Navigation – Plan du site

A paraglacial rock-slope failure origin for cirques: a case study from Northern Iceland

Des mouvements de masse paraglaciaires à l’origine des cirques : une étude de cas au nord de l'Islande
Julien Coquin, Denis Mercier, Olivier Bourgeois et Armelle Decaulne
p. 117-136

Résumés

La contribution des mouvements de masse rocheuse paraglaciaires au creusement de cirques dans des paysages glaciaires est explorée dans la montagne de Tindastóll, dans la péninsule de Skagi, dans le nord de l'Islande. Nous analysons 8 cirques qui se sont développés pendant le Pléistocène et 13 cavités de mouvements de masse rocheuse paraglaciaires qui se sont développées pendant l'Holocène dans ce paléo-plateau. À partir d'une reconstruction de la surface pré-quaternaire du plateau, nous calculons les volumes excavés des cirques et des cavités des mouvements de masse rocheuse et quantifions leur contribution au cours de l'Holocène à l’élargissement du cirque. En extrapolant cette contribution à l'ensemble du Quaternaire, nous constatons que les mouvements de masse rocheuse paraglaciaires sont des contributeurs de premier ordre au développement des cirques. Cette contribution est double : premièrement, les mouvements de masse rocheuse paraglaciaires créent des cavités le long des versants de la vallée, dans lesquels des glaciers de cirque peuvent se développer lors de glaciations ultérieures ; deuxièmement, les mouvements de masse rocheuse paraglaciaires le long des parois de cirque préexistants favorisent l’élargissement et l’approfondissement du cirque lui-même. Nos résultats révèlent également que les taux d’érosion glaciaire / paraglaciaire du Quaternaire varient de 0,02 à 0,17 mm par an dans les cirques étudiés. Nous en déduisons que les glaciers sont (i) des facteurs préparatoires efficaces pour la déstabilisation des pentes par les mouvements de masse rocheuse paraglaciaires, (ii) des convoyeurs efficaces pour évacuer les dépôts produits par les mouvements de masse rocheuse paraglaciaire dans les cirques, mais (iii) des agents non nécessairement prédominants dans l'excavation des planchers du cirque.

Haut de page

Notes de la rédaction

Article soumis le 26 février 2019, reçu sous sa forme révisée le 17 juin 2019 et définitivement accepté le 26 juin 2019.

Texte intégral

We thank Susan Conway and the NERC Airborne Research and Survey Facility for obtaining the air photography and LiDAR data on which part of this paper relies. The LiDAR data and aerial photography were collected by the UK's NERC ARSF (Natural Environment Research Council) using financial support from the EUFAR (European Facility for Airborne Research) project “ICELAND_DEBRISFLOWS”, in collaboration with Susan Conway (Open University, UK). The authors are grateful to Helgi Páll Jónsson for his logistic assistance and valuable help in the field. They also thank Carol Robins for editing the English text. Funding was received from OSUNA (CNRS UMS 3281), supporting the project “Icelandic and Martian paraglacial mass movements: comparative analyses”. Financial support was also provided by the Institut Universitaire de France through the program “Arctic facing climate change: analyses of geomorphological paraglacial crises” and by the CNRS-GDR 3062 “Polar mutations”. The authors also thank Samuel McColl, Olav Slaymaker and Takashi Oguchi for their thoughtful reviews and constructive comments that improved the manuscript.

1. Introduction

1Cirques are erosional flat-floored depressions surrounded by steep walls that are ubiquitous in alpine glacial landscapes (Evans, 1974). The role of glaciers on the shaping of these landscapes has been recognized long ago (Helland, 1877; Penck and Brückner, 1904), but the influence of glaciers on cirque formation is still a controversial issue. Most glacial models of cirque development assume a dominance of plucking and abrasion (Galibert, 1962; White, 1970), while periglacial models give primary importance to frost weathering along rockwalls (Johnson, 1904; Gardner 1987; Hooke, 1991; Sanders et al., 2012) and to nivation (Matthes, 1900; Lewis, 1939; Embleton and King, 1975). Both cases are based on the assumption that cirques form strictly during glaciations and imply that their shapes and sizes could somehow reflect the intensity and duration of these glaciations. Accordingly, numerous studies have used the geometry of cirques as a proxy for reconstructions of Pleistocene equilibrium line altitudes or paleo-climates (Olyphant, 1977; Meierding, 1982; Porter, 2000; Brook et al., 2006; Delmas et al., 2014; Principato and Lee, 2015; Crest et al., 2017).

2Many uncertainties remain however in this theory, which postulates rather than demonstrates the glacial origin of cirques. The most significant problem with the assumption that plucking and abrasion are the dominant processes in the development of cirques may be the rates of the processes involved. Erosion rates in excess of 1 mm.yr¹ have been reported in glacial landscapes (Delmas et al., 2009), but these may include erosion by periglacial, paraglacial and interglacial processes because they derive from measurements of sediment volumes (Glasser and Hall, 1997) or meltwater sediment fluxes (Hallet et al., 1996) integrated over whole glacial catchments and/or over long time intervals. Based on a review of erosion rates strictly related to glacial processes, Benn and Evans (1998) found that the rates of cirque deepening and enlargement are comprised between 0.008 and 0.8 mm.yr¹. Turnbull and Davies (2006) argued that these rates are generally too low to explain the sizes and shapes of cirques, given the cumulated duration of the Quaternary glaciations. To overcome this issue, they proposed that cirques could originate from topographic depressions formed by deep-seated gravitational mass-movements.

3Here we test the hypothesis that the misfit between the time required by glacial erosion to carve out cirques and the cumulated duration of the Quaternary glaciations can be solved by integrating the contribution of mass-movements in their development. We focus on the role of paraglacial rock slope failure (RSF), a specific kind of mass-movement that affects slopes and ridges adjacent to receding glaciers (Bovis, 1990; Ballantyne, 2002; Ballantyne, 2013; Ballantyne et al., 2014; Jarman, 2006; Hewitt, 2009; Kellerer-Pirklbauer et al., 2010; McColl, 2012). RSF includes slow deep-seated gravitational slope deformation (DSGSD) (Savage and Varnes, 1987; Chigira, 1992; Ambrosi and Crosta, 2006; Agliardi et al., 2009; Bachmann et al., 2009; Mège and Bourgeois, 2011; McColl, 2012; Crosta et al., 2013; Jaboyedoff et al., 2013; Coquin et al., 2014) and rapid landslides and rock-avalanches (Ballantyne and Stone, 2004; Ballantyne et al., 1998; Jarman, 2006; Ostermann et al., 2012; Mercier et al., 2013). RSF can deeply modify the geometry of deglaciated slopes (Arsenault and Meigs, 2005; Jarman, 2009; Ballantyne, 2013) and produce cavities in which glaciers can subsequently develop (Turnbull and Davies, 2006). To assess the role of paraglacial RSF on the excavation of cirques, we developed a methodology based on: (i) reconstruction of pre-incision surfaces in Quaternary glacial landscapes, (ii) identification and mapping of cirques and RSF cavities, (iii) measurement of excavated volumes of cirques and RSF cavities, and (iv) calculation of the corresponding erosion rates.

4The coastal mountains of Northern Iceland, with their widespread plateaus dissected by numerous cirques, provide an ideal location to apply this methodology. First, plateau relicts are sufficiently well preserved to allow reconstructions of pre-incision surfaces and calculations of excavated volumes. Second, paraglacial RSFs are widespread and have been related to the shrinking of the Weichselian ice-sheet at the onset of the Holocene (Jónsson, 1957; Pétursson and Sæmundsson, 2008; Mercier et al., 2013 and 2017; Cossart et al., 2014; Feuillet et al., 2014; Coquin et al., 2014 and 2016; Decaulne et al., 2016), Third, the relatively homogeneous lithology and the simple tectonic structure provide an excellent opportunity to isolate glacial and paraglacial controls on cirque development.

2. Study area

5The study area is part of a coastal mountain range in the Skagi Peninsula, Northern Iceland, which is bound by the Skagafjörður fjord in the east and the Húnaflói bay in the west. The Skagi peninsula strikes NNW–SSE and extends over 55 km. This study examines the Tindastóll Mountain along the easternmost part of the Skagi Peninsula. The ridge extends over 16 km and strikes NNW–SSE between Skagafjörður in the east and the glacial valley Laxárdalur in the west. The ridge culminates at 1,000 m a.s.l. in the southern part and its elevation decreases to about 600 m a.s.l. in the northern part. The ridge is a plateau with a stair-cased aspect, dissected by cirques alternately located on its east-facing and west-facing slopes (fig. 1).

Fig. 1 – Study area.
Fig. 1 – Zone d’étude.

Fig. 1 – Study area.   Fig. 1 – Zone d’étude.

On the western side of Skagafjörður (A), within Tindastóll Mountain (B); Three sites are studied in details (C-E). Panoramic view of the east-facing slope of the Tindastóll Mountain together with an interpretative sketch of the main geomorphic features (F); 1. Rock slope failure scarp; 2. Rock slope failure deposit; 3. Detritic deposit.
Sur la rive ouest de Skagafjörður (A), sur la crête de Tindastóll (B) ; trois sites sont étudiés en détail (C-E). Vue panoramique de Tindastóll et schéma interprétatif des principales caractéristiques géomorphologiques (F) ; 1. Cicatrice d’arrachement des mouvements de masses rocheuses paraglaciaires ; 2. Dépôts des mouvements de masse rocheuse ; 3. Dépôts détritiques.

6The bedrock of the Tindastóll Mountain belongs to an upper Tertiary sequence of basic, intermediate and acid volcanic products with thin interbedded sedimentary layers, and corresponds to the outer margin of a dismantled rhyolitic central volcano (Sæmundsson, 1979; Långbacka and Gudmundsson, 1995; Kristjansson et al., 2004; Decaulne et al., 2007). Along the eastern slope of the ridge, the lava pile is affected by an assemblage of dykes and faults that developed mainly 2–7 Ma ago, when oceanic rifting was active in Skagafjörður (Garcia et al., 2003; Bourgeois et al., 2005).

7In Northern Iceland, glacial erosional and depositional features are widespread over most valley and cirque floors but are not observed on the intervening plateaus (Bourgeois et al., 2000). The sharp rim separating cirques from plateaus also suggests that glacial erosion left the plateaus barely touched. Based on these observations, we follow the classical assumption that, during the Quaternary glaciations, the valleys and cirques of Northern Iceland were occupied by glaciers while the intervening plateaus were either ice free or occupied by low erosive cold-based ice domes. Therefore, we assume that the present-day topographic surface of the plateaus is inherited from pre-Quaternary times and provides a convenient benchmark to estimate the rates of cirque shaping processes during the Quaternary.

3. Methodology

3.1. Mapping cirques and RSF cavities

8The cirques and RSF cavities of the Tindastóll Mountain were mapped from high-resolution color aerial orthophotographs from Loftmyndir ehf. available online, from a 1 m resolution DEM generated from airborne laser altimeter (LiDAR) measurements, and from field investigations. As the northernmost part of the study area was not covered by the LiDAR survey, the 30 m resolution ASTER DEM (METI and NASA) was used in this area. Topographic maps of Iceland produced by Mál og Menning were also used to complement the dataset.

9Thanks to the high resolution of the dataset and the excellent state of freshness of the cirques and RSF cavities, an accurate delimitation of their contours was possible. Interpretative aspects were nonetheless part of the mapping process:

• The cirques were mapped following the most commonly used topographic criteria based on the three break-in-slopes generally observed along a longitudinal cirque profile: the upper cirque edge, the cirque headwall footslope, and the shoulder step connecting the hanging cirque to the downward valley (Evans and Cox, 1974, 1995). As the cirques are located on the plateau edges, their upper proximal parts were easily recognizable and were drawn with accuracy. Difficulties arose mostly at distal locations where the topographic break-in-slope connecting a hanging cirque floor to the adjacent valley side was unclear. If the bedrock step was partly dismantled by gullies connecting the hanging cirque floor to the adjacent valley floor, the drawing of the cirque followed the upper rim of these gullies. In the case of a valley shoulder buried under sediments smoothing the break-in-slope, the cirque was delimited by the closest contour line from the topographic map of Iceland.

• The RSF cavities were mapped mainly from the DEMs. RSF cavities look similar to cirques, with a generally steep break-in-slope along the upper rim. However, the accurate determination of RSF contours was difficult where the basal parts of their cavities remained buried by their deposits.

3.2. Morphometric variables

10Morphometric variables for cirques and RSFs are defined after Evans and Cox (1974) and Delmas et al. (2014). The length l is measured along the bisector, a straight line that bisects the cirque from its proximal headwall to its distal shoulder step. The width w is the maximal distance between the opposite sidewalls, measured along a straight line orthogonal to the bisector. The amplitude a for a cirque is the differential elevation between the cirque headwall at the intersection point with the bisector and the lowest point on the cirque floor, while the amplitude for a RSF cavity corresponds to the greatest differential elevation between the headscarp rim and the cavity floor.

3.3. Reconstruction of pre-incision surface and calculation of excavated volumes

11To compute the excavated volumes of cirques and RSF cavities, we first reconstructed the pre-Quaternary surface of the paleo-plateau. This reconstruction relies on several assumptions: (i) erosion rates on the plateaus were low during glaciations, (ii) present day plateaus between cirques are preserved relicts of the pre-Quaternary surface, (iii) the “missing” surface above cirques and RSF cavities is assumed to have been smooth and aligned on the present-day elevation of the surrounding plateaus, and (iv) erosion processes involved in the “missing” volumes of cirques were active during the Quaternary only.

12The first step of the reconstruction (fig. 2) consisted in deriving elevation contours from the present-day DEMs, with respect to their spatial resolution (with 10-m intervals for the LiDAR DEM and 20-m intervals for the ASTER DEM). Sections of elevation contours coinciding with cirques and RSF cavities were then replaced by artificial contours using white dashed lines (fig. 2). These artificial contours were drawn so, that their curvatures were continuous with those of adjacent contours using black lines (fig. 2) to preserve the general aspect of the contour lines (fig. 3B).

