Navigation – Plan du site

Optically stimulated luminescence dating: procedures and applications to geomorphological research in France

Les datations par luminescence optique : procédures et applications à la recherche géomorphologique en France
Stéphane Cordier
p. 21-40

Résumés

L’objectif de cet article est de proposer une présentation d’ensemble des méthodes de datations par luminescence stimulée optiquement (OSL) et de leurs applications dans le champ des recherches en géomorphologie. Après une présentation des principes physiques généraux, les procédures mises en œuvre aussi bien sur le terrain qu’en laboratoire sont abordées, à partir de l’exemple des analyses réalisées sur des alluvions sableuses issues d’une basse terrasse de la Moselle. La place des datations OSL dans la recherche géomorphologique en France et son potentiel pour les recherches futures sont décrits à travers une présentation de la diversité des environnements sédimentaires sensu lato et des problématiques pouvant être traités. Ainsi, l’article souligne l’importance de la méthode pour les recherches en géomorphologie, notamment dans le cadre du développement de la géomorphologie quantitative.

Haut de page

Notes de la rédaction

Article soumis le 13 juin 2009, accepté le 18 décembre 2009

Texte intégral

Introduction

1Absolute dating methods have been developed over the last five decades (Jull and Scott, 2007). They are now largely used to date not only palaeontological or organic remains, but also minerals that characterise detrital clastic sedimentary material. The most common methods applied to minerals are cosmogenic radionuclides, electron spin resonance (ESR) and luminescence techniques. The latter were first applied to burned minerals from archaeological artefacts [thermoluminescence (TL) method]. Improvements of this technique led to the development, for more than twenty years, of the optical dating method [commonly referred to as Optically Stimuled Luminescence (OSL)] which is now applied to sediments from various origins (Wintle, 2008). The aim of this paper is to provide people involved in geomorphological research a global overview about the principles and procedures of optical dating, from the field sampling to the age interpretation. Most of the publications actually focus on one part of either the method (e.g., laboratory procedures, statistical treatment) or its applications, and available general literature for non-specialists is still uncommon. The general principles of the method are described first. The paper then explains how OSL dating is applied to obtain a depositional age, through the field and laboratory procedures employed. These procedures are described as clearly as possible in order to provide useful information for geomorphologists interested in the method, and illustrated by a case study that has involved luminescence dating of fluvial sands (samples LUM 975 and LUM 978) from the lower alluvial terrace of the Moselle River (M1 terrace as defined by S. Cordier et al., 2006). However, it is essential to keep in mind that the protocols (preparation of the sediments, measurements) may vary from one laboratory to the other: the presentation does not aim to be exhaustive, but to reference the main procedures at each step of the dating. Finally the paper reviews the recent applications of OSL dating in France, and assesses the potential of applying the luminescence dating technique to a range of geomorphic research.

General physical principles

2Luminescence methods (TL and OSL) are based on the estimation of the impact of radiation on the crystalline structure of minerals while they are shielded from light (Aitken, 1985; Wintle, 1997; Aitken, 1998; Duller, 2004; Vandenberghe, 2004). The main minerals studied are quartz and K-rich feldspar, which can be found in almost all sedimentary environments. The radiation (α, β, γ) comes from radionucleides which are present in the mineral and its natural environment, mainly uranium, thorium (and their decay products), potassium, and for a small proportion from cosmic particles (Aitken, 1985). They lead to the emission of electrons which are subsequently trapped in crystalline lattice defects. Some of the traps are considered ‘unstable’ (“shallow traps”), which means that an electron inside will not remain trapped for the whole duration of burial. On the contrary, defects situated deeper inside the lattice have a higher thermal lifetime. These “deep traps” (stable traps associated with high energy levels) can adequately be used for dating. The total amount of trapped electrons within a crystal is proportional to the total energy absorbed and retained by the crystal (or dose), hence the time it was exposed to radiation. As soon as the mineral is exposed to sunlight, for example during its transport, trapped electrons absorb the photon energy (from the Sun), and are released from the traps.

3The rate of release depends on four main parameters: i) the kind of mineral: the eviction occurs faster for quartz than for feldspars (reduction of the luminescence signal by a factor of 100 in ~ 20 s for the former, but a few minutes for the latter; Godfrey-Smith et al., 1988; Mercier, 2008); ii) the intensity of the light flux; iii) the wavelength of the stimulating light: quartz is especially sensitive to short wavelengths from the visible spectrum (e.g., UV, blue, green), i.e., these are more efficient to stimulate and release the electrons. Feldspars have the specificity of being sensitive both to short and to near-infrared or infrared wavelength (800-950 nm; Bøtter-Jensen et al., 1994); iv) the sensitivity of the trap to light. Released electrons can recombine with another kind of crystalline defects (“holes” reflecting electrons vacancies). The recombination generates the emission of light (the luminescence signal) which can be measured in laboratory through heating (for TL) or through light stimulation for OSL (Huntley et al., 1985). The intensity of the signal is proportional to the amount of released electrons. The wavelength of the signal is allocated to the nature of the mineral: the OSL from quartz is typically measured in ultra-violet (340-370 nm wavelength), while quartz also emits in blue (460-500 nm wavelength) and in orange-red (600-650 nm wavelength; Huntley et al., 1991). These latter wavelengths are however less used to avoid interactions with the stimulating light. Feldspars mainly emit in the 330-550 nm wavelength (Huntley et al., 1991; Stokes, 1999) and are typically measured in the 400-550 nm wavelength (blue light).

4Based upon these physical principles it is possible to recognise the importance of luminescence dating for sediments (fig. 1): i) prior to sediment transport (e.g., when the mineral is stored within the host-rock), natural radiation generates the trapping of electrons and the build-up of a latent luminescence signal; ii) when a grain is produced by mechanical erosion and transported (at the Earth’s surface, in the air, or in a river), it is exposed to sunlight. This results in the release of trapped electrons (“bleaching”) and the emission of the luminescence signal. This phenomenon is associated with a resetting (“zeroing”) of the dosimetric clock; iii) after one or several transport phases the grain is “definitively” buried under a sedimentary cover. It is exposed again to radiation and accumulates trapped electrons. Much later the grain is sampled in the field and stimulated in the laboratory using either visible light for quartz (OSL stricto sensu, typically performed with blue or green LEDs), or infrared light for feldspars (Infrared Stimulated Luminescence IRSL).

Fig. 1 – Schematic representation of the history of a mineral from the host-rock to the depositional event and its dating in the laboratory.
Fig. 1 Représentation schématique de l’histoire d’un minéral depuis la roche mère jusqu’à son dépôt dans une formation sédimentaire et la datation de ce dépôt en laboratoire.

Fig. 1 – Schematic representation of the history of a mineral from the host-rock to the depositional event and its dating in the laboratory.Fig. 1 – Représentation schématique de l’histoire d’un minéral depuis la roche mère jusqu’à son dépôt dans une formation sédimentaire et la datation de ce dépôt en laboratoire.

5As shown in fig. 2, the luminescence signal obtained by stimulation corresponds with a decay curve which shows the progressive emptying of the electrons from the traps. The observed OSL signal from quartz is however made up of several exponentials related to different levels of traps (Bailey et al., 1997; Lian and Roberts, 2006). This reflects the existence of a fast component (associated with the emptying of the most light-sensitive traps), a medium component, and several slow components (corresponding with less bleachable traps; Bailey, 2000). The analysis of the OSL signal makes it possible to estimate the time elapsed since burial, i.e., the age of the sediment. This age is derived using the following equation:

Age (in a) = P (in grays Gy or J/kg) / Dr (in Gy/a) (1)

6P is the palaeodose (the total amount of dose absorbed by the mineral throughout its burial). It is estimated in the laboratory by the determination of the equivalent dose De, i.e., the artificial dose which is necessary to induce a luminescence signal similar to the natural one. Dr is the dose rate (the rate at which the sediment is exposed to natural radiations). However, the measured signal does not always reflect the time elapsed since the burial of the sediment (fig. 1): i) if the grain is not completely bleached during transport (e.g., because of mass transport, or transport in turbid water), the measured signal includes a part of the initial (non bleached) signal, leading to an overestimate of the equivalent dose, and consequently of the age (Duller, 2008); ii) after a very long burial every crystalline defect will be filled by electrons, thus preventing additional trapping although the mineral still absorbs radiation. The mineral becomes saturated, and its equivalent dose will consequently underestimate the event of interest. Quartz generally saturates at lower doses than feldspars. Therefore it is possible to date feldspars that are several hundreds of thousands  of years old (Juschus et al., 2007; Martins et al., in press), with attention paid to the “anomalous fading” (see below). On the contrary, it is difficult to date quartz beyond 100-150 ka (except when the dose rate is low, i.e., less than 1 Gy/ka; Huntley et al., 1993). Recent research focusing on the recuperated signal of quartz (Re-OSL) however opens the way to overcome satisfactorily the problem of quartz saturation (Wang et al., 2006; Wintle, 2008); iii) feldspars are affected by anomalous fading (Wintle, 1973; Huntley and Lamothe, 2001). This phenomenon, which might be explained by a quantum mechanic tunnelling process, is associated with a spontaneous eviction of electrons in deep traps, at room temperature, and without exposure to light. Anomalous fading generates an underestimate of the equivalent dose, and consequently of the burial age estimate. It is highly recommended to apply measurement procedures in the laboratory to detect the fading. A successful mathematical model was proposed by D.J. Huntley and M. Lamothe (2001) to compensate for anomalous fading and get more reliable age estimates (see also M. Auclair et al., 2003).

Fig. 2 – Example of decay curves (quartz sands from sample LUM 978) showing the progressive eviction of the electrons from their traps during the first seconds of optical stimulation and the correlative emission of photons associated either with the natural signal or with artificial signals obtained for various irradiation doses. This decay is not a single exponential curve as several levels of traps are involved, some of them being more rapidly bleached (fast component) than the others (medium and slow components).
Fig. 2 Exemple de courbes (sables quartzeux issus de l’échantillon LUM 978) montrant la libération progressive des électrons dans les premières secondes de la stimulation optique. Les signaux OSL correspondent d’une part au signal naturel, d’autre part aux signaux obtenus en laboratoire après irradiation à des doses variables. La courbe ne correspond pas à une simple décroissance exponentielle de la libération des électrons en fonction du temps car plusieurs niveaux de piège interviennent, certains étant plus rapidement blanchis que d’autres.

Fig. 2 – Example of decay curves (quartz sands from sample LUM 978) showing the progressive eviction of the electrons from their traps during the first seconds of optical stimulation and the correlative emission of photons associated either with the natural signal or with artificial signals obtained for various irradiation doses. This decay is not a single exponential curve as several levels of traps are involved, some of them being more rapidly bleached (fast component) than the others (medium and slow components).Fig. 2 – Exemple de courbes (sables quartzeux issus de l’échantillon LUM 978) montrant la libération progressive des électrons dans les premières secondes de la stimulation optique. Les signaux OSL correspondent d’une part au signal naturel, d’autre part aux signaux obtenus en laboratoire après irradiation à des doses variables. La courbe ne correspond pas à une simple décroissance exponentielle de la libération des électrons en fonction du temps car plusieurs niveaux de piège interviennent, certains étant plus rapidement blanchis que d’autres.

  

Field procedures

7Two grain-sizes of sediment are commonly used for optical dating: the “coarse grain analysis” uses fine to medium sand (90-250 m); the “fine grain analysis” uses fine silt (4-11 m). The choice of these sizes is justified by the procedures used for the dose rate determination (see next section). The best sampling strategies are those that involve communication between a geomorphologist and luminescence dating expert prior to and during the field sampling process. The sampling is best undertaken from a pit (or a natural section), or using cores. In a pit, samples should be taken from an homogeneous layer more than 30-40 cm thick, away from coarse deposits, which may complicate the dose rate calculation. Sampling should also be avoided for sedimentary layers affected by bioturbation or pedogenic activity, as it may generate in situ bleaching and variations in the dose rate through time (fig. 3). The sampled sediment should be cleaned first by removing the outermost 10 cm. A ~ 10 cm diameter metal or PVC tube (so that the sample is not exposed to sunlight during sampling) is then forced into the sediment, removed, capped, and if possible put into a black plastic bag. It is possible to operate by night (especially for consistent and hardened sediments which can be dug with a shovel) but also by day since the outer parts of the sample will be removed before the measurements. Once the sample has been removed the remaining hole can be used for an in situ determination of the dose rate (see below). The dose rate may also be determined in the laboratory. This is achieved by taking a representative sample (typically ± 700 g for coarse grain, 50 g for fine grain) of the same material, in close proximity to the luminescence tube. When the profile presents several sedimentary units, further samples should also be taken so that the luminescence ages can be compared. This comparison is however not sufficient to assess the accuracy of the age, which can be only obtained using an independent technique (see below). A second possibility is to collect luminescence samples from sedimentary cores. It is essential to protect from light the sediment inside the core, until it is opened in laboratory under subdued light conditions. The sedimentary layering within the core should be checked to ensure that sedimentary horizons have not been disturbed during the coring process. As for other analyses it is especially important not to sample the area between different cores junctions. A part of the sample should also be used for dose rate estimates.

