NAME¶
r.statistics - Calculates category or object oriented statistics.
KEYWORDS¶
raster, statistics
SYNOPSIS¶
r.statistics
r.statistics help
r.statistics [-
c]
base=
name cover=
name
method=
string [
output=
name] [--
overwrite]
[--
verbose] [--
quiet]
Flags:¶
- -c
-
Cover values extracted from the category labels of the cover map
- --overwrite
-
Allow output files to overwrite existing files
- --verbose
-
Verbose module output
- --quiet
-
Quiet module output
Parameters:¶
- base=name
-
Name of base raster map
- cover=name
-
Name of cover raster map
- method=string
-
Method of object-based statistic
Options:
diversity,distribution,average,mode,median,avedev,stddev,variance,skewness,kurtosis,min,max,sum
- output=name
-
Resultant raster map (not used with 'distribution')
DESCRIPTION¶
r.statistics is a tool to analyse exploratory statistics of a "cover
layer" according to how it intersects with objects in a "base
layer". A variety of standard statistical measures are possible (called
"zonal statistics" in some GIS). All cells in the base layer are
considered one object for the analysis. For some applications, one will first
want to prepare the input data so that all areas of contiguous cell category
values in the base layer are uniquely identified, which can be done with
r.clump.
The available methods are the following (english - german):
- distribution - Verteilung in Prozent
- average - Durchschnitt
- mode - Modalwert
- median - Median
- average deviation - Durchschnittliche Abweichung
- standard deviation - Standardabweichung
- variance - Varianz
- skewness - Schiefe der Verteilung
- kurtosis - Relative Abflachung (-) oder Zuspitzung (+)
- minimum - Minimum
- maximum - Maximum
- sum - Summe
The calculations will be performed on each area of data of the cover layers
which fall within each unique value, or category, of the base layer.
Setting the
-c flag the category lables of the covering raster layer will
be used. This is nice to avoid the GRASS limitation to interger in raster maps
because using category values floating point numbers can be stored.
All calculations except "distribution" create an output layer. The
output layer is a reclassified version of the base layer with identical
category values, but modified category labels - the results of the
calculations are stored in the category labels of the output layer.
For distributions, the output is printed to the user interface (stdout). If an
output file name was specified, it will be ignored. The result will be a text
table with three columns. In the first column are the category values of the
base layer (a), in the second column the associated value of the cover layers
(b), and in the third column the percentage of area in that base layer
category (a) that falls into that row's value in the cover layer(b). Example:
1 124 23
1 201 47
1 273 30
2 101 5
2 152 16
2 167 60
2 187 19
.
.
.
So for the first line in the output above, we see that 23% of the cells of the
base layer category 1 have a value of 124 in the cover layer. To transfer the
values stored as category labels into cell values,
r.mapcalc can be
used ('@' operator).
EXAMPLES¶
Calculation of average elevation of each field in the Spearfish region:
r.statistics base=fields cover=elevation.dem out=elevstats method=average
r.category elevstats
r.mapcalc "fieldelev=@elevstats"
r.univar fieldelev
SEE ALSO¶
r.average, r.clump, r.mode, r.median,
r.mapcalc, r.neighbors, r.univar r.category
AUTHOR¶
Martin Schroeder, Geographisches Institut Heidelberg, Germany
Last changed: $Date: 2011-11-08 12:29:50 +0100 (Tue, 08 Nov 2011) $
Full index
© 2003-2014 GRASS Development Team