Fig. 2 – Paleo-isolines are reconstructed on the basis of 10-m isoline contours derived from LiDAR and ASTER DEMs., of surfaces dissected by cirques (A), and by RSF cavities (B).
Fig. 2 – Reconstruction des paléo-isohypses sur la base des isohypses à équidistance de 10 m dérivées des données LiDAR et MNT ASTE, des surfaces disséquées par les cirques (A) et des mouvements de masse rocheuse (B).

Fig. 2 – Paleo-isolines are reconstructed on the basis of 10-m isoline contours derived from LiDAR and ASTER DEMs., of surfaces dissected by cirques (A), and by RSF cavities (B).  Fig. 2 – Reconstruction des paléo-isohypses sur la base des isohypses à équidistance de 10 m dérivées des données LiDAR et MNT ASTE, des surfaces disséquées par les cirques (A) et des mouvements de masse rocheuse (B).

13As we will discuss in section 4.2, the present-day stair-cased topography of the plateau is related to the presence of faults that were active during the Quaternary. We restored the vertical offset of the topographic surface across these faults, with the same methods as for the cirques.

14Paleo-DEMs corresponding to the reconstructed pre-erosion surface were then generated with a cell size of 5 m (fig. 3C), using the Topo to Raster tool of ArcGIS™, designed for local interpolations based on contour lines (Hutchinson, 1993). A grid of elevation changes was produced by subtracting the present-day DEM (resampled at a cell size of 5 m) from the paleo-DEM, using the Raster Calculator tool of ArcGIS™ (fig. 3D). Finally, zonal statistics and excavated volumes were computed from the grid of elevation changes, for each cirque and RSF cavity (tab. 1-2).

Fig. 3 – Estimation of eroded volumes from cirques based on the reconstruction of the paleo-plateau.
Fig. 3 – Estimation des volumes érodés des cirques fondée sur la reconstruction du paléo-plateau.

Fig. 3 – Estimation of eroded volumes from cirques based on the reconstruction of the paleo-plateau.   Fig. 3 – Estimation des volumes érodés des cirques fondée sur la reconstruction du paléo-plateau.

Each cirque is delimited on the present-day LiDAR-DEM (A). Paleo-isolines corresponding to the pre-incision of glacial cirques are drawn (B). The paleo-DEM is derived from reconstructed isolines with a cell resolution of 5x5 m (C). Volume of eroded material is then calculated for each cell by subtracting cell values of the present-day LiDAR DEM from the reconstructed paleo-DEM (D).
Chaque cirque est délimité selon le MNT ASTER actuel (A). paléo-isohypses correspondant aux cirques glaciaires avant l’incision (B). Le paléo-MNT est dérivé de la reconstruction des isohypses avec une résolution de cellule 5 x 5 m (C). Le volume de matériel érodé est ensuite calculé pour chaque cellule en soustrayant les valeurs des cellules de l’actuel MNT LiDAR de celles du paléo-MNT reconstruit (D).

Tab. 1 – Morphometric criteria and volume of cirques.
Tab. 1 – Critères morphométriques et volume des cirques.

Tab. 1 – Morphometric criteria and volume of cirques.   Tab. 1 – Critères morphométriques et volume des cirques.

Tab. 2 – Morphometric criteria and volumes of paraglacial rock-slope failure (RSF).
Tab. 2 – Critères morphométriques et volumes des mouvements de masse paraglaciaires.

Tab. 2 – Morphometric criteria and volumes of paraglacial rock-slope failure (RSF).   Tab. 2 – Critères morphométriques et volumes des mouvements de masse paraglaciaires.

S1. RSF source area; S2. RSF deposit area; S3. RSF entire body; a. Volume calculation was performed on the deposit area of RSF 4 by using a density correction assuming that RSF deposit is 63% of the bedrock density; b. In spite of its great size, the volume of displaced material was not calculated for RSF 1 given the absence of high-resolution DEM in this location. * The values used to calculate the volume of RSFs.
S1. Zone source RSF; S2Zone de dépôt RSF; S3. RSF tout l'ensemble; a. Un calcul de volume a été effectué sur la zone de gisement de RSF 4 en utilisant une correction de densité supposant que le gisement RSF représente 63 % de la densité du substrat rocheux; b. Malgré sa grande taille, le volume de matériau déplacé n’a pas été calculé pour RSF 1 en raison de l’absence de MNT haute résolution à cet endroit. * Valeurs utilisées pour calculer le volume de RSF.

15The volume, V, of a cirque or RSF cavity is taken as:

            

V = 

⎡⎲i = n

⎣⎳i = 1

(zinitial,i – zfinal,i)

w²                  [1]

16where zinitial,i is the elevation value of cell i on the paleo-DEM, zfinal,i is the elevation value of the same cell on the present-day DEM, n is the number of cells in the cirque or cavity and w is the cell size (5 m).

3.4. Estimation of Quaternary erosion rates from excavated volumes of cirques

17We postulate that the geometry of cirques results from a combination of glacial and paraglacial denudation processes, which alternated during the succession of Quaternary glaciations and deglaciations. Based on this assumption, the sum of glacial and paraglacial Quaternary denudation rates, E1, can be derived from the excavated volume of a cirque, if the cumulated duration of Quaternary glacial and paraglacial intervals is known:

            

E1 = V/At                                  

                [2]

18where V and A are the volume and area of the cirque and t the cumulated duration of Quaternary glacial and paraglacial intervals.

19The first glaciations in Iceland occurred about 5 to 3 Ma ago but probably consisted of minor and local ice-fields as their deposits have only been identified over restricted highland areas (Geirsdóttir et al., 2007). Based on the 2.5 Ma age estimated for the oldest glacial deposits in the Tjörnes peninsula (Geirsdóttir and Eiríksson, 1994), we assume an upper boundary of 2.5 Ma for the onset of cirque formation in the Tindastóll Mountain Since then, around 18 major glaciations have been recorded in Iceland (Geirsdóttir et al., 2007) and the Tjörnes stratigraphy indicates that the duration of the glaciations increased after 1 Ma, with the development of 100 ka glacial sequences interrupted by 10 to 30 ka interstadial sequences (Eiríksson, 1985; Raymo et al., 1989; Geirsdóttir and Eiríksson, 1994, 1996; Paillard, 2015). It is therefore reasonable to assume that interstadial conditions predominated between 2.5 and 1 Ma while glacial and paraglacial conditions dominated after 1 Ma. On the whole, glacial and paraglacial conditions may thus have prevailed during approximately half of the 2.5 Ma Pleistocene sequence, which gives t = 1.25 Ma in Equation 2.

3.5. Assessing the respective contributions of paraglacial RSF and glacial erosion to cirque development

20We define the contribution of Holocene paraglacial RSF to the present-day geometry of a given cirque as the ratio, C, of the cumulated volume of RSF cavities included in the perimeter of this cirque to the total volume of the cirque. Then we extrapolate the value of C to obtain the value of P, the contribution of RSF to cirque development over the full duration of the Quaternary:

            

P = Cn                                  

                  [3]

21Where the value of n is 18, the number of glaciation and deglaciation cycles that presumably occurred during the Quaternary (Geirsdóttir et al., 2007). This extrapolation is based on the assumption that the contribution of RSF to the development of cirques is the same at each paraglacial sequence, which is obviously reductive but is convenient for first-order estimations.

22The contribution, G, of glacial erosion to the development of cirques can then be computed from the value of P:

            

G = 1 - P                                  

                 [4]

23In addition, the glacial erosion rate in cirques, E2, is given by:

            

E2 = GV/At                                  

             [5]

24Finally, the time, T, required to carve out a cirque by glacial erosion alone can be estimated by:

            

T = V/AE2                                  

            [6]

4. Results

4.1. A Plateau dissected by cirques

25In planform, the cirques of the Tindastóll Mountain exhibit a variety of shapes and sizes (tab. 1), from simple and small ones (A < 1.2.10 m³ – cirques 2, 3 and 8) to large ones (A > 2.5·10 m³ – cirques 1, 4, 5, 6 and 7) (fig. 4). This variety is also reflected by their l and w values, which range from 910 to 1,910 m and 790 to 2,340 m, respectively (tab. 1). The cirques are shallow with respect to their length and width (180 ≤ a ≤ 600 m, 2.3 ≤ l/a ≤ 6.7 m and 3.1 ≤ w/a ≤ 5.5 m). This suggests that the erosional processes involved in both lateral and backwall retreat were more efficient than those involved in downwearing.

26The series of transverse topographic profiles on Figure 4 illustrates that the pre-Quaternary surface has been unequally preserved along the Tindastóll Mountain. It has been well preserved in the vicinity of cirques 3 (profile B–B’), 2 (profile C–C’) and 8 (profile H–H’), which have the smallest volumes (tab. 1). By contrast, it has been deeply incised in the vicinity of cirques 4 (profile D–D’) (fig. 4), 5 (profile F–F’) and 7 (profile G–G’), which have larger volumes (tab. 1).

Fig. 4 – Present-day transversal profiles (in dark gray) from the ASTER and LiDAR DEMs together with pre-incision profiles (in light gray) obtained from reconstructed paleo-DEMs.
Fig. 4 – profils transverses actuels (gris foncé) extraits des MNT ASTER et LiDAR profilant également les surfaces préalables à l’incision (gris clair), obtenus selon la reconstruction des paléo-MNT.

Fig. 4 – Present-day transversal profiles (in dark gray) from the ASTER and LiDAR DEMs together with pre-incision profiles (in light gray) obtained from reconstructed paleo-DEMs.   Fig. 4 – profils transverses actuels (gris foncé) extraits des MNT ASTER et LiDAR profilant également les surfaces préalables à l’incision (gris clair), obtenus selon la reconstruction des paléo-MNT.

4.2. A Plateau affected by gravitational spreading

4.2.1. Gravitationally-driven reactivation of faults

27The topographic surface of the Tindastóll Mountain is divided into a stair-cased assemblage of flat surfaces, separated by fault scarps parallel to the ridge axis. These brittle deformation features are outlined on the topographic profiles of Figure 4 (profiles D–D’, E–E’ and H–H’). The fault scarps are partly traceable on the plateau, on either side of cirque walls, which suggests that the faults developed before the cirques (Fig. 5, 6B-C, 7C-D).

Fig. 5 – Overall physiography together with a geomorphological map of the Tindastóll Mountain (A) with location of pictures taken during field investigations (fig. 6) and feature references for which volumes have been estimated (Tables 1 and 2). (B) Geomorphological map of the Tindastóll Mountain showing the main glacial-paraglacial related features.
Fig. 5 – Physiographie générale de la zone d’étude (A) accompagnée de la localisation des photographies de terrain (fig. 6) et carte géomorphologique de la crête de Tindastóll (B) montrant les formes glaciaires et paraglaciaires liées.

Fig. 5 – Overall physiography together with a geomorphological map of the Tindastóll Mountain (A) with location of pictures taken during field investigations (fig. 6) and feature references for which volumes have been estimated (Tables 1 and 2). (B) Geomorphological map of the Tindastóll Mountain showing the main glacial-paraglacial related features.   Fig. 5 – Physiographie générale de la zone d’étude (A) accompagnée de la localisation des photographies de terrain (fig. 6) et carte géomorphologique de la crête de Tindastóll (B) montrant les formes glaciaires et paraglaciaires liées.

1. Fault; 2. Dyke; 3. Landslide scarp without deposit; 4. Landslide scarp with deposit; 5. Detritic material; 6. Cirque.
1. Faille ; 2. Dyke ; 3. Escarpement d’arrachement sans dépôt ; 4. Escarpement d’arrachement avec dépôt ; 5. Matériel détritique ; 6. Cirque.

28Similar assemblages of ridge-parallel faults have been described in many regions of the World that were formerly glaciated, including the Skagafjördur area. They are generally attributed to ridge collapse by paraglacial gravitational spreading (Beck, 1968; Bovis, 1982; Ballantyne, 1986, 1991, 1992, 1997, 2002, 2008; Savage and Varnes, 1987; Chigira, 1992; Jarman, 2006; Mege and Bourgeois, 2011; McColl, 2012; Ballantyne and Stone, 2013; Crosta et al., 2013; Jaboyedoff et al., 2013; Coquin et al., 2015, 2016). We suspect that the faults of the Tindastóll Mountain have a similar origin. The orientation of these faults however coincides with that of an assemblage of dykes and faults that were active mainly 2–7 Ma ago, when oceanic rifting was active in Skagafjörður (Långbacka and Gudmundsson, 1995; Bourgeois et al., 2005; Johannesson and Sæmundsson, 2009). This suggests that paraglacial gravitational spreading of the Tindastóll Mountain may be controlled by a pre-existent tectonic assemblage. However, the vertical displacements expressed in the present-day plateau topography remain hard to reconcile with a purely tectonic origin as the tectonic activity of the Skagafjörður paleo-rift ceased 3 Ma ago (Garcia et al., 2003) before the onset of Quaternary glaciations (Geirsdóttir et al., 2007). The deformations on the Tindastóll Mountain are consistent with recent and local reactivations of a fault system inherited from rifting before 3 Ma. Such local reactivations can be triggered by gravitational spreading induced by deglaciation. The inherited rifting fault assemblage is thus a major predisposing factor for later gravitationally-driven paraglacial deformation.

4.2.2. Gravitational spreading signature in cirques

29The ridge deformation related to gravitational spreading has also influenced the shaping of cirques. This can be illustrated in the case of cirque 1, which displays an N–S orientation similar to that of the previously described gravitational deformation: cirque 1 first developed into an initial crestal graben related to gravitational spreading (fig. 8B-C). Close links between cirque shape and inherited crestal graben are also observed in cirques 4 and 7. While these display a main W–E orientation, second generation of cirques with an N–S orientation have developed in close association with the N–S fault scarp system: the paraglacial ridge deformation guided glacial erosion along weaknesses during subsequent glaciations (fig. 5, 7C).

Fig. 6 – Geomorphological features along the Tindastóll Mountain and their explanatory sketches.
Fig. 6 – Éléments géomorphologiques le long de la chaîne Tindastóll et ses schémas explicatifs.

Fig. 6 – Geomorphological features along the Tindastóll Mountain and their explanatory sketches.   Fig. 6 – Éléments géomorphologiques le long de la chaîne Tindastóll et ses schémas explicatifs.

