The role of inherited extensional fault segmentation
and linkage in contractional orogenesis: a
reconstruction of Lower Cretaceous inverted rift
basins in the Eastern Cordillera of Colombia
Andre?s Mora,n Tatiana Gaona,w Jonas Kley,z Diana Montoya,w Mauricio Parra,n Luis Ignacio
Quiroz,? German Reyesw and Manfred R. Streckern
nInstitut f?r Geowissenschaften, Universitt Potsdam, Golm, Germany
wIngeominas, Bogota? , Colombia
zInstitut f?r Geowissenschaften, Universitt Jena, Jena, Germany
?SmithsonianTropical Research Institute, Balboa, Republic of Panama?
ABSTRACT
Lower Cretaceous early syn-rift facies along the eastern ?ank of the Eastern Cordillera of Colombia,
their provenance, and structural context, reveal the complex interactions between Cretaceous
extension, spatio-temporal trends in associated sedimentation, and subsequent inversion of the
Cretaceous Guatiqu|? a paleo-rift. South of 41300N lat, early syn-rift alluvial sequences in former
extensional footwall areas were contemporaneous with fan-delta deposits in shallow marine
environments in adjacent hanging-wall areas. In general, footwall erosionwas more pronounced in
the southern part of the paleorift. In contrast, early syn-rift sequences in former footwall areas in the
northern rift sectors mainly comprise shallow marine supratidal sabkha to intertidal strata, whereas
hanging-wall units display rapid transitions to open-sea shales. In comparisonwith the southern
paleo-rift sector, fan-delta deposits in the north are scarce, and provenance suggests negligible
footwall erosion.The southern graben segment had longer, and less numerous normal faults, whereas
the northern graben segment was characterized by shorter, rectilinear faults.To the east, the graben
systemwas bounded by major basin-margin faults with protracted activity and greater throw as
comparedwith intrabasinal faults to the west. Intrabasinal structures grew through segment linkage
and probably interacted kinematically with basin-margin faults. Basin-margin faults constitute a
coherent fault system that was conditioned by pre-existing basement fabrics. Structural mapping,
analysis of present-day topography, and balanced cross sections indicate that positive inversion of
extensional structureswas focused along basin-bounding faults, whereas intrabasinal faults remained
una?ected andwere passively transported by motion along the basin-bounding faults.Thus, zones of
maximum subsidence in extension accommodated maximum elevation in contraction, and former
topographic highs remained as elevated areas.This documents the role of basin-bounding faults as
multiphased, long-lived features conditioned by basement discontinuities. Inversion of basin-
bounding faults was more e?cient in the southern than in the northern graben segment, possibly
documenting the inheritance and pivotal role of fault-displacement gradients. Our observations
highlight similarities between inversion features in orogenic belts and intra-plate basins,
emphasizing the importance of the observed phenomena as predictive tools in the spatiotemporal
analysis of inversion histories in orogens, as well as in hydrocarbon and mineral deposits exploration.
INTRODUCTION
Abetter understanding of the processes and styles govern-
ing the inversion of inherited normal faults in orogenic
belts is a fundamental problem in structural geology. For
example, in orogens the presence of structural and strati-
graphic traps related to former rift tectonics may guide the
migration of fossil fuels and the emplacement of mineral
resources (e.g. Uliana et al., 1995). In such settings, the
origin and evolution of inverted structures are therefore
important in improving exploration. Furthermore, if pro-
nounced pre-existent anisotropies related to extensional
Correspondence: Andre? s Mora, Institut f?r Geowissenschaften.
Universitt Potsdam, Karl Liebnecht-Str 24, D14476 Potsdam-
Golm,Germany. E-mail: andres.mora@geo.uni-potsdam.de
BasinResearch (2009) 21, 111?137, doi: 10.1111/j.1365-2117.2008.00367.x
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
111
tectonics are favorably orientedwith respect to the greatest
horizontal stress during a subsequent contractile phase,
reactivation of such structures may signi?cantly impact
the spatiotemporal evolution of deformation in the entire
mountain belt (e.g. Lowell, 1995). Hence the width of an
orogen may be a direct consequence of this type of interac-
tion and its growth may not be governed by the principles
of wedge mechanics (e.g. Davis et al., 1983) documented in
regions where such inherited structures do not exist (Hil-
ley et al., 2005). In addition, the polarity of normal faults,
associated disparities in syntectonic sediment thickness,
and existence of extensional transfer structures may exert
an important control on orogenic evolution during con-
traction (e.g. Hayward &Graham, 1989).
Positive basin inversion involves the partial or total re-
versal of motion on extensional faults subjected to super-
imposed compressional stresses (Williams et al., 1989).
The phenomenon of basin inversion has been documen-
ted worldwide. For example, mild to moderate inversion
in intraplate basins has been observed in numerous North
Atlantic basins (Cartwright, 1989; Chapman, 1989; Hay-
ward & Graham, 1989; Roberts et al., 1993; Sinclair, 1995;
Thomas & Coward, 1995) or in the North German Basin
(Kossow et al., 2000; Kossow & Krawczyk, 2002; Gemmer
et al., 2003). In addition, strongly shortened half grabens
have been documented in the Alps (Gillcrist et al., 1987),
with prominent examples found in the Central Andes
(Kley&Monaldi, 2002) and Atlas mountains (Beauchamp
et al., 1999;Teixell et al., 2003).
Huyghe & Mugnier (1995) suggested that the nature of
inversion in mildly inverted intraplate basins is di?erent
from inversion in orogenic belts where extensional struc-
tures are typically fully reversed, creating topography and
exposure of former basin strata (e.g. Badley, 2001). The
time span between the extensional phase and the onset of
subsequent contraction is often larger in orogenic belts
compared with intraplate basins (Huyghe & Mugnier,
1995).Therefore, by the time a formerly rifted lithosphere
undergoes contraction in an orogenic setting, it may have
recovered its pre-extensional strength (Huyghe & Mug-
nier, 1995). In this context, it is an important question if
di?erent crustal properties in?uence variations in inver-
sion styles in fully developed orogenic belts in comparison
with intraplate basins.
In intraplate basins undergoing mild to moderate inver-
sion, high-quality seismic images have improved our 3D
understanding of the behavior of individual normal faults
and linked graben systems during subsequent compres-
sion (Cartwright, 1989; Chapman, 1989). Furthermore,
laboratory 3D analog models of inversion have helped
visualize the ?rst-order principles and structures asso-
ciated with these processes (Eisenstadt & Withjack, 1995;
McClay, 1995).
In the more severely inverted basins of many orogens,
however, detailed ?eld-based accounts regarding the
mechanical interaction of segmented former extensional
faults, the nature of displacement transfer, and how such
structures are incorporated into the orogenic realm are
rare (Hayward & Graham, 1989; Garc|? a Senz, 2002; Kley
et al., 2005; Monaldi et al., 2008). These problems are of-
ten compounded by a general lack of good seismic images
with unequivocal extensional features. Furthermore, due
to the high degree of superimposed shortening, a precise
assessment of extensional features is made di?cult at the
outcrop level, as in the case of theWesternAlps (e.g.Gillcr-
ist et al., 1987; Coward et al., 1989; Graciansky et al., 1989;
Hayward&Graham,1989). In these extreme cases, a palin-
spastic reconstruction may help decipher the principal
extensional features and original location of ancestral
extensional structures (e.g. Nemcok et al., 2001). Finally,
prominent along-strike changes in the patterns and styles
of deformation have been observed in many inversion oro-
gens (Grier et al., 1991; Carrera et al., 2006). However, few
studies have unambiguously demonstrated that those
changes are indeed determined by along-strike variations
in the modes of previous extensional deformation, fault
interaction, and linkage.
The Eastern Cordillera of Colombia (Fig.1) has experi-
enced much less deformation than many other inversion
orogens (e.g. Colletta etal.,1990;Mora etal., 2006). It there-
fore has the advantage of furnishing valuable information
on ancient normal fault arrays, the characteristics of asso-
ciated syn-rift depositional settings, and the lateral evolu-
tion before contraction.
In this study, we document a previously unrecognized
pattern of extensional fault interaction, deduced mainly
from early syn-rift stratigraphic sequences associatedwith
former normal faults. First, we determine the local charac-
teristics and lateral extent of syn-rift units in light of the
Early Cretaceous paleotopography in order to use the
sedimentary facies associations as proxies for past normal
fault evolution and linkage. Second, we evaluate the e?ects
of the pre-existing extensional fault arrays on Cenozoic
contractional deformation. In contrast with the study of
Mora et al. (2006), we use sedimentology to document that
a pattern of selective fault reactivation is not onlydue to the
orientation of the compressive stress ?eld with respect
to the pre-existent faults (Mora et al., 2006), but is also
relatedwith the pattern of extensional fault growth.
GEOLOGIC SETTING OF THE EASTERN
CORDILLERA
The Colombian Eastern Cordillera is an atypical Andean
foreland fold and thrust belt. In stark contrast with other
Andean foreland fold and thrust belts it appears to be more
in?uenced by inherited structures rather than subduction
related orogenesis (e.g. Colletta etal., 1990).The spatial co-
incidence between the Eastern Cordillera (Fig. 1) and a
Neocomian extensional province was proposed byColletta
et al. (1990), although it has also been suggested that this
region may represent a simple ramp basin that subsided
more than adjacent regions during the Early Cretaceous
(Roeder & Chamberlain, 1995), similar to the Cameros
Basin in northern Spain (Guimera et al., 1995). Indeed, in
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
112
A. Moraet al.
the Eastern Cordillera it had been impossible to recognize
and precisely locate former rift-controlling faults. How-
ever, basin-analysis studies of this region favor a rift model
(Sarmiento-Rojas et al., 2006). In addition, structural
reconstructions (Mora et al., 2006) unambiguously show
that the Early Cretaceous basin along the eastern ?ank
of the Eastern Cordillera was a rift zone. These authors
proposed a reactivation mechanism restricted to an in-
verted frontal master fault (Servita? fault), where the entire
eastern ?ank of the EC was passively uplifted while the
internal structures underwent virtually no contractile
reactivation. The main reason for selective fault reactiva-
tion in the Cretaceous extensional province was inferred
to be related to the orientation of the tectonic stress
?eld with respect to the pre-existing anisotropies (Mora
et al., 2006).
CD
GD
SA
EF
SF
SjF
NF
GF
LF
TF
GT
73?W
4?
ECNAZCA
PLATE
SOUTH
AMERICAN
PLATE
CARIBBEAN 
PLATE
0?
6?
?73?
STRATIGRAPHY
Terrace
Alluvium
Quaternary
STRUCTURES
Reverse fault
Normal fault Anticline
Syncline
Pre-Devonian Quetame Group
Upper Paleozoic 
sedimentary units
Lower Cretaceous basal units 
Macanal Formation
Caqueza Formation
F?meque Formation
Une Formation
Chipaque  Formation
Guadalupe Group
Paleocene-Middle Eocene units
Carbonera Formation
Le?n Formation
Lower Guayabo Formation
Upper Guayabo Formation
Pre-
Cretaceous
Lower 
Cretaceous
Upper
Cretaceous
Paleogene
Neogene
Alluvial Fan
Strike-slip fault
CD: Chingaza Dome
EF: Esmeralda Fault
FA: Farallones Anticline
GD: Guavio Dome
GF: El Garabato Fault
GT: Guaicaramo Thrust
LF: Lengup? Fault
LMF: Los Medios Fault
NF: Naranjal Fault
SA: Santa Mar?a Anticline
SjF: San Juanito Fault
SF: Servit? Fault
TF: Tesalia Fault
Fig 16c
Fig 16b
Fig 16a
4?N
Fig 17
Villavicencio
20 km
5?N
Fig.1. (a) Geodynamic setting and simpli?ed topography of the northern Andes.The study area is indicated by the box. (b) Geological
map of the Farallones deMedina area with location of the cross sections in Fig.15.Map depicts areas covered in Figs12, 13, 17, 18 and
locations of balanced structural pro?les.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
113
Role of inherited extensional fault segmentation and linkage
The investigated area of this study is located in the
Farallones de Medina range of the Eastern Cordillera. This
area (Figs1and 2) is located along the eastern foothills of the
mountain belt and exposes lower Cretaceous syn-rift
facies in di?erent tectonic blocks.Although the structural re-
lationships in this region are now well known (e.g. Mora
etal., 2006), the lateral distribution of syn-rift facies and their
relationship to the structures is not yet fully understood.
