Age-depth models/sedimentation rates - new features in OxCal

[christoph...@rlaha.ox.ac.uk]

There have been a lot of requests recently for being able to output the age-depth models at specific depths.

I have now made this easier within the OxCal user interface.  Here is how it works:

1. First click on the table () symbol next to the P_Sequence you want the age depth model for.

2a. You can still paste in the depths you want as a single column in the left hand yellow window with one of your columns being the depth, headed by z.  Then press the [Get dates <<] button.  This is useful if you want to add ages to a table of proxy values.

2b. Alternatively, under the direct outputs from the age-depth model in the right hand yellow window you will see three boxes which let you set the start, end and increments in depths that you want.  You can change these to whatever values you want end then if you press the [<< Depths] button next to this, this will give the age depth model in a grid at this resolution.  By default this will give outputs as requested in the P_Sequence command covering the whole range of the model.

3. Above the left hand pane you can then either save the output to a csv file, copy and paste it from the window, or plot any proxy data against age using the [Plot] button.

The interpolated age depth model will have the following columns which are always given:

intcal13_t - this is the date in fractional year format as used internally by OxCal

intcal13_dt - this is the one sigma error in the date

z - this is the depth

sed - this is the sedimentation rate as calculated during the MCMC run.

In addition whatever other reported values (mu, sigma, median, 1, 2, 3 sigma ranges etc) are also given in whatever age format you have chosen.

Finally if you are wishing to use the ensemble of age-depth realisations which are generated with each run (assuming you have requested interpolation) you can also select the number to display under [Format > Alter] and at the bottom of the page "Ensembles".  This will give a column for each realisation, headed ens_0, ens_1, etc.  These will also have whatever date format you have chosen.

I hope this will help those of you who are trying to get the most out of your age-depth models.

Best wishes

Christopher

[Natalia Piotrowska]

Dear Christopher,
this new feature made my Monday brighter! Hopefully things will work smoothly with my new models :-)
Thank you,
Natalia

[Erik Marsh]

Love these features! Thanks so much. Now I can fairly easily recalibrate Bayesian depth models for core data from old climate papers.

A quick clarification on sedimentation rate. My P_Sequence uses cm, so the sedimentation rate output is showing me cm/year, correct?