52
Dortz, 2007) that contrast with the very few earthquakes recorded in the region
53
(Ambraseys & Jackson, 1998). Destructive earthquakes have occurred close to or along
54
the Lut faulted borders only, and, according to the historical and instrumental records
55
(Ambraseys & Melville, 1982; Ambraseys & Jackson, 1998), the prominent right‐lateral
56
strike‐slip faults inland remained quiescent for millenniums. Although the absence of
57
historical record of earthquakes in remote and uninhabited desert does not mean an
58
absence of earthquakes, knowledge of the behaviour of such faults and assessment of
59
the regional seismic hazard requires paleoseismic studies and depends on the selection
60
of suitable trenching sites. This is the case for the region of the Dehshir and Anar faults.
61
While a recent paleoseismic study stated the occurrence of large and infrequent
62
earthquakes on the Dehshir fault (Nazari et al., 2009; Fattahi et al., 2010), the seismic
63
behaviour of the Anar fault is still to be assessed.
iew
ev
rR
hal-00681707, version 1 - 14 May 2012
34
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 2 of 28
Page 3 of 28
The Anar fault is a 200‐km‐long NS strike‐slip fault located nearby 55°E, in the
65
middle of the Central Iran Plateau (Figure 1). The northern part of the fault is located
66
within a mountainous region where several closely spaced splays cut across the Kuh-e
67
Kharanaq range. The southern portion of the fault runs along the Kuh-e Bafq range and cuts
68
right across the western piedmont of the range and across the Anar salt flat. The fault goes
69
through the populated city of Anar and further bends eastwards to reactivate a thrust fault
70
to the south. According to
71
luminescence (OSL) dating of cumulative offset of alluvial fans, the southern portion of the
72
fault slips at a minimum rate of 0.8 mm/year (Le Dortz et al., 2009). Both the sharpness
73
of these cumulative offsets and the absence of along fault microseismicity suggest these
74
offsets have accrued through large and infrequent earthquakes rather than by creeping.
75
At a few places, the offset of small gullies that ranges between 2.5 and 3.5 m (Figure 2)
76
may be interpreted as the amount of coseismic slip during the last earthquake but clear
77
evidence for a continuous and distinctive surface break is missing. Trenching appears
78
therefore necessary to document earthquakes and access the Recent seismic history of
79
the fault.
rR
81
Be cosmic ray exposure (CRE) and optically-stimulated
ee
80
10
rP
Site selection and trench stratigraphy
ev
82
We scrutinized the fault on Quickbird satellite imagery and in the field to select
83
the most favourable place to conduct paleoseismic investigations. The selected site is
84
located 35 km north of the city of Anar along a very clear portion of the N175°E fault
85
trace that is highlighted by an E‐facing scarp (Figure 3). This fault scarp is readily seen
86
cutting across numerous rills and ephemeral streams that incise a fan surface
87
characterized by a subdued bar‐and‐swale morphology. The abandoned alluvial fan
88
surface comprises a loose desert pavement of varnished clasts separated by sandy‐silty
89
material. The smooth scarp has a minimum height of 35 cm near the excavation (Figure
90
3b). South of the trench site, the scarp height increases progressively to reach one meter
91
where small streams, which have been dammed and channelled along the fault,
92
rejuvenate this scarp. North of the trench site, the scarp height also increases, but the
93
streams are wider and deeper than to the south and have been able to maintain their
94
courses to flow across the fault trace. At the trench site, the possibility of damming small
95
intermittent streams during the emplacement of the fan material is high and the
iew
hal-00681707, version 1 - 14 May 2012
64
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
Geophysical Journal International
possibility of subsequent erosion low, maximizing the chances to record the earthquakes
97
coeval with alluvial aggradation and subsequent colluvial deposition. The next
98
paragraphs report the observations gathered at the selected site, where trenching
99
allowed us to distinguish three unambiguous events.
100
The eastern tip of the excavated trench is located about 5 km west of the Kuh‐e
101
Bafq range at 31.1953°N; 55.1536°E, and 1544‐m‐altitude above sea level (Figures 1 and
102
3b). The trench strikes N85°E, it has a 28‐m‐length, a depth between 4.3 and 4.7 m, and
103
a width of some 1.5 m (Figures 3, 4, and 5). Trench walls expose a total deposit thickness
104
of 5.9 m with fairly flat, 0.1‐to 1‐m‐thick beds that correspond to medium‐distal, alluvial
105
fan facies. Trench stratigraphy is relatively straightforward as these beds are continuous
106
and easily correlated across the fault zone located approximately in the middle of the
107
trench (Figure 5). Overall the beds found in the trench are composed of sands and
108
gravels mixed with variable amounts of silt and clay, these deposits correspond mostly
109
to a debris‐flow dominated alluvial fan. Although debris‐flows and sheet‐floods appears
110
to dominate, sediment dynamics is not easy to determine precisely for each bed due to
111
the medium‐distal deposit conditions.
rR
ee
rP
112
A total of 17 units have been defined in the excavated trench (see Table 1 and
113
Figures 4 and 5). The uppermost part of the trench, approximately the last meter, is
114
made of the thinner units 13 to 17 containing smaller pebbles than the alluvial units 1 to
115
8 and an almost clay‐free matrix contrary to the units 9 to 11. This last meter hence is
116
made of runoff deposits. The only evidence for a significant break in alluvial
117
sedimentation is the γ0.15‐m‐thick, gypsiferous calcrete that is located toward the top of
iew
ev
hal-00681707, version 1 - 14 May 2012
96
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
118
unit 11. This latter unit and units 9 and 10 correspond to reddish brown, mud‐
119
dominated debris flows. OSL was used to date the alluvial layers seen in the trench.
120
121
OSL dating
122
Six lenses of fine sandy‐silts intercalated between fanglomerates at various depths
123
below the surface were sampled for OSL in opaque tubes (Figure 5). The Single Aliquot
124
Regeneration (SAR) protocol (e.g.; Murray and Wintle, 2000) was employed for the
125
Equivalent dose (De) measurement once quartz had been extracted and cleaned from each
126
sample. The analytical procedures employed are identical to that applied to similar samples
Page 4 of 28
Page 5 of 28
from the neighbouring Sabzevar, Doruneh, and Dehshir faults (Fattahi et al, 2006; 2007;
128
2010; Le Dortz et al., 2011). In order to make all data consistent, three samples with ages that
129
were previously published in Le Dortz et al. (2009), including one sample collected
130
approximately 8 km farther north (OSL-2 on Figure 1 and in Table 2), have been refined in
131
the light of procedural development outlined by Fattahi et al. (2010). Table 2 provides the
132
relevant information for OSL ages in years from present (2010) with 1 sigma errors.