Fig. 3 – Location of the case study.
Fig. 3 Localisation du site étudié.

Fig. 3 – Location of the case study.Fig. 3 – Localisation du site étudié.

A: General map of the study area. B: Map of the Wintrange basin and location of the gravel-pit Rem-VI. C: Log of the studied profiles. D: Sampling of LUM 978.
A : Carte générale du secteur d’étude. B : Carte du bassin de Wintrange montrant l’emplacement de la carrière Rem VI. C : Log des profils étudiés. D : Prélèvement de l’échantillon LUM 978.

Determination of the dose rate

8For both pit and core sampling, the coordinates (x, y, z) and depth of the sample have to be measured to determine the dose rate. Evaluation of the time-averaged water content present in the sediment being sampled throughout its burial is also an important, yet difficult parameter to assess. Water actually absorbs part of the radiation, hence modifying the dose rate and the age estimate of the sample: in the case study (sample LUM 975 and LUM 978), an increase of 10% of the water content changes the OSL age by a similar proportion. Field observations for water content by a geomorphologist may therefore be useful, e.g., to recognise groundwater level changes which may be highlighted by manganese or iron oxides levels. Good estimates of modern moisture content can also be achieved by comparing the weight of the dried sample with that of the natural sample. However, the value is typically not indicative of the entire period of burial, especially if the sediment is thousands of years old. It is also important to mention that the effective dose rate varies with the grain size. The fine grains receive the full range of energy from , and radiations (Wintle 2008). In contrast, particles (penetrating range: 25 m) are absorbed in the outer part of coarse grains, while particles (range: 2 mm) as well as rays (range: 50 cm) are only attenuated. The dose rate may be measured in situ using a radiation dosimeter or a field gamma spectrometer (Aitken, 1998; Hossain, 2003). It may also be evaluated in the laboratory from the representative sediment sampled close to the luminescence one. The fundamental assumption is that the dose rate remained constant during the whole burial period, i.e., the system was ‘closed’. This assumption can be considered as valid if the decay series show a radioactive equilibrium, i.e., all the radionuclides present the same activity (this activity being a function of the number of nuclei and the half-life of each nuclide; see details in S.M. Hossain, 2003). The contents of radionuclides within the sediment as well as information about their decay behaviour can be obtained by high resolution gamma spectroscopy. In that case the sediment has to be dried, homogenised by grinding, cast inside an air-tight container (to prevent from the loss of 222Rn by gaseous diffusion) and stored for one month. Such a storage is required to establish a radioactive equilibrium between 222Rn and 226Ra. Then the activity of several key radionuclides (for example 234Th, 214Bi, 214Pb and 210Pb for the 238U chain) is measured during at least 24 h. It is shown by the spectrum of energy (in keV) of gamma emissions and must be equal for all the measured radionuclides to assume a radioactive equilibrium. This activity is proportional to the dose rate which can then be calculated using conversion factors (Adamiec and Aitken, 1998). Other main methods for dose rate calculation include the thick source alpha counting (TSAC), which consists of counting the alpha particles emitted by the U and Th decay series; the GM-beta counting to measure the beta dose rate from decay of U, Th and K; and the instrumental neutron activation analysis (INAA) which consists of irradiating the sample with neutrons prior to measuring the rate at which gamma-rays are emitted from an element, this rate being proportional to its concentration (Hossain, 2003). It is important to note that most of these determinations are indirect, i.e., the , and doses are not measured directly but derived from the radionuclide concentrations. In the same way the cosmic dose is calculated using mathematical equations (Prescott and Hutton, 1994).

Laboratory preparations for the luminescence measurements

9While the preparation for the dose rate is made under normal light, sample preparations for the luminescence measurements must be performed in the laboratory under subdued red or orange light, to allow the manipulation of sediments without affecting the latent luminescence. After the removal of the top and the bottom of the sampling tube (since both parts were exposed to light) or sampling in the core, the sediment is sieved in order to select a representative grain size to be studied (fig. 4). As the energy absorbed by a grain throughout burial highly depends on its size, it is important that the grain size is as homogeneous as possible. For the case study presented here a grain size of 100-150 m was isolated. We subsequently exposed this grain size fraction to HCl, sodium-oxalate, and H2O2 to dissolve or remove carbonates, aggregates and organic materials, respectively. Following these treatments, heavy liquid solutions were used to separate plagioclase, heavy minerals, quartz, and K-rich feldspars, the latter two minerals being commonly employed in OSL dating. Coarse quartz grains were then treated with concentrated HF to dissolve feldspar grains and to etch the outer layer of the grains (± 10 m). Since alpha radiation is less efficient and not reproduced in the laboratory, etching is important to reduce the outer part of the grains which absorbs the α particles. Finally the purified quartz was sieved again to remove grains whose diameter was significantly reduced by the etching. Heavy liquid solutions are less efficient for fine grain sediments (4-11 m), and strong etching may obviously not be performed, as it would etch the whole grain. Alternative chemical treatments exist (Roberts, 2007), for example using fluorosilisic acid (H2SiF6). Another solution is to analyse ‘polymineralic samples’ (i.e., with quartz and feldspars being not separated) and to stimulate them first with IR light (to reduce the signal from feldspars) prior to the measurement (with blue LEDs) of an OSL signal mainly derived from quartz grains (double SAR protocol; Banerjee et al., 2001). After the chemical preparation, the quartz and feldspar grains were mounted and stuck on 10 mm diameter discs in order to obtain several subsamples (termed “aliquots”). The number of grains is highly variable: “large” and “medium” aliquots contain several thousands of sand-sized grains and several hundreds of sand-sized grains, respectively. In case of partial bleaching, even fewer grains can be used, corresponding either with small aliquots (several tens of sand-sized grains) or with single-grain aliquots. Single grain aliquots can also be prepared with a purposed-built single-grain disc (which is made up of a regular array of 100 holes, with one grain per hole; Duller, 2008). The size of the aliquot is actually an important parameter (Duller, 2008): medium or large aliquots may lead to an averaging of the palaeodose estimate in case of partial bleaching, since most of the aliquots will be likely to include both well bleached and poorly bleached grains. They are however successfully employed for fully bleached sediments. On the contrary, measuring an insufficient number of grains may lead to the obtaining of too weak a luminescence signal, influenced also by the background noise of the measuring system. Despite significant improvements (Rittenour, 2008), the use of single grain aliquots is still subject to ongoing debates: on the one hand, single grain dating provides a more precise equivalent dose and allows for partial bleaching to be detected. On the other hand, since only a very few quartz grains emit luminescence, up to several hundreds of grains have sometimes to be measured to record a sufficient OSL signal. The use of single grain aliquots is difficult when the exposure to sunlight is unequal from grain to grain (e.g., with mass-transport or fluvial sediments), as the average value is unreliable: statistical models have thus to be applied in the hope of isolating a population of grains that were fully zeroed prior to burial, which further limits its routine application. S.B. DeLong and L.J. Arnold (2007) found significant differences in age estimates between small aliquots and single grains for only very young samples (less than 1 ka), the results being very similar for older sediments. It is finally important to note that the choice of aliquot size depends both upon the depositional environment to be dated (e.g., sediments fully or partially bleached) and the age (since older material will be able to provide a brightest luminescence signal).

Fig. 4 – Schematic diagram showing the laboratory procedures for the equivalent dose and dose rate determinations.
Fig. 4 Schéma décrivant le protocole opératoire de préparation des échantillons en vue de la détermination de la dose équivalente et du débit de dose.

Fig. 4 – Schematic diagram showing the laboratory procedures for the equivalent dose and dose rate determinations.Fig. 4 – Schéma décrivant le protocole opératoire de préparation des échantillons en vue de la détermination de la dose équivalente et du débit de dose.

  

The SAR procedure

10Several techniques were developed over the last two decades to measure the equivalent dose from the OSL and IRSL signals (Vandenberghe, 2004). Two types of procedures can be distinguished: additive and regenerative. The additive procedures (such as the “Multiple Aliquot Additive Dose” procedure, MAAD) derive from the TL techniques and were largely employed during the 1990s. However, their precision is restricted due to methodological limits (the use of several aliquots to obtain only one equivalent dose, obtained by extrapolation of the natural signal on a curve derived from various laboratory doses). On the contrary, the regenerative procedures provide more precise ages as they make it possible to obtain by interpolation a similar number of equivalent dose measurements to aliquot measurements. The main procedure which revolutionised optical dating is the “Single Aliquot Regenerative” (SAR) protocol. It was recently developed for quartz grains (Murray and Wintle, 2000 and 2003) and applied also to feldspars (Wallinga et al., 2000). The SAR procedure is now a standard protocol used in luminescence dating (Wintle and Murray, 2006). We consequently describe in the present paper its general principles and its application to the case-study samples (LUM 975 and LUM 978).

11The SAR protocol (as for any OSL measurement) is commonly performed using a Risø system (fig. 5). It includes the measurement of several OSL signals for a single aliquot (tab. 1). The signals correspond first to the natural signal, then to signals obtained after artificial irradiation of the aliquot at determined doses (termed “regenerative doses”). Basically the comparison of the natural signal with the artificial luminescence signals should make it possible to interpolate the palaeodose directly. The procedure is however more complicated, as artificial irradiation leads to the trapping of electrons in unstable traps of the mineral crystal lattice. Since the measurement of the luminescence signal occurs immediately after irradiation, these electrons are very likely to be released during the optical stimulation, leading to an overestimate of the signal. The unstable traps must consequently be emptied before the measurement. This is achieved by a preliminary heating (“preheat”) of the aliquot. The temperature of the preheat has to be chosen by a preliminary test that requires exposing the aliquots to different preheat temperatures or by exposing the aliquots to a given temperature during various time-spans.

Fig. 5 – The system used to perform the SAR protocol.
Fig. 5 Vue d’ensemble du lecteur utilisé pour la réalisation du protocol SAR.

Fig. 5 – The system used to perform the SAR protocol.Fig. 5 – Vue d’ensemble du lecteur utilisé pour la réalisation du protocol SAR.

The aliquots are placed on a wheel in the reader (black box on the right). The black tube above the reader is the photomultiplier tube used to record the photons emitted by quartz and feldspar during a stimulation (i.e., the luminescence signal). The stimulation sources (blue and IR LEDs) are nested below the photomultiplier tube. The heater element is located in the lower part of the reader. The grey cylinder is the irradiator accommodating a 90Sr/90Y source. In the middle of the picture, the controller is used to select the parameters (OSL or IRSL stimulation, etc.). The results can be seen on the screen left of the picture (photo: T. Lauer).
Les aliquotes sont placées sur une roue insérée dans le lecteur stricto sensu, à droite de l’image. Au-dessus du lecteur se trouve le tube photomultiplicateur (en noir) utilisé pour la détection du signal luminescent. Les diodes utilisées pour la stimulation (bleues ou infra-rouges) se trouvent près de la base du tube. Le cylindre gris correspond à la source irradiante (irradiation obtenue à partir d’une source 90Sr/90Y). Au centre de l’image, le contrôleur permet de sélectionner notamment le type de stimulation (OSL ou IRSL). Les résultats peuvent être observés sur l’écran à gauche (photo : T. Lauer).

Tab. 1 – The SAR protocol performed for the fluvial sands from the Luxembourgian Moselle.
Tab. 1 Le protocole SAR appliqué dans l’étude des sables de la Moselle luxembourgeoise.

The SAR protocol applied to coarse-grain quartz
1. Preheat at 260°C for 10 s, heating rate 5°C/s.
2. Blue-LED stimulation at 125°C for 40 s.
3. Give test-dose = 100 s.
4. Cut-heat at 200°C for 0 s, heating rate 5°C/s.
5. Blue-LED stimulation at 125°C for 40 s.
6. Blue-LED stimulation at 280°C for 40 s.
7. Give various regenerative doses (300 s, 600 s, 900 s and 1200 s) and repeat step 1 to 6.
Checking of recuperation and recycling:8. Give a regenerative dose = 0 s and repeat step 1 to 6.
9. Give a repeated regenerative dose and repeat step 1 to 6.
Checking of feldspar contamination:
10. Give the same regenerative dose 11. Preheat at 240°C for 10 s, heating rate 5°C/s.
12. IR diodes stimulation at 125°C for 100 s.
13. Repeat step 2 to 5.

