www.elsevier.com/locate/geomorph

Geomorphology 69 (

Geomorphology and geochronology of sackung features

(uphill-facing scarps) in the Central Spanish Pyrenees

F. Gutierrez-Santolallaa,T, Enrique Acostab, Santiago Rıosb,

Jesus Guerreroa, Pedro Luchaa

aEdificio Geologicas, C/. Pedro Cerbuna, 12, Universidad de Zaragoza, Zaragoza 50009, SpainbInstituto Geologico y Minero de Espana, Oficina de Proyectos de Zaragoza, C/. Fernando El Catoloico, 59, 4 8C, Zaragoza 50006, Spain

Received 12 August 2004; received in revised form 27 January 2005; accepted 29 January 2005

Available online 17 March 2005

Abstract

The paper analyses uphill-facing scarps and associated troughs developed in the oversteepened slopes of two neighbouring

glacial valleys in the central Spanish Pyrenees. Previous studies of sackung landforms in the Pyrenees have argued for deglacial

unloading as the genetic mechanism, but this causal and temporal relationship has not been proved due to the lack of

chronological data. The antislope scarps in the two studied locations, Vallibierna and Estos, are developed in Palaeozoic

metasedimentary rocks (parallel to the contour lines and the structural grain), occur in the intermediate sector of the hillslope,

and are up to 0.5 km long and several meters high. A trench was excavated in a sackung trough fill in each of the valleys in

order to gain information about their chronology and genesis. Charcoal from the lowermost unit in Vallibierna provided an age

of 5.9 cal. ka for the sackung and extrapolation of the three dates obtained in Estos indicates that the trough formed ca. 7.6–7.8

cal. ka. Deglaciation of the studied sectors of the valleys occurred between 16 and 13 ka. The time lag of N5 ka suggests that

glacial erosion and the subsequent debutressing of the oversteepened valley walls created slopes predisposed to sackung

development, but did not initiate the movement. Seismic shaking is proposed as a probable triggering factor. This hypothesis,

although supported by the sudden deformation event recorded by a failure plane exposed in Vallibierna trench, and by the

seismic and neotectonic activity of the area, cannot be proved due to the lack of chronological information about

paleoearthquakes.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Sackung; Uphill-facing scarps; Gravitational spreading; Trenching; Seismic activity; Pyrenees

0169-555X/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.geomorph.2005.01.012

T Corresponding author. Tel.: +34 979 761090; fax: +34 976

761106.

E-mail address: fgutier@unizar.es (F. Gutierrez-Santolalla).

1. Introduction

The term sackung, German word for sagging, was

first introduced by Zischinsky (1966, 1969) to

designate the surface manifestations of deep-seated

rock creep in slopes on foliated bedrock. In the

2005) 298–314

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 299

subsequent literature, sackung (plural sackungen)

generically refers to linear geomorphic features

produced by gravitational spreading in slopes (Varnes

et al., 1989; Crosta, 1996; Ward, 2003). The most

characteristic sackung-type landforms are uphill-fac-

ing scarps (also called antislope scarps, counter

scarps, or antithetic scarps) associated with linear

depressions lying at the upslope side of the scarps

(Radbruch-Hall, 1978; Varnes et al., 1989). These

peculiar landforms commonly form complexes paral-

lel to the contours and ridge tops in the upper sector of

steep slopes. The uphill-facing scarps that split the

ridge tops into two crests give place to double or twin

ridges (dopplegrate) and ridge-top depressions. These

depressions show a graben-like appearance when

scarps with opposite orientations affect the ridge

crests. Some authors note arched uphill-facing scarps

with a downslope convexity (Jahn, 1964; Beck, 1968;

Varnes et al., 1989). The asymmetric troughs asso-

ciated with these scarps frequently host closed

depressions with ephemeral ponds where deposition

of fine-grained material takes place. The presence of

small shallow sinkholes has also been reported along

these depressions despite the absence of soluble rocks

(Bovis, 1982; McCalpin and Irvine, 1995). McCalpin

and Irvine (1995), from a compilation of published

sackung scarp dimensions carried out by McCleary et

al. (1978), give 15–300 m and 1–9 m as typical ranges

of scarp length and height, respectively. Other

frequent geomorphic features found in slopes affected

by sackungen are downhill-facing scarps with accom-

panying linear depressions and a bulge at the toe of

the slope (talzuschub) (Nemcok, 1972; Mahr, 1977;

Radbruch-Hall, 1978; Varnes et al., 1989). In some

cases, the upper sector of the slopes shows a curved

downhill-facing scarp resembling the head scar of a

landslide (Soeters and Rengers, 1983; Bordonau and

Vilaplana, 1986; McCalpin and Irvine, 1995; Agliardi

et al., 2001). Some authors also mention saddles in

secondary ridges (McCalpin and Irvine, 1995) or

different types of mass movements in the lower parts

of the slope (Radbruch-Hall, 1978; Crosta, 1996;

Agliardi et al., 2001).

Examples of sackung-type landforms have been

documented from most of the important mountain

belts (Radbruch-Hall, 1978; Crosta, 1996). They are

particularly frequent in steep-sided ridges flanked by

deep glacial valleys. The sackungen inventory elabo-

rated by McCleary et al. (1978) yields 400–1200 m

and 25–508 as typical ranges of slope height and

gradient. According to Varnes et al. (1989), massive

ridges with rounded crests are more favourable for

sackungen development than narrow ridges. Sackung

scarps have been observed in a wide variety of

lithologies including metamorphic, sedimentary, vol-

canic, and plutonic rocks (Radbruch-Hall, 1978;

McCalpin and Irvine, 1995). The persistent orienta-

tion of the sackung features parallel to the contour

lines indicates that topography is the main factor that

controls their trend. In numerous cases, a clear

parallelism has been found between the direction of

the sackungen and the strike of any set of disconti-

nuity planes (bedding, jointing, cleavage, and folia-

tion) (Jahn, 1964; Tabor, 1971; Bovis, 1982; Soeters

and Rengers, 1983; Varnes et al., 1989; Corominas,

1990; McCalpin and Irvine, 1995; Kellogg, 2001;

Agliardi et al., 2001; Di Luzio et al., 2004).

Several mechanisms have been proposed to explain

the origin of the uphill-facing scarps. The earliest

researchers (e.g., Paschinger, 1928) and some other

authors quoted by Tabor (1971) ascribed the sackung

scarps to structurally controlled differential erosion,

primarily due to frost action, nivation processes, and

aeolian deflation. Subsequent theories attribute sack-

ung features to deformational processes that involve

different modes of lateral spreading in rock masses

(Fig. 1). Jahn (1964), for example, suggested that the

uphill-facing scarps result from the combination of

gravity-induced spreading of fractured ridges and

erosion in the uphill sides of the resulting tension

cracks (Fig. 1A). That author stresses the downward

removal of particles through the cracks (ravelling) in

the development of the trenches. Other authors

suggest that the uphill-facing scarps and troughs are

produced by the differential downslope bend of rocks

affected by steeply dipping discontinuity planes (Ter-

Stepanian, 1966; Zischinsky, 1966, 1969; Tabor,

1971) (Fig. 1B). In this model, the troughs correspond

to linear depressions developed between the rocks

rotated valleyward and the undisturbed rocks. Tabor

(1971) believes that the zone of creep may extend as

much as 200 m below the surface. Bovis (1982)

proposes that the sackung features are produced by a

combination of flexural slip toppling of block slabs

defined by joints steeply dipping into the slope and

erosion of the upslope side of the resulting linear

A

Spreading of fracturesand erosion in uphill sides

Downslope bending ofrocks with steeply dippingplanes

Flexural toppling anderosion in uphill sides

Normal "fault" steeplydipping into the slope.

