v.what.rast.1grass man page

v.what.rast — Uploads raster values at positions of vector points to the table.


vector, sampling, raster, position, querying, attribute table, surface information


v.what.rast --help
v.what.rast [-ip] map=name [layer=string] [type=string[,string,...]] raster=name [column=name] [where=sql_query] [--help] [--verbose] [--quiet] [--ui]


Interpolate values from the nearest four cells
Print categories and values instead of updating the database
Print usage summary
Verbose module output
Quiet module output
Force launching GUI dialog


map=name [required]
Name of vector points map for which to edit attributes
Or data source for direct OGR access
Layer number or name
Vector features can have category values in different layers. This number determines which layer to use. When used with direct OGR access this is the layer name.
Default: 1
Input feature type
Options: point, centroid
Default: point
raster=name [required]
Name of existing raster map to be queried
Name of attribute column to be updated with the query result
WHERE conditions of SQL statement without ’where’ keyword
Example: income < 1000 and inhab >= 10000


v.what.rast retrieves raster value from a given raster map for each point or centroid stored in a given vector map. It can update a column in the linked vector attribute table with the retrieved raster cell value or print it. The column type needs to be numeric (integer, float, double, ...).

If the -p flag is used, then the attribute table is not updated and the results are printed to stdout.

If the -i flag is used, then the value to be uploaded to the database is interpolated from the four nearest raster cells values using an inverse distance weighting method (IDW). This is useful for cases when the vector point density is much higher than the raster cell size.

Points and centroid with shared category number cannot be processed. To solved this, unique categories may be added with v.category in a separate layer.


If multiple points have the same category, the attribute value is set to NULL. If the raster value is NULL, then attribute value is set to NULL.

v.what.rast operates on the attribute table. To modify the vector geometry instead, use v.drape.

Categories and values are output unsorted with the print flag. To sort them pipe the output of this module into the UNIX sort tool (sort -n). If you need coordinates, after sorting use v.out.ascii and the UNIX paste tool (paste -d’|’). In the case of a NULL result, a "*" will be printed in lieu of the value.

The interpolation flag is only useful for continuous value raster maps, if a categorical raster is given as input the results will be nonsense. Since the search window is limited to four raster cells there may still be raster cell-edge artifacts visible in the results, this compromise has been made for processing speed. If one or more of the nearest four raster cells is NULL, then only the raster cells containing values will be used in the weighted average.


Transferring raster values into existing attribute table of vector points map

Reading values from raster map at position of vector points, writing these values into a column of the attribute table connected to the vector map:

# work on copy of original geodetic points map
g.copy vector=geodetic_pts,mygeodetic_pts
# set computational region to raster map to be queried
g.region raster=elev_state_500m -p
# add new column to existing table
v.db.addcolumn map=mygeodetic_pts column="height double precision"
v.what.rast map=mygeodetic_pts raster=elev_state_500m column=height
# compare official geodetic heights to those of elevation model
v.db.select map=mygeodetic_pts columns=Z_VALUE,height separator=comma

Transferring raster values into new vector points map

In case of a vector map without attached attribute table, first add a new attribute table. This table is then populated with values queried from the raster map:

# create new random vector points map
v.random pnts n=100
# add new table, link to map
v.db.addtable map=pnts column="height double precision"
# set computational region to raster map to be queried
g.region raster=elevation -p
# query raster map and upload values to vector table into specified column
v.what.rast map=pnts raster=elevation column=height
# verify new attribute table:
v.db.select pnts
# verify statistics of uploaded values:
v.univar map=pnts column=height type=point

See Also

v.category, v.db.addtable, v.db.select, v.drape, v.univar, v.rast.stats, v.what.vect


Radim Blazek
Hamish Bowman (interpolation)

Last changed: $Date: 2015-09-10 12:51:40 +0200 (Thu, 10 Sep 2015) $

Main index | Vector index | Topics index | Keywords index | Full index

© 2003-2016 GRASS Development Team, GRASS GIS 7.0.4 Reference Manual


GRASS 7.0.4 Grass User's Manual