The SAR protocol applied to coarse-grain feldspars
1. Preheat at 260°C for 10 s, heating rate 5°C/s.
2. IR stimulation at 125°C for 100 s.
3. Give test-dose = 100 s.
4. Cut-heat at 260°C for 0 s, heating rate 5°C/s.
5. IR stimulation at 125°C for 100 s.
6. Give various regenerative doses (300 s, 600 s, 900 s and 1200 s) and repeat step 1 to 6.
Checking of recuperation and recycling:
7. Give a regenerative dose = 0 s and repeat step 1 to 5.
8. Give a repeated regenerative dose and repeat step 1 to 5.

12These procedures allow the recognition of a “plateau” (fig. 6), corresponding with the range of temperatures (or time-spans) for which the luminescence per unit of dose remains constant, i.e., there is no significant sensitivity change. Laboratory treatments may actually induce sensitivity changes in the grains, which means that the signal obtained after a subsequent stimulation may be affected by the preheat and irradiation conditions. To monitor this the SAR protocol includes the administration of a constant “test-dose”, followed by a short heating (“cutheat”) to deplete electrons in shallow traps, but without inducing a sensitivity change in the grains (Mercier, 2008), prior to the measurement of the induced signal (“test signal”). At this step, the SAR protocol can be summarised as follows (tab. 1): i) preheat and measurement of the natural luminescence signal (Ln); ii) administration of a test-dose, heating, and measurement of the induced signal (Tn); iii) several cycles corresponding with irradiations at various regenerative doses, leading to the measurement of several artificial luminescence signals (Lx). Each measurement is followed by the administration of the test dose and measurement of the test signal (Tx). The equivalent dose is then calculated from the normalised signals (Lx/Tx), by interpolating the Ln/Tn ratio on the growth curve formed with the Lx/Tx ratios (fig. 7).

Fig. 6 – Results of the preheat test performed for sample LUM 978.
Fig. 6 Résultats du test de préchauffe réalisé sur l’échantillon LUM 978.

Fig. 6 – Results of the preheat test performed for sample LUM 978.Fig. 6 – Résultats du test de préchauffe réalisé sur l’échantillon LUM 978.

The preheat test show the presence of a small plateau between 240°C and 260°C, the former being used for subsequent measurements. The recycling ratio is constantly close to 1, and the recuperation always lower than 3%.
Ce test met en évidence l’existence d’un plateau entre 240°C et 260°C, la première température étant retenue pour les mesures ultérieures. Le « rapport de recyclage » (recycling ratio) montre des valeurs systématiquement proches de 1, la récupération étant toujours inférieure à 3 %.

Fig. 7 – Example of palaeodose determination using the SAR protocol for the same aliquot as in fig. 2.
Fig. 7 Détermination de la paléodose à partir du protocole SAR pour la même aliquote que celle présentée fig. 2.

Fig. 7 – Example of palaeodose determination using the SAR protocol for the same aliquot as in fig. 2.Fig. 7 – Détermination de la paléodose à partir du protocole SAR pour la même aliquote que celle présentée fig. 2.

The y-axis shows the ratios obtained for each of these artificial doses (Lx/Tx) and for the natural sample (Ln/Tn). The palaeodose value for this aliquot from sample LUM 978 is 69.3 ± 4.3 Gy.
L’axe des ordonnées montre les ratios obtenus pour chacune de ces doses (Lx/Tx) ainsi que pour le signal naturel (Ln/Tn). La paléodose pour cette aliquote de l’échantillon LUM 978 est de 69,3 ± 4,3 Gy.

13The SAR protocol must however include complementary tests to validate the results. These tests may vary from one laboratory to the other but basically correspond with: i) a repeat of the very first regenerative dose which was given after the Ln and Tn measurement. The ratio between the normalised signal and that obtained after the first cycle (or “recycling ratio”) has to be closed to 1 (with a common tolerance of ± 10%; Murray and Wintle, 2000). This shows that sensitivity changes were corrected using the test dose. A recycling ratio significantly different with 1 means that for a similar dose the two signals are not the same: the aliquot is consequently discarded for the equivalent dose determination; ii) a luminescence measurement for a regenerative dose equal to zero. If the normalised signal is theoretically equal to zero, a weak signal is often induced by the transfer of electrons during the preheat process. This recuperated signal must be small compared to the natural one (with a tolerance of 5%; Murray and Wintle, 2000; Roberts, 2008), otherwise it means that this thermal transfer is too high and the aliquot has to be rejected; iii) a measurement of feldspar contamination, performed on OSL measurements of quartz. This detection is undertaken using infrared diodes. This test is important because feldspars are not only stimulated by infrared light, but also by the blue or green light used for quartz. Hence, the presence of high and the aliques the (Aitken, 1998; H doshes to record from quartz. An aliquot with a high feldspar contamination (beyond a 10% tolerance; Duller, 2003) is systematically discarded from the subsequent calculations. A similar test is not necessary when the analyses focus on feldspar grains, because the quartz grains which may be present in the aliquots are insensitive to infrared stimulation; iv) a measurement of anomalous fading for feldspar. This test may be performed using a SAR protocol including variable delays between irradiation and measurement of the signal to estimate the fading to be estimated. Accurate ages are then obtained by inserting this fading in a correction model (Huntley and Lamothe, 2001; Auclair et al., 2003).

Determination of the equivalent dose (De) using statistical models

14The SAR protocol creates as many equivalent doses as aliquots, with the exception of those which had to be discarded after the tests. In the case of aeolian sediments, all of the analysed grains are assumed to be well bleached, and all the Dehave a similar value, which can be used to calculate the age of the sediment. However, partial (or incomplete) bleaching is common, especially if the transport history was short or the exposure to sunlight was insufficient, as can be the case for fluvial sediments. This partial bleaching can be homogeneous (all the grains being incompletely bleached in the same proportion) or heterogeneous (differential bleaching). In this latter case the Dedistribution shows a scattering (fig. 8). Some aliquots can present a very high palaeodose, which greatly overestimate the age of the last transport event. This explains why the mean is not appropriate in estimating the accurate equivalent dose. It is therefore necessary to use a statistical model. Several models have recently been developed. The “Central Age Model” (CAM) calculates the weighted mean but takes into account the additional dispersion resulting in the measurement uncertainties (Galbraith et al., 1999). It will also overestimate the equivalent dose in the presence of a partially bleached sediment. In contrast, the “Minimum Age Model” (Galbraith and Laslett, 1993) makes it possible to identify the most bleached aliquots, while the “Finite Mixture Model” (Galbraith and Green, 1990; Galbraith, 2005) is useful when the observed distributions of equivalent doses present discrete populations. As for the sampling strategy the choice of the model depends upon the kind of sediments and presupposes a discussion between the field and luminescence specialists (Bailey and Arnold, 2006). Comparison with independent age control may also be very useful, as shown by H. Rodnight et al. (2006) for fluvial sediments. The relevance of these models increases with the number of aliquots. The number of 50 aliquots is sometimes considered as a minimal value to ensure a reliable equivalent dose determination (Rodnight, 2008), but it is important to keep in mind that the number of aliquots to be measured depends on the sample and increases with the scattering.

Applications and place of OSL in geomorphological research in France

15The physical principles of the optical dating method, and its reliability for quartz and for feldspars from silty to sandy sediments, have resulted in optical dating being applied to a diverse range of sedimentary environments, as described in several journal papers (see for example special issue of Boreas 1, 2008). The aim of this section is to review the applied representative studies dealing with OSL in France. As in other countries, the first dating of sediments was based on thermoluminescence (Wintle et al., 1984; Balescu et al., 1988 and 1992; Rousseau et al., 1998). While OSL developed as an improvement of TL, some studies combined TL and OSL dating on the same material (Balescu and Lamothe, 1994; Balescu et al., 1997; Antoine et al., 2003), before the latter became more widely employed. The first OSL applications (tab. 2) focused on aeolian and coastal sediments, which are likely to be fully bleached prior to deposition and cannot be easily dated by other methods (Jacobs, 2008; Roberts, 2008; Singhvi and Porat, 2008). Loess deposits were successfully dated especially in NW France. Several loess-palaeosol sequences (Engelmann et al., 1999; Folz, 2000; Antoine et al., 2003; Bahain et al., in press; Cliquet et al., 2009b; Guette-Marsac et al., 2009) were dated in the Paris basin, confirming the occurrence of several sedimentation periods linked to Pleistocene cold stages and enabling supraregional correlations. Most of the research focused on the last interglacial-glacial cycle (Antoine et al., 1999), integrating the dating of interglacial formations (Antoine et al., 2006). Sequences covering longer periods (up to MIS 12) were dated either using IRSL in Normandy (Cliquet et al., 2009a) or using the Re-OSL protocol in the middle Palaeolithic site of Bonneval (western Paris basin; Sun et al, in press). Coastal sands from the North Sea or Channel coastlines were also optically dated for more than one decade. The dating of raised beaches (Balescu et al, 1997; Regnault et al, 2003; Coutard et al., 2006; Cliquet et al., 2009b) or coastal dunes (Folz, 2000; Van Vliet-Lanoë et al., 2006) led to the allocation of their formation to interglacials MIS 7 or 5, enabling a reconstruction of the coastal dynamics during the last glacial-interglacial cycles. At the same time the dating of young Holocene dunes (Clarke et al., 1999 and 2002) allowed establishment of a connection between aeolian dynamics and the frequency of storms along the Aquitaine coast. The first OSL dating of fluvial sediments from French rivers also started at the end of the 1990s, despite the problem of differential bleaching being for a long time considered as a major hindrance in applying luminescence dating to fluvial sediments (Wallinga, 2002). Improvements in the detection of the partial bleaching (e.g., use of smaller aliquots) and statistical models allowed accurate ages to be obtained (Rittenour, 2008). Fluvial deposits of the Seine River have been locally dated both in its lower reach (middle Pleistocene estuarine silts; Balescu and Lamothe, 1994; Balescu et al., 1997; tab. 2) and in an upper Palaeolithic site near Paris (Folz et al., 2001). Dating was also performed in SW France, focusing on the MIS 4 Tarn terrace near Millau and associated travertines (Vernet et al., 2008) and on the Lateglacial to Holocene alluvial floodplain of the Vézère (Mol et al., 2004). A more extensive dataset was provided for the Loire basin (Straffin et al., 1999; Colls et al., 2001; Stokes et al., 2001), enabling a reliable chronostratigraphy of the youngest fluvial sediments (from ca. 150 ka to the Middle Ages) to be proposed. The same sediments were subsequently used to improve the dating method using the far-red IRSL of high ans (Arnold et al., 2003).

Tab. 2 – Overview of OSL dating studies performed in France (readily available publications).
Tab. 2 Liste des travaux intégrant des datations OSL portant sur des sites français (sont référencées ici pour chaque site les publications les plus facilement accessibles).

Reference

Study area

Sediment

Dating technique

Age range(in ka)

Main topics

Loess and palaeosoils

Engelmann et al. (1999)

Somme valley

palaeosoil

quartz OSL

120-70

geoarchaeology, palaeoenvironmental reconstructions

Folz (2000)

Normandy and Brittany

loess

quartz OSL

25-13

geoarchaeology, palaeoenvironmental reconstructions

Van-Vliet-Lanoë et al. (2000)

NW France

loess

TL/IRSL

105-25

palaeoenvironmental reconstructions

Antoine et al. (2003)

Paris region

loess-palaeosoils

TL/IRSL

160-20

palaeoenvironmental reconstructions

Regnauld et al. (2003)

N Brittany

reworked loess

IRSL

6.4

reconstruction of coastline evolution

Balescu and Tuffreau (2004) in Bahain et al. (2009)

Somme valley

loess

IRSL

195

palaeoenvironmental reconstructions

Sun et al. (2009)

Loir valley

loess-palaeosoils

quartz Re-OSL

450-95

methodological improvements, geoarchaeology

Cliquet et al. (2009a)

Normandy

loess-palaeosoils

IRSL

475-50

palaeoenvironmental reconstructions

Cliquet et al. (2009b)

Normandy

loess

IRSL

140

palaeoenvironmental reconstructions

Guette-Marsac et al. (2009)

Normandy

loess and reworked loess

quartz OSL IRSL

45-20

geoarchaeology

Coastal sediments

Balescu et al. (1992 and 1997); Van-Vliet-Lanoë et al. (2000)

Pas-de-Calais

raised beach

IRSL/TL

230-205

internal forcing on sea-level change

Clarke et al. (1999, 2002)

Aquitaine

dune sands

IRSL

3.7 to present

relations between dune formation and storminess

Folz (2000); Van-Vliet-Lanoë et al. (2006)

Normandy

aeolian sands

quartz OSL

115-100

Geoarchaeology, palaeoenvironmental reconstructions

Regnauld et al. (2003)

N Brittany

raised beach

quartz OSL

>90 to present

reconstruction of coastline evolution

Coutard et al.