133
Initial attempts to use single grains for De determination failed due to the dimness of
134
OSL signal. As a result, De measurements were undertaken on 9.6-mm-diameter aliquots
135
containing approximately 1500-2000 grains. Whilst normally this might result in averaging
136
out of any multiple dose component within a sample, here it is assumed that for these dim
137
samples the luminescence signal from each aliquot was produced by a relatively few number
138
of bright grains and thus may be considered as almost measuring at single grain level.
139
Therefore, the De distribution of the single aliquot De measurements is considered to be
140
almost a true reflection of the actual De distribution within a sample. For samples showing
141
unimodal, apparently normally distributed De’s with a low over-dispersion (Ant-I and Ant-IV
142
in Table 2), which are interpreted as having had the OSL signal reset (bleached) prior to
143
burial, Central Age Model (CAM) was employed for calculation purposes. For the remaining
144
samples (Ant-II, Ant-III, Ant-V, Ant-VI and OSL-2 in Table 2), the depositional setting, field
145
sedimentary logs, and the scatter of the replicate aliquot De data indicated that prior to burial,
146
full resetting (bleaching) of the OSL signal had not taken place and/or that the sediments had
147
undergone some post-depositional disturbance (Bateman et al., 2007). Indeed, the small size
148
of the catchment area at the trench site, less than 20 km2, is indicative of a rapid transport
149
before the emplacement of the fan (Le Dortz et al., 2009). The great majority of the Anar
150
trench deposits come from high-discharge depositional events with limited surface exposure,
151
which favour partial bleaching (Rittenour, 2008). As a consequence, Finite Mixture Model
152
(FMM; Roberts et al. 2000) was used where samples showed skewed, scattered, or
153
multimodal distributions and the dominant De component was used for age calculations (as in
154
Fattahi et al., 2010). This approach yielded ages in accordance with site stratigraphy except
155
for sample Ant-III whose age was too young. This sample has exhibited the highest over-
156
dispersion value (51%, Table 2) of all measured samples and three De component were
157
extracted by FMM for this sample, the highest representing 30% of the signal and the smallest
158
47%. It is possible to get an age that fits with stratigraphy using the higher two De
159
components extracted using FMM, corresponding to ages of 7.3±0.5 ka and 10.7±0.8 ka,
iew
ev
rR
ee
rP
hal-00681707, version 1 - 14 May 2012
127
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
Geophysical Journal International
respectively. However, there are no good reasons to accept components representing only
161
23% or 30% of the data over the 47% of the data incorporated into the dominant smallest
162
FMM component. The latter cannot be ignored on the basis of partial bleaching as it has a
163
lower De and is unlikely to represent post-depositional disturbance as it is the dominant De
164
component but also based on observed bedding within the unit. However, unlike most other
165
samples studied, Ant-III was sampled actually on a stratigraphic boundary (between units 8
166
and 9). As the dose-rate for this sample is based only on the activity from unit 9, a difference
167
in the gamma radiation dose received from unit 8 might help explain the apparent under-
168
estimation of age based on the lowest FFM De component. Unfortunately field-based gamma-
169
spectrometer readings or material for analyzing the radioactive elements within unit 8 were
170
not available to test this and so the age of Ant-III has not been included within the subsequent
171
site interpretation. Thus, the ages of the trench units have been constrained based on only five
172
OSL ages (Ant-I, Ant-II, Ant-IV, Ant-V, and Ant-VI).
173
174
Ages of trench units
ee
rP
175
The samples collected within the trench define a time range spanning from 5.8 to
176
14.9 ka. The oldest age Ant‐I (13.6±1.3 ka) stands at the bottom of the trench and
177
belongs to one of the oldest stratigraphic units (unit 2), it indicates that all the trench
178
units should be younger than 15 ka. As the youngest OSL sample Ant‐VI (6.2 ± 0.4 ka) is
179
located at 0.8 m below the surface (unit 14), these ages define a maximum time interval
180
between 0 and 14.9 ka and a minimum one between 6.6 and 12.3 ka, indicating the
181
trench deposits aggraded during the uppermost Marine Isotopic Stage 2 (MIS‐2γͳʹ‐
182
22 ka) and part, or the entirety of MIS‐1. The age of the fan surface at the trench site
183
appears poorly constrained between 0 and 6.6 ka.
iew
ev
rR
hal-00681707, version 1 - 14 May 2012
160
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
184
A proxy to the surface age may be obtained by estimating average sedimentation
185
rates. Assuming this fan surface is still active, the whole 5.9 m of sedimentary units seen
186
in the trench should have been aggraded during the last 14.9 or 12.3 ka at an overall
187
sedimentation rate ranging between 0.4 and 0.48 mm/yr. Deposit thicknesses above the
188
samples Ant‐I (13.6±1.3 ka) and Ant‐II (12.4±0.6 ka) yield slightly lower rates ranging
189
between 0.28 and 0.36 mm/yr. Conversely, if one assumes the top of the fan surface was
190
abandoned just after 6.6 ka, the oldest possibility for the youngest sample (Ant‐VI), the
Page 6 of 28
Page 7 of 28
sedimentation rate may have reached 1 mm/yr. These two extreme hypotheses raise the
192
problem of the abandonment age of the alluvial fan at the precise location where the
193
trench was excavated. The refined age of OSL‐2 (10.1±0.6 ka versus 11.8 ±6.5 ka in Le
194
Dortz et al., 2009), which is sampled some 8 km north of the trench (see location on
195
Figure 1, see also Figures 4a and 7 in Le Dortz et al., 2009) at 0.8 m below the surface of
196
197
γͶǤͷ , suggests its surface was
198
be older than 6.6 ka. These ages are in agreement with the late Pleistocene and Holocene
199
regional climate scenario, for the central and eastern Iran, as recently proposed by
200
Walker & Fattahi (2011). In fact, the satellite image in the vicinity of the trench site
201
(Figure 3a) shows this location corresponds to an area where the fan surface has been
202
exceptionally preserved from the backward erosion of deeply incised dry streams.
