Accepted Manuscript

Magnetic stratigraphy of the Bucaramanga alluvial Fan: Evidence for a ≤ 3 mm/yr slip
rate for the Bucaramanga-Santa Marta Fault, Colombia

Giovanny Jiménez Diaz , Fabio Speranza , Claudio Faccena , German Bayona ,

Andres Mora

PII: S0895-9811(14)00154-0
DOI: 10.1016/j.jsames.2014.11.001
Reference: SAMES 1341

To appear in: Journal of South American Earth Sciences

Received Date: 13 March 2014

Revised Date: 24 October 2014
Accepted Date: 7 November 2014

Please cite this article as: Diaz, G.J., Speranza, F., Faccena, C., Bayona, G., Mora, A., Magnetic
stratigraphy of the Bucaramanga alluvial Fan: Evidence for a ≤ 3 mm/yr slip rate for the Bucaramanga-
Santa Marta Fault, Colombia, Journal of South American Earth Sciences (2014), doi: 10.1016/
j.jsames.2014.11.001.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 1
ACCEPTED MANUSCRIPT

1 Magnetic stratigraphy of the Bucaramanga alluvial Fan: Evidence for a ≤ 3

2 mm/yr slip rate for the Bucaramanga-Santa Marta Fault, Colombia

4 Giovanny Jiménez Diaz 1, 2, Fabio Speranza 2, Claudio Faccena 1, German

PT
5 Bayona 3, Andres Mora 4

RI
7 1 Università di Roma TRE, Roma, Italy

SC
2 Istituto Nazionale di Geofisica e Vulcanologia, Roma, Italy

9 3 Corporación Geológica ARES, Bogota, Colombia

U
10 4 Instituto Colombiano del Petróleo, Ecopetrol-ICP, Bucaramanga, Colombia
AN
11

12
M

13

14
D

15
TE

16
EP

17

18
C

19
AC

20

21

22

23

24

25
 2
ACCEPTED MANUSCRIPT

26

27

28 Abstract

29

PT
30 The 550 km long Bucaramanga-Santa Marta Fault is one of the main active tectonic

31 features of NW South America. It is a left-lateral strike-slip fault bounding the

RI
32 Maracaibo block, and straddling northern Colombia from the Caribbean Sea to the

33

SC
Eastern Cordillera. Variable total displacement values (from 40 to 110 km), and

34 present-day slip rates (from 0.01 to 10 mm/yr) have been proposed so far for the

U
35 Bucaramanga Fault. Here we report on the paleomagnetic investigation of a Plio-
AN
36 Pleistocene (?) continental alluvial fan juxtaposed to the Bucaramanga Fault, and

37 horizontally displaced by 2.5 km with respect to its feeding river. Nine (out of
M

38 fourteen) reliable paleomagnetic directions define a succession of six different

39 magnetic polarity zones that, lacking additional age constraints, can be correlated
D

40 with several tracks of the Plio-Pleistocene magnetic polarity time scale. If the
TE

41 youngest age model is considered, most recent sediments of the fan can be
EP

42 reasonably dated at 0.8 Ma (Brunhes-Matuyama chron transition), translating into a

43 maximum 3 mm/yr slip rate for the Bucaramanga Fault. Older age models would
C

44 obviously yield smaller slip rates. Our paleomagnetic sites, located at 4-10 km from
AC

45 the fault, do not show significant rotations, implying weak fault coupling and/or

46 ductile upper crust behavior adjacent to the Bucaramanga Fault.

47

48 Keywords: Paleomagnetism, Bucaramanga Fan, Bucaramanga Fault, Slip rate.

49

50
 3
ACCEPTED MANUSCRIPT

51 Introduction

52

53 The NW margin of South America is characterized by a complex pattern of

54 mountain ranges (Western, Central, and Eastern Cordillera of Colombia, Fig. 1),

PT
55 and by a puzzle of crustal blocks undergoing independent movements (Kellogg et

56 al., 1982; Taboada et al., 2000; Trenkamp et al., 2002). The eastward subduction of

RI
57 the Nazca Plate beneath South America at a rate of 6 cm/yr is well documented

58

SC
(Trenkamp et al., 2002), while the existence of an ESE-ward directed subduction of

59 the Caribbean plate is debated (Pindell et al., 1988). A well-developed accretionary

U
60 wedge (Fig. 1) is exposed at the boundary between the Caribbean and South
AN
61 American plate (e.g., Flinch et al., 2003).

62 The Maracaibo block, the biggest semi-rigid block of the Colombian tectonic
M

63 puzzle, is bounded by three major wrench faults: the dextral Boconó and Oca

64 faults, and the sinistral Bucaramanga-Santa Marta Fault (Fig. 1). The relevance,
D

65 age, and displacement rate of these major strike-slip faults have been widely
TE

66 discussed in the past (Irving, 1971; Toro, 1990; Ujueta, 2003; Mora and Garcia,
EP

67 2006; Lopez et al., 2008), and a consensus on the whole block and fault kinematics

68 has not been reached so far.

C

69 The Bucaramanga-Santa Marta Fault system juxtaposes basement rocks to the east
AC

70 and sedimentary Jurassic to Cenozoic rocks to the west. This fault system extends

71 for a distance of 550 km from the Caribbean coast to the Eastern Cordillera (Fig.

72 1), where it crosses the city of Bucaramanga (ca. 500,000 inhabitants). It can be

73 divided along strike into three major zones (Northern, Central and Southern zone),

74 where it bounds several distinct geological provinces. The Northern zone, or Santa

75 Marta section (Paris, 2000; Ingeominas, 2001a), corresponds to a 140 km long

 4
ACCEPTED MANUSCRIPT

76 topographic lineament juxtaposing old crystalline rocks of the Santa Marta Massif

77 to Neogene deposits of the Lower Magdalena Valley basin (Fig. 1). The 100 km

78 long Central zone is covered by alluvial deposits of the Lower Magdalena Valley

79 basin (Ariguani graben) and Cesar Valley basin (Mora and García, 2006; Fig.1).

PT
80 The Southern zone corresponds to Bucaramanga section (Paris, 2000; Ingeominas,

81 2001a), a 310 km long outstanding linear topographic feature separating the

RI
82 Santander Massif from the Middle Magdalena Valley basin (Fig. 1). At the city of

83

SC
Bucaramanga, the fault clearly left-laterally displaces a 10-15 km wide continental

84 alluvial fan of presumable Plio-Pleistocene age, produced by debris carried by the

U
85 Suralá river from the adjacent Santander Massif (Fig. 2). The zone is well known
AN
86 for the high seismic activity of the “Bucaramanga nest” (Taboada et al, 2000; Zarifi

87 et al, 2007; Fig. 3), but earthquakes are deep (from 100 to 200 km), thus they
M

88 should not be related to the activity of the Bucaramanga Fault.