-2006

Normandy

raised beach

quartz OSL

125-110

reconstruction of coastline evolution

Cliquet et al. (2009b)

Normandy

raised beach

quartz OSL, IRSL

190-80

palaeoenvironmental reconstructions, geoarchaeology

Fluvial sediments

Balescu and Lamothe (1994); Balescu et al. (1997)

lower Seine valley

estuarine silts

IRSL

315-200

sea-level change

Folz et al. (2001)

middle Seine valley

fluvial sands

quartz OSL

~20

geoarchaeology, methodological improvements

Straffin et al. (1999); Colls et al. (2001); Stokes et al. (2001); Arnold et al. (2003)

Loire and Arroux valleys

fluvial sands

quartz OSL, IRSL

140 to present

methodological improvements, fluvial response to climate change

Cordier et al. (2005, 2006 and 2009)

Moselle and Meurthe valleys

fluvial sands

quartz OSL, IRSL

160-25

fluvial response to climate and tectonic forcing

Vernet et al. (2008)

Tarn valley

fluvial sands

quartz OSL

75

palaeoenvironmental reconstructions

Tufa and travertines

Antoine et al. (2006);

Bahain et al. (2009)

Somme valley

tufa

quartz OSL

125-90

palaeoenvironmental reconstructions
methodological improvements

Vernet et al. (2008)

Tarn valley

travertines

quartz OSL

275

palaeoenvironmental reconstructions

16A similar reconstruction of the terrace staircase chronostratigraphy was more recently developed in the Moselle basin in order to estimate the fluvial incision rates and to improve the correlations with the Rhine terrace system (Lauer et al., in press). About 20 samples from the Meurthe and Moselle fluvial terraces in France and in surrounding countries (Germany and Luxembourg) were dated (Cordier et al., 2005, 2006 and 2009), and in particular the samples LUM 975 and LUM 978 illustrating the present paper. These samples originate from a section exposing the sediments from the lower terrace M1. The section (called “Rem VI”; see details in Naton et al., 2009) is located on the Luxembourgian (western) side of the Moselle River close to the French border (fig. 3). The fluvial sedimentary sequence exposed here is typical for the M1 alluvial terrace in this area, exhibiting a succession of three units: a coarse lower unit (unit A) up to 5 m thick, a sandy unit  (unit B) 1 to 3 m thick, and an upper unit (unit C) less than 2 m thick, which corresponds to an alternation of sandy and silty thin laminae. The samples LUM 975 and 978 were taken from a sandy horizon on top of unit A and in unit B, respectively. We applied the typical SAR protocol (tab. 1) to both quartz and feldspar sand-sized grains extracted from these samples (Cordier et al., in press). The equivalent dose distributions are presented in fig. 8. They reflect incomplete bleaching of some aliquots, which is typical for fluvial sediments. In our case study, however, quartz surprisingly shows a greater scattering than feldspars. This may be explained by the presence of two distinct populations of quartz grains: one being fully bleached since it originates from the Vosges Massif (ca. 250 km upstream), and one with a proximal origin (Taunus quartzites, outcropping a few kilometers upstream from Remerschen). In contrast, all the feldspars are assumed to come from the Vosges Massif. The consequence of the scattering is that De values (fig. 8) and the derived ages are significantly different according to the statistical models, especially for quartz (tab. 3).

Fig. 8 – Palaeodose distribution for quartz and feldspars aliquots from samples LUM 975 and 978.
Fig. 8 Distribution des paléodoses pour les quartz et feldspaths des échantillons LUM 975 et 978.

Fig. 8 – Palaeodose distribution for quartz and feldspars aliquots from samples LUM 975 and 978.Fig. 8 – Distribution des paléodoses pour les quartz et feldspaths des échantillons LUM 975 et 978.

Each De value is represented by a point. The precision is indicated by the x-axis and increases to the right. The De value is plotted on the y-axis as the number of standard deviations away from a chosen central value. The presence of a majority of points within the grey band (2 band) for the feldspars from sample LUM 975 reflects that the aliquots are well bleached. In contrast, the radial plots for quartz from both samples show the presence of high palaeodoses corresponding with incompletely bleached grains.
Chaque point correspond à une valeur de De (indiquée sur l’axe des ordonnées). En abscisse est indiquée la marge d’erreur. La présence d'une majorité de points dans la bande grisée (bande des 2 ) pour les feldspaths de l'échantillon LUM 975 suggère que les aliquotes sont bien blanchies. Les diagrammes soulignent pour les quartz la présence de paléodoses très élevées reflétant un blanchiment incomplet des grains.

Tab. 3 – Estimation of the palaeodose using several statistical parameters and models. While the results are more or less similar for feldspars, the values are highly variable for quartz. Due to the fluvial origin of the studied sediments, the “Minimum Age Model” (right column) was used to calculate the age.
Tab. 3 Estimation de la paléodose à partir de plusieurs paramètres et modèles statistiques. Les valeurs obtenues sont assez homogènes pour les feldspaths, alors que des écarts importants apparaissent pour les quartz. Compte tenu du type de sédiments étudié (alluvions), le « Modèle de l’Age Minimum » (colonne de droite) a été retenu pour le calcul de l’âge du dépôt.

De (in Gy)

Sample

Number of aliquots

Mean

Median

Weighted mean

Central age model

Minimum age model

Quartz

975

55

77.2±3.4

73.5

60.4±0.5

73.3±3.0

48.6±2.8

978

35

87.0±5.9

80.1

73.9±0.8

80.9±3.1

61.9±3.5

Feldspar

975

54

54.1±0.4

53.9

53.7±0.3

54.0±0.4

52.4±1.0

978

12

61.0±0.6

61.3

60.8±0.6

60.8±1.9

60.5±1.0

17The selection of the most appropriate model was complicated by the absence of independent age control.  As for current luminescence dating of Rhine sediments where such an age control is available (Lauer et al., in press), the ages obtained using the “Minimum Age Model” are here considered as the most accurate (tab. 4), while those obtained with the latter models are interpreted as the maximum ages of the depositional event. For feldspars the age was corrected using the model of D.J. Huntley and M. Lamothe (2001) to take into consideration the anomalous fading (tab. 4). This example demonstrates that the palaeodose calculation should take into consideration the geomorphological context of the sample in order to understand the mechanisms for the scattering. Based on the Minimum Age Model, the age estimates show that the sediments from below the lower terrace M1 were deposited during the Weichselian upper Pleniglacial. This chronological framework is consistent with the age estimate for the present river floodplain M0 (Lateglacial to Holocene ages; Carcaud, 1992), and with the previous OSL dating of higher fluvial terraces of the French Moselle River and its mains tributaries:  the Meurthe River (fig. 9; Cordier et al., 2005 and 2006) and more recently the Sarre River. These results still have to be improved (especially by obtaining an independent age control). However, they emphasise the climate control on terrace formation for the Moselle River and its tributaries, each terrace formation being allocated to a glacial-interglacial cycle. They also confirm the sedimentological results, which show that the younger terraces of the rivers Meurthe-Moselle have age similarities between the Paris basin and the Rhenish Massif, suggesting tectonic stability along the river valleys during the middle and upper Pleistocene.

Fig. 9 – Chronological frame of the Meurthe-Moselle alluvial terrace system based on the optical datings. The generalised results (see also Cordier et al., 2005, 2006) demonstrate the correlation between terrace formation and Pleistocene climate cycles.
Fig. 9 Cadre chronologique des terrasses alluviales de la Meurthe-Moselle obtenu à partir des datations par luminescence optique. Ces résultats portant sur l'ensemble de la vallée (voir aussi Cordier et al., 2005, 2006) soulignent les corrélations entre la formation des terrasses alluviales et les cycles climatiques du Pléistocène.

Fig. 9 – Chronological frame of the Meurthe-Moselle alluvial terrace system based on the optical datings. The generalised results (see also Cordier et al., 2005, 2006) demonstrate the correlation between terrace formation and Pleistocene climate cycles.Fig. 9 – Cadre chronologique des terrasses alluviales de la Meurthe-Moselle obtenu à partir des datations par luminescence optique. Ces résultats portant sur l'ensemble de la vallée (voir aussi Cordier et al., 2005, 2006) soulignent les corrélations entre la formation des terrasses alluviales et les cycles climatiques du Pléistocène.

Tab. 4 – Age estimate for samples LUM 975 and LUM 978.
Tab. 4 Les résultats pour les échantillons LUM 975 et LUM 978

The results show the presence of “anomalous fading” affecting the feldspars (underestimation of the age). The fading was estimated using the model of D.J. Huntley and M. Lamothe (2001) to calculate the fading rate. This reflects the percent decrease of intensity per decade, with a decade being a factor of 10 in time since irradiation. Corrected feldspars ages are in good agreement with the quartz results for both samples.
Les résultats pour les échantillons LUM 975 et LUM 978 soulignant l’existence d’un « fading anomal » des feldspaths (âge sous-estimé). Ce fading est corrigé à partir du modèle de D.J. Huntley et M. Lamothe (2001) permettant d’évaluer le taux de « fading » (taux de réduction de l’intensité du signal en fonction du temps). Les âges ainsi corrigés pour les feldspaths sont en bonne corrélation avec ceux obtenus sur les quartz.

Sample

Number
of aliquots

De (in Gy)

Da (in Gy/ka)

Age (in ka)

Fading rate
(in %/decade)

Age after fading correction
(in ka)

Quartz

975

55

48.6±2.8

2.0±0.2

24.6±1.9

-

-

978

35

61.9±3.5

2.5±0.2

24.6±1.9

-

-

Feldspars

975

54

52.4±1.0

2.5±0.1

21.2±1.2

2.7

27.4±1.3

978

12

60.5±1.0

3.0±0.2

19.9±1.2

2.7

25.7±1.2

18As shown in this review, OSL dating is being applied in France to several kinds of sediments covering various time-spans. Loess or alluvial sediments that are several hundreds of thousands years have been dated using the IRSL or the Re-OSL technique. At a shorter time scale (e.g., several ka or tens of ka), optical dating was successfully applied, for example to reconstruct the main periods of loess accumulation or glacial fluctuations during the last climate cycle or during the Lateglacial and Holocene periods. Finally, historical phases of sediment aggradation were dated, with the accuracy of the age results being confirmed by historical archives. The method was integrated into several research topics (tab. 1), including palaeoenvironmental reconstructions, but also geoarchaeology (dating of sediments including artefacts, as abundant Palaeolithic sites lack material suitable for radiocarbon dating), and estimates of external (climate cycles, storminess), or internal (tectonic uplift) forcing. In parallel it allowed methodological improvements (e.g., application of new protocols: Sun et al., in press; test of the method on new material: Antoine et al., 2006; evaluation of the bleaching during fluvial transport: Stokes et al., 2001; and use of the far-red IRSL emission in spite of the blue IRSL: Arnold et al., 2003). However, it is worth noting that most of the data are used for local reconstructions, and that quantitative large-scale studies (e.g., sediment budgets, estimates of regional aggradation or erosion rhythms) remain uncommon. Despite OSL being applied to several kinds of sediments, it is obviously not the most used dating method in France, where the available chronological dataset is commonly based on radiocarbon (for young sediments trapping organic material), U/Th, TL, and for the Paris basin on Electron-Spin-Resonance (ESR). This might be explained by the methodological limits of the OSL method, which in particular relate to the anomalous fading of feldspars, the partial bleaching, and the age uncertainty. This latter derives from the addition of uncertainties which concern both the dose rate (e.g., water content) and the equivalent dose (e.g., error during the OSL measurement, for the growth curve fitting; Duller, 2007). The final age uncertainty typically ranges between 5% and 10%. As a consequence it is important (especially for old sediments) not to exclude it from the interpretation. For example, an age of 116 ± 10 ka cannot be allocated firmly to the Eemian interglacial without palaeoenvironmental evidence or independent age control. Despite these limits the weak number of publications including OSL dating in France should however not be interpreted in terms of the accuracy of the method, as it may be successfully compared with other common dating methods: i) the thermoluminescence methods are commonly used for archaeological or volcanic materials, especially because of the mineral composition of the studied material. For other sediments OSL dating is typically used as it focuses on the light-sensitive part of the signal (which is faster and more completely bleached) while TL also measures the non-bleachable signal (Duller, 2004); hence OSL dating makes it possible to date younger sediments, and to reduce the occurrence of partial bleaching; ii) in opposition to radiocarbon, OSL dating is applied to sediment sequences that contain no organic material. It can also be used for longer time-scales (up to several hundreds of ka), while radiocarbon cannot be used for materials older than 35-40 ka). From a geomorphological point of view optical luminescence also presents the advantage that it dates directly the deposited material; iii) cosmogenic dating methods are used principally for age estimates of surfaces (e.g., deglaciated areas or fault scarps), although it has also been successfully applied to flights of terraces (Brocard et al., 2003). Furthermore, it is more difficult to obtain several ages on a given vertical profile, while the OSL dating methods makes it possible to get a high-resolution chronological reconstruction; iv) the Electron Spin Resonance method is applicable to material that is up to several hundreds of thousands years old, but is typically applied to palaeontological remains (bones or teeth). Even if accurate ages have been obtained for aeolian or fluvial sediments (Antoine et al., in press), its use in geomorphological research worldwide is still uncommon.

