wtConfigureSolution: Solver-related parameter values

In KWB-R/kwb.wtaq: Interface to WTAQ Drawdown Model (http://water.usgs.gov/ogw/wtaq/)

Solver-related parameter values

Description

Define WTAQ parameters related to the solver algorithms

Usage

wtConfigureSolution(isoln = 2, rerrnr = 1e-10, rerrsum = 0, nmax = 0, 


    ntms = 30, ns = 8, error = 1e-04, nnn = 6, method = 3)

Arguments

isoln	Numerical-inversion solution type: ISOLN = 1: Solution by the Stehfest algorithm (must use this option for confined aquifers). ISOLN = 2: Solution by the de Hoog algorithm (must use this option for drainage with unsaturated-zone characterization [wtConfigureDrainage()$idra == 2]).
rerrnr	Relative error for Newton-Raphson iteration and finite summations of drawdown for water-table aquifers. A value of 1.0E-10 is suggested. Enter 0.0D0 for AQTYPE = CONFINED.
rerrsum	Only used if isoln = 1. Relative error for finite summations of drawdown for confined aquifers. Suggested value is 1.0E-07 to 1.0E-08. Enter 0 if AQTYPE = WATER TABLE.
nmax	Only used if isoln = 1. Maximum number of terms permitted in the finite summations of drawdown for confined aquifers. Suggested value is 200. Enter 0 if AQTYPE = WATER TABLE.
ntms	Factor used to determine number of terms in the finite summations for drawdown for water-table aquifers. Suggested values are 20 or 30. Enter 0 if AQTYPE = CONFINED.
ns	Only used if isoln = 1. Number of terms used in the Stehfest algorithm. This must be an even integer, the value of which depends upon computer precision. If the computer holds 16 significant figures in double precision, let NS = 6 to 12. A value of 8 is recommended.
error	Only used if isoln = 2. Relative error sought for the accuracy of the numerical inversion. A value of 1.0E-04 is suggested.
nnn	Only used if isoln = 2. Number of terms used in the summation of the Fourier series of the approximation to the inverse Laplace transform. A value of 6 is suggested.
method	Only used if isoln = 2. Indicates which method will be used to accelerate convergence of the Fourier series. Options are 1, 2, or 3. Only METHOD = 3 has been tested and was found to be satisfactory. Users can consult de Hoog and others (1982) and John Knight's subroutine LAPADC for additional details if needed.

Value

list with elements isoln, rerrnr, rerrsum, nmax,

ntms, ns, error, nnn, method,

representing WTAQ parameters related to the solving algorithms used in WTAQ.

See descriptions in Arguments section.

Author(s)

Hauke Sonnenberg

See Also

wtConfigure

KWB-R/kwb.wtaq documentation built on June 17, 2022, 3:05 a.m.