89 In this paper we report on a paleomagnetic investigation of the Bucaramanga Fan to

D

90 constrain its age by magnetic stratigraphy, one of the few viable method to
TE

91 dategravel-size alluvial fan deposits. The results allow the documentation of when
EP

92 fan displacement occurred, and unravel recent and present-day slip rate of the

93 Bucaramanga Fault. The city of Bucaramanga has not been struck by devastating
C

94 earthquakes since when historical accounts are available (XVII century AD), thus it
AC

95 is unclear whether we are living an inter-seismic interval preceding a future strong

96 earthquake, or in fact there is no significant present-day activity of the

97 Bucaramanga Fault, as GPS data of Trenkamp et al. (2002) might suggest (Fig. 1).

98

99

100
 5
ACCEPTED MANUSCRIPT

101 Characteristics of the Bucaramanga-Santa Marta Fault

102

103 Previous work around the City of Bucaramanga has been done in the frame of

104 regional and local cartography projects (Ward et al., 1973; Ingeominas, 2001b;

PT
105 Diedrix et al., 2009; Fig. 2). In the Santander Massif, which is bounded to the west

106 by the Bucaramanga Fault system, Precambrian, Paleozoic, and Mesozoic intrusive

RI
107 and metamorphic rocks (schists and gneisses) are exposed. According to Cediel et

108

SC
al. (2003), the Bucaramanga-Santa Marta Fault was active during the Precambrian

109 Grenville - Orinoco continental collision, and reactivated in the Aptian-Albian

U
110 times. The Bucaramanga Fault bounds a thick deposition of upper Jurassic and
AN
111 Cretaceous rocks to the south and west (Kammer and Sanchez, 2006), whereas to

112 the north and east the deposition of Jurassic rocks is more condensed and in some
M

113 areas lowermost Cretaceous strata are lacking (Ward et al, 1973).

114 Provenance analysis indicates that the Santander Massif and its sedimentary cover
D

115 were uplifted since Paleogene time (Ayala et al, 2012). Paleocene movements may
TE

116 be related to other structures in the Cobardes Anticline (Parra et al, 2012, Fig. 1).
EP

117 Mora and García (2006) using extensive seismic profile analysis and well data in

118 the Central zone, proposed that the Bucaramanga-Santa Marta Fault was
C

119 responsible for the intense structural deformation that affected pre-Oligocene units,
AC

120 indicating that it was an active boundary between the Lower Magdalena Valley and

121 Cesar basins during the late Eocene-early Oligocene. They also suggested that fault

122 activity ended after Oligocene, as the Lower Magdalena Valley and Cesar basins

123 have been connected without tectonic disturbances.

124

125 Ross et al. (2009) using Apatite fission track analysis (AFTA) of Precambrian to
 6
ACCEPTED MANUSCRIPT

126 Neogene rocks from the Santander Massif and Middle Magdalena Valley proposed

127 three paleothermal episodes (interpreted as erosional phases) in the Paleocene (65

128 to 60 Ma), Early Miocene (20 to 18 Ma) and Middle Miocene (12 to 9 Ma). This

129 evidence provides age clues for the uplift of the Santander Massif with respect to

PT
130 the adjacent Magdalena Valley that likely occurred along the Bucaramanga Fault.

131 As the Bucaramanga Fault is a recent geomorphological feature, it was likely

RI
132 activated in the last phase of deformation of late Miocene age, as suggested by

133

SC
AFTA data. Finally, Pindell et al. (1988) proposed that the Bucaramanga – Santa

134 Marta fault is the western limit of the Maracaibo Block since Miocene times.

U
135
AN
136 Reported total displacement values of the Bucaramanga – Santa Marta fault vary

137 from 45 to 110 km. Toro (1990) using structural reconstructions proposed for the
M

138 southern fault portion a displacement of 45 km. According to Campbell (1968), the

139 total displacement is more than 100 km, relying on a regional scale correlation
D

140 between rocks from the Central Cordillera and the Santa Marta Massif. Other
TE

141 authors (Irving, 1971; Tschantz, 1974) proposed 110 km of displacement after the
EP

142 analysis of samples drilled in oil wells. Montes et al. (2010) and Bayona et al.

143 (2010) considered the Bucaramanga-Santa Marta Fault system as western boundary
C

144 of the Maracaibo block clockwise rotation during the Cenozoic. Reported values
AC

145 for the fault slip rate range between 0.01-0.2 mm/yr, calculated on the basis of

146 geomorphic offset features (Paris, 2000; Ingeominas, 2001a), to 10 mm/yr,

147 evaluated after regional geology considerations (Rivera, 1989).

148

149 López et al., (2008) reported the occurrence of Pseudotachylite associated to

150 cataclasites in the southern sector of the Bucaramanga Fault, that they interpreted
 7
ACCEPTED MANUSCRIPT

151 as an evidence for paleoseismicity along the fault. Diedrix et al. (2009) using

152 paleoseismological investigation and radiocarbon dating of paleosols trenched

153 adjacent to the fault proposed eight Holocene seismic events during the last 8300

154 years and magnitudes in the order of 6.5-7.0, associated to the Bucaramanga Fault

PT
155 activity.

156

RI
157 Ujueta (2003) made an extensive review of the previous works on the Bucaramanga

158

SC
Santa Marta Fault available since 1933, and observed that some authors consider

159 the Bucaramanga Santa Marta fault as a reverse fault, others as left lateral strike-

U
160 slip fault, while few works propose two different and independent faults.
AN
161 Concerning the age of the Bucaramanga Fault activity, a late Mesozoic, Paleocene,

162 Eocene or Pliocene-Quaternary age are proposed (Ujueta, 2003). Finally, using
M

163 seismic profiles, he discussed the continuity of the fault Central zone, and proposed

164 two different but related faults, with absence of connection between the southern
D

165 and the northern zone.

TE

166 The deep seismicity of the Bucaramanga Nest

EP

167

168 No significant shallow crustal seismicity has been recorded along the Santa Marta-
C

169 Bucaramanga Fault (Fig. 3), but at depth a remarkable cluster of subcrustal
AC

170 earthquakes in the 100-200 km depth interval is observed in the so-called

171 “Bucaramanga nest”, centered at 7°N 73°W (Fig 3). It differs from other worldwide

172 nests by its high rate of activity in a volume much smaller than the other nests at

173 similar intermediate depths (Schneider et al., 1987; Zarifi and Harsco, 2003; Prieto

174 et al., 2012). The Bucaramanga nest has been mostly related to tectonic processes

175 occurring in the oceanic subducting plates beneath Colombia (Zarifi and Havskov,
 8
ACCEPTED MANUSCRIPT

176 2003). Taboada et al. (2000) proposed that the Bucaramanga Nest is due to the

177 interaction between the subducting Nazca and paleo-Caribbean plates, while Zarifi

178 et al. (2007) suggested a simultaneous subduction process and collision between

179 two subducted slabs. Conversely, Cediel et al. (2003) proposed that the seismic

PT
180 activity recorded in the Bucaramanga Nest represents a zone of tectonic detachment

181 associated with NW-ward migration of the Maracaibo subplate, while Rivera

RI
182 (1989) proposed that there is a relationship between the Bucaramanga Fault and

183

SC
Nest.

