banner.jpg (24775 bytes)
HOME.gif (109 bytes) SCHEDULE-1.gif (136 bytes) PERSONNEL-1.gif (148 bytes) PUBLICATIONS-1.gif (155 bytes) MDDProg.gif (146 bytes) ARCHIVE.gif (136 bytes) CONTACTUS-1.gif (147 bytes)

AGEMON_MULTI.F PROGRAM

 

INTRODUCTION

  • This program was developed to calculate a set of cooling history from the set of domain distributions obtained from the arrmulti.f program.   (domain distributions corresponding to pairs of E,log(Do/ro2) normally distributed arount the best least square fit. 

  • The program is similar to autoage-mon.f except it now by default calculates 5 best fit cooling histories for each pair E,log(Do/ro2) and corresponding domain distribution.
  • The temperature at the initial age is randomly pick between 350°C to 600°C. Accordingly, the requirements of starting the modelling at high temperature have been removed. However, it introduces the implicit assumption that the starting time of the model corresponds to an age where the sample has been completely reset (zero initial argon concentration). Therefore, we suggest that the model be started at an age at least 5Ma or 25% older than the max. age shown in the sample age spectrum.
  • The input and output file names have also been personalized to the sample name to allow running of multiple samples in one single step and in the same directory.
  • The program only models the heating steps that were degassing at temperatures below 1100°C (below the onset of melting).

.

PROGRAM DESCRIPTION
The agemon_multi.f routine was develop to automatically estimate best fit cooling histories to a k-feldspar age spectrum (40Ar/39Ar data), once the diffusion and distribution parameters of the sample had been estimated by modeling its 39Ar data with the arrmulti.f program. The program starts using an initial guessed cooling history (CH). The first derivative of the cooling history is approached using an expansion in Chebyshev polynomials. It allows restricting the family of possible CH approximations to only monotonic cooling. Subsequent better estimates of the CH are obtained adjusting the coefficients of the Chebyshev polynomials by a iterative variational process (Levenberg-Marquardt method). The iteration stop when an acceptable minimum of chi-square (square difference between the model and the data) is obtained. The best fitting CH is recovered by integration of its first derivative.

The program requires the following input files.

  • N13KSP_TMP.IN: Input file containing the heating schedule identical to temstep.in (File created by the mddfiles_lst.f program). Same file used as input in the arrmulti.f program.
  • N13KSP_AGE.IN: Two columns list (X,Y) of the Cumulative % of 39Ar released vs. the Apparent Age (Ma) for each step heating.
  • N13KSP_SIG.IN One column list of 1-sigma uncertainty of the apparent age for each step heating.
  • N13KSP.AME Diffusion and distribution parameters as estimated by the arrmulti.f routine. Output file of the arrmulti.f program.
  • LIST.TXT (optional) Listing file to run several samples in the same directory on a single step.  The listing file comprise a two column list of the samples names and max. plateau ages separated by comma or spaces. If spaces are used, make sure that the max. plateau ages
    appear at the 11 row (see example).

The program creates the following output files.

  • N13KSP_MCH-OUT.DAT: All best fit CHs obtained from each  E,log(Do/ro2)pair (5 CHs for each pair).
  • N13KSP_MAGES-OUT.DAT: Best fit age spectra obtained from each  E,log(Do/ro2)pair (5 calculation for each pair).

The program also output the following files for debugging use only

  • N13KSP_E#_MPAR.OUT: Output the running parameters used during the variational calculation. (# is the number of each E,log(Do/ro2) pair, subsequently number from 1 to 10)
  • N13KSP_E#_MCHISQ.DAT: List of best chi-squares obtained for each run for the # pair.
  • N13KSP_E#_MCH-OUT.DAT: All best fit CHs obtained from the # pair.
  • N13KSP_E#_MAGES-OUT.DAT: All best fit age spectra obtained from the # pair.
  • STATUS:  Comprise information about the status of the process (# runs and energy done so far).

Note: Cooling history and age data are given in a two-column X|Y format. Each set of data (age or cooling history) is separated by '&' at the end of each set. 

RUNNING THE PROGRAM
To run the program follows these steps:
  • If you have run the program mddfiles_lst.f then your input files are ready.  Otherwise, you will need to manually create the files samplename_age.in and samplename_sig.in following the example files aboved (sample n13ksp).
  • The program arrmulti.f should have been run, check that its output file samplename.ame and the input files samplename_tmp.in exist.
  • Run the program from the directory where your input files are. NOTE: Sometimes is better to run the program in a less busy machine, To do that, first make a rlogin to that machine and change to the directory you are working, then run the program.
  • After you start the program it will prompt you some inputs (read the running example below for examples answers and hints)

Running example 1:  (running a single sample with default values)

   Sample name MUST have 10 or less characters
Do you want enter samples from a list (yes/no)?
no
Enter sample name (stop to exit)
12h
Insert max. plateau age
75
Do you want to run the default values
yes
42.0250440000000
40.8924500000000

..........

Note:  As the max. plateau age enter a value somewhat higher than the highest radiogenic age registered on your age spectru (without Excess of argon contamination).  This is the age at which the model will start. 

 
Running example 1:  (running a multiple samples from a list file)

Sample name MUST have 10 or less characters
Do you want enter samples from a list (yes/no)?
yes
---------------------------------------
The list file must have sample names separated
from their max. plateau age by comma or spaces, no tabs,
If spaces are used, make sure that the max. plateau ages
appear at the 11 row
---------------------------------------
Enter listing file name
list.txt
12h
42.0250440000000
40.8924500000000

.......

Note: the max. plateau age input is incorporated on the listing file (list.txt), see above.
Running example 1:  (running a single sample with optional values)

Sample name MUST have 10 or less characters
Do you want enter samples from a list (yes/no)?
no
Enter sample name (stop to exit)
12h
Insert max. plateau age
75
Do you want to run the default values
no
Enter # of Es from E-gaussian distribution use to
calculate the CHs (Max.= 10 )
4
Enter # CHs calculate per E (Max=50))
10
42.0250440000000
40.8924500000000

......

Note: Using the extra options you can run a subset (or only the peak E) from the gaussian set of pairs, or increase the # of cooling histories calculates per pair.   Note that the first pair corresponds with the gaussian peak, best-fit E,Do values.
HINTS and COMMENTS
  • The program autoage-mon.f is still in its experimental stage and had not been fully neither tested nor optimized. The running could take a long time, from hours to days depending on the number of runnings you select. If you run the program for the first time or are not sure about a new sample, just start by checking the program with a few numbers of runnings (~5).
  • Steps with anomalous age (i.e. excess of argon at initial steps), produce larges value of chi-square sometimes preventing the program of getting better fit over others more important portions of the age spectrum. A way to avoid this problem is to increase the value of the uncertainty in such anomalous steps modifying the file sig.in, or by changing the file age.in. In general, the absolute chi-square values are meaningless since not all source of errors are considered in the 1sd age uncertainty.