B C D

Fig. 1. Diagrams illustrating the mechanisms involving lateral spreading proposed by several authors for the generation of uphill-facing scarps.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314300

trenches (Fig. 1C). The most widely accepted

interpretation attributes the uphill-facing scarps to

the dip-slip displacement of failure planes steeply

dipping into the slope (Beck, 1968; Mollard, 1977;

Radbruch-Hall et al., 1976; Radbruch-Hall, 1978;

Varnes et al., 1989; McCalpin and Irvine, 1995;

Pasuto and Soldati, 1996; Agliardi et al., 2001) (Fig.

1D). These tensional structures are related to the

lateral spreading of the ridge flanks and the conse-

quent subsidence of the ridge crest (Radbruch-Hall,

1978; Varnes et al., 1989). In this model, the slope

troughs correspond to half-graben depressions devel-

oped on the downthrown uphill side of the fault scarp

(hanging wall). The ridge-top depressions bounded by

fault scarps with opposite orientations are the topo-

graphic manifestation of graben structures. The lateral

spreading of the rock mass perpendicular to the ridge

axis explains the bulges frequently observed in the

lower sector of the slopes affected by sackungen. The

existence of nearly vertical failure planes at the toe of

uphill-facing scarps has been corroborated at some

locations by trenching (McCleary et al., 1978;

McCalpin and Irvine, 1995; Tibaldi et al., 2004) and

in natural exposures of sackung trough fills (Beget,

1985). Very little is known about the depth and

geometry of the sackung failure planes due to the

difficulty of obtaining subsurface data. Some authors

suggest that these brittle structures are accommodated

downwards by a broad zone of continuous deforma-

tion (Mahr, 1977; Mahr and Nemcok, 1977; Rad-

bruch-Hall, 1978), whereas others believe that these

planes are linked to deep-seated discrete and contin-

uous sliding surfaces (Beck, 1968; Agliardi et al.,

2001; Tibaldi et al., 2004; Di Luzio et al., 2004). A

complete spectrum may exist between both extreme

situations. In spite of the abundant literature, numer-

ous authors consider that the mechanisms that produce

sackung features are not well understood (Radbruch-

Hall, 1978; Varnes et al., 1989; Pasuto and Soldati,

1996; Agliardi et al., 2001). Possibly the sackung

geomorphic features, although they look remarkably

similar, are a case of morphologic convergence or

equifinality.

Concerning the dynamics, several sources of

stresses have been proposed to explain the lateral

spreading involved in the generation of sackungen. (1)

The majority of the investigators attribute spreading to

the removal of the load and lateral support in glacially

oversteepened slopes by the retreat of valley glaciers.

This hypothesis has been substantiated in a particular

example of the Italian Alps with geomechanical

numerical simulations (Agliardi et al., 2001). Some

authors like Kellogg (2001) or Agliardi et al. (2001)

indicate that the unloading-induced lateral spreading

may be favoured by increased pore water pressures

during deglaciation periods. (2) Tensional stresses

caused by gravity forces acting on the upper sectors of

steep-sided ridges are considered to be capable of

producing sackung scarps (Radbruch-Hall, 1978).

This theory is supported by analytical modelling of

stresses in symmetric linear ridges and elongated

shield volcanoes (Radbruch-Hall, 1978; Savage and

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 301

Swolfs, 1986; Varnes et al., 1989; Pan and Amadei,

1994). (3) Earthquake shaking is proposed as a likely

cause of sackungen generation in active seismic areas

(Beck, 1968; McCleary et al., 1978; Radbruch-Hall,

1978; Solonenko, 1977; McCalpin, 1999). The

dynamic loading produced by earthquakes may trigger

discrete episodes of rapid movement (Beck, 1968) or

may accelerate the long-term and low-rate creep

movement of the sackung structures (Pasuto and

Soldati, 1996). (4) Uphill-facing scarps have also

been interpreted as the result of the coseismic

displacement of deep-seated tectonic faults (fault

scarps) (Cotton, 1950; Plafker, 1967; Forcella and

Orombelli, 1984; McCalpin, 1999). (5) Subsidence

due to evaporite dissolution is also a possible cause

for the generation of sackung scarps. In Teruel

Neogene Graben (Iberian Range, Spain), the localized

interstratal karstification of Triassic evaporites has

produced a monocline structure concordant with the

topography in the overlying Neogene sediments. The

conspicuous uphill-facing scarps affecting the inclined

limb of the monocline have been related to the

tensional forces caused by the dissolution-induced

subsidence phenomena (Gutierrez, 1998; Calvo et al.,

1999).

Little is known about the regime (continuous or

episodic) and rate of movement of the sackung

features. In the Southern Rocky Mountains, from a

sackung trough fill exposed by an artificial trench,

McCalpin and Irvine (1995) deduce a slow and

continuous displacement for a bfaultQ plane overlain

by fine-grained deposits affected by synsedimentary

ductile deformation and devoid of colluvial wedges.

In the North Cascades, Beget (1985) infers three to

four episodes of displacement from dated tephra

layers cut by sackung failure planes. Tibaldi et al.

(2004) infer two episodes of deformation from

trenches excavated across sackung features developed

in low relief slopes (190 m) in the Western Alps. The

episodic displacement of sackung scarps is also

claimed by those who support a causal link between

sackung features and earthquake activity (i.e., Beck,

1968). Regarding the rates of movement, in Bald

Eagle Mountain (Rocky Mountains of Colorado),

accurate geodetic measurements carried out during

more than two decades yield maximum rates of 4 mm/

year of horizontal displacement (Varnes et al., 1990,

2000). Bovis and Evans (1995) have measured geo-

detically continuous slip rates of up to 10 mm/year in

antislope scarps in the British Columbia. McCalpin

and Irvine (1995), from the stratigraphy of a sackung

trough fill, calculate mean rates of 0.14–0.75 mm/year

and 0.43 mm/year of vertical and horizontal displace-

ment, respectively.

The sackung geomorphic features have some

implications from the applied point of view. The

uphill-facing scarps may be easily misinterpreted as

recent tectonic fault scarps and as evidences of large

earthquakes, since gravitational movements com-

monly takes place along pre-existing brittle tectonic

structures (Radbruch-Hall, 1978; McCalpin, 1996).

As Beck (1968) and Varnes et al. (1989) point out, a

diagnostic criterion is that sackungen orientation is

generally determined by the local topography. McCal-

pin (1999) reviews several criteria and methods to

differentiate between nontectonic sackung scarps and

tectonic fault scarps. Despite that, in active seismic

areas, the sackung scarps may undergo episodes of

rapid movement triggered by earthquakes. Although

the probability of these slope movements to turn into

dangerous catastrophic failures is believed to be low

(Crosta, 1996), examples of catastrophic failures from

slopes affected by sackung scarps have been reported

in Japan (Chigira and Kiho, 1994) and Canada (Evans

and Couture, 2002). On the other hand, the continuous

(Varnes et al., 1990, 2000) or intermittent (Beget,

1985) differential movements in the slopes affected by

sackung pose a constraint to engineering structures

(Mahr, 1977; Radbruch-Hall, 1978; McCalpin and

Irvine, 1995).

Data about the age of sackung scarps are very

scarce. Most of the information is based on relative

chronological criteria. Beget (1985) dated three to

four episodes of spreading movement in Washington

State by means of tephra layers cut by sackung failure

planes in natural exposures. Agliardi et al. (2001)

dated an uphill-facing scarp that offsets a rock glacier

in the Alps based on the minimum age estimated for

this periglacial accumulation (5000–770 years BP). In

Bristish Columbia, Bovis (1982), applying lichen-

ometry, gives an age of 110 years for a 1 m high

sackung scarp affecting a Neoglacial moraine. Beck

(1968) reports scarps crossing active scree slopes in

the Southern Alps of New Zealand. The only radio-

metric age of an uphill-facing scarp known by the

authors has been provided by McCalpin and Irvine

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314302

(1995) from the Southern Rocky Mountains of

Colorado. These authors date the base of a sackung

trough fill (11,000–11,500 years BP) exposed by a

backhoe trench extrapolating to the base of the fill the

radiocarbon ages of three paleosoils.

