[R-sig-Geo] subregion of a SpatialGridDataFrame

Paul Hiemstra p.hiemstra at geo.uu.nl
Wed Feb 6 14:11:39 CET 2008


Hi,

Loading the grid and telling R that it is a grid can also be done by:

library(gstat)
data(meuse.grid)
gridded(meuse.grid) =~x+y # Now it is a SPixelDF
# The former is shorthand for:
# coordinates(meuse.grid) =~x+y
# gridded(meuse.grid) = TRUE
fullgrid(meuse.grid) = TRUE # Make it a SGridDF

cheers,
Paul


Edzer Pebesma wrote:
> Ingo Holz wrote:
>   
>> Hi,
>>
>>  this sounds like a stupid question, however:
>>
>>  I have a SpatialGridDataFrame (SGDF) and want to cut a rectangular 
>> subregion from this SGDF to a new sub-region-SGDF.
>>
>>  Is it possible to define a new bounding-box with two points and cut the 
>> subregion out of the SGDF?
>>
>> Ingo
>>
>> _______________________________________________
>> R-sig-Geo mailing list
>> R-sig-Geo at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>   
>>     
>  > data(meuse.grid)
>  > coordinates(meuse.grid)=~x+y
>  > meuse.grid=as(meuse.grid, "SpatialGridDataFrame")
> Error in as(meuse.grid, "SpatialGridDataFrame") :
>   no method or default for coercing "SpatialPointsDataFrame" to 
> "SpatialGridDataFrame"
> # this surprised me, but the following does work:
>  > meuse.grid=as(meuse.grid, "SpatialPixelsDataFrame")
>  > meuse.grid=as(meuse.grid, "SpatialGridDataFrame")
>  > image(meuse.grid[40:60,20:50,1])
>
> # meaning the first index selects rows (top to bottom), and the second 
> columns; the third attribute
>
>  > bbox(meuse.grid[40:60,20:50,1])
>      min    max
> x 179200 180440
> y 331360 332200
>  > x1 = bbox(meuse.grid[40:60,20:50,1])[1,1]
>  > x2 = bbox(meuse.grid[40:60,20:50,1])[1,2]
>  > y1 = bbox(meuse.grid[40:60,20:50,1])[2,1]
>  > y2 = bbox(meuse.grid[40:60,20:50,1])[2,2]
>  > p = Polygon(cbind(c(x1,x2,x2,x1,x1),c(y1,y1,y2,y2,y1)))  
>  > sps = SpatialPolygons(list(Polygons(list(p),"bbox")))
>  > meuse.grid=as(meuse.grid,"SpatialPixelsDataFrame")
>  > image(meuse.grid[!is.na(overlay(meuse.grid,sps)),1])
>
> # this second form needs the grid represented as pixels, as the overlay 
> considers each grid cell, and retuns a vector the length of the number 
> of pixels. The image should be identical to the former.
> --
> Edzer
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>   


-- 
Drs. Paul Hiemstra
Department of Physical Geography
Faculty of Geosciences
University of Utrecht
Heidelberglaan 2
P.O. Box 80.115
3508 TC Utrecht
Phone: 	+31302535773
Fax:	+31302531145
http://intamap.geo.uu.nl/~paul




More information about the R-sig-Geo mailing list