184

U
185 The Bucaramanga Fan
AN
186

187 The Bucaramanga Fan is limited to the east by the Santander Massif and the
M

188 Bucaramanga Fault, and to the west by the Suárez Fault (Fig. 2). The Suárez Fault

189 is a west-dipping reverse fault with sinistral displacement and Quaternary activity
D

190 (Page, 1986; Paris et al., 2000; Ingeominas, 2001a) that terminates northward
TE

191 against the Bucaramanga Fault. In the study area, the Suárez Fault juxtaposes
EP

192 Jurassic and Cretaceous strata against the Bucaramanga Fan. Locally, the

193 Bucaramanga Fan is folded with vertical to overturned beds by the effect of the
C

194 Suárez Fault (Ingeominas, 2008).

AC

195

196

197 De Porta (1959) defined the Bucaramanga Fan as a sedimentary deposit of

198 Quaternary age with an alluvial fan morphological shape (Fig. 2).

199 According to Julivert (1958) The E-W creeks and the shape of the Bucaramanga

200 Fan can be related, but the present-day drainage network can't explain the
 9
ACCEPTED MANUSCRIPT

201 Bucaramanga Fan location. Ingeominas (2001b and 2008) suggested that the Suratá

202 River is the feeder of the Bucaramanga Fan, although it is presently shifted by 2.5

203 km with respect to fan apex. According to Ingeominas (2008) the southern part of

204 the Bucaramanga Fan has been more faulted than the northern part, and seems to be

PT
205 diachronous. Thus the progressive northward migration of the fan apex has been

206 related to the sinistral shear along the Bucaramanga Fault.

RI
207

208

SC
According to Ingeominas (2001b) the thickness of the Bucaramanga Fan increases

209 from east to west with an average thickness of 250 m, and from base to the top it

U
210 has been subdivided into, the Organos, Finos, Gravoso, and Limos Rojos member
AN
211 (Fig. 2).

212 The Organos Member was first defined by Hubach (1952), and according to Bueno
M

213 and Solarte (1994) is a monotonous sequence of conglomerates with intercalations

214 of fine sandstones. Conglomerate clasts are 10 - 30 cm in diameter (and in some

D

215 cases more than 1 m) and include gneisses and schists fragments. This member is
TE

216 clast supported, contains a clay matrix (Ingeominas, 2001b), and crops out to the
EP

217 west of the study area with a maximum thickness of 180 m (Mancera and

218 Salamanca, 1994). The Finos Member (Hubach, 1952) is a continuous sub-
C

219 horizontal 15 m thick bed overlying the Organos member along a net discontinuity,
AC

220 and it consists of clays evolving upsection into fine-grained sandstone beds

221 (Ingeominas, 2001b). The Gravoso Member (Vargas and Niño, 1992) is an 8 to 30

222 m thick, matrix supported level of gravel-size clasts overlying the Finos Member

223 along a net contact. The clasts are on average 15 cm in diameter (cobble size), and

224 the matrix includes a mixture of clay- and sand-size fragments. Finally, the

225 Gravoso Member transitionally passes upward into the 15 m thick Limos Rojos
 10
ACCEPTED MANUSCRIPT

226 Member (Julivert, 1963); this unit consists of argillaceous sandstones and

227 conglomeratic sandstones, interbeded with layers of siltstones. Locally, block

228 fragments (meter scale in diameter) of sandstone texture are supported by the red

229 silty matrix (Ingeominas, 2001b).

PT
230 The Bucaramanga Fan is generally referred to a Plio-Pleistocene age (Ingeominas,

231 2001b), although no conclusive age constraints were gathered from the continental

RI
232 succession so far.

233

SC
234

U
235
AN
236 Sampling and Methods

237
M

238 We collected sandy-silty samples for paleomagnetic analysis in fourteen sites (154

239 cores, Table 1) from the Bucaramanga Fan, using a petrol-powered portable drill
D

240 cooled by water. Ten sites were gathered in the Organos Member, two in the Finos
TE

241 Member, and two in the Gravoso Member (Figure 2 and Table 1). At each site we
EP

242 collected 6–12 cores (11 on average), spaced in at least two outcrops in order to try

243 to average out secular variation of the geomagnetic field. All samples were oriented
C

244 using a magnetic compass, corrected to account for the local magnetic field
AC

245 declination value at the sampling area (about 7º W) according to NOAA’s National

246 Geophysical Data Center (http://www.ngdc.noaa.gov).

247 Cores were prepared into standard cylindrical specimens of 22 mm height, and

248 rock magnetic and paleomagnetic measurements were done in the shielded room of

249 the paleomagnetic laboratory of the Istituto Nazionale di Geofisica e Vulcanologia

250 (Roma, Italy). All samples were thermally demagnetized through 11–12 steps up to
 11
ACCEPTED MANUSCRIPT

251 680°C by a shielded oven, and the natural remanent magnetization (NRM) of the

252 specimens was measured after each step with a DC- SQUID cryogenic

253 magnetometer (2G Enterprises, USA). Thermal demagnetization data were plotted

254 on orthogonal diagrams (Zijderveld, 1967), and the magnetization components

PT
255 were isolated by principal component analysis (Kirschvink, 1980). The site-mean

256 directions were evaluated by Fisher’s (1953) statistics.

RI
257 On a set of selected specimens, magnetic mineralogy analyses were carried out to

258

SC
identify and characterize the main magnetic carriers using the thermal

259 demagnetization of a three-component isothermal remanent magnetization (IRM)

U
260 imparted on the specimen axes, according to the method of Lowrie (1990). Fields
AN
261 of 2.7, 0.6, and 0.12 T were successively imparted on the z, y, and x sample axes

262 (respectively) with a pulse magnetizer (Model 660, 2G Enterprises). The

M

263 generalized stratigraphic column of the Bucaramanga Fan was made using field

264 data and previous geological maps (Ingeominas, 2001b). The stratigraphic position
D

265 for all sampled sites was calculated using a cross section.
TE

266
EP

267 Results

268
C

269 Magnetic Mineralogy

AC

270

271 The thermal demagnetization of a three-component IRM shows that for most of the

272 samples both the medium coercivity and the hard fractions are demagnetized

273 between 600 and 680°C (Fig. 4), pointing to hematite as the main magnetic carrier

274 of our samples. In almost all sites, a drop of the medium coercivity and hard

275 fraction occurring before 100º shows that a small amount of goethite is also
 12
ACCEPTED MANUSCRIPT

276 associated to hematite.