A: Snout of cirque 1 toward SW. Vertical displacements and RSFs on the slope, with a significant downward displacement of block 2. B: N-facing sidewall of cirque 6, looking S. Normal faulting striking north, underlined by a stepped topographic surface. C: The same cirque wall viewed from an upper location on the opposite sidewall, looking SW. The backwall receding triggered by a rotational landslide (RSF-6) gives an arcuate aspect to the S part of the cirque. D: Cirque 3 viewed toward S. The post-glacial translational mass-movement (RSF-4) alters the linear aspect of the cirque headwall. E: The W-facing slope at the snout of cirque 1, viewed from the opposite sidewall. Two RSF cavities alter the linear aspect of the cirque wall. The absence of deposit at the base of the left cavity suggests either that the RSF occurred during a previous deglaciation sequence when the cirque was still occupied by a glacier, which evacuated the supraglacial RSF deposit, or during a previous interstadial stage, the deposit being then removed by subsequent glacier readvances. F: The N-facing wall of cirque 5 shows a linear headscarp and the scarp of an old RSF event that occurred before re-glaciation, which has partly eroded the RSF lateral scarp and has removed the RSF deposits.
A : Extrémité du cirque 1 vers le SO. Déplacements verticaux et dépôts de mouvements de masse sur le versant, avec un mouvement descendant du bloc 2. B : La paroi exposée au nord du cirque 6, vue vers le sud, montre une faille normale à rejet vers le nord, soulignée par une marche topographique. C : La même paroi de cirque vue de plus haut, vers le SO ; le retrait de la paroi, dû à un glissement de terrain rotationnel, procure un aspect arqué à la paroi. D : Cirque 3 vu vers le S ; le mouvement de masse translationnel post-glaciaire (RSF 4) altère la linéarité de la paroi du cirque. E : Versant exposé vers l’O à l’extrémité du cirque 1 ; deux cavités de mouvements de masse altèrent la linéarité de la paroi ; l’absence de dépôt à la base de la cicatrice d’arrachement à gauche suggère soit que le glissement de terrain a eu lieu lors d’une séquence de déglaciation antérieure lorsque le cirque était encore englacé, qui a alors évacué le matériel du glissement, soit durant une phase interglaciaire antérieure, le matériel étant évacué lors de la période glaciaire subséquente. F : La paroi exposée au N du cirque 5 montre une nette cicatrice d’arrachement sans dépôt, celui-ci s’étant produit avant la dernière glaciation puisque celle-ci a en partie érodé l’escarpement latéral et a évacué le matériel.

Fig. 6 – (continuing)
Fig. 6 – (suite)

Fig. 6 – (continuing)   Fig. 6 – (suite)

G: Dyke lineament incorporated in the basaltic trap observed along the main E-facing valley slope, where dipping between dyke and lava pile are contrastive. H: The S-facing wall of cirque 4, where the cirque incision reveals sheared rock material within a fault zone area. I-J: Succession of stepped flat surfaces separated by a scarp perpendicular to the main slope.
G : Dyke dans l’empilement basaltique le long de la face est de la chaîne de Tindastóll. H : Parois sud du cirque 4 montrant la roche cisaillée et fracturée dans une zone de faille. I-J : Successions de surfaces horizontales séparées par des escarpements perpendiculaires au versant principal.

Fig. 7 – Simplified sketch of the Tindastóll relief evolution over successive glacial-paraglacial stages (in planform).
Fig. 7 – Schéma simplifié de l’évolution du relief sur la chaîne Tindastóll après plusieurs cycles glaciaires et paraglaciaires (en plan).

Fig. 7 – Simplified sketch of the Tindastóll relief evolution over successive glacial-paraglacial stages (in planform).   Fig. 7 – Schéma simplifié de l’évolution du relief sur la chaîne Tindastóll après plusieurs cycles glaciaires et paraglaciaires (en plan).

A: Paleo-plateau before the onset of the Quaternary glaciations. B: First incision of the plateau edge by large-scale RSFs driven by gravitational spreading rock slope deformations, which affect the basaltic plateau in the context of deglaciation (paraglacial sequence). C: Glacial erosion concentrates along depressions generated by previous RSF events and crestal grabens. Such an initial depression allows expansion of glacial troughs and cirques (glacial sequence). D: Present-day relief of the plateau deeply dissected by large cirques formed by the complex interplay of glacial and paraglacial processes. 1. Fault (certain); 2. Fault (questionable); 3. Fault (inferred).
A : Paléo-plateau avant les glaciations quaternaires. B : Premières incisions en bordure du plateau par des mouvements de masses rocheuses en contexte de déglaciation (séquence paraglaciaire). C : Concentration de l’érosion glaciaire dans les dépressions générées par les premiers mouvements de masse durant le stade interglaciaire précédent (séquence glaciaire). D : Relief actuel du plateau largement incisé par de larges cirques formé par les jeux combinés des processus glaciaires et paraglaciaires. 1. Faille (certain) ; 2. Faille (discutable) ; 3. Faille (déduite).

30Ridge deformations with significant vertical displacement along inherited faults are in some cases associated with highly erosive RSF events. These may favor cirque walls destruction instead of shaping them. For example, when the overall backwall recedes along main slopes, given the highly erosive paraglacial readjustments that outweigh the efficiency of processes involved in the shaping of the cirque, the cirque incision and further development cannot keep pace with the widening of valleys of higher order. Such differences in rates of destructive processes offer an explanation to the absence or destruction of cirques in the northern part of the study area (profiles A–A’ and B–B’) (fig. 4).

4.3. RSF signature in cirques

4.3.1. Post-glacial RSFs in cirques

31Evidence of RSFs is widespread over the study area (fig. 5, 6A, C-E). Fifteen RSF cavities connected to landslide debris were identified, 59‑5567 × 10³ m² in surface area and 0.08‑12.88 × 10 m³ in volume (tab. 2), indicating highly varied RSF sizes. The preservation of the RSFs suggests that the events occurred after complete deglaciation. Six of the post-glacial RSF cavities are located along the cirque walls, affecting the overall excavated volume of cirques.

32Despite high landslide susceptibility in an area affected by DSGSD (Capitani et al., 2013; Coquin et al., 2016), establishing a genetic link between the RSFs in the study area and the more diffuse brittle deformations is difficult. Nonetheless, the spatial distribution of the RSFs shows that their cavities are mostly emplaced along valley walls striking north, i.e. that the failure planes of the RSFs are controlled by the pre-existing tectonic assemblage (Fig. 8-9B-C).

Fig. 8 – 3D sketches showing the role slope destabilization processes in trough/cirque genesis and further development. The present-day footprint of glacial cirques along the Tindastóll Mountain is interpreted as being controlled by initial topographic depressions formed under the action of large-scale paraglacial ridges and slope destabilization processes (DSGSDs and RSFs).
Fig. 8 – Schémas 3D montrant le rôle des déstabilisations de versant dans la genèse des cirques et leur développement. L’empreinte actuelle des cirques glaciaires de Tindastóll est contrôlée par des dépressions topographiques formées sous l’action de crêtes paraglaciaires et des déstabilisations de versant (DSGSD et mouvements de masses rocheuses).

Fig. 8 – 3D sketches showing the role slope destabilization processes in trough/cirque genesis and further development. The present-day footprint of glacial cirques along the Tindastóll Mountain is interpreted as being controlled by initial topographic depressions formed under the action of large-scale paraglacial ridges and slope destabilization processes (DSGSDs and RSFs).   Fig. 8 – Schémas 3D montrant le rôle des déstabilisations de versant dans la genèse des cirques et leur développement. L’empreinte actuelle des cirques glaciaires de Tindastóll est contrôlée par des dépressions topographiques formées sous l’action de crêtes paraglaciaires et des déstabilisations de versant (DSGSD et mouvements de masses rocheuses).

A: Full-glacial condition. B: Gravitational spreading process coeval with deglaciation (first step of paraglacial readjustment sequence). C: Cirque initiation assisted by deep-seated RSF. D: Cirque development assisted by glacial denudation processes. E: Cirque development assisted by paraglacial RSFs; the activation of second-order rock-slope failures in the context of deglaciation favours the development of cirques. F: Source-area depression produced by 2nd-order rock-slope failures together with ridge-top depression areas initiated by gravitational spreading of the mountain ridge promote snow and ice accumulation, assisting cirque development under the action of glacial erosion processes. G: Large and deep cirques are produced by the gravitational spreading and rock slope failures initiated under paraglacial conditions together with the action of glacial erosion processes under glacial conditions.
A : Pléniglaciaire. B : Processus d’étalements gravitationnels contemporains de la glaciation. C : Initiation des cirques assistés par des mouvements de masse rocheuse profonds. D : Développement des cirques assisté des processus de dénudation glaciaire. E : Développement des cirques assistés de mouvements de masse rocheuse paraglaciaires ; l’activation de mouvements de masse secondaires favorise le développement des cirques. F : Les dépressions des zones sources des glissements de masse de second ordre et les dépressions au long des crêtes sommitales liées aux étalements gravitaires favorisent les accumulations de neige et de glace, assistant la formation du cirque sous l’action des processus d’érosion glaciaire. G : Cirques larges et profonds initiés durant les périodes paraglaciaires associés à l’action des processus d’érosion glaciaire durant les périodes glaciaires.

4.3.2. Evidence of older RSF events in cirques

33Several depressions similar to RSF cavities do not display deposits in cirques. They are more likely related to RSF events, and the absence of deposits is ascribable to removal by subsequent glacier readvances. Such an old RSF activities are documented along the S-facing slope of cirque 1 (fig. 5-6E) and the N-facing slopes of cirques 4 and 5 (fig. 5-6F). In cirques 4 and 5, the irregular and arcuate shape of the rim together with the steep slope gradient their paraglacial origin, even if it’s been partially reshaped during subsequent glacier advances. In cirque 1, the scarp is well preserved and only the RSF deposit is lacking, i.e. the RSF event probably occurred while the cirque floor was still occupied by a glacier during the final stage of deglaciation.

4.4. Reconstructed cirque genesis and development under the action of paraglacial instabilities

34Slope and ridge instabilities reach their maximal intensity during the recession of glaciers. The stress redistribution along the debuttressed valley slopes favors the opening of new tension cracks and joints. The whole ridge may experience large-scale rock mass deformation induced by gravitational spreading, producing crestal grabens, scarps and counterscarps (fig. 8B). At a regional scale, the glacial unloading induced by ice-sheet melting may release an increase in seismic activity due to viscoelastic adjustments of the mantle and subsequent crustal deformations (Jull and McKenzie, 1996). Acting together, stress debuttressing and seismic shaking promote the development of a wide variety of instabilities along deglaciated slopes, from minor rock falls to large RSFs. The gravitationally-driven fault reactivation also acts as a precursor for large scale RSFs along the main valley slope adjacent to Skagafjörður and Laxárdalur (fig. 8C) (Cossart et al., 2017).

35The openings of large cavities by the RSFs provide a good location for glaciers to develop during subsequent glaciation. Once the initial cavity is formed by the RSF, glaciers can shape the cirque. The development of tension cracks, joints and fractures along debuttressed rockwall also facilitate glacial erosion (plucking) during subsequent glaciation, facilitating rapid cirque widening and lengthening (fig. 8D). Once cirques are shaped, a second generation of RSF 2nd-order RSFs can develop along cirque walls during further deglaciation sequences (fig. 8E). Then, the RSF material accumulated on the cirque floor will be eroded and evacuated by subsequent glacier readvances (fig. 8F).

Fig. 9 – RSF contribution ratio for the eight cirques of the Tindastóll Mountain given 1, 18 or 30 paraglacial sequences.
Fig. 9 – proportion de la contribution des mouvements de masse rocheuse pour la formation de 8 cirques de la montagne Tindastóll selon 1, 18 ou 30 séquences paraglaciaires.

Fig. 9 – RSF contribution ratio for the eight cirques of the Tindastóll Mountain given 1, 18 or 30 paraglacial sequences.   Fig. 9 – proportion de la contribution des mouvements de masse rocheuse pour la formation de 8 cirques de la montagne Tindastóll selon 1, 18 ou 30 séquences paraglaciaires.

1. Cirque volume; 2. Paraglacial contribution to cirque volume measured for 1 sequence; 3. Extrapolated for 18 sequences; 4. Extrapolated for 30 sequences.
1. Volume du cirque ; 2. Contribution paraglaciaire au volume du cirque mesurée pour une séquence paraglaciaire ; 3. Extrapolation à 18 séquences paraglaciaires ; 4. Extrapolation à 30 séquences paraglaciaires.

4.5. Measuring the RSF contribution to the shaping of cirques

4.5.1. Unsteady contribution of RSFs to the shaping of cirques

36The inferred influence of paraglacial RSFs on the making of cirques, based on the RSF contribution ratio, highlighted an unsteady effect during the last Weichselian paraglacial sequence. The greatest values were reported in cirques 1 and 3, with RSF cavities explaining 3.5% and 4% of the whole volume of cirques, respectively, while cirques 2, 4 and 5 displayed no contribution (fig. 9). This ratio describes the influence of the last paraglacial sequence compared to the whole volume of cirques. Thus, the seemingly low values point out large disparities between its two components: the first (i.e. the volume of RSF cavity) reflects the action of one local process (the RSF event) occurring during a single paraglacial whereas the volume of the second one (i.e. the whole volume of cirque) involved a complex combination of many processes (mainly glacial, periglacial and paraglacial) operating during a succession of many glacial and paraglacial sequences throughout the Quaternary. Hence, the total contribution of paraglacial RSFs requires the influence of RSF activities during earlier paraglacial sequences to be taken into account. Presumably, a minimum of 18 major glaciation and deglaciation cycles occurred in the last 2.5 Ma (Geirsdóttir, 2007), and the overall RSF contribution to the carving of the cirques ranges between 1% and 71% (fig. 10). The influence of RSFs is negligible for cirques 7 and 8, displaying values of 0.5% and 1.5%, respectively, but predominant for cirques 1 and 3 with values of 64% and 71%, respectively.

