Elsevier

Geomorphology

Volume 330, 1 April 2019, Pages 81-88
Geomorphology

River incision and migration deduced from 36Cl cosmic-ray exposure durations: The Clue de la Cerise gorge in southern French Alps

https://doi.org/10.1016/j.geomorph.2019.01.011Get rights and content

Abstract

We estimate recent (20 ka) incision and lateral migration rate on a spectacular ~N-S striking gorge cross-cutting N-dipping limestones in a fold-and-thrust belt of the external Southern French Alps. Upstream of the gorge, the river makes a right angle turn and follows the limestone surface parallel to the bedding direction (~EW). Eighteen samples were gathered along two vertical river polished profiles and three others on the denuded north-dipping surface layer, in order to determine river incision and lateral migration timing and rates. Surface age exposures of polished river cliff and bedding surface have been obtained from 36Cl cosmogenic nuclide dating and allow us to constrain the rate of vertical incision and of surface denudation of the Jurassic limestone layer. Fifteen 36Cl exposure ages have been obtained and range from ~21 ka to ~3 ka. Samples clearly define two age/height linear trends, which slopes correspond to incision rates of 1 and 0.3 mm/yr, respectively. The transition from low to high incision rate occurs at about ~12 ka, i.e. during the Younger Dryas glacial phase. Data also indicate that the vertical incision of the N-S branch of the gorge is coeval with the lateral (northward) river migration of its E-W branch associated with the denudation of the limestones surface. Extrapolating incision rates deduced from 36Cl exposure ages through time suggests that the incision of the Clue and denudation of the Jurassic limestones started about 500–600 ka ago.

Introduction

Vertical superposition of layers with different resistances to fluvial incision result in complex incision rate patterns in space and time (Shtober-zisu et al., 2018). For instance, rivers flowing through rock sequences bearing large lithological contrasts and a strong tectonic structuration often form so-called rectangular or transverse drainage patterns (Oberlander, 1965; Stokes et al., 2008; Twidale, 2004): some branches of the streams run parallel to tectonic structures, while others cross-cut them at high angles to follow the average slope of the topography. Rectangular drainage patterns are particularly well developed in folded or tilted sedimentary formations with contrasted lithologies (Howard, 1967). They can also occur on rocks of relatively uniform lithology, in which case they are controlled by the presence of faults and joints (Anton et al., 2015). Incision mechanisms and corresponding eroded volumes ought to be different depending on the stream direction with respect to the tectonic and stratigraphic structure. For instance, on a dipping stratigraphic pile with a soft rock layer (marls) overlying a more resistant one (limestones), the erodibility contrast will cause a dipward migration of bedding-parallel streams and progressive denudation of the strong layer surface. Meanwhile, bedding-perpendicular streams will incise vertically both layers, resulting in the upstream (retrogressive) migration of a knickpoint controlled by the lithological contrast (Fig. 1). Quantifying the rate at which rivers incise and migrate on their substrate has a potential interest for landscape modelling and mass budget estimates, as the presence or absence of lateral channel migration may lead to large differences in the amount of eroded material (Fig. 1). Moreover, recent studies evidenced that gorge formation can be extremely rapid even with moderate discharge (Anton et al., 2015), which points out the need for a better quantification of bedrock incision rates.

In this study, the recent incision and migration rates were estimated on a spectacular (1000 m long, 400 m high) ~N-S striking gorge cross-cutting the N-dipping Jurassic and Cretaceous sediments in the southernmost part of the external Alps fold-and-thrust belt. In this area, the Esteron River, a large tributary of the Var River, displays a typical rectangular drainage pattern with ENE and NNW striking branches (Fig. 2A and B). The objective of this study was to examine at which rate bedding-perpendicular and bedding-parallel river branches have incised and migrated through time, in order to infer how such transverse drainage patterns develop.

Section snippets

Geological, tectonic and geomorphological setting

The study area encompasses the southernmost units of the external Castellane fold-and-thrust belt, which consists of a Lower Jurassic to Miocene sedimentary cover partly decoupled from its underlying Paleozoic to Permian basement by Upper Triassic evaporites (Fig. 2A). West of the Var River, Miocene shortening was accommodated by both thin- and thick-skin thrusts and ~E-W trending folds (e.g., Jourdon et al., 2014 and references therein), while some right-lateral faulting occurred along the Var

Recent river incision rates in the var catchment

Recent studies based on Cosmic Ray Exposure (CRE) dating of river polished surfaces in the Var catchment have highlighted periods of large incision rates (several mm/yr) since the last 15–20 ka (Rolland et al., 2017; Saillard et al., 2014). In the Vésubie River, which is the next large tributary of the Var River upstream the Esteron, CRE ages range from 2 to 18 ka, leading to an average incision rate of 2.2 mm/yr (Saillard et al., 2014). In the Tinée River, another tributary draining the

Methods

We sampled the rectangular drainage pattern in the upper part of the ‘Clue de la Cerise’ gorge, where Jurassic limestones are deeply incised by the Esteron River. There, the top surface of the Jurassic limestones is well preserved, as well as polished incised surfaces on the gorge cliffs (Fig. 3B and C). 18 samples were gathered along two vertical river polished profiles and three others on the denuded north-dipping surface layer of the Jurassic limestones, on the upstream segment of the gorge.

