Mathieu Duval

Mathieu Duval

I am an ARC Research Fellow in ARCHE at Griffith University. I am specialized in Quaternary Geochronology and in the dating of the early human occupations around the Mediterranean.

Location Brisbane, Australia

Activity

  • This is an excellent conclusion Catherine, and I must say that I have reached the same!

  • @AlvaroA @RobBruce ahahah, I hope so too!

  • Thank you very much

  • from outside AND from inside ;-). There is an internal component that has to be considered too.

    C-14 and U-series are usually called radiometric methods, based on the evaluation of the radioactive decay or growth of selected radioelements. ESR is instead a palaeodosimetric method, for which the effects of radioactivity are quantified instead.

    You are...

  • @ErinTaylor Not really, there are indeed environments with very low radioactivity, but still, it always remains measurable.
    Any kind of sediment has naturally radioactive elements such as Uranium, Thorium and Potassium, although their concentrations may be quite variable depending on the environment considered.

  • Most of the fossils are quite fragmented, and not often found in anatomical connection, so this situation is not frequently faced.

  • @ChristianeGuinot Merci pour l'information, c’est intéressant. Les thématiques abordées semblent un peu différentes et donc complémentaires du nôtre.

  • Hello Drosia, that's right. You know the current amount of C-14 left in the fossil and the decay rate, so it is possible to estimate when the organism died. Then, you will have to put this (uncalibrated) C-14 age on the calibration curve in order to obtain the calibrated age.

  • The age uncertainty in ESR is indeed larger than for U-series and C-14. U-series and C-14 are both radiometric methods based on the measurement of the decay or growth of radioactive elements in a sample. So, basically the number of sources of uncertainty is quite limited, and the resulting age error is therefore much smaller than ESR.
    In comparison, in ESR...

  • @DROSIACHARISI excellent question! Well, U-series method is clearly easier and faster than ESR. In particular, laser ablation U-series does not require any chemical preparation of the sample. And several tens of U-series dates can be obtained in a day. In contrast ESR is much more time consuming. It usually takes me months to process a tooth and obtain an ESR...

  • @HARRYOWHODO That's a good question. There are several ways to identify uranium leaching in a tooth.

    The first and most obvious evidence of uranium leaching is when the activity ratio Th-230/U-234 is higher than secular equilibrium. In other words it means that there are more daughter products (Th-230) in the tooth than parents (U-234), which is...

  • Question 2: We do not use a calibration curve per se. The enamel sample is divided into several sub-smaples that are gamma irradiated in the laboratory. So for each aliquot we know the gamma irradiation dose that has been given. The unknown here is the radiation dose absorbed by the enamel over geological times. Further details about the dose reconstruction...

  • Hello Cian,
    apologies for this late answer.

    Question 1: Knowing the size of the sample is essential in order to evaluate how ionizing radiation is attenuated when it goes through the enamel layer. Alpha particles have an approx. range of penetration in the enamel of a few tens of micrometers, while it is of about 2 mm for beta particles. Consequently,...

  • And other methods like palaeomagnetism or biochronology can be used to, although they do not provide a numerical age (a number).

    Palaeomagnetism can indicate the magnetic polarity of the sediment, which can then be tentatively correlated with the Global Polarity Time Scale.

    Biochronology is more like a relative dating method. By evaluating the absence...

  • Good question Rachel. Stratigraphy is a useful tool, because it enables to obtain a relative chronology. But there are several dating tools that can be used to date Miocene fossils (older than 2.6 Ma).

    Argon-Argon is one of them for example. It is a very precise method but it is applied to volcanic minerals. So you can only use it in volcanic context. This...

  • Thank you Walter

  • This is an excellent summary Pete!

  • @AmeliaHare Good question ! and my apologies if I have been a bit confusing in the video. I was just giving an example here, to emphasize the importance of considering not only the age result (100,000 years) but also the associated error (10,000 years). In my example, I just wanted people to understand that the true age was somewhere between 90,000 years and...