277

278 Paleomagnetism

279

PT
280 Only 9 (out of 14) sites yielded reproducible paleomagnetic directions during

281 cleaning, while the remaining 5 sites showed scattered demagnetization diagrams

RI
282 (Figure 5 and Table 1). For most of the sites, a characteristic magnetization

283

SC
component (ChRM) was isolated between 550 and 680°C, confirming that hematite

284 represents the main magnetic carrier. For about 10% of the samples, a ChRM is

U
285 isolated between 380 and 680°C, suggesting the coexistence of hematite and
AN
286 maghaemite (Fig. 5). Mean paleomagnetic directions are reasonably well

287 constrained, the α95 values being comprised between 8.4° and 22.4° for all sites but
M

288 one yielding a value of 48.2 (Fig. 6 and Table 1). Three sites are of normal polarity

289 (D = 356.1, I = 9.4, K = 136.76 and α95 = 10.6 ) and the other six are of reverse
D

290 polarity (D = 178.2, I =1.8, K = 36.20, and α95 = 11.3 ); the reversal test (according
TE

291 to McFadden and McElhinny (1990), is positive of Class “c”.

EP

292 The fact that our results yielded reverse polarity directions indicate that the

293 Bucaramanga Fan should be older than the Bruhnhes polarity chron (0 to 0.8 Ma)
C

294 When plotted according to their stratigraphic position, the sites from the
AC

295 Bucaramanga Fan yield a succession of three normal- and three reverse-polarity

296 magnetozones (Fig. 7). Each magnetozone is defined by one paleomagnetic

297 direction, except the youngest magnetozone R3, which is corroborated by four

298 different site results. As the age of the Bucaramanga Fan is vaguely inferred to be

299 Plio-Pleistocene, several different correlations with the global magnetic polarity

300 time scale (GPTS, Cande and Kent, 1995) are possible (Fig. 8). If we assign the
 13
ACCEPTED MANUSCRIPT

301 youngest possible age to the succession (Fig. 8a), R3 correlates with upper

302 Matuyama polarity chron (C1r.1r), and N3, N2, and N1 with the Jaramillo,

303 Olduvai, and Gauss chron, respectively. In a slightly older age model (Fig. 8b), R3

304 correlates with the mid Matuyama chron (C1r.2r), N3 with Olduvai chron, and N2

PT
305 and N1 fall both within the Gauss chron. It is not possible at present to indicate a

306 preferred age model using calculated sedimentation rates that vary irregularly

RI
307 (considering both age models of Figure 8) between 30 and 300 mm/kyr. In fact in

308

SC
this kind of deposits the sedimentation is expected to be highly irregular and

309 discontinuous, and the thickness of some magnetozones (such as N3) is poorly

U
310 constrained due to the occurrence of several failed sites located below and above it
AN
311 (Fig. 7).

312
M

313

314 Discussion and Conclusions

D

315
TE

316 The slip rate of the southern segment of the Bucaramanga Fault can be in principle
EP

317 calculated by dividing its along-fault displacement by the age of the displaced

318 rocks. Left-lateral displacement is well known, and equates the 2.5 km distance
C

319 from the mouth of the Suralá river to the apex of the Bucaramanga Fan
AC

320 (Ingeominas 2008, Figs. 9 and 10). Conversely, the age of the displaced sediments

321 is not firmly constrained, as several different age models can equally fit the

322 magnetozone succession gathered from the Bucaramanga Fan (Fig. 7).

323

324 However, a maximum possible slip rate can be calculated assigning the youngest

325 possible age of the succession, i.e. inferring that magnetozone R3 corresponds to
 14
ACCEPTED MANUSCRIPT

326 the top of the Matuyama polarity chron (Fig. 8a). In this case, a 0.8 Ma age

327 (Brunhes-Matuyama boundary, Cande and Kent, 1995) can be reasonably assigned

328 to the youngest sediments of the Bucaramanga Fan, by considering that: 1) we did

329 not reach the normal-polarity magnetozone above R3, thus the youngest sampled

PT
330 sediments (site GJ192) might fall in the lower-medium part of chron C1r.1r, and be

331 0.9-1 Ma old; 2) the upper member of the Bucaramanga Fan is the Limos Rojos

RI
332 Member, but the youngest paleomagnetic directions were obtained in the

333

SC
underlying Gravoso Member; 3) the Limos Rojos Member is only 15 m thick, and

334 shares the same fan morphology with the underlying Finos-Gravoso members (Fig.

U
335 2). This suggests that the Limos Rojos Member was rapidly emplaced, and that no
AN
336 significant time gap (and displacement of the fan feeding source) occurred with

337 respect to the Gravoso Member.

M

338 A 0.8 Ma age of the top of the Bucaramanga Fan translates into a maximum slip

339 rate of 3 mm/yr for the Bucaramanga Fault. Obviously, progressively smaller slip
D

340 rates would arise by considering progressively older age models for the
TE

341 Bucaramanga Fan. Our slip rate estimate is in rough agreement with conclusions by
EP

342 Diedrix et al. (2009), who documented by paleosismology eight main earthquakes

343 in the last 8300 years, with 400 to 1300 years recurrence intervals, and estimated
C

344 magnitude values of 6.5 to 7.0. Average recurrence interval documented by Diedrix
AC

345 et al. (2009) is 1000 years, which would translate into an accumulated displacement

346 of 3 m, if our 3 mm/yr spreading rate is considered. A 3 m displacement along a

347 seismogenic fault is expected to be associated with 7-7.5 M earthquakes (Wells and

348 Coppersmith, 1994), slightly greater than Diederix et al. (2009) estimates. We note

349 however that older age models for the Bucaramanga Fan would give smaller slip

350 rates, thus better agreement with Diederix et al. (2009) calculated values.
 15
ACCEPTED MANUSCRIPT

351 Considering that the city of Bucaramanga has a documented history of 390 years

352 without any catastrophic earthquake occurrence, a 1.2 m displacement could have

353 been accumulated since then, if the 3 mm/yr slip rate is considered.

354 Sediments of the Bucaramanga Fan have been clearly displaced by left-lateral fault

PT
355 shear, thus counterclockwise rotations are expected at fault walls, as already

356 observed in several strike-slip fault settings (Sonder et al., 1994; Kimura et al.,

RI
357 2011). On the other hand, our sites from the Bucaramanga Fan, located at 4-10 km

358

SC
from the fault, do not show significant rotations (Fig. 6).

359 Using the power law rheology model (England et al., 1985), in which the relation

U
360 between distance from a fault trace y (km) and vertical-axis relative rotation θ
AN
361 (rad.) is expressed as:

362
M

363 θ=arctan (((4Dsπ √n)/L) exp((−4π √ny)L))

364
D

365 Ds (km) is the displacement in one side of the fault, L/2 (km) is the length of the
TE

366 fault, and n is the stress exponent and describes the average mechanical behavior of
EP

367 the lithosphere (Sonder and England, 1986). Clearly, the lack of observed rotations

368 may imply that 1) fault locking is extremely weak, whereas fault walls are
C

369 extremely rigid and undergo minimal internal deformation, and/or 2) the upper
AC

370 crust west of the Bucaramanga Fault has a ductile behavior. Additional geophysical

371 data (e.g. a high-resolution GPS network across fault zone) are needed to fully

372 understand this issue, and better constrain the seismic hazard assessment related to

373 the Bucaramanga Fault activity.

