how to model an unconformity in a lake core?

[Erik Marsh]

Hi all – 

What's the best way to model an unconformity of no deposition in a lake core?

I'm running dates from this article:

Arnold, T. Elliott, Aubrey L. Hillman, Mark B. Abbott, Josef P. Werne, Steven J. McGrath, and Elizabeth N. Arkush 2021Drought and the collapse of the Tiwanaku Civilization: New evidence from Lake Orurillo, Peru. Quaternary Science Reviews 251:106693. DOI:10.1016/j.quascirev.2020.106693.

There is an unconformity at a depth of 116.5 cm., indicating a lapse with no deposition of sediment. But we have almost no information on how long this lapse was – dates near the unconformity are clear outliers. In the paper, the authors use interpolated medians from just above and below the unconformity to estimate its lapse. Is this a sound approach? Is a boundary a better approach, perhaps? In theory, this would estimate the mid-point of the lapse without sediment deposition, correct? It seems to me there is too little information to effectively estimate a start and end date for the unconformity.

My first attempt at the code is below.

Thanks for your feedback!

Erik

 Plot()

 {

  Curve("IntCal20","IntCal20.14c");

  Curve("SHCal20","SHCal20.14c");

  Mix_Curve("Mixed","IntCal20","SHCal20",U(0,100));

  P_Sequence("",1,0.4,U(-2,2))

  {

   Boundary();

   R_Date("UCI-164754",4285,40)

   {

    z=287;

   };

   R_Date("UCI-224579",3370,220)

   {

    z=261.5;

   };

   R_Date("UCI-209788",3670,60)

   {

    z=257.5;

   };

   R_Date("UCI-224585",3020,150)

   {

    z=252.5;

   };

   R_Date("UCI-164753",2620,30)

   {

    z=228.5;

   };

   R_Date("UCI-172624",2300,20)

   {

    z=195;

   };

   R_Date("UCI-224578",2250,100)

   {

    z=183;

   };

   R_Date("UCI-224584",1780,180)

   {

    z=157.5;

   };

   R_Date("UCI-164752",1655,30)

   {

    z=142.5;

   };

   //All four outliers fall in this section of the core

   //   R_Date("UCI-209787",2275,20){z=142.5; };

   //   R_Date("UCI-172623",2850,15){z=124.5;};

   //   R_Date("UCI-224583",1530,110) {z=121; };

   //   R_Date("UCI-224582",1640,240) {z=118; };

   Boundary("Unconformity")

   {

    z=116.5;

   };

   R_Date("UCI-224581",760,180)

   {

    z=103;

   };

   R_Date("UCI-224580",970,160)

   {

    z=100;

   };

   R_Date("UCI-224577",470,170)

   {

    z=92.5;

   };

   R_Date("UCI-224576",420,270)

   {

    z=78.5;

   };

   R_Date("UCI-224575",460,170)

   {

    z=62;

   };

   R_Date("UCI-172622",420,30)

   {

    z=47.5;

   };

   Curve("Bomb13SH12","bomb13sh12.14c");

   R_Date("UCI-172621",-30,15)

   {

    z=25;

   };

   Boundary();

  };

 };

[MILLARD, ANDREW R.]

Hi Erik,

 

A single boundary still implies continuous deposition. I’d use two P_sequences each with their own boundaries for the two sections of the core, and place them within a Sequence so that they are constrained to follow one another. Modelled like this, the two P_sequences have independent deposition regimes, so do not learn from one another about the k value. This is a disadvantage if you think the deposition process was the same before and after the unconformity. You might also consider two boundaries in your P_sequence separated by a Gap with a suitable prior.

 

Best wishes

Andrew

--

Dr. Andrew Millard

Associate Professor of Archaeology, and

Designated Individual under the Human Tissue Act,

Durham University, UK

Email: A.R.M...@durham.ac.uk 

Personal page: https://www.dur.ac.uk/directory/profile/?id=160

Scottish Soldiers Project: https://www.dur.ac.uk/scottishsoldiers

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

 

 

From: ox...@googlegroups.com <ox...@googlegroups.com> On Behalf Of Erik Marsh
Sent: 02 February 2021 01:56
To: OxCal <ox...@googlegroups.com>
Subject: how to model an unconformity in a lake core?

 

[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 on the web visit https://groups.google.com/d/msgid/oxcal/02db36be-4f74-405a-84f2-8cbc7bc70d05n%40googlegroups.com.

[sofia...@gmail.com]

Dear Andrew,

I am a beginner at age-depth models and I have a similar problem with Erik, 

could you please give me a small code example on how I put two P_sequences with their own boundaries within a Sequence? I have to model unconformities of no deposition of a fluvial environment.

Thank you in advance,

Sofia

[Erik Marsh]

Thanks for that suggestion, Andrew, right on the money. Sofia, here's the code I'm using. Hope this helps. –Erik

 Plot()

 {

  Curve("IntCal20","IntCal20.14c");

  Curve("SHCal20","SHCal20.14c");

  Mix_Curve("Mixed","IntCal20","SHCal20",U(0,100));

  Sequence()

  {

   P_Sequence("Below unconformity",1,0.7,U(-2,2))

   {

    Boundary("Bottom of core");

    R_Date("UCI-164754",4285,40) { z=287; };

    R_Date("UCI-224579",3370,220) { z=261.5; };

    R_Date("UCI-209788",3670,60) { z=257.5; };

    R_Date("UCI-224585",3020,150) { z=252.5; };

    R_Date("UCI-164753",2620,30) { z=228.5; };

    R_Date("UCI-172624",2300,20) { z=195; };

    R_Date("UCI-224578",2250,100) { z=183; };

    R_Date("UCI-224584",1780,180) { z=157.5; };

    R_Date("UCI-164752",1655,30) { z=142.5; };

    Boundary("117 cm") { z=117; };

   };

   //   All four outliers fall in this section of the core

   //   R_Date("UCI-209787",2275,20) {z=142.5; };

   //   R_Date("UCI-172623",2850,15) {z=124.5;};

   //   R_Date("UCI-224583",1530,110) {z=121; };

   //   R_Date("UCI-224582",1640,240) {z=118; };

   P_Sequence("Above unconformity",1,1,U(-2,2))

   {

    Boundary("116 cm") {z=116; };

    R_Date("UCI-224581",760,180) { z=103; };

    R_Date("UCI-224580",970,160) { z=100; };

    R_Date("UCI-224577",470,170) { z=92.5; };

    R_Date("UCI-224576",420,270) { z=78.5; };

    R_Date("UCI-224575",460,170) { z=62; };

    R_Date("UCI-172622",420,30) { z=47.5; };

    Curve("Bomb13SH12","bomb13sh12.14c");

    R_Date("UCI-172621",-30,15) { z=25; };

    Boundary("Top of core");

   };

  };

 };

[sofia...@gmail.com]

Thank you so much!