DESCRIPTION

r.univar calculates the univariate statistics of one or several raster map(s). This includes the number of cells counted, minimum and maximum cell values, range, arithmetic mean, population variance, standard deviation, coefficient of variation, and sum. Statistics are calculated separately for every category/zone found in the zones input map if given. If the -e extended statistics flag is given the 1st quartile, median, 3rd quartile, and given percentile are calculated. If the -g flag is given the results are presented in a format suitable for use in a shell script. If the -t flag is given the results are presented in tabular format with the given field separator. The table can immediately be converted to a vector attribute table which can then be linked to a vector, e.g. the vector that was rasterized to create the zones input raster.

When multiple input maps are given to r.univar, the overall statistics are calculated. This is useful for a time series of the same variable, as well as for the case of a segmented/tiled dataset. Allowing multiple raster maps to be specified saves the user from using a temporary raster map for the result of r.series or r.patch.

NOTES

As with most GRASS raster modules, r.univar operates on the raster array defined by the current region settings, not the original extent and resolution of the input map. See g.region.

This module can use large amounts of system memory when the -e extended statistics flag is used with a very large region setting. If the region is too large the module should exit gracefully with a memory allocation error. Basic statistics can be calculated using any size input region.

Without a zones input raster, the r.quantile module will be significantly more efficient for calculating percentiles with large maps.

For calculating univariate statistics from a raster map based on vector polygon map and uploads statistics to new attribute columns, see v.rast.stats.

EXAMPLE

In this example, the raster polygon map basins in the North Carolina sample dataset is used to calculate raster statistics for zones for elevation raster map:

g.region raster=basins -p

This will set and print computational region in the format:

projection: 99 (Lambert Conformal Conic)
zone:       0
datum:      nad83
ellipsoid:  a=6378137 es=0.006694380022900787
north:      228500
south:      215000
west:       630000
east:       645000
nsres:      10
ewres:      10
rows:       1350
cols:       1500
cells:      2025000

Check basin's IDs using:

r.category basins

This will print them in the format:

Visualization of them underlying elevation map can be created as:

d.mon wx0
d.rast map=elevation
r.colors map=elevation color=grey 
d.rast map=basins
r.colors map=basins color=bgyr 
d.legend raster=basins use=2,4,6,8,10,12,14,16,18,20,22,24,26,28,30
d.barscale

Figure: Zones (basins, opacity: 60%) with underlying elevation map for North Carolina sample dataset.

Then statistics for elevation can be calculated separately for every zone, i.e. basin found in the zones parameter:

r.univar -t map=elevation zones=basins separator=comma \
         output=basin_elev_zonal.csv

This will print information in the format:

zone,label,non_null_cells,null_cells,min,max,range,mean,mean_of_abs,
stddev,variance,coeff_var,sum,sum_abs2,,116975,0,55.5787925720215,
133.147018432617,77.5682258605957,92.1196971445722,92.1196971445722,
15.1475301152556,229.447668592576,16.4433129773355,10775701.5734863,
10775701.57348634,,75480,0,61.7890930175781,110.348838806152,
48.5597457885742,83.7808205765268,83.7808205765268,11.6451777476995,
135.610164775515,13.8995747088232,6323776.33711624,6323776.33711624
6,,1137,0,66.9641571044922,83.2070922851562,16.2429351806641,
73.1900814395257,73.1900814395257,4.15733292896409,17.2834170822492,
5.68018623179036,83217.1225967407,83217.12259674078,,80506,
0,67.4670791625977,147.161514282227, ...

Comma Separated Values (CSV) file is best viewed through a spreadsheet program such as Microsoft Excel, Libre/Open Office Calc or Google Docs:

Figure: Raster statistics for zones (basins, North Carolina sample dataset) viewed through Libre/Open Office Calc

TODO

To be implemented mode, skewness, kurtosis.

AUTHORS

Hamish Bowman, Otago University, New Zealand
Extended statistics by Martin Landa
Multiple input map support by Ivan Shmakov
Zonal loop by Markus Metz

Last changed: $Date$

DESCRIPTION

NOTES

EXAMPLE

TODO

SEE ALSO

AUTHORS