Phyllites and green schists of the pre-Devonian Que-
tame Group constitute the oldest rocks in the area and
are overlain by the sedimentary Farallones Group. These
units comprise Devonian nonmarine to shallow marine
sandstones (Areniscas de Gutierrez Formation); Carboni-
ferous continental red-bed sequences (Capas del Valle del
Guatiqu|? a Formation), and Permian shallow marine car-
bonates (Gachala? Formation).
The upper Paleozoic rocks are superseded by ca. 5-km-
thick Cretaceous sequence (Fig. 3).The lowermost Berria-
sian and Valanginian units overlying the Paleozoic rocks
are the main target units of this study, as they represent
the early syn-rift strata. However, the Berriasian strata
cannot be grouped into one single, basal lithostratigraphic
unit as these strata display rapid lateral facies changes. For
example, calcareous units overlying the pre-Cretaceous
basement have been grouped into the Calizas del Guavio
Formation (Ulloa & Rodr|? guez, 1979); coeval strata com-
posed of polymictic gravels, breccias, ?uvial sandstones
and associated overbank ?nes constitute the Brechas de
Buenavista Formation (Dorado, 1992). In other locations,
shallow marine terrigenous lithic sandstones, polymictic
conglomerates, and subordinate amounts of gray siltstones
of Early Cretaceous age that overly Paleozoic rocks belong
to the Bata? Formation (B?rgl, 1961; Geyer, 1973; Ulloa &
Rodr|? guez, 1979; Etayo-Serna et al., 2003, Fig. 2). All of
these units are therefore lateral equivalents and are over-
lain by an ammonite-dominated shale unit (Lutitas de
Macanal Formation, Fig. 2).
METHODS
Our study focuses on Berriasian and Valanginian strata re-
cording the earliest syn-rift sedimentation in the Farallones
de Medina area.We produced a ?eld structural map of the
studyareabased on1: 25 000and1:10 000 topographicmaps
and aerial photographs (Fig.1). Careful ?eld structural map-
ping helped detect and de?ne the best exposures of the early
syn-rift units.Where possible, we chose pro?les displaying
well exposed contactswith the underlyingPaleozoic substra-
tum and used this unconformity as a marker horizon.We
measured detailed (1: 200) stratigraphic columns using a Ja-
cob sta?. In the gravel intervals, we analyzed clast composi-
tions in order to determine provenance (Table1). Additional
quantitative petrologic data is mostly abstent in those non-
conglomeratic intervals. Paleocurrent measurements of
groove castsweremainly taken in turbiditic sandstone strata.
However, paleocurrent data is scarce in most of the local-
ities. In addition, we integrated our datawith published sec-
tions (Stibane, 1968; Dorado, 1992; Parra, 2000; Geostratos,
2005; Rinco? n & Ta? mara, 2005). The measured sections
(Fig. 3) were described according to10 di?erent facies asso-
ciations. The graphic symbols in the pro?les, representing
various lithologic and primary features observed in the ?eld
and reported in our stratigraphic pro?les can be seen in
Fig. 4. Sections were correlated by following key horizons in
the ?eld or on aerial photographs based on prominent
topographic features. Using these data, we constructed a
map of early syn-rift facieswith the interpreted sedimentary
environments in this extensional setting during the Early
Cretaceous. To expand our analysis of superimposed con-
tractional deformation, we also constructed three balanced
cross sections based on outcrop and seismic data.
SEDIMENTARYCHARACTERISTICS OF
THE EARLYCRETACEOUS RIFT-BASIN
FILL
Buenavista formation
Facies association B1
This facies association constitutes the basal conglomerate
at localities 5, 6, 7 (Fig. 6) 11,12 (Fig. 7) and16, and most of
the sequence at localities1and 2 (Fig. 5). Its thickness typi-
cally ranges from10 to 40m.This association is character-
ized by a set of very thick beds, either with lenticular
geometry or irregular geometry, occasionally lacking in-
ternal strati?cation.The individual beds are composed of
poorly sorted cobble-sized, subrounded conglomerates (in
localities 5, 11and12 some 25^60 cm boulders are also pre-
sent), mostly supported by a sandy to silty matrix. The
cobble-size fraction is predominantly monomictic quart-
zose. Sorting is commonly poor. Often, there are inter-
bedded pebble-conglomerates and pebbly conglomeratic
sandstones in well-de?ned beds with variable thickness
and inverse grading.
We interpret these rocks to have been deposited in an
alluvial-fan system. Matrix-supported conglomeratic
beds with sandy matrix and irregular geometry may corre-
spond to debris- ?ow deposits, whereas conglomeratic
beds with well-de?ned geometry re?ect con?ned channel
?ows during stream-dominated deposition (e.g. Rust &
Koster, 1984; Nemec & Postma, 1993). At locality 1 (Fig. 5),
there is a clear lateral change toward marine facies (i.e.
black shales with ammonites) suggesting that in some
localities these conglomeratic beds are transitional be-
tween ?uvial and coastal environments.
Facies association B2
This facies association is characterized by poorly sorted
breccias and coarse matrix-supported conglomerates lack-
ing internal organization or well de?ned bedding surfaces.
This facies association is a lateral equivalent of shallow
marine units because it normally outcrops adjacent to
ammonite bearing black shales. The thickness is highly
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
114
A. Moraet al.
Fig. 2. (a) Reconstructed distribution of Early Cretaceous faults.The numbers are elevations above sea level in certain localities.
(b) Locations of the stratigraphic sections quoted in the text.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
115
Role of inherited extensional fault segmentation and linkage
variable and the lateral continuity is restricted.This facies
association overlies the basal Cretaceous unconformity at
localities 3 (Fig. 5), 2, 8, 10 and 15 (Fig. 8), and at about
375m above this unconformity, at locality 7 (Fig. 6).Typi-
cally, this facies consists of boulders (460 cm in diameter)
and cobbles in a pebbly to sandy matrix. Some boulders
exceed 2m in diameter; angular to sub-angular clasts
dominate. Owing to a general lack of organization and a
high fraction of metamorphic basement clasts this facies
association di?ers from the basal conglomerates in
facies association B1 (Table 1, Quetame Group phyllites),
except at outcrop locality 10, where these clasts are
virtually absent.
Owing to the poor strati?cation and sorting, abundant
angular fragments, and limited lateral continuity, the ma-
trix-supported conglomerates of facies association B2 are
interpreted as debris ?ow deposits (Rust & Koster, 1984;
Nemec & Postma, 1993; Blair & McPherson, 1998).
The rapid lateral transition into marine facies (e.g. facies
Table1. Clast count method
Section/m
above base
Composition of fragments
Sedimentary
rocks
Unstable-
basement
rocks
1. Puente Quetame
68 100 0
70 100 0
61 100 0
48 100 0
25 100 0
19 100 0
16 100 0
2. Alto del Tigre
437 100 0
426 100 0
415 100 0
405 17 83
387 24 76
363 57 43
354 52 48
263 67 33
137 71 29
24 56 44
13 100 0
3. Paujil
Lower
segment
o30 470
Upper
segment
65 35
4. Florida Creek
450 Observed
5. Calvario
2 100 0
6. La Esfondada
6 100 0
5 (n) 45 55
7. San Isidro-Lajitas
525 71 29
523 89 11
515 62 38
503 57 43
497 68 32
484 97 3
447 85 15
422 74 26
1 100 0
8. Carpanta
118 53 47
90 87 13
45 64 34
17 95 5
9. El Tigre Creek
10. La Ardita Creek
248 100 0
224 100 0
195 100 0
164 100 0
142 100 0
118 100 0
102 100 0
Table1. (Continued)
Section/m
above base
Composition of fragments
Sedimentary
rocks
Unstable-
basement
rocks
72 100 0
29 92 8
24 90 10
18 80 20
11. Palomas
12 100 0
15 100 0
18 100 0
12. Gachaluno River
14 100 0
9 100 0
73 100 0
55 100 0
159 100 0
13. Cascadas
14. Chivor Alto
15. Batatas
59 100 0
43 100 0
18 100 0
10 100 0
16.Malacara
17. Santa Rosa
18.Mina El Porvenir
After a general inspection of the outcrop and rough analysis of the clasts
present, an arbitrary face was selected in order to count the composition
of the clasts in every single gravel size grain (up to boulder size).The frac-
tions were separated in clasts of sedimentary rocks, milky quartz from
veins, feldespars (never found) and unstable or clearly discernible base-
ment lihologies.We took in this table the sedimentary clasts plus the
basement rocks as100.Then put in the table the amount of each fraction
with respect to the total. Our goal was only to detect the presence or ab-
sence of basement rocks.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
116
A. Moraet al.
associations M1) indicates a fan delta in a shallow marine
setting.
Facies association B3
These units comprise tabular to lenticular beds, 0.3 to sev-
eral meters thick, with medium-to coarse-grained lithic
sandstones and laterally equivalent lithic conglomeratic
sandstones and conglomerates. Medium to thick beds (up
to 1m thick) of siltstones and mudstones are interbedded
with the upper sandstones of this facies association. It is
well developed at localities 2 (intervals between 30 and
83m and, between 325 and 383m. Fig. 5) and 6 (lower seg-
ment of the basal syn-rift unit, Fig. 6).The lower 40m at
locality 6 (Fig. 6) have a ratio of sand to mud layers of about
90 : 10. Amalgamated sandstone beds with lenticular geo-
metry are abundant, with frequent inverse grading and
occasionally with planar cross bedding. However, in the
upper 150m at locality 6 the sand to mud layer ratio is
40 : 60. This section contains frequent thin to very thick-
bedded (sensu Ingram,1954) red-colored siltstones to sandy
siltstones with pervasive mottling.These are interbedded
with medium to thick-bedded lithic sandstones (petrogra-
phy after Parra, 2000) with normal grading and channel
geometry. Amalgamated sandstone beds very often display
lateral accretion surfaces and have thin basal conglomera-
tic lags.
The amalgamated lithic sandstone beds in the lower
40m at locality 6 and 2 are interpreted as bars in a braided
river system (e.g.Collinson,1996;Miall,1996). In the upper
150m of locality 6, the channelized sandy ?ning-upward
sequences with lateral accretion surfaces suggest a mean-
dering river environment. Accordingly, interbedded silt-
stones and sandy siltstone beds with pervasive mottling
represent ?oodplain deposits and overbank ?nes (Miall,
1985).
Calizas Del Guavio formation
Facies association C1
Facies association C1 is well developed at localities 18
(between 160 and 215m), 15 (between 95 and 160m) and 14
Stratigraphic interval 
of interest in this studyBATA`
Fig. 3. Meso-Cenozoic stratigraphy of the eastern ?ank of the Eastern Cordillerawith the stratigraphic interval of interest highlighted
(Modi?ed fromMora et al., 2006).