203
Therefore, the late aggradation on the fan surface at the trench site (1544 m a.s.l.),
204
resulting from surface runoff, lasted until more recently than some 8 km further north
205
on a more proximal alluvial fan (1748 m a.s.l.) where Le Dortz et al. (2009) collected
206
the sample OSL‐2. In conclusion, the best estimate for the aggradation rate of the trench
207
sediments is provided by the depth difference of 3.3 m between Ant‐I (13.6 ± 1.3 ka) and
208
Ant‐VI (6.2 ± 0.4 ka) samples, this rate ranges between 0.36 and 0.58 mm/yr.
209
Interestingly, this rate is comparable to the late Pleistocene‐Holocene aggradation rates
210
derived from alluvial sediments in similar deposition settings at southwestern Nebraska,
211
Negev desert and Central Iran plateau (Daniels et al., 2003; Guralnik et al., 2011;
212
Schmidt et al., 2011). Consequently, the best averaged net sedimentation rate, if
213
meaningful in such environment, amounts to 0.47±0.11 mm/yr suggesting that the final
214
aggradation of the trench units ended between 3.6 and 5.2 ka to the west of the fault
215
zone and approximately one thousand years later to the east of the fault zone. The
216
abandonment age of the fan surface at the trench site (4.4±0.8 ka) is comparable with
217
late Holocene drought cycles at 4.2 ka, proposed by Staubwasser et al. (2003) to the
218
southeast of the Persian Gulf.
abandoned some 9‐10 ka ago at this site while the fan surface at the trench site cannot
iew
ev
rR
ee
rP
hal-00681707, version 1 - 14 May 2012
191
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
219
220
Seismic event identification
221
The surface geomorphology, particularly, the characteristics of the fault scarp,
222
detailed geological and structural analyses of the trench walls (bed unconformities,
Geophysical Journal International
sealed fault strands, and fissure fills or sand‐blows), and restoration of the trench log
224
permit to trace three individual event horizons that correlate with three destructive past
225
earthquakes, which ruptured the ground surface. The evidence for these three seismic
226
events, which are labelled A, B, and C from the youngest to the oldest, is presented
227
below.
228
Event A, the youngest event is responsible for the 35‐cm‐high, E‐facing scarp,
229
which is observed at the surface in the vicinity of the trench (Figures 3b and 3c). The
230
trench log shows this scarp is located above a nearly 1‐m‐wide, steep faulted zone
231
associated with an E‐facing flexure of the upper trench units; i.e., downthrown to the
232
east (Figure 5). Units 16 and 17 are not observed west of the fault zone nor warped
233
eastward. They rest unconformably on unit 15 dipping about 25°E (Figure 6), hence
234
indicating the faulting and flexuring deformation postdate unit 15 and predate units 16
235
and 17. As there are no significant thickness differences of the upper trench units on
236
both sides of the fault zone, the faulting and flexuring deformation appear to result from
237
a sudden and discrete slipping event on the fault zone instead of a continuous shear; i.e.,
238
creep slip (Figure 7‐step 1). Inasmuch as the upper trench units are seen faulted up to
239
unit 15, the discrete slipping event should relate to an earthquake that ruptured the
240
surface producing the scarp. The event horizon of this earthquake locates on the top
241
boundary of unit 15, so that it stands between this latter and units 16 and 17 to the east
242
of the fault and corresponds approximately with the present‐day topographic surface to
243
the west of the fault. This suggests that the western block has been slightly eroded as
244
well as partly concealed by units 16 and 17 so that its present‐day height
245
underestimates its original height at the time of the earthquake (Figure 7‐step 2). Then,
246
unit 17 is a local sag pond deposit caused by event A.
iew
ev
rR
ee
rP
hal-00681707, version 1 - 14 May 2012
223
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
247
Backstripping of event A suggests its vertical throw is of some 60 cm (Figure 7‐
248
step 3), confirming that the 35‐cm high scarp at the surface has been partly degraded.
249
Taking into account the Anar fault is dominantly a strike‐slip fault (Figure 2), the 60 cm
250
of vertical displacement should be associated with at least a couple of meters of right‐
251
lateral displacement. Then, the magnitude of event A should be at least on the order of
252
Mw 7. The age of event A is poorly constrained as the uppermost OSL sample (Ant‐VI),
253
which is at 0.8‐m‐depth, comes from unit 14. This indicates only that event A is younger
254
than 6.6 ka. Tentatively, the sediment thickness, which stands between the sample Ant‐
Page 8 of 28
Page 9 of 28
VI and the event horizon of event A, may be used to propose a rough estimate of the
256
youngest possible age for this seismic event based on the aggradation rate between Ant‐
257
VI and Ant‐I. Since aggradation of unit 15 ended between 3600 and 5200 years (see
258
previous section) event A should not be younger than 3600 years and might be as old as
259
5200 years. This time interval will be considered as the preferred age of event A in the
260
following discussion.
261
Event B, the penultimate event, is easily determined from analysis of the trench
262
walls. Several filled fissures (unit 12), which postdate unit 11 and predate unit 13, are
263
seen (Figures 5 and 7‐step 4). A careful analysis of these filled fissures permits to
264
conclude they correspond indeed to either open cracks or sand‐blows (Figures 8a and
265
8b). It is possible to observe these features formed suddenly as they disrupt the hard
266
gypsiferous calcrete (e.g., Figure 8b), which is located in the top part of unit 11. Thus, the
267
interface between units 11 and 13 corresponds to the event horizon of event B. The
268
backstripping analysis of the trench log (Figure 7), removing the effects of event A and
269
the units aggraded between events A and B, allow reconstructing the event B horizon
270
just after the penultimate earthquake (Figure 7‐step 4). This reconstruction suggests
271
that the rupture of event B should have occurred on two parallel fault strands of the
272
Anar fault zone, totalling a vertical throw of some 40 cm, distributed into 25 cm on the
273
eastern strand and 15 cm on the western one. Thus, the penultimate earthquake appears
274
to have a vertical throw close to that of the last one, hence a magnitude lightly
275
comparable with the one of event A. The age of event B is well constrained by three OSL
276
ages (Ant‐V, ‐IV, and ‐VI) and bracketed between 5.8 ka and 7.8 ka yielding an average
277
age of 6.8±1 ka for this earthquake (Figure 9).
iew
ev
rR
ee
rP
hal-00681707, version 1 - 14 May 2012
255
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
278
Event C, the oldest seismic event, can be safely identified on the trench walls.