Fig. 10 – Potential influence of RSF activity on the shaping of cirques given the number of glaciation-deglaciation cycles. Note that for an RSF activity equivalent to the last paraglacial sequence (in volume of eroded material), around 25 to 30 cycles can explain the whole volume of cirques 1 and 3 and around 25% of the overall volume of Tindastóll cirques.
Fig. 10 – Influence potentielle de l’activité des mouvements de masse rocheuse au façonnement des cirques selon le nombre de cycles de glaciation-déglaciation. À noter que pour une activité de mouvements de masse rocheuse équivalente à celle de la dernière séquence paraglaciaire (en volume de matériel érodé), entre 25 et 30 cycles peuvent expliquer la totalité des volumes des cirques 1 et 3, mais seulement 25 % du volume de l’ensemble des cirques de Tindadtóll.

Fig. 10 – Potential influence of RSF activity on the shaping of cirques given the number of glaciation-deglaciation cycles. Note that for an RSF activity equivalent to the last paraglacial sequence (in volume of eroded material), around 25 to 30 cycles can explain the whole volume of cirques 1 and 3 and around 25% of the overall volume of Tindastóll cirques.   Fig. 10 – Influence potentielle de l’activité des mouvements de masse rocheuse au façonnement des cirques selon le nombre de cycles de glaciation-déglaciation. À noter que pour une activité de mouvements de masse rocheuse équivalente à celle de la dernière séquence paraglaciaire (en volume de matériel érodé), entre 25 et 30 cycles peuvent expliquer la totalité des volumes des cirques 1 et 3, mais seulement 25 % du volume de l’ensemble des cirques de Tindadtóll.

1. Cirque 1; 2. Cirque 3; 3. Cirque 6; 4. Cirque 7; 5. Cirque 8; 6. Total.
1. Cirque 1 ; 2. Cirque 3 ; 3. Cirque 6 ; 4. Cirque 7 ; 5. Cirque 8 ; 6. Total.

4.5.2. An underestimated RSF contribution ratio in cirque shaping

37At the Tindastóll scale, the calculation of the overall RSF contribution ratio in the making of cirques is based on the assertion that the last post-glacial RSF activity is representative of RSF activity of former deglaciation sequences, implying that the RSF contribution ratio was constant throughout glaciation and deglaciation cycles of the Quaternary. An intuitive view is that the volume of RSFs is maximal during the last paraglacial sequence when cirques are fully-developed and provide more potential failure sites than during previous ones. However, in case of cirques 5 and 7, our observations indicate that the RSF activity was null or negligible during the last paraglacial sequence but these cirques experienced older RSF activity during previous paraglacial sequences (fig. 6F). Indeed, RSF activity must have varied from one paraglacial sequence to another. As old RSF cavities and ridge deformations have been partly reshaped by the glacial erosion of subsequent glacier readvances, it is impossible to determine whether the overall excavated volume of these cirques is related to glacial or paraglacial processes. Consequently, the RSF contribution on their shaping is largely underestimated by the RSF contribution ratio. These observations thus illustrate the limits of an approach based on the assumption of a constant rate of RSF activity throughout the Quaternary. Moreover, the paraglacial processes involved on the shaping of cirques are not limited to the action of mass-movements. These paraglacial processes also include the more diffuse backwall retreat in the form of discrete rock-fall events and gravitationally-driven ridge deformation. The foot-slopes of cirque wall are covered by scree deposits which suggest that the paraglacial rock-fall events have exerted a non-negligible influence on the carving of cirques (fig. 6D, 6F, 6K). Our observations also indicate that paraglacial ridge deformation has exerted a significant role in the shaping of cirques 1, 4 and 7. Further investigations on the influence of these ridge deformation or rock-fall activity would be required to embrace the role of all of these paraglacial processes in the carving of cirques.

4.6. Estimation of strictly glacial erosion rates

38The erosion rates estimated for the eight cirques of the Tindastóll Mountain range from 0.08 to 0.17 mm.yr¹ with a mean value of 0.11 mm.yr¹ (tab. 3). As rates correspond to the missing volume of eroded material compared to a pre-incision surface benchmark, these values reflect an overall rate including both paraglacial and glacial processes, which does not provide insights into the individual contribution of each of them.

39Following previous speculation to distinguish the respective parts of glacial and RSF processes from the overall missing volume, it becomes possible to derive the strictly glacial rates associated with the shaping of cirques. In cirques where the RSF activity is different from 0, the strictly glacial rates vary from 0.02 to 0.13 mm.yr¹ (tab. 3). By assuming that both glacial and paraglacial processes contributed to the whole present-day cirque eroded volume, the greater the volume related to paraglacial RSF, the smaller the volume generated by strictly glacial erosion processes. Hence, glacial erosion rates are particularly low in cirques 1 and 3 while they logically display values very similar to those of overall erosion rates in cirques 6, 7 and 8 where the calculated RSF contribution is negligible.

40Such an approach involves defining a stable time duration applied to all the cirques (in this case, 1.25 Ma or half of the Quaternary), assuming that all eight cirques have similar ages. However, if cirques started to form in RSF cavities, they probably have different ages. The excavated cirque volume is probably primarily controlled by (i) the age of the initial activation of the RSF event in which the cirque was formed and (ii) the intensity of RSF activity affecting its cirque wall during subsequent paraglacial sequences. Cirques 2, 3 and 8 are probably smaller because the RSF events in the cavities of which cirques started to develop occurred during a later paraglacial sequence than for the large ones and/or because the initial cavity of the RSF was smaller than for the large ones.

Tab. 3 –Mixed and unmixed paraglacial and glacial erosion rates in the shaping of cirques together with extrapolated time required to carve out cirques on the basis of strictly glacial processes.
Tab. 3 – Taux d'érosion mélangé et non mélangés du paraglacaire et des glaciers dans la formation des cirques, ainsi que le temps extrapolé nécessaire pour sculpter des cirques sur la base de processus strictement glaciaires.

Tab. 3 –Mixed and unmixed paraglacial and glacial erosion rates in the shaping of cirques together with extrapolated time required to carve out cirques on the basis of strictly glacial processes.  Tab. 3 – Taux d'érosion mélangé et non mélangés du paraglacaire et des glaciers dans la formation des cirques, ainsi que le temps extrapolé nécessaire pour sculpter des cirques sur la base de processus strictement glaciaires.

E1. Overall erosion rates including both glacial and paraglacial processes; E2. Strictly glacial erosion rates; t. Time required to carve out cirques only on the basis of glacial erosion rates.
E1. Taux d'érosion globaux incluant les processus glaciaires et paraglaciaires. E2. Taux d'érosion strictement glaciaire. t. Temps requis pour sculpter les cirques uniquement sur la base des taux d’érosion glaciaire.

5. Discussion

5.1. Predominant control of paraglacial processes in the shaping of cirques

41In cold landscape geomorphology, whether glacial or periglacial processes are the most erosive agents in the carving of cirques is still an issue. In the classic glacial theory of cirques, the deepening of the cirque floor is mainly controlled by glacial processes (plucking and the work of basal meltwater under temperate-based glaciers) while the periglacial ones (action of frost or weathering) control the supra-glacial cirque-wall retreat (Evans, 2008). Numerous morphometric analyses have been carried out to support the dominance of glacial or periglacial processes in the shaping of cirques: deep-seated cirques confirm the predominant influence of glacial processes while elongated ones support the primary role of periglacial processes. The cirques in the Tindastóll Mountain are thus closely related to those elsewhere that display elongated shapes but with the major difference that they have a paraglacial origin, not a periglacial one. The hypothesis of the major role of non-glacial processes (paraglacial or periglacial) in the carving of cirques is also in good agreement with several studies highlighting that the rates of horizontal lengthening of cirques are significantly higher than their rates of vertical deepening. In the Southern Alps of New Zealand, Brook et al. (2006) calculated a rate of 0.44 mm.yr¹ for backwall retreat and 0.29 mm.yr¹ for vertical deepening. In the Himalayas, along the north face of Annapurna, Heimsath and McGlynn (2008) found 0.42 ± 0.16 mm.yr¹ for vertical deepening and 1.2 ± 0.5 mm.yr¹ for horizontal retreat.

42Assuming that cirques are strictly related to glacial erosional processes also implies a coupling between denudation rates and glacier dynamics throughout a glaciation cycle, with rates increasing during glaciation and decreasing during deglaciation. However, in the eastern Pyrenees, Delmas et al. (2009) found a very different pattern with low rates during pleniglacial conditions (0.05 mm.yr¹) contrasting with the peak erosion rate measured during the Late Glacial transition to ice-free conditions (0.6 mm.yr¹). These results were in agreement with results form Crest et al. 2017 and suggest that erosion processes are mainly active during deglaciation, in good agreement with the hypothesis of a predominant control of paraglacial processes in the shaping of cirques. Such observations suggest that the main role of glaciers is not necessarily their presumed ability to erode bedrock and to excavate cirques during glaciation.

43Whereas the role of glacial processes in carving out cirques remains hard to demonstrate, numerous studies have pointed out the role of glacial debuttressing in the development of major ridge and slope instabilities in the context of deglaciation (Bovis, 1990; Ballantyne, 2002; Ballantyne et al., 2014; Jarman, 2006; Hewitt, 2009; Kellerer-Pirklbauer, 2010; McColl, 2012). Paraglacial deformations can occur while the adjacent valleys are still partly buttressed by glaciers, suggesting that only a small lowering of glaciers can promote ridge and slope instabilities (Bovis, 1982, 1990; Mège and Bourgeois, 2011; McColl and Davies, 2013; Gourronc et al., 2014). On the eastern side of the Skagafjörður, the ridge of Hnjúkar experienced paraglacial gravitational spreading in an early stage of the late Weichselian deglaciation whereas the adjacent valleys were still occupied by valley glaciers at least 300 meters thick, confirming that the basaltic outcrops are highly sensitive to minor changes in ice dynamics (Coquin et al., 2015). Such observations provide good evidence that only minor changes in glacier dynamics can provoke significant paraglacial instabilities.

5.2. Role of glaciers as conveyors to evacuate debris

44Numerous studies have also outlined the role of deglaciation in the activation of mass-movements (Ballantyne et al., 1998; Ballantyne, 2013; Ballantyne and Stone, 2013; Wilson, 2005; Wilson and Smith, 2006; Peras et al., 2016). In the Skagafjörður, many RSF deposits are still emplaced on the valley floor, suggesting that the most recent RSF activity occurred after the Late Weichselian deglaciation (Jónsson, 1957; Pétursson and Sæmundsson, 2008; Cossart et al., 2014; Coquin et al. 2016) or during concomitant to previous deglaciation periods if RSF scars are visible but not their deposits. This highlights the significant role of the glacier in cirque development by conveying RSF deposits. These deposits are evacuated either synchronously or during a further glacier advance depending on whether the cirque/valley floor is still partly glaciated or already ice-free at the time of the RSF event. These observations emphasize the positive feedback loop of the glacier in the shaping of the cirque during glaciation and deglaciation cycles: high rates of debris supplied by slope instabilities are induced by a receding glacier during deglaciation; debris are then conveyed and evacuated by the glacier during a subsequent glacier readvance; promoting both the erosion and transport conditions are promoted during such cycles, contributing to carve out the cirque. The Tinsdastóll cirques illustrate such a positive loop.

5.3. Implication of a paraglacial origin of cirques

45Evolutionary models assume, rather than demonstrate, that cirques are glacial landforms, meaning that the predominant controls in the making of cirques are purely glacial processes, active when glaciers occupy the cirques. Such an assertion also provides the tempting opportunity to use morphometric attributes of cirques as proxies for paleo-climate or Quaternary glaciation reconstructions (Sugden, 1969; Aniya and Welch 1981; Haynes 1998; De Blasio 2002; Brook et al. 2006; Principato and Lee, 2014). However, the potential impact of non-glacial processes, specifically paraglacial ones occurring in the context of deglaciation, has been neglected. In the case of a paraglacial origin of cirques, the frequency of glacier fluctuations during the Quaternary primarily controls the shape and size of present-day cirques rather than the intensity or the duration of glaciation: the larger the size of a cirque, the greater the number of glacier fluctuations and thus the greater impact of paraglacial processes involved in its shaping. Analyzing the distribution of cirques compared to the Quaternary ice-sheet distribution in the eastern Pyrenees, Delmas et al. (2014) noted that the largest cirques were mainly found in relatively low areas where glaciers were less abundant and short-lived. In the case of dominant glacial erosion in shaping cirques, the largest cirques might be observed in areas that have experienced maximal glaciations during the Quaternary. The location of these cirques is, however, consistent with cirques having a mass movement origin. In addition, Delmas et al. 2014 reported that the large features displayed high w/a and l/a morphometric ratios, similar to those observed in the Tindastóll Mountain, suggesting that they probably underwent great paraglacial activity enhanced by a high frequency of glacier fluctuations. These observations support the shape and size of cirques providing insight into the intensity of paraglacial activity and the frequency of Quaternary glacier fluctuations rather than the duration of glaciations. In the Westfjords of Iceland, Principato and Lee (2014) observed a positive correlation between cirque distribution and the distance to the present-day coastline with the number of cirques increasing seaward. The authors concluded that the proximity of open water provides the moisture supply favorable to the development of cirque glaciers. It is interesting to note that such a spatial correlation also coincides with the amount of post-glacial vertical uplift, which seems to increase from the interior of Iceland to the outer parts (Stewart et al., 2000; Dauteuil et al., 2005). This is particularly the case in the Westfjords where raised shorelines (documenting the post-glacial rebound) reach the maximal elevation of 150 m a.s.l. recorded in Iceland (Ingólfsson and Norddahl, 2001). Additionally, in the Skagafjörður area, Cossart et al. (2014) noted that RSFs were more abundant in the outer parts of the fjord where post-glacial uplift had reached its maximum activity, suggesting that RSFs might be triggered by the seismic shaking induced by post-glacial crustal deformations. Perhaps the high level of post-glacial crustal deformations in the outer parts of the Westfjords has promoted high RSF activity, which could explain the correlation between cirque distribution and distance to the seashore.

