bar_about.gif (903 bytes)bar_gallery.gif (575 bytes)bar_home.gif (658 bytes)bar_links.gif (469 bytes)bar_members.gif (607 bytes)bar_projects.gif (687 bytes)bar_services.gif (552 bytes)


General Overview of Dating Methods

 

Radiation Exposure Dating Methods: Dating by electron spin resonance (ESR), thermoluminescence (TL) , and optically stimulated luminescence (OSL) are all based on the measurement of trapped electronic charges that accumulate in crystalline materials as a result of low-level natural radioactivity present at sites (Rink; 1997, 1998). The three methods require materials whose trapped charges were zeroed at the date of site occupation, either by heating (as in the case of TL dating of burned flint), by exposure to daylight (OSL dating) or by growth de novo, as in the case of ESR dating of teeth. The total radiation exposure ("dose") received by the sample is estimated by laboratory measurements, while the levels of radioactivity (annual "doses") in sites are determined by testing in areas near finds of datable materials . Ages are obtained as the ratio of total dose to annual dose, where the annual dose is measured at the site and assumed to have been similar in the past. By use of an isochron method we may in some cases be able to determine the true average annual dose over the entire burial history, and avoid problems of variation in annual dose with time.

 

Electron Spin Resonance Dating (ESR): Fossil teeth are a ubiquitous component of prehistoric sites, and as a consequence, ESR dating of tooth enamel is very widely applicable in archaeology and palaeoanthropology. Since publication of the first papers on dating of sites in Israel (Schwarcz et al., 1988; Schwarcz et al., 1989) electron spin resonance (ESR) dating of tooth enamel has been recognized as a useful tool for chronometric dating in the time range beyond the 40 ka limit of radiocarbon and up to at least 2 Ma (Schwarcz et al. 1994). It takes its place alongside U-series and luminescence dating as one of the principal dating methods for this time range. General descriptions of ESR dating of tooth enamel are given by Grün et al. (1987), Grün and Stringer (1991), and Schwarcz and Grün (1992). Significant levels of uncertainty in the ESR age arise through problems associated with uranium uptake into the enamel, which in many cases can be reduced through additional dating of the teeth using MS/U-series dating. Here, the rate of change of radioactivity inside the tooth can be better understood, and appropriate corrections to the ages can be made.

 

ESR Dating: Tooth samples are obtained from loci in sites at which we can also determine the local annual radiation dose. Enamel is irradiated to different doses using 60Co g -rays, and the ESR spectrometer is used to measure the increase in signal with dose, which is plotted as a function of dose. The past dose is estimated from the equivalent dose (DE), which is given by the dose intercept of the line showing the incremental growth of the signal as a function of dose. The age is effectively given by the ratio of DE to the average annual dose, d, during burial; d consists of an internal and external component. The internal, and part of the external annual dose is determined by the content of uranium (U) in the enamel and dentine, which we measure by neutron activation analysis: (NAA). Since living teeth contain no U, we must take into account the way in which U has been taken up since burial. Two limiting models are commonly used: EU = early uptake; and LU = linear, continuous uptake. By measurement of the U-series isotope ratios of the teeth (230Th/234U, 234U/238U) using thermal ionization mass spectrometry we can estimate the mode of U uptake, and the consequent self-irradiation history of the tooth. The external part of d can be determined in various ways (see below). The coupled ESR/MS/U-series age is computed using these data, other site information (water content, depth, etc.), as well as corrections for self-absorption of radiation by the enamel. Although MS/U-series dating of the teeth adds significantly to the amount of work (and cost) of the dating, we and other workers (R. Grün, ANU; C. Falgueres, Inst. de Paleontologie Humaine, Paris) have found that this is necessary whenever the internal component of d is a significant fraction of the total (and consequently, the EU and LU ages differ significantly). Generally it suffices to do U-series analyses on a representative subset of the teeth from a site, as the uptake histories of all teeth in a given stratigraphic level are usually similar. However, teeth from different levels in a site can have significantly different uptake histories, as we found at Krapina (Rink et al., 1995).

 

Optically-Stimulated Luminescence Dating (OSL): Although thermoluminescence dating of sediments was well underway in the 1970's, problems associated with incomplete zeroing made the technique difficult to apply. Huntley et al. (1985) and Hütt et al. (1988) opened a new era of optical dating, showing that quartz and feldspar grains in sediments exhibit luminescence signals which are rapidly and fully zeroed by a few minutes of day light, thus avoiding the chief problem of TL dating. This led to widespread application of OSL techniques to windblown and waterlain sediments, which are of special importance to archaeological sites in the time range between 1 and 800 ka. Quartz grains can be dated from the older sites, while feldspar grains may only have a usable range out to about 150 ka. These methods depend on using the present-day annual doses as estimates of the external annual dose in the past.