279
Evidence for a third earthquake rests on the facts that lower trench units (5 to 8) exhibit
280
higher vertical throws than middle ones (9 to 11). Different reconstructions have been
281
tested to restore the trench lower units, the best fit is obtained when considering the
282
interface between units 8 and 9 as the event horizon of a surface rupturing earthquake
283
(Figure 7‐step 6). This reconstruction favours that event C ruptured along the eastern
284
fault strand with a vertical displacement of about 25 cm. Nevertheless, this vertical
285
displacement is poorly constrained and higher estimates might be obtained according to
286
the amount of erosion chosen (dashed line on Figure 7‐step 6). Even if this vertical
Geophysical Journal International
throw is poorly determined and appears smaller than the ones of the subsequent
288
earthquakes, this does not imply a much smaller magnitude as the amount of coseismic
289
vertical displacement is highly variable along a strike‐slip fault (e.g., Barka, 1996; Barka
290
et al., 2002). The age of event C is not well constrained, it occurred prior to the OSL age
291
of sample Ant‐V (7.1±0.7 ka) and after that of Ant‐II (12.4±0.6 ka) so that it is bracketed
292
within the interval between 6.4 and 13 ka, with a minimum interval of 4 ka (Figure 9).
293
We have investigated the possibility for a fourth event in the lowermost part of the
294
trench (Figure 5). However, the lowermost units (1 to 4) cannot be followed on the
295
eastern side of the trench precluding the possibility to document an additional event.
296
297
Discussion and Conclusions
rP
298
Due to the reduced size of the surface sag pond, a 3D trench exploration was not
299
undertaken at the trench site so that the slip per event was not directly measurable for
300
the three identified seismic events, A, B, and C. Accounting for the lack of morphological
301
segmentation along the Anar fault (Le Dortz et al., 2009) and the excavation of a single
302
trench, there are no possibilities to calculate confidently any magnitude. The restored
303
vertical throws per event suggest only that the related magnitudes are more likely on
304
305
the order of Mwγ (see previous section). The right‐lateral offsets observed at the
306
streams are seen offset by 8±0.5 m (Le Dortz et al., 2009 and Figure 3a) while several
307
weakly incised small dry gullies show offsets of 3±0.5 m only (Figure 2). Considering
308
these lower offsets were more likely formed during the most recent earthquake, they
309
provide the best estimate for the coseismic horizontal slip of event A. Such a coseismic
310
slip at the surface agrees with a magnitude close to 7. Thus the higher offsets, which are
311
only observed where the alluvial surface is older, should represent the cumulative
312
horizontal slip of the last three events that are seen in the trench. Therefore, the three
313
events, which are observed within the trench to the north of Anar city, should have had
314
slip per event on the order of 3 m, suggesting they are of similar magnitudes.
ev
rR
ee
surface (Figures 2 and 3a) permit to reinforce this inference. Deeply incised intermittent
iew
hal-00681707, version 1 - 14 May 2012
287
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 10 of 28
315
Overall, the new data presented in this paper give evidence for at least three
316
seismic events of similar sizes on the Anar fault (Figure 9). OSL age of Ant‐I sample
317
indicates that the aggradation of the trench units did not started much before 14.9 ka or
Page 11 of 28
to the latest by 12.3 ka. Since only 3 events have occurred during the last 15 ka, this
319
provides a rough estimate of at most 5 ka for the average maximum time interval
320
between two earthquakes. The youngest age possibility may reduce this average time
321
322
interval between two earthquakes γͶ ka. Considering the best estimates of average
323
respectively (Figure 9), the time interval between two subsequent earthquakes is ill‐
324
defined and might vary significantly. Nevertheless, the average ages for events A, B, and
325
C indicate that the intervals between two subsequent earthquakes should be of 2400
326
and 2900 years between events A‐B and B‐C, respectively. As the elapsed time since the
327
last earthquake is 3600 years at least and 5200 years at most, this suggests we are
328
getting close to the end of the seismic cycle and may anticipate a destructive earthquake
329
for the Anar city in the near future.
ages for the three seismic events A, B, and C are: 4.4±0.8 ka, 6.8±1 ka, and 9.7±3.3 ka,
rP
330
Finally, the cumulative offset of 8±0.5 m (Le Dortz et al., 2009) postdating fan
331
aggradation and the refined age of fan abandonment given by OSL‐2 sample
332
(10.1±0.6 ka) confirms a minimum slip rate estimate of 0.8±0.1 mm/yr for the Anar
333
fault. Then, the westernmost prominent right‐lateral faults of the Central Iran plateau,
334
namely the Dehshir and Anar faults, which are active though void of historical and
335
instrumental earthquakes, are characterized by slip rates close to 1 mm/yr (Le Dortz et
336
al., 2009 and 2011). Such faults have repeatedly produced destructive earthquakes with
337
large magnitudes (Mwγ) and long recurrence interval of several thousands of years
ev
rR
ee
338
during the Holocene (Nazari et al., 2009; Fattahi et al., 2010 and this paper). This
339
demonstrates that the Central Iran plateau does not behave totally as a rigid block and
340
that its moderate internal deformation is nonetheless responsible for a significant
341
seismic hazard.
iew
hal-00681707, version 1 - 14 May 2012
318
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
342
343
Acknowledgements
344
This study benefited of the logistic and financial assistance from Geological
345
Survey of Iran. Université Pierre et Marie Curie and INSU‐CNRS provided
346
complementary funding for the fieldwork and OSL measurements. M. Foroutan
347
acknowledges a grant from the French Embassy in Tehran for part of his PhD thesis and
348
complementary support from UPMC‐ISTeP. David Lambert, Attaché Scientifique et
Geophysical Journal International
Culturel, is thanked for his continuing support to the cooperation between UPMC and
350
GSI. M. Hosseini from Kerman GSI office helped with the logistics of the fieldwork. B.
351
Oveisi, M. Nazem Zadeh, A. Agha Hosseini, and M.A. Shokri are thanked for various
352
contributions to the fieldwork, H. Bani Assadi and M. Adhami for their safe driving in the
353
field. M. Fattahi would like to thank the research department of the Tehran University.