374

375
 16
ACCEPTED MANUSCRIPT

376

377

378

379 Acknowledgements: Many thanks to Ecopetrol-ICP and the project: “Cronologia de

PT
380 la deformación en las cuencas Subandinas” for funding and supporting the field

381 work. Danilo Seccia (INGV) helped to realize Figure 3. GB acknowledges

RI
382 Colciencias for continuing support ARES. Thanks to Omar Montenegro for

383

SC
logistical and field work. We are grateful to two anonymous referees and to JSAES

384 Editor James Kellogg for providing careful reviews of our work.

U
385
AN
M
D
TE
C EP
AC
 17
ACCEPTED MANUSCRIPT

386 References

387

388 Ayala-Calvo, C., G. Bayona, A. Cardona, C. Ojeda, O. Montenegro, C. Montes, V.

389 Valencia and C. Jaramillo (2012), The Paleogene synorogenic succession in the

PT
390 northwestern Maracaibo block: Tracking intraplate uplifts and changes in

391 sediment-delivery systems. Journal of South America Earth Sciences, Special

RI
392 Edition on Tectonic and climatic shaping of the northern Andes and southern

393

SC
Caribbean margin 39, 93-111. Doi: 10.1016/j.jsames.2012.04.005

394 Bayona, G., Jimenez, G., Silva, C., Cardona, A., Montes, C., Roncancio, J., and

U
395 Cordani, U., (2010), Paleomagnetic data and K.Ar ages from Mesozoic units of
AN
396 the Santa Marta Massif: A preliminary interpretation for block rotations and

397 translations: Journal of South American Earth Sciences, v. 29, p. 817-831.

M

398 Bueno. E and A. Solarte (1994), Geología, Geotecnia y Comportamiento Erosivo

399 del Área de Reserva Forestal de Bucaramanga. Tesis para optar al título de
D

400 Geólogo. Escuela de Geología, UIS, Bucaramanga.

TE

401 Cande, S.C. and D.V. Kent (1995), Revised calibration of the geomagnetic polarity
EP

402 timescale for the late Cretaceous and Cenozoic. Journal Geophys. Res., 100,

403 6,093-6,095.
C

404 Campbell, C. J (1965), The Santa Marta wrench fault of Colombia and its regional
AC

405 setting. Fourth Caribbean Geological Conference. Memoir: 247-261. Trinidad.

406 Cediel, F., R. Shaw, and C. Cáceres (2003), Tectonic assembly of the northern

407 Andean Block, in Bartolini, C., Buffler, R. T., and Blickwede, J., eds., The

408 Circum-Gulf of Mexico and the Caribbean: Hydrocarbon habitats, basin

409 formation and plate tectonics, American Association of Petrolum Geologists

410 Memoir 79, p. 815-848.

 18
ACCEPTED MANUSCRIPT

411 De Porta. J (1959), La Terraza de Bucaramanga. UIS. Bol. de Geología No. 3. p.p:

412 5-13, Bucaramanga.

413 Diederix. H, C. Hernandez, M. T. Eliana, J. A. Osorio, P. Botero (2009), Memorias

414 Congreso Colombiano de Geología, Paipa- Boyaca.́ Resultados preliminares del

PT
415 primer estudio paleosismológico a lo largo de la falla de Bucaramanga.

416 England. P, G. Houseman and L. Sonder (1985), Length scales for continental

RI
417 deformation in convergent, divergent, and strike–slip environments: analytical

418

SC
and approxi- mate solutions for a thin viscous sheet model. J. Geophys. Res. 90,

419 3551–3557.

U
420 Fisher, R. A. (1953), Dispersion on a sphere, Proc. R.Soc. London, 217, 295–305.
AN
421 Flinch, J.F. (2003), Structural evolution of the Sinu-Lower Magdalena area(

422 Northern Colombia), in Bartolini, C., Buffler, R., and Blickwede, J., eds., The
M

423 Circum-Gulf of Mexico and the Caribbean: Hydrocarbon Habitats, Basin

424 Formation and Plate Tectonics, American Association of Petroleum Geologists

D

425 Memoir 79, p. 776-796.

TE

426 Hubach. E (1952), Interpretación Geológica de la erosión y de los deslizamientos

EP

427 en Bucaramanga y medidas de defensa, Serv. Geol. Nal. Bogotá. Informe 867

428 (inédito), 2-4.

C

429 Ingeominas (2008), Modelo de evolución morfotectónica cuaternaria basado en

AC

430 evidencias estructurales, neotectónicas y paleosismológicas de los principales

431 sistemas de falla en la región de Bucaramanga. Diederix. H, Torres. E. M,

432 Hernández. C and Botero. P. A. Bogotá

433 Ingeominas (2001a), Base de datos de fallas activas. Recopilación Bibliográfica.

434 Proyecto Compilación y Levantamiento de la Información Geodinámica, Montes

435 N., Sandoval A., Bogotá.

 19
ACCEPTED MANUSCRIPT

436 Ingeominas (2001b), Zonificación Sismo Geotécnica Indicativa del Área

437 Metropolitana de Bucaramanga, Fase II. Convenio realizado entre la CDMB e

438 Ingeominas, Bucaramanga.

439 Irving. E. M (1971), La evolución estructural de los Andes mas septentrionales de

PT
440 Colombia. Boletín Geológico, Ingeominas, Bogotá. 19 (2), 1-90

441 Julivert. M (1958), La morfoestructura de la zona de Mesas al SW de Bucaramanga.

RI
442 UIS. Bol. de Geología No. 1. p.p: 7-43, Bucaramanga.

443

SC
Julivert, M. (1963), Nuevas observaciones sobre la estratigrafía y tectónica del

444 Cuaternario de los alrededores de Bucaramanga. UIS. Bol. de Geología No. 15.

U
445 p.p: 41-56, Bucaramanga.
AN
446 Kammer, A., and J. Sanchez (2006), Early Jurassic rift structures associated with

447 the Soapaga and Boyacá faults of the Eastern Cordillera, Colombia:
M

448 Sedimentological inferences and regional implications: Journal of South

449 American Earth Sciences, v. 21, p. 412-422.

D

450 Kellogg. J. N. and W. E. Bonini (1982), Subduction of the Caribbean Plate and
TE

451 basement uplifts in the over- riding South American Plate: Tectonics, v. 1, no. 3,
EP

452 p. 251–276.

453 Kimura, H. N. Ishikawa, and H. Sato (2011), Estimation of total lateral

C

454 displacement including strike-slip offset and broader drag deformation on an

AC

455 active fault; tectonic geomorphic and paleomagnetic evidence on the Tanna fault

456 zone in central Japan Tectonophysics, 501(1-4):87-97 doi:

457 10.1016/j.tecto.2011.01.016

458 Kirschvink, J. L (1980), The least-squares line and plane and the analysis of

459 paleomagnetic data, Geo- phys. J. R. Astron. Soc., 62, 699–718.

