Combining TL ages

[Chris Harpel]

Hi all,

I have two published TL ages that I am trying to incorporate into a broader model.  I seldom deal with TL dating, so have run into a couple of questions about how to do so.

The two TL ages are from a tephra layer and ostensibly should both date the unit's emplacement/burial.  So from a theoretical perspective, I assume that it should be acceptable to combine them to try to better constrain the unit's eruption age.  Am I off base in this?

In reading through the OxCal info online about combining ages, there is a statement that TL ages should be combined before the dose rate error is applied.  However, I have not yet been able to find an example of how to do so.  Would anyone have an example of how to do this that they would be willing to share?

Cheers,
Chris

[Erik Marsh]

Hi Chris – 

The two TL ages are from a tephra layer and ostensibly should both date the unit's emplacement/burial.  So from a theoretical perspective, I assume that it should be acceptable to combine them to try to better constrain the unit's eruption age.  Am I off base in this?

Yes, that's right.

 

In reading through the OxCal info online about combining ages, there is a statement that TL ages should be combined before the dose rate error is applied. 

Can you point us to what you read? I don't remember seeing anything like that for TL dates. I normally apply all the corrections first, then model the dates.

Are the dates from a region that produces reliable TL ages?

Erik

[Christopher Ramsey]

There is some information on the type of information you might include at:

https://c14.arch.ox.ac.uk/oxcalhelp/hlp_analysis_inform.html#paramdate

But what goes into these calculations requires expertise in the TL/OSL method as applied on the site. It would be best to work on this with the lab.

As Erik says you can just put in the final estimated date. The reason for parameterising it is primarily to deal with components of the uncertainty which are shared between measurements.

Best wishes

Christopher

> --
> You received this message because you are subscribed to the Google Groups "OxCal" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to oxcal+un...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/oxcal/98e8fb32-db95-467c-83d8-17fc9f36b9dfn%40googlegroups.com.

[Chris Harpel]

Christopher and Erik,

Many thanks for your responses to my queries.

Erik, the specific section that I was referring to is under the "Other Dates" section at this link: https://c14.arch.ox.ac.uk/oxcal3/arch_cmb.htm#calendar

It sounds like it makes more sense to just apply the ages as reported, which is my preference, and not get into the details of the dose rate or anything along those lines.  The way I have it in my preliminary model is simply:

Combine("TL")
     {
      Age("TL-S1", N(28700,2900))
      {
       Year=2017;
      };
      Age("TL-S2", N(32300,3100))
      {
       Year=2017;
      };

Rather than use C_Combine and C_Date, I used Combine and Age and assigned a normal distribution to the error values.  Is this reasonable or would it be better to stick to C_Combine and C_Date?  The year that the samples were collected and dated is 2017.

Christopher, thanks for pointing me to the information on parameterisation.  It might be something that actually comes in handy in the future, so I will probably start playing around with it just to become more familiar with it.

Erik, your question regarding whether I am working in a region that produces reliable TL ages is a good one, but complicated.  The simple answer is that I do not know.  The long answer is that I have read through some of the posts here that you have participated in and it sounds like TL dates in South America are complex.  I am working in southern Peru, around Arequipa, so you may have some helpful input on whether partial bleaching is a problem where I am working.  Your question, however, also drove me to go back to some of the published ages from the area.  The scant TL ages do seem to be coming in younger than 14C ages from the same sites.  There are, however, currently no 14C ages that can be directly compared with the two specific TL dates that I am working with here.  Another major caveat is that we are also finding that certain samples yield 14C ages that are significantly older than is realistic for their stratigraphic position.  As a result, there is some potential that the TL ages are correct and a few of the comparative 14C ages are skewed old.  To try to tease out which specific ages may be problematic, however, would require a thorough review of the published ages and our new ages all constrained within a tephrostratigraphic framework.  Such an undertaking is actually the broader project I mentioned, but there is still a lot to do and we are working on it in pieces with these two TL ages in question being part of the first chunk.

Thanks again for both of your responses.

Cheers,
Chris

[Susanne Lindauer]

Hi Chris,

 

just a short note from a C14 person also working on OSL and TL.  You´re right, just model the ages as they were reported. A discussion on dose rate etc. is not part of an Oxcal model.