  • Hello Reid, one of the main interests of ESR compared with other dating methods is that it can be applied to a wide variety of Quaternary materials (e.g. carbonates, phosphates or silicates), such as: Cave deposits (speleothem, stalagmite, flowstone), Terrestrial mollusc shell or snail, fossil teeth, authigenic terrestrial apatite veins, Optically bleached...

  • Hello Charles, good question. The answer is quite easy: as often as needed. I mean, there have been an increasing number of fossil discoveries over the last couple of decades, and many of these fossils are older than 50,000 years. In other words, they lie beyond the limits of C-14. Consequently, ESR appears to be one of the very few numerical dating methods...

  • Careful here Enid, because U-series dating also provides a "+/-" error, like C-14. However, C-14 dates the death of the organism, while U-series dates the incorporation of uranium into the bone/tooth. This incorporation can happen a long time after the death of the organism. Therefore, we interpret the U-series result as a minimum age constraint for the fossil...

  • Agreed!

  • Good, means that we have reached our goal with this MOOC ;-)

  • Hello Maya, good question.
    In general, the radiation dose absorbed by modern enamel is negligible compared with that accumulated over thousands, tens of thousands, or hundreds of thousands of years. In other words, the ESR signal measured in fossil teeth is much bigger than in modern teeth.

    That said, please also note that ESR can be used to evaluate the...

  • Sometimes they do :-)!
    For example, last year we dated a human fossil bone found in Saudi Arabia to about 90 ka by laser ablation U-series. We also used other methods, like OSL (Optically Stimulated Luminescence) applied to the sediment surrounding the fossil, and ESR on a hippo fossil tooth found in association with the human fossil. They all returned very...

  • I agree with that Beth

  • Sometimes, no reasonable explanations can be proposed, and we just have to openly acknowledge it. However, please keep in mind that apparent U-series ages for teeth or bone are minimum ages for the fossils. We date the incorporation of uranium into the bone/tooth, not the formation of the tooth. Consequently, if we have a 100 ka old fossil (1 ka = 1,000...

  • @AmeliaHare Yes we do!
    Enamel is the most mineralized tissue of the body. It is much less porous than bones or dentine tissues. Consequently, uranium migration into enamel is significantly slower than in dentine or a bone. We usually observe that Uranium migrates into the enamel layer from the dentine side (the dentine-enamel junction), while there is...

  • @AlvaroA. In a way yes, this is what would happen, the apparent U-series age should be younger in the centre of the sample, although it is a bit more complicated in reality because there are many other parameters to take into account. For example, the preservation state of the tooth is also playing an active role in uranium diffusion, as uranium will...

  • 1 standard deviation = 68% probability; 2 standard deviation = 95% probability.
    Repeated measurement of a sample should enable to obtain a mean value and associated standard deviation, which may be used as a proxy to evaluate the variability/scatter of the results.

  • Theoretically, we can cover the whole Pleistocene (the last 2.6 Million years - Ma). The ESR signal is very stable in time, over several millions of years, but the limiting factor is usually around the uranium uptake modelling. The older the samples, and the higher probability to have uranium leaching from dental tissues, which simply precludes the application...

  • ... and please note that the growth rate is highly variable between caves, and even within a given cave.

  • Hello J Scott, the growth of the calcite deposits on top of the rock art can indeed be estimated by collecting various samples across the calcite crust.
    When the calcite crust is thick enough (a few millimeters), it is then technically possible to sample the base of the crust (the closest to the painting, which corresponds to the oldest domain) as well the...

  • @AmeliaHare @AlvaroA : In the case of ESR for example, the first dating application study was published in 1975 (see: https://www.nature.com/articles/255048a0), that is indeed much later than C-14.
    Please note that EPR (Electron Paramagnetic Resonance) and ESR (Electron Spin Resonance) both refer to the same technique. By convention, the acronym ESR is...

  • Hello Pat,
    the truth is that without the context and the sediment surrounding the tooth, we can simply not get an ESR age. This is quite different to C-14 or U-series, for which only the sample is required to calculate an age estimate. In ESR, the sample alone is not sufficient.

  • Hello Rob,
    an aliquot is a sub-sample of a given sample. Basically, a tooth enamel sample is divided into several sub-samples of the same weight.