This paper analyses the sackung features found in

two neighbouring glacial valleys of the central

Spanish Pyrenees. The main aspects addressed in this

work are: (1) the geomorphology of the sackung-type

features, (2) the chronology of the uphill-facing scarps

and their kinematics by means of the study of trough

fills exposed in artificial trenches, and (3) the

temporal and causal relationships between the degla-

ciation of the valleys and the gravitational spreading

phenomena.

2. General geological and geomorphological setting

The two studied locations with sackung scarps are

located in the glacial valleys of Estos and Vallibierna,

both tributaries of the Esera River Valley, a major

transverse drainage of the central Spanish Pyrenees

(Fig. 2). These areas are situated in the so-called Axial

Pyrenean Zone, a structural unit formed by strongly

deformed Paleozoic sedimentary and metasedimen-

Benasque

F R A N C E

Mt Aneto (3 404 m)

Sackungen

Kilometers

N

242o36'0''N

0o37'48''E0o32'48''E0o27'0''E

42o36'0''N

42o39'36''N

0o27'0''E 0o32'24''E 0o37'48''E

42o39'36''N

Fig. 2. Topography of the Benasque Valley headwaters and location of the studied sackung features.

tary rocks (affected by the Hercynian and Alpine

orogenies), and Late Hercynian syntectonic grano-

dioritic plutons (Garcıa-Sansegundo, 1991, 1992;

Barnolas and Pujalte, 2004). The glaciers sculptured

the main topographic features in the Esera Valley

headwaters in Pleistocene times. The area shows the

typical inherited erosional and depositional landforms

of glaciated alpine mountain environments including

horns, aretes, cirques, overdeepened basins, deep

glacial valleys flanked by steep and high-relief slopes,

moraines, and moraine-dammed glaciolacutrine

basins (Martınez de Pison, 1989; Garcıa-Ruiz et al.,

1992) (Fig. 2). Several authors have established a

general chronology for the Late Pleistocene–Holocene

glacial phases in the Pyrenees based on absolute

datings and the morpho-topographic correlation of

glacial deposits (Bordonau, 1992a; Bordonau et al.,

1992). According to this regional chronology, the

Pyrenean glaciers reached the maximum extent

between 50,000 and 45,000 years BP, long before

the Last Glacial Maximum recorded in the majority of

mountain ranges (Garcıa-Ruiz et al., 2003). At that

time, the Esera glacier was around 36 km long, from

3400 to 900 m in elevation, and reached more than

900 m in thickness in Benasque area (Martınez de

Pison, 1989; Bordonau, 1992b, 1993). Subsequent to

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 303

the Post-Maximum Stabilization Phase (31,000–

45,000 years BP) and the Valley Glaciers Phase

(N26,000 years BP), the Phase of High Altitude

Glaciers is divided into the Episode of High Altitude

Valley Glaciers and the Episode of Cirque Glaciers.

During the first episode, dated at 16,000–15,000 years

BP, the sectors of Vallibierna and Estos valleys, where

the studied sackungen are located, were still occupied

by glaciers (Copons and Bordonau, 1997). In the

Episode of Cirque Glaciers, situated at 14,000–13,000

years BP, the front of the glaciers were upstream of

these sectors. Thus, the two studied sackung slopes

were deglaciated between 16,000 and 13,000 years

BP. The present-day glaciers in the Esera watershed

are restricted to some cirques carved in the highest

granodioritic massifs (Martınez de Pison and Arenil-

las, 1988). As the glaciers have retreated towards

topographically higher positions, the profile of the

oversteepened slopes carved by the glaciers has been

partially attenuated by periglacial, fluvial, and mass

wasting processes creating talus slopes and cones,

alluvial fans, and a wide variety of slope movements

including sackungen.

The authors of this work know of four areas in the

Pyrenees where sackung type landforms have been

documented, all of them developed in outcrops of

Paleozoic metasediments of the Axial Pyrenean Zone:

(a) Bohı area (Soeters and Rengers, 1983), (b) Vielha

area (Bordonau and Vilaplana, 1986), both in the

headwaters of the Noguera Ribagorzana watershed;

(c) the Valira d’Orient Valley in Andorra (Corominas,

1990); and (d) Vallibierna Valley, one of the locations

studied in this work which was briefly described in a

previous work by Lampre (1998). In all of these cases,

the sackung features have been attributed to gravita-

tional spreading phenomena caused by deglacial

unloading in oversteepened slopes. However, this

causal and temporal relationship has not been proved

due to the lack of geochronological data (Moya et al.,

1992).

3. Sackungen in Vallibierna Valley

3.1. Geology and geomorphology

The sackung landforms are located in the north-

eastern flank of the Sierra Negra Range, which forms

the southern slopes of Vallibierna glacial valley (Fig.

2). The orientation of this relief is parallel to a WNW–

ESE trending anticlinorium composed of SW verging

folds. The metasedimentary rocks that form the slope

affected by the sackung scarps strike parallel to the

topographic grain and show a general dip towards the

valley (Rıos et al., 1997). Most of the slope is

underlain by tightly folded and densely jointed

Silurian black slates. These rocks have a high content

of carbonaceous material and pyrite. The lowermost

part of the slope and a relatively small patch located in

the middle–upper sector of the slope are formed by a

sequence of dark limestone and black slates Devonian

in age (Garcıa-Sansegundo, 1991, 1992; Rıos et al.,

1997).

The slopes in Vallibierna Valley show a composite

profile with a relatively gentle upper segment and a

steep lower segment. The upper segments reflect the

preglacial topography of the valley, whereas the lower

segments delineate the trough excavated in glacial

times. This geomorphic evidence indicates that the ice

reached more than 300 m thick in this sector of the

valley during the Last Glacial Maximum (LGM).

Considering that the density of the glacial ice is

around 0.9 g/cm3 (Benn and Evans, 1998), a stress

release in excess of 27 kg/cm2 (2.7 MPa) can be

estimated for the rocks located in the valley bottom

since the LGM (50,000–45,000 years BP). Terminal

moraines have been mapped in the valley about 1 km

downstream and 2–3 km upstream of the slope

affected by sackungen (Martınez de Pison, 1989;

Garcıa-Ruiz et al., 1992; Copons and Bordonau,

1997; Lampre, 1998). Copons and Bordonau (1997),

based on morpho-topographic criteria, correlate these

deposits to the Episode of High Altitude Valley

Glaciers (16,000–15,000 years BP) and the Episode

of Cirque Glaciers (14,000–13,000 years BP), respec-

tively. Thus, according to this chronology, the sector

of the valley where the sackungen are located was

deglaciated sometime between 16,000 and 13,000

years BP.

The slope with sackung scarps has a local relief of

750 m, from 2691 to 1940 m, and an average gradient

of 19.58 (Figs. 3 and 4). A concave ridge defines the

crest of the slope with a conspicuous downhill facing

scarp (Lampre, 1998) up to 50 m high that resembles

the head scar of a landslide. A similar situation has

been described by several authors in slopes with

Fig. 3. Geomorphological sketch of the slopes affected by antislope scarps in Vallibierna and Estos valleys and location of the trenches

excavated in the sackung trough fills.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314304

counter scarps (Soeters and Rengers, 1983; Bordonau

and Vilaplana, 1986; McCalpin and Irvine, 1995;

Agliardi et al., 2001). An elongated depression with

an ephemeral pond has been mapped at the foot of this

scarp north of Estibafreda Peak. The depression

hosting the Ardones Pond constitutes a NW–SE

ridge-top graben-like morphostructure flanked by the

main downhill-facing scarp and two uphill-facing

scarps (Fig. 5C). On the other hand, the profile of the

upper sector of the slope shows an undulating

topography. All these features give the slope the

appearance of being affected by a landslide; however

no obvious geomorphic features allow identifying the

margins and toe of such type of slope movement. In

addition, the lower portion of the slope and the trace

of Vallibierna River do not show any evidence of

postglacial outward bulging.

