Recent developments in Quaternary dating methods

Abstract. An overview of methods to numerically date continental Quaternary deposits is provided including radiocarbon, Uranium series, cosmogenic nuclides, luminescence and electron spin resonance. Physical background and methodology are briefly described and potential problems as well as recent developments in the field are highlighted.


Recent developments in Quaternary dating methods
Irka Hajdas, Susan Ivy-Ochs, Zürich, Robyn Picker¬ ing, Frank Preusser, Bern 1 Introduction Determining numerical ages for artefacts, sediments and biological remains is a major aspect of Quaternary research. Both accurate and precise numerical ages are important as, in contrast to older geological periods, the Quaternary comprises of a very short period of time 2.6 Million years). Over such a brief time, species evolution is generally too small to allow for biostratig¬ raphy establishment as a ubiquitous tool for correla¬ tion. This is particularly pertinent as the time spans of interest are often relatively short on a geological scale: often a few hundred years or less. Furthermore, conti¬ nental deposits are of particular interest in Quaternary research, perhaps more so than in most fields of classical geology. Firstly, a major focus of Quaternary research is on the human genetic and cultural evolution, as well as our ancestors interaction with climate and environ¬ ment. Findings related to human activity are, however, mainly restricted to continental settings. Secondly, Quaternary topics, such as palaeoclimate and palaeo¬ seismic research, are often carried out in the context of past and potential future hazards for mankind, hence, in areas that are or have been populated. Most con¬ tinental records are, due to their sedimentary nature, fragmentary and bear only rarely fossil remains that fulfil the necessary requirements of biostratigraphic markers. Secure correlation of different findings is thus often possible by direct numerical dating only.
Quaternary geochronology was established with the advent of radiocarbon in the 1950 s and several other methods have been developed since then. Methodologi¬ cal developments have been rapid, especially during the past two decades. It is the aim of this article to give a brief overview of the most relevant of those methods available in the context of continental Quaternary research Table  1). Its purpose is to introduce the non-specialist to poten¬ tial methods and their limitations with regard to establish¬ ing chronologies for continental records. A much more detailed summary is provided by a series of recently pub¬ lished review papers edited by Preusser et al. 2008).
2 Brief overview of different methods

Radiocarbon
The cosmogenic radioactive isotope of carbon 14C or radiocarbon) is produced in the atmosphere, quickly oxidized to CO and CO 2 and well mixed within the atmosphere. The constant exchange between the atmos¬ phere and other carbon reservoirs builds the basis of the 14C dating method. Living organisms biota) incor¬ porate the 14C during photosynthesis plants) or via the food chain animals). Equilibrium between input and decay of 14C exists in the living tissue. The cutoff of 14C uptake death or deposition of the inorganic carbon) marks the time zero of the radiocarbon clock and 14C activity measured in the material is used to calculate the time that elapsed since that event. The half-life of 5730 ± 40 yr sets the limit of the 14C dating method when approaching ages older than 50 000 BP.
Calculation of the radiocarbon ages requires informa¬ tion about initial 14C content at the time of deposition or death. Due to the temporal variability of the atmos¬ pheric 14C content effects of variable production rates and exchange rates in global carbon reservoirs) this value is only assumed to be constant; it must later be corrected in a calibration process that is using an experimentally constructed calibration curve radio¬ carbon ages or activities in the past are placed on the calendar time scale). Calibration of radiocarbon ages is a mathematical procedure, applied in calibration programs, that finds the calendar ages corresponding to the measured 14C age. Dependent on the nature of the calibration curve this may result in a wide range of calendar ages and/ or multiple intervals, thus reflecting another limitation of the 14C method.
Nearly sixty years after first radiocarbon ages were published radiocarbon dating is still a dynamic field with a wide range of applications mainly in geo¬ chronology but also in environmental and biomedical studies. Recent advances in sample preparation and measurements techniques allow new applications and much higher throughput of samples measured. These developments lead to higher resolution chronologies of records, providing the option for dating various frac¬ tions of carbon in studied material, which allows better understanding of contamination problem. Examples, which are of direct interest to Quaternary studies of the last 50 000 years, include new methods of prepa¬ ration bones or treatment of wood and charcoal sam¬ ples Hajdas et al. 2007, and references therein). The advance in preparation techniques helps to reduce the contamination of sample with modern or old carbon that could interfere with measurement of the correct 14C activity remaining in a dated sample.
The changes in the atmospheric content of 14C Figure  1 shows the present INTCAL04 calibration curve Reimer et al. 2004). The tree ring based chronology is constantly pushed back in time beyond the Holocene into the Late Glacial. Records such as trees, lake sedi¬ ments, deep sea sediments, corals and stalagmites have formation of carbonate to be dated by other than radiocarbon dating methods. As a consequence of a growing interest in the calibra¬ tion of very old radiocarbon ages 20 000 to 50 000 BP) Ramsey et al. 2006) the future update of the INTCAL04 data set will provide an estimated calibra¬ tion curve for the last 50 000 to 55 000 years.
2.2 Uranium series The Uranium and Thorium decay series contain radioactive isotopes of many elements, among them U, Th, Pa and Ra, and end in stable Pb. In a closed system the intermediate isotopes exist in so-called secular equilibrium, that is, their abundance ratios are such that each isotope decays at exactly the rate AD* ca. 50 000 years at which it is produced. Various natural processes cause the fractionation of these elements, disturbing these equilibria among isotopes. The gradual return to equilibrium conditions allows for a quantification of the time passed since the fractionation causing event. In the case of U-Th dating, Th is separated from parent U due to the vast difference in the sol¬ ubility between these elements in water. Thus, cal¬ cite precipitates with negligible amounts of Th. The subsequent in situ decay of 234U to 230Th can be used to date material by measuring the present levels of both isotopes by mass spectrometry Bourdon et al. 2003). Uranium series has been shown to be a pre¬ cise dating tool on both marine coral) and terres¬ trial stalagmites, flowstones, tufa) carbonates dating back to 500 000 years before present Edwards et al. 1987). U-Th dating has been critical in palaeoclimate studies as it can provide ages for past sea level high stands from corals. High resolution records of ter¬ restrial climate change recorded in stalagmites are greatly enhanced by the ease of U-Th dating of such material.  are produced in exposed minerals in rock surfaces and sediment due to interactions with secondary cosmic ray particles. The length of time that a rock surface has been exposed can be determined by measuring the concentrations of these nuclides Gosse & Phillips 2001). 10Be, 14C, 21Ne, and 26Al are measured in quartz, 3He and 21Ne in olivine and pyroxene, while 36Cl is measured in whole rock samples of any rock type.
An amazing array of landforms can be surface expo¬ sure dated with cosmogenic nuclides making them an indispensable tool in geomorphology. In addition to the determination of exposure ages of boulders on moraines and landslide deposits, which in many cases can be dated by no other method, innovative ongo¬ ing and future studies include the dating of terraces using depth profiling, elucidation of differential rates of bedrock erosion under past glaciers and the dating of deeply buried sediments with multiple cosmogenic nuclides burial dating). Surface exposure dating relies heavily on detailed field mapping and careful observa¬ tion during sample selection. Focus is on rock surfaces that are the most representative of the age of the land¬ form and the least affected by post-depositional dis¬ turbances. Difficulties such as rock surface weathering, post-depositional boulder instability or pre-exposure, nevertheless, must always be kept in mind. Data inter¬ pretation is made in light of regional morphostratig¬ raphy and results from other dating methods for the same or related sites.
With respect to methodology, reduction of back¬ grounds reflecting improvements in both sample preparation procedures and measurement techniques accelerator mass spectrometry) allows the dating of rock surfaces that have been exposed for only a few hundred years, for example those created during his¬ torical events such as landsliding or glacier advance). At the other end of the spectrum, in slowly weather¬ ing environments exposure ages up to many millions of years are measured Schaefer et al. 1999). In the Alpine realm obtaining exposure ages older than sev¬ eral hundred thousand years is unlikely Ivy-Ochs et al. 2007). The limit is given by the rate of rock surface weathering, which is several mm/ ka in the Alps. Geo¬ morphologists await the addition of 14C measured in quartz to the set of available cosmogenic nuclides, which will prove especially useful for young and/ or rapidly weathering landforms.

