FIFTY YEARS OF LUMINESCENCE DATING*

Luminescence techniques for dating both heated materials (e.g., pottery and burnt stone) and sediments have gone through three phases in the past 50 years. The first 22 years, from 1957 to 1979, were devoted to thermoluminescence (TL) techniques applied to heated material. In the next six years, from 1979 to 1985, it was found that TL dating could be applied to sediments. However, the TL signals of both quartz and feldspar grains observed for modern sediments were small, but were not zero, thus making them useful only for older sediments. A new luminescence signal that was totally zeroed by exposure to sunlight for a relatively short period of time, a few hours or less, was found; this led to the development of optically stimulated luminescence (OSL) dating techniques in 1985. Further developments have continued to the present day. Within the past 22 years, 1999 was also a year with major developments. A reliable procedure for single aliquots of quartz that have a rapidly bleached (‘fast’) OSL component was formalized and an instrument that allowed rapid measurement of equivalent doses for single grains was constructed. These developments have led to OSL becoming a major dating tool in Quaternary geology, at least for the past 100 000 years, and in archaeology, particularly as related to the dispersal of modern humans.     

OPTICAL DATING OF SINGLE AND MULTIPLE GRAINS OF QUARTZ FROM JINMIUM ROCK SHELTER,NORTHERN AUSTRALIA: PART II,RESULTS AND IMPLICATIONS*

Quartz sediments from the floor deposit at Jinmium rock shelter have been investigated using the multiple-grain and single-grain optical dating methods described by Galbraith et al. (1999, this volume). Here we present the results of this dating programme and argue that the artefact-bearing sediments were deposited within the last 10000 years. This time interval is consistent with the radiocarbon chronology but is much younger than previous claims for initial human occupation during the Pleistocene. Analysis of individual grains revealed also that the characteristic saturation doses of some grains are unusually high, which may permit dating of deposits older than a few hundred thousand years. Such grain-to-grain differences raise doubts, however, about the validity of using multiple-grain samples to investigate the phenomenology of quartz luminescence.     

OPTICAL DATING OF SINGLE AND MULTIPLE GRAINS OF QUARTZ FROM JINMIUM ROCK SHELTER,NORTHERN AUSTRALIA: PART I,EXPERIMENTAL DESIGN AND STATISTICAL MODELS*

Jinmium rock shelter is famous for the claims made by Fullagar et al. (1996) for the early human colonization and ancient rock art of northern Australia. These claims were based on thermo-luminescence ages obtained for the artefact-bearing quartz sediments that form the floor deposit at the site. In this paper, we outline the background to the optical dating programme at Jinmium, and describe the experimental design and statistical methods used to obtain optical ages from single grains of quartz sand. The results, interpretations, and implications of this dating programme are reported in a companion paper (Roberts et al. 7999, this volume).     

DATING OF THE HOMINID (HOMO NEANDERTHALENSIS) REMAINS ACCUMULATION FROM EL SIDRÓN CAVE (PILOÑA,ASTURIAS, NORTH SPAIN): AN EXAMPLE OF A MULTI‐METHODOLOGICAL APPROACH TO THE DATING OF UPPER PLEISTOCENE SITES

The age of Neanderthal remains and associated sediments from El Sidrón cave has been obtained through different dating methods (¹⁴CAMS, U/TH, OSL, ESR and AAR) and samples (charcoal debris, bone, tooth dentine, stalagmitic flowstone, carbonate‐rich sediments, sedimentary quartz grains, tooth enamel and land snail shells). Detrital Th contamination rendered Th/U dating analyses of flowstone unreliable. Recent ¹⁴C contamination produced spurious age‐values from charcoal samples as well as from inadequately pretreated tooth samples. Most consistent ¹⁴C dates are grouped into two series: one between 35 and 40 ka and the other between 48 and 49 ka. Most ESR and AAR samples yielded concordant ages, ranging between 39 and 45 ka; OSL dating results permitted adequate bracketing of the sedimentary layer that contained the human remains. Our results emphasize the value of multi‐dating approaches for the establishment of reliable chronologies of human remains.     

Expectations of scatter in equivalent-dose distributions when using multi-grain aliquots for OSL dating

In the OSL dating of sediment, the scatter in equivalent dose (D _e) between grains is almost always larger than would be expected due to counting statistics alone. Some scatter may be caused by insufficient (partial) bleaching of some of the grains prior to deposition. In order to date partially bleached sediment, it is essential to estimate the amount of scatter caused by other processes (e.g. grain-to-grain variability in the natural dose rate). Measurements of such scatter are performed at the single-grain level; by contrast, most OSL dating is performed on multi-grain subsamples, for which grain-to-grain scatter is reduced through averaging.     