The main complex of uphill-facing scarps is

located in the intermediate sector of the slope,

between 2500 and 2100 m (Figs. 3, 4, and 5A).

This distribution contrasts with the majority of the

documented sackung scarps, commonly located in

the crestal area of the slopes (Radbruch-Hall, 1978;

Varnes et al., 1989). These geomorphic features

follow a WNW–ESE trend parallel to the structural

and topographic grain (Lampre, 1998) and some of

them show a slight valleyward convexity. Very

likely, the failures have occurred along pre-existing

discontinuity planes. The length of the antislope

scarps ranges from a few tens of meters to around

500 m. Some of the uphill-facing scarps are

accompanied by linear depressions with a flat and

turf-covered aggradational surface underlain by fine-

grained deposits (Fig. 5B). The scarps, with heights

varying form 0.4 to 3.6 m, do not show a fresh

appearance, indicating that they are not active or that

the degradation rate exceeds the vertical component

of the deformation rate. Regarding the hydrology of

Fig. 5. Photographs of the sackung features in Vallibierna Valley. (A) General view of the antislope scarps. (B) Excavated sackung trough. (C)

Ridge-top depressions flanked by downhill and uphill-facing scarps in Ardones Pond area. (D) Trench in a trough fill.

Fig. 4. Stereoscopic images of the sackung area in Vallibierna Valley.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 305

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314306

the slope, it shows well-defined creeks which dissect

some sackung scarps. In the lower portion of the

slope, there is also a permanent spring of Fe-rich

Bedrock slope

ColluviumTrough Sackung ridge

Bedrock slopeTrench

ENEWSW

0

2

4

6

8

10

12

100 20 30 40 50 60 70 m

20°

11°

Bedrockslope

Trough

Sackung ridge

Trench

NESW

0

2

4

6

8

10

12

100 20 30 40 50 60 m

23°

Bedrock slope

20,5°

29°

Weatheredbedrock

Angulargravel

Marls withclasts

Organic soilhorizon

Oxidationhorizon

Sackung "fault" N 116 E 76 STrench limitWeathered bedrock

Weathered bedrock

Weatheredbedrock

Angularclasts

Clay andscattered clasts

Organic soilhorizon

WSW ENEVALLIBIERNA

c2

b2c1

b1

a2

a15850 (5890) 5900cal yr BP

5580 (5640) 5750cal yr BP

2700 (2750) 2850cal yr BP

m 6 5 4 3 2 1 0

0 m

1

6400 (6650) 6790cal yr BP

5860 (5890) 5910cal yr BP

3460 (3480) 3570cal yr BP

Weathered bedrock

ESTÓSSW NE

Trench limit

m 6 5 4 2 1

0 m

1

2

z

y

x

03

ig. 6. Diagrams of the trenches dug in the sackung trough fills of Vallibierna and Estos and longitudinal profiles of the slopes showing the

cation of the trenches. The datings cited are calendar-corrected ages (see Table 1 for conventional radiocarbon ages).

F

lo

waters that give place to an actively forming

accumulation ferruginous sinter (Fig. 3). Probably,

the Fe content of the water is derived from the

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 307

alteration of the pyrite of the Silurian slates under

oxidizing conditions. This spring reveals that a

relatively large quantity of water flows through the

rock mass forming the slope.

3.2. Trench stratigraphy and geochronology

The trench of Vallibierna was excavated in a

sackung trough at about 2260 m situated next to the

tree line (Figs. 3, 4, and 5). The reasons for selecting

this location include the existence of an area affected

by frequent ponding in the closed depressions and the

presence of trees in the vicinity. Both circumstances

suggested that datable material like charcoal and/or

organic horizons could be found in the trough fill. The

trench was oriented perpendicular to the sackung

ridge and trough, and its downslope end was located

at the slope break between the ridge and the

depression (Figs. 5D and 6). The excavation, 1.5 m

deep and 5.5 m long, was performed with pick, hoe,

and shovel. The longitudinal profile of the slope

across the sackung ridge and trough is shown in Fig.

6. It shows that the sackung ridge does not have an

obvious scarp next to the flat-bottomed depression,

which grades into a colluvium slope. In addition, the

bedrock hillslope below the sackung is ca. 98 steeperthan that above the sackung, suggesting that the ridge

has experienced an outward toppling (Jahn, 1964;

Bovis, 1982; McCalpin and Irvine, 1995).

The artificial trench exposes three fining-upward

sequences, each made of two units (a, b, and c) lying

on weathered bedrock (Fig. 6). The lower units (a1,

b1, and c1), dark brown-grey in colour, are composed

of pebble-sized angular clasts with a clay matrix. The

upper units (a2, b2, and c2) consist of black, dark grey

marly clay with scattered pebble-sized clasts. In the

Table 1

Summary of radiocarbon ages provided by Beta Analytic, from charcoal s

Laboratory number

(Beta Analytic)

Method Unit (location

h-171813 Radiometric b2 (Vallibierna

h-171812 Radiometric a2 (Vallibierna

h-171811 AMS a1 (Vallibierna

h-171810 AMS z (Estos)

h-171809 AMS y (Estos)

h-171808 Radiometric x (Estos)

Flanking calendar ages represent 1r limits.

upper sequence (c2), two stacked organic soil

horizons have been identified. The lower units of

each of the sequences, up to 20 cm thick and with no

distinctive bedding planes, are interpreted as sheet-

flow deposits derived from the upper slope and

accumulated in single rapid events. The upper units

represent low energy and slow sheetwash deposition

in a closed depression subjected to ponding. The

presence of only two soil horizons at the top of the

trough fill suggests a continuous aggradation with no

significant depositional hiatus. Radiocarbon ages have

been obtained from small pieces of charcoal found in

three stratigraphic units (a1, a2, and b2) of the

exposed trough fill. A summary of the conventional

radiocarbon ages and the calendar ages is shown in

Table 1. Units a1 and a2 yield 5890 and 5640 cal.

years BP, respectively, and unit b2 dates at 2750 cal.

years BP. The date of unit a1 (5890 cal. years BP),

lying directly on weathered bedrock, may be consid-

ered the age of the sackung trough and scarp since it

represents an instantaneous depositional event that

took place just after the uphill-facing scarp formed the

depression.

Regarding the structure, the lower sequence (a) is

affected by a normal fault that dips 768 into the slope

and has a N116E strike (Fig. 6). The fault shows 22

cm of dip slip displacement and about 23 cm of

vertical offset subsequent to deposition of the lower

sequence. The movement of the fault has produced the

bending (drag folds) of the lower unit on both sides of

the structure. The upper unit of this pre-fault sequence

is only preserved on the hanging wall. This sackung

fault is truncated by an erosional surface and

fossilized by the two upper sequences showing an

onlap contact with the upthrown block. This geo-

metrical configuration indicates that the fault was

amples

) Conventional

age (14C years BP)

Calendar age

(cal. years BP)

) 2620F150 2700 (2750) 2860

) 4920F130 5580 (5640) 5750

) 5080F40 5850 (5890) 5900

3290F40 3460 (3480) 3570

5090F40 5860 (5890) 5910

5830F160 6440 (6650) 6790

Fig. 7. Sketch showing the inferred evolution of the sackung scarp

and trough excavated in Vallibierna.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314308

generated later than the deposition of unit a2 and just

before the accumulation of unit b1. With the available

datings, this deformation can be situated between

Fig. 8. Stereoscopic images of the

5640 and 2750 cal. years BP. The fact that the bedrock

in the footwall is overlain by deposits of the lower

sequence suggests that the depression was initially

formed by a master sackung scarp located beyond the

trench towards the downslope side (Fig. 7). This scarp

was formed soon before the deposition of unit a1

(5890 cal. years BP). Subsequent to deposition of unit

a2, a new sackung scarp formed within the depression

between 5640 and 2750 cal. years BP, causing the

erosion of unit a2 in the upthrown side. Once this

scarp was eliminated, the accumulation of the two

upper sequences took place fossilizing the fault and

onlapping the previously uplifted block.