46In addition, the number of glaciation and deglaciation cycles is probably underestimated in Icelandic cirques as stratigraphic and sedimentological studies have mostly documented the highest magnitude glaciations of the Quaternary (Geirsdóttir et al., 2007). This hypothesis is supported by morphological observations together with paleo-climatic reconstructions. In Iceland, the relief reflects competition between constructive agents (volcanism) and denudation processes, carving out a range of erosional landforms, from first generation and simple features such as cirques or valley-heads to main troughs and fjords. If the rate of destructive processes is considered, first cirque generation can only develop if their erosion rates outweigh the rate of processes involved in fjord and trough widening. Therefore, the presence of first cirques generation indicates that the rate of backwall retreat is greater for cirque walls than for adjacent second generation of troughs and fjords, at least during more recent times. This observation suggests that cirques have probably experienced many more glaciation and deglaciation cycles during the Quaternary than valleys of higher order. This is in good agreement with paleo-ELA reconstructions, which point out the great sensitivity of cirque glaciers to climate change (Dahl and Nesje, 1992). While a short cold event is sufficient for a glacier to develop in a cirque, the glaciation of a large fjord requires the several tens of thousands of years of stadial conditions that prevailed during the Quaternary. Moreover, paleo-climatic reconstructions from glacial, oceanic and terrestrial proxies have revealed the high sensitivity of oceanic and atmospheric conditions in the central north Atlantic, supporting the idea that Iceland probably underwent a high frequency ice-sheet and/or ice-cap fluctuation during the Quaternary in response to minor climate changes (Geirsdóttir et al., 2009). Hence, small climatic variations during the Quaternary may have encouraged rapid and significant changes in the accumulation-zone of regional ice sheets and/or local ice-fields because of the topographic configuration of ice-catchments, mostly characterized by widespread plateaus. In this context, cirques were probably occupied by rapidly-fluctuating outlets fed by ice-fields located on surrounding plateaus. Thus, they probably experienced significantly more glaciation and deglaciation cycles than the 18 main glaciations recorded for 2.5 Ma. Extrapolation must therefore take into account that cirques may have experienced many more paraglacial sequences than previously estimated. At a constant rate of paraglacial RSF activity, around 25 to 30 paraglacial sequences would be sufficient to complete the overall shaping of cirques 1 and 3 (fig. 10).

6. Conclusion

47In this study, we have proposed a methodology based on paleo-surface reconstruction and calculation of missing volumes of cirques and RSF cavities to assess the contribution of RSFs on the carving of cirques of the Tindastóll Mountain (at the time scale of the last paraglacial sequence and throughout the Quaternary) and to estimate Quaternary denudation rates. According to the specific geological setting of the Tindastóll where cirques and RSFs coexists, this study highlights the potentially significant impact of paraglacial mass-movements on the shaping of these cirques.

(i) The results show that the paraglacial RSFs exert an unstable influence, from null or negligible in some cases to predominant in others.

(ii) The observations indicate that several RSF deposits have been evacuated during glacier readvances, highlighting both that cirques have experienced mass-movements during previous paraglacial sequences and that glaciers are efficient to transport and evacuate the debris produced during non-glacial stages.

(iii) Our observations have outlined that RSF activity has been in some instances more important during previous paraglacial sequence, suggesting that the real paraglacial activity in the carving of cirques is underestimated by the RSF contribution ratio only related to the last deglaciation stage.

(iv) The results highlight that Quaternary denudation rates of the cirques along the Tindastóll Mountain are ranging between 0.08 to 0.17 mm.yr¹.

(v) The strictly glacial rates (without the contribution of RSFs) for cirques where the RSF activity is not null are ranging between 0.02 to 0.13 mm.yr¹.

(vi) The observations carried out in the study area suggest that the shape and size of cirques probably better reflect on the frequency of glacier fluctuations during the Quaternary rather than the duration of glaciations.

Even if the RSF contribution ratio provides an interesting tool to roughly evaluate the potential influence of mass-movements on the carving of cirques at the time scale of the last paraglacial sequence, difficulties arise to accurately estimate the impact of mass-movements which occurred during previous paraglacial sequences. Hence, further investigations are still required to clarify the potential influence of mass-movements in cirque enlargement throughout the Quaternary. Such an influence could probably be partly validated through other approaches such as surface exposure dating of the bedrock surfaces. For example, cosmogenic exposure ages at various parts of the cirque bedrock floor might inform about the rates of vertical versus lateral/headward erosion, thus providing support for mass movement dominating the development of the cirque.

Haut de page

Bibliographie

Agliardi F., Crosta G., Zanchi, A. (2001) – Structural constraints on deep-seated slope deformation kinematics. Engineering Geology, 59, 83‑102.
DOI :
10.1016/S0013-7952(00)00066-1

Agliardi F., Crosta G.B., Zanchi A., Ravazzi C. (2009) – Onset and timing of deep-seated gravitational slope deformations in the eastern Alps, Italy. Geomorphology, 103, 113‑129.
DOI :
10.1016/j.geomorph.2007.09.015

Ambrosi C., Crosta, G.B. (2006) – Large sackung along major tectonic features in the Central Italian Alps. Engineering Geology, 83, 183‑200.
DOI :
10.1016/j.enggeo.2005.06.031

Aniya M., Welch R. (1981) – Morphometric analyses of Antarctic cirques from photogrammetric measurements. Geografiska Annaler, 63, 41‑53.
DOI :
10.1080/04353676.1981.11880017

Arsenault A.M., Meigs A.J. (2005) – Contribution of deep-seated bedrock landslides to erosion of a glaciated basin in southern Alaska. Earth Surface Processes and Landforms, 30, 1111‑1125.
DOI :
10.1002/esp.1265

Bachmann D., Bouissou S., Chemenda A. (2009) – Analysis of massif fracturing during Deep-Seated Gravitational Slope Deformation by physical and numerical modeling. Geomorphology, 103, 130‑135.
DOI :
10.1016/j.geomorph.2007.09.018

Ballantyne C.K. (1986) – Landslides and slope failures in Scotland: A review. Scottish Geographical Magazine, 102, 134‑150.
DOI :
10.1080/00369228618736667

Ballantyne C.K. (1991) – Scottish landform examples ‑ 2: The landslides of Trotternish, Isle of Skye. Scottish Geographical Magazine, 107, 130‑135.
DOI :
10.1080/00369229118736821

Ballantyne C.K. (1992) – Rock slope failure and debris flow, Gleann na Guiserein, Knoydart: Comment. Scottish Journal of Geology, 28, 77‑80.
DOI :
10.1144/sjg28010077

Ballantyne C.K. (1997) – Holocene rock slope failures in the Scottish Highlands. Paläoklimaforschung, 19, 77‑80.

Ballantyne C.K. (2002) – Paraglacial geomorphology. Quaternary Science Reviews, 21, 1935‑2017.
DOI :
10.1016/S0277-3791(02)00005-7

Ballantyne C.K. (2008) – After the Ice: Holocene Geomorphic Activity in the Scottish Highlands. Scottish Geographical Journal, 124, 8‑52.
DOI :
10.1080/14702540802300167

Ballantyne C.K. (2013) – Late glacial rock-slope failures in the Scottish Highlands. Scottish Geographical Journal, 129, 67‑84.
DOI :
10.1080/14702541.2013.781210

Ballantyne C.K., Stone J.O., Fifield L.K. (1998) – Cosmogenic Cl-36 dating of postglacial landsliding at The Storr, Isle of Skye, Scotland. The Holocene, 8, 347‑351.
DOI :
10.1191/095968398666797200

Ballantyne C.K., Sandeman G.F., Stone J.O., Wilson P. (2014) – Rock-slope failure following Late Pleistocene deglaciation on tectonically stable mountainous terrain. Quaternary Science Reviews, 86, 144‑157.
DOI :
10.1016/j.quascirev.2013.12.021

Ballantyne C.K., Stone J.O. (2004) – The Beinn Alligin rock avalanche, NW Scotland: cosmogenic 10Be dating, interpretation and significance. The Holocene, 14, 448‑453.
DOI :
10.1191/0959683604hl720rr

Ballantyne C.K., Stone J.O. (2013) – Timing and periodicity of paraglacial rock-slope failures in the Scottish Highlands. Geomorphology, 186, 150‑161.
DOI :
10.1016/j.geomorph.2012.12.030

Beck A.C. (1968) – Gravity faulting as a mechanism of topographic adjustment. New Zealand Journal of Geology and Geophysics, 11, 191‑199.
DOI :
10.1080/00288306.1968.10423684

Benn D.I., Evans D.J.A. (1998) – Glaciers & Glaciation. Arnold, London, 734 p.

Boulton G. (1982) – Processes and Patterns of Glacial Erosion, In Coates D. (Ed.): Glacial Geomorphology. Springer Netherlands, pp. 41‑87.

Bourgeois O., Dauteuil O., Van Vliet-Lanoë B. (2000) – Geothermal control on flow patterns in the Last Glacial Maximum ice sheet of Iceland. Earth Surface Processes and Landforms, 25, 59‑76.
DOI :
10.1002/(SICI)1096-9837(200001)25:1<59::AID-ESP48>3.0.CO;2-T

Bourgeois O., Dauteuil O., Hallot E. (2005) – Rifting above a mantle plume: structure and development of the Iceland Plateau. Geodinamica Acta, 18, 59‑80.
DOI :
10.3166/ga.18.59-80

Bovis M.J. (1982) – Uphill-facing (antislope) scarps in the Coast Mountains, southwest British Columbia. Geological Society of America Bulletin, 93, 804‑812.
DOI :
10.1130/0016-7606(1982)93<804:UASITC>2.0.CO;2

Bovis M.J. (1990) – Rock-slope deformation at Affliction Creek, southern Coast Mountains, British Columbia. Canadian Journal of Earth Sciences 27, 243‑254.
DOI :
10.1139/e90-024

Brook M.S., Kirkbride M.P., Brock B.W. (2006) – Cirque development in a steadily uplifting range: rates of erosion and long-term morphometric change in alpine cirques in the Ben Ohau Range, New Zealand. Earth Surface Processes and Landforms, 31, 1167‑1175.
DOI :
10.1002/esp.1327

Capitani M., Ribolini A., Federici P.R. (2013) – Influence of deep-seated gravitational slope deformations on landslide distributions: A statistical approach. Geomorphology, 201, 127‑134.
DOI :
10.1016/j.geomorph.2013.06.014

Chigira M. (1992) – Long-term gravitational deformation of rocks by mass rock creep. Engineering Geology, 32, 157‑184.
DOI :
10.1016/0013-7952(92)90043-X

Coquin J., Mercier D., Bourgeois O., Cossart E., Decaulne A. (2015) – Gravitational spreading of mountain ridges coeval with Late Weichselian deglaciation: impact on glacial landscapes in Tröllaskagi, northern Iceland. Quaternary Science Reviews, 107, 197‑213.
DOI :
10.1016/j.quascirev.2014.10.023

Coquin J., Mercier D., Bourgeois O., Feuillet T., Decaulne A. (2016) – Is the gravitational spreading a precursor of the landslide of Stífluhólar (Skagafjörður, Northern Iceland). Géomorphologie : Relief, Processus, Environnement, 22 (1), 9‑24.
DOI :
10.4000/geomorphologie.11295

Cossart E., Mercier D., Coquin J., Decaulne A., Feuillet T., Jónsson H.P., Sæmundsson Þ. (2017) – Denudation rates during a postglacial sequence in Northern Iceland: example of Laxárdalur valley in the Skagafjörður area. Geografiska Annaler, 99 (3), 240‑261.
DOI :
10.1080/04353676.2017.1327320

Cossart E., Mercier D., Decaulne A., Feuillet T., Jónsson H.P., Sæmundsson Þ. (2014) – Impacts of post-glacial rebound on landslide spatial distribution at a regional scale in northern Iceland (Skagafjörður). Earth Surface Processes and Landforms, 39 (3), 336‑350.
DOI :
10.1002/esp.3450

Crest Y., Delmas M., Braucher R., Gunnell Y., Calvet M., ASTER Team (2017) – Cirques have growth spurts during deglacial and interglacial periods: Evidence from 10Be and 26Al nuclide inventories in the central and eastern Pyrenees. Geomorphology, 278, 60‑77.
DOI :
10.1016/j.geomorph.2016.10.035

Crosta G.B., Frattini P., Agliardi F. (2013) – Deep seated gravitational slope deformations in the European Alps. Tectonophysics, 605, 13‑33.
DOI :
10.1016/j.tecto.2013.04.028

Dahl S.O., Nesje A. (1992) – Paleoclimatic implications based on equilibrium-line altitude depressions of reconstructed Younger Dryas and Holocene cirque glaciers in inner Nordfjord, western Norway. Palaeogeography Palaeoclimatology Palaeoecology, 94, 87‑97.
DOI :
10.1016/0031-0182(92)90114-K

Dauteuil O., Bouffette J., Tournat F., Van Vliet-Lanoë B., Embry J., Quété Y. (2005) – Holocene vertical deformation outside the active rift zone of north Iceland. Tectonophysics, 404, 203‑216.
DOI :
10.1016/j.tecto.2005.04.009

De Blasio F.V. (2002) – Note on simulating the size distribution of glacial cirques. Earth Surface Processes and Landforms, 27, 109‑114.
DOI :
10.1002/esp.307

Decaulne A., Cossart É., Mercier D., Coquin J., Feuillet T., Jónsson H.P. (2016) – An early Holocene age for the Vatn landslide (Skagafjörður, central northern Iceland): Insights into the role of postglacial landsliding on slope development. The Holocene, 26, 1304‑1318.
DOI :
10.1177/0959683616638432

Decaulne A., Sæmundsson Þ., Jónsson H.P., Sandberg O. (2007) – Changes in deposition on a colluvial fan during the upper Holocene in the Tindastóll Mountain, Skagafjörður district, North Iceland: preliminary results. Geografiska Annaler, 89, 51‑63.
DOI :
10.1111/j.1468-0459.2007.00307.x

Delmas M., Calvet M., Gunnell Y. (2009) – Variability of Quaternary glacial erosion rates – A global perspective with special reference to the Eastern Pyrenees. Quaternary Science Reviews, 28, 484‑498.
DOI :
10.1016/j.quascirev.2008.11.006

Delmas M., Gunnell Y., Calvet M. (2014) – Environmental controls on alpine cirque size. Geomorphology, 206, 318‑329.
DOI :
10.1016/j.geomorph.2013.09.037

Einarsson P. (1991) – Earthquakes and present-day tectonism in Iceland. Tectonophysics, 189, 261‑279.
DOI :
10.1016/0040-1951(91)90501-I

Eiríksson J. (1985) – Facies analysis of the Breidavík group sediments on Tjörnes, North Iceland. Acta naturalia Islandica 31, 56 p.

