How to interpret a strongly negative -517 ΔR?

[Erik Marsh]

Hello all – 

I am using legacy dates from northern Chile to calculate ΔR, to update Ortlieb et al.'s 2011 estimate for the site La Chimba 13. This code runs, following the stratigraphy (though details are sparse), but I'm not sure how to interpret the strongly negative ΔR of -517 years (during the site's occupation 10–12 cal BP). This ΔR seems to imply that shells and otoliths had less carbon that coeval terrestrial samples – which seems backward to me. The model assumes the ΔR was similar over the site's occupation, but of course that might not be true.

There are three sets of pairs from roughly the same context (using Combine, two fail squared tests, so I've used Phase). In two (150-160 cm & 180-190 cm), the shell 14C age is younger than the carbon; in one it's older (hearth A), leaving me to also suspect errors in the publications. How is the model resolving a single ΔR for these opposed trends? Must be the large error ranges.

These are conventional dates run at Pennsylvania, Toronto, and Queensland run from the late 70 to 90s. The two unexpectedly young shell dates are the only two from Queensland.

I'm trying to make the most of imperfect legacy data, but is this result of -517 useable at all? If so, what could explain such a strongly negative ΔR?

Thanks for your input,

Erik

 Plot()
 {
  Curve("SHCal20","shcal20.14c");
  Curve("Marine20","marine20.14c");
  Delta_R("ΔR", U(-1200,1200));
  Sequence("Dated occupation phases")
  {
   Boundary("Start");
   KDE_Plot("Overall site occupation")
   {
    Span("Span");
    Sequence("Hearths N02/W14")
    {
     Delta_R("=ΔR");
     R_Date("TO-6325 hearth C shell", 10280, 90);
     R_Date("TO-6319 hearth B shell", 10030, 90);
     Phase("hearth A")
     {
      Delta_R("=ΔR");
      R_Date("TO-6320 hearth A shell", 9800, 80);
      Curve("=SHCal20");
      R_Date("TO-6527 hearth A charcoal", 9260, 90);
     };
    };
    Sequence("N23/W04")
    {
     Curve("=SHCal20");
     R_Date("Beta-94994 190-210cm charcoal", 10120, 70);
     Phase("180-190 cm cm")
     {
     Curve("=SHCal20");
      R_Date("TO-6528 180-190cm charcoal", 9910, 90);
      Delta_R("=ΔR");
      R_Date("UQ-2157 180-190cm shell", 9163, 140);
     };
     Phase("N23/W04, 150-160 cm")
     {
      Curve("=SHCal20");
      R_Date("TO-6526 150-160cm charcoal", 9460, 90);
      Delta_R("=ΔR");
      R_Date("UQ-2156 150-160cm shell", 9088, 120);
     };
    };
    Sequence("N54/W05")
    {
     Curve("=SHCal20");
     R_Date("P-2702a 30-40cm charcoal", 9680, 160);
     R_Date("P-2702b 20-30cm charcoal", 9400, 160);
     Delta_R("=ΔR");
     R_Date("TO-5631 0-10cm otoliths", 9170, 80);
    };
   };
   Boundary("End");
  };
 };

[MILLARD, ANDREW R.]

Hi Erik,

 

There are multiple possible reasons:

1. your model is poorly constrained as there are few terrestrial samples bracketing the marine samples, and this allows ΔR to follow the majority of the data as two samples in a phase can be of very different ages
2. Perhaps the species identifications or assumed ecologies are wrong  and these are in fact freshwater shells with an atmospheric 14C content.
3. Given that these are legacy dates, the shells were probably not checked for recrystallisation and you may be seeing diagenetic carbonate in some of them which can be more recent than the charcoal from the same contexts.

Running with everything on SHCal gives an acceptable Amodel=69.6 which suggests numbers 2 and 3 can’t be ruled out.

 

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 Erik Marsh
Sent: 20 August 2026 14:20
To: OxCal <ox...@googlegroups.com>
Subject: How to interpret a strongly negative -517 ΔR?

 

[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/1ec9ef05-7c35-4157-a5eb-69af7dc8d902n%40googlegroups.com.

[Erik Marsh]

Andrew – 

I think that's exactly it – recrystallization and diagenesis. I haven't worked with shell dates before, but this makes complete sense in this case, especially given the 10 millennia of burial time. Interestingly, this paper shows these processes can be quite quick and came up with a time function.

As for option 2, it's true that we don't know the species, but this is quite close to the coast. There is one otolith sample, which has the same issue. While there are no sample-specific details, a separate paper identified 13 fish species from hundreds of otoliths, which were 99% from marine-exclusive environments. That sample is probably more likely to be correct, since it wouldn't recrystallize and diagenesis would be less of an issue. Too bad there's not stratigraphic constraint to check it separately from the shell dates. Oh well. As you mention in option 1, the model is poorly constrained overall.

Anyway – I think that I did what I could with these legacy dates. We'll have to exclude the shell dates, now with a much better justification – including your point about the model converging with SHCal20. At least the site still has useful charcoal dates.

Thanks! – Erik