4. Sackungen in Estos Valley

4.1. Geology and geomorphology

The sackung scarps of Estos are situated on a slope

on the northeastern flank of the valley with a NW–SE

orientation (Fig. 2). In this case, the topographic grain

is also controlled by the structure, a NW–SE trending

synclinorium composed of south verging folds

affected by north dipping thrusts. The axis of the

synclinorium runs roughly along the main ridge and

the strata show a general NE dip into the slope (Rıos

et al., 1997). The upper half of the slope is formed by

Devonian slates with intercalated arenaceous beds and

sackung area in Estos Valley.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 309

the lower half is underlain by Devonian slates and

limestone (Garcıa-Sansegundo, 1991, 1992; Rıos et

al., 1997). These formations are affected by a dense

network of discontinuity planes. Several geomorphic

features like the composite profile of the valley slopes

and truncated bedrock spurs indicate that the thickness

of the Estos glacier was more than 350 m during the

Last Glacial Maximum (50,000–40,000 years BP).

Thus deglaciation resulted in unloading in excess of

31.5 kg/cm2 (3.15 MPa) for the rocks located in the

valley bottom. Although there are no specific studies

about the glacial chronology in the Estos Valley, it can

be assumed that deglaciation took place approxi-

mately at the same time as in Vallibierna Valley

(16,000–13,000 years BP), given the marked topo-

graphic similarity and proximity between both valleys

(Martınez de Pison, 1989) (Fig. 2).

The slope affected by the uphill-facing scarps has

an average gradient of 228 and a local relief of about

900 m, between 2420 and 1500 m. The upper sector

of the slope is carved by a shallow glacial cirque

(Garcıa-Ruiz et al., 1992) that hosted an ice body once

linked to the main Estos glacier (Figs. 3 and 8). To the

west of Estos Peak there are two tongue-shaped talus

rock glaciers in direct connection with active talus

Fig. 9. Photographs of the sackung-type landforms in Estos Valley. (A an

Excavated sackung trough. (D) Close-up view of trench.

slopes (Barsch, 1996). Several characteristics, such as

the subdued and subsided topography of the inner part

of the rock glaciers and the dense turf vegetation of

the front and outer sides, suggest that these are relict

accumulations free of internal ice (Barsch, 1996).

Only one antislope scarp has been identified in the

crestal area. As in Vallibierna, most of the uphill-

facing scarps are located in the middle sector of the

slope, between 2070 and 1800 m (Figs. 3 and 8). The

upper ones clearly disrupt the glacial erosional

topography in the cirque bottom. These morphostruc-

tures and the accompanying troughs show a NW–SE

direction parallel to the topographic contours and the

structural grain. This is also the strike of a penetrative

set of cleavage planes dipping into the slope measured

in some outcrops located close to the front of the

largest rock glacier. Although probable, it is not

possible to ascertain whether the sackung features

correspond to pre-existing bedding or cleavage planes

since the dip of the failure surfaces is unknown. The

uphill-facing scarps range in length from 80 to 500 m

and the height varies from 0.6 to 2.75 m (Fig. 9).

Although the scarps do not show a fresh appearance,

they are more conspicuous than the scarps of

Vallibierna. This is probably due to the higher

d B) General view of a sackung antislope scarps and troughs. (C)

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314310

resistance to erosion of the bedrock in Estos.

Concerning the hydrology, the upper half of the slope

underlain by slates is devoid of gullies, suggesting

that most of the surficial water infiltrates in the slope.

The lower and steeper half of the slope is dissected by

two gully systems, which show some stretches

controlled by the sackung troughs.

4.2. Trench stratigraphy and geochronology

The sackung trough excavated in Estos Valley is

located at about 2000 m in elevation, close to the tree

line, although in this sector the upper limit of the

forest has been lowered for grazing. The flat bottom of

this closed depression is covered by grass and shows

evidence of frequent ponding. As in Vallibierna, the

trench was aligned perpendicular to the antislope

scarp with the downslope end located at the foot of the

scarp face (Fig. 9C and D). The excavation, dug with

manual tools, reached weathered bedrock at a depth of

2.1 m (Fig. 6). Similarly to the studied sackung in

Vallibierna, the longitudinal profile of the slope shows

that the bedrock hillslope located below the trough is

steeper than the slope segment above the trough,

probably due to outward toppling of the sackung ridge

(Jahn, 1964; Bovis, 1982; McCalpin and Irvine, 1995)

(Fig. 6). In this case, the sackung shows a well-

defined 1.1 m high uphill-facing scarp (Fig. 9C).

The 2.1 m thick trough fill exposed in the trench

shows a horizontal structure with no evidence of

synsedimentary deformation. It is dominantly com-

posed of massive beds of dark brown-grey marls with

Fig. 10. Plot representing the calendar age and the depth of the dated beds

fast and slow rates of deposition to the bottom of the trough provides an

scattered granule and pebble-sized subangular clasts

(Fig. 6). Some organic soil horizons and two

oxidation horizons have been identified within the

marl beds. Two layers of angular clasts are interca-

lated in the marl sequence. The lower bed, between

1.53 and 1.75 m below the ground surface, is made up

of angular cobbles with a clast-supported texture and a

marly matrix. The upper coarse-grained bed, between

0.85 and 1.01 m, consists of clast-supported fine

angular boulders with a marly matrix, following Blair

and McPherson’s (1999) grain size terminology. The

lack of well-developed paleosoils suggests a contin-

uous deposition with no significant hiatus. The

massive dark marl beds represent deposition in a

ponded area under reducing conditions. The angular

clasts embedded with the marls were probably trans-

ported by sheet flow or dry ravel from the adjacent

slope. The thin organic soil horizons record short

sedimentary interruptions and the oxidation horizons

periods of desiccation. The angular gravel beds are

interpreted as the product of rapid high-energy sheet

flow events probably caused by high intensity storm

events.

Small pieces of charcoal have been collected from

three layers of the trough fill (x, y, and z) located at

1.75–1.9, 1.53–1.75, and 0.85–1.01 m below the

ground surface. Layer x is composed of marls with

scattered clasts, and y and z layers correspond to the

two gravel beds. The calibrated ages provided for x, y,

and z units are 6650, 5890, and 3480 cal. years BP,

respectively. The conventional radiocarbon ages are

supplied in Table 1. The graph representing the

in the sackung trough fill of Estos. The downward projection of the

estimate of the age of the sackung.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 311

calendar age of the dated beds versus the depth below

the ground surface provides two rates of deposition:

0.29 mm/year between 3480 and 5890 cal. years BP

and 0.24 mm/year between 5890 and 6650 cal. years

BP (Fig. 10). The projection of the plot using both

rates yield an age range of 7795–7589 years BP for

the bottom of the sackung, situated 2.1 m below the

ground surface. Considering that the trough fill shows

a relatively continuous and homogenous deposition

and that the lowest dated bed is very close to the

bottom of the trough, this temporal range can be

considered as an acceptable estimate of the age of the

sackung trough and scarp.