Etlicher B. (1986) – Les massifs du Forez, du Pilat et du Vivarais: régionalisation et dynamique des héritages glaciaires et périglaciaires en moyenne montagne cristalline. Thèse de doctorat d’État, Université de Lyon II, Centre d’études foréziennes, 681 p.

Embleton C., King C.A.M. (1975) – Glacial and periglacial geomorphology. volume 1: 573 p. & volume 2 : 203 p.

Evans I.S. (1974) – The geomorphology and asymmetry of glaciated mountains with special reference to the Bridge River District, British Columbia. St. Catherine's College, Cambridge, 449 p.

Evans I.S. (2008) – Glacial erosion processes and forms: mountain glaciation and glacier geography. In Burt T.P., Chorley R.J., Brunsden D., Cox N.J., Goudie A.S. (Eds.): The history of the study of landforms or the development of geomorphology, vol. 4: Quaternary and recent processes and forms (189081965) and the mid-century revolutions. The Geological society of London, 413‑494.

Evans I.S., Cox N. (1974) – Geomorphometry and the operational definition of cirques. Area, 6, 150‑153.

Evans I., Cox N. (1995) – The form of glacial cirques in the English-Lake district, Cumbria. Zeitschrift für Geomorphologie, 39, 175‑202.

Feuillet T., Coquin J., Mercier D., Cossart E., Decaulne A., Jónsson H.P., Sæmundsson Þ. (2014) – Focusing on the spatial non-stationarity of landslide predisposing factors in northern Iceland: Do paraglacial factors vary over space Progress in Physical Geography, 38, 354‑377.
DOI : 10.1177/0309133314528944

Galibert G. (1962) – Recherches sur les processus d'érosion glaciaires de la Haute Montagne Alpine. Bulletin de l'Association de géographes français, 303-304, 8‑46.
DOI : 10.3406/bagf.1962.5577

Garcia S., Arnaud N.O., Angelier J., Bergerat F., Homberg C. (2003) – Rift jump process in Northern Iceland since 10 Ma from 40Ar/39Ar geochronology. Earth and Planetary Science Letters, 214, 529‑544.
DOI :
10.1016/S0012-821X(03)00400-X

Gardner J.S. (1987) – Evidence for headwall weathering zones, Boundary Glacier, Canadian Rocky Mountains. Journal of Glaciology, 33, 60‑67.
DOI :
10.3189/S0022143000005359

Geirsdóttir Á., Eiriksson J. (1994) – Growth of an intermittent ice sheet in Iceland during the late Pliocene and early Pleistocene. Quaternary Research, 42, 115‑130.
DOI :
10.1006/qres.1994.1061

Geirsdóttir Á., Eiriksson J. (1996) – A review of studies of the earliest glaciation of Iceland. Terra Nova 8, 400‑414.
DOI :
10.1111/j.1365-3121.1996.tb00765.x

Geirsdóttir Á., Miller G.H., Andrews J.T. (2007) – Glaciation, erosion, and landscape evolution of Iceland. Journal of Geodynamics 43, 170‑186.
DOI :
10.1016/j.jog.2006.09.017

Geirsdóttir Á., Miller G.H., Axford Y., Sædís Ó. (2009) – Holocene and latest Pleistocene climate and glacier fluctuations in Iceland. Quaternary Science Reviews, 28, 2107‑2118.
DOI :
10.1016/j.quascirev.2009.03.013

Glasser N.F., Hall A.M. (1997) – Calculating Quaternary glacial erosion rates in northeast Scotland. Geomorphology, 20, 29‑48.
DOI :
10.1016/S0169-555X(97)00005-6

Gourronc M., Bourgeois O., Mège D., Pochat S., Bultel B., Massé M., Le Deit L., Le Mouélic S., Mercier D. (2014) – One million cubic kilometers of fossil ice in Valles Marineris: Relicts of a 3.5 Gy old glacial landsystem along the Martian equator. Geomorphology, 204, 235‑255.
DOI :
10.1016/j.geomorph.2013.08.009

Hallet B., Hunter L., Bogen J. (1996) – Rates of erosion and sediment evacuation by glaciers: A review of field data and their implications. Global and Planetary Change, 12, 213‑235.
DOI :
10.1016/0921-8181(95)00021-6

Haynes V.M. (1998) – The morphological development of alpine valley heads in the Antarctic Peninsula. Earth Surface Processes and Landforms, 23, 53‑67.
DOI :
10.1002/(SICI)1096-9837(199801)23:1<53::AID-ESP817>3.0.CO;2-O

Hebdon N.J., Atkinson T.C., Lawson T.J., Young I.R. (1997) – Rate of Glacial Valley Deepening During the Late Quaternary in Assynt, Scotland. Earth Surface Processes and Landforms, 22, 307‑315.
DOI :
10.1002/(SICI)1096-9837(199703)22:3<307::AID-ESP759>3.0.CO;2-U

Heimsath A.M., McGlynn, R. (2008) – Quantifying periglacial erosion in the Nepal high Himalaya. Geomorphology, 97, 5‑23.
DOI :
10.1016/j.geomorph.2007.02.046

Helland A. (1877) – On the ice-fjords of North Greenland and on the formation of fjords, lakes and cirques in Norway and Greenland. Quarterly Journal of the Geological Society, XXXIII, 161 et seq.
DOI :
10.1144/GSL.JGS.1877.033.01-04.10

Hewitt K. (2009) – Glacially conditioned rock-slope failures and disturbance-regime landscapes, Upper Indus Basin, northern Pakistan, In Knight J., Harrison S. (Eds.): Periglacial and paraglacial processes and environments. Geological Society, London, pp. 235‑255.
DOI :
10.1144/SP320.15

Hooke R. (1991) – Positive feedbacks associated with erosion of glacial cirques and overdeepenings. Geological Society of America Bulletin, 103, 1104‑1108.
DOI :
10.1130/0016-7606(1991)103<1104:PFAWEO>2.3.CO;2

Hughes P., Gibbard P., Woodward J. (2007) – Geological controls on Pleistocene glaciation and cirque form in Greece. Geomorphology, 88, 242‑253.
DOI : 10.1016/j.geomorph.2006.11.008

Huguet F. (2008) – Cirques glaciaires et étagement des formes dans le massif du Feldberg, dans le sud de la Forêt Noire (Allemagne). Géomorphologie: Relief, Processus, Environnement, 13 (4), 309‑318.
DOI : 10.4000/geomorphologie.4372

Hutchinson M. (1993) – Development of a continent-wide DEM with applications to terrain and climate analysis. In M.F. Goodchild, B.O. Parks, L.T. Streyaert (Eds): Environmental modeling with GIS, New York Oxford University Press, 392‑399.

Ingólfsson Ó., Norddahl H. (2001) – High relative sea level during the Bolling Interstadial in western Iceland: a reflection of ice-sheet collapse and extremely rapid glacial unloading. Arctic, Antarctic, and Alpine Research, 33 (2), 231‑243.
DOI :
10.1080/15230430.2001.12003426

Jaboyedoff M., Penna I., Pedrazzini A., Baroň I., Crosta G.B. (2013) – An introductory review on gravitational-deformation induced structures, fabrics and modeling. Tectonophysics, 605, 1‑12.
DOI :
10.1016/j.tecto.2013.06.027

Jarman D. (2006) – Large rock slope failures in the Highlands of Scotland: characterisation, causes and spatial distribution. Engineering Geology, 83, 161‑182.
DOI :
10.1016/j.enggeo.2005.06.030

Jarman D. (2009) – Paraglacial rock slope failure as an agent of glacial trough widening. Geological Society, London, Special Publications 320, 103‑131.
DOI :
10.1144/SP320.8

Jóhannesson H., Sæmundsson K. (2009) – Geological Map of Iceland. 1:600,000. Tectonics. Icelandic Institute of Natural History, Reykjavik.

Johnson W.D. (1904) – The Profile of Maturity in Alpine Glacial Erosion. The Journal of Geology, 12, 569‑578.
DOI :
10.1086/621181

Jónsson Ó. (1957) – Skriðuföll og snjóflóð. Bókaútgáfan norðri, p. 141.

Jull M., McKenzie D. (1996) – The effect of deglaciation on mantle melting beneath Iceland. Journal of Geophysical Research: Solid Earth, 101, 21815‑21828.
DOI :
10.1029/96JB01308

Kellerer-Pirklbauer A., Proske H., Strasser V. (2010) – Paraglacial slope adjustment since the end of the Last Glacial Maximum and its long-lasting effects on secondary mass wasting processes: Hauser Kaibling, Austria. Geomorphology, 120, 65‑76.
DOI :
10.1016/j.geomorph.2009.09.016

Kristjansson L., Gudmundsson A., Hardarson B. (2004) – Stratigraphy and paleomagnetism of a 2.9-km composite lava section in Eyjafjördur, Northern Iceland: a reconnaissance study. International Journal of Earth Science, 93, 582‑595.
DOI :
10.1007/s00531-004-0409-4

Långbacka B.O., Gudmundsson A. (1995) – Extensional tectonics in the vicinity of a transform fault in north Iceland. Tectonics, 14, 294‑306.
DOI :
10.1029/94TC02904

Le Breton E., Dauteuil O., Biessy G. (2010) – Post-glacial rebound of Iceland during the Holocene. Journal of the Geological Society, 167, 417‑432.
DOI :
10.1144/0016-76492008-126

Lewis W. (1939) – Snow-patch erosion in Iceland. Geographical Journal, 94 (2), 153‑161.
DOI :
10.2307/1787251

Matthes F.E. (1900) – Glacial sculpture of the Bighorn Mountains. Wyoming. US Geological Survey Twenty-first Annual Report Part II, 167‑190.

McColl S.T. (2012) – Paraglacial rock-slope stability. Geomorphology, 153‑154, 1‑16.
DOI :
10.1016/j.geomorph.2012.02.015

McColl S.T., Davies T.R.H. (2013) – Large ice-contact slope movements: glacial buttressing, deformation and erosion. Earth Surface Processes and Landforms, 38, 1102‑1115.
DOI :
10.1002/esp.3346

Mège D., Bourgeois O. (2011) – Equatorial glaciations on Mars revealed by gravitational collapse of Valles Marineris wallslopes. Earth and Planetary Science Letters, 310, 182‑191.
DOI :
10.1016/j.epsl.2011.08.030

Meierding T.C. (1982) – Late pleistocene glacial equilibrium-line altitudes in the Colorado Front Range: A comparison of methods. Quaternary Research, 18, 289‑310.
DOI :
10.1016/0033-5894(82)90076-X

Mercier D., Coquin J., Feuillet T., Decaulne A., Cossart E., Jónsson H.P., Sæmundsson Þ. (2017) – Are Icelandic rock-slope failures paraglacial« Age evaluation of seventeen rock-slope failures in the Skagafjörður area, based on geomorphological stacking, radiocarbon dating and tephrochronology. Geomorphology, 296, 45–58.
DOI :
10.1016/j.geomorph.2017.08.011

Mercier D., Cossart E., Decaulne A., Feuillet T., Jónsson H.P., Sæmundsson Þ. (2013) – The Höfðahólar rock avalanche (sturzström): Chronological constraint of paraglacial landsliding on an Icelandic hillslope. The Holocene, 23, 432‑446.
DOI :
10.1177/0959683612463104

Olyphant G.A. (1977) – Topoclimate and the depth of cirque erosion. Geografiska Annaler, 59, 3‑4, 209‑213.
DOI :
10.1080/04353676.1977.11879952

Ostermann M., Sanders D., Ivy-Ochs S., Alfimov V., Rockenschaub M., Römer A. (2012) – Early Holocene (8.6 ka) rock avalanche deposits, Obernberg valley (Eastern Alps): Landform interpretation and kinematics of rapid mass movement. Geomorphology, 171‑172, 83‑93.
DOI :
10.1016/j.geomorph.2012.05.006

Paillard D. (2015) – Quaternary glaciations: from observations to theories. Quaternary Science Reviews, 107, 11‑24.
DOI :
10.1016/j.quascirev.2014.10.002

Penck A., Brückner E. (1904‑1905‑1906) – Die Alpen im Eiszeitalter. Tauchnitz ed, Leipzig. 3 volumes, 1,200 p.

Peras A., Decaulne A., Cossart E., Coquin J., Mercier D. (2016) – Distribution and spatial analysis of rockslides failures in the Icelandic Westfjords: first results. Géomorphologie : Relief, Processus, Environnement, 22 (1), 25‑35.
DOI : 10.4000/geomorphologie.11303

Pétursson H.G., Sæmundsson Þ. (2008) – Skriðuföll í Skagafirði, In Sæmundsson Þ., Decaulne A., Jónsson H.P. (Eds.): Skagfirsk náttúra 2008. Náttúrustofa Norðurlands vestra NNV, Sauðárkrókur, pp. 25‑20.

Porter S.C. (2000) – Snowline depression in the tropics during the Last Glaciation. Quaternary Science Reviews, 20, 1067‑1091.
DOI :
10.1016/S0277-3791(00)00178-5

Principato S.M., Lee J.F. (2014) – GIS analysis of cirques on Vestfirðir, northwest Iceland: implications for palaeoclimate. Boreas, 43, 807‑817.
DOI :
10.1111/bor.12075

Raymo M., Ruddiman W., Backman J., Clement B., Martinson D. (1989) – Late Pliocene variation in Northern Hemisphere ice sheets and North Atlantic deep water circulation. Paleoceanography, 4, 413‑446.
DOI :
10.1029/PA004i004p00413

Rundgren M., Ingólfsson Ó., Björck S., Jiang H., Haflidason H. (1997) – Dynamic sea-level change during the last deglaciation of northern Iceland. Boreas, 26, 201‑215.
DOI :
10.1111/j.1502-3885.1997.tb00852.x

Saemundsson K. (1979) – Outline of the geology of Iceland. Jökull, 29, 7‑28.