BAYESIAN ANALYSIS OF PLEISTOCENE CHRONOMETRIC METHODS*

Bayesian analysis of radiocarbon dates on Holocene archaeological sites has become well established. Application to Pleistocene sites dated by multiple techniques would be advantageous. This paper develops the necessary mathematical apparatus in the form of likelihoods for luminescence dating, and uranium‐series dating, and considers the possibility for amino‐acid racemization dating. Application of the new methods is illustrated using the stratigraphic sequences of dates from the sites of Saint‐Césaire and La Chaise‐de‐Vouthon. For application to amino‐acid racemization dating, fuller publication of data is found to be required.     

THE BRITISH ARCHAEOMAGNETIC CALIBRATION CURVE: AN OBJECTIVE TREATMENT

This paper examines the limitations and deficiencies of the current archaeomagnetic calibration curve in use in Britain and explores a different approach to its construction, drawing on recent advances in this subject, in particular the use of a weighted moving window method of averaging. The main advantage of this method is shown to be its ability to take into account uncertainties in both the magnetic direction and the archaeological date of data used in the calibration curve, leading to a secular variation curve with an associated error estimate. Hence, when dating a magnetic direction, the uncertainties in the calibration curve can be taken into account, as well as the uncertainties in the magnetic direction to be dated.     

The Potential for Dating the Old Scatness Site,Shetland, By Optically Stimulated Luminescence

A series of deposits from the agricultural infield of the multiperiod settlement mound, Old Scatness, were investigated for their potential to yield optically stimulated luminescence dates. Luminescence properties of quartz grains were found to vary through the sequence, but dates were successfully obtained from five deposits, including anthropogenic soils, windblown sands and sands within midden deposits. Single‐aliquot equivalent dose measurement was found to be the most appropriate method for dating the deposits. The OSL dates obtained accorded well with the dates provided by archaeological evidence and included the post‐medieval, Iron Age, Bronze Age and Neolithic periods of Shetland, while a substantial midden was dated to the Bronze/Iron Age transition.     

On the applicability of post-IR IRSL dating to Japanese loess

Recent work on infrared stimulated luminescence (IRSL) dating has focussed on finding and testing signals which show less or negligible fading. IRSL signals measured at elevated temperature following IR stimulation at 50°C (post-IR IRSL) have been shown to be much more stable than the low temperature IRSL signal and seem to have considerable potential for dating. For Early Pleistocene samples of both European and Chinese loess natural post-IR IRSL signals lying in the saturation region of the laboratory dose response curve have been observed; this suggests that there is no significant fading in nature. As a contribution to the further testing of post-IR IRSL dating, we have used 18 samples from two Japanese loess profiles for which quartz OSL and tephra ages up to 600 ka provide age control. After a preheat of 320°C (60 s), the polymineral fine grains (4–11 μm) were bleached with IR at 50°C (200 s) and the IRSL was subsequently measured at 290°C for 200 s. In general, the fading uncorrected post-IR IRSL ages agree with both the quartz OSL and the tephra ages. We conclude that the post-IR IRSL signal from these samples does not fade significantly and allows precise and accurate age determinations on these sediments.     

Luminescence dating of anthropogenic deposits from Tall Zarʿa in the Jordan Valley

This study uses quartz-based optically stimulated luminescence dating to determine when a massive fortification wall (W11186) was constructed in the archaeological site of Tall Zarʿa in the Jordan Valley. A total of 11 samples of sediments were taken from a trench on the north side of the wall (extramuros). The extracted quartz grains had good luminescence characteristics and were identified as well bleached by comparison with feldspar infrared-stimulated luminescence (IRSL) ages. The optical ages showed that the deposition against the wall occurred in two different periods. The first was around 3.20 ± 0.07 ka ago (Iron Age), and the second was around 2.28 ± 0.08 ka (Hellenistic period). Human activity was evident because of the intermixing of cultural material with the presumed naturally deposited units.     

INFRARED STIMULATED LUMINESCENCE DATING OF ARCHAEOSEDIMENTS

In this study the feasibility of infrared stimulated luminescence (IRSL) for dating loessic colluvial sediments, deposited as a result of human activity, is investigated. It is demonstrated that the ‘IRSL-clock’is reset after a few minutes of daylight exposure even under foggy weather conditions. Samples of independently known age were taken at the Neolithic Bruchsal Aue site in south-west Germany. Different laboratory procedures are tried and a method is suggested (narrow wave-length band in the blue, preheat at 220° C for 5 min) to overcome age underestimates. This laboratory technique yielded archaeologically reasonable ages for colluvial sediments and hollow fillings. The IRSL-ages demonstrate that the deposition of the colluvia is related to soil erosion triggered by human activities, such as wood clearing and agriculture. The successful dating of archaeosediments has a great potential for applications in environmental archaeology as well as in geomorphology and in particular for the reconstruction of man-landscape interactions in central Europe.     