5. Discussion and conclusions

The available data suggest that the studied uphill-

facing scarps and the adjacent depressions were

generated by the combination of at least two types

of movement. On the one hand, dip-slip displacement

occurred along failure planes steeply dipping into the

slope. The sackung fault identified in the trench of

Vallibierna proves the existence of this type of brittle

structure. On the other hand, the higher inclination of

the bedrock slopes located below the sackung troughs

compared to that of the hillslopes above the troughs

suggests that blocks defined by the sackung faults

have undergone a forward toppling with the conse-

quent horizontal extension. Although the obtained

dating provides the age of an individual antislope

scarp and depression for each location, it is likely that

the whole complex of sackung features in each slope

formed during a relatively short time span, given their

remarkably similar appearance.

Regarding the chronology, the Estos and Valli-

bierna uphill-facing scarps formed at about 7.6–7.8

and 5.9 cal. ka BP, respectively, at least 5000–6000

years after the deglaciation of the valleys that occurred

between 16 and 13 ka BP (1 ka=1000 years). A

considerable lag between deglaciation (14–15 ka BP)

and the onset of sackung generation (11–11.5 ka BP)

was also found by McCalpin and Irvine (1995) in the

Southern Rocky Mountains. The temporal distance

(N5 ka) between deglaciation and the age of the

studied sackungen indicates that the unloading of the

oversteepened slopes produced by withdrawal of the

valley glaciers was not the direct cause of the uphill-

facing scarps in Estos and Vallibierna, as has been

suggested for some sackung features studied else-

where in the Pyrenees (Soeters and Rengers, 1983;

Bordonau and Vilaplana, 1986; Corominas, 1990). In

our case, it seems that glacial erosion and the

subsequent debutressing of the valley sides generated

slopes predisposed to sackung development but did

not initiate the lateral spreading phenomena. There-

fore, a different triggering factor must be invoked to

explain the temporal occurrence of the antislope

scarps. Two possible options are climate change and

earthquake shaking.

Numerous investigations demonstrate a good corre-

lation between the concentration in time of landslides,

including large deep-seated slope movements, and

climatic changes. The available data from several

European regions concerning these temporal relation-

ships have been recently reviewed in several papers

(Borgatti et al., 2001; Borgatti and Soldati, 2002;

Soldati et al., 2004). Obviously, in our case, the

possible causal relationship between the lateral spread-

ing movements and climate is merely speculative since

we only have two dates. On the other hand, the

triggering effect of climatic factors on these type of

deep-seated large movements is much more limited

than on shallow and small mass wasting processes.

However, it may be interesting to find out if the

sackungen of Vallibierna and Estos were formed during

periods of high landslide frequency. The initial gravita-

tional spreading in both valleys took place during the

Atlantic period, characterized by a relatively warm and

dry climate (Lamb, 1977). In the Pyrenees, the scarce

number of datings does not permit identification of

clustering periods of landslides (Moya et al., 1992). In

the Cantabrian Mountains, the prolongation of the

Pyrenees to the west, a period with a relatively high

frequency of landslides, has been identified from the

Preboreal to the beginning of the Atlantic periods

(10.2–7 ka BP) (Gonzalez Dıez et al., 1996, 1999),

which partially overlaps the time interval when our

sackung were generated (7.8–7.6 and 5.9 ka).

Concerning seismic activity, Beck (1968) proposed

that the sackung scarps he studied in New Zealand were

formed by discrete episodes of rapid movement

triggered by earthquakes. Radbruch-Hall (1978)

reports that the generation of fresh antislope scarps in

Alaska and California has been attributed to historic

earthquakes. McCalpin (1999) quotes several studies

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314312

that document the generation or rejuvenation of

sackung-like landforms during historic earthquakes.

Several arguments support the hypothesis that earth-

quake-induced dynamic loading was involved in the

development of the sackung scarps. The studied slopes

are located in a seismically and neotectonically active

area. In 1373, an earthquake with a MSK intensity of

VIII–IX destroyed buildings in this area of the

Pyrenees (Olivera et al., 1994). According to the IGN

data base, five earthquakes with maximum MSK

intensities higher than V have been felt during the last

century in areas located at a distance less than 25 km

from the study area. The 1923 earthquake of Viella, a

village located 14 km to the NE of the study area,

reached a maximum MSK intensity of VIII (IGN,

1982) and caused severe damage to structures includ-

ing Romanasque churches. Susagna et al. (1994), based

on the analysis of macroseismic and instrumental data,

attribute a magnitude ML=5.6 and MW=5.3 to this

event. Ortuno et al. (2004) propose the Northern

Maladeta Fault as the most likely source of this

earthquake. This ENE–WSW structure crosses the

headwaters of Benasque Valleys. Ortuno et al. (2004)

identify a planation surface offset up to 475 m on both

sides of this fault and estimate a slip rate between 0.04

and 0.07 mm/year for the last 10 Ma based on the

elevation of the base of lacustrine sediments of

probable Upper Miocene age deposited in a tectonic

basin generated in the downthrown block. According to

the INQUA intensity scale based on seismically

induced ground effects (Michetti et al., 2003), earth-

quakes with intensities of VII and IX may trigger large

mass movements (105–106 m3). The largest instrumen-

tally recorded earthquake in Benasque Valley reached

4.1 in Richter magnitude. The epicentre of this event,

which occurred in 1982, was located less than 10 km to

the south of the studied sackung features. Several

earthquakes with magnitudes higher than 5 have been

recorded in the eastern sector of the Pyrenees (IGN,

2004). Some other geomorphic evidences of neo-

tectonic activity have been documented in the area. In

Barrancs and Escaleta Valleys, at about 9 km from the

studied sackung features, Moya and Vilaplana (1992)

reported conspicuous normal faults offsetting glacial

erosional landforms. Evidence of post-glacial faulting

(b30 ka BP) has also been documented in the adjacent

Noguera Ribagorzana Valley (Bordonau andVilaplana,

1986). The hypothetical participation of the seismic

activity is also supported by the sudden deformation

event recorded by the fault identified in Vallibierna

trench. Once the sackungen are initiated by an earth-

quake, the subsequent development could result from

earthquake-triggered incremental movements, or from

a continuous slow creep interrupted by coseismic

episodes of rapid displacement. However, although

the seismic hypothesis seems to be more probable than

the climate change option, the lack of chronological

information about palaeoearthquakes precludes the

corroboration of this interpretation (Jibson, 1996;

McCalpin, 1999).

Acknowledgements

The authors would like to thank Drs. James

McCalpin (Crestone Science Center, Colorado) and

Jose Marıa Garcıa-Ruiz (Pyrenean Institute of Ecology,

Spain) for the their corrections and suggestions, which

have helped to substantially improve the original

manuscript. The paper has also been benefited from

the thorough review of Drs. Pablo Silva and Martin

Thorp. We are also grateful to Mr. Inocencio Altuna

(Head of Posest-Maladeta Natural Park) for providing

the permit to dig the trenches in the Posets-Maladeta

protected area. The investigation has been financially

supported by the DAMOCLES project (N.EVG1-CT-

1999-00007) founded by the European Union.

References

Agliardi, F., Crosta, G., Zanchi, A., 2001. Structural constraints on

deep-seated slope deformation kinematics. Engineering Geology

59, 83–102.

Barsch, D., 1996. Rock-glaciers. Springer, Berlin. 331 pp.

Barnolas, A., Pujalte, V. (Ed.), 2004. La Cordillera Pirenaica. In:

Vera, J.A. (Ed.), Geologıa de Espana. IGME-SGE, Madrid,

pp. 233–343.

Beck, A.C., 1968. Gravity faulting as a mechanism of topographic

adjustment. New Zealand Journal of Geology and Geophysics

11, 191–199.

Beget, J.E., 1985. Tephrochronology of antislope scarps on an

alpine ridge near Glacier Peak, Washington, U.S.A. Arctic and

Alpine Research 17, 143–152.