  • The standard is typically a known-age material that has similar characteristics to the samples dated.
    For example, when we date fossil teeth by U-series, we use another fossil tooth as a standard (in our case, a Rhino fossil tooth from a Chinese site), because it has a similar density and composition to the samples that are being dated.
    A good standard...

  • Glad to hear that Pat. This is one of the reasons we did this MOOC, so that people can get to know other dating methods :-)

  • @RachelWelton This is correct.

  • These are important concepts indeed, John. We will cover some of them during week #1.

  • Yes, it should!

  • The truth is, without the additional indirect age constraints from the ESR and OSL methods, our interpretation would be very limited: we could only say that the fossil finger bone is at least 90,000 years old. This shows the interest to combine different dating methods (direct and indirect).

    You may find further info about this discovery...

  • Another example. While excavating a site in the middle of the desert in Saudi Arabia, some colleagues found a human finger bone. Once the finger was attributed to Homo sapiens, it rapidly became crucial to know its age.

    Direct dating options were limited. The fossil was most likely older than the C-14 time range. ESR does not work on bones, but on teeth...

  • Hello Setareh,
    sometimes there are not so many options available. Typically, the C-14 is the preferred method to date fossils whose age is somewhere between today and 50,000 years ago. Beyond this time range, C-14 doesn't work.

    But sometimes, the sample preservation is bad and fossils cannot be dated with C-14, even though their age is expected to be...

  • @MariaGrigg
    Hello Maria, it's quite possible indeed. Volunteers are always welcome. My suggestion would be to try to contact my colleagues archaeologists, anthropologists or paleontologists in ARCHE and ask them directly. They should be able to help you with that.

    Please have a look at the ARCHE website and go to the Research themes...

  • This is indeed an additional source of uncertainty.

  • Glad you liked the videos of the labs Gail ! That was quite a work to summarize the analytical process in only a few minutes...

  • @JohnLateano
    That's right John, although it might seem somewhat paradoxical in first instance.

  • Thank you Fiona

  • @MariaGrigg Hello Maria, Australian archaeology is definitely not my area of expertise (I am more focused on the earliest occupations in the Mediterranean area), so you are taking me out of my comfort zone :-).

    The article you mention refers to the recent discoveries made in the North of Australia at Madjedbebe rockshelter.
    The dating performed using C-14...

  • Sorry for that Maria, but I am still around if you have questions I can answer.

  • Thank you Maria

  • @AndrewHainsworth Hello Andrew,
    I am not an anthropologist so I am afraid I cannot say much about Homo naledi, sorry. However, I would recommend you to have a look at the article written by John Hawks for The Conversation: https://theconversation.com/more-secrets-of-human-ancestry-emerge-from-south-african-caves-77352.
    I am hoping it might bring some further...

  • Hello Patricia,
    actually when we calculate an age, we get this kind of number (let's say 350,216 years old). However, given the age uncertainty (let's say 10%), it doesn't make sense to report it with such a level of precision.
    So, we round up the number to 350,000 +/- 35,000 years.
    216 years would represent here less than 0.1% compared to 350,000 years ,...

  • @GillianWatt You are right Gillian. With the technological development in the last decades, the detection levels are now much lower (which means we can detect much smaller signals than before), and measurement precision is most likely much better now. That said, there will be still some uncertainty that cannot be reduced, such as the evaluation of the...

  • To give you an idea, the magnitude of the radiation dose measured in modern enamel may be of a few milliGrays (mGy), while the geological dose accumulated by fossil tooth enamel may be of several tens of Gy (factor of 10,000 higher) up to a few thousands Gy for 1 million-year fossil teeth.

  • That said, to directly answer your question:

    - U-series dating on a modern tooth would not bring anything new, simply because (hopefully) there is no measurable uranium inside. The uranium incorporation usually starts when the sample is buried in the sediment and in contact with groundwater.
    - ESR can actually be used not to date modern teeth but as a...

  • @SandyJ Hello Sandy, very interesting question here!
    We also perform a lot of methodological investigation with the aim to improve the reliability and accuracy of the dating method. To do so, we work on known-age samples (those are samples dated via another independent method), and we see if we are able to get a similar age. This is very useful to actually...

