Electron Spin Resonance (ESR) Dating Method And Its Contribution To Chronological Study Of Early Pleistocene Archaeological Sites In China

The study of Early Pleistocene archaeological sites in East Asia can help to solvethe questions of when hominins first left Africa and how they might have settled inAsia. Some earliest hominin evidence in China were not widely accepted mainly dueto chronological uncertainty, since most of them are beyond the classic radioisotopicdating range and lack of volcanic materials for K-Ar and Ar-Ar dating. CombinedESR/U-series dating could solve this problem, as its application to fossil tooth has apotential of dating the archeological sites older than1Ma, depending however on theaccuracy of reconstructing uranium uptake history. An U-diffusion parameter pshould be defined for each dental tissue and the obtained US-ESR model age is lessuncertain than the conventional EU (early uptake) and LU (linear uptake) ESR modelages.In this work, combined ESR and U-series dating method of fossil tooth has beenused to date two Early Pleistocene sites of China, Longgupo, Chongqing, SouthwestChina and Donggutuo, Hebei, North China. These two sites may represent the earliesthuman settlement in each region of China. The Longgupo site, once a cave, has acomplex deposition environment which makes the previous dating work problematic.The purpose of this thesis includes establishing a precise chronological framework ofthe stratigraphic sequence of Longgupo which based on the recent joint Chinese-French excavation in2003-2006. For the Donggutuo site, the dating work of thisthesis was expected to be checked by the well-established paleomagnetic date in orderto evaluate the dating method used in this study.This thesis presents the feasibility of applying combined ESR/U-series method ondating Chinese early Pleistocene sites, and the result of this study places thechronology of the Longgupo site between1.4and1.8Ma. With respect to thetechniques development of the stone artifacts unearthed from the site, Longgupo isstill one of the earliest evidence of human settlement in southern China. This study shows that this combined method provides the availability of dating theold archaeological sites, but some challenges are also brought forward and addressed:(1) Paleodose determinationThe paleodose reconstruction is the main problem not only for tooth dating but alsofor quartz minerals studied by the ESR method. In combined ESR/U-series dating, theuncertainty on DEis the main error source of the final result. The determination of DEof the samples from old sites is more difficult because of the high dose valueaccumulated during the history. As regeneration protocol used in luminescence studiescannot be applied on tooth enamel without the destruction of the ESR traps, theextrapolated additive dose method is used for DEdetermination, and the dose responsecurve is simulated with certain fitting functions. Which function is morerepresentative and close to the true value of paleodose of the sample, especially forthe old tooth samples were studied in this thesis.Three different fitting functions were investigated: single saturation exponential(SSE), exponential plus linear (EPL) and double saturation exponential (DSE), inorder to find the better solution for DEdetermination of old tooth samples. This studyshow that conventional SSE function is unambiguous not suitable for fitting doseresponse curves of old tooth samples from Longgupo and Donggutuo sites and itsfitting curve shows a large deviation with analytical points and higher fitting errorsthan the other two functions. EPL and DSE fittings are the extensions of SSE, ofwhich both have their own possible physical meanings, and have been appliedpreviously to some other dating materials. EPL function was used to provide higherDEvalues than DSE one, but the results are in general close to each other in the errorrange. The DSE function gives the lowest DEamong the three fitting but relative largeerrors because more fitting parameters were introduced in this function. The adjustedcoefficient of determination (adjusted R2) which indicates the fitting effect shows thatthe DSE function fit best to the experimental dose points. According to the work ofthis thesis, it could conclude that the conventional paleodose fitting method (SSE) isnot applicable to date the old tooth. Double saturation exponential function is more acceptable and exponential plus linear one is sometimes a compromise solution in thecase that the number of aliquots is not enough to reduce the uncertainty of DES.(2) External gamma dose rateThe importance of external gamma dose rate was addressed in this study, especiallyfor the Longgupo site. This issue is quite essential to ESR and TL dating studies ofarchaeological sites, which are often enriched with fossils and stone artifacts andcorrespond then to typical “lumpy” site. The external gamma dose rate measuredtoday has always a discrepancy with the past value because of the complex variationsof sedimentological history and hydrological environment changes.In some cases, when the sample layer is full of bone fragments, the external doserate is variable because the surrounding bones would absorb uranium from theenvironment. In this study, the external dose rate contributes greatly to the total doserate of Longgupo tooth samples (more than50%in general) and the measuredexternal dose rate values have a big influence on the age calculation. Unfortunately,the in situ gamma dose rate was not available for some of the studied samples leadingto an inverse correlation of the calculated ages with the stratigraphic sequence and toa discrepancy with biostratigraphy evidence. In view of the above-mentioned facts, Iwent to Longgupo again in October,2011in order to make a more precisemeasurement of the external dose rate on the field. Although this work is moredifficult nowadays after five years of previous excavation when the samples werecollected, it is necessary not only for the dating study but also may have animplication of the depositional environment changes of the site.(3) Uranium uptake historyThe uranium uptake history in the tooth is the key issue for the determination of thedose rate. The advantage of combined ESR/U-series dating is to reconstruct it by theuse of a diffusion parameter rather than by artificial assumptions. In the present workthe uranium uptake was not the main problem for Longgupo samples because the dominant contribution of the external dose rate was related with the recent uptakehistory. In contrary, uranium uptake and/or leaching have a key effect on the agecalculation of Donggutuo samples. This is partly due to the high uranium content ofthe samples and to the high isotopic230Th/234U ratios.The two old sites studied here correspond to two types of sites: a cave site(Longgupo) and an open site (Donggutuo). The difference of uranium uptake historybetween these two sites can probably be explained by the different depositionalhistories. Longtupo was systematically excavated after its discovery and, although thestratigraphy of the site is complicated, the sedimentary stack was not disturbed a lotbecause of the protection of the carpeted breccias on the slope. In contrast, theDonggutuo site and a couple of other localities in the Nihewan basin were exposed inthe open air by recent erosion. With a changing erosion system, there is also a changein the hydrological environment and the erosion process is commonly accompaniedby drainage re-activation and lowering of the ground water table. This goes alongwith renewed uranium mobilization. Uranium leaching always makes a difficulty foruranium reconstruction. For the moment, there is no better solution for dating thefossil tooth with uranium leaching, and a practical solution is to check systematical U-series mapping to identify the most promising domains for combined ESR/U-seriesdating.Dating Early Pleistocene sites is always a challenge, the combined ESR/U-seriesdating method provides a possibility and it has the best potential for dating old fossilsamples from old sites. For the estimation of a reliable dating result, a number ofquestions have to be addressed other than the issues mentioned above, such as theestimation of large dose values and the difference between natural and laboratoryirradiation. The environmental dose rate is always a problem for dating “lumpy” sites,and it should be measured in situ when such measurements are possible. This studymay provide a choice of dating the old sites in China. In combination with otherdating methods, such as the paleomagnetism or cosmogenic nuclides method, thecombined ESR/U-series fossil dating technique will help to create a more precisechronological framework of ancient human settlement in China.