URANIUM‐SERIES AND LUMINESCENCE DATING OF VOLCANIC LITHIC ARTEFACTS*

U‐series dating was used to determine the growth rate of a feldspar‐to‐clay weathering rind in a mid‐Holocene Cascade–Olcott tradition andesite core, and luminescence dating (last exposure to sunlight) was used to date fine‐grained feldspars scraped from the surfaces of similar buried artefacts from a ¹⁴C‐sediment‐dated archaeological site (45KI464) on the wet western slope of the Cascade Mountains of Washington. For U‐series dating, we measured ²²⁶Ra excess (²²⁶Ra excess =²²⁶Ra –²³⁰Th) in five stratigraphic depth controlled rind scrapings. ²³⁸U, ²³²Th and ²³⁰Th were counted by alpha spectrometry, and ²²⁶Ra and ²¹⁰Pb were counted by gamma spectrometry on each sub‐sample.     

Luminescence dating of volcanogenic outburst flood sediments from Aso volcano and tephric loess deposits, southwest Japan

Luminescence dating has been applied to volcanogenic outburst flood sediments (Takuma gravel bed) from Aso volcano, Japan, and tephric loess deposits overlying the gravel bed. The poly-mineral fine grains (4–11 μm) from loess deposits were measured with pulsed optically stimulated luminescence (pulsed OSL) and post-IR infrared stimulated luminescence (pIRIR) methods, whereas the Takuma gravel bed containing no quartz, was measured with IRSL and pIRIR methods using sand sized (150–200 μm) plagioclase. The loess deposits date back at least to ～50 ka by consistent IRSL, pIRIR and pulsed OSL ages from the lowermost part of the loess deposits from one section. The ages obtained from the bottom part of the other loess section are not consistent each other. However, we consider that the pIRIR age (72±6 ka) which showed negligible anomalous fading is most reliable, and regard as a preliminary minimum age of the Takuma gravel bed. The equivalent doses (D_e) for the plagioclase from the Takuma gravel bed have a narrow distribution and the weighted mean of the three samples yield an age of 89±3 ka. This age is in agreement with the last caldera-forming eruption of Aso volcano (～87 ka) and it is likely that the pIRIR signal has not been bleached before the deposition. IRSL dating without applying pIRIR using small aliquots was also conducted, however, the IRSL signal shows no clear evidence of an additional bleaching during the event of outburst flood from the caldera lake.     

Investigations on the degree of bleaching of quartz OSL signals using modern aeolian dust from western Loess Plateau, China

Optically Stimulated Luminescence signal of quartz extracted from modern aeolian dust with known maximum age (about decades) was analyzed in terms of degree of bleaching. The results of dose recovery tests show that the modified double single-aliquot regenerative-dose protocol with the early background subtraction is robust for dating these modern dusts using small aliquots. Bleaching of these dusts is discussed based on the distribution of De values in histograms, scatter plots of D_e versus sensitivity corrected natural OSL signal and comparison between measured D_e and expected D_e. The results indicate that most dusts were completely bleached but some dusts were not completely bleached. For those incompletely bleached dusts in Lanzhou area, the maximum OSL age overestimation is up to ～1 ka, which might be caused by fast deposition accompanied by heavy sand/dust storms. The research suggests that cautions should be given to OSL ages younger than 1 ka in the western China close to deserts.     

EVALUATING CONTEMPORANEITY AND POST‐MORTEM AGE OF MALACOLOGICAL REMAINS USING SCLEROCHRONOLOGY AND DENDROCHRONOLOGY*

Sclerochronology and dendrochronology are based on series of incremental skeletal and cambial growths that have been correctly aligned in time via cross‐dating (i.e., incremental dating through synchronization of the growth variability on a visual and statistical basis). Here we report a study of the shell increments of two anthropogenic death assemblages of freshwater pearl mussel (Margaritifera margaritifera) originating from Finnish Lapland. First, we cross‐dated a new sclerochronology from an assemblage of river pearl mussel shells with unknown post‐mortem age. Second, this chronology was cross‐dated with the previously published sclerochronology of the same species and geographical source. Third, the composite of the two sclerochronologies was compared to dendrochronologies from Lapland. Temporal association of the two types of chronologies was markedly high, as confirmed by t‐values of 7.1 and 8.9, which are well above the acceptable limit of 3.5. Our study demonstrates the potential of sclerochronological and dendrochronological materials and techniques in assessing the contemporaneity (i.e., temporal overlap) and post‐mortem age of archaeomalacological deposits. In the palaeontological context, these results also show the wider possibility of dating shells with a large number of annual increments from natural deposits.     