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
117
Role of inherited extensional fault segmentation and linkage
(discontinuous interval between 35 and135mFigs 2 and 8),
and it consists of two distinct inter?ngering cycles with a
thickness of 0.5^2m. First, laminated dolomitic mud-
stones (at locality 15) or laminated calcareous mudstones
(at localities 14 and 18) alternate with thin-to medi-
um-bedded nodular calcareous mudstones, often with
chicken-wire growths, enterolithic folding, and fenestrae.
Desiccation cracks occur on top of some of the beds in this
unit.The second unit comprises laminated structures in-
terpreted to be stromatolites (Kalkowsky, 1908; Logan
et al., 1964; Tucker & Wright, 1990; Riding, 2000). In
addition, we found a clothed texture re?ecting syn-
depositional processes at locality15, also interpreted to re-
?ect microbial origin (trombolites sensu Riding, 2000).
At localities 18 and 15a these associations are also inter-
bedded with very thick black shale and black marl hori-
zons, occasionally with laminae rich in pyrite.
The ?rst unit is characteristic of evaporite pseudo-
morphs. For example, nodules and enterolithic structures
(Fig. 9) resemble diagenetic structures formed in modern
Limestone
Dolomites
Sandy limestones
Stromatolites with hemispheroidal fabric
Stromatolites with laminated fabric
Marlstones
Sandstone
Conglomerate
Conglomeratic sanstone
Claystone
Silstones
Evaporites pseudomorphs with nodular fabric
Pyrite nodules
Calcareous concretions
Pyrite laminae
Breccia
SILICICLASTIC SEDIMENTARY ROCKS
CARBONATE SEDIMENTARY ROCKS
Micrite interlayering
Chickenwire structures
Sandy mudstone
Shale
Red mudstones
Muddy sanstone
Evaporites pseudomorphs with enterolithic structures
Muddy limestones
Calcareous sandstone
Corals
Echinoderms
Brachiopods
Crinoids
Bivalves
Gasteropods
Plant remains
Trunk remains
Parallel lamination
Continuous Wavy bedding
Discontinuos wavy bedding
Crude bedding
Bioturbation
Rootlets
Organic matter
Fault
MO
GypsumY
Synsedimentary folding
Ripples
Amonites
Erosion surface
Imbrication
Slumps
Groove cast
Normal grading
Inverse grading
Angular Unconformity
Planar cross bedding with
tangential foresets
Planar cross bedding
Foraminifera
Lenticular bedding
Trough cross bedding
Flaser bedding
Fig.4. List of symbols and abbreviations and their meaning.These symbols and abbreviations were used in Figs 5, 6, 7, 8, and10.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
118
A. Moraet al.
anhydrites (Kendall & Harwood, 1996; Alsharhan & Ken-
dall, 2003). Alternations of dolomitic mudstones and an-
hydrite are common under hypersaline conditions when
calcium sulfate precipitates and residual waters are en-
riched in magnesium (Alsharhan & Kendall, 2003). Facies
association C1 thus suggests deposition under conditions
typical of coastal sabkhas (Warren&Kendall,1985; Alshar-
han&Kendall, 2003). In combinationwith the desiccation
cracks, these characteristics re?ect arid to hyperarid cli-
mate conditions (Tucker &Wright, 1990).
The second, stromatolite-bearing unit is interpreted to
indicate a temporary changeover from supratidal sabka to
intertidal conditions, similar to present-day environ-
ments reported inDubai (Tucker &Wright,1990).The pre-
sence of black shales and marls interbeddedwith the other
units suggests subtidal conditions, possibly associated
2. Alto del Tigre
Naranjal Fault Servit? Fault
W E
3. Paujil Creek
B2
B1
M2
B1
B3
B1
M3
B1
B3
B1
B3
1. Puente Quetame
B1
B3
B1
M3
Fig. 5. See locations in Fig. 2. Stratigraphic logs located in the southernmost segment of the study area. Contrasting facies in the
di?erent pro?les coincides with di?erent faulted domains.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
119
Role of inherited extensional fault segmentation and linkage
with an euxinic basin.Overall, these deposits appear to in-
dicate oscillating sea level in a sustained arid environment.
Facies association C2
This facies association is characterized by tabular beds of
mudstones, marls, and intraclastic wackestones, well de-
veloped at locality 15c (Fig. 8, between 0 and 95m). The
beds commonly contain disseminated bivalve shells (bio-
clastic wackestones). Locally bivalve packstones are found.
In this facies association, well- sorted bioclasts generally
do not occur in growth position and are sub-horizontal
with respect to bedding. Occasionally, poorly sorted bio-
clasts are found in di?erent positions.
We interpret this facies association to be related to in-
ternal to intermediate parts of a carbonate platform envir-
6.Esfondada 7a. Lajitas
San Juanito Fault
W E
20
40
60
80
100
120
260
140
160
180
200
220
240
280
300
320
340
360
380
400
420
440
460
480
500
520
B1
C3
M3
M3
B2
Si Sa Gr
Md Pc
Wc Gr Bd
Cl Si Sa Gr
Md Pc
Wc Gr Bd
Cl
280
300
320
340
360
380
400
420
440
460
480
500
20
40
60
80
100
120
140
160
180
200
220
240
260
Cretaceous
Paleozoic
B1
B3
B3
M1
M1
Si Sa Gr
Md Pc
Wc Gr Bd
Cl Si Sa Gr
Md Pc
Wc Gr Bd
Cl
Fig. 6. See locations in Fig. 2. Stratigraphic pro?les in the footwall and hanging wall blocks of the San Juanito fault depicting marked
contrasts between early syn-rift facies in both domains.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
120
A. Moraet al.
onment. Abundantwackestones and some bioclastic pack-
stones with horizontal, well- sorted, and often disarticu-
lated bivalves may correspond to proximal storm- ?ow
shell concentrations (sensu F?rsich, 1995). In contrast, the
less frequent wacke- and packstone levels with rarely bro-
ken, often articulated bivalve shells in di?erent positions,
and poor selection,may indicate storm-wave action (storm
wave shell concentration sensu F?rsich, 1995). Overall, we
11. Palomas 12.. Gachaluno River
W E
220
240
260
280
300
320
340
360
380
400
420
440
M3
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
20
40
60
80
100
120
140
160
Interval not to scale.
Poorly exposed ca.500m
shaly interval
Paleozoic
Cretaceous
180
200
M2
B1
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
20
40
60
80
100
120
140
160
180
200
220
240
260
Paleozoic
Cretaceous
B1
B3
Bt-1
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
280
300
320
340
360
380
400
420
440
460
480
500
520
Bt-1
Bt-1
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
Gachaluno Fault
0
Fig.7. See locations in Fig. 2. Stratigraphic logs east andwest of the Gachaluno fault showing di?erent early syn-rift facies in each
domain controlled by the Gachaluno fault.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
121
Role of inherited extensional fault segmentation and linkage
interpret the tabular wacke- and mudstone layers
to have been deposited above the basal level of storm
in?uence and below the basal level of fair-weather wave
in?uence.
Facies association C3
Tabular beds of quartzose wackestones, calcarenites,
carbonate-cemented sandstones, and minor calcareous
mudstones characterize this facies association.The C3 fa-
cies association occurs at locality 14 (between 145 and
190m, Fig. 8) and in lower non-conglomeratic segments
of locality 7a and 7b (Figs 2 and 6). Interestingly, at both lo-
calities there are occasional horizons where the bulk of the
rock is composed of calcareous mudstone but most of the
sand-size components consisting of quartz or terrigenous
fragments (petrography after Parra, 2000). At locality 7a,
Parra (2000) observed bioclasts of agglutinated benthic
foraminifera (Textularia sp.,Glomospira sp.), aminor amount
of planktonic species (Praeglobulina sp.), and ostracods. In
addition, gastropods occur in the lowermost calcareous
beds at locality 7b.
The coarser grain size of this facies association repre-
sents a higher-energy version of facies association C2.We
interpret theC3 association as internal carbonate platform
deposits. In addition, the presence of quartz and other ter-
rigenous components indicates an adjacent terrigenous
sediment source. Ostracods, agglutinated foraminifera,
and gastropods may indicate a restricted, probably high
salinity, internal platform environment.
San Isidro Fault
Paleozoic
Cretaceous
20
40
60
80
100
120
140
160
C1
C2
MO
MO
MO
MO
Y
20
40
60
80
100
120
140
160
180
200
Paleozoic
Cretaceous
C1
C3
B2
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
20
40
60
80
B2
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
20
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
Si Sa Gr
Md Pc
Wc Gr Bd
Cl
Fig. 8. See locations in Fig. 2. Early syn-rift facies controlled by the San Isidro fault, northern segment of the study area.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
122
A. Moraet al.
Bata? formation
Facies association Bt-1
Facies association Bt-1 is well represented at localities 12
and 19. It constitutes a continuously exposed horizon
between both localities. The lower section in locality 12
overlying Paleozoic rocks consists of a 30-m-thick con-
glomerate included in facies association B1. This unit is
overlain by ca. 70m of strata corresponding to facies asso-
ciation B3. This interval is superseded by a ca. 400-
m-thick sandstone unit that constitutes facies association
Bt-1 (Fig. 7). The dominant lithology consists of
amalgamated medium to thick beds of medium- to
coarse-grained quartz sandstones. These layers are often
interbeddedwith thin, ?ne-grained sandstones exhibiting
?aser and lenticular bedding. Occasionally, bioturbation
(burrows) characterizes these beds.The medium to thick
quartz-sandstone layers are commonly interbedded with
upward ?ning, very thick to massive lithic to sub-lithic
sandy to conglomeratic layers with erosional bases marked
by conglomeratic lags. Large-scale cross bedding with
conglomeratic laminae often accompanies layers exhibit-
ing erosional bases. These strata are well developed be-
tween 30 and190m at locality12 (Fig. 7).
At this location, the upper stratigraphic horizons of this fa-
cies association are characterizedbydecreasing sandstone,ab-
sence of upward?ning conglomeratic sandstone layers, and an
increasing proportion of interbedded mud- and siltstones.
Between290and360m, transitions betweenmuddyand sandy
facies are commonlybioturbated (burrows) and ?aser, lenticu-
lar, and wavy laminations are typical. However, these features
are less common in the upper portion of the facies association
above 360m,which is a dominantly muddy interval.
The lower monomineralic sandstones with interbedded
thin ?ne-grained sand to siltstone layerswith ?aser bedding
are interpreted as shoreline deposits (Galloway & Hobday,
1996; Reading & Collinson, 1996). However, the lithic sand-
stones and ?ne-grained conglomerates with sharp basal
contacts and ?ning-upward sequences indicate a periodic
in?uence of?uvial processes in this near-shore environment
(Dalrymple & Choi, 2007). In this context, the large-scale
cross bedding with conglomeratic laminae may correspond
to ?uvial lateral accretion surfaces.We suggest that this ?u-
vial in?uence disappears progressively upsection as lithic
sandstones and conglomerates with ?ning upward se-
quences disappear. In the upper part of this facies associa-
tion, an increasing proportion of shales and siltstones, a
decreasing sandstone granulometry, and the presence of
bioturbated intervals suggest amiddle to lower shoreface en-
vironment (Galloway &Hobday,1996).
Macanal formation
Facies associationM1
This association is characterized by laminated beds of vari-
able thickness with sandy mudstones, siltstones, and ?ne-
grained sandstones that are exposed over a ca. 300m interval
in the upper section of the pro?le at locality 6 (Fig. 6) and in
(a) (b)
(c) (d)
Fig.9. Photographs of facies association C1at locality15. (a) Enterolithic structures interlayered in a laminated sequence of dolomites
and mudstones. (b) Detail of entherolitic structure above interlaminations of dolomites and mudstones. Notice below the laminated
beds the presence of calcite nodular growths (anhydrite pseudomorphs?). (c) Interlaminations of dolomites (light layers) and mudstones
(dark layers) of interpreted microbial origin. (d) From left to right. Small enterolithic growths; ?rst laminated dolomites; layer with
abundant nodular growth; second laminated dolomites.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
123
Role of inherited extensional fault segmentation and linkage
the basal segment at locality13 (Fig.10) where it attains 60m
thickness. Sandstones are mostly quartzose. Internally, the
beds exhibit ?aser, lenticular, wavy, and planar lamination.