354
We thank two anonymous reviewers for helpful and constructive comments.
355
356
References
357
Ambraseys, N. & Melville, C., 1982. A history of Persian earthquakes. Cambridge
358
University Press.
rP
359
Ambraseys, N.N. & Jackson, J.A, 1998. Faulting associated with historical and recent
360
earthquakes in the Eastern Mediterranean region. Geophys. J. Int. , 133, 390‐406.
361
Barka, A.A., 1996. Slip distribution along the North Anatolian Fault associated with the
362
large earthquakes of the period 1939 to 1967. Bull. Seismol. Soc. Am., 86 (5), 1238–
363
1254.
rR
ee
364
Barka, A.A., Akyüz, H.S., Altunel, E., Sunal, G., Çakir, Z., Dikbas, A., Yerli, B., Armijo, R.,
365
Meyer, B., De Chabalier, J.B., Rockwell, T., Dolan, J.R., Hartleb, R., Dawson, T.,
366
Christofferson, S., Tucker, A., Fumal, T., Langridge, R., Stenner, H., Lettis, W.,
367
Bachhuber, J. & W. Page, 2002. The Surface Rupture and Slip Distribution of the 17
368
August 1999 Izmit Earthquake (M 7.4), North Anatolian Fault. Bull. Seismol. Soc. Am.,
369
92, 43–60.
iew
ev
hal-00681707, version 1 - 14 May 2012
349
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 12 of 28
370
Bateman, M.D., Boulter, C.H., Carr, A.S., Frederick, C.D., Peter, D. & M. Wilder, 2007.
371
Preserving the palaeoenvironmental record in Drylands: Bioturbation and its
372
significance for luminescence derived chronologies. Sedimentary Geology, 195(1‐2),
373
5‐19.
374
375
376
Daniels, J.M., 2003. Floodplain aggradation and pedogenesis in a semiarid environment.
Geomorphology, 56, 225‐242.
Fattahi, M., Walker, R., Hollingsworth, J., Bahroudi, A., Nazari, H., Talebian, M., Armitage,
Page 13 of 28
S., Stokes, S., 2006. Holocene slip‐rate on the Sabzevar thrust fault, NE Iran,
378
determined using Optically‐stimulated Luminescence (OSL). Earth and
379
Planetary Science Letters, 245, 673‐684.
380
Fattahi, M., Walker, R.T., Khatib, M.M., Dolati, A. & A. Bahroudi, 2007. Slip‐rate estimate
381
and past earthquakes on the Doruneh fault, eastern Iran. Geophy. J. Int., 168,
382
691‐709.
383
Fattahi, M., Nazari, H., Bateman, M.D., Meyer, B., Sébrier, M., Talebian, M., Le Dortz, K.,
384
Foroutan, M., Ahmadi Givi, F. & M. Ghorashi, 2010. Refining the OSL age of the
385
last earthquake on the Dheshir fault, Central Iran. Quaternary Geochronology, 5,
386
286‐292, doi:10.1016/j.quageo.2009.04.005.
rP
387
Guralnik, B., Matmon, A., Avni, Y., Porat, N., Fink, D., 2011. Constraining the evolution of
388
river terraces with integrated OSL and cosmogenic nuclide data. Quaternary
389
Geochronology, 6, 22‐32.
390
Jackson, J. & D. McKenzie, 1984. Active tectonics of the Alpine‐Himalayan Belt between
rR
391
ee
western Turkey and Pakistan. Geophys. J. R. Astron. Soc., 77, 185‐264.
392
Le Dortz, K., Meyer, B., Sébrier, M., Nazari, H., Braucher, R., Fattahi, M., Benedetti, L.,
393
Foroutan, M., Siame, L., Bourles, D., Talebian, M., Bateman, M.D. & Ghorashi, M.,
394
2009. Holocene right‐slip rate determined by cosmogenic and OSL dating on the
395
Anar fault, Central Iran. Geophys. J. Int., 179, 700‐710.
iew
ev
hal-00681707, version 1 - 14 May 2012
377
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
396
Le Dortz, K., B. Meyer, M. Sébrier, R. Braucher, H. Nazari, L. Benedetti, M. Fattahi, D.
397
Bourles, M. Foroutan, L. Siame, A. Rashidi, & M. Bateman, 2011. Dating inset
398
terraces and offset fans along the Dehshir fault combining cosmogenic and OSL
399
methods. Geophys. J. Int., 185, 1147‐1174.
400
401
Meyer, B., Mouthereau, F., Lacombe, O. & P. Agard, 2006. Evidence of Quaternary activity
along the Deshir Fault, Geophy. J. Int., 164, 192‐201.
402
Meyer, B. & K. Le Dortz, 2007. Strike‐slip kinematics in Central and Eastern Iran :
403
estimating fault slip‐rates averaged over the Holocene, Tectonics, 26, TC5009,
404
doi:10.1029/2006TC002073.
Geophysical Journal International
406
Murray, A.S. & A.G. Wintle, 2000. Luminescence dating of quartz using an improved
single‐aliquot regenerative‐dose protocol., Radiation Measurements, 32, 57‐73.
407
Nazari, H., M. Fattahi, B. Meyer, M. Sébrier, M. Talebian, M. Foroutan, K. Le Dortz, M. D.
408
Bateman & M. Ghorashi, 2009. First evidence for large earthquakes on the
409
Deshir Fault, Central Iran Plateau. Terra Nova, 21, 417–426.
410
411
Rittenour, T.M., 2008. Luminescence dating of fluvial deposits: applications to
geomorphic, palaeoseismic and archaeological research. Boreas, 37, 613‐635.
412
R.G. Roberts, R.F. Galbraith, H. Yoshida, G.M. Laslett, J.M. Olley, 2000. Distinguishing dose
413
populations in sediment mixtures: a test of single‐grain optical dating
414
procedures using mixtures of laboratory‐dosed quartz. Radiation Measurements,
415
32 , 459‐465.
rP
416
Schmidt, A., Quigley, M., Fattahi, M., Azizi, Gh., Maghsoudi, M. and H. Fazeli, 2011.
417
Holocene settlement shifts and palaeoenvironments on the Central Iranian Plateau:
418
investigating
419
DOI:10.1177/0959683610385961.
linked
ee
systems.