 

Regarding the reliability of TL ages, usually the sampling is the problem. If it is not done by a luminescence specialist (which is often the case) these people usually claim the lab is not reliable or the dates are wrong. This is an endless discussion. TL dates and C14 dates do NOT have to return the same results. You answer different questions. With C14 you never date the accumulation of sediments, but OSL does. TL dates when something was heated, ideally over 500 °C. TL and OSL share some electron traps but not all therefore even when using TL and OSL on the same material might give different results. But C14 and OSL/TL  look at different materials and aspects.

Regarding the partial bleach aspect, you might not detect it with TL, but with OSL. You could use different minerals for this as quartz bleaches faster than feldspar, but with volcanic tephra this becomes increasingly difficult as this directly influences the mineral genesis. There are some papers from colleagues who worked at the German Eifel tephra as far as I remember. If you need some more hints feel free to get in touch J

 

Best wishes

Susanne

 

Dr. Susanne Lindauer

 

Research Associate

 

 

Curt-Engelhorn-Zentrum Archäometrie gGmbH Büro C4,8 (3.OG) | 68159 Mannheim T:  +49 621 293 3826 www.ceza.de

 

        

 

 

Diese E-Mail kann Betriebs- oder Geschäftsgeheimnisse oder sonstige vertrauliche Informationen enthalten. Sollten Sie diese E-Mail irrtümlich erhalten haben, ist Ihnen eine Kenntnisnahme des Inhalts, eine Vervielfältigung oder Weitergabe der E- Mail ausdrücklich untersagt. Bitte benachrichtigen Sie uns und vernichten Sie die empfangene E-Mail. Vielen Dank.

 

This e-mail may contain trade secrets or privileged, undisclosed, or otherwise confidential information. If you have received this e-mail in error, you are hereby notified that any review, copying, or distribution of it is strictly prohibited. Please inform us immediately and destroy the original transmittal. Thank you for your cooperation.

To view this discussion on the web visit https://groups.google.com/d/msgid/oxcal/b14183f4-bbe9-4c86-8acb-5daf5b33f749n%40googlegroups.com.

[Chris Harpel]

Hi Susanne,

Thanks for the added input on this.  I appreciate it.  I agree that it is absolutely necessary to consider what the dates are related to rather than just blindly using them.  I will definitely check into the Laacher See papers and may reach out at some point for more details about the TL dating.

Cheers,

Chris

[Susanne Lindauer]

Hi Chris,

 

great J The authors that come to my mind regarding Laacher See tephra are Ludwig Zöller, Hendrik Blanchard and Christoph Schmidt has also done quite some research on this. I´m looking forward to reading your paper unless we get in touch earlier J

 

Cheers

To view this discussion on the web visit https://groups.google.com/d/msgid/oxcal/166f6c60-a182-444e-b5bb-af043ceb4f58n%40googlegroups.com.

[Erik Marsh]

Hi Chris – 

You code looks fine to me. There are different ways to code this in OxCal, but I don't think it will affect the results.

You might have seen that we have a recent paper comparing 14C and TL dates in the Andes (Marsh et al. 2021), but it is on archaeological ceramics and you are looking at much older sediment samples. Susanne is right – without a targeted study comparing dating methods, the methods debates go in circles. There is often mismatch between dates and it is usually quite difficult to explain why; there are just too many factors. You're right to keep a skeptical eye on both 14C and TL dates, and keep in mind the caveats that Susanne mentioned.

Yes, luminescence dating in the Andes is not straightforward – our paper does have a short section that mentions efforts to improve the method with geological samples. These papers also note unexpected dates (some younger, some older) and different lab methods – there are multiple proposals for different regions and sediment types. I would start with those references, to see if there are similar situations to compare to (for example, northern Chile). Since you're looking at a tephra, you can always hope there are other dates from the same eruption (but that region is so active it may be hard to identify which tephras are from the same eruption without geochemistry on lots of regional samples). Just thinking out loud here, but the fact that it's a tephra make me wonder if you could do single-aliquot dates on the obsidian bits (also not straightforward, but it would be great! see this paper).

Perhaps slight mismatches do not affect your research questions – if you're dealing with geological time then a few hundred (or thousand) years either way doesn't matter, looking at the full 95% error range. Your error ranges are large but are probably honest – and they may be more than precise enough to build a geological sequence, especially if your 14C dates are coming in too old (there are another set of potential problems there!).

Good luck,

Erik