Sanders J.W., Cuffey K.M., Moore J.R., MacGregor K.R., Kavanaugh J.L. (2012) – Periglacial weathering and headwall erosion in cirque glacier bergschrunds. Geology, 40, 779‑782.
DOI :
10.1130/G33330.1

Savage W.Z., Varnes D.J. (1987) – Mechanics of gravitational spreading of steep-sided ridges («sackung»). Bulletin of the International Association of Engineering Geology, 35, 31‑36.
DOI :
10.1007/BF02590474

Sparrow G.W.A. (1974) – Non-glacial cirque formation in southern Africa. Boreas 3, 61‑68.
DOI :
10.1111/j.1502-3885.1974.tb00828.x

Stewart I.S., Sauber J., Rose J. (2000) – Glacio-seismotectonics: ice sheets, crustal deformation and seismicity. Quaternary Science Reviews, 19, 1367‑1389.
DOI :
10.1016/S0277‑3791(00)00094‑9

Sugden D.E. (1969) – The age and form of corries in the Cairngorms. The Scottish Geographical Magazine, 85, 34‑46.
DOI :
10.1080/00369226908736110

Turnbull J.M., Davies T.R.H. (2006) – A mass movement origin for cirques. Earth Surface Processes and Landforms, 31, 1129‑1148.
DOI :
10.1002/esp.1324

Wilson P. (2005) – Paraglacial rock-slope failures in Wasdale, western Lake District, England: morphology, styles and significance. Proceedings of the Geologists' Association, 116, 349‑361.
DOI :
10.1016/S0016-7878(05)80052-5

Wilson P., Smith A. (2006) – Geomorphological characteristics and significance of late quaternary paraglacial rock–slope failures on Skiddaw Group terrain, Lake District, northwest England. Geografiska Annaler, 88, 237‑252.
DOI :
10.1111/j.1468-0459.2006.00298.x

White W. (1970) – Erosion of cirques. Journal of Geology, 78, 123‑126.

Haut de page

Annexe

Version française abrégée

Les cirques sont des modelés en creux, à fond plat, entourées de parois escarpées et omniprésents dans les paysages glaciaires (Evans, 1974). Le rôle des glaciers dans leur formation est reconnu depuis longtemps (Helland, 1877 ; Penck et Brückner, 1904), mais leur influence sur la formation des cirques reste un sujet de controverse. En effet, la plupart des modèles glaciaires de développement de cirques supposent une prédominance de l’arrachement et de l'abrasion (Galibert, 1962 ; White, 1970), tandis que les modèles périglaciaires accordent une importance primordiale à l’érosion par le gel le long des parois (Johnson, 1904 ; Gardner, 1987 ; Hooke, 1991 ; Sanders et al., 1991). Les deux cas sont basés sur l'hypothèse que les cirques se forment uniquement pendant les glaciations et impliquent que leurs formes et leurs tailles pourraient en quelque sorte refléter l'intensité et la durée de ces glaciations. De plus, les taux d’érosion glaciaires sont généralement trop faibles pour expliquer la taille et la forme des cirques, compte tenu de la durée cumulée des glaciations du Quaternaire. Nous testons ici l'hypothèse selon laquelle le décalage entre le temps requis par l'érosion glaciaire pour sculpter des cirques et la durée cumulée des glaciations du Quaternaire peut être résolu en intégrant la contribution des mouvements de masse paraglaciaires (RSF) à leur développement.

Pour évaluer le rôle des RSF paraglaciaires sur l’évidemment des cirques, nous avons développé une méthodologie basée sur : (i) la reconstruction des surfaces de plateau avant leur incision glaciaire quaternaire, (ii) l'identification et la cartographie des cirques et des cavités liées aux RSF, (iii) la mesure des volumes excavés de cirques et des RSF, et (iv) le calcul des taux d’érosion correspondants.

La zone d'étude examine la montagne Tindastóll qui fait partie d'une chaîne de montagnes côtières de la péninsule de Skagi, dans le nord de l'Islande, en rive gauche du Skagafjörður. La crête s'étend sur 16 km et culmine à 1 000 m dans la partie sud alors que son altitude diminue à environ 600 m dans la partie nord. La crête est un plateau en escalier, composé de basaltes d’âge Tertiaire, disséqué par des cirques situés alternativement sur ses versants est et ouest (fig. 1).

Les cirques et les RSF ont été cartographiés à partir d'orthophotographies aériennes couleur haute résolution de Loftmyndir ehf. disponibles en ligne, à partir d’un MNT de 1 m de résolution généré à partir de mesures LiDAR et des mesures sur le terrain. La partie la plus au nord de la zone d'étude n'étant pas couverte par le levé LiDAR, la résolution ASTER DEM de 30 m (METI et NASA) a été utilisée. Les cartes topographiques de l'Islande produites par Mál og Menning ont également été utilisées pour compléter l'ensemble de données. Les variables morphométriques pour les cirques et les RSF ont été définies d'après Evans et Cox (1974) et Delmas et al. (2014).

La première étape de la reconstruction de la surface pré-quaternaire du plateau a consisté à déduire les contours d'élévation des MNT actuels (fig. 2). Les sections des contours d'élévation qui coïncidaient avec les cirques et les cavités des RSF étaient ensuite remplacées par des contours artificiels suivant des lignes pointillées blanches (fig. 2). Ces contours artificiels ont été dessinés de manière à ce que leurs courbures soient continues avec celles des contours adjacents suivant des lignes noires (fig. 2) afin de préserver l'aspect général des lignes de contour (fig. 3B).

Les résultats montrent que les cirques de la montagne Tindastóll présentent une variété de formes et de tailles (tab. 1), allant des plus simples aux plus petites (A < 1,2 10⁶ m³ - cirques 2, 3 et 8) aux grandes (A > 2,5 10⁶ m³ - cirques 1, 4, 5, 6 et 7) (fig. 4). Cette variété est également reflétée par leurs valeurs l et w, comprises entre 910 et 1 910 m et entre 790 et 2 340 m (tab. 1). Les cirques sont peu profonds en ce qui concerne leur longueur et leur largeur. Cela suggère que les processus d'érosion impliqués dans le recul latéral et de retrait des parois ont été plus efficaces que ceux impliqués dans l’abaissement du plancher (downwearing).

La surface topographique de Tindastóll est divisée en marches d’escalier, séparées par des failles parallèles à l'axe de la crête (fig. 4-5, 6B-C et 7C-D), que l’on retrouve de part et d'autre des cirques, ce qui suggère que les failles se sont développées avant eux (fig. 8). Ces failles sont cohérentes avec les réactivations récentes et locales d’un système de failles hérité du rifting antérieur à 3 Ma. Ces réactivations locales peuvent être déclenchées par un étalement gravitaire induit par la déglaciation et ont également influencé la formation des cirques. Ceci peut être illustré dans le cas du cirque 1, qui présente une orientation N–S similaire à celle de la déformation gravitaire qui s'est d'abord développé à partir d’un graben sommital initial lié à l'étalement gravitationnel (fig. 8B-C). Des liens étroits entre la forme de cirque et le graben sommital hérité sont également observés dans les cirques 4 et 7. Bien que ceux-ci affichent une orientation principale W–E, la deuxième génération de cirques orientés N–S s'est développée en association étroite avec le système hérité de faille N–S. La déformation paraglaciaire du plateau a donc guidé l’érosion glaciaire le long des faiblesses au cours des glaciations ultérieures (fig. 5, 7C).

Les preuves de RSF sont largement répandues dans la zone d'étude (fig. 5 et 6A, C-E). Quinze cavités RSF connectées à des dépôts de glissements de terrain ont été identifiées, d'une surface de 59 à 5567 × 10³ m² et d'un volume de 0,08 à 12,88 × 10⁶ m³ (Tableau 2), indiquant des tailles de RSF très variées. La préservation des RSF suggère que les événements se sont produits après la déglaciation complète. Six des cavités post-glaciaires liées aux RSF sont situées le long des parois de cirque, affectant le volume total excavé des cirques. La distribution spatiale des RSF montre que leurs cavités sont principalement mises en place le long des parois des vallées orientées vers le nord, c’est-à-dire que les plans de rupture des RSF sont contrôlés par le canevas tectonique préexistant (fig. 7, 8B-C).

L’ouverture des grandes cavités par les RSF fournit un bon emplacement pour que les glaciers se développent au cours de la glaciation suivante. Une fois que la cavité initiale est formée par le RSF, les glaciers peuvent façonner le cirque. Le développement des discontinuités le long de la paroi rocheuse facilite également l'érosion glaciaire au cours de la glaciation ultérieure (fig. 8D). Une fois les cirques modelés, une seconde génération de RSF de second ordre peut se développer le long des parois du cirque lors de séquences de déglaciation ultérieures (fig. 8E). Ensuite, les matériaux des RSF accumulés sur le plancher du cirque seront évacués par les ré-avancées ultérieurs des glaciers (fig. 8F).

Vraisemblablement, au moins 18 cycles majeurs de glaciation et de déglaciation ont eu lieu au cours des derniers 2,5 Ma en Islande (Geirsdóttir, 2007), et la contribution globale des RSF à l’évidemment des cirques varie entre 1 % et 71 % (fig. 9-10). Le volume excavé des cirques est probablement principalement contrôlé par (i) l'âge de l'activation initiale de l'événement des RSF dans lequel le cirque s'est formé et (ii) l'intensité de l'activité des RSF affectant la paroi de cirque lors de séquences paraglaciaires ultérieures.

Selon le cadre géologique spécifique de Tindastóll où coexistent des cirques et des RSF, cette étude met en évidence l'impact potentiellement important des mouvements de masse paraglaciaires sur la formation de ces cirques.

(i) Les résultats montrent que les RSF paraglaciaires exercent une influence variable, allant de nulle ou négligeable dans certains cas à prédominante dans d'autres.

(ii) Les observations indiquent que plusieurs dépôts de glissements de masse ont été évacués au cours de séquences glaciaires, soulignant à la fois que les cirques ont connu des mouvements de masse lors de précédentes séquences paraglaciaires et que les glaciers sont efficaces pour transporter et évacuer les débris produits lors des interglaciaires.

(iii) Nos observations ont montré que l'activité des RSF avait parfois été plus importante lors de la précédente séquence paraglaciaire, suggérant que la véritable activité paraglaciaire dans l’excavation de cirques est sous-estimée par le ratio de contribution de RSF lié uniquement au dernier stade de déglaciation.

(iv) Les résultats montrent que les taux de dénudation au Quaternaire des cirques situés le long de montagne Tindastóll sont compris entre 0,08 et 0,17 mm an.

(v) Les taux strictement glaciaires (sans la contribution des RSF) pour les cirques où l'activité de RSF n'est pas nulle sont compris entre 0,02 et 0,13 mm an.

(vi) Les observations effectuées dans la zone d'étude suggèrent que la forme et la taille des cirques reflètent probablement mieux la fréquence des fluctuations des glaciers pendant le Quaternaire que la durée des glaciations.

Même si le ratio de contribution de RSF constitue un outil intéressant pour évaluer de manière approximative l’influence potentielle des mouvements de masse sur l’excavation des cirques à l’échelle de la dernière séquence paraglaciaire, il est difficile d’estimer avec précision l’impact des mouvements de masse survenus au cours des séquences précédentes.