Luminescence
This method utilizes a signal in natural minerals such as quartz and feldspar that is sensitive to heat and light. The luminescence signal used for dating originates from the interaction of naturally occurring radioactivity with the crystal lattice of minerals cf.
Aitken 1998). Radiation energy is stored within the lattice during burial and is erased once the mineral is exposed to certain stimulation energy. The event being dated is i) mineralisation, ii) last heating to ca. 450°C, or iii) last daylight exposure of the mineral grains. Problems associated with luminescence dating arise from incomplete zeroing of the luminescence signal during the resetting event e. g. daylight exposure), changes in sediment radioactivity caused by radioac¬ tive disequilibrium e. g. in some carbonate or organic¬ rich sediments), changes in sediment moisture and instability of the latent luminescence signal, in particu¬ lar in a variety of feldspars. The dating range is high depending on the properties of the material being investigated, in particular the amount of radioactivity within the sediment, and the potential of minerals to store the radiation damage.
While a decade ago these problems left the reliability of some dating results rather questionable, the per¬ formance of luminescence dating has significantly improved during the last decade with the introduc¬ tion of new measurement techniques single grain laser-stimulation) as well as improvements in meas¬ urement protocols single-aliquot regenerative-dose).
Recent work is focusing on how to further improve the accuracy as well as the reproducibility of lumines¬

Electron spin resonance ESR)
The origin of the ESR signal is similar to lumines¬ cence and both methods are partially closely related with regard to methodology. The major differences are the approach to signal stimulation in ESR a magnetic field is being used) and the material being investigated. Most applications of ESR have been either carbonates such as corals and molluscs Schellmann et al. 2008) or tooth enamel Grün 2006). The event being dated is the formation of the mineral. With these materials, uptake and loss of radioactive elements, in particu¬ lar from the Uranium decay chain, and re-crystallisa¬ tion from aragonite to calcite are specific problems. Recent research has also been focussing on dating quartz, especially to use the lightsensitive part of the ESR signal in a similar way as is used in luminescence dating. First results look promising but it may need a few more years before all methodological problems are fully explored; the method should be hence con¬ sidered as still in an experimental stage. Despite all limitations, a variety of papers have reported a good consistency of ESR ages for the different materials mentioned above with independent age control. If the problems are carefully addressed the method allows dating from a few hundred or even less) to several hundredthousand years, and possibly even beyond one million years.

Conclusions
The progress in dating methods during the past two decades has significantly improved the possibility to constrain the timing of geological processes, climate, environmental change and human evolution during the Quaternary. Due to the availability of modern dating approaches parts of Quaternary history have had to be re-written and we now have a much better understanding of the interaction of environment, climate and humans in the past. Nevertheless, all of the methods mentioned here still have potential for improvement with regard to precision, accuracy, age range and materials suitable for dating and these are presently being explored.