The
Holocene,
21
(4),
583‐595,
rR
420
Staubwasser, M., Sirocko, F., Grootes, P.M. and M. Segl, 2003. Climate change at the 4.2 ka
421
BP termination of the Indus valley civilization and Holocene south Asian monsoon
422
variability. Geophys. Res. Lett., 30(8), 1425, doi:10.1029/2002GL016822.
iew
ev
hal-00681707, version 1 - 14 May 2012
405
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 14 of 28
423
Vernant, P., Nilforoushan, F., Hatzfeld, D., Abbassi, M.R., Vigny, C., Masson, F., Nankali, H.,
424
Martinod, J., Ashtiani, A., Bayer, R., Tavakoli, F. & J. Chéry, 2004. Present‐day
425
crustal deformation and plate kinematics in the Middle East constrained by GPS
426
measurements in Iran and northern Oman. Geophy. J. Int., 157, 381‐398.
427
Walker, R. & M. Fattahi, 2011. A framework of Holocene and Late Pleistocene
428
environmental change in eastern Iran inferred from the dating of periods of
429
alluvial fan abandonment, river terracing, and lake deposition, Quaternary
430
Science Reviews, 30, 1256‐1271, doi:10.1016/j.quascirev.2011.03.004.
431
432
Walker, R. & J. Jackson, 2004. Active tectonics and late Cenozoic strain distribution in
central and eastern Iran, Tectonics, 23, TC5010, doi:10.1029/2003TC001529.
Page 15 of 28
434
Figure captions
435
Figure 1. Landsat mosaic of the Anar fault area. White squares for location of
436
photograph in Figure 2, Quickbird extract in Figure 3, and location of OSL‐2 sampling
437
site. Upper right inset locates the area within a simplified seismotectonic map of Iran. K,
438
D, A, KB, N and G, respectively for Kashan, Dehshir, Anar, Kuh Banan, Nayband, and
439
Gowk faults.
440
Figure 2. Field photograph of the Anar fault (vertical arrows, top panel) taken
441
towards the north from 31.2764°N and 55.1304°E with emphasis on a 3 m right‐lateral
442
offset rill (horizontal arrow, bottom panel).
443
Figure 3. (a) Quickbird imagery of the Anar fault trace (red arrows) centred on
444
the paleoseismological site (rectangle). Circle indicates the 8±0.5 m cumulative dextral‐
445
offset riser described as site 1 in Le Dortz et al. (2009). (b) Topographic DGPS map (top)
446
and profiles (bottom) of the paleoseismological site. Contour interval is 5 cm (survey
447
data not tied to absolute elevation). The smooth and subdued E‐facing fault scarp is
448
indicated by a red overprint and the location of the trench by a rectangle. Violet and
449
green dots locate the DGPS data points used for the topographic profiles. (c) Field
450
photograph, looking south, of the E‐facing fault scarp. The upper part of the southern
451
trench wall is in the foreground. Three white labels that are 1‐m spaced show scale on
452
trench wall.
iew
ev
rR
ee
rP
hal-00681707, version 1 - 14 May 2012
433
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
453
Figure 4. Composite photomosaic of the sediments exposed in the Anar trench
454
excavation, with top four meters (between 11 and 12 meters) east and bottom two
455
meters (between 19 and 20 meters) west of the main fault zone (a), and corresponding
456
stratigraphic log of the units with vertical positions of dated samples (b). Numbers
457
indicate units described in detail in Table 1. Three event horizons (EH‐A, ‐B and ‐C) are
458
shown as thick black lines (see text for discussion). Stars locate the positions (see exact
459
location in Figure 5) of the six samples with OSL ages given (c).
460
Figure 5. (Top) Photomosaic of southern trench wall (see Figure 3 for location).
461
(Bottom) Corresponding log, with labels marking the stratigraphic units. Faults and
462
fractures are shown with red and dashed red lines, respectively. Short black lines
Geophysical Journal International
represent bedding of sedimentary layers. Event horizons discussed in text are labelled
464
as EH‐A, EH‐B and EH‐C. Labelled black stars locate the OSL samples. Sample Ant‐I,
465
collected west of the fault zone at 4.1‐m‐deep within unit 2, yields an age of 13.6±1.3 ka.
466
Sample Ant‐II, collected east of the fault zone at 4.2‐m‐deep within unit 5, yields an age
467
of 12.4±0.6 ka. Sample Ant‐III, collected at 1.5‐m‐deep in a sandy lens of unit 9, yields a
468
poorly constrain ages between 11.5 and 3.3 ka. Sample Ant‐IV, collected at 1.5‐m‐deep at
469
the base of a fissure filled with eolian and runoff sand deposits within unit 12, yields an
470
age of 6.2±0.6 ka. Sample Ant‐V, collected at 1‐m‐deep within unit 11, yields an age of
471
7.1±0.7 ka. Sample Ant‐VI, collected at 0.8‐m‐deep in a sandy lens within unit 14, yields
472
an age of 6.2±0.4 ka.
473
Figure 6. Photomosaic (top) and interpretation (bottom) of the main faulted
474
zone showing evidence for the most recent earthquake (event A). Units 11 to 15,
475
affected by faulting, are warped downward, while undeformed units 16 and 17 lie
476
unconformably atop. Faults and tiny fractures are shown as red and dashed red lines,
477
respectively.
ee
rP
478
Figure 7. Schematic view of possible restoration of trench log showing sequence
479
of faulting and depositional phases from present‐day (step 1) to prior to the oldest
480
paleoearthquake (step 7). The trench log has been simplified for clarity. Dashed black
481
lines show inferred ground surface prior to the erosion. Step 1) is present‐day situation.
482
Step 2) is E‐facing fault scarp formed during the most recent earthquake. Step 3) is
483
restoration of the ground surface to its position prior to event A showing a vertical
484
displacement of about 60 cm created by faulting on the eastern branch. Step 4) is the
485
penultimate earthquake denoted by open fissures and cracks formed near the main
486
faulted zone. Sand‐blows also emplaced in both sides and away of the main faulted zone.