Haut de page

Table des illustrations

Titre Fig. 1 – Study area. Fig. 1 – Zone d’étude.
Légende On the western side of Skagafjörður (A), within Tindastóll Mountain (B); Three sites are studied in details (C-E). Panoramic view of the east-facing slope of the Tindastóll Mountain together with an interpretative sketch of the main geomorphic features (F); 1. Rock slope failure scarp; 2. Rock slope failure deposit; 3. Detritic deposit. Sur la rive ouest de Skagafjörður (A), sur la crête de Tindastóll (B) ; trois sites sont étudiés en détail (C-E). Vue panoramique de Tindastóll et schéma interprétatif des principales caractéristiques géomorphologiques (F) ; 1. Cicatrice d’arrachement des mouvements de masses rocheuses paraglaciaires ; 2. Dépôts des mouvements de masse rocheuse ; 3. Dépôts détritiques.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-1.png
Fichier image/png, 6,0M
Titre Fig. 2 – Paleo-isolines are reconstructed on the basis of 10-m isoline contours derived from LiDAR and ASTER DEMs., of surfaces dissected by cirques (A), and by RSF cavities (B). Fig. 2 – Reconstruction des paléo-isohypses sur la base des isohypses à équidistance de 10 m dérivées des données LiDAR et MNT ASTE, des surfaces disséquées par les cirques (A) et des mouvements de masse rocheuse (B).
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-2.png
Fichier image/png, 904k
Titre Fig. 3 – Estimation of eroded volumes from cirques based on the reconstruction of the paleo-plateau. Fig. 3 – Estimation des volumes érodés des cirques fondée sur la reconstruction du paléo-plateau.
Légende Each cirque is delimited on the present-day LiDAR-DEM (A). Paleo-isolines corresponding to the pre-incision of glacial cirques are drawn (B). The paleo-DEM is derived from reconstructed isolines with a cell resolution of 5x5 m (C). Volume of eroded material is then calculated for each cell by subtracting cell values of the present-day LiDAR DEM from the reconstructed paleo-DEM (D). Chaque cirque est délimité selon le MNT ASTER actuel (A). paléo-isohypses correspondant aux cirques glaciaires avant l’incision (B). Le paléo-MNT est dérivé de la reconstruction des isohypses avec une résolution de cellule 5 x 5 m (C). Le volume de matériel érodé est ensuite calculé pour chaque cellule en soustrayant les valeurs des cellules de l’actuel MNT LiDAR de celles du paléo-MNT reconstruit (D).
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-3.png
Fichier image/png, 2,7M
Titre Tab. 1 – Morphometric criteria and volume of cirques. Tab. 1 – Critères morphométriques et volume des cirques.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-4.png
Fichier image/png, 15k
Titre Tab. 2 – Morphometric criteria and volumes of paraglacial rock-slope failure (RSF). Tab. 2 – Critères morphométriques et volumes des mouvements de masse paraglaciaires.
Légende S1. RSF source area; S2. RSF deposit area; S3. RSF entire body; a. Volume calculation was performed on the deposit area of RSF 4 by using a density correction assuming that RSF deposit is 63% of the bedrock density; b. In spite of its great size, the volume of displaced material was not calculated for RSF 1 given the absence of high-resolution DEM in this location. * The values used to calculate the volume of RSFs. S1. Zone source RSF; S2Zone de dépôt RSF; S3. RSF tout l'ensemble; a. Un calcul de volume a été effectué sur la zone de gisement de RSF 4 en utilisant une correction de densité supposant que le gisement RSF représente 63 % de la densité du substrat rocheux; b. Malgré sa grande taille, le volume de matériau déplacé n’a pas été calculé pour RSF 1 en raison de l’absence de MNT haute résolution à cet endroit. * Valeurs utilisées pour calculer le volume de RSF.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-5.png
Fichier image/png, 15k
Titre Fig. 4 – Present-day transversal profiles (in dark gray) from the ASTER and LiDAR DEMs together with pre-incision profiles (in light gray) obtained from reconstructed paleo-DEMs. Fig. 4 – profils transverses actuels (gris foncé) extraits des MNT ASTER et LiDAR profilant également les surfaces préalables à l’incision (gris clair), obtenus selon la reconstruction des paléo-MNT.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-6.png
Fichier image/png, 1,3M
Titre Fig. 5 – Overall physiography together with a geomorphological map of the Tindastóll Mountain (A) with location of pictures taken during field investigations (fig. 6) and feature references for which volumes have been estimated (Tables 1 and 2). (B) Geomorphological map of the Tindastóll Mountain showing the main glacial-paraglacial related features. Fig. 5 – Physiographie générale de la zone d’étude (A) accompagnée de la localisation des photographies de terrain (fig. 6) et carte géomorphologique de la crête de Tindastóll (B) montrant les formes glaciaires et paraglaciaires liées.
Légende 1. Fault; 2. Dyke; 3. Landslide scarp without deposit; 4. Landslide scarp with deposit; 5. Detritic material; 6. Cirque. 1. Faille ; 2. Dyke ; 3. Escarpement d’arrachement sans dépôt ; 4. Escarpement d’arrachement avec dépôt ; 5. Matériel détritique ; 6. Cirque.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-7.png
Fichier image/png, 2,2M
Titre Fig. 6 – Geomorphological features along the Tindastóll Mountain and their explanatory sketches. Fig. 6 – Éléments géomorphologiques le long de la chaîne Tindastóll et ses schémas explicatifs.
Légende A: Snout of cirque 1 toward SW. Vertical displacements and RSFs on the slope, with a significant downward displacement of block 2. B: N-facing sidewall of cirque 6, looking S. Normal faulting striking north, underlined by a stepped topographic surface. C: The same cirque wall viewed from an upper location on the opposite sidewall, looking SW. The backwall receding triggered by a rotational landslide (RSF-6) gives an arcuate aspect to the S part of the cirque. D: Cirque 3 viewed toward S. The post-glacial translational mass-movement (RSF-4) alters the linear aspect of the cirque headwall. E: The W-facing slope at the snout of cirque 1, viewed from the opposite sidewall. Two RSF cavities alter the linear aspect of the cirque wall. The absence of deposit at the base of the left cavity suggests either that the RSF occurred during a previous deglaciation sequence when the cirque was still occupied by a glacier, which evacuated the supraglacial RSF deposit, or during a previous interstadial stage, the deposit being then removed by subsequent glacier readvances. F: The N-facing wall of cirque 5 shows a linear headscarp and the scarp of an old RSF event that occurred before re-glaciation, which has partly eroded the RSF lateral scarp and has removed the RSF deposits. A : Extrémité du cirque 1 vers le SO. Déplacements verticaux et dépôts de mouvements de masse sur le versant, avec un mouvement descendant du bloc 2. B : La paroi exposée au nord du cirque 6, vue vers le sud, montre une faille normale à rejet vers le nord, soulignée par une marche topographique. C : La même paroi de cirque vue de plus haut, vers le SO ; le retrait de la paroi, dû à un glissement de terrain rotationnel, procure un aspect arqué à la paroi. D : Cirque 3 vu vers le S ; le mouvement de masse translationnel post-glaciaire (RSF 4) altère la linéarité de la paroi du cirque. E : Versant exposé vers l’O à l’extrémité du cirque 1 ; deux cavités de mouvements de masse altèrent la linéarité de la paroi ; l’absence de dépôt à la base de la cicatrice d’arrachement à gauche suggère soit que le glissement de terrain a eu lieu lors d’une séquence de déglaciation antérieure lorsque le cirque était encore englacé, qui a alors évacué le matériel du glissement, soit durant une phase interglaciaire antérieure, le matériel étant évacué lors de la période glaciaire subséquente. F : La paroi exposée au N du cirque 5 montre une nette cicatrice d’arrachement sans dépôt, celui-ci s’étant produit avant la dernière glaciation puisque celle-ci a en partie érodé l’escarpement latéral et a évacué le matériel.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-8.png
Fichier image/png, 2,5M
Titre Fig. 6 – (continuing) Fig. 6 – (suite)
Légende G: Dyke lineament incorporated in the basaltic trap observed along the main E-facing valley slope, where dipping between dyke and lava pile are contrastive. H: The S-facing wall of cirque 4, where the cirque incision reveals sheared rock material within a fault zone area. I-J: Succession of stepped flat surfaces separated by a scarp perpendicular to the main slope. G : Dyke dans l’empilement basaltique le long de la face est de la chaîne de Tindastóll. H : Parois sud du cirque 4 montrant la roche cisaillée et fracturée dans une zone de faille. I-J : Successions de surfaces horizontales séparées par des escarpements perpendiculaires au versant principal.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-9.png
Fichier image/png, 3,2M
Titre Fig. 7 – Simplified sketch of the Tindastóll relief evolution over successive glacial-paraglacial stages (in planform). Fig. 7 – Schéma simplifié de l’évolution du relief sur la chaîne Tindastóll après plusieurs cycles glaciaires et paraglaciaires (en plan).
Légende A: Paleo-plateau before the onset of the Quaternary glaciations. B: First incision of the plateau edge by large-scale RSFs driven by gravitational spreading rock slope deformations, which affect the basaltic plateau in the context of deglaciation (paraglacial sequence). C: Glacial erosion concentrates along depressions generated by previous RSF events and crestal grabens. Such an initial depression allows expansion of glacial troughs and cirques (glacial sequence). D: Present-day relief of the plateau deeply dissected by large cirques formed by the complex interplay of glacial and paraglacial processes. 1. Fault (certain); 2. Fault (questionable); 3. Fault (inferred). A : Paléo-plateau avant les glaciations quaternaires. B : Premières incisions en bordure du plateau par des mouvements de masses rocheuses en contexte de déglaciation (séquence paraglaciaire). C : Concentration de l’érosion glaciaire dans les dépressions générées par les premiers mouvements de masse durant le stade interglaciaire précédent (séquence glaciaire). D : Relief actuel du plateau largement incisé par de larges cirques formé par les jeux combinés des processus glaciaires et paraglaciaires. 1. Faille (certain) ; 2. Faille (discutable) ; 3. Faille (déduite).
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-10.png
Fichier image/png, 597k
Titre Fig. 8 – 3D sketches showing the role slope destabilization processes in trough/cirque genesis and further development. The present-day footprint of glacial cirques along the Tindastóll Mountain is interpreted as being controlled by initial topographic depressions formed under the action of large-scale paraglacial ridges and slope destabilization processes (DSGSDs and RSFs). Fig. 8 – Schémas 3D montrant le rôle des déstabilisations de versant dans la genèse des cirques et leur développement. L’empreinte actuelle des cirques glaciaires de Tindastóll est contrôlée par des dépressions topographiques formées sous l’action de crêtes paraglaciaires et des déstabilisations de versant (DSGSD et mouvements de masses rocheuses).
Légende A: Full-glacial condition. B: Gravitational spreading process coeval with deglaciation (first step of paraglacial readjustment sequence). C: Cirque initiation assisted by deep-seated RSF. D: Cirque development assisted by glacial denudation processes. E: Cirque development assisted by paraglacial RSFs; the activation of second-order rock-slope failures in the context of deglaciation favours the development of cirques. F: Source-area depression produced by 2nd-order rock-slope failures together with ridge-top depression areas initiated by gravitational spreading of the mountain ridge promote snow and ice accumulation, assisting cirque development under the action of glacial erosion processes. G: Large and deep cirques are produced by the gravitational spreading and rock slope failures initiated under paraglacial conditions together with the action of glacial erosion processes under glacial conditions. A : Pléniglaciaire. B : Processus d’étalements gravitationnels contemporains de la glaciation. C : Initiation des cirques assistés par des mouvements de masse rocheuse profonds. D : Développement des cirques assisté des processus de dénudation glaciaire. E : Développement des cirques assistés de mouvements de masse rocheuse paraglaciaires ; l’activation de mouvements de masse secondaires favorise le développement des cirques. F : Les dépressions des zones sources des glissements de masse de second ordre et les dépressions au long des crêtes sommitales liées aux étalements gravitaires favorisent les accumulations de neige et de glace, assistant la formation du cirque sous l’action des processus d’érosion glaciaire. G : Cirques larges et profonds initiés durant les périodes paraglaciaires associés à l’action des processus d’érosion glaciaire durant les périodes glaciaires.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-11.png
Fichier image/png, 783k
Titre Fig. 9 – RSF contribution ratio for the eight cirques of the Tindastóll Mountain given 1, 18 or 30 paraglacial sequences. Fig. 9 – proportion de la contribution des mouvements de masse rocheuse pour la formation de 8 cirques de la montagne Tindastóll selon 1, 18 ou 30 séquences paraglaciaires.
Légende 1. Cirque volume; 2. Paraglacial contribution to cirque volume measured for 1 sequence; 3. Extrapolated for 18 sequences; 4. Extrapolated for 30 sequences. 1. Volume du cirque ; 2. Contribution paraglaciaire au volume du cirque mesurée pour une séquence paraglaciaire ; 3. Extrapolation à 18 séquences paraglaciaires ; 4. Extrapolation à 30 séquences paraglaciaires.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-12.png
Fichier image/png, 2,6M
Titre Fig. 10 – Potential influence of RSF activity on the shaping of cirques given the number of glaciation-deglaciation cycles. Note that for an RSF activity equivalent to the last paraglacial sequence (in volume of eroded material), around 25 to 30 cycles can explain the whole volume of cirques 1 and 3 and around 25% of the overall volume of Tindastóll cirques. Fig. 10 – Influence potentielle de l’activité des mouvements de masse rocheuse au façonnement des cirques selon le nombre de cycles de glaciation-déglaciation. À noter que pour une activité de mouvements de masse rocheuse équivalente à celle de la dernière séquence paraglaciaire (en volume de matériel érodé), entre 25 et 30 cycles peuvent expliquer la totalité des volumes des cirques 1 et 3, mais seulement 25 % du volume de l’ensemble des cirques de Tindadtóll.
Légende 1. Cirque 1; 2. Cirque 3; 3. Cirque 6; 4. Cirque 7; 5. Cirque 8; 6. Total. 1. Cirque 1 ; 2. Cirque 3 ; 3. Cirque 6 ; 4. Cirque 7 ; 5. Cirque 8 ; 6. Total.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-13.png
Fichier image/png, 32k
Titre Tab. 3 –Mixed and unmixed paraglacial and glacial erosion rates in the shaping of cirques together with extrapolated time required to carve out cirques on the basis of strictly glacial processes. Tab. 3 – Taux d'érosion mélangé et non mélangés du paraglacaire et des glaciers dans la formation des cirques, ainsi que le temps extrapolé nécessaire pour sculpter des cirques sur la base de processus strictement glaciaires.
Légende E1. Overall erosion rates including both glacial and paraglacial processes; E2. Strictly glacial erosion rates; t. Time required to carve out cirques only on the basis of glacial erosion rates. E1. Taux d'érosion globaux incluant les processus glaciaires et paraglaciaires. E2. Taux d'érosion strictement glaciaire. t. Temps requis pour sculpter les cirques uniquement sur la base des taux d’érosion glaciaire.
URL http://journals.openedition.org/geomorphologie/docannexe/image/13057/img-14.png
Fichier image/png, 9,7k
Haut de page

Pour citer cet article

Référence papier

Julien Coquin, Denis Mercier, Olivier Bourgeois et Armelle Decaulne, « A paraglacial rock-slope failure origin for cirques: a case study from Northern Iceland », Géomorphologie : relief, processus, environnement, vol.25 - n°2 | 2019, 117-136.

Référence électronique

Julien Coquin, Denis Mercier, Olivier Bourgeois et Armelle Decaulne, « A paraglacial rock-slope failure origin for cirques: a case study from Northern Iceland », Géomorphologie : relief, processus, environnement [En ligne], vol.25 - n°2 | 2019, mis en ligne le 04 juillet 2019, consulté le 19 novembre 2019. URL : http://journals.openedition.org/geomorphologie/13057 ; DOI : 10.4000/geomorphologie.13057

Haut de page

Auteurs

Julien Coquin

Laboratoire LETG-Nantes, UMR 6554, France | Observatoire des Sciences de l’Univers Nantes Atlantique (OSUNA, CNRS UMS 3281), France | CNRS – GDR 2012 "AREES", Meudon, France (julien.coquin@wanadoo.fr).

Articles du même auteur

Denis Mercier

CNRS – GDR 2012 "AREES", Meudon, France | Sorbonne Université, CNRS Laboratoire LGP, UMR 8591, Paris, France (denis.mercier@sorbonne-universite.fr). Tel : +33 (0)1 44 32 14 36.

Articles du même auteur

Olivier Bourgeois

Observatoire des Sciences de l’Univers Nantes Atlantique (OSUNA, CNRS UMS 3281), France | Université de Nantes, CNRS, UMR 6112, Laboratoire de Planétologie et Géodynamique de Nantes, France (olivier.bourgeois@univ-nantes.fr).

Articles du même auteur

Armelle Decaulne

Laboratoire LETG-Nantes, UMR 6554, France | Observatoire des Sciences de l’Univers Nantes Atlantique (OSUNA, CNRS UMS 3281), France | CNRS – GDR 2012 "AREES", Meudon, France (armelle.decaulne@univ-nantes.fr).

Articles du même auteur

Haut de page

Droits d’auteur

© Groupe français de géomorphologie

Haut de page
  • Logo CNRS - Institut des sciences humaines et sociales
  • Logo Groupe français de géomorphologie
  • OpenEdition Journals