Phases

[Lee McAuliffe]

Hi,

I was wondering if I could recieve some feedback on my use of OxCAL
4.1 using radiocarbon samples to date paleoearthquakes from a trench
wall. Initially I followed the example given in the Lienkaemper and
Ramsey, 2009 Primer paper, however after reading the manual available
online I am unsure if I have modelled my dates correctly.

I currently have 20 radiocarbon dates from a 3.5m deep trench which I
want to use to carefully date three paleo events. This is my current
configuaration:

Sequence(
Boundary
Phase (R_Date1, R_Date2, R_Date3, R_Date4)
R_Date5
R_Date6
R_Date7
R_Date8
Date (Event1)
R_Date9
R_Date10
Date (Event2)
Phase (R_Date11, R_Date12, R_Date12)
R_Date
Date (Event3)
Phase (R_Date14, R_Date15, R_Date16)
Phase (R_Date17, R_Date18)
R_Date19
R_Date20
Boundary


I have placed samples taken from identical units into phases where the
exact stratigraphic order within those units is unclear. I have used
the 'Date' function to date my three events, and I have used the
'Boundary' function to mark the base and the top of my trench.

To a first degree I seem to get reasonable ages for my events, however
after reading into the use of Phases I am unsure if I should be using
a contiguous phase, sequential phase or overlapping phase. I also
came across a posting where it was mentioned that the events should be
input as "boundary's" and not "Date's". Please could you give me some
feedback on my ordering model.

Thank you

Lee

[Christopher Ramsey]

Lee

Thanks for this question.

I think that what you have done makes sense but there are some subtleties to the model are worth considering in each case.

The model you have below is fine if the radiocarbon dates are essentially randomly drawn from the profile and there is no reason to suppose that they are more concentrated in one time period than another - this is an assumption of random deposition of 'radiocarbon-suitable' samples. In this model, all things being equal (which they won't be with radiocarbon dates in there), the model would be such that the time between Event1 and Event2 is likely to be half as long as the time between Event2 and Event3, just because the number of dates between Event2 and Event3 is double. The model will be pretty flexible, so it will allow for other possibilities too - but that will be the most likely outcome.

Boundaries are useful when there is a definite change in the depositional process. So if for example, between Event1 and Event2 there was a sterile layer with no organic material at all it would certainly be worth marking these events with Boundaries. Under a model with Boundaries, the number of dates between each Boundary does not affect the expected age in the model at all. In general a model with Boundaries will give wider ranges - covering a greater range of possibilities.

One approach that you might take is to try both types of model and see how robust the conclusions are to the different assumptions.

Christopher

> --
> You received this message because you are subscribed to the Google Groups "OxCal" group.
> To post to this group, send an email to ox...@googlegroups.com.
> To unsubscribe from this group, send email to oxcal+un...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/oxcal?hl=en-GB.
>

[Lee McAuliffe]

Hi Christopher Ramsey,

I am revisiting this issue that I had e-mailed you about last year and have a few questions that I was hoping you could help me with.

I am a little confused with when it is necessary to use a 'boundary' and how to use them.  Since I know that deposition at my site was not continuous, do I just want to place the boundaries where I believe there may have been a hiatus in deposition?  Also, in my sequence of commands I have placed a boundary at the beginning and end of my sequence and I am not sure if that is the correct thing to do. In your 2009 primer paper you have placed a 'boundary' as your base layer and I just followed that chain of commands.  Am I correct to use a 'boundary' to mark the top of my trench? 

Secondly, because I am dating charcoal samples from a trench, should I be using the depth function as well?  What are your thoughts?

Thank you for all your help.

Lee McAuliffe

[Christopher Ramsey]

The boundaries define the groups of dates within a model. Thus you need at least two (for one group of dates) in any model. If you have a hiatus you would add two boundaries:

Boundary
.... elements of phase one
Boundary
... hiatus
Boundary
.... elements of phase two
Boundary

and so on. The [Tools > Models > Phases] tool sets up commonly used models for you and helps to give an idea of how to do this.

I think it only makes sense to use a depth model if the deposition is fairly well behaved (like a lake sediment) otherwise a straight sequence is probably more appropriate.

Best wishes

Christopher