The persistent ?aser, lenticular and wavy laminations
indicate periodic tidal ?ow in a siliciclastic tidal ?at (Read-
ing & Collinson, 1996).These strata are commonly located
in the upper part of the described facies associations (e.g.
Locality 6, Figs 2 and 6) between transitional to marginal
marine environments represented by facies associations
B2 and Bt-1, and open-sea shales of facies association
M3.Therefore, they re?ect an intermediate part in an up-
ward deepening succession.
Facies associationM2
Typically, this facies association comprises a cyclic repeti-
tion of medium to thin mudstone beds with ?ne-grained
to very ?ne-grained sandstones displaying normal grading
and erosive to irregular basal contacts. It includes an inter-
val of about 150m at locality 11 (Fig. 7).These sandstones
commonly have tangential cross bedding and tool marks
at the base of individual beds.The top of each sandstone
cycle is transitional with thin to medium beds of sandy
black shales and shales with abundant plant remnants.
The beds with basal erosive contacts, often with tool
marks, normal grading, and upper transitional contacts
suggest deposition by density currents (e.g. Shanmugam,
1997). A nearby continental source for the turbidites can
be inferred from the abundant plant remnants in the inter-
bedded shales. Paleocurrent directions measured on the
sole marks (Fig. 7) and a westward pinchout of the sand-
stones, indicate general ?ow from continental sources
in the east.Turbidites in section 11probably represent the
lateral equivalents of ?uvial conglomerates and sandstones
in locality 12 to the east.
Facies associationM3
This facies consists of monotonous, well-laminated, am-
monite-bearing shale beds and rare medium-grained
sandstones. Occasionally, the shales contain pyritic lami-
nae or nodules. In some localities there are abundant plant
remains interbedded with the shales (e.g. Locality 11).
Facies association M3 occurs on top of all other facies as-
sociations or represents their lateral equivalent. Because it
constitutes a major part of the up to 3-km-thick Berria-
sian-Valanginian Macanal Formation, its most important
feature compared with all other facies associations is its
thickness of many hundreds of meters. It is also the only
unitwith a large lateral extent, covering virtually the entire
study area.
Based on the predominant ?ne-grained lithologies and
ubiquitous ammonites, this facies association is inter-
preted to re?ect an open-sea platform environment. The
absence of indicators ofwave in?uence or storm beds sug-
gest deposition in the deepest portions of the platform.
This is compatible with the pyritic laminae, which is an
indicator of anoxic conditions. The presence of plant re-
mains suggests continental proximity.
EARLY CRETACEOUS EXTENSIONAL
TECTONICS AND FACIES
DISTRIBUTIONS
Our facies analysis demonstrates that the Berriasian
depositional settings were spatially highly variable and
disparate over short distances, emphasizing the important
role of tectonism.Thus, to further understand Berriasian
paleogeography, it is necessary to view our analysis in a
structural context to evaluate fault controls on local and re-
gional scales. In the following sections, we ?rst describe
the main structures in the study area, group them into re-
lated segments, and then compare them with the spatial
distribution of facies.
MO
MO20
40
60
80
100
Paleozoic
Cretaceous
Poorly exposed
shaly interval
M3
M1
Cl Si Sa Gr
Md Pc
Wc Gr Bd
Fig.10. See location in Fig. 2.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
124
A. Moraet al.
The structural pattern shows more segmented and per-
vasive faulting in the northern part of the study area. In
contrast, in the south faulting is reduced to less numerous
and longer structures.Therefore, we de?ned two di?erent
segments that constitute the Guatiqu|? a paleo-rift. A
northern segment comprising the short and rectilinear
Gachaluno, Esmeralda, Garabato, Malacara, Miralindo,
San Isidro, and Manizales faults (Fig. 2), and a southern
segment with the longer Servita? , San Juanito, Naranjal
and LosMedios faults (Fig. 2).
Based on the dominant facies association for the ?rst
100m above the basal unconformity at each location
(excluding the ubiquitous lowermost basal conglomerates,
if they are thinner than 20m) and mapping the lateral ex-
tension of the facies associations, we compiled a facies
map (Figs 2 and 11), in which we depict the earliest stages
of Neocomian sedimentation. We also exclude the very
local debris- ?ow deposits in locality 15. We extrapolated
the units of interest on the geological map to areas covered
by younger strata to the extent admissible from the
mapped contacts. In the facies map it is evident that
the lateral extent of many depositional systems appears
to be bounded by the traces of the main faults (Fig. 11) as
discussed below.
Structural controls on early Cretaceous
sedimentation along the different graben
segments
In the southern segment shallow marine facies east of the
San Juanito fault (facies association C3; localities 5 and 7,
Figs 2, 6 and 11) contrast with an up to 300-m-thick basal
Cretaceous alluvial sequence west of the same structure
(facies association B3; locality 6, Figs 6 and 11). Thus,
di?erential subsidence along this fault presumably con-
trolled these di?erent depositional systems. In addition,
the mean grain size of alluvial sequences west of the San
Juanito fault decreases southward until they are replaced
by shallow marine deposits almost at the southern tip of
the San Juanito fault. Therefore, a paleo-high related to
Fig. 8 
34
2
1 5
7
6 8
10
12
11
13
19
18
14
15
16 17
9
Tertiary
Cretaceous
Upper Paleozoic
Pre-Devonian metamorphic 
basement
Fluvial sandy facies
Fan Delta and gravel facies
Carbonate inner platform 
facies
Sabhka and lagoon facies
Siliciclastic shallow marine 
facies
Lower Cretaceous facies 
associations
4?30?
74?
5?
75? 0 10 20 km
Fig. 5
Fig. 6
Fig. 7
Fig.11. Map of early syn-rift Cretaceous facies distribution.Notice that the reconstructed distribution of facies does not coincide with
the geological map in Fig.1 as the lateral extent of early syn-rift units has been extrapolated to covered areas. However, post-rift
shortening has not been removed, but the normal nature of the faults during the early Cretaceous is shown in the ?gure.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
125
Role of inherited extensional fault segmentation and linkage
the San Juanito fault is inferred.We term this paleo-high
the Chingaza Dome (Fig. 12). The direction of spatial
changes in facies and grain size suggests southward-direc-
ted axial drainage west of the San Juanito fault. More sig-
ni?cantly, in the southern segment shallow marine facies
are restricted to a central area bounded by the Servita? fault
to the east and the San Juanito fault to the west (Fig.11). In
this area and all other sectors with marine deposition in
the southern segment, laterally restricted fan-delta de-
posits pinch out (Facies association B2) in a direction per-
pendicular to all faults in this segment.This pattern can be
interpreted as a transverse drainage from the fault-
bounded paleo-topographic highs toward the adjacent
subsiding areas with marine deposition.
In the northern segment, the Garabato and Malacara
faults are associated with an up to 200-m-thick sequence
of supratidal evaporites, intertidal stromatolites, and shal-
low water carbonates (facies associations C1, C2 and C3),
which are restricted to the west of these structures (local-
ities14,15,17 and18, Figs 8,11and13). In contrast, a 20-m-
thick Lower Cretaceous siliciclastic tidal- ?at unit above
Upper Paleozoic rocks is rapidly superseded up section
by distal-platform black shales with ammonites in the
eastern block of the Garabato fault (facies associationsM1
and M3; Locality 13, Figs 10, 11 and 13). Again, fault con-
trolled di?erential subsidence can be inferred from these
patterns. Shallow-water to supratidal units to the west
pinch out near the tips of the Garabato and Miralindo
faults, suggesting that these depositional systems also
occupied a fault- controlled paleo-high that we term Rio
Guavio Dome (Fig. 13). This assessment is corroborated
by emerald-bearing mineralizations which are restricted
to facies association C2 in the Rio Guavio Dome.
These mineralizations have not been reported east of the
Garabato and Malacara faults. Other structures like the
Manizales and San Isidro faults can be also reinterpreted
?73.7?
4.5?
4.4?
Sa
n J
ua
nit
o
Fa
u
lt
Ch
ing
az
a D
om
e
Cross section in Figure 12B
Metamorphic basement
Devonian (?) shales
Facies association B3
Facies association M1
Facies association C2
Facies association B2Locality 6
Locality 7
5 km
2000
0
?
4000
Locality 6
Locality 7
San
 Juanito Fault
(a)
(b)
Fig.12. Detailed geological map and cross section of the ChingazaDome.Notice that the San Juanito Fault separates a footwall domain
to the west where braid and meandering river depositional systems dominate from a domain to the east with fan deltas adjacent to early
syn-rit shallow marine systems. See location in Fig.1.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
126
A. Moraet al.
Sa
n 
Is
id
ro
 F
a
u
lt
R
io
 G
ac
ha
lu
no
 F
a
u
lt
2000m
0
El
 G
ar
a
ba
to
 F
a
u
lt
Locality 14 Locality 15
Locality 11
Locality 11
Map
NW SE
El
 G
ara
ba
to 
Fa
ult
Ma
lac
ar
a 
Fa
ult
Ma
niz
ale
s F
au
lt
Sa
n 
Isi
dr
o 
Fa
ul
t
Shales (Facies Association M3))
Sandy Limestones Facies (Facies association C3)
Upper level of estromatolites un evaporites pseudomorphs
(Facies Association C1)
Upper level of Mudstones and Wackestones
(Facies Association C2)
Marls and evaporite pseudomorphs
(Facies Association C1)
Upper Paleozoic Units
Locality 15
Locality 14
Locality 16
Locality 13
Locality 17
Locality 18
4? 45`
73?30`
Lower level of estromatolites and evaporite pseudomorphs
(Facies association C1)
Lower Level of Mudstones and Wackestones
(Facies Association C2)
Conglomerates and limestones interbedded
Basal conglomerate (B1)
Cross section in Figure 13B
(a)
(b)
0 5 km
Fig.13. Detailed geological map and cross section of theGuavioDome.Notice the map relationships in the area highlighted in the box
south of Locality14; the activity of the normal fault near locality15 is contemporaneous or even post-dates the deposition of the level of
conglomerates and sandstones but pre-dates the deposition of the lower level of stromatolites and evaporite pseudomorphs (Facies
association C1). See location in Fig.1.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
127
Role of inherited extensional fault segmentation and linkage
as Neocomian faults, because they separate di?erent early
syn-rift facies in their hanging and footwalls (Figs 8 and
13). Similarly, the Gachaluno and Esmeralda faults restrict
the extent of the sandy-shorezone facies (facies association
Bt-1, Fig. 6) of the Bata Formation at localities 11 and 19
(Fig. 11), contrasting with the open-sea shaly facies to the
west. In general, faults in the northern segments separate
marine facies of di?erent bathymetric a?nity. Transverse
alluvial to fan-delta depositional systems are only locally
present west of the Gachaluno fault.
Taken together, the character and extent of the analyzed
facies sequences and structures suggest that virtually all
faults in both segments are Neocomian in age. Deposition
is therefore interpreted to have been determined by
fault activity during the initial phase of rifting. In this ex-
tensional scenario, the San Juanito fault in the south was
kinematically linkedwith theGachaluno fault in the north
by an accommodation zone, here called Claraval linea-
ment (Fig. 14). However, the Servita? fault and associated
faults to the east are di?erent from all other structures.