Benn, D.I., Evans, D.J.A., 1998. Glaciers and Glaciation. Arnold,

London. 734 pp.

Blair, T.C., McPherson, J.G., 1999. Grain-size and textural

classification of coarse sedimentary particles. Journal of

Sedimentary Research 69, 6–19.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314 313

Bordonau, J., 1992a. La cronologıa del ultimo ciclo glaciar en los

Pirineos. VIII Congreso Latinoamericano de Geologıa Sala-

manca, Simposios, vol. 2, pp. 48–53.

Bordonau, J., 1992b. Els complexos glacio-lacustres relacionats amb

el darer cicle glacial als Pirineus. Geoforma, Logrono. 251 pp.

Bordonau, J., 1993. The Upper Pleistocene ice-lateral till complex

of Cerler (Esera valley, Central Southern Pyrenees; Spain).

Quaternary International 18, 5–14.

Bordonau, J., Vilaplana, J.M., 1986. Geomorphologie et tectonique

recente dans le Val d’Aran (zone axiale des Pyrenees Centrales,

Espagne). Revue de Geologie Dynamique et de Geographie

Physique 27, 303–310.

Bordonau, J., Serrat, D., Vilaplana, J.M., 1992. Las fases glaciares

cuaternarias en los Pirineos. In: Cearreta, A., Ugarte, F.M.

(Eds.), The Late Quaternary in the Western Pyrenean region.

Universidad del Paıs Vasco, Bilbao, pp. 303–312.

Borgatti, L., Soldati, M., 2002. The influence of Holocene climatic

changes on landslide occurrence in Europe. In: Rybar, J.,

Stemberk, J., Wagner, P. (Eds.), Lanslides. A.A. Balkema

Publishers, Lisse, pp. 111–116.

Borgatti, L., Soldati, M., Surian, N., 2001. Rapporti tra frane e

variazioni climatiche: una bibliografia ragionata relativa al

territorio europeo. Il Quaternario 14 (2), 137–166.

Bovis, M.J., 1982. Uphill-facing (antislope) scarps in the Coast

Mountains, southwest British Columbia. Geological Society of

America Bulletin 93, 804–812.

Bovis, M.J., Evans, S.G., 1995. Rock slope movements along the

Mount Currie bfault scarp,Q Southern Coast Mountains, British

Columbia. Canadian Journal of Earth Science 32, 2015–2020.

Calvo, J.P., Alcala, L., Alonso-Zarza, A.M., van Dam, J., Gutierrez

Santolalla, F., 1999. Estratigrafıa y estructura del area de Los

Mansuetos (Cuenca de Teruel). Precisiones para la definicion

del estratotipo del Turoliense. Geogaceta 25, 55–58.

Chigira, M., Kiho, K., 1994. Deep-seated rockslide avalanches

preceded by mass rock creep of sedimentary rocks in Akaishi

Mountains, central Japan. Engineering Geology 38, 221–230.

Copons, R., Bordonau, J., 1997. El ultimo ciclo glalciar (Pleistone

superior—Holoceno) en el Macizo de la Maladeta (Pirineos

Centrales). Revista de la SociedadGeologica de Espana 10 (1–2),

55–66.

Corominas, J., 1990. Influencia del glaciarismo cuaternario en la

estabilidad de las laderas del Valle del Valira d’Orient (Andorra).

In: Gutierrez, M., Pena, J.L., Lozano, M.V. (Eds.), I Reunion

Nacional de Geomorfologıa, vol. 2. Sociedad Espanola de

Geomorphlogıa, Teruel, pp. 521–532.

Cotton, C.A., 1950. Tectonic scarps and fault valleys. Geological

Society of America Bulletin 61, 717–757.

Crosta, G., 1996. Landslide, spreading, deep seated gravitational

deformation: analysis, examples, problems and proposals. Geo-

grafia Fisica e Dinamica Quaternaria 19, 297–313.

Di Luzio, E., Saroli, M., Esposito, C., Bianchi-Fasani, G., Cavinato,

G.P., Scarascia-Mugnozza, G., 2004. Influence of structural

framework on mountain slope deformation in the Maiella anti-

cline (Central Apennines, Italy). Geomorphology 60, 417–432.

Evans, S.G., Couture, R., 2002. The 1965 Hope Slide, British

Columbia; catastrophic failure of a sagging rock slope. Abstracts

with Programs-Geological Society of America, 16–26.

Forcella, F., Orombelli, G., 1984. Holocene slope deformations in

Valfurva, Central Alps, Italy. Geographıa Fısica e Dinamica

Quaternaria 7, 41–48.

Garcıa-Ruiz, J.M., Bordonau, J., Martınez de Pison, E., Vilaplana,

J.M., 1992. Mapa Geomorfologico de Benasque. Geoforma,

Logrono. 39 pp.

Garcıa-Ruiz, J.M., Valero-Garces, B.L., Martı-Bono, C., Gonzalez-

Samperiz, P., 2003. Asynchroneity of maximum glacier

advances in the central Spanish Pyrenees. Journal of Quaternary

Science 18 (1), 61–72.

Garcıa-Sansegundo, J., 1991. Estratigrafıa y estructura de la Zona

Axial pirenaica en la transversal del Valle de Aran y de la Alta

Ribagorza (Parte I). Boletın Geologico y Minero 102, 781–829.

Garcıa-Sansegundo, J., 1992. Estratigrafıa y estructura de la Zona

Axial pirenaica en la transversal del Valle de Aran y de la Alta

Ribagorza (Parte II). Boletın Geologico y Minero 103, 42–93.

Gonzalez Dıez, A., Salas, L., Dıaz de Teran, J.R., Cendrero, A.,

1996. Late Quaternary climate changes and mass movement

frequency and magnitude in the Cantabrian region, Spain.

Geomorphology 15, 291–309.

Gonzalez-Dıez, A., Remondo, J., Diaz de Teran, J.R., Cendrero, A.,

1999. A methodological approach for the analysis of the

temporal occurrence and triggering factors of landslides. Geo-

morphology 30, 95–113.

Gutierrez, F., 1998. Fenomenos de subsidencia por disolucion de

formaciones evaporıticas en las fosas neogenas de Teruel y

Calatayud (Cordillera Iberica). PhD Thesis, University of

Zaragoza, Zaragoza. 569 pp.

IGN, 1982. Catalogo general de isosistas de la Penınsula Iberica.

Instituto Geografico Nacional, Madrid. Publ. 202. 59 pp.

IGN, 2004. Boletın de sismos proximos. Ministerio de Fomento,

Madrid. (web site: http://www.geo.ign.es).

Jahn, A., 1964. Slope morphological features resulting from

gravitation. Zeitschrift fqr Geomorphologie. Supplementband

5, 59–72.

Jibson, R.W., 1996. Using landslides for paleoseismic analysis. In:

McCalpin, J. (Ed.), Paleoseismology. Academic Press, London,

pp. 397–438.

Kellogg, K.S., 2001. Tectonic controls on a large landslide complex:

William Fork Mountains near Dillon, Colorado. Geomorphol-

ogy 41, 355–368.

Lamb, H.H., 1977. Climate. Present, Past and Future, vol. 2.

Methuen & Co. Ltd., London. 835 pp.

Lampre, F., 1998. Estudio geomorfologico de Ballibierna.

Consejo de Proteccion de La Naturaleza de Aragon,

Zaragoza. 200 pp.

Mahr, T., 1977. Deep-reaching gravitational deformations of high

mountain slopes. Bulletin of the International Association of

Engineering Geology 16, 121–127.

Mahr, T., Nemcok, A., 1977. Deep-seated creep deformations in the

crystalline cores of the Tatry Mts. Bulletin of the International

Association of Engineering Geologists 16, 104–106.

Martınez de Pison, E., 1989. Morfologıa glaciar del Valle de

Benasque (Pirineo Aragones). Eria, 51–64.