487
Step 5) is restoration of the ground surface to its position prior to event B showing
488
vertical displacement created during penultimate earthquake on the eastern and
489
western fault branches about 25 and 15 cm, respectively. Step 6) is the oldest
490
earthquake, it has produced some surficial fissures within downthrown (eastern) block
491
and a 25‐cm‐height E‐facing fault scarp. Step 7) is restoration of the ground surface to its
492
position prior to event C, showing a vertical displacement of about 25 cm.
iew
ev
rR
hal-00681707, version 1 - 14 May 2012
463
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 16 of 28
493
Figure 8. Evidence for the penultimate earthquake (event B) encountered in a
494
fissure fill and a liquefaction feature (sand‐blow) at 16 and 22 meters of the trench log,
Page 17 of 28
respectively. Red lines correspond to tiny cracks postdating the penultimate earthquake.
496
(a) Photomosaic of fissure filled with eolian sands and silts and interpretative sketch.
497
The fissure is ~85‐cm‐deep and up to ~55‐cm‐wide, tapering downward through unit 9.
498
Colours denote stratigraphy as shown in Figures 4 and 5. The material at the base of the
499
fissure includes a collapsed piece from sidewall and grades into interbedded sand and
500
silt layers. The fissure is sealed by a grey, well‐stratified, surficial runoff deposit (unit
501
13). (b) Photomosaic and interpretative sketch of a sand‐blow (unit 12) made of sandy
502
material of unit 8 injected into and deforming units 9 and 11. Hydraulic fractures,
503
vertical and oblique alignments of dragged sands and gravels deform the host sediments
504
close to the liquefaction pillar.
505
Figure 9. Five well‐constrained OSL ages place bounds on the ages of the three
506
paleoearthquakes (A, B, and C) identified in the Anar trench exposure. Light grey areas
507
represent the maximum time windows for the past earthquakes, dark grey pointing the
508
minimum time window for event C. For event A, hatched domain shows the preferred
509
time interval of the earthquake (4.4±0.8 ka). Colour codes are similar to units shown in
510
Figures 4 and 5. Question mark illustrates the ambiguity due to the lack of ages in units
511
postdating the most recent earthquake (16 and 17).
rR
ee
rP
512
ev
513
Table 1. Detailed description of the units observed in the trench excavated across the
514
Anar fault, corresponding stratigraphic column and log are shown in Figures 4 and 5,
515
respectively.
516
iew
hal-00681707, version 1 - 14 May 2012
495
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Geophysical Journal International
517
Table 2. Calculated OSL ages on the Anar samples with their parameters. Ages have
518
been calculated for quartz grains with size ranging between 90 and 150 microns.
519
Geophysical Journal International
Table 1
522
Unit
523
17
Light grey to cream, poorly stratified, silty sand with some angular gravels (20%, 0.5-2 cm).
524
525
16
Light cream, fairly stratified, medium consolidated, angular gravels (60%, 0.5-6 cm), with sand and silt
matrix and some gypsiferous cementation.
526
527
15
Buff to light grey, well stratified, medium consolidated, angular gravels (50%, 0.5-10 cm), with sand and silt
matrix and some gypsiferous cementation.
528
529
14
Grey to buff grey, thinly stratified, medium consolidated sand and angular granules (0.5-4 cm) with crossbedding.
530
531
13
Grey, well stratified, medium to well consolidated, coarse sand with angular granules and few pebbles (30%,
0.2-6 cm). The unit bottom is erosive and rests both on units 12 and 11.
532
12
Light cream to grey, sandy fissure fills and sand-blows.
533
534
11
Brown to olive cream, roughly stratified, medium to well consolidated, silts and sands with some channels of
angular gravels (40%, 0.1-4 cm). The top exhibits a hard gypsiferous calcrete.
535
536
10
Cream to grey, fairly stratified, poorly consolidated, coarse sands and angular gravels (50%, 0.1-3 cm) with
sandy-silty matrix and some cross-bedding.
537
538
9
Reddish brown to brownish grey, non-stratified, semi-consolidated coarse sand, with clayish silt matrix,
including some channels of angular gravels (90%, 2-15 cm) with silt matrix.
539
540
8
Cream to brownish grey, fairly stratified, well consolidated, angular granules and pebbles (60%, 0.2-3 cm),
with silty sand matrix, including some small cobbles and traces of sulfate calcrete within the coarser levels.
541
542
7
Grey to brownish grey, poorly stratified, poorly sorted angular pebbles (70%, 0.5-10 cm), with sandy silt
matrix.
543
544
6
Cream to brown, fairly stratified, angular granules and pebbles (80%, 0.2-4 cm), with silty sand matrix,
including some sparse big cobbles and silty sand lenses.
545
5
Buff to light grey, fairly stratified, loose silt and sand.
546
4
Grey to light grey, poorly stratified, well sorted, silty sands with some cross-bedding and gravelly lenses.
547
548
3
Grey, fairly stratified, well consolidated, medium to well sorted, coarse angular gravels (90%, 0.5-10 cm)
with silty sand matrix, including some sparse bigger cobbles and clayish silty sand lenses.
549
2
Grey, stratified, loose, silty sand, with some cross-bedding.
550
1
Grey, poorly stratified, well consolidated, angular pebbles (>20%, 0.5-3 cm) with silty sand matrix.
iew
ev
rR
ee
551
Detailed Stratigraphic Explanation
rP
hal-00681707, version 1 - 14 May 2012
520
521
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 18 of 28
Page 19 of 28
hal-00681707, version 1 - 14 May 2012
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
552
Geophysical Journal International
Table 2
553
Sample
Latitude
°N
Longitude
°E
Depth
(m)
Water
(%)
K
(%)
U
(%)
Th
(%)
Annual dose
rate (Gy/ka)
Ant‐I (OSL1‐b)*
31.1952
55.1534
4.1
2
0.93±0.03
1.16±0.05
4.0±0.1
Ant‐II
31.1952
55.1534
4.2
0.4
0.77±0.03
0.97±0.05
Ant‐III
31.1952
55.1534
1.7
1.1
0.97±0.01
Ant‐IV
31.1952
55.1534
1.5
0.6
Ant‐V
31.1952
55.1534
1.0
Ant‐VI (OSL1‐a)*
31.1952
55.1534
OSL‐2
31.2697
55.1337
554
Skewness
CAM De
(Gy)
1.64±0.05
over‐
dispersion
(%)
24
0.69
22.3±1.90
3.4±0.1
1.42±0.04
27
4.09
16.1±0.90
17.5±0.47
12.4±0.6b
4.3±0.1
1.75±0.04
51
2.1
9.7±1.09
6.1±0.24
3.5±0.2b
0.94±0.01
1.44±0.05
3.7±0.1
1.78±0.14
20
0.01
11.0±0.61
0.3
0.99±0.01
1.72±0.05
4.2±0.1
1.96±0.14
25
0.35
13.5±0.90
13.9±0.70
7.1±0.7b
0.8
rP
1.23±0.05
0.6
1.21±0.01
1.60±0.05
5.9±0.1
2.20±0.05
42
0.94
11.6±1.39
13.7±0.56
6.2±0.4c
0.8
1.1
1.06±0.01
1.33±0.05
1.90±0.05
41
2.39
20.1±2.16
19.3±0.76
10.1±0.6c
Fo
ee
rR
4.0±0.1
ev
555
a
556
b Age
based on De determined using the dominant component of Finite Mixture Modeling (FMM; Roberts et al., 2000).