 

 

Optically-Stimulated Luminescence Dating (OSL): Here the same technique as TL is applied except the sample grains are exposed to light rather than heated. Higher precision on laboratory determinations of the total radiation exposure (dose) have been achieved (Rees-Jones, 1995) using very small quartz grains (.004 to .011 mm diameter) and by use of a single aliquot technique for feldspar (Duller, 1994). The latter requires less sample preparation. For colluvial samples and certain waterlain samples that might be insufficiently zeroed, a test for complete zeroing is available (Li, 1994). Single-aliquot procedures for quartz are also under development (Murray et al., 1996).

 

Thermoluminescence dating (TL) of burned flint has gained wide acceptance as one of the most reliable dating methods for prehistoric archaeological sites (e.g. Mercier et al, 1995). It can be used over a time range from about 5 to 1000 ka. We are collaborating with N. Mercier and H. Valladas on TL dating of burned flint from sites in Israel and the Crimea, using their method for age determinations. Although burned flint is relatively rare in many archaeological sites, we have found that it is an invaluable tool for crosschecking our ESR ages. In sites that have plenty of teeth for ESR dating, but which show complex U-uptake histories, age estimates will have large uncertainties. These can be significantly reduced using burned flint, even if only a few samples are available.

 

 

Thermoluminescence Dating (TL): (Valladas, 1994) Larger pieces of flint which appear to be burned are tested by TL to determine whether the temperature and duration of the ancient heating was sufficient to completely zero the trapped charge throughout the entire sample. After doses of radiation (as above) the crushed grains of flint (primarily quartz) are heated and the emitted light (thermoluminescence) is measured and plotted as a function of dose. DE is measured as above. The same aspects of dosimetry apply as above except there is no uranium uptake problem. Both alpha and beta(or gamma) radiation dose response are routinely measured since the chemical composition of flint is not homogeneous

 

Mass Spectrometric U-Series (MS/U-series) Dating of Calcite and Tooth Materials: This method is not based on radiation exposure, but rather on the decay of uranium into thorium over time. Changes in soil radioactivity or moisture content do not lead to uncertainties in ages calculated by this method. The difference between its use in calcite and on tooth materials is explained in Appendix 1.

 

Mass-Spectrometric Uranium Series Dating (MS/U-series) of Calcite and Tooth Materials: Calcites which are intercalated with archaeological layers are excellent time markers in the time range 1-400 ka. Natural calcite, a uniquely valuable material for precise age determinations, is formed by evaporation of calcium-rich waters which drip into archaeological contexts. Two different isotopes of uranium in the dripping water are incorporated into the newly-formed crystals, but without any of the daughter thorium atoms. The decay rates of these isotopes are well known. The mass spectrometric approach (Edwards et al., 1996) determines the parent-daughter ratios with much higher precision than the older technology of alpha spectrometry. Therefore, precise ages (± 1% 2s ) can be obtained with MS/U-series dating. In the case of tooth enamel (see above), we obtain the average age of the uranium which may enter teeth slowly over time and in some cases be partially leached back out. Therefore the significance of the U-series "age" depends on the mode of U-uptake.

 

Isochron variant of ESR and Thermoluminescence Dating: In some cases the internal radioactivity of the sample itself is an important component of the annual dose. This is used to advantage by selecting samples with variable internal dose. This allows one to extract the true average external annual radiation dose which might vary in the site over time because of changes in moisture content, or because of changes in soil mineralogy. Geochemical alteration is now known to mobilize radioactive potassium through some sites rich in ash, through a process which leads to changes in soil mineralogy (Schiegl et al., 1992). This added uncertainty in the external annual dose can be avoided by studying the mineralogy and using the isochron technique. We can thus avoid the assumption of constant moisture content and fixed mineralogy, as well as problems associated with "lumpiness" in the radiation field. Application of this method to teeth (Blackwell and Schwarcz, 1993) and burned flint (Aitken and Valladas, 1992) are discussed in Appendix 1.

 

Advantages of Combined Use of Methods

The intercomparisons between ESR and other dating methods at over 30 sites (see above) shows that the basic assumptions of the ESR dating method are reliable. However, further research is needed on site-specific problems such as "lumpy" radioactivity in sites and changes in annual radiation doses over time due to 1) uranium uptake or loss, and 2) fluctuations in water content.

