Constraining model for recent material

[Chloe Stringer]

Hi all,

I have two ages on freshwater shells collected from a natural death assemblage (i.e. not archaeological, just a collection of dead shells we are trying to date). The material was collected in 2014CE and we expect the deposit dates prior to the 1980s as that is when this species became locally extinct. I created an OxCal (v4.4.4) model (pasted below). The End Boundary stretches past 2014 (which is not possible). Is there a way to put 2014 into the model as the maximum date? I have tried adding it as a Before date, but it doesn't seem to have any effect.

Any help is much appreciated!

Best,

Chloe

Plot()

 {

  Curve("SHCal20","shcal20.14c");

  Outlier_Model("General",T(5),U(0,4),"t");

  Sequence()

  {

   Boundary("Start 1");

   Phase("1")

   {

    R_Date("OZAZ54",220,25)

    {

     Outlier(0.5);

    };

    R_Date("Wk-53750",296,13)

    {

     Outlier(0.5);

    };

    Before( )

    {

     C_Date("Collection Year",2014,0);

    };

   };

   Boundary("End 1");

   Span("seq");

  };

 };

[MILLARD, ANDREW R.]

Hello Chloe,

 

The Before command indicates that things stratigraphically below this must precede 2014 but this date has no known temporal relationship to things stratigraphically later, including Boundary("End 1") in your model.

 

As your collection year is after the end of the period of deposition, I think it comes after Boundary("End 1") in your model. I would make the last part of the sequence this:

 

    R_Date("Wk-53750",296,13)

    {

     Outlier(0.5);

    };

   };

   Boundary("End 1");

   C_Date("Collection Year",2014,0);

   Span("seq");

 

 

Outlier(0.5) is strong prior, suggesting that before the shells were dated you thought there was a 50% chance they were out of place. Is that what you meant? 0.05 is a more typical prior outlier probability.

 

 

 

Best wishes 

Andrew 

-- 

Prof. Andrew Millard 

Department of Archaeology, Durham University, UK 

Email: A.R.M...@durham.ac.uk  

Personal page | Dunbar 1650 MOOC 

StratoBayes: De-risking subsurface decision making

 

From: ox...@googlegroups.com <ox...@googlegroups.com> On Behalf Of Chloe Stringer
Sent: 09 July 2026 19:29
To: OxCal <ox...@googlegroups.com>
Subject: Constraining model for recent material

 

You don't often get email from chloe.ann...@gmail.com. Learn why this is important

[EXTERNAL EMAIL]

--
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 visit https://groups.google.com/d/msgid/oxcal/dd6bf44d-a3bb-4335-8c93-490c219b8769n%40googlegroups.com.

[Chloe A Stringer]

Dear Andrew,

Thank you for your help. Moving the C_Date to below the End Boundary appears to have constrained model as I hoped!

Also, thank you for picking up on the fact that it was Outlier(0.5) than my intended Outlier (0.05). I was obviously too close to the data.

Best,

Chloe

You received this message because you are subscribed to a topic in the Google Groups "OxCal" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/oxcal/AVBdjix_1LQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to oxcal+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/oxcal/LO4P265MB422692E9A3B28486A419F0CB81FD2%40LO4P265MB4226.GBRP265.PROD.OUTLOOK.COM.

[John Meadows]

Dear Chloe

did I miss something? your samples are freshwater mollusc shells, which would usually incorporate a reservoir effect. I don't see anything in your model that covers this. I am not sure how I would proceed here, because if I knew that the material was post-bomb, I would be calibrating with the appropriate bomb curve, not SHCal20. However, if the shells date to the 1960s or 70s, the reservoir effect would have to be enormous to produce the measured 14C ages (think hot springs), so I guess the assemblage is older. If the material is pre-1950, the local FRE must be quite minor or even negligible (<200y), but that does not help to narrow down the date of death. 

Here's a bit of code to play around with, to see what sort of 14C ages you might get from known-age shells with known or negligible reservoir effects (FREs).  

 Plot()
 {
  Curve("SHCal20","shcal20.14c");

  R_Simulate("AD1800 no FRE",1800,1);
  Delta_R("FRE",1600,60);
  R_Simulate("AD1800 with FRE",1800,1);
  Curve("Bomb21SH3","bomb21sh3.14c");
  R_Simulate("AD1980 no FRE",1980,1);
  Delta_R("bomb FRE",1600,60);
  R_Simulate("AD1980 with FRE",1980,1);
 };

good luck

John

To view this discussion visit https://groups.google.com/d/msgid/oxcal/CAEbDL%3DrXx7-P6sfy%2BbrkBY4dBjEuJ-EG%3DBaHgit%3Dji8ysx49sg%40mail.gmail.com.