19The OSL dating method thus shows several advantages in comparison with other geochronological approaches. It is however important to try and get as often as possible an independent age-control to improve the reliability of the results. It may derive from archaeological or historical data, from the recognition of palaeoenvironmental evidence (pollens, periglacial features; Wintle, 2008), or from comparison with other absolute dating techniques (e.g., radiocarbon; Folz et al., 2001; Rittenour, 2008). It is however important to note that the comparisons between optical dating of quartz and independent age controls often show good correlations (e.g., Murray and Olley, 2002; Rittenour, 2008), demonstrating the accuracy of the OSL method. The above-mentioned methodological limits should not be considered as fixed in regard to the constant improvements which have characterised the OSL method for more than twenty years (SAR protocol, development of the small aliquots and single grain analyses, improvements in the anomalous fading estimate and in the statistical treatment, applications to new kinds of sediments). Hence the reason justifying the rarity of studies using OSL should very likely be attributed to the lack of laboratories in France in comparison with other countries (2 laboratories in France but ca. 15 in the UK). This fact is all the more regrettable in that research worldwide shows that the optical dating method represents one of the most useful geochronological tools in geomorphological studies.

Sneaking a look worldwide: potential for future applications in France

20OSL dating has for a few years been applied worldwide to most sediments: not only aeolian, coastal or fluvial, but also marine and tsunamigenic (Jacobs, 2008), glaciogenic lato sensu (Tsukamoto et al., 2002; Fuchs and Owen, 2008), slope-deposits (Fuchs and Lang, 2009) and even volcanic materials (e.g., Tsukamoto et al., in press). However, it is important to keep in mind that not all sediments from all sections can be successfully dated: geomorphological field expertise before sampling remains essential in order to target the sediments showing a higher probability for complete bleaching, since only these sediments will allow an accurate age to be obtained. Methodological improvements have enlarged the field of applications, but have also supported the development of quantitative geomorphology. Optical dating allows the recognition of high-resolution sedimentary discontinuities, as demonstrated for the aeolian environments where it favoured the passage from a “constant sedimentation rate” model to a model including numerous hiatuses and erosive periods (Singhvi and Porat, 2008). In the same way it is an essential tool for research focusing on the existence of morphological thresholds, or dealing with the influence of forcing (e.g., climate, tectonic, anthropogenic forcing). For example L.B. Clemmensen et al. (2007) showed that human mediaeval deforestation in Denmark combined with increased wind activity, led to the formation of sand dunes between the 15th and the 19th centuries AD. Based on these statements, the perspectives for applying the OSL method to geomorphological research in France are various. The main foci may be summarised as follows: i) the recent Re-OSL protocol can be applied to loess deposits which are often dated using IRSL (or only using TL as in the Achenheim section; Rousseau et al., 1998) to improve the chronological framework. This can subsequently be used for sediment budget estimation during the Pleistocene climatic cycles, completing the results presented by M. Frechen et al. (2003) for the last cycle; ii) the dating of coastal sediments has to be completed (e.g., in the Mediterranean coast) and updated. The reconstruction of coastal dune formation may actually help us to understand their dynamics, which in turn can be used to assess their response to an increase of storms and their management in a context of sea-level rise; iii) in fluvial environments OSL may be applied both to sediments which lack age control (as it is the case for many terrace staircases) and to sediments which have already been dated using other methods such as ESR (e.g., Somme and Yonne valleys; Antoine et al., in press) or radiocarbon (e.g., alluvial floodplain of the Moselle valley; Carcaud, 1992), such a comparison between geochronological approaches being essential for further methodological improvements. From an applied point of view, the acquisition of an extensive dataset is essential for a better understanding of fluvial response to environmental changes and for sediment budget reconstructions (e.g., Van Balen et al., 2000); iv) the application of the method to sediments which still lack unquestionable absolute dating. This is especially the case of glaciogenic deposits (especially fluvio-glacial to glacio-lacustrine) which are preserved in all the French mountains and often include sand-sized deposits potentially suitable for OSL dating; v) the dating of slope-deposits lato sensu may be used to achieve a better understanding of the influence of human societies on slope dynamics, as well as for hazard management.

Conclusion

21The optical dating method may be considered as a principal tool for geomorphological research, as it offers the possibility to date directly the depositional event, using two common minerals (quartz and feldspars) that are preserved in a wide range of sedimentary environments, and for various time-spans. Evidence for this is shown by the worldwide increase in the number of laboratories and publications. A similar observation has been made in the general review of S. Stokes (1999), but associated with the persistence of methodological problems. Ten years later, most of these problems seem to have been overcome: even if the operating procedures may vary from one laboratory to another (e.g., the use of single-grain or of small aliquots, preheat conditions, etc.), improvements enhance the possibility of new applications especially in France where studies remain sparse. The method should therefore contribute to the development of quantitative geomorphology (Singhvi and Porat, 2008), as it makes it possible to recognise sedimentary discontinuities (identification and dating of the thresholds separating erosive and aggradational periods), or to estimate aggradation rates. These results an then be placed within the general dynamics of the Earth’s system (tectonics, human forcing, climate change; Roberts, 2008), leading to a better understanding of its past, present and future functioning.

Acknowledgements
The author thanks Manfred Frechen, Sumiko Tsukamoto, and the PhD students and technicians of the section Geochronology and Isotope Hydrology of the Leibniz Institute of Applied Geophysics  (LIAG) in Hannover, for their constant help and support during the laboratory work. He also wishes to gratefully acknowledge the four anonymous reviewers for their very constructive comments and advice on the first version of the manuscript. Martin Stokes (university of Plymouth), Tobias Lauer (LIAG) and Martin Williams (university of Adelaide) are acknowledged for their improvements of the manuscript and especially for having polished the English. Access to the Rem VI gravel pit and sampling were made possible thanks to Alphonse and Jean-Pierre Hein, Marc Wiltzius and Fernando Soares (Hein company S.A.), and to Henri-Georges Naton and Laurent Brou (Musée National d'Histoire et d'Art du Luxembourg), respectively. This work has been made possible thanks to a partial support of the DAAD (German Academic Exchange Service).

Haut de page

Bibliographie

References

Adamiec G., Aitken M.J. (1998) – Dose rate conversions factors: update. Ancient TL 16, 37-50.

Aitken M.J. (1985)Thermoluminescence dating. Academic Press, London, 359 p.

Aitken M.J. (1998)An introduction to optical dating. Oxford University press, Oxford, 267 p.

Antoine P., Rousseau D.D., Lautridou J.-P., Hatté C. (1999) – Last interglacial-glacial climatic cycle in loess-palaeosol successions of north-western France. Boreas 28, 551-563.

Antoine P., Bahain J.-J., Debenham N., Frechen M., Gauthier A., Hatté C., Limondin-Lozouet N., Locht J.-L., Raymond P., Rousseau D.D. (2003) – Nouvelles données sur le Pléistocène du Nord du Bassin parisien : les séquences loessiques de Villiers-Adam (Val d’Oise, France). Quaternaire, 14, 219-235.

Antoine P., Limondin-Lozouet N., Auguste P., Locht J.-L., Ghaleb B., Reyss J.-L., Escudé E., Carbonel P., Mercier N., Bahain J.-J., Falguères C., Voinchet P. (2006) – Le tuf de Caours (Somme, France) : mise en évidence d’une séquence éémienne et d’un site paléolithique associé. Quaternaire, 17, 281-320.

Antoine P., Auguste P., Bahain J.-J., Chaussé C., Falguères C., Ghaleb B., Limondin-Lozouet N., Locht J.-L., Voinchet P. (in press) – Chronostratigraphy and palaeoenvironments of Acheulean occupations in Northern France (Somme, Seine and Yonne valleys). Quaternary International 1-6.

Arnold L., Stokes S., Bailey R., Fattahi M., Colls A., Tucker G. (2003) – Optical dating of potassium feldspar using far-red (λ>665 nm) IRSL emissions: a comparative study using fluvial sediments from the Loire River, France. Quaternary Science Reviews 22, 1093-1098.

Auclair M., Lamothe M., Huot S. (2003) – Measurement of anomalous fading for feldspars IRSL using SAR. Radiation Measurements 37, 487-492.  

Bahain J.-J., Falguères C., Dolo J.-M., Antoine P., Auguste P., Limondin-Lozouët N., Locht J.-L., Tuffreau A., Tissoux H., Farkh S. (in press) – ESR/U-series dating of teeth recovered from well-stratigraphically age-controlled sequences from Northern France. Quaternary Geochronology 1-5.

Bailey R.M., Smith B.W., Rhodes E.J. (1997) – Partial bleaching and the decay form characteristics of quartz OSL. Radiation Measurements 27, 123-136.

Bailey R.M. (2000) – The slow component of quartz optically stimulated luminescence. Radiation Measurements 32, 233-246.

Bailey R.M., Arnold L.J. (2006) – Statistical modelling of single grain quartz De distributions and an assessment of procedures for estimating burial doses. Quaternay Science Reviews 25, 2475-2502.

Balescu S., Dupuis C., Quinif Y. (1988) – TL stratigraphy of pre-weichselian loess from NW Europe using feldspar coarse grain. Quaternary Science Reviews 7, 309-313.

Balescu S., Packman S., Wintle A.G., Grün R. (1992) – Thermoluminescence dating of middle pleistocene raised beach of Sangatte (Northern France). Quaternary Research 35, 390-396.

Balescu S., Lamothe M. (1994) – Comparison of TL and IRSL age estimates of feldspar coarse grains from waterlain sediments.  Quaternary Geochronology 13, 437-444.

Balescu S., Lamothe M., Lautridou J.-P. (1997) – Luminescence evidence for two middle Pleistocene interglacial events at Tourville, northwestern France. Boreas 26, 61-72.

Balescu S., Tuffreau A. (2004) – La phase ancienne du paléolithique moyen dans la France septentrionale (stades isotopiques 8 à 6): apports de la datation par luminescence des séquences loessiques. Archeologicheskii almanach, Donetsk, 16, 5-22.

Banerjee D., Murray A.S., Bøtter-Jensen L., Lang A. (2001) – Equivalent dose estimation using a single aliquot of polymineral fine grains. Radiation Measurements 33, 73-94.

Bøtter-Jensen L., Duller G.A.T., Poolton N.R.J. (1994) – Excitation and emission spectrometry of stimulated luminescence from quartz and feldspars. Radiation Measurements 23, 613-616.

Brocard G.Y., van der Beek P.A., Bourlès D.L., Siame L.L., Mugnier J.-L. (2003) – Long-term fluvial incision rates and postglacial river relaxation time in the French Western Alps from 10Be dating of alluvial terraces with assessment of inheritance, soil development and wind ablation effects. Earth and Planetary Science Letters 209, 197-214.

Carcaud N. (1992)Remplissage des fonds de vallée de la Moselle et de la Meurthe en Lorraine sédimentaire. PhD thesis, University of Nancy 2, 281 p.

Clarke M.L., Rendell H.M., Pye K., Tastet J.-P., Pontee N.I., Massé L. (1999) – Evidence for the timing of dune development on the Aquitaine coast, southwest France. Zeitschrift für Geomorphologie N.F. Supplement Band 116, 147-163.

Clarke M.L., Rendell H.M., Tastet J.-P., Clavé B., Massé L. (2002) – Late-Holocene sand invasion and North Atlantic storminess along the Aquitaine Coast, southwest France. The Holocene 12, 231-238.