  • Hello Fiona,
    relative errors usually do not change much for a given method, but the absolute error will obviously increase with the age. For example, we know that ESR typically has a relative error of 10%. So, for an age of 50,000 years, the absolute error would be 5,000 years. In comparison, an age of 500,000 years would have an associated error of 50,000....

  • Hello Andrew,
    you are right about your first comment. In first instance we would use the approx. age of 300 ka because it is easier to report, but if we want to be correct we should actually report the age range.

    Not sure to fully understand your second comment. Can you please clarify?

  • + H. naledi, H. floresiensis...
    That's a good summary Gail.

  • Hello Marianne, to my knowledge, as of now, the Barium should not impact ESR or Luminescence dating results. We currently have no evidence to suggest that.
    However, that said, there is still the possibility that future investigations might prove the opposite ;-) !

  • Hello everyone,
    I would like to thank all of you for joining us on this course.
    I have truly enjoyed the interaction with you and it is great to have the possibility to share a bit of our knowledge in Dating and Human Evolution with people interested in the topic.
    Not always easy to clearly explain our work, and this has been a great exercise in that...

  • That's a good question John! Could be the entire African continent instead...
    A couple of months ago we reported the discovery of lithic tools dated to 2.4 million years in Algeria: https://theconversation.com/stone-tools-date-early-humans-in-north-africa-to-2-4-million-years-ago-107617.
    This date is actually very close to that of the oldest Oldowan lithic...

  • I don't think so. In first instance I don't see how the Barium could impact U-series or ESR results.

  • Hello Michael, this is actually one of the reasons for us to do this course: try to explain to the public what is behind a date/number. Now you have some keys to critically evaluate by yourself some of the dating results published over the recent years.

    Interestingly, the new US-ESR results obtained by Rainer of around 285,000 years overlap within error...

  • Hello Philip, seems like your knowledge in Human Evolution is really up to date. This study was only published a few days ago, so it is a bit too early to evaluate whether this is accepted or not by the scientific community.

    However, the work has been published in a prestigious journal and went through a peer-review process, which means that it has been...

  • Than you Sue

  • Thank you Ella

  • Hello Ann, you made a very important comment here. There is indeed no perfect dating method that can be used for all samples in any sites or contexts.

  • Hello Ella,
    you are right, the number of ESR dating laboratories and specialists in the world is very limited, especially when compared with C-14, the most popular and standardized method.
    About research funding, we usually get funded through competitive calls for proposals from the state agencies (e.g. Australian Research Council), Regions, private...

  • With these experimental points we are able to describe the specific behavior of the radiation-induced ESR signal for this enamel sample (all enamel samples do not behave exactly the same).
    So we can fit a function (with a known equation) through the experimental data points. This function describes the behavior of the ESR signal with the dose. The more...

  • We cannot irradiate the sample like in nature (because the intensity of the radioactivity is too low and we cannot reproduce the different types of radiations (alpha, beta, gamma, cosmic)). So, in the laboratory we typically use gamma radiations, and at a much higher intensity than in nature. And this is where we use either the single aliquot or multiple...

  • So, let's go back now to the De value. It is obtained by using ESR spectroscopy. In other words, ESR is a technique that is used to estimate the total dose absorbed by a sample. (It can work on modern enamel as well, in order to evaluate the radiation absorbed by people accidentally exposed to radioactivity). So, my point here is that ESR does not directly...

  • Hello Michael,
    we need to determine 2 parameters in order to be calculate an ESR age: the equivalent dose (which is the total dose absorbed by the enamel since the tooth has been buried) and the dose rate (which is an estimate of the radiation dose absorbed by the sample every year).

    The equivalent dose (De) is obtained with ESR, by a building dose...

  • Hello Fiona,
    apologies for the confusion:
    - ESR is mostly applied to fossil teeth rather than bones. Some attempts have been performed on bones in the early 80s, but didn't work quite well, so this material is not considered as being suitable for the method.
    - ESR is actually less precise than other radiometric methods, mostly because of the large number...

  • My turn to learn something today :-). Thanks to everybody