Search

Rock luminescence dating has grown in popularity in recent years, and potentially offers several advantages over traditional sediment luminescence dating. One key advantage when determining burial ages is the potential for an individual rock surface or clast to preserve clear evidence for bleaching prior to the burial event of interest and the subsequent accumulation of dose, and in some circumstances these materials may even preserve a history of multiple exposure and burial events. This previous exposure and burial history is inferred from the study of variation in age with increasing depth away from the rock surface. But such patterns in age-depth profiles in rock may also potentially be influenced by inhomogeneous radionuclide distributions causing spatial variations in the dose rate [1, 2]. In unconsolidated sediment dating it is recognised that various factors such as accuracy and reproducibility of heating [3] and grain size [4] can also influence the assessment of equivalent dose, and the same is anticipated to be true of rock slices. Minimising sources of variability within and between rock slices is important if rock age-depth profiles are to be interpreted appropriately. This study investigates the influence of various parameters on the reproducibility of assessments of equivalent dose from rock slices. The factors explored include the thickness of the materials being measured, beta source calibration, the reproducibility and efficacy of heating rock slices, and variations in optical attenuation. The ability to cut reproducibly thin and flat rock slices minimises the likelihood of variability within and between rock slices for many parameters, and gives greater confidence in the equivalent dose values determined and in their interpretation as part of a rock age-depth profile describing the bleaching and burial history. A method to achieve this routine preparation of uniformly thin and flat rock slices is described. Working with thin rock slices also means that the number of slices that can be prepared over a given depth into rock is obviously greater than can be achieved using thicker slices, which can be particularly advantageous for rock types which permit limited penetration of light, improving their potential to provide information on prior bleaching and burial events. [ABSTRACT FROM AUTHOR]

Devices for imaging luminescence emissions have been used for a number of years [e.g. 1], but where analysis of a sequence of measurements of the luminescence signal following different treatments is required a significant challenge has been to accurately register the images so that a signal measured from a region of interest defined in one image is compared with the corresponding part of the other images. This is particularly an issue with automated instruments where sample repositioning varies from one measurement step to the next, resulting in samples moving laterally and/or rotationally with respect to the EMCCD field of view. The calcitic opercula of Bithynia tentaculata are of interest for their TL emission which grows to doses in excess of 8,000 Gy, meaning it is potentially able to date the entire Quaternary period. The opercula are teardrop shaped and typically 2-4 mm on their longest axis. Imaging using an EMCCD permits spatially resolved equivalent dose determination, but registering the images has been a major impediment to analysis of sequences of measurements. Previous work [2] manually determined the translation and rotation of the opercula using distinctive features on the periphery of the opercula as markers. However, this approach is time consuming, especially where between 30 and 50 images of each opercula may need to be aligned for a sequence of measurements; it is also prone to human error. An automated process of image registration has been developed using ImageJ [3] capable of tracking each opercula. This procedure identifies the outline of the opercula in each image and then registers them. Using the translation and rotation of each image, any region of interest defined on one image looks at the same part of the opercula on all other images. Additionally, since the process identifies the outline of the opercula it allows a series of morphometric measurements to be made automatically, and for a series of regions of interest to be defined within the confines of the outline shape. The new approach is described and illustrated. A number of techniques are used to assess the accuracy of the registration process, and the impact upon the determination of De or other measurements such as pulse annealing are shown. This approach could also have applications for other types of object, such as rock slices, although they would need to be non-circular to be able to assess rotation of the object. It could also be adapted to enable the comparison of images captured by different devices mounted on the same instrument. [ABSTRACT FROM AUTHOR]