460 López, J. A, M. A. Cuéllar, J. A Osorio, L. E. Bernal and E. Cortés (2008),

 20
ACCEPTED MANUSCRIPT

461 Pseudotaquilitas y el carácter paleosísmico de un segmento del Sistema de Fallas

462 de Bucaramanga (SFB), noreste del Muncipio de Pailitas, Departamento del

463 Cesar, Colombia. Boletín de Geología, Bucaramanga. 30 (2), 79-92.

464 Lowrie, W. (1990), Identification of ferromagnetic minerals in a rock by coercivity

PT
465 and unblocking temperature properties, Geophys. Res. Lett., 17, 159 – 162.

466 Marcera. M and P. Salamanca (1993), Cartografía geológica y estratigráfica a

RI
467 detalle y Zonificación geotécnica del sector oriental del Área Metropolitana de

468

SC
Bucaramanga. Tesis para optar al título de Geólogo. Escuela de Geología, UIS,

469 Bucaramanga.

U
470 McFadden, P. L., and M. W. McElhinny (1990), Classification of the reversal test
AN
471 in paleomagnetism, Geophys. J. Int., 103, 725–729.

472 Montes, C., Guzman, G., Bayona, G., Cardona, A., Valencia, V., and C. Jaramillo
M

473 (2010), Clockwise Rotation of the Santa Marta Massif and Simultaneous

474 Paleogene to Neogene Deformation of the Plato-San Jorge and Cesar-Ranchería

D

475 Basins: Journal of South American Earth Sciences, v. 29, p. 832-848.

TE

476 Mora, A. and Garcia A. (2006), Cenozoic Tectono-Stratigraphic Relationships

EP

477 between the Cesar Sub-Basin and the Southeastern Lower Magdalena Valley

478 Basin of Northern Colombia. AAPG 2006 Annual Convention. Houston, Texas.
C

479 Parra, M., Mora, A., Lopez, C., Rojas, L.E., and K.B. Horton (2012), Detecting
AC

480 earliest shortening and deformation advance in thrust-belt hinterlands: Example

481 from the Colombian Andes: Geology, v. 40, p. 175-178.

482 París, G. M, Machette. R, Dart and K. Haller (2000), Map and Database of

483 Quaternary Faults and Folds in Colombia and its Offshore Regions. U.S.

484 Department of the Interior U.S. Geological Survey. 66 p.

485 Pindell, J. L., S. C. Cande, W. C. Pitman III, D. B. Rowley, J. F. Dewey, J.

 21
ACCEPTED MANUSCRIPT

486 LaBrecque, and W. Haxby (1988), Plate- kinematic framework for models of

487 Caribbean evolution: Tectonophysics, v. 155, p. 121–138.

488 Prieto. G. A, Gregory C. Beroza . G. Cb, Barrett. S. A, López. G.A and M. Florez

489 (2012), Earthquake nests as natural laboratories for the study of intermediate-

PT
490 depth earthquake mechanics. Tectonophysics 570-571, 42–56

491 Rivera. L. A (1989), Inversion du Tenseur des Contraintes et des Mechanismes au

RI
492 Foyer a partir des Donnees de Polarite pour une Population de Seismes:

493

SC
Applicaton a l’Etude du Foyer de Sismicite Intermediaire de Bucaramanga

494 (Colombie): Doctoral Thesis, Universite de Strasbourg, France, 266 p.

U
495 Ross, I., P. Parra, C. Mora and C. Pimentel (2009), AFTA apatite fission track
AN
496 analysis constraints on the Mesozoic to Quaternary thermal and tectonic evolution

497 of the Middle Magdalena Basin and Santander Massif, Eastern Cordillera,
M

498 Bucaramanga area, Colombia. X Simposio Bolivariano: Exploración Petrolera en

499 Cuencas Subandinas. Cartagena, Colombia. Memoir.

D

500 Schneider, J., Pennington, W., Meyer, R., (1987), Microseismicity and focal
TE

501 mechanisms of the intermediate depth Bucaramanga nest, Colombia. J. Geophys.

EP

502 Res. 92, 13913–13926.

503 Sonder. L and P. England (1986), Vertical averages of rheology of the continental
C

504 lithosphere: relation to thin sheet parameters. Earth Planet. Sci. Lett. 77, 81–90.
AC

505 Sonder. L, J. Jones, C.H. Salyards, S.L and K. M. Murphy (1994), Vertical axis

506 rotation in the Las Vegas Valley shear zone, southern Nevada: paleomagnetic

507 constraints on kinematics and dynamics of block rotations. Tectonics 13, 769–

508 788.

509 Taboada, A., L. A. Rivera, A. Fuenzalida, A. Cisternas, H. Philip, A. Bijwaard, J.

510 Olaya, and C. Rivera (2000), Geodynamics of the northern Andes: Sub- ductions
 22
ACCEPTED MANUSCRIPT

511 and intracontinental deformation (Colombia), Tectonics, 19, 787–813.

512 Toro, J. (1990), The termination of the Bucaramanga Fault in the Cordillera

513 Oriental, Colombia: Masters Thesis, University of Arizona, Department of

514 Geosciences, Tucson: 60.

PT
515 Trenkamp, R., J. N. Kellogg, J. T. Freymueller, and H. P. Mora (2002), Wide plate

516 deformation, south- ern Central America and northwestern South America, CASA

RI
517 GPS observations, J. S. Am. Earth Sci., 15, 157–171.

518

SC
Tschanz. C, R. Marvin, J. Cruz, H. Mennert, and E. Cebula (1974), Geologic

519 evolution of the Sierra Nevada De Santa Marta. Geol. Soc. Am. Bull., 85, 269-

U
520 276.
AN
521 Ujueta, G. (2003), La Falla de Santa Marta-Bucaramanga no es una sola falla; son

522 dos fallas diferentes: La Falla de Santa Marta y la Falla de Bucaramanga:

M

523 Geología Colombiana No. 28, Universidad Nacional de Colombia, Bogotá.

524 Vargas, A., G and A. Niño (1992), Patrones de fracturamiento asociados a la Falla
D

525 Bucaramanga. Tesis, Universidad Industrial de Santander, Bucaramanga.

TE

526 Ward, D., Goldsmith, R., Cruz, J., and H. Restrepo (1973), Geología de los
EP

527 Cuadrángulos H-12, Bucaramanga y H-13, Pamplona, Departamento de

528 Santander, Boletín Geológico Ingeominas 21(1-3), 132 p.

C

529 Wells, D.L and K. J. Coppersmith (1994), New Empirical Relationships among
AC

530 Magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface

531 Displacement. Bulletin of the Seismological Society of America, Vol. 84, No. 4,

532 pp. 974-1002

533 Zijderveld, J. D. A. (1967), A.C. demagnetization of rocks: Analysis of results, in

534 Methods in Palaeomagnetism, edited by D. W. Collinson, K. M. Creer, and S. K.

535 Runcorn, pp. 254–286, Elsevier, New York.

 23
ACCEPTED MANUSCRIPT

536 Zarifi, Z., and J. Havskov (2003), Characteristics of dense nests of deep and

537 intermediate-depth seismicity. Adv. Geophys. 46, 237–278.