Clemmensen L.B., Bjørnsen M., Murray A., Pedersen K. (2007) – Formation of aeolian dunes on Anholt, Denmark since AD 1560: A record of deforestation and increased storminess. Sedimentary Geology 199, 171-187.

Cliquet D., Lautridou J.P., Antoine P., Lamothe M., Leroyer M., Limondin-Lozouet N., Mercier N. (2009a) – La séquence loessique de Saint-Pierre-lès-Elbeuf (Normandie, France) : nouvelles données archéologiques, géochronologiques et paléontologiques. Quaternaire, 20, 321-343.

Cliquet D., Lautridou J.P., Lamothe M., Mercier N., Schwenninger J.L., Alix P., Vilgrain G. (2009b) – Nouvelles données sur le site majeur d'Ecalgrain : datations radiométriques et occupations humaines de la Pointe de la Hague (Cotentin, Normandie). Quaternaire, 20, 345-359.

Colls A.E., Stokes S., Blum M.D., Straffin E. (2001) – Age limits on the late quaternary evolution of the upper Loire River. Quaternary Science Reviews 20, 743-750.

Cordier S., Frechen M., Harmand D., Beiner M. (2005) – Middle and Upper Pleistocene fluvial evolution of the Meurthe and Moselle valleys in the Paris basin and the Rhenish Massif. Quaternaire, 16, 201-215.

Cordier S., Harmand D., Frechen M., Beiner M. (2006) – Fluvial system response to Middle and Upper Pleistocene climate change in the Meurthe and Moselle valleys (Eastern Paris basin and Rhenish Massif). Quaternary Science Reviews 25, 1460-1474.

Cordier S., Frechen M., Harmand D. (2009) – The Pleistocene fluvial deposits of the Moselle and Middle Rhine valleys: new correlations and compared evolutions. Quaternaire, 20, 35-47.

Cordier S., Frechen M., Tsukamoto S. (in press) – Methodological aspects on luminescence dating of fluvial sands from the Moselle basin, Luxembourg. Geochronometria.

Coutard S., Lautridou J.-P., Rhodes E., Clet M. (2006) – Tectonic, eustatic and climatic significance of raised beaches of Val de Saire, Cotentin, Normandy, France. Quaternary Science Reviews 25, 596-611.

DeLong S.B., Arnold L.J. (2007) – Dating alluvial deposits with optically stimulated luminescence, AMS 14C and cosmogenic techniques, western Transverse Ranges, California, USA. Quaternary Geochronology 2, 129-136.

Duller G.A.T. (2003) – Distinguishing quartz and feldspars in single grain luminescence measurements. Radiation measurements 37, 161-165.

Duller G.A.T. (2004) – Luminescence dating of Quaternary sediments: recent advances. Journal of Quaternary Science 19, 183-192.

Duller G.A.T. (2007) – Assessing the error on equivalent dose estimates derived from single aliquot regenerative dose measurements. Ancient TL 25, 15-23.

Duller G.A.T. (2008) – Single-grain optical dating of Quaternary sediments: why aliquot size matters in luminescence dating. Boreas 37, 589-612.

Engelmann A., Frechen M., Antoine P. (1999) – Chronostratigraphie frühweichselselzeitlicher kolluvialer Sedimente von Bettencourt-Saint-Ouen (Nord-Frankreich). In Becker-Hauman R., Frechen M. (Eds.): Terrestrische Quatärgeologie. Logabook, Koln, 12-22.

Folz E. (2000)La luminescence stimulée optiquement du quartz : développements méthodologiques et applications à la datation de séquences du Pléistocène supérieur du Nord-Ouest de la France. PhD thesis, University of Paris 7, 267 p.

Folz E., Bodu P., Bonte P., Joron J.-L., Mercier N., Reyss J.-L. (2001) – OSL dating of fluvial quartz from Le Closeau, a Late Paleolithic site near Paris – comparison with 14C chronology. Quaternary Science Reviews 20, 927-933.

Frechen M., Oches E.A., Kohfeld K.E. (2003) – Loess in Europe - Mass accumulation rates during the last glacial period. Quaternary Science Reviews 22, 1835-1857.

Fuchs M., Owen L.A. (2008) – Luminescence dating of glacial and associated sediments: review, recommendations and future directions. Boreas 37, 636-659.

Fuchs M., Lang A. (2009) – Luminescence dating of hillslope deposits – a review. Geomorphology 109, 17-26.

Galbraith R.F. (2005)Statistics for fission tracks analysis. Chapman and Hall/CRC Press, Boca Raton, FL, 192 p.

Galbraith R.F., Green P.F. (1990) – Estimating the component age in a finite mixture. Nuclear Tracks and Radiation Measurements 17, 197-206.

Galbraith R.F., Laslett G. (1993) – Statistical models for mixed fission tracks ages. Radiation Measurements 21, 459-470.

Galbraith R.F., Roberts R.G., Laslett G.M., Yoshida H., Olley J.M. (1999) – Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: part I. Experimental design and statistical models. Archaeometry 41, 339-364.

Godfrey-Smith D.I., Huntley D.J., Chen W.H. (1988) – Optical dating studies of quartz and feldspar sediment extracts. Quaternary Science Reviews 7, 373-380.

Guette-Marsac C., Lautridou J.P., Cliquet D., Lechevalier C., Schwenninger J.L., Lamothe M., Mercier N., Fosse G. (2009) – Les occupations du Paléolithique moyen et supérieur d'Epouville (Pays de Caux) en contexte loessique. Quaternaire, 20, 389-404.

Hossain S.M. (2003)A critical comparison and evaluation of methods for the annual radiation dose determination in the luminescence dating of sediments. PhD thesis, Universiteit Gent, 209 p.

Huntley D.J., Godfrey-Smith D.I., Thewalt M.L.W. (1985) – Optical dating of sediments. Nature 313, 105-107.

Huntley D.J., Godfrey-Smith D.I., Haskell E.H. (1991) – Light induced emission spectra from some quartz and feldspars. International Journal of Radiation Applications and Instrumentation, Part D. Nuclear Tracks and Radiation Measurements 18, 127-131.

Huntley D.J., Hutton J.T., Prescott J.R. (1993) – Optical dating using inclusions within quartz grains. Geology 21, 1087-1090.

Huntley D.J., Lamothe M. (2001) – Ubiquity of anomalous fading in K-feldspars and the measurement and correction for it in optical dating. Canadian Journal of Earth Sciences 38, 1039-1106.

Jacobs Z. (2008) – Luminescence chronologies for coastal and marine sediments. Boreas 37, 508-535.

Jull A.J.T., Scott A.E. (2007) – Dating Techniques. Encyclopedia of Quaternary Science, Elsevier, Oxford, 453-459.

Juschus O., Preusser F., Melles M., Radtke U. (2007) – Applying SAR-IRSL methodology for dating fine-grained sediments from Lake El’gygytgyn, north-eastern Siberia. Quaternary Geochronology 2, 187-194.

Lauer T., Frechen M., Hoselmann C., Tsukamoto S. (in press) – OSL dating of Upper Pleistocene aeolian and fluvial deposits from the Heidelberg Basin. Proceedings of Geologists’ Association.

Lian O.B., Roberts R.G. (2006) – Dating the Quaternary: progress in luminescence dating of sediments. Quaternary Science Reviews 25, 2449-2468.

Martins A.A., Cunha P.P., Buylaert J.-P., Huot S., Murray A.S., Dinis P., Stokes M. (in press) – K-feldspar IRSL dating of a Pleistocene river terrace staircase sequence of the Lower Tejo River (Portugal, western Iberia). Quaternary Geochronology.

Mercier N. (2008) – Datation des sédiments quaternaires par luminescence stimulée optiquement : un état de la question. Quaternaire, 19, 195-204.

Mol J., Roebroeks W., Kamermans H., Van Kolfschoten T., Turq A. (2004) – Weichselian and Holocene fluvial evolution of the Vezere River valley (Dordogne, France). Quaternaire, 15, 187-193.

Murray A.S., Olley J.M. (2002) – Precision and accuracy in the optically stimulated luminescence dating of sedimentary quartz: a status review. Geochronometria 21, 1-16.

Murray A.S., Wintle A.G. (2000) – Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol. Radiation Measurements 32, 57-73.

Murray A.S., Wintle A.G. (2003) – The single aliquot regenerative dose protocol: potential for improvements in reliability. Radiation Measurements 37, 377-381.

Naton H.G., Cordier S., Brou L., Damblon F., Frechen M., Hauzeur A., Le Brun-Ricalens F., Valotteau F. (2009) – Fluvial evolution of the Moselle valley in Luxembourg during Late Pleistocene and Holocene: palaeoenvironment and human occupation. Quaternaire, 20, 81-92.

Prescott J.R., Hutton J.T. (1994) – Cosmic ray contributions to dose rates for luminescence and ESR dating: Large depths and long-term time variations. Radiation Measurements 23, 497-500.

Regnault H., Mauz B., Morzadec-Kerfourn M.T. (2003) – The last interglacial shoreline in northern Brittany, western France. Marine Geology 194, 65-77.

Rittenour T.M. (2008) – Luminescence dating of fluvial deposits: applications to geomorphic, paleoseismic and archaeological research. Boreas 37, 613-635.

Roberts H. (2007) – Assessing the effectiveness of the double-SAR protocol in isolating a luminescence signal dominated by quartz. Radiation measurements 42, 1627-1636.

Roberts H.M. (2008) – The development and application of luminescence dating to loess deposits: a perspective on the past, present and future. Boreas 37, 483-507.

Rodnight H. (2008) – How many equivalent dose values are needed to obtain a reproducible distribution? Ancient TL 26, 3-9.

Rodnight H., Duller G.A.T., Wintle A.G., Tooth S. (2006) – Assessing the reproducibility and accuracy of optical dating of fluvial deposits. Quaternary Geochronology 1, 109-120

Rousseau D.D., Zöller L., Valet J.-P. (1998) – Late Pleistocene climatic variations at Achenheim, France, based on a MAgnetic Susceptibility and TL chronology of loess. Quaternary Research 49, 255-263.

Singhvi A.K., Porat N. (2008) – Impact of luminescence dating on geomorphological and palaeoclimate research in drylands. Boreas 37, 536-558.

Stokes S. (1999) – Luminescence dating applications in geomorphological research. Geomorphology 29, 153-171.

Stokes S., Bray H.E., Blum M.D. (2001) – Optical resetting in large drainage basin: tests of zeroing assumptions using single aliquots procedures. Quaternary Science Reviews 20, 879-885.

Straffin E.C., Blum M.D., Colls A., Stokes S. (1999) – Alluvial stratigraphy of the Loire and Arroux Rivers (Burgundy, France). Quaternaire 10, 271-282.

Sun X., Mercier N., Falgueres C., Bahain J.-J., Despriee J., Bayle G., Lu H. (in press) – Recuperated optically stimulated luminescence dating of middle-size quartz grains from the Palaeolithic site of Bonneval (Eure-et-Loir, France). Quaternary Geochronology 1-6.

Tsukamoto S., Asahi K., Watanabe T., Rink W.J. (2002) – Timing of past glaciations in Kanchenjunga Himal, Nepal by optically stimulated luminescence dating of tills. Quaternary International 97-98, 57-67.

Tsukamoto S., Duller G.A.T., Wintle A.G., Frechen M. (in press) – Optical dating of a Japanese marker tephra using plagioclase. Quaternary Geochronology.

Van Balen R.T., Houtgast R.F., Van DER Wateren F.M., Vandenberghe J., Bogaart P.W. (2000) – Sediment budget and tectonic evolution of the Meuse catchment in the Ardennes and Roer Valley Rift System. Global and Planetary Change 27, 113-129.

Vandenberghe D. (2004)Investigation of the optically stimulated luminescence dating method for application to young geological sediments. PhD thesis, Universiteit Gent, 348 p.

Van Vliet-Lanoë B, Laurent M., Bahain M., Balescu S., Falguères C., Field M., Hallégouët B., Keen D.H. (2000) – Middle Pleistocene raised beach anomalies in the English Channel : regional and global stratigraphic implications. Journal of Geodynamics 29, 15-41.

Van Vliet-Lanoë B., Cliquet D., Auguste P., Folz E., Keen D., Schwenninger J.-L., Mercier N., Alix P., Roupin Y., Meurisse M., Seignac H. (2006) – L’abri sous-roche du Rozel (France, Manche) : un habitat de la phase récente du Paléolithique moyen dans son contexte géomorphologique. Quaternaire, 17, 207-258.

Vernet J.-L., Mercier N., Bazile F., Brugal J.-P. (2008) – Travertins et terrasses de la moyenne vallée du Tarn à Millau (Sud du Massif Central, Aveyron, France). Quaternaire, 19, 3-10.

