spc: Derive Spatial Predictive Components

Description Usage Arguments Value Note Author(s) See Also Examples

Description

Derives Spatial Predictive Components for a given set of covariates. It wraps the stats::prcomp method and predicts a list principal components for an object of type "SpatialPixelsDataFrame".

Usage

1
2
3
4
5
6
## S4 method for signature 'SpatialPixelsDataFrame,formula'
spc(obj, formulaString, scale. = TRUE, 
      silent = FALSE, ...)
## S4 method for signature 'list,list'
spc(obj, formulaString, scale. = TRUE, 
      silent = FALSE, ...)

Arguments

obj

object of class "SpatialPixelsDataFrame" (must contain at least two grids) or a list of objects of type "SpatialPixelsDataFrame"

formulaString

object of class "formula" or a list of formulas

scale.

object of class "logical"; specifies whether covariates need to be scaled

silent

object of class "logical"; specifies whether to print the progress

...

additional arguments that can be passed to stats::prcomp

Value

spc returns an object of type "SpatialComponents". This is a list of grids with generic names PC1,...,PCp, where p is the total number of input grids.

Note

This method assumes that the input covariates are cross-correlated and hence their overlap can be reduced. The input variables are scaled by default and the missing values will be replaced with 0 values to reduce loss of data due to missing pixels. This operation can be time consuming for large grids.

Author(s)

Tomislav Hengl

See Also

stats::prcomp, SpatialComponents-class

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
# load data:
library(plotKML)
library(sp)

pal = rev(rainbow(65)[1:48])
data(eberg_grid)
gridded(eberg_grid) <- ~x+y
proj4string(eberg_grid) <- CRS("+init=epsg:31467")
formulaString <- ~ PRMGEO6+DEMSRT6+TWISRT6+TIRAST6
eberg_spc <- spc(eberg_grid, formulaString)
names(eberg_spc@predicted) # 11 components on the end;
## Not run: # plot maps:
rd = range(eberg_spc@predicted@data[,1], na.rm=TRUE)
sq = seq(rd[1], rd[2], length.out=48)
spplot(eberg_spc@predicted[1:4], at=sq, col.regions=pal)

## End(Not run)

GSIF documentation built on May 2, 2019, 5:44 p.m.

Related to spc in GSIF...