538 Zarifi, Z., J. Havskov and A. Hanyga (2007), An insight into the Bucaramanga nest

539 Tectonophysics 443, 93–10.

PT
540

RI
U SC
AN
M
D
TE
C EP
AC
 24
ACCEPTED MANUSCRIPT

541 Figure captions

542

543 Figure 1. Major tectonic and structural features of the NW margin of South

544 America. Plate velocity vectors with respect to stable South America are from

PT
545 Trenkamp et al. (2002). WC = Western Cordillera, CC = Central Cordillera, EC =

546 Eastern Cordillera, SMF = Santa marta Fault, BF = Bucaramanga Fault.

RI
547

548

SC
Figure 2. Geological map of the Bucaramanga Fan (scale 1:25,000, modified from

549 Ingeominas, 2001b) and location of the 14 paleomagnetic sites. See Figure 1 for

U
550 location.
AN
551

552 Figure 3. Seismicity in NW South American Plate recorded during 2006-2009 by

M

553 the Red Sismica Nacional de Colombia (RSNC)-Ingeominas.

554
D

555 Figure 4. Thermal demagnetization of a three-component IRM according to the

TE

556 method of Lowrie (1990) for six representative specimens.

EP

557

558 Figure 5. Orthogonal vector diagrams of typical demagnetization data, in situ

C

559 coordinates. Solid and open dots represent projections on the horizontal and vertical
AC

560 planes, respectively. Demagnetization step values are in °C.

561

562 Figure 6. Equal-area projections of the site-mean paleomagnetic directions from the

563 study area. Solid (open) symbols represent projection onto the lower (upper)

564 hemisphere. Open ellipses are the projections of the α95 cones about the mean

565 directions. The star represents the normal polarity geocentric axial dipole (GAD)
 25
ACCEPTED MANUSCRIPT

566 field direction (D=0°, I=14°) calculated for the Bucaramanga latitude (7.1° N).

567

568 Figure 7. Stratigraphic column of the Bucaramanga Fan, and stratigraphic variation

569 of magnetic polarity zones. Magnetic polarity zones are progressively numbered

PT
570 from the section bottom. Uncertain polarity intervals are grey and cross-hatched.

571 The crosses indicate failed sites.

RI
572

573

SC
Figure 8. Stratigraphic distance versus age correlation plots. Magnetic polarity

574 zones are correlated to polarity chrons from the GPTS (Cande and Kent, 1995).

U
575 Sloping lines are average sediment accumulation rates. Two different possible
AN
576 correlations are shown (see text for explanation).

577
M

578 Figure 9. (a) Geomorphological context in the study area, where the Suárez and

579 Bucaramanga faults are located. A distance of 2.5 km occurs between the
D

580 Bucaramanga Fan apex and the Suratá River. (b) Elevation contours around the
TE

581 area. Three zones are well defined (from E to W), the eastern wall of the
EP

582 Bucaramanga Fault with high topographic expression, the Bucaramanga Fan, and

583 the hanging wall of the Suárez Fault. Finally, three topographic profiles across the
C

584 Suratá River, the apex and Bucaramanga Fan are shown. BF = Bucaramanga Fault,
AC

585 SF = Suárez Fault.

586

587 Figure 10. Scheme showing the displacement of the Bucaramanga Fan apex from its

588 feeding Suratá River along the Bucaramanga Fault.

589 Table 1. Paleomagnetic results

 ACCEPTED MANUSCRIPT

PT
Table 1

RI
Paleomagnetic results from Bucaramanga Fan

SC
Geographic coordinates In Situ
Site Member Longitude ºW Latitude ºN N/n D (º) I (º) k α95 (º)
GJ186 Finos 73.132033 7.075063 10/8 179.0 13.4 7.75 21.2

U
GJ187 Finos 73.132031 7.075061 11/6 176.4 9.0 14.9 17.9
GJ188 Gravoso 73.143939 7.075496 12/0 - - - -

AN
GJ189 Organos 73.168091 7.036684 12/0 - - - -
GJ190 Organos 73.174056 7.049284 11/9 353.5 4.0 9.14 18.0
GJ191 Organos 73.138192 7.067906 6/0 - -

M
GJ192 Gravoso 73.132332 7.097189 12/5 176.7 9.2 12.61 22.4
GJ193 Organos 73.133936 7.090153 12/0 - - - -
GJ194 Organos 73.138862 7.087891 11/0 - - - -

D
GJ195 Organos 73.138862 7.087891 12/12 188.1 -19.0 12.79 12.6

TE
GJ196 Organos 73.162533 7.088209 11/8 181.5 -0.6 11.63 18.5
GJ197 Organos 73.161192 7.101659 12/8 357.9 7.6 44.45 8.4
GJ198 Organos 73.165264 7.100704 12/8 356.9 16.7 13.37 15.7
GJ199 Organos 73.148797 7.066336
EP
11/5 167.9 -1.3 3.47 48.2
C

The geographic coordinates are referred to WGS84 datum. Bedding is sub-horizontal. D and I are in situ declination and inclination; K and α95 are statistical
AC

parameters after Fisher (1953).

 75º 74º 73º
ACCEPTED MANUSCRIPT 80º 75º 70º 65º
15º

N
n accretionary wedge
ea
ri bb
h Ca
ut
So
OF
M a r a c a ib o

SMFNothern
10º
11º S a n ta M a rta B lo c k
M a s s if

PT
VDUP
CART C o c o s C C
P e r ija

nt
ro
P la te

f
Z one
W C E C

st
R a n g e 5º

ru
Th
RI

ra
L o w e r

ille
10º

rd
M a g d a le n a C R B

Co
rn
V a lle y Ariguani

ste
SC

Ea
graben
B a s in S o u th A m e r ic a n
Cen

P l a t e 500
0º N a z c a 0 1000
t ra l

U
P la te Km
El Ca
F aul
Z on

AN
9º
t
e

r me n

MONT
M a r a c a ib o
B lo c k
M c o
n o
F a
u l
t

nd
es 0 75 Km 150 Km
BF thern Z

D
o
aA
Sou

B
ri d
Me
8º
TE

S a n ta n d e r C u c u ta
Plate velocity Vectors
M a s s if URIB 1 cm/yr
Reverse fault
one

EP

strike-slip fault
BUCM Seismic lines
C h
(Ujueta, 2003)
C

B u c a ra m a n g a u c
7º M id d le a ri
m
M a g d a le n a a F
AC

FIGURE 2 a u
C e n tr a l C o r d ille r a V a lle y lt
B a s in
A n tic o b a r d e s
lin e
L o s C