Results

Along the highest vertical profile, CRE ages range from ~21 ka at the top of it to ~11 ka at the bottom located higher than 5 m above the surface of the river (Table 1 and Fig. 4). Along the lower vertical profile, CRE ages range from 7 to 3 ka, while surface samples have CRE ages between 2 and 8 ka. All samples but 3 align on two age/height linear trends which slopes correspond to mean incision rates of 1 and 0.3 mm/yr, respectively, although the latter is only constrained by 4 points (Fig. 4

Climatic significance of CRE 36Cl ages and incision rates

As detailed in Section 5, the obtained CRE 36Cl ages in this study show age/height linear trends implying incision rates of 0.3 and 1 mm/yr before and after the Younger Dryas glacial phase at ~12 ka, respectively (Fig. 4). These new data acquired in the SW part of the Alpine range are consistent with data collected in other parts of the Alpine belt (Brocard et al., 2003; Montgomery and Korup, 2010; Valla et al., 2010), which all large incision rates the Late Quaternary. These results obtained

Conclusion

We sampled and measured in situ-produced 36Cl concentration in 12 samples along a ~15 m high river polished cliff in a thick, homogeneous limestone layer and on the bedding surface at the top of it. Samples from the bedding surface and from the cliff provide consistent age/height relationships, allowing us to precise a geometrical and chronological model for the formation of a rectangular fluvial drainage pattern. According to this model (Fig. 1) the bedding-parallel part of the channel

Acknowledgments

This study has been funded by the French CNRS INSU Syster and Aleas Research programs and by the UCA-IDEX RiTMICA project. We are grateful to Martin Stokes, and anonymous reviewer and to Editor Markus Stoffel for their insightful comments.

References (30)

  • C.R. Twidale

    River patterns and their meaning

    Earth Sci. Rev.

    (2004)
  • L. Anton et al.

    Exceptional river gorge formation from unexceptional floods

    Nat. Commun.

    (2015)
  • V. Bauve et al.

    Pliocene to Quaternary deformation in the Var Basin (Nice, SE France) and its interpretation in terms of “slow-active” faulting

    Swiss J. Geosci.

    (2012)
  • F. Bigot-Cormier et al.

    Déformations pliocènes de la marge nord-Ligure (France): les conséquences d' un chevauchement crustal sud-alpin

    Bull. Soc. Geol. Fr.

    (2004)
  • G. Clauzon et al.

    Alternate interpretation of the Messinian salinity crisis: controversy resolved?

    Geology

    (1996)
  • Cited by (7)

    • Fluvial bedrock gorges as markers for Late-Quaternary tectonic and climatic forcing in the Southwestern Alps

      2022, Geomorphology
      Citation Excerpt :

      The discrepancy between these studies and ours might be the result of transient sediment storage, either on land, at the mouth of the Var River, or on the submarine slope (Petit et al., 2022). Trend 2 (orange lines, Fig. 7A), which includes data from Esteron (Petit et al., 2019), Bévéra and Pérouré (this study) gorges, shows incision rates of ~0.5 mm/yr that increase up to ~1 mm/yr after 10 ka. While we see the influence of deglaciation in rivers that are directly connected to formerly-glaciated areas, we can observe an acceleration of incision after 10 ka in rivers that are tributaries of the “connected” rivers.

    • Interplay of fluvial incision and rockfalls in shaping periglacial mountain gorges

      2021, Geomorphology
      Citation Excerpt :

      These observations therefore suggest that the CdB Gorge formation results from a vertical fluvial incision process, and is later widened by glacial and postglacial processes related to climate phases. Several studies based on CRE dating have been carried out on gorges and valleys of the SW Alps (Saillard et al., 2014; Rolland et al., 2017; Petit et al., 2019; Fig. 7). The comparison of fluvial incision at the scale of SW Alps from gorges formed in valleys with or without glacial influence shows the dominance of high and variable incision rates in formerly glaciated catchments, while catchments devoid of glacial influence have lower incision rates since the LGM.

    View all citing articles on Scopus
    View full text