MAN PAGE

r.1Dcorrelogram - generates one dimensional correlograms and variograms from surfaces.

(GRASS Raster Program)

SYNOPSIS

r.1Dcorrelogram
r.1Dcorrelogram help
r.1Dcorrelogram [-q] input=name(s) lags=value(s) lagnames=name(s) [max_distance=value]

DESCRIPTION

r.1Dcorrelogram generates one dimensional correlogram(s) and variograms of a surface. Non-masked cell values are used to determine the correlogram(s) and variogram(s). Correlograms and variograms are used to determine the spatial correlation of a map. Cells close to a location are more likely to be similar or identical to that location: r.1Dcorrelogram provides a measure of similarity of cells at various distances. The output, placed in file(s) named in lagnames, contains the correlation of pairs of cells grouped into ranges of distances. For example, all pairs of cells with a distance between 0 and 100 meters will be grouped together if the lags value is 100.

OPTIONS

-q This flag prevents the program from informing user of program's progress.

input Input map(s): Map(s) for which correlation is calculated. Map cells should contain continuous values. Spatial correlation of categorical data can be done by reclassifying categories of interest to 1 and other categories to 0. Cell values that should not be included in the correlation should be masked out using r.mask.

lags Input values(s): Distance increment of lags. Each lag value will generate a different lag table of correlation values. Each table will contain correlation of pairs of cells grouped by distance. If a particular lag value is x, groups will be 0-x, x-2x, ..., and nx-max_distance.

lagnames Input name(s): Name(s) of ASCII file(s) containing lag table(s). There must be a name for each lags value.

max_distance Optional input value: max_distance is the maximum distance that r.1Dcorrelogram will calculate correlations out to. If max_distance is not used, r.1Dcorrelogram will calculate correlations out to the extent of the current region.

NOTES

r.1Dcorrelogram uses the correlogram function and the variogram function to determine the spatial correlation of a map. The correlogram function is sensitive to data outliers making it unsuitable for some applications. The correlogram function was chosen by the author for error modeling applications.

SEE ALSO

Isaaks, E. and Srivastava, R., 1989, An Introduction to Applied Geostistics, Oxford University Press, New York.

r.random.surface, r.random.model, r.2Dcorrelogram, r.2Dto1D, r.mapcalc, r.mask, r.random

AUTHOR

Charles Ehlschlaeger, Department of Geography, University of Cincinnati.