Rift lakes are perfect places for sediment accumulation and hence potentially contain continous paleoclimate records. Located in the eastern African rift valley, in southern Ethiopia, Lake Chew Bahir preserves a record of environmental change which is pertinent to the story of hominin evolution and dispersal. Independent chronology from various dating techniques is available throughout this 293 m composite core record, from which a coherent Bayesian age-depth model was generated spanning the last 620 ka and providing one of the longest directly-dated lacustrine sediment sequences in eastern Africa [1]. However, there are regions of the core where the age depth model could be improved if further datable materials were available. For instance, beyond ~50 m depth, the quartz OSL signal appears to be saturated and there is a large gap to the first 40Ar/39Ar age. The luminescence signal from feldspars saturates at a greater De value than that from quartz and therefore offers the potential to extend the maximum age range of the luminescence-derived chronology in the core, and to fill in gaps within the quartz-derived chronology where quartz is not available. The luminescence chronology for the Lake Chew Bahir core was refined and extended using the fading corrected post-IR IRSL225 signal from polymineral fine grains. However, there are challenges and complications regarding both assessment of fading and the subsequent application of the fading rates. The duration of the delay times used to assess fading, and the given laboratory dose (two critical factors in measurement of fading rates), were the same for all samples, but because of variability in lithology in the core, it is not clear whether sample specific fading rates should be used or if the mean fading rate of all samples is appropriate. Moreover, selection of the most appropriate fading correction model is critical. To circumvent the complications of fading rate measurement and its application, the recently developed post-Isothermal (pIt) IRSL protocol [2] was therefore also explored. Application of a modified pIt-IRSL225 protocol to polymineral fine grains from this core, demonstrates that this signal is successful in providing accurate ages for this material without any fading correction. Additionally, uncertainty on the derived ages were reduced in comparison with the fading corrected post-IR IRSL225 ages. However, since the calculated De from this protocol relies on the IR50/pIRIR225 De ratio, it seems to be limited to the saturation limit of the IR50 signal. This may be the reason that the pIt-IRSL225 age for one of the older samples in the core (~300 ka) does not agree with its fading corrected post-IR IRSL225 age. Possible modifications to the pIt-IRSL225 protocol to overcome this limitation will be discussed. Furthermore, the suitability of the isothermal annealing step in this protocol to account for the athermal process of signal loss (i.e. fading) was investigated. The data suggest there is a correlation between the isothermal annealing time (for which the De in the pIt- IRSL225 protocol is reached) and the g-values determined using IR50 and pIRIR225 signals. [ABSTRACT FROM AUTHOR]

A significant advantage of rock surface luminescence dating for determining the burial age of cobbles is that well-bleached samples can be identified according to the variation in calculated luminescence age with depth. A plateau in the luminescence age-depth profile extending from the outer surface into the cobble that also gives a lower value than is seen closer towards the interior of the cobble is conclusive evidence of sufficient daylight exposure. The age of this sub-surface plateau represents the time elapsed since the cobble was last exposed to daylight, prior to burial. But there is speculation regarding whether this age of last exposure represents the true age of the formation of the features / landforms of interest for dating. Over 300 cobbles were investigated in this study, originating from different glacial valleys in Tibet. Samples were taken from a range of depositional settings, including different types of moraines and from glaciofluvial sediments. Independent age control and expected ages were provided by cosmogenic nuclide exposure dating and by collection of modern samples. The Ln/Tn signal or equivalent dose were determined using post-IR IRSL protocols. The diverse shapes of luminescence-depth profiles indicate various bleaching / transport histories for different cobbles. Those considered as well-bleached samples could be found from all sampling sites. However, different sites show different proportions of well-bleached cobbles. We found that some cobbles show sub-surface plateaus that were lower than the values toward the centre of the cobble, but the sub-surface ages are apparently older than the independent or expected ages. One explanation of this is that those cobbles were exposed to sufficient daylight in the past to be wellbleached, but not at the time immediately prior to their last deposition which formed the feature or landform sampled for dating, but instead during a previous event since which time they had not been re-exposed to daylight. This means that the age determined for a feature could be overestimated even though a luminescence-depth profile with a low sub-surface dose plateau was obtained from a cobble indicating that the cobble was well-bleached. Unlike traditional luminescence dating of sediment grains, there are practical and logistical reasons why it is unrealistic, if not impossible, to routinely measure a large number of cobbles from every single site and apply statistical models to extract the true age for the feature of interest for dating. The data show us how confident we are in obtaining the true age for a feature or landform when applying luminescence dating on buried cobbles. Recommendations are provided regarding applying rock surface luminescence dating in environments where conditions mean that not all cobbles will have had any previous luminescence signals (including records of previous bleaching events) removed prior to formation of a feature that is subsequently sampled for dating. In such environments, different levels of low sub-surface age plateau recorded in different cobbles from the same feature demonstrate the potential of tracking the pre-depositional history of these sediments by dating multiple cobbles. [ABSTRACT FROM AUTHOR]

