surface.Krig {fields}R Documentation

Plots a surface and contours


Creates different plots of the fitted surface of a Krig object. This is a quick way to look at the fitted function over reasonable default ranges.


surface.Krig(obj, grid.list = NA, extrap = FALSE, graphics.reset =
                 NULL, xlab = NULL, ylab = NULL, main = NULL, zlab =
                 NULL, zlim = NULL, levels = NULL, type = "C", nx =
                 80, ny = 80, ...)


obj A Krig object.
grid.list A list with as many components as variables describing the surface. All components should have a single value except the two that give the grid points for evaluation. If the matrix or data frame has column names, these must appear in the grid list. If grid.list is missing an the surface has just two dimensions the grid is based on the ranges of the observed data.
extrap Extrapolation beyond the range of the data. If false only the convex hull of the observations is plotted. Default is false.
graphics.reset Reset to original graphics parameters after function plotting.
type Type of plot as a character. "p" perspective plot (persp). "c" contour plot (contour). "b" a two panel figure with perspective and contour plots. "I" image plot with legend strip (image.plot). "C" image plot with contours overlaid. Image with contour is the default.
main Title of plot
xlab x axis label
ylab y axis label
zlab z axis label if "p" or "b" type is used.
zlim Z limits passed to persp
levels Contour levels passed to contour.
nx Number of grid points to evaluate surface on the horizontal axis (the x-axis).
ny Number of grid points to evaluate surface on the vertical axis (the y-axis).
... Any other plotting options.


This function is essentially a combination of predict.surface and plot.surface. It may not always give a great rendition but is easy to use for checking the fitted surface. The default of extrap=F is designed to discourage looking at the estimated surface outside the range of the observations.

NOTE: that any Z covariates will b edropped and only the spatial part of the model will be evaluated.

See Also

Krig predict.surface, plot.surface, image.plot


fit<- Krig(ozone$x,ozone$y, theta=30)  # krig fit 

#Image plot of surface with nice, smooth  contours and shading

surface(fit, type="C", nx=128, ny=128) 

[Package fields version 3.3.1 Index]