Questions about OxCal age-depth modeling and modern sample calibration

[xiannong Song]

Hello everyone, How are you?

I am a beginner using OxCal to build an age-depth model for a peat core. I have some questions about my model and would appreciate any guidance:

1. For my modern sample at 30 cm depth (labelled as core_30cm), OxCal gives a warning: "Date may extend out of range --32+/-19BP." Is this a normal warning for samples post-1950? How should I interpret and handle this warning?  

2. I compared the calibrated age of this modern sample with results from rbacon, and noticed the OxCal age is about 30 years older than the rbacon age (around -40 cal yr BP for core_30cm). Could this discrepancy be due to errors in my OxCal code or settings for modern samples?

3. The age-depth model plot I obtained looks not very smooth and has some abrupt segments. Could this be caused by the limited number of dated points? Are there any recommended ways to improve the smoothness?

4.  Is my K-value set correctly?  Where can I check what the appropriate K-value found by the model is? I apologize that my understanding of the K-value is still limited, especially regarding the “[interpolation]” and “[log10(k/k0) expression]” options. 

PS: The tephra layers were identified by other methods, and I want to estimate their approximate ages through the age-depth model.  

Attached are screenshots of my code and model output.

Thank you very much for your help and advice.

code:

Plot()
 {
  P_Sequence("core_age_depth_model",100,1,U(-2,2))
  {
   Curve("SHCal20","shcal20.14c");
   Boundary("Bottom")
   {
    z=0.89;
   };
   R_Date("core_88cm", 2435, 30)
   {
    z=0.88;
   };
   Date("Tephra_74cm")
   {
    z=0.74;
   };
   Date("Tephra_58cm")
   {
    z=0.58;
   };
   R_Date("core_55cm", 1730, 30)
   {
    z=0.55;
   };
   Date("Tephra_44cm")
   {
    z=0.44;
   };
   Curve("Bomb21SH12","bomb21sh12.14c");
   R_F14C("core_30cm", 1.0039466, 0.0023783)
   {
    z=0.30;
   };
   Boundary("Top",AD(2021))
   {
    z=0;
   };
  };
  Curve("SHCal20");
 
  R_Date("Commerson eruption", 1890, 55);
  R_Date("U3 tephra", 2140, 35);
  R_Date("Bert1 eruption", 2285, 25);
 };

[MILLARD, ANDREW R.]

Dear Xiannong,

 

I’ve attempted to answer your questions below

1. For my modern sample at 30 cm depth (labelled as core_30cm), OxCal gives a warning: "Date may extend out of range --32+/-19BP." Is this a normal warning for samples post-1950? How should I interpret and handle this warning?  
   

1. 
   This is normal for dates close to 0 BP. If you examine the Bomb21SH12 curve you will see that the F14C value does not extend out of range, and you can ignore the warning.
   
   

1. I compared the calibrated age of this modern sample with results from rbacon, and noticed the OxCal age is about 30 years older than the rbacon age (around -40 cal yr BP for core_30cm). Could this discrepancy be due to errors in my OxCal code or settings for modern samples?
   

1. 
   rbacon does not use the bomb curve, so the probability distributions of dates will be truncated at the end of SHCal20 causing the means to be earlier. Rbacon also uses a different approach to sedimentation rates which can lead to differences.
   
   

1. The age-depth model plot I obtained looks not very smooth and has some abrupt segments. Could this be caused by the limited number of dated points? Are there any recommended ways to improve the smoothness?
   

1. 
   This is because you have set the [interpolation] parameter frequency at 1 per metre of depth and your core is less than a metre long, causing OxCal to only interpolate depths at the positions with dates. Setting this to 10 or 20 would give smoother output.
   
   

1.  Is my K-value set correctly?  Where can I check what the appropriate K-value found by the model is? I apologize that my understanding of the K-value is still limited, especially regarding the “[interpolation]” and “[log10(k/k0) expression]” options. 
   

1. 
   The k-value is in the first line of the output table, and clicking on the word P_Sequence will show the probability distribution. With a posterior range of -1.12 to -0.87 it seems that your log10(k/k0) prior is not constraining the result.
   
   

Best wishes 

Andrew 

-- 

Prof. Andrew Millard 

Department of Archaeology,

Durham University, UK 

Email: A.R.M...@durham.ac.uk  

Personal page: https://www.durham.ac.uk/staff/a-r-millard/ 

Dunbar 1650 MOOC: https://www.futurelearn.com/courses/battle-of-dunbar-1650 

 

 

From: ox...@googlegroups.com <ox...@googlegroups.com> On Behalf Of xiannong Song
Sent: 11 July 2025 10:40
To: OxCal <ox...@googlegroups.com>
Subject: Questions about OxCal age-depth modeling and modern sample calibration

 

You don't often get email from s9281...@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/1a622c5b-204a-4495-ae95-5b295b1d0ab1n%40googlegroups.com.