University of FloridaDepartment of Agricultural & Biological Engineering

 

VGPest3


Description

This program estimates the 3 parameters (a,n,r) for the van Genuchten's Mualem soil-water characteristic curve formula by using the optimization subroutine LMDIF1 from the math library MINPACK (Argonne National Laboratory). VGPest3 reads the Soil and Water Characteristic Curve (SWCC) laboratory values, water content (, cm3/cm3) vs suctions (h, -cm water) and calculates the parameters such as:

  • θ_c(h) = θr + (θs-θr)×[1+(a.h)^n]^(-m); m = 1 -1/n (1)

and then calculates the unsaturated conductivity curve as:

  •           θ_c(h) - θr
    Se = ---------------- (2)
                  θs - θr
  • Kr(Se) = Ku/Ks = Se^0.5 [1-(1-Se^(1/m))^m]^2 (3)

where:

  • θ(h), θ_c(h) = Measured and predicted water contents
    Se(i,2) = effective saturation (j = 2) for each pressure step (j = 1)
    SWCC(i,2) = Water content (j = 2) for each suction step (j = 1)
    θs, θr, θr_c = Saturated and Residual water content (_c = predicted)
    Ks (cm/h) = Saturated hydraulic conductivity.

Program Usage & Output

The program runs in DOS mode and is written in Fortran (main engine) and C++ (interface). Command line UNIX and DOS versions are available from the authors.

Unpack the contents of the ZIP package on a directory. To run the program write "vg" at the command line and a graphical interfase will open in the screen. Follow instructions as given. The program allows to select the initial parameters for the optimization. After convergence is reached, the program shows the results on a graphic mode.

Input File Example

Before starting the program the input file must be written. The program input is contained in the input file swcc.in written in ASCII format. The user should edit this file replacing the contents with the values from the case to analyze. Please notice that this file must be saved in ACSII format for the program to run properly. This is a sample input file:

 

Title (30 characters max): "Lysimeter Soil 2-35cm 28/4/94"
alpha,n,r: 1.000 1.000 0.1
Ks(cm/h): 4.8
(hi,θi); NOTE: For i=1 → (0,θs): 0. 0.475108
3.81 0.442734
13.81 0.401439
28.81 0.372662
53.81 0.343165
103.81 0.315827
203.81 0.292086
403.81 0.274101
15000 0.2056

Output Example

To run the program in the command line type the program name followed by the name of the input file ASCII text file with the contents shown above:

> VGpest3 swcc.in

This will produce the complete screen output shown below (it can be redicted to a text file (i.e. VGpest3 swcc.in> scwcc.out).

              \      --   ----   ----    ----   ---  --- ----
              |    /    /       /    /  /     /        /
              |   /    /  ---  /----   ----   ---     /
              |  /    |    /  /       /          /   /
              |/       ---   /       -----  ----    /

          ************************************************
          *        VGPEST-Version 1.1, 15/Spet/97        *
          *>>>>>>3 PARAMETER VERSION (Wr, alpha, n)<<<<<<*
          *           Rafael Munoz-Carpena               *
          * U. of Florida (USA). e-mail:carpena@ufl.edu  *
          *                                              *
          ************************************************

 ----Press any letter and return to continue---------

                     INTRODUCTION
 =======================================================
program estimates the parameters (a, n,Wr) for the van soil-water characteristic curve model.It uses the Levenberg-Marquardt non-linear optimization algorithm (subroutine LMDIF1, MINPACK math library, Argonne National Laboratory).The program reads in the SWCC lab values, water content(cm3/cm3) vs pressure heads (-cm water), and fits the parameters such as:

    W_c(h) = Wr + (Ws-Wr)*[1+(a.h)^n]^(-m) ;m=1 -1/n  (1)

It then calculates the unsaturated conductivity curve as:

           W_c(h)- Wr
    Se(h)= ---------- = [1+(a.h)^n]^m                 (2)
            Ws - Wr

    Kr(Se)=Ku/Ks= Se^0.5 [1-(1-Se^(1/m))^m]^2        (3)
where:
W(h),W_c(h)=Soil-water characteristic curve (_c=predicted)
Se(h)      = Effective saturation for each pressure step
Ws,Wr      = Saturated and Residual water content
Ks(cm/h)   = Saturated conductivity

 ----Press any letter and return to continue---------

                    INPUTS
 =======================================================
SAMPLE DATA SET: Lysimeter soil 2-35cm 28/4/94    
 
   h(cm)   W(cm3/cm3)
 ----------------------
    0.00  0.475108
    3.81  0.442734
   13.81  0.401439
   28.81  0.372662
   53.81  0.343165
  103.81  0.315827
203.81  0.292086
  403.81  0.274101
15000.00 0.205600

 Ws= 0.475108000000000    
 Ks(cm/h)=  4.80000000000000    
 Tol.= 9.999999999999999E-021  ;Initial guess (alpha,n,Wr)=
  1.00000000000000        1.00000000000000       0.470000000000000    
 
 ----Press any letter and return to continue---------

                   RESULTS
   =======================================================
SAMPLE DATA SET: Suelo Lisimetro 2-35cm 28/4/94    

 Number of function evaluations performed:       84
 Algorithm estimates that the relative error in the
 sum of squares is at most tol, where FTOL=  0.10000E-11
 Parameters=0.18493E+00 0.12222E+01 0.14891E+00
               alpha         n        Wr
 
  h(cm)         W         W_c        Error        Kr      Kuns(cm/h)
---------------------------------------------------------------
    0.00    0.475108    0.475108    0.000000    1.000000    4.800000
    3.81    0.442734    0.446663   -0.003929    0.023106    0.110909
   13.81    0.401439    0.400803    0.000636    0.002105    0.010104
   28.81    0.372662    0.368924    0.003738    0.000393    0.001888
   53.81    0.343165    0.342622    0.000543    0.000086    0.000415
  103.81    0.315827    0.317282   -0.001455    0.000017    0.000081
  203.81    0.292086    0.294239   -0.002153    0.000003    0.000015
  403.81    0.274101    0.273907    0.000194    0.000001    0.000003
15000.00   0.205600    0.204950    0.000650    0.000000   0.000000
 
 Root mean sum of square errors for W(h) (RMSE)=  2.026878865761420E-003
 
                                  1/3 bar   1 bar   5 bar  15 bar
 Standard water content values:    0.28    0.25    0.22    0.20

 

Program License

This program is distributed as Freeware/Public Domain under the terms of GNU-License. If the program is found useful the authors ask that acknowledgment is given to its use in any resulting publication and the authors notified. The source code is available from the authors upon request:

  • Rafael Muñoz-Carpena
    Professor, Hydrology & Water Quality
    Department of Agricultural
     & Biological Engineering
    University of Florida
    P.O. Box 110570
    287 Frazier Rogers Hall
    Gainesville, FL  32611-0570

    (352) 392-1864 x287
    622-1864 x287 (Suncom)
    (352) 392-4092 (fax)
    carpena@ufl.edu

© Copyright 1996  Rafael Muñoz-Carpena & Javier Álvarez-Benedí


Return to top

This page was last updated on September 16, 2024.