The rectilinear trace of the Servita? fault contrasts
with the discontinuous trace of faults farther west. The
Servita? fault system along the eastern boundary of the
study area also constitutes the depositional boundary of
Neocomian strata to the east. We therefore di?erentiate
the Servita? fault as a basin-boundary fault contrasting
with the intrabasinal faults to the west. In the northern
graben segment an equivalent principal Neocomian
boundary fault is not discernible from ?eld data, although
it should be located along strike of the Servita? fault, as can
be inferred from seismic and gravimetry data (e.g. Mora
et al., 2006).
West of the Servita? fault, irrespective of the structural
position, and therefore along all of the intrabasinal struc-
tures of the northern and southern segments, there is
evidence for a ubiquitous rapid vertical transition from
shallow marine and ?uvial facies to more distal marine to
deep marine shaly facies. In addition, Berriasian toValan-
ginian black shales often attain more than 2 km thickness
(e.g. locality 17) whereas underlying alluvial and shallow
marine sequences are rarely thicker than 500m. Total
accommodation space may apparently have increased after
the deposition of shallow-marine and alluvial sequences as
intrabasinal footwall blocks were covered by marine sedi-
ments, either due to decreased activity or locking of faults.
In contrast, pre-Aptian rocks are absent east of the
Servita? fault and associated basin boundary structures
(e.g. Figs 16 and 17). Therefore, these faults continued to
be active until theAptian.Figure13 (see normal fault south
of locality 15), 17 and18 show structural relationships doc-
umenting some cases where the intrabasinal faults to the
west of the basin bounding faults were active only during
short periods, as mapped Berriasian horizons clearly
post-date the faults.
Footwall uplift patterns
Structural mapping reveals that in the southern segment
of the Guatiqu|? a paleo-rift paleozoic rocks are absent di-
rectly to the east of the Servita' fault (Figs1and11, Locality
3). However, at locality 10 west of the Servita' fault, a Ber-
riasian debris- ?ow deposit pinches out to the west and
evolves into sandy turbiditic facies (Fig. 17), indicating a
higher source area to the east. The provenance of these
debris- ?ow deposits identi?es a source region in Paleo-
zoic sedimentary rocks (Table 1 see Locality 10). Phyllite
clasts do not occur, although the faulted contact of the
debris- ?ow deposits to the east is against phyllites of the
QuetameGroup (Fig.17).The structural and stratigraphic
relationships at locality 10 (Fig. 17) thus show that Paleo-
zoic rocks were present east of the Servita? fault but eroded
before, and during the onset of Cretaceous deposition.
Erosion of Paleozoic rocks before, and during the onset of
Cretaceous deposition is also documented in the western
block of the San Juanito fault by map patterns (Fig. 12;
Mora et al., 2006) and sandstone petrography from Parra
(2000). Interestingly, except localities 1 and 10, all sites in
the southern segment of the graben contain numerous
clasts of phyllites coming from the Quetame Group (see
Fan Delta and gravel facies
Carbonate inner platform facies
Fluvial sandy facies
Sabkha and lagoon facies
Siliciclastic platform facies Exposed areas
Transitional to marginal marine facies
Fig.14. Idealized con?guration of the Lower Cretaceous graben setting (Guatiqu|? a graben) located in the Farallones deMedina region.
Not to scale.Notice alluvial units and erosion in the footwalls to the south (southern graben segment) and fan deltas in the hanging wall
areas.The supratidal and intertidal units in the footwalls of the northern segment contrast with siliciclastic platform facies in the
hanging walls.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
128
A. Moraet al.
Table 1 and Fig. 11).This observation, and the presence of
debris- ?ow units adjacent to the fault-bounded paleo-to-
pographic highs, document that there was footwall uplift
along the faults bounding the southern graben.This uplift
must have been pronounced, creating high terrain from
which basement clasts were eroded and transported to the
hanging-wall basins. These processes were ultimately
responsible for the removal of ca. 3 km of upper Paleozoic
rocks before the onset of Cretaceous deposition on these
structural highs (see generalized con?guration in lower
part of Fig.14).
In contrast, in the northern segment and north of local-
ity 10 there are no surface exposures of the Quetame
Group phyllites and quartzites immediately below the
Cretaceous section (Fig. 11). Underlying units belong to
upper Paleozoic sedimentary strata. Interestingly, meta-
morphic clasts are absent in the basal conglomerates in
sections 10^19 in the northern segment (Table 1). In addi-
tion, local debris- ?ow sequences were only observed in
the hanging-walls of the San Isidro and Gachaluno faults
(Figs 7 and 8).Therefore, the provenance and areal extent
of sediments document that footwall upliftwas either neg-
ligible or absent in the northern segment (compare lower
and upper part of Fig.14).
Footwall-uplift patterns are observed along faults dip-
ping both west and east (Figs 2, 14 and 16) in both seg-
ments. Therefore, we hypothesize that the eastern faults
with the footwall to the east, were active at the same time
and probably interacted mechanically with the corre-
sponding western faults and their footwalls. However, it is
not clear whether the southern and northern graben seg-
ments were active coevally. Since the work of Jackson &
McKenzie (1983), footwall uplift in many extensional
provinces has been documented to be proportional to
adjacent hanging-wall subsidence. If that is also true for
our study area, hanging-wall subsidence in the southern
segment should have been higher than in the north.
Accordingly, we do not interpret the early deposition of
marine sequences in the footwall blocks of the northern
segment to re?ect an increase in total tectonic subsidence
with respect to the southern segment, but rather reduced
local footwall uplift.
STYLES OF REACTIVATION
Superimposed Cenozoic contraction along the exten-
sional structures generated many signi?cant structural
features in the Eastern Cordillera. At all scales, former ex-
tensional basement highs appear to have con?ned contrac-
tional folding structures (Fig. 15). A clear example is the
Rio Los Medios syncline, where the former extensional
hanging wall of the Los Medios normal fault has been
folded in contraction against the former extensional foot-
wall uplift (Fig. 18). This behavior, de?ned as basement
buttressing, has been identi?ed as a typical form of reacti-
vation by Mora et al. (2006) in the Eastern Cordillera.The
most representative large-scale examples of basement
buttressing are the Farallones and Santa Maria anticlines
(Figs 2 and 15). All of the documented Neocomian faults
con?ne the areal extent of both structures (Figs 2, 13 and
15). The Farallones anticline coincides with the hanging-
wall sectors in front of the normal faults at the southern
segment of theGuatiqu|? a graben,whereas theSantaMar|? a
anticline coincides with downthrown areas in the north-
ern segment, clearly identifying them as inversion anticli-
nes. In contraction, the Claraval lineament also forms a
transfer zone as the Farallones and SantaMar|? a anticlines
are distributed in en-echelon manner along this structure.
Therefore, there is a fundamental basement-related pre-
disposition of faulting and folding in this part of the
northern Andes.This has resulted in di?erences in inver-
sion-generated structural relief and reactivation along
pre-existent fault planes, which we will discuss for the
southern and northern segments.We illustrate these dif-
ferences by comparing paleo-topography deduced from
the facies map (Fig. 11) with present-day topography
(Fig. 2), and by employing serial balanced cross-sections
(Fig.15).
Reactivation in the southern graben segment
In the southern graben segment, the basal Cretaceous un-
conformity in the hinge zone of the Farallones anticline
lies at ca. 4000m above sea level, the maximum elevation
of this horizon in the study area (Figs 2 and 15).The loca-
tion of the Farallones anticline coincides with the zone of
Metamorphic basement
Upper Paleozoic
Lower Cretaceous
Facies association B2
Lower Creataceous
Facies association M2
Lower Cretaceous
shale units
Upper Cretaceous units
Neogene
2 Km
0
1000
2000m Servit? Fault
Fig.15. Detailed cross section along the Ardita Creek. Notice the westward granulometric change in the Early Cretaceous facies to the
west and the presence of a blockwithout Paleozoic cover below the Early Cretaceous units. See ?Discussion? in the text.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
129
Role of inherited extensional fault segmentation and linkage
0?4
4
Upper Paleozoic Units
0
4
0
(c)
(b)
Neogene Molasse
Oligocene-Mid-Miocene 
Paleocene-Eocene
Upper Cretaceous
Lower Cretaceous
Tesalia Fault
NW SE
NW SE
0
? 4
4
NW SE
0
4
?4
(a)
San Juanito Fault Servit? Fault
Lengup? Fault
Tesalia Fault Guaicaramo Thrust
Fig.16. Balanced cross sections along the di?erent segments of the study area. (a) Cross section along the northern segment. Note that
the present day active foreland deformation front is also included.Below a retrodeformedEocene state is included. (b)Cross-section along
the central segment of the study area. (c) Cross section along the southern segment of the study area, in the lower part aMaastrichtian-
Paleocene retrodeformed state is included.Notice how fault propagation folding is minor along theServita? fault comparedwith prominent
fault propagation folding in the central and northern cross-sections associatedwith theTesalia and Lengupa? thrusts. In contrast,
shortening in the central and northern cross-sections is absorbed mostly by the Guaica? ramoThrust. See locations in Fig.1.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
130
A. Moraet al.
maximum extensional subsidence during the Cretaceous.
Despite this, many paleo-topographic highs in extension
have retained their character as topographic highs during
contraction. For example, the alluvial facies in the footwall
of the San Juanito fault (locality 6) or the southernmost
exposures of syn-rift rocks in the Farallones anticline
(locality 2) coincide with structural and topographic
culminations at elevations near 4000m above sea level
(compare Figs 2 and11).We therefore posit that maximum
inversion-generated structural relief appears to coincide
with zones of maximum extensional subsidence, although
adjacent paleo-topographic highs have remained as topo-
graphic highs in contraction.
The contractional reactivation (displacement) in the
southern segment along the faults themselves appears to
be concentrated along the reactivated basin-boundary
faults and associated shortcuts (Figs 2 and 11). For exam-
ple, it is evident that reverse displacement along the Servi-
ta? fault during contraction is generally limited. However,
the Servita? fault was only reactivated in contraction at its
southern termination and remained in net extension near
its northern termination (Fig. 15). In addition, most of
the contractional fault displacement and shortening on
the eastern ?ank of the Eastern Cordillera is due to the
Mirador shortcut thrust (Fig. 1). This shortcut branches
from the Servita? fault, an interpretation supported by seis-
mic pro?les and the fact that in map view this structure is
always parallel to theServita? fault, except near its tipswhich
converge with the trace of the Servita? fault (Figs 1 and 15).
Therefore, these map patterns may represent an oblique
view of converging thrusts at depth, as has been observed
in many thrust systems (e.g., Boyer and Elliot, 1982). The
Servita? fault and Mirador shortcut account for ca. 20 km
of shortening in the Eastern Cordillera as deduced from
line-length balanced cross-sections (Fig. 15). In contrast,
intra-basinal faults with their footwalls to the west, such
as the San Juanito Fault, only accommodated minor reacti-
vation (Mora et al., 2006, Fig. 15). It therefore appears that
10 km
MoT
LF
SjF
Ch
A AA
LMF
FA
MAT
SF
MzT
BT
NF
SF
NF Villavicencio
MAT
AA: Apiay Anticline
BT: Boa Thrust
ChA: Chichimene Anticline
FA: Farallones Anticline
LMF: Los Medios Fault
MAT: Mirador-Agualinda Thrust
MoT: Monserrate Thrust
MzT: Manzanares Thrust
NF: Naranjal Thrust
SF: Servit? Fault
Reverse fault
Normal fault
Strike-slip fault
Anticline
Syncline
4?N
74? W
Fig.17. Geological map of the southernmost part of the study area. Symbols are the same as in Fig.1.The Lower Cretaceous Servita?
basin bounding fault can be identi?ed as the structure that bounds the area where Upper Paleozoic sequences are preserved above the
pre-Devonian metamorphic basement (QuetameGroup). In contrast, east of this structure Lower Cretaceous rocks rest on top of pre-
Devonian metamorphic basement.Therefore structures east of the Servita? fault can be interpreted as footwall shortcuts. Symbols are
the same as in Fig.1. See location in Fig.1.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
131
Role of inherited extensional fault segmentation and linkage
the intrabasinal Neocomian highs like San Juanito footwall
remained in high topographic positions, mainly because
they were passively transported by the Servita? fault.
