Class for spatial attributes that have spatial locations on a regular grid.
class
Objects from the Class
Objects can be created by calls of the form as(x, "SpatialPixelsDataFrame"), where x is of class SpatialPointsDataFrame-class , or by importing through rgdal. Ordered full grids are stored instead or unordered non-NA cells;
Slots
bbox:: Object of class "matrix"; bounding box
proj4string:: Object of class "CRS"; projection
coords:: see SpatialPoints ; points slot
coords.nrs: see SpatialPointsDataFrame
grid:: see GridTopology-class ; grid parameters
grid.index:: integer; index of points in the list to points in the full (ordered) grid. x cycles fastest; all coordinates increase from low to hight except y, which decreases from high to low
data:: Object of class data.frame, containing the attribute data
Extends
Class "SpatialPixels", directly. Class "Spatial", by class "SpatialPixels".
[: signature(x = "SpatialPixelsDataFrame"): selects row(s) and/or attribute(s), and returns an object of class SpatialPixelsDataFrame; rows refer here to the pixel numbers, not grid lines. For selecting a square block in a grid, coerce to a SpatialGridDataFrame-class first, and use [ on that object
as.matrix: signature(x = "SpatialPixelsDataFrame"): coerce to matrix
SpatialPixels-class, which does not contain the attribute data
Examples
data(meuse.grid)# only the non-missing valued cellscoordinates(meuse.grid)= c("x","y")# promote to SpatialPointsDataFramegridded(meuse.grid)<-TRUE# promote to SpatialPixelsDataFramemeuse.grid[["idist"]]=1- meuse.grid[["dist"]]# assigns new attributeimage(meuse.grid["idist"])# note the single [# toy example:df = data.frame(z = c(1:6,NA,8,9), xc = c(1,1,1,2,2,2,3,3,3), yc = c(rep(c(0,1.5,3),3)))coordinates(df)=~xc+yc
gridded(df)=TRUEimage(df["z"])# draw labels to verify:cc = coordinates(df)z=df[["z"]]zc=as.character(z)zc[is.na(zc)]="NA"text(cc[,1],cc[,2],zc)