Luminescence-dating studies produce vast bodies of primary data that usually remain inaccessible or incompatible with future studies or adjacent scientific disciplines. To facilitate open data exchange and long-term data preservation in luminescence dating studies, we propose a new XML-based structured data format called XLUM [1]. The design applies a hierarchical data storage concept consisting of a root node (node 0), a sample (node 1), a sequence (node 2), a record (node 3) and a curve (node 4). The curve level holds information on the technical component (e.g., photomultiplier, thermocouple). A finite number of curves represent a record (e.g., an optically stimulated luminescence curve). Records are part of a sequence measured for a particular sample. Our format proposal allows the user to retain information from the measurement process on a technical component level. The additional storage of related metadata fosters future data mining projects on large datasets. The XML-based format is less memory efficient than binary formats; however, in focus is data exchange, preservation and hence XLUM long-term format stability. XLUM is inherently stable to future updates and backwards compatible. We support XLUM through two new software packages for R and Python, each called 'xlum' [2, 3]. Both packages facilitate the import of XLUM files into the respective environment. The R package 'xlum' converts existing luminescence-data file formats (e.g., *.bin/binx, *.psl, *.xsyg) into the new XLUM format. XLUM is licensed under the MIT license and hence available for free to be used in open and closed-source commercial and non-commercial software and research projects. Our presentation will outline the basic concept of XLUM, discuss advantages and disadvantages and sketch possible future format progress. [ABSTRACT FROM AUTHOR]

Calcitic snail opercula are a promising new material for thermoluminescence (TL) dating with the potential to cover the last ~3 Ma [1]. The ERC-funded Bridging Europe: a Quaternary Timescale for the Expansion and Evolution of Humans (EQuaTe) project aims to produce a secure dating framework for the earliest human occupation and expansion across Europe using this TL signal. Thermal lag is a key consideration for TL measurements and assessing the reproducibility of heating samples during TL measurement is an important but challenging issue. The potential magnitude of the problem is illustrated by Schmidt et al. [2] who reported a variation of ~60°C in the temperature of the "110 °C" TL peak from quartz grains when heating at 5 °C/s. It is unclear what proportion of this variability is due to thermal lag or incorrect calibration of hotplates. Duller et al. [3] showed that across eight different readers it was possible to reduce the variability in the "110 °C" TL peak from quartz to 2°C, and they suggested the temperature at which the TL peak was observed could be used to routinely assess the reproducibility of sample heating during measurement protocols. This relatively simple procedure involves measuring the TL glow curve and fitting the 110 °C TL peak with a Weibull function [see 4] thereby monitoring the peak position during a measurement sequence. The TL signal from calcitic opercula displays one low temperature peak (~100 °C, peak I) and two high temperature peaks (~275 °C and 350 °C, peak II and peak III respectively). Lifetimes of 7.4 x 107 and 1.4 x 1011 years have been reported for peaks II and III [1], and these are the signals that are of value for dating [e.g. 5, 6]. Little work has been published for peak I (~100°C). Here we present our investigation of the low temperature TL peak produced by Bithynia tentaculata opercula, and whether it can be used to investigate thermal lag and monitor temperature during TL measurements. The approach is used to assess the reproducibility of heating between different TL measurements during a single aliquot regenerative dose (SAR) procedure, and also to examine variability in heating between different opercula from a single sample. [ABSTRACT FROM AUTHOR]