Reactivation in the northern graben segment
As in the south, in the former northern graben segment,
maximum inversion-generated structural relief coincides
with areas of maximum Neocomian subsidence, such as
the hinge zone of the Santa Maria anticline (Fig. 2).
Accordingly, paleo-highs in the northern segment, (e.g.
the Rio Guavio paleo-high) have retained their character
as structural culminations in contraction with respect to
the elevation of the basal Cretaceous unconformity. How-
ever, with elevations of about 2200m, structural culmina-
tions in the north lie at substantially lower elevations than
in the south.
Maximum fault displacement in the northern segment
is concentrated along structures that lie along strike north
of the Servita? fault, including the Lengupa? and Tesalia
faults. Neither one of these structures have been unequi-
vocally documented as a Lower Cretaceous structures.
There is also no evidence that they constitute shortcuts
branching out from inherited structures, and the amount
of slip is minor comparedwith equivalent reactivated faults
along strike in the southern segment (e.g., Servita? fault.
Fig. 15). Adjacent normal faults like the Gachaluno fault
have not been reactivated in contraction (Fig. 2). In con-
trast, it is striking that a vertical to overturned panel
of Tertiary and Cretaceous rocks, associated with the
Lengupa? andTesalia faults, bounds the entire SantaMaria
anticline to the east. No comparable panel exists in the
southern segment (Figs 2 and 15). Therefore, most of the
basin inversion in the northern segment is due to fault-
propagation folding, rather than direct displacement along
reactivated faults or footwall shortcuts. Furthermore,
shortening accumulated by faulting and fault-propagation
folding in the forelimb of theSantaMaria anticline is close
to 20 km, similar to the ca. 20 km of shortening to the south
in that case almost only due to faulting (Fig.15).
Interestingly, if the displacement trends of the paleoto-
pographic highs like the Rio Guavio and Chingaza domes
Metamorphic basement
Devonian (Sandstones)
Devonian (Shales)
Carboniferous
Lower Cretaceous 
(Pre-Barremian)
Barremian to Albian
Upper Cretaceous
A
0 5 km
74?
500
2000
3000m
NW(b)
(a)
SE
Cross section in Figure 18B
4?20?
Lo
s 
M
ed
io
s 
Fa
ul
t
Fig.18. (a)Detailed geological map of theLosMedios fault.Notice the map expression of pronounced footwall uplift before the onset of
deposition of Lower Cretaceous strata. On the other hand Barremian units post-date the activity of the LosMedios fault.
(b) Reactivation along the plane of the LosMedios fault has been limited but tight folding west of LosMedios fault attests for the Los
Medios footwall acting as a stress riser buttressing the folds to the west.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
132
A. Moraet al.
are compared along strike, a displacement de?cit results
in the area between both, in extension as well as in
contraction. This is probably related with displacement
transfer towards basin bounding structures through
the Claraval accommodation zone, both in extension
and in contraction. Conversely, reactivated faults
bounding the Farallones and Santa Maria anticlines to
the east show whatWalsh et al. (2003) called a sympathetic
increase and decrease in displacement, without displace-
ment gaps.
DISCUSSION AND CONCLUSIONS
Careful mapping and analysis of Neocomian facies asso-
ciations coupled with structural investigations suggest
that faults in the Farallones anticline domain originated
as extensional faults during the early Cretaceous. Facies
and provenance analysis enable us to distiguish former
southern and northern rift segments that formed an inte-
gral part of the Guatiqu|? a paleo-rift. In the southern seg-
ment, during early syn-rift deposition (Facies associations
C and B), tectonic subsidence was higher compared with
the northern segment. Early syn-rift deposition in the
southern segment was associated with adjacent areas of
high paleo-topography in footwall areas (Fig. 14). This is
true for both intrabasinal and basin boundary faults in this
segment where shallow marine facies and fan deltas char-
acterize subsiding areas, while alluvial facies were depos-
ited in paleo-highs. In contrast, north of about 41300N a
more segmented northern rift sector existed during rift
initiation (Facies associations C, B, Bt). Here, as deduced
from footwall uplift, intrabasinal and basin-boundary
faults accrued less displacement than structures in the
southern segment. Shallow marine facies dominate in hang-
ing-wall areas, whereas sabkha facies characterize footwall
sectors. Intrabasinal faults in both the southern and north-
ern segments account for a highly variable early syn-rift
facies distribution due to isolated depocenters and paleo-
highs.This is exempli?ed by the Chingaza and Rio Guavio
domes, which restrict the areal extent of Cretaceous deposi-
tional systems. Segmented faults, isolated depocenters, and
paleo-highs during rift initiation show that intrabasinal fault
systems evolved from localized areas of extension and grew
along strike through segment linkage (e.g. Cartwright et al.,
1995;Dawers&Anders,1995). In the northern part of the rift,
individual faults are more numerous, shorter and rectilinear,
re?ecting less advanced fault linkage (e.g. Gawthorpe &Lee-
der, 2000). Accordingly, a more internally structured north-
ern rift, coupled with the spatial patterns of footwall uplift,
may have resulted from either rift propagation from south
to north or northward extensional displacement gradients
(Cowie et al., 2000) (Fig. 2). From the relationships between
faulting and sedimentation observed here, we propose that
east-dipping intra-basinal faults (e.g., San Juanito andGara-
bato faults) interacted mechanically with faults farther east
(e.g. Servita? fault) during rift initiation.
Subsequently, the rapid overall facies transition from
?rift initiation? (sensu Pr?sser, 1993. Facies associations B,
C and Bt) to the ?ne grained ?rift climax? sedimentation
(sensu Pr?sser,1993. Facies associationsM)marks the onset
of an episode where subsidence was rapid adjacent to the
Servita? fault. Footwall uplift along this structure was pro-
tracted, contrasting with intrabasinal faults whose footwall
areas were already submerged.Thus, footwall uplift of the
intrabasinal faults was subdued comparedwith basin sub-
sidence associated with the eastern boundary faults.This,
and the widespread complete Cretaceous sequence west of
the major boundary faults demonstrate that intrabasinal
faults remained inactive until their Cenozoic inversion. In
contrast, the onset of Cretaceous deposition east of the
Servita? fault only occurred during the Albian.Thus, total
syn-rift displacement along the basin-boundary faults
appears to have been greater than along the intrabasinal
faults. If true, the early linkage of the intrabasinal faults
was interrupted during the rift climax.Therefore, the fault
pattern that we observe in the intrabasinal faults repre-
sents the pattern of growth during the rift initiation stage.
In this context, it is worth noting that in the southern
segment the Servita? fault is rectilinear and displays no evi-
dence for earlier coalescing segments that would suggest
growth through linkage (Trudgill & Cartwright, 1994;
Cartwright et al., 1995; Dawers & Anders, 1995).Therefore,
we propose that the facies transition from?rift initiation? to
?rift climax? is not related to enhanced subsidence after
linkage (Gupta et al., 1999; Cowie et al., 2000), but rather
to an increased displacement along the principal basin-
boundary faults.The overall structural evolution suggests
that the intrabasinal Neocomian faults may represent
antithetic collapse structures associatedwith a major roll-
over anticline in the hanging-wall of the eastern boundary
faults, analogous to the central Kenya Rift (Roessner &
Strecker, 1997), the Viking Graben (Roberts & Yielding,
1991) or the collapse grabens in analog models of listric
faults (Mitra, 1993; McClay, 1995; Withjack et al., 1995; Ya-
mada &McClay, 2004; Seyferth &Henk, 2006).
Cenozoic shortening in the Eastern Cordillera of
Colombia is fundamentally in?uenced by the inherited
extensional structures. During contraction, the principal
bounding faults experienced a higher degree of reactiva-
tion, with throw reversing the extensional o?set (Fig. 15).
There is more e?cient reactivation along the Servita?
fault/Mirador shortcut pair in the southern segment com-
paredwith the northernLengupa? andTesalia faults. In the
northern segment, fold propagation dominates over direct
slip. Interestingly, footwall uplift was absent or negligible
during the Neocomian in the northern segment, whereas
in the southern segment footwall topography was pro-
nounced.Thus, if footwall uplift is a proxy for extensional
subsidence in the adjacent hanging wall (Jackson &
McKenzie, 1983), areas undergoing maximum extensional
displacement also tend to focus reverse displacement
along former border faults (Fig. 15). Accordingly, zones of
maximum Cenozoic inversion-generated uplift (i.e.
reverse fault displacement plus anticlinal fold amplitude)
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
133
Role of inherited extensional fault segmentation and linkage
appear to coincide with zones of maximum Neocomian
extensional displacement and subsidence, like the Faral-
lones anticline in the southern segment. These observa-
tions are similar to those of Underhill & Paterson (1998)
in the Wessex basin. In addition, we also ?nd similarities
with 3D analog inversion models (Yamada & McClay,
2004), suggesting inheritance of along-strike displace-
ment gradients in the extensional province.
In contrast to the major basin bounding faults, reactiva-
tion of the much more segmented intrabasinal faults was
less pronounced. As a consequence, syn-extensional
paleo-topographic highs were passively uplifted and per-
sisted as topographically elevated areas in contraction,
with similar segmentation and displacement gaps. It is
likely that focused faulting along rectilinear basin-bound-
ing structures in both extension and in contraction identi-
?es these structures as long-lived, predisposed zones of
weakness in the crust, similar to observations in theGam-
toos basin (Paton & Underhill, 2004) of South Africa. In
such a scenario, fault length in extension is de?ned early
during rifting (Walsh et al., 2002). If fault segments exist,
they de?ne a kinematically coherent system that merges
into a single structure at depth (Walsh et al., 2003), con-
trasting with growth through linkage. Interestingly, in the
foreland east of the basin-boundary faults, upper Paleo-
zoic rocks were not encountered in exploratory wells, and
Andean phyllitic basement rocks are replaced by Guyana
shield basement (e.g. Duen as, 2002). In addition, the exis-
tence of the basin-bounding faults as pre-Mesozoic mul-
ti-phased structures is supported by many studies (e.g.
Forero-Suarez, 1990). This could provide a mechanism
for the initial rapidMesozoic trace lengthening and subse-
quent preferential fault reactivation. In this context, it is
clear why intrabasinal faults that grew through linkage are
subordinate anisotropies comparedwith the longer basin-
bounding structures.
These observations raise the question as to which factor
de?ned the observed growth mechanisms for the intrabas-
inal structures.Walsh et al. (2003) argue that an incidental
linkage of otherwise isolated faults is more likely where a
pre-existent coherent weakness is absent and faulting has
to randomly break through a mechanically heterogeneous
multilayer. Interestingly, in contractionwe document either
direct reactivation or shortcuts branching from the basin
margin faults while intrabasinal faults remained undis-
turbed.We interpret this pattern to indicate that, irrespec-
tive of long-lived basement fabrics, segmented extensional
structures that purely grew through linkage of individual
faults may be less prone to inversion than normal faults that
propagated upward from a coherent common plane.