Wallinga J., Murray A., Wintle A. (2000) – The single-aliquot regenerative-dose (SAR) protocol applied to coarse-grain feldspar. Radiation Measurements 32, 529-533.

Wallinga J. (2002) – Optically stimulated luminescence dating of fluvial deposits: a review. Boreas 31, 303-322.

Wang X.L., Lu Y.C., Wintle A.G. (2006) – Recuperated OSL dating of fine grained quartz in chinese loess. Quaternary Geochronology 1, 89-100.

Wintle A.G. (1973) – Anomalous fading of thermoluminescence in mineral samples. Nature 245, 143-144.

Wintle A.G., Shackleton N.J., Lautridou J.-P. (1984) – Thermoluminescence dating of periods of loess deposition and soil formation in Normandy. Nature 310, 491-493.

Wintle A.G. (1997) – Luminescence dating: laboratory procedures and protocols. Radiation Measurements 27, 769-817.

Wintle A.G. (2008) – Fifty years of luminescence dating. Archaeometry 50, 276-312.

Wintle A.G., Murray A.S. (2006) – A review of quartz optically stimulated luminescence characteristics and their relevance in single aliquot regeneration dating protocols. Radiation Measurements 41, 369-391.

Haut de page

Annexe

Version française abrégée

Cet article vise à présenter aux chercheurs impliqués dans des recherches géomorphologiques sensu lato l’ensemble de la méthode de datation OSL (pour Optically Stimulated Luminescence) depuis les principes physiques jusqu’aux procédures (du terrain à l’interprétation de l’âge) et aux applications. La présentation est illustrée par la datation de sables fluviatiles issus de la vallée de la Moselle.

Les méthodes de datation par luminescence reposent sur la mesure de l’impact de radiations ionisantes sur un minéral cristallin (quartz ou feldspath potassique notamment) à l’abri de la lumière solaire (Aitken, 1985, 1998). Ces radiations (, , ) proviennent de radionucléides présents dans l’environnement (uranium, thorium et les éléments issus de leur désintégration, potassium) et dans une moindre mesure du rayonnement cosmique. Elles provoquent l’émission d’électrons qui se trouvent piégés dans les défauts du cristal, à des profondeurs (ou niveaux d’énergie) variables. On peut opposer les pièges peu profonds, d’où les électrons pourront rapidement être éjectés, aux pièges « stables », où les électrons peuvent rester piégés pendant plusieurs centaines de milliers d’années, et qui sont utilisés à ce titre pour la datation. La quantité d’électrons piégée dépend de la quantité de rayonnement reçue par le cristal (ou dose), donc du laps de temps durant lequel celui-ci est soumis au rayonnement. Dès que le minéral se retrouve exposé à la lumière, notamment lors de sa prise en charge par un agent de transport, les électrons piégés absorbent l’énergie des photons incidents et sont libérés. Ils peuvent se recombiner avec des « trous » (vacances d’électrons) du cristal. Cette recombinaison s’accompagne de l’émission d’un signal luminescent dont l’intensité est fonction du nombre d’électrons libérés (fig. 1). A l’issue du transport, le grain ainsi « blanchi » est déposé et recouvert par des sédiments qui le protègent de la lumière solaire. Il subit de nouveau une irradiation naturelle permettant l’acquisition d’un signal dit luminescent, lequel peut, après prélèvement de l’échantillon, être mesuré en laboratoire lors d’une stimulation (fig. 2). Cette stimulation, obtenue par chauffage dans le cas de la thermoluminescence, se fait dans le cas de l’OSL par exposition à une lumière dont la longueur d’onde dépend du minéral considéré : généralement bleu-vert pour les quartz, infrarouges pour les feldspaths (Infrared Stimulated Luminescence, IRSL). Le signal obtenu est fonction de la durée de l’enfouissement et correspond à une courbe montrant la libération progressive des électrons (fig. 2). Les études sur les quartz ont montré que cette courbe est composite et associe différents niveaux de pièges plus ou moins rapides à vider (« fast, medium, slow components »). L’âge du sédiment est obtenu suivant l’équation fondamentale suivante : Age (en a) = P (en grays ou J/kg) / Dr (en grays/a). P est la paléodose, soit la quantité totale de dose absorbée par le minéral depuis son enfouissement. Elle est estimée en laboratoire par la détermination de la dose équivalente De (dose artificielle nécessaire pour obtenir un signal luminescent similaire au signal naturel). Dr correspond pour sa part au débit de dose, c’est-à-dire à l’intensité du rayonnement. Si en principe la paléodose est reflétée par le signal luminescent, plusieurs cas particuliers doivent être évoqués (fig. 1) : 1) lorsque le transport ne permet pas un blanchiment complet du grain (transport en masse ou dans un cours d’eau turbide), le signal mesuré inclut une partie du signal initial, de sorte que la dose équivalente est plus grande que la paléodose. L’âge est alors surestimé ; 2) si tous les défauts du cristal sont occupés par des électrons, le processus de piégeage s’interrompt bien que le minéral continue à être irradié. Le minéral est saturé, d’où une sous-estimation de la dose équivalente et donc de l’âge. Les quartz saturent à des doses plus faibles que les feldspaths : alors que pour ces derniers il est possible de dater des sédiments vieux de plusieurs centaines de milliers d’années, il a longtemps été difficile de dater des quartz au-delà de 100-150 ka. Cependant, aujourd’hui, on peut espérer dater des quartz plus anciens grâce à la procédure dite du signal de récupération du quartz (Re-OSL ; Wang et al., 2006) ; 3) les feldspaths sont affectés par le phénomène du « fading anomal » (anomalous fading), qui correspond à la libération spontanée d’électrons sans exposition à la lumière (Wintle, 1973). Ce fading aboutit également à une sous-estimation de la dose équivalente, d’où la nécessité d’effectuer une correction (modèle de D.J. Huntley et M. Lamothe, 2001) pour parvenir à un âge fiable.

Les sédiments les plus fréquemment utilisés pour les datations OSL sont les sables fins à moyens (90-250 m) et les limons fins (4-11 m). Le prélèvement se fait après rafraîchissement de la coupe, à l’aide d’un tube opaque, dans une couche la plus homogène possible, et dans la mesure du possible à l’écart de lits grossiers et/ou pédogénéisés (fig. 3). Un échantillon témoin (au moins 700 g pour des sables) peut être prélevé dans la même couche en vue de la détermination du débit de dose en laboratoire. Lorsque la coupe est suffisamment épaisse, plusieurs échantillons peuvent être prélevés afin de vérifier la cohérence des résultats obtenus. Il est également possible d’effectuer le prélèvement dans une carotte, dès lors que celle-ci est restée à l’abri de la lumière, et que les différentes strates sont bien préservées. La détermination du débit de dose implique la prise en compte de deux paramètres : la localisation de l’échantillon (coordonnées x, y et z ; profondeur) et sa teneur moyenne en eau durant l’enfouissement. En effet, l’eau absorbe une partie des radiations, influant donc sur le débit de dose effectivement reçu par le sédiment (une variation de 10 % de la teneur en eau peut changer l’âge final dans une proportion similaire). Le débit de dose est aussi influencé par la taille des sédiments étudiés : alors que les grains fins reçoivent toute l’énergie issue des rayonnements , et , les sables sont surtout sensibles aux rayonnements et . Le débit de dose peut être mesuré de plusieurs manières (Aitken, 1998), soit sur le terrain à l’aide d’un dosimètre ou d’un spectromètre portable, soit en laboratoire. La détermination de la dose équivalente (fig. 4) est réalisée en laboratoire sous lumière tamisée rouge ou orange, pour ne pas engendrer de blanchiment significatif des grains. L’échantillon est d’abord tamisé et débarrassé des carbonates et de la matière organique. Des liqueurs de densité différentes permettent ensuite d’isoler les feldspaths potassiques et les quartz. Lorsque l’étude porte sur la fraction sableuse, ces derniers subissent ensuite un traitement à l’acide fluorhydrique (etching) destiné à éliminer les feldspaths résiduels ainsi que la partie externe des grains exposée au rayonnement : ce dernier, moins efficace que les radiations et à produire un signal de luminescence, n’est pas reproduit en laboratoire. Dans le cas de la fraction fine, la séparation à l’aide des liqueurs de densité est moins efficace et le traitement à l’acide fluorhydrique, trop agressif, est remplacé par un traitement à l'acide fluorosilicique. La mesure du signal OSL peut également se faire sur des grains « polyminéraux », qui sont d’abord stimulés à l’aide de diodes infrarouges - pour réduire le signal issu des feldspaths - puis à l’aide de diodes bleues permettant d’enregistrer un signal majoritairement issu des quartz (méthode dite « double SAR » ; Banerjee et al., 2001). Une fois préparés, les grains sont répartis en plusieurs sous-échantillons ou aliquotes. Le nombre de grains par aliquote varie selon la fraction granulométrique analysée mais aussi selon les choix opératoires : les aliquotes peuvent être « mono-grain », « petites » (quelques dizaines de grains pour les sables fins à moyens), « moyennes » ou « larges » (plusieurs milliers de grains de sable). Le choix de la taille des aliquotes n’est pas anodin : l’utilisation d’aliquotes moyennes ou larges peut s’avérer délicate en cas de blanchiment incomplet, chaque aliquote mêlant potentiellement des grains bien blanchis à des grains imparfaitement blanchis, d’où une surestimation de la dose équivalente. A l’inverse, l’étude d’aliquotes trop petites risque d’aboutir à l’obtention d’un signal OSL trop faible pour être interprété avec précision. Le choix de la taille des aliquotes dépend donc à la fois du type de sédiments et de leur âge, sachant que des sédiments anciens donnent un signal OSL plus important. Les techniques employées en vue de la détermination de la dose équivalente peuvent être regroupées en deux types : les techniques additives et les techniques régénératives. Les premières ont été largement utilisées durant les années 1990, avant le développement des techniques régénératives, plus précises. La principale repose sur le protocole dit « Single Aliquot Regenerative » (SAR) qui a été développé sur les quartz (Murray et Wintle, 2000 et 2003 ; Wintle et Murray, 2006) puis appliqué aux feldspaths. Ce protocole a été utilisé pour dater les sables fluviatiles de la Moselle illustrant cet article. Il consiste à mesurer successivement pour une même aliquote plusieurs signaux OSL (fig. 5 et tab. 1) correspondant d’une part au signal naturel, d’autre part aux signaux obtenus après irradiations à des doses variables connues (« doses de régénération »). En principe, la comparaison du signal naturel avec les signaux artificiels devrait permettre de déterminer la dose équivalente par simple interpolation. Cependant, l’irradiation en laboratoire aboutit à un remplissage des pièges instables, qui sont vidés au moment de la stimulation lumineuse et viennent donc intensifier le signal OSL, ce qui rend délicate la comparaison avec le signal naturel. Il faut donc vider avant mesure ces pièges instables grâce à une préchauffe de l’aliquote (fig. 6). La préchauffe pouvant provoquer des changements dans la sensibilité de l’aliquote, c’est-à-dire aboutir à mesurer un signal OSL différent pour une même dose de rayonnement, ces changements sont évalués par l’administration d’une « dose test » suivie d’une courte préchauffe permettant le vidage des pièges instables. Ainsi, le protocole SAR inclut (tab. 1) : une préchauffe et la mesure du signal naturel (Ln), l’administration d’une dose test et la mesure après préchauffe du signal induit (Tn), puis une série d’irradiations à des doses de régénération variables permettant la mesure de plusieurs signaux OSL artificiels (Lx). Chaque mesure de Lx est suivie de l’administration de la dose test et de la mesure du signal induit Tx. Dès lors, la dose équivalente peut être calculée en interpolant le rapport Ln/Tn sur la courbe formée à partir des rapports Lx/Tx (fig. 7). Cette démarche est validée par l’ajout, dans le protocole, de tests supplémentaires (tab. 1) afin d’établir que les changements de sensibilités ont bien été pris en compte. En outre, lorsque le protocole SAR est appliqué à des grains de quartz, une détection effectuée à l’aide de diodes infrarouges permet d’identifier et d’exclure les aliquotes contenant des feldspaths, ce qui est toujours possible malgré le traitement à l’acide fluorhydrique et peut avoir une incidence sur la mesure du signal de luminescence. En revanche, dans le cas de l’étude de feldspaths, il importe d’évaluer le fading anomal par des procédures spécifiques (Huntley et Lamothe, 2001). Au final, la procédure SAR aboutit à l’obtention d’autant de doses équivalentes que d’aliquotes. Dans le cas de sédiments bien blanchis (cas des dépôts éoliens), les doses équivalentes présentent généralement des valeurs similaires. Pour les sédiments transportés dans des conditions ne permettant pas une exposition suffisante à la lumière solaire (cas des alluvions), la distribution est souvent hétérogène (fig. 8) : la moyenne ou la médiane ne permettent pas d’obtenir une estimation fiable de la dose équivalente. Des modèles statistiques, dont le « Minimum Age Model », qui permet d’identifier les sédiments les mieux blanchis d’une population, et le « Finite Mixture Model », utile lorsque l’échantillon comprend plusieurs populations bien distinctes, ont donc été développés récemment (Galbraith et al., 1999). De même que pour la taille des aliquotes, le choix du traitement statistique dépend du type de sédiment et du contexte de chaque échantillon.

