Posterior predicting into the future?

[Joanna Tindall]

Hi, 

I have some 210Pb ages and SCP ages from a core taken in 2019. I have imported the ages as C_Dates, with uncertainty and depth to produce an age depth model using a p sequence. However, when I include the posterior function to extrapolate to my bottom boundary, the model moves the known ages to be much younger (e.g. a 1958 CE date becomes 2022 - which is well, impossible) and then predicts into the future (ie. I know the top of my core is 2019, but the model suggests 2043! 

Any ideas?

Thank you, 

Joanna

[Erik Marsh]

Hi Joanna,

Yes, OxCal is so great it can even predict the future! This can happen with models – if you post your code, the forum will be able to help you with it.

Erik

[Joanna Tindall]

Hi Erik, 

No longer going into the future (although I don't know what I have changed to stop that!) but still producing lots of errors and not extending to the bottom boundary. Code below:

Options()

 {

  BCAD=TRUE;

 };

 Plot()

 {

  P_Sequence("alre13";)

  {

   Boundary("TOP", 2013)

   {

    C_Date("0", 2013, 0);

    z=0;

   };

   C_Date("top", 2013, 0)

   {

    z=0.5;

   };

   C_Date("8.5", 2004.4,5)

   {

    z=8.5;

   };

   C_Date("12.5", 1987.2,5)

   {

    z=12.5;

   };

   C_Date("16.5", 1970,5)

   {

    z=16.5;

   };

   C_Date("24.5", 1965.5,5)

   {

    z=24.5;

   };

   C_Date("32.5", 1963.3,5)

   {

    z=32.5;

   };

   C_Date("40.5", 1961.7,5)

   {

    z=40.5;

   };

   C_Date("48.5", 1960.5,5)

   {

    z=48.5;

   };

   C_Date("56.5", 1957.8,5)

   {

    z=56.5;

   };

   C_Date("64.5", 1953.7,5)

   {

    z=64.5;

   };

   C_Date("72.5", 1944.7,5)

   {

    z=72.5;

   };

   C_Date("80.5", 1935.1,5)

   {

    z=80.5;

   };

   C_Date("84.5", 1924.8,5)

   {

    z=84.5;

   };

   C_Date("88.5", 1917.6,5)

   {

    z=88.5;

   };

   C_Date("92.5", 1913.3,5)

   {

    z=92.5;

   };

   C_Date("96.5", 1908.6,5)

   {

    z=96.5;

   };

   C_Date("104.5", 1901.2,5)

   {

    z=104.5;

   };

   C_Date("108.5", 1895.2,5)

   {

    z=108.5;

   };

   C_Date("112.5", 1889.4,5)

   {

    z=112.5;

   };

   C_Date("120.5", 1880.1,5)

   {

    z=120.5;

   };

   C_Date("128.5", 1872.4,5)

   {

    z=128.5;

   };

   C_Date("136.5", 1866.3,5)

   {

    z=136.5;

   };

   C_Date("144.5", 1861.8,5)

   {

    z=144.5;

   };

   C_Date("152.5", 1859.6,5)

   {

    z=152.5;

   };

   C_Date("160.5", 1859.6,5)

   {

    z=160.5;

   };

   C_Date("168.5", 1858.8,5)

   {

    z=168.5;

   };

   C_Date("176.5", 1850,5)

   {

    z=176.5;

   };

   C_Date("184.5", 1841.2,5)

   {

    z=184.5;

   };

   C_Date("192.5", 1832.5,5)

   {

    z=192.5;

   };

   Boundary("Bottom")

   {

    z=296.5;

   };

  };

 };

Thanks!

Joanna

[Joanna Tindall]

Although this one is still time travelling forwards!

 Options()

 {

  BCAD=TRUE;

  PlusMinus=FALSE;

 };

 Plot()

 {

  P_Sequence(, )

  {

   Boundary("Top")

   {

    z=0;

   };

   C_Date("0", 2019, 0)

   {

    z=0;

   };

   C_Date(2019, 0)

   {

    z=0.5;

   };

   C_Date("5.5", AD(2013), 3)

   {

    z=5.5;

   };

   C_Date("10.5", AD(2003), 5)

   {

    z=10.5;

   };

   C_Date("15.5",AD(1991), 7)

   {

    z=15.5;

   };

   C_Date("15.5", AD(1972), 10)

   {

    z=20.5;

   };

   C_Date("20.5",AD(1963), 10)

   {

    z=25.5;

   };

   C_Date("40", AD(1961.75), 10)

   {

    z=40;

   };

   C_Date("40", AD(1957.76), 10)

   {

    z=57;

   };

   C_Date("57",AD(1953.71), 10)

   {

    z=65;

   };

   Boundary("Bottom")

   {

    z=260;

   };

  };

 };

[Erik Marsh]

Joanna – OxCal expects dates within sequences to start with the deepest and oldest dates and work your way towards the top with younger ages (in order of deposition). Give that try and it should work better, though no guarantees it will still be able to predict the future.

Erik