In summary, our paleogeographic reconstruction of the
early syn-rift setting in the northern Andes provides a un-
ique comparison between extensional tectonics and subse-
quent orogenic contraction.As in other inversion provinces
(e.g. Coward et al., 1989; Graciansky et al., 1989; Bailey et al.,
2002;Kley etal., 2005; Carrera etal., 2006), the fundamental
imprint of pre-existent basement fabrics has generated di-
rect reactivation,withbuttresses and shortcuts as typical in-
version features. However, our observations show that
inheritance of lateral displacement gradients and the me-
chanism of extensional fault growth itself are additional,
important factors also in?uencing inversion styles in con-
tractional orogens. These phenomena may be more com-
mon than previously thought. Importantly, only those
extensional structures which coincide with basement fea-
tures that existed before the extensional phase profoundly
determine the width of the contractional orogen. In addi-
tion selective fault reactivation in contraction appears to be
related to the style of extensional fault growth, and not only
to the orientation of the compressional stresseswith respect
to the pre-existent normal faults as previously proposed in
this region (e.g.Mora et al., 2006).
Our observations are useful in the extrapolation of fea-
tures observed in analog models and mild to moderately
inverted intra-plate basins to the structures of inverted oro-
gens. In the latter case, shortening is often more intense and
the relationships between ancestral extensional segmenta-
tion and superimposed past or ongoing contraction are less
clear. However, geometries and patterns of inversion in oro-
gens and intra-plate areas are comparable. Our analysis
helps understand andpredict the behavior ofmany structur-
al and stratigraphic traps of extensional origin incorporated
in contractional settings. The documented features are
therefore important for fossil fuels or mineral deposits ex-
ploration, particularly in frontier intra-plate or orogenic
areas that normally rely on limited subsurface data sets.
ACKNOWLEDGEMENTS
The authors are indebted to Arndt Peterhansel, Estelle
Mortimer, Gabriela Marcano, Cornelius Uba and Jessica
Zamagni for interesting discussions that improved the
ideas in this paper. Cornelius Uba and Jessica Zamagni
graciously read an earlier version of this paper.
Careful reviews by Nestor Cardozo, William Cavazza
andBrianHorton improved the ?nal version of this manu-
script. Funding for ?eld work came from Leibniz funds to
M. Strecker (DFG STR373/16-1) and the Geological Sur-
vey of Colombia (Ingeominas, project Cinturones Esmer-
ald|? feros de la Cordillera Oriental). A. Mora and M. Parra
are grateful to the DAAD for funding their stay at the
Geosciences Departament at the University of Potsdam.
Birgit Fabian made Figs 2 and11. Dr Fernando Etayo-Ser-
na (Ingeominas) determined the relative ages of the am-
monites available in our pro?le locations. Oscar
Fernandez from Midland Valley provided a 2dMove li-
cense for balancing our cross sections. Sara Moro? n and
Jaime Corredor assisted our work in the ?eld.
REFERENCES
Alsharhan, A.S. & Kendall, C.G.St.C. (2003) Holocene
coastal carbonates and evaporites of the southernArabianGulf
and their ancient analogues. Earth Science Rev., 61, 191^243.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
134
A. Moraet al.
Badley, M.E. (2001) Late tertiary faulting, footwall uplift and
topography inWestern Ireland. In:The Petroleum Exploration of
Ireland?sO?shoreBasins (Ed. by P.M.Shannon, P.D.W.Haughton
&D.V. Corcoran,Geol. Soc. Spec. Publ., 188, 201^207.
Bailey, C.M., Giorgis, S. & Coiner, L. (2002) Tectonic inver-
sion and basement butressing: an example from the central
Appalachian Blue Ridge Province. J. Struct. Geol., 24, 277^304.
Beauchamp,W.,Allmendinger,R.W.,Barazangi,M.,Dem-
nati, A., El Alji, M. & Dahmani, M. (1999) Inversion tec-
tonics and the evolution of the High Atlas Mountains,
Morocco, based on a geological-geophysical transect.Tectonics,
18, 163^184.
Blair, C.T. &McPherson, J.G. (1998) Recent debris- ?ow pro-
cesses and resultant form and facies of the dolomite alluvial
fan, OwensValley, California. J. Sediment. Res., 68, 800^818.
Boyer,S.E.&Elliot,D. (1982)Thrust systems.AAPGBulletin,
66, 1196^1230.
B?rgl,H. (1961)El Jura? sico e Infracreta? ceo delR|? oBata? , Boyaca? .
Bol. Geol. Inst. Geol. Nac., 6, 169^211.
Carrera, N., Mun oz, J.A., Sa' bat, F., Mon, R. & Roca, E.
(2006) The role of inversion tectonics in the structure of the
Cordillera Oriental (NWArgentinean Andes). J. Struct. Geol.,
18, 1921^1932.
Cartwright, J.A. (1989) The kinematics of inversion in the
Danish Central Graben. In: Inversion Tectonics (Ed. by M.A.
Cooper &G.D.Williams),Geol. Soc. Spec. Publ., 44, 153^175.
Cartwright, J.A.,Trudgill, B.D. & Mansfield, C.S. (1995)
Fault growth by segment linkage: an explanation for scatter in
maximum displacement and trace length data from the Ca-
nyonlands grabens of SE Utah. J. Struct. Geol., 17, 1319^1326.
Chapman,T.J. (1989)ThePermian toCretaceous structural evo-
lution of the Western Approaches Basin (Melville sub-basin),
UK. In: Inversion Tectonics (Ed. by M.A. Cooper & G.D.
Williams),Geol. Soc. Spec. Publ., 44, 177^200.
Colletta,B.,Hebrard,F.,Letouzey, J.,Werner,P.&Rudki-
weicz, J.L. (1990) Tectonic style and crustal structure of the
Eastern Cordillera, Colombia from a balanced cross section.
In: Petroleum and Tectonics in Mobile Belts (Ed. by J. Letouzey),
pp. 81^100. EditionsTechnip, Paris.
Collinson, J.D. (1996) Alluvial sediments. In: Sedimentary
Environments: Processes, Facies and Stratigraphy (Ed. by H.G.
Reading), pp. 37^82. Blackwell Science, Oxford.
Coward, M.P., Enfield, M.A. & Fischer, M.W. (1989) Devo-
nian basins of Northern Scotland: extension and inversion re-
lated to Late Caledonian-Varisican tectonics. In: Inversion
Tectonics (Ed. by M.A. Cooper & G.D.Williams, Geol. Soc. Spec.
Publ., 44, 275^308.
Cowie, P.A.,Gupta, S.&Dawers,N.H. (2000) Implications of
fault array evolution for synrift depocentre development:
insights from a numerical fault growth model. Basin Res., 12,
241^261.
Dalrymple, R.W. & Choi, K. (2007) Morphologic and facies
trends through the ?uvial-marine transition in tide-domi-
nated depositional systems: a schematic framework for envir-
onmental and sequence-stratigraphic interpretation. Earth
Sci. Rev., 81, 135^174.
Davis, D., Suppe, J. & Dahlen, F.A. (1983) Mechanics of fold-
and-thrust belts and accretionary wedges. J. Geophys. Res., 88,
1153^1172.
Dawers, N.H. & Anders, M.H. (1995) Displacement-length
scaling and fault linkage. J. Struct. Geol., 17, 607^614.
Dorado, J. (1992) Contribucio? n al conocimiento de la Estrati-
graf|? a de la formacio? n Brechas de Buenavista (l|? mite Jura? sico-
Creta? cico), regio? n noroeste de Villavicencio (Meta). Geol.
Colombiana, 17, 7^40.
Duen as,H. (2002) Paleozoic palynological assemblages from the
Colombian Llanos Basin. 34th. Annual Meeting of the American
Association of Stratigraphic Palynologists, 2001, San Antonio,
Texas. Palynology, 26, p. 264.
Eisenstadt,G.&Withjack,M.O. (1995) Estimating inversion:
results from clay models. In:Basin Inversion (Ed. by J.G.Bucha-
nan & P.G. Buchanan,Geol. Soc. Spec. Publ., 88, 119^136.
Etayo-Serna, F., Sole? -De Porta, N.S., De Porta, J. & Gao-
na,T. (2003) The Bata? formation of Colombia is truly Cretac-
eous, not Jurassic. J. South Am. Earth Sci., 16, 113^117.
Forero-Suarez,A. (1990) The basement of the Eastern Cordil-
lera, Colombia: an allochthonous terrane in northwestern
South America. J. South Am. Earth Sci., 3, 141^151.
F?rsich, F.T. (1995) Shell concentrations. Eclogae Geol. Helv., 88,
643^655.
Galloway, W.E. & Hobday, D.K. (1996) Terrigenous Clastic
Depositional Systems: Applications to Fossil Fuel and Groundwater
Resources, 2nd edn. Springer, Berlin.
Garc|? a Senz, J. (2002) Cuencas extensivas del Creta? cico Inferior
en los Pirineos centrales: formacio? n y subsecuente inversio? n.
PhDThesis, Universitat de Barcelona, Barcelona.
Gawthorpe,R.L.&Leeder,M.R. (2000)Tectono-sedimentary
evolution of active extensional basins.Basin Res., 12, 195^218.
Gemmer, L., Nielsen, S.B. & Bayer, U. (2003) Late
Cretaceous-Cenozoic evolution of the North German basin-
results from 3-D geodynamic modelling. Tectonophysics, 373,
39^54.
Geostratos (2005)Columna estratigra? ?caR|? o Chivor, Ingeomi-
nas. Bogota? , Colombia. Internal Report, 21p.
Geyer,O.F. (1973) Das prkretazischeMesozoikumvonKolum-
bien.Geol. Jahrbuch B, 5, 1^155.
Gillcrist,R.,Coward,M.&Mugnier, J.-L. (1987) Structural
inversion and its controls: examples from the Alpine foreland
and the French Alps.Geodinam. Acta, 1, 5^34.
Graciansky,P.C.D.,Dardeau,G.,Lemoine,M.&Tricart,P.
(1989) The inverted margin in the French Alps and foreland
basin inversion. In: Inversion Tectonics (Ed. by M.A. Cooper &
G.D.Williams,Geol. Soc. Spec. Publ., 44, 87^104.
Grier, M.E., Salfity, J.A. & Allmendinger, R.W. (1991)
Andean reactivation of theCretaceousSaltaRift, northwestern
Argentina. J. South Am. Earth Sci., 4, 351^372.
Guimera, J., Alonso, A. & Ramon Mas, J. (1995) Inversion of
an extensional-ramp basin by a newly formed thrust: the Ca-
meros Basin (N. Spain). In: Basin Inversion (Ed. by J.G. Bucha-
nan & P.G. Buchanan,Geol. Soc. Spec. Publ., 88, 433^453.
Gupta, S., Underbill, J.R., Sharp, I.R. & Gawthorpe, R.L.
(1999)Role of fault interactions in controlling synrift sediment
dispersal patterns: Miocene, Abu Alaqa Group, Suez Rift, Si-
nai, Egypt.Basin Res., 11, 167^189.
Hayward, A.B. & Graham, R.H. (1989) Some geometrical
characteristics of inversion. In: Inversion Tectonics (Ed. by
M.A. Cooper & G.D. Williams, Geol. Soc. Spec. Publ., 44,
17^39.
Hilley, G.E., Blisniuk, P.M. & Strecker, M.R. (2005) Me-
chanics and erosion of basement cored uplift provinces.
J. Geophys. Res., Solid Earth, 110, B12409, doi: 10. 1029/
2005JB0003704, 2005.