En France, les premières applications de la méthode OSL ont porté sur les dépôts éoliens et littoraux (plages, dunes), potentiellement les mieux blanchis et de surcroît difficiles à dater par d’autres méthodes (tab. 2). Généralement associée à des vestiges préhistoriques, l’étude de séquences loess-paléosols a permis de rattacher des formations loessiques du nord-ouest de la France aux diverses phases froides du Quaternaire (Engelmann et al., 1999 ; Folz, 2000 ; Antoine et al., 2003 ; Bahain et al., sous presse ; Cliquet et al., 2009b). Alors que la plupart des études portent surtout sur les deux derniers cycles climatiques, certaines datations IRSL (Cliquet et al., 2009a) et, plus récemment, l’application de la méthode Re-OSL (Sun et al., sous presse), ont permis de dater des loess du stade isotopique 12 (450 ka) dans l’ouest du Bassin parisien. Dans un autre contexte, la datation de sables prélevés sur des plages perchées (Balescu et al., 1997 ; Regnault et al., 2003 ; Coutard et al., 2006) ou dans des dunes littorales (Folz, 2000 ; Van Vliet-Lanoë et al., 2006) de la Manche et de la Mer du Nord a conduit à rapporter leur mise en place aux périodes interglaciaires (MIS 7 ou 5). Par ailleurs, sur la côte Aquitaine, l’étude de dunes de la deuxième moitié de l’Holocène a montré l’existence d’un lien entre la dynamique éolienne et une augmentation de la fréquence des grandes tempêtes (Clarke et al., 1999 et 2002). Les améliorations successives apportées à la méthode ont également facilité son application aux sédiments fluviatiles, pourtant sujets à un blanchiment différentiel des grains. Des études ponctuelles ont été réalisées dans les vallées de la Seine (Balescu et al., 1997 ; Folz et al., 2001), de la Vézère (Mol et al., 2004) et du Tarn (Vernet et al., 2008). Des études plus systématiques ont été menées dans les bassins de la Loire (Straffin et al., 1999 ; Stokes et al., 2001) et de la Moselle (Cordier et al., 2005, 2006), aboutissant à l’obtention d’un calage chronostratigraphique pour les terrasses alluviales les plus récentes et à l’évaluation des vitesses d’incision depuis le Saalien (fig. 9, tab. 3 et tab. 4). La méthode a donc été appliquée à des sédiments d’origines diverses et d’âges variables (de plusieurs centaines de ka jusqu’à l’actuel). Cependant, elle a surtout été utilisée en vue de reconstitutions paléoenvironnementales à grande échelle (tab. 2) alors que les études quantitatives restent rares. En outre, la place des datations OSL dans les recherches géomorphologiques en France reste limitée. Ce fait s’explique avant tout par le faible nombre de laboratoires dans le pays. En effet, même si la méthode OSL présente plusieurs limites (fading anomal des feldspaths, risque d’un blanchiment différentiel, marges d’erreur liées à la détermination de la teneur en eau ou à l’instrumentation), les améliorations récentes lui permettent de soutenir la comparaison avec d’autres méthodes de datation : toutes les méthodes présentant des limites, seule l’utilisation de plusieurs méthodes sur un même dépôt peut permettre d’obtenir un âge fiable. La faible place accordée aux datations OSL en France est d’autant plus regrettable qu’au-delà des travaux évoqués, ses applications sont multiples : les études récentes menées dans le monde entier soulignent la possibilité de dater directement des sédiments marins, glaciaires sensu lato (Fuchs et Owen, 2008), des dépôts de versant voire certains dépôts « tsunamigéniques » (Jacobs, 2008) ou volcaniques, à condition au préalable d’identifier les sédiments potentiellement les mieux blanchis, seuls à même de pouvoir donner un âge exploitable. La multiplication des laboratoires équipés et des publications confirme de fait que les datations OSL constituent malgré leurs limites un outil majeur pour les recherches en géomorphologie.

De nombreuses perspectives existent donc en France, incluant la datation par les nouvelles procédures de sédiments déjà étudiés (notamment par d’autres méthodes) et la réalisation de nouvelles datations portant sur une grande variété de sédiments (glaciaires, colluviaux, fluviatiles,etc.). Ces travaux ne peuvent que favoriser le développement d’une géomorphologie quantitative, facilitant une meilleure appréhension du fonctionnement du système Terre et des réflexions prospectives sur son évolution à plus ou moins long terme.

Haut de page

Table des illustrations

Titre Fig. 1 – Schematic representation of the history of a mineral from the host-rock to the depositional event and its dating in the laboratory.Fig. 1 Représentation schématique de l’histoire d’un minéral depuis la roche mère jusqu’à son dépôt dans une formation sédimentaire et la datation de ce dépôt en laboratoire.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-1.jpg
Fichier image/jpeg, 336k
Titre Fig. 2 – Example of decay curves (quartz sands from sample LUM 978) showing the progressive eviction of the electrons from their traps during the first seconds of optical stimulation and the correlative emission of photons associated either with the natural signal or with artificial signals obtained for various irradiation doses. This decay is not a single exponential curve as several levels of traps are involved, some of them being more rapidly bleached (fast component) than the others (medium and slow components).Fig. 2 Exemple de courbes (sables quartzeux issus de l’échantillon LUM 978) montrant la libération progressive des électrons dans les premières secondes de la stimulation optique. Les signaux OSL correspondent d’une part au signal naturel, d’autre part aux signaux obtenus en laboratoire après irradiation à des doses variables. La courbe ne correspond pas à une simple décroissance exponentielle de la libération des électrons en fonction du temps car plusieurs niveaux de piège interviennent, certains étant plus rapidement blanchis que d’autres.
Crédits   
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-2.jpg
Fichier image/jpeg, 1,1M
Titre Fig. 3 – Location of the case study.Fig. 3 Localisation du site étudié.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-3.jpg
Fichier image/jpeg, 1,8M
Titre Fig. 4 – Schematic diagram showing the laboratory procedures for the equivalent dose and dose rate determinations.Fig. 4 Schéma décrivant le protocole opératoire de préparation des échantillons en vue de la détermination de la dose équivalente et du débit de dose.
Crédits   
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-4.jpg
Fichier image/jpeg, 540k
Titre Fig. 5 – The system used to perform the SAR protocol.Fig. 5 Vue d’ensemble du lecteur utilisé pour la réalisation du protocol SAR.
Légende The aliquots are placed on a wheel in the reader (black box on the right). The black tube above the reader is the photomultiplier tube used to record the photons emitted by quartz and feldspar during a stimulation (i.e., the luminescence signal). The stimulation sources (blue and IR LEDs) are nested below the photomultiplier tube. The heater element is located in the lower part of the reader. The grey cylinder is the irradiator accommodating a 90Sr/90Y source. In the middle of the picture, the controller is used to select the parameters (OSL or IRSL stimulation, etc.). The results can be seen on the screen left of the picture (photo: T. Lauer).Les aliquotes sont placées sur une roue insérée dans le lecteur stricto sensu, à droite de l’image. Au-dessus du lecteur se trouve le tube photomultiplicateur (en noir) utilisé pour la détection du signal luminescent. Les diodes utilisées pour la stimulation (bleues ou infra-rouges) se trouvent près de la base du tube. Le cylindre gris correspond à la source irradiante (irradiation obtenue à partir d’une source 90Sr/90Y). Au centre de l’image, le contrôleur permet de sélectionner notamment le type de stimulation (OSL ou IRSL). Les résultats peuvent être observés sur l’écran à gauche (photo : T. Lauer).
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-5.jpg
Fichier image/jpeg, 232k
Titre Fig. 6 – Results of the preheat test performed for sample LUM 978.Fig. 6 Résultats du test de préchauffe réalisé sur l’échantillon LUM 978.
Légende The preheat test show the presence of a small plateau between 240°C and 260°C, the former being used for subsequent measurements. The recycling ratio is constantly close to 1, and the recuperation always lower than 3%.Ce test met en évidence l’existence d’un plateau entre 240°C et 260°C, la première température étant retenue pour les mesures ultérieures. Le « rapport de recyclage » (recycling ratio) montre des valeurs systématiquement proches de 1, la récupération étant toujours inférieure à 3 %.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-6.jpg
Fichier image/jpeg, 136k
Titre Fig. 7 – Example of palaeodose determination using the SAR protocol for the same aliquot as in fig. 2.Fig. 7 Détermination de la paléodose à partir du protocole SAR pour la même aliquote que celle présentée fig. 2.
Légende The y-axis shows the ratios obtained for each of these artificial doses (Lx/Tx) and for the natural sample (Ln/Tn). The palaeodose value for this aliquot from sample LUM 978 is 69.3 ± 4.3 Gy.L’axe des ordonnées montre les ratios obtenus pour chacune de ces doses (Lx/Tx) ainsi que pour le signal naturel (Ln/Tn). La paléodose pour cette aliquote de l’échantillon LUM 978 est de 69,3 ± 4,3 Gy.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-7.jpg
Fichier image/jpeg, 152k
Titre Fig. 8 – Palaeodose distribution for quartz and feldspars aliquots from samples LUM 975 and 978.Fig. 8 Distribution des paléodoses pour les quartz et feldspaths des échantillons LUM 975 et 978.
Légende Each De value is represented by a point. The precision is indicated by the x-axis and increases to the right. The De value is plotted on the y-axis as the number of standard deviations away from a chosen central value. The presence of a majority of points within the grey band (2 band) for the feldspars from sample LUM 975 reflects that the aliquots are well bleached. In contrast, the radial plots for quartz from both samples show the presence of high palaeodoses corresponding with incompletely bleached grains.Chaque point correspond à une valeur de De (indiquée sur l’axe des ordonnées). En abscisse est indiquée la marge d’erreur. La présence d'une majorité de points dans la bande grisée (bande des 2 ) pour les feldspaths de l'échantillon LUM 975 suggère que les aliquotes sont bien blanchies. Les diagrammes soulignent pour les quartz la présence de paléodoses très élevées reflétant un blanchiment incomplet des grains.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-8.jpg
Fichier image/jpeg, 376k
Titre Fig. 9 – Chronological frame of the Meurthe-Moselle alluvial terrace system based on the optical datings. The generalised results (see also Cordier et al., 2005, 2006) demonstrate the correlation between terrace formation and Pleistocene climate cycles.Fig. 9 Cadre chronologique des terrasses alluviales de la Meurthe-Moselle obtenu à partir des datations par luminescence optique. Ces résultats portant sur l'ensemble de la vallée (voir aussi Cordier et al., 2005, 2006) soulignent les corrélations entre la formation des terrasses alluviales et les cycles climatiques du Pléistocène.
URL http://journals.openedition.org/geomorphologie/docannexe/image/7785/img-9.jpg
Fichier image/jpeg, 560k
Haut de page

Pour citer cet article

Référence papier

Stéphane Cordier, « Optically stimulated luminescence dating: procedures and applications to geomorphological research in France », Géomorphologie : relief, processus, environnement, vol. 16 - n° 1 | 2010, 21-40.

Référence électronique

Stéphane Cordier, « Optically stimulated luminescence dating: procedures and applications to geomorphological research in France », Géomorphologie : relief, processus, environnement [En ligne], vol. 16 - n° 1 | 2010, mis en ligne le 01 avril 2012, consulté le 17 décembre 2017. URL : http://journals.openedition.org/geomorphologie/7785 ; DOI : 10.4000/geomorphologie.7785

Haut de page

Auteur

Stéphane Cordier

Université Paris-Est-Créteil-Val-de-Marne (UPEC) - Département de Géographie - Laboratoire de Géographie Physique - UMR 8591 CNRS - 61, avenue du Général de Gaulle - 94010 Créteil Cedex - France; Leibniz Institut for Applied Geophysics - Section Geochronology and Isotope Hydrology - Hannover - Germany (stephane.cordier@u-pec.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