In the time range below 40 ka, ESR and luminescence methods should now be considered as alternatives to and comparators against 14C. Despite the excellent AMS 14C dates which now exist for many younger Palaeolithic sites, we believe it is important to date sites using other methods, especially where AMS 14C ages are near 40 ka. No calibration for dates in this time range is available and there can be serious problems due to contamination of samples with younger carbon. Our recent work (see Starosele and El Castillo above) clearly shows that ESR and coupled ESR/MS/U-series dating can be reliably used between 30-45 ka. Beyond the 14C time range, a combination of any of the four techniques mentioned in the objectives provides a very powerful dating approach. Long term changes in both moisture content and in sediment radioactivity, which previously could not be detected through standard application of these methods, can now be handled with ESR or TL isochron dating . Where materials needed for these approaches cannot be obtained, the next best approach is to use a combination of methods which depend to different degrees on moisture content and soil chemistry. Wherever possible, one should use MS/U-series on archaeological calcites, since it does not depend at all on these variables. Since most sites lack this valuable material, they would be dated through a combination of radiation exposure methods. When doing so, it is better to choose one where the internal dose is large, because any external annual dose fluctuations due to moisture will then have a smaller effect on the age calculation. A good example of this is the use of OSL on potassium feldspar, because the radioactive potassium content insures a large internal dose to the grains. For sites where single methods based on radiation exposure are the only option, our philosophy is to calculate the ages using different assumed moisture content values (Rink et al., 1996 a, 1996b, 1997), so as not to inflate the significance of ages calculated based on present day moisture content.

 

 

Isochron Variant of ESR and Thermoluminescence Dating: This method is applicable wherever multiple samples with different internal dose rates have all been exposed to the same external dose rate d-ex. The total dose DE is given by DE = t ([d-ex] + [d-in]) = [DE-ex] + t(d-in).

This is a linear relation; a graph of d-in vs. DE should have a slope = t (age), and an intercept of D-ex = t(d-ex) at d-in = 0. At DE = 0, the d-in intercept is -(d-ex), the external annual dose. This is the time-averaged annual external dose. This technique allows us to avoid the assumption of an external annual dose based only on present-day measurements in the site, which might not account for "lumpiness" or changes in water content over time. This method would be applicable to multiple enamel subsamples from a single tooth or to burned flints buried close together in homogeneous sediments

 

Objectives

* To help archaeologists and anthropologists determine the ages of Lower and Middle Palaeolithic cultural materials and hominid remains from Europe, West Asia, Africa and East Asia.

* To apply the multiple state-of the-art dating methods of coupled ESR/MS/U-series dating of tooth enamel, TL dating of burned flint, OSL dating of quartz and feldspar in windblown and waterlain sediments, and MS/U-series dating of archaeological calcites.

* To pursue basic research on identification and characterization of geochemically-altered tooth enamel, which should provide better criteria for sample choice in ESR dating.

* To improve the methods for uranium uptake modelling in ESR dating through continued application of the coupled ESR/MS/U-series approach in sites and through improvements in computer software.

 

 

Significance: The significance of this proposal is that we are uniquely able to provide multiple dates in sites which remained impossible to date before the advent of these techniques. Although the combination of methods is limited by the materials available in a site or group of sites, we show how particular combinations reduce the overall uncertainty associated with age estimates. This would allow a more secure chronology to be erected for a particular archaeological or anthropological question.

As practitioners of these methods, we recognize their tremendous capabilities and their limitations. Although a growing number of geologists have accepted the general reliability of these new techniques because of evidence from dating intercomparisons, the anthropological and archaeological communities remain somewhat more sceptical. This is not surprising since many of the findings have challenged long-standing views about human evolution. (e.g. Valladas, 1988 or Swisher et al., 1996). Therefore we continue to test and refine the method through research on "problem" samples, and we propose to apply the very latest variants of the techniques which should reduce the number of assumptions involved.

The ESR and luminescence methods are "in-situ-intensive", thus concurrent dating and excavation are crucial. Witness sections, if present at all, are often inaccessible or poorly documented, or unstable, and dating has often proven problematic because few museum samples can be reliably referred to the radioactivity levels in usable witness sections. Working at sites under current excavation allows us to 1) observe detailed relationship between dated material and archaeological elvers, 2) obtain radioactivity surveys at points closest to dated samples and 3) discover all possible relevant datable material at the site. Furthermore, radiation surveys may prove invaluable for future application of improved variants of the radiation exposure dating methods.