Geoarchaeological studies on Roman time harbour sediments in Cologne — comparison of different OSL dating techniques

Due to the construction of a new North-South subway in Cologne, Roman time harbour sediments were exposed and were sampled for luminescence dating. A very good independent age control was given by the precise knowledge of the chronology of Roman activity and by radiocarbon ages of charcoal samples. Hence, different methodological approaches within luminescence dating were applied for Holocene heterogeneously bleached fluvial samples and were compared to the known ages. For one sample, optically stimulated luminescence (OSL) dating was applied to coarse-grained quartz using a single aliquot regenerative-dose (SAR) protocol. After D_e-measurements, different statistical approaches were tested (i.e. arithmetic mean, median, minimum age model, finite mixture model, leading edge method and the Fuchs and Lang approach). It is demonstrated that the Fuchs and Lang approach along with the leading edge method yielded the best matching OSL ages with respect to the known ages.     

DIRECT GAMMA-RAY SPECTROMETRIC DATING OF FOSSIL BONES: PRELIMINARY RESULTS

A non-destructive gamma-ray spectrometry was used for direct dating of fossil animal and human bones taken from an alluvial Quaternary site in northern Italy. The ages were obtained using the U-Pa method: the chronological results are in agreement with palaeofaunal indicators and taphonomic data. The ages of two samples showed concordance with the U-Th method; in a third case good agreement was also found with the radiocarbon date, thus lending credence to our modified gamma spectrometric method.     

AUTO-REGENERATIVE THERMOLUMINESCENCE DATING USING ZIRCON INCLUSIONS

Fired ceramics containing zircon inclusions have been dated by allowing the zircons to regenerate their own thermoluminescence (TL) signal, hence auto-regenerative TL dating. The technique is conceptually straightforward. One first measures the TL accrued since the last heating of the material. The zircon grains are then stored for six months and the TL signal regenerated through self-irradiation is measured. Since the internal dose-rate for zircon is dominated by the internal component the age of the sample is simply given by the ratio of the natural to auto-regenerated signal times the laboratory storage period. The technique, however, requires the measurement of a very small auto-regenerated signal, which introduces a number of experimental and physical complications. The methods for overcoming these problems and successfully dating zircons by auto-regeneration are described.     

The effect of fluvial environments on sediment bleaching and Holocene luminescence ages — A case study from the German Alpine Foreland

This study investigates the potential of luminescence to date deposits from different fluvial sedimentary environments; namely point bar deposits, sandy and silty channel fills and floodplain sediments. Samples were taken from Holocene (<5 ka) terraces of the Lech and Danube rivers, for which independent age constraint is available through ¹⁴C ages, archaeological data and historical maps. OSL-ages were obtained using small aliquots of coarse grain quartz for the majority of samples. Two further samples were dated by the IRSL-signals of polymineral fine grain extracts, as no sufficient number of coarse grains could be extracted from these sediments. In order to detect and ac-count for incomplete bleaching, we used the decision process suggested by Bailey and Arnold [Statistical modelling of single grain quartz De distributions and an assessment of procedures for estimating burial dose. Quaternary Science Reviews 25, 2475–2502, 2006]. Although their model was designed for single grains of quartz, our study shows that it is also applicable to multiple grains of quartz, pro-vided that a low number of luminescent grains is present on one aliquot. Luminescence ages of point bar deposits and a sandy channel fill correspond most closely to the independent age control. In the floodplain, sand-striped floodplain channel deposits were incompletely bleached to a moderate degree, yielding ages with acceptable overestimations, while fine-grained floodplain deposits were worst bleached. One crevasse splay deposit was so severely incompletely bleached that none of the age models was able to yield accurate ages.     

METHODOLOGICAL DEVELOPMENTS IN THE LUMINESCENCE DATING OF BRICK FROM ENGLISH LATE‐MEDIEVAL AND POST‐MEDIEVAL BUILDINGS*

Fired clay brick samples, obtained from a group of seven high‐status late‐medieval and post‐medieval buildings in England ranging in age from c. ad 1390 to 1740, were dated by the luminescence method using an optically stimulated luminescence (OSL) technique. The results obtained indicate that, when applied to quartz extracted from brick, the technique is capable of producing dates that are in consistently good agreement with independent dating evidence for the buildings. For six samples taken from a group of four dating ‘control’ buildings the mean difference between the central values of luminescence and assigned ages was 5 ± 10 years (SD, n = 6). The methodology used is appropriate for application to other standing buildings in other temporal and geographic regions, and may be used with confidence where conventional dating methods are less certain. The study also examines the luminescence characteristics of quartz and the characteristics of the lithogenic radionuclides in brick samples and identifies various aspects related to the assessment of experimental uncertainty in testing the reliability of the method.