557
c Age
previously published, now refined using Finite Mixture Modeling (FMM; Roberts et al., 2000).
558
* Parenthesis contains the sample label used by Le Dortz et al. (2009)
559
Age based on De determined using Central Age Model (CAM; Galbraith et al., 1999).
iew
FMM De
(Gy)
Age
(ka)
13.6±1.3a
6.2±0.6a
Geophysical Journal International
iew
hal-00681707, version 1 - 14 May 2012
ev
rR
ee
rP
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Page 20 of 28
W
Page 21 of 28
iew
v
Re
hal-00681707, version 1 - 14 May 2012
er
Pe
S
E
r
Fo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
Geophysical Journal International
N
55.15°E
3452710 3452720
a
3452700
D
C
15
3452690
44
15
0
44
15
.25
43
3452670
ee
rR
.50
10 m
324040
Elevation (m)
1544.9
1544.7
1544.5
1544.3
1544.1
31.19
East
200 m
.5
3452680
rP
hal-00681707, version 1 - 14 May 2012
A
c
324060
324050
324070
324080
324090
44
15
31.195
Fo
Page 22Bof 28
1543.75
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
b
1544.75
31.2°N
Geophysical Journal International
324100
West
B
VE = 20
ev
35cm
A
C
324040
D
ie
324060
324080
324100
distance (m)
w
West
Page 23 of 28
Calender Age (Ka)
3.5±0.2
17
16
4
15
5
9
8
A
10
ie
w
6
11
Ant-II
Ant-II
12.4±0.6
5
9
ev
7
7
8
Ant-III
8
Ant-V
7.1±0.7
11
6
t
Ant-IV
rR
EH-C
12
10
ee
9
13
Ant-V
Ant-VI
6.2±0.4
I
14
11
10
I
15
I
7.3±0.5
13
10.7±0.8
EH-B
Ant-VI
n
14
Ant-IV
6.2±0.6
EH-A
4
12
13
3
Ant-I
13.6±1.3
Stratigraphic Depth
(a)
rP
hal-00681707, version 1 - 14 May 2012
Present-day
(b)
Fo
1
2
3
4
5
0m
6
7
8
9
10
11
12
13
14
15
-1m
16
17
18
19
20
21
22
23
24
-2m
25
26
27
28
29
30
31
32
33
-3m
34
35
36
37
38
39
40
41
42
43
-4m
44
45
46
47
48
49
50
51
52
-5m
53
54
55
56
57
58
59
60
(c)
Geophysical Journal International
2
14
Ant-I
1
15
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
East
hal-00681707, version 115- 14 May 2012
10
5
Meter 0
20
25
West
28
N085°
4
4
3
3
Pee
0
rR
-1
4
evi
Location: N31.1953
E055.1536
Elevation: 1544.5m
ew
EH-A
3
EH-B
2
EH-C
100
12
14
10
5
4
3
6
2
1
-1
Ant-I
Ant-II
0
5
200 cm
1
0
-1
4
3
2
7
9
8
0
0
Ant-III
17
15
13
11
Ant-IV
Ant-VI
Page 24 of 28
1
16
Ant-V
2
Geophysical Journal International
1
For
2
10
15
20
25
28
1
0
-1
Page 25 ofEast
28
hal-00681707, version 1 - 14 May 2012
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
West
Geophysical Journal International
Fo
rP
ee
Meter 12
rR
50
0
100
ev
East
CENTIMETERS
14
West
ie
17
16
w
15
11 +13 +14
Step 1: Present-day
E
Page 26 of 28W
Geophysical Journal International
Ant-VI
Ant-V
Ant-I
Ant-II
0
100
200 cm
W
0
100
200 cm
W
ee
rP
Event A=60cm
0
rR
100
200 cm
W
0
100
200 cm
W
w
ie
ev
hal-00681707, version 1 - 14 May 2012
Ant-III
Ant-IV
Fo
1
2
3
4
5
6
7
8Step 2: Just after Event A
E9
10
11
12
13
14
15
16
17
Step 3: Just before Event A
18
E
19
20
21
22
23
24
25
26
27
Step 4: Just after Event B
E28
29
30
31
32
33
34
35
Step 5: Just before Event B
E36
37
38
39
40
41
42
43
44
Step 6: Just after Event C
E45
46
47
48
49
50
51
Step 7: Just before Event C
E52
53
54
55
56
57
58
59
Event B=25cm+15cm
0
100
200 cm
W
0
100
200 cm
W
Event C=25cm
0
100
200 cm
(a)
Page 27 of 28
hal-00681707, version 1 - 14 May 2012
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
(b)
Geophysical Journal International
Fo
rP
ee
0
50
15
14
13
11
CENTIMETERS
rR
100
0
50
ev
15
ie
13
12
11
12
9
8
3/23
14
w
10
100
CENTIMETERS
9
8
OSL Samples
Yrs X 1000
Ant-I
Ant-II
Event A
5.8
5.6
6.6
6.8
6.4
Event B
rR
7.8
Event C
ev
11.8
ie
hal-00681707, version 1 - 14 May 2012
?
ee
12.3
13.0
w
12
23
3
44
5
65
76
8
97
10
8
11
12
9
13
10
14
15
11
16
12
17
18
13
19
20
14
21
15
22
23
Ant-V
rP
Fo
1
Ant-IV
Geophysical Journal International
Page 28 of 28
Ant-VI
0
14.9