Martınez de Pison, E., Arenillas, M., 1988. Los glaciares actuales

del Pirineo espanol. La nieve en el Pirineo espanol. MOPU,

Madrid, pp. 29–98.

F. Gutierrez-Santolalla et al. / Geomorphology 69 (2005) 298–314314

McCalpin, J., 1996. Paleoseismology. Academic Press, San Diego.

583 pp.

McCalpin, J., 1999. Criteria for determining the seismic signifi-

cance of sackungen and other scarplike landforms in mountain-

ous regions. Techniques for Identifying Faults and Determining

their Origins. U.S. Nuclear Regulatory Commission, Washing-

ton, pp. 2.55–2.59.

McCalpin, J.P., Irvine, J.R., 1995. Sackungen at the Aspen

Highlands ski area, Pitkin County, Colorado. Environmental

and Engineering Geoscience 1, 277–290.

McCleary, J., Dohrenwend, J., Cluff, L., Hanson, K. 1978., 1872

earthquake studies, Washington public power supply system

Nuclear Project Numbers 1 and 4. Woodward-Clyde Consul-

tants, San Francisco. 75 pp., unpublished report.

Michetti, A.M., Esposito, E., Mohammadioun, J., Mohammadioun,

B., Gqrpinar, A., Porfido, S., Rogozhin, E., Serva, L.,

Tatevossian, R., Vittori, E., Audemard, F., Comerci, V., Marco,

S., McCalpin, J., Morner, N.A., 2003. An innovative approach

for assessing earthquake intensities: the proposed INQUA scale

based on seismically-induced ground effects in the environment.

XVI INQUA Congress, Reno. 97 pp., http://www.apat.gov.it/

site/en-GB/Projects/INQUA_Scale/Documents/.

Mollard, J.D., 1977. Regional landslide types in Canada.

Reviews in Engineering Geology 3. Geological Society of

America, pp. 29–56.

Moya, J., Vilaplana, J.M., 1992. Tectonica reciente en el Macizo

de la Maladeta, sector del alto Esera (Pirineo Central). In:

Cearreta, A., Ugarte, F.M. (Eds.), The Late Quaternary in the

Western Pyrenean Region. Universidad del Paıs Vasco, Bilbao,

pp. 385–403.

Moya, J., Vilaplana, J.M., Corominas, J., 1992. Late Quaternary and

historical landslides in the south-eastern Pyrenees. In: Mat-

thews, J.A., Brunsden, D., Frenzel, B., Gl7ser, B., Weih, M.M.

(Eds.), Rapid Mass Movement as a Source of Climatic Evidence

for theHolocene, Pal7oklimaforschung—Paleoclimate Research,

vol. 19, pp. 55–73.

Nemcok, A., 1972. Gravitational slope deformation in high

mountains. Proceedings of the 24th international geological

congress, Montreal, sect. 13, 132–141.

Olivera, C., Riera, A., Lambert, J., Banda, E., Alexandre, P., 1994.

Els terratremols de l’any 1373 al Pirineu. Effectes a Espanya i

Franca. Servei Geologic de Catalunya. 220 pp.

Ortuno, M., Perea, H., Masana, E., Santanach, P., 2004. La falla

Norte de la Maladeta, Sfuente sısmica del terremoto de Biela (11

de Noviembre de 1923)? Geotemas 6 (3), 171–174.

Pan, E., Amadei, B., 1994. Stresses in an anisotropic rock mass with

irregular topography. Journal of Engineering Mechanics 120,

97–119.

Paschinger, V., 1928. Untersuchungen qber dopplegrate. Zeitschriftfqr Geomorphologie 3, 204–236.

Pasuto, A., Soldati, M., 1996. Rock spreading. In: Dikau, R.,

Brunsden, D., Schrott, L., Visen, M.-L. (Eds.), Landslide

recognition. Identification, movement and causes. Wiley, Chi-

chester, pp. 122–136.

Plafker, G., 1967. Surface faults on Montague Island associated

with the 1964 Alaska earthquake. U.S. Geological Survey

Professional Paper 543-G (42 pp.).

Radbruch-Hall, D.H., 1978. Gravitational creep of rock masses on

slopes. In: Voight, B. (Ed.), Rockslides and Avalanches, 1.

Natural Phenomena. Developments in Geotechnical Engineering.

vol. 14A. Elsevier, Amsterdam, pp. 607–657.

Radbruch-Hall, D.H., Varnes, D.J., Savage, W.Z., 1976. Gravita-

tional spreading on steep sided ridges (bSackungQ) in western

United States. Bulletin of the International Association of

Engineering Geology 14, 23–35.

Rıos, L.M., Galera, J.M., Barettino, D., 1997. Mapa Geologico de

Espana, Hoja de Benasque (180). Instituto Geologico y Minero

de Espana, Madrid. 78 pp.

Savage, W.Z., Swolfs, H.S., 1986. Tectonic and gravitational

stresses in long symmetric ridges and valleys. Journal of

Geophysical Research 91, 3677–3685.

Soeters, R., Rengers, N., 1983. Dos ejemplos de hundimientos

gravitacionales en los alrededores de Caldes de Bohi (Pirineos

Leridanos). In: Rıos, J.M. (Ed.), Libro Jubilar, vol. 3. IGME,

Madrid, pp. 193–202.

Soldati, M., Corsini, A., Pasuto, A., 2004. Landslides and climate

change in the Italian Dolomites since Late glacial. Catena 55,

141–161.

Solonenko, V.P., 1977. Landslides and collapses in seismic zones

and their prediction. Bulletin of the International Association of

Engineering Geology 15, 4–8.

Susagna, T., Roca, A., Goula, X., Batllo, J., 1994. Analysis of

macroseismic and instrumental data for the study of the 19

November 1923 earthquake in the Aran Valley (Central

Pyrenees). Natural Hazards 10, 7–17.

Tabor, R.W., 1971. Origin of the ridge-top depressions by large-

scale creep in the Olympic Mountains, Washington. Geological

Society of America Bulletin 82, 1811–1822.

Ter-Stepanian, G., 1966. Types of depth creep on slopes in rock

masses. First International Congress of the International Society

of Rock Mechanics, vol. 2, pp. 157–160.

Tibaldi, A., Rovida, A., Corazzato, C., 2004. A giant deep-seated slope

deformation in the Italian Alps studied by paleoseismological and

morphometric techniques. Geomorphology 58, 27–47.

Varnes, D.J., Radbruch-Hall, D.H., Savage, W.Z., 1989. Topo-

graphic and structural conditions in areas of gravitational

spreading of ridges in the western United States. U.S. Geo-

logical Survey Professional Paper 1496 (28 pp.).

Varnes, D.J., Radbruch-Hall, D.H., Varnes, K.L., 1990. Measure-

ment of ridge spreading movements (sackungen) at Bald Eagle

Mountain, Lake County, Colorado, 1975–1989. U.S. Geological

Survey, Open File Report 90-543 (13 pp.).

Varnes, D.J., Coe, J.A., Godt, J.W., Savage, W.Z., Savage, J.E.,

2000. Measurement or ridge-spreading movements (sackungen)

at Bald Eagle Mountain, Lake County, Colorado, II: continu-

ation of the 1975–1989 measurements using a Global Position-

ing System in 1997 and 1999. U.S. Geological Survey, Open file

report 00-205, on-line edition.

Ward, S., 2003. Sackung. In: Goudie, A.S. (Ed.), Encyclopedia of

Geomorphology, vol. 2. Routledge, London, pp. 890–892.

Zischinsky, U., 1966. On the deformation of high slopes. First

Congress of the International Society of Rock Mechanics.

Proceedings vol. 2, pp. 179–185.

Zischinsky,U., 1969.Uber sackungen. Rock Mechanics 1 (1), 30–52.