Huyghe, P. & Mugnier, J.-L. (1995) A comparison of inverted
basins of the southern North Sea and inverted structures of
the external Alps. In: Basin Inversion (Ed. by J.G. Buchanan &
P.G. Buchanan,Geol. Soc. Spec. Publ., 88, 339^353.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
135
Role of inherited extensional fault segmentation and linkage
Ingram,R.L. (1954) Terminology for the thickness of strati?ca-
tion and parting units in sedimentary rocks. Geol. Soc. Am.
Bull., 65, 937^938.
Jackson, J. & McKenzie, D. (1983) The geometrical evolution
of normal fault systems. J. Struct. Geol., 5, 471^482.
Kalkowsky, E. (1908) Oolith und stromatholit im nord-
deutschen Buntsandstein. Z.Deut. Geol. Gesellsch., 60, 68^125.
Kendall, A.C. & Harwood, G.M. (1996) Marine evaporites:
arid shorelines and basins. In: Sedimentary Environments: Pro-
cesses, Facies and Stratigraphy (Ed. by H.G. Reading), pp.
281^324. Blackwell Science, Oxford.
Kley, J.&Monaldi,C.R. (2002)Tectonic inversion in the Santa
Barbara System of the central Andean foreland thrust belt,
northwestern Argentina. Tectonics, 21, 1061, doi: 10.1029/
2002TC902003.
Kley, J., Rossello, E.A., Monaldi, C.R. & Habighorst, B.
(2005) Seismic and ?eld evidence for selective inversion of
Cretaceous normal faults, Salta rift, northwest Argentina.
Tectonophysics, 399, 155^172.
Kossow,D.,Krawczyk,C.,McCann,T., Strecker,M.&Ne-
gendank, J.F.W. (2000) Style and evolution of salt pillowsand
related structures in the northern part of theNorthestGerman
Basin. Int. J. Earth Sci., 89, 652^664.
Kossow, D. & Krawczyk, C.M. (2002) Structure and quanti? -
cation of processes controlling the evolution of the inverted
NE-German Basin.Mar. Petrol. Geol., 19, 601^618.
Logan,B.W.,Rezak,R.&Ginsburg,R.N. (1964)Classi?cation
and environmental signi?cance of algal stromatolites. J. Geol.,
72, 68^83.
Lowell, J.D. (1995)Mechanics of basin invesion fromwolrdwide
examples. In: Basin Inversion (Ed. by J.G. Buchanan & P.G.
Buchanan.Geol. Soc. Lond. Spec. Publ., 88, 339^353.
McClay,K.R. (1995)The geometries and kinematics of inverted
fault systems: a review of analogue model studies. In: Basin
Inversion (Ed. by J.G. Buchanan & P.G. Buchanan, Geol. Soc.
Spec. Publ., 88, 97^118.
Miall, A. 1985 Architectural- element analysis: a new method
for facies analysis applied to ?uvial deposits. Earth Sci. Rev.,
22, 261^238.
Miall, A.D. (1996) The Geology of Fluvial Deposits: Sedimentary
facies, Basin Analysis, and PetroleumGeology. Springer, NewYork.
Mitra,S. (1993)Geometry and kinematic evolution of inversion
structures.Am. Assoc. Petrol. Geol. Bull., 77, 1159^1191.
Monaldi, C.R., Salfity, J.A. & Kley, J. (2008) Preserved ex-
tensional structures in an inverted Cretaceous rift basin,
nothwestern Argentina: outcrop examples and implications
for fault reactivation. Tectonics, 27, TC1011, doi: 10.1029/
2006TC001993.
Mora, A., Parra, M., Strecker, M.R., Kammer, A., Dimate? ,
C. & Rodr|? guez, F. (2006) Cenozoic contractional reactiva-
tion ofMesozoic extensional structures in the Eastern Cordil-
lera of Colombia. Tectonics, 25, TC2010, doi: 10.1029/
2005TC001854.
Nemcok,M.,Nemcok, J. &Wojtaszek,M. (2001) Recosntruc-
tion of Cretaceous rifts incorporated in the Outer West Car-
pathian wedge by balancing.Mar. Petrol. Geol., 18, 39^63.
Nemec, W. & Postma, G. (1993) Quaternary alluvial fans
in southwestern Crete: sedimentation processes and
geomorphic evolution. In: Alluvial Sedimentation (Ed. by M.
Marzo & C. Puidgefabregas, Int. Assoc. Sedimentol. Spec. Publ.,
17, 235^276.
Parra, M. (2000) Estratigraf|? a y Petrograf|? a del Creta? cico Infer-
ior en el Parque Natural Chingaza y la Cuenca Alta del R|? o
Guatiqu|? a, Cundinamarca y Meta, Colombia. B.Sc. Thesis,
Universidad Nacional de Colombia, Bogota? .
Paton,D.A.&Underhill, J.R. (2004) Role of crustal anisotro-
py in modifying the structural and sedimentological evolution
of extensional basins:TheGamtoos Basin, South Africa.Basin
Res., 16, 339^359.
Pr?sser, S. (1993) Rift-related linked depositional systems and
their seismic expression. In:Tectonics and Seismic Sequence Stra-
tigraphy (Ed. byG.D.Williams &A.Dobb,Geol. Soc. Spec. Publ.,
71, 35^66.
Reading,H.G.&Collinson, J.D. (1996) Clastic coasts. In:Se-
dimentary Environments: Processes, Facies and Stratigraphy (Ed. by
H.G. Reading), pp.154^231. Blackwell Science, Oxford.
Riding,R. (2000)Microbial carbonates: the geological record of
calci?ed bacterial-algal mats and bio?lms. Sedimentology, 47,
179^214.
Rinco? n,A.&Ta? mara, J. (2005) La Falla deMirados y su signif-
icado para la sedimentacio? n del Titoniano-Neocomiano (Vil-
lavicenciio, Colombia). BSc Thesis, Universidad Nacional de
Colombia, Bogota? .
Roberts, A.M. & Yielding,G. (1991) Deformation around ba-
sin-margin faults in the North Sea/mid-Norway rift. In:The
Geometry of Normal Faults (Ed. by A.M. Roberts, G.Yielding &
B. Freeman,Geol. Soc. Spec. Publ., 56, 61^78.
Roberts, A.M.,Yielding,G. & Badley,M.E. (1993) Tectonics
and bathymetric controls on stratigraphic sequences within
evolving half-graben. In:Tectonics and Seismic Sequence Stratigra-
phy (Ed. byG.D.Williams & A. Dobb,Geol. Soc. Spec. Publ., 87^
121.
Roeder, D. & Chamberlain, R.L. (1995) Eastern Cordillera of
Colombia: Jurassic-Neogene crustal evolution. In: Petroleums
Basins of South America (Ed. by A.J.Tankard, S.R. Suarez & H.J.
Welsink,Am. Assoc. Petrol. Geol.Memoir, 62, 633^645.
Roessner, S. & Strecker,M.R. (1997) Late Cenozoic tectonics
and denudation in the Central Kenya Rift: quanti?cation of
long-term denudation rates.Tectonophysics, 278, 83^94.
Rust, B. & Koster, E. (1984) Coarse alluvial deposits. In: Facies
Models, Geoscience Canada Reprint Series, 1. (Ed. by R.Walker)
Geological Association of Canada,Toronto.
Sarmiento-Rojas, L.F., Van Wess, J.D. & Cloetingh, S.
(2006) Mesozoic transtensional basin history of the Eastern
Cordillera, Colombian Andes: inferences from tectonic mod-
els. J. South Am. Earth Sci., 21, 383^411.
Seyferth, M. & Henk, A. (2006) A numerical sandbox: high
resolution distinct element models of halbgraben formation.
Int. J. Earth Sci., 95, 189^203.
Shanmugam,G. (1997) The Bouma sequence and the turbidite
mind set. Earth-Sci. Rev., 42, 201^229.
Sinclair, I.K. (1995) Transpressional inversion due to episodic
rotation of extensional stresses in the Jeanne d?Arc Basin, o?-
shoreNewfoundland. In:Basin Inversion (Ed. by J.G. Buchanan
& P.G. Buchanan,Geol. Soc. Spec. Publ., 88, 249^271.
Stibane, F.R., ed. (1968) Zur Geologie von Kolumbien, S?damerika:
Das Quetame- und Garzon-Massiv. Geotektonische Forschun-
gen. E. Schweizerbart, Stuttgart.
Teixell,A.,Arboleya,M.-L., Julivert,M.&Charroud,M.
(2003) Tectonic shortening and topography in the central
High Atlas (Morocco). Tectonics, 22, 1051, doi: 10.1029/2002T
C001460.
Thomas, D.W. & Coward, M.P. (1995) Late Jurassic-Early Cre-
taceous inversion of the northern East Shetland Basin, north-
ernNorth Sea. In:Basin Inversion (Ed. by J.G. Buchanan &P.G.
Buchanan,Geol. Soc. Spec. Publ., 88, 275^306.
r 2009 TheAuthors
Journal Compilationr Blackwell Publishing Ltd, EuropeanAssociation of Geoscientists & Engineers and International Association of Sedimentologists
136
A. Moraet al.
Trudgill, B. & Cartwright, J. (1994) Relay-ramp forms and
normal-fault linkages, CanyonlandsNational Park,Utah.Geol.
Soc. Am. Bull., 106, 1143^1157.
Tucker, M.E. & Wright, V.P. (1990) Carbonate Sedimentology.
Blackwell Scienti?c Publications, Oxford.
Uliana,M.A., Arteaga,M.E., Legarreta, L., Cerda? n, J.J. &
Peroni, G.O. (1995) Inversion structures and hydrocar-
bon ocurrence in Argentina. In: Basin Inversion (Ed. by J.G.
Buchanan & P.G. Buchanan.Geol. Soc., Lond. 88, 211^233.
Ulloa, C. & Rodr|? guez, E. (1979) Geolog|? a del Cuadra? ngulo
K12, Guateque.Bol. Geol. Ingeom., 22, 3^55.
Underhill, J.R. & Paterson, S. (1998) Genesis of tectonic
inversion structures: seismic evidence for the development of
key structures along the Purbeck-Isle of wight disturbance. J.
Geol. Soc., 155, 975^992.
Walsh, J.J., Bailey,W.R., Childs, C., Nicol, A. & Bonson,
C.G. (2003)Formation of segmented normal faults: a 3-Dper-
spective. J. Struct. Geol., 25, 1251^1262.
Walsh, J.J.,Nicol,A.&Childs,C. (2002)An alternative model
for the growth of faults. J. Struct. Geol., 24, 1669^1675.
Warren, J.K. & Kendall, C.G.S.C. (1985) Comparison of
sequences formed in marine sabkha (subaerial) and salina
(subaqueous) settings: modern and ancient. Am. Assoc. Petrol.
Geol. Bull., 69, 1013^1023.
Williams, G.D., Powell, C.M. & Cooper, M.A. (1989) Geo-
metry and kinematics of inversion tectonics. In: Inversion
Tectonics (Ed. by M.A. Cooper & G.D.Williams, Geol. Soc. Spec.
Publ., 44, 3^15.
Withjack,M.O., Islam, Q.T. & La Pointe, P.R. (1995) Normal
faults and their hanging-wall deformation: an experimental
study.Am. Assoc. Petrol. Geol. Bull., 79, 1^18.
Yamada,Y. & McClay, K.R. (2004) Analog modeling of inver-
sion thrust structures: experiments of 3D inversion structures
above listric fault systems. In: Thrust Tectonics and Petroleum
Systems (Ed. by K.R. McClay), Am. Assoc. Petrol. Geol. Memoir,
82, 276^302.
Manuscript received 13 April 2007;Manuscript accepted 15 May
2008.
r 2009 The Authors
Journal Compilationr Blackwell Publishing Ltd, European Association ofGeoscientists & Engineers and International Association of Sedimentologists
137
Role of inherited extensional fault segmentation and linkage