M e d e llin

RION lin
e
6º A n
tic
c o
lt

E a s te rn b u
au

r c a
lF

A
C o r d ille r a
p a
Y o

Figure 1
 73º 11´0" W 73º 9´0" W 73º 7´0" W 73º 5´0" W

0 2 Km 4 Km

7º 12´0" N

Santander
Massif
ACCEPTED MANUSCRIPT

7º 10´0" N

er
á Riv
ra t
Su
m
te
ys

PT
lt s
au

Bu
zF

ca
re

ra m

RI
a
Su

ang
7º 8´0" N
Apex aF
SC
aul
tS
yst
U
em
AN

GJ197
M

7º 6´0" N
GJ198 GJ192
D

GJ195 GJ193
GJ196 GJ194
TE

GJ188
GJ187
EP

GJ186
7º 4´0" N
GJ191
F lo

GJ199
C
Suarez Fault system

ri d
aF
AC

au
lt

GJ190

GJ189 7º 2´0" N

Cretaceous Flow deposits

Triassic-Jurassic Limos rojos Member

Finos and Gravoso Bucaramanga Fan
Precambric to Paleozoic Member
Fault Organos Member
Covered Fault or lineament
Paleomagnetic sites

Figure 2
 79º W ACCEPTED MANUSCRIPT
75º 70º

200 km
depth[km]
0

PT
40
10º N

RI
80

U SC
120

AN
160
M B u c a ra m a n g a
N e s t
200
D

N a z c a
TE

5º
P la te
EP
C
AC

S o u th A m e r ic a n
P la te
0º
Figure 3
 ACCEPTED MANUSCRIPT
0.6 T < coer.< 2.7 T
0.12 T < coer. <0.6 T
coer. < 0.12 T
10E-1 A/m 1 A/m 10E-1 A/m
4 4

1
3 3

PT
GJ194 GJ197 GJ194
2 2

RI
SC
1 1

U
AN
0 0 0
20 100 200 300 400 500 600 20 100 200 300 400 500 600 20 100 200 300 400 500 600
T (xC) T (xC) T (xC)

M
10E-1 A/m 10E-1 A/m 10E-2 A/m
7

D
6 5 TE 3
5 4
EP

GJ186 GJ190 GJ192

4 2
3
C

3
2
AC

2 1
1
1

0 0 0
20 100 200 300 400 500 600 20 100 200 300 400 500 600 20 100 200 300 400 500 600
T (xC) T (xC) T (xC)

Figure 4
 N Up N Up
ACCEPTED
N MANUSCRIPT
Up
200º

500º

600º

Unit= 537.e-06 A/m

GJ190-2 Unit= 411.e-06 A/m
Unit= 508.e-06 A/m

600º
GJ192-9 200º
GJ195-1
250º

500º
W 680º
500º E W E W E

PT
680º
W E W 600º E W 680º 600º E
200º

(a) (b) (c)

RI
500º

600º

SC
380º

500º

U
200º 250º

S Down S Down S Down

AN
N Up200º N Up N Up

460º

M
600º 200º

D
380º

600º 460º

GJ197-3 Unit= 266.e-06 A/m TE

GJ198-2 680º Unit= 258.e-06 A/m GJ199-3
380º Unit= 83.9e-06 A/m
W 680º E W E W E
EP
W E W E W E
680º 550º
460º
380º
200º (d) (e) (f)
C

460º
AC

200º

S Down S Down S Down

Figure 5
 ACCEPTED N
MANUSCRIPT

190 197

198

PT
RI
U SC
270 90

AN
M
D
TE
EP

195
186
C

192
187 199
196
AC

180
Geocentric axial dipole field direction
expected at Bucaramanga (Lat 7.13º N)
Mean paleomagnetic direction
(D = 357.5 I = 2.0 K = 41.63 a 95 = 8.1)

Figure 6
Limos Rojos
Member
220

188
Gravoso
Member

200 192
ACCEPTED MANUSCRIPT

186
Member

R3
Finos

180 187

PT
160 195
191
RI
SC
140
190 N3
U
AN

120 193
M

194
D

189
100 199
TE

R2
EP

197 N2
Organos Member

80
C

196 R1
AC

60
198 N1

40

20

0
Figure 7
 ACCEPTED MANUSCRIPT

A B
Limos Rojos

Limos Rojos
Member

Member
220 220

PT
188 188
Gravoso

Gravoso
Member

Member
200 192 200 192

RI
186 186
Member

Member
R3 R3
Finos

Finos
180 187 180 187

SC
160 195 160 195
191 191

U
140 140
190 N3 190 N3

AN
120 193 120 193
194 194

M
189 189
100 199 R2 100 199 R2

197 N2 197 N2
Organos Member

Organos Member
D
80 80
196 R1 196 R1

TE
60 60
198 N1 198 N1

EP
40 40

20 20

C
0 AC SYSTEM
0

Chrono
P LI OC E NE P LE I S TOC E N E SYSTEM

Chrono
P LI OC E NE P LE I S TOC E N E
E ar l y Pl i oc e n e Middle Pliocene Late Pliocene Early Pleistocene M. Pleistocene E ar l y Pl i oc e n e Middle Pliocene Late Pliocene Early Pleistocene M. Pleistocene
Stages Stages
Za nc l e an Piacenzian Gelasian Santern. Emilian Sicilian Za nc l e an Piacenzian Gelasian Santern. Emilian Sicilian
5

Ma

1
Ma
POLARITY POLARITY
G I L B E R T G A U S S M A T U Y A M A BRUNHES CHRONS G I L B E R T G A U S S M A T U Y A M A BRUNHES CHRONS
POLARITY POLARITY
Thvera

Sidufjall

Nunivak

Cochiti

Mammoth

Kaena

Olduvai

Jaramillo

Thvera

Sidufjall

Nunivak

Cochiti

Mammoth

Kaena

Olduvai

Jaramillo
Polarity Polarity
Subchrons Subchrons

Figure 8
 ACCEPTED MANUSCRIPT

(a) (b)
73º 9' W 73º 5' W 73º 1' W
7º 10' N C
A`

PT
2.5 km

RI
B`
7º 8' N

U SC
A

AN
7º 6' N

M
B

D
C`

TE
7º 4' N
A A’
1050 m

EP
950 m
850 m
750 m SF BF
650 m
C
Reverse fault B’
AC

B Apex
Strike-slip fault
1050 m
950 m
850 m
750 m BF
650 m

1050 m C C’
950 m
850 m
750 m
650 m

Figure 9
 r
ve
ACCEPTED MANUSCRIPT

Ri
á
at
Stage A

r
Su
Bucaramanga Fault

PT
D

RI
U SC
AN
M
r

Stage B
ve
Ri
á

D
at
r
Su

TE

2.5 Km Bucaramanga Fault

C EP
D
AC

Figure 10
 ACCEPTED MANUSCRIPT
Highlights

Paleomagnetic investigation of a Plio-Pleistocene (?) continental alluvial fan

juxtaposed to the Bucaramanga Fault, and horizontally displaced by 2.5 km with
respect to its feeding river. Nine reliable paleomagnetic directions define a succession
of magnetic polarity zones, correlated with several tracks of the Plio-Pleistocene

PT
magnetic polarity time scale. If the youngest age model is considered, most recent
sediments of the Bucaramanga Fan can be reasonably dated at 0.8 Ma (Brunhes-

RI
Matuyama chron transition), translating into a maximum 3 mm/yr slip rate for the
Bucaramanga Fault.

U SC
AN
M
D
TE
C EP
AC