NAME
m.svfit - Semivariogram model fitting.
(GRASS Data Import/Processing Program)
SYNOPSIS
m.svfit
m.svfit help
m.svfit [-pqw]
[input=name]
model=value
range=value
[graph=name]
DESCRIPTION
m.svfit calculates a sample semivariogram and
either plots it in the GRASS graphics window or writes the
estimated parameters to standard output, or both.
For more information, refer to the
tutorial or see
the example below.
OPTIONS
Flags:
- -q
- Quiet. Cut out the chatter.
- -p
- Plot model and sample semivariogram (requires
g.gnuplot).
- -w
- Use weighted least squares (default is general least squares)
Parameters:
sites=name
UNIX file containing sample semivariogram (see NOTES).
Default is standard input.
model=value
Integer Model Index, one of
- Linear,
- Spherical,
- Exponential,
- Gaussian,
- Quadratic, or
- Wave or Hole Effect.
range=value
Range of semivariogram.
graph=name
Basename to save graphing data/commands files.
Graphs are saved in the current working directory with
the extensions .gp and .dat. Implies
the -p flag.
NOTES
Three columns of data are expected as input: lag distance
(h), semivariogram value (gamma), and the
number of data pairs used to compute it (N(h)).
This may either be from a UNIX file, entered from the
command line (terminated by control-d), or via a pipe or
redirection.
EXAMPLE
m.svfit was designed to be used in conjunction
with
s.sv,
a GRASS sites program for calculating sample
semivariograms. The following example calculates a sample
semivariogram of the sites list wells with a
nominal lag distance of 5 and then fits a linear model with
a range of 100. The sample semivariogram and model are
plotted in the GRASS graphics monitor and the graphing
instructions and data are saved to files with the basename
svwells in the current working directory:
- s.sv -q wells lag=5 | m.svfit -p m=1 r=100 g=svwells
By saving the graphing instructions and data, the
semivariogram may be plotted again later by the following
command:
- g.gnuplot svwells.gp
SEE ALSO
s.univar,
s.normal,
g.gnuplot,
s.sv
and
Semivariogram Modeling -
A GRASS Tutorial on Exploratory
Data Analysis and Semivariogram Modeling.
BUGS
Please send all bug fixes and comments to the author.
AUTHOR
James Darrell McCauley,
Agricultural
Engineering,
Purdue University