snapKrig (version 0.0.2)

sk_plot_pars: Plot the covariance structure of a snapKrig model

Description

Visualization of the footprint of a covariance kernel as a heatmap of approximate size dim(g), where each grid cell is colored according to its covariance with the central grid point.

Usage

sk_plot_pars(pars, g = NULL, simple = FALSE, ...)

Value

the same as sk_plot

Arguments

pars

list of the form returned by sk_pars with entries 'y', 'x', ('eps', 'psill')

g

any object understood by sk

simple

logical, if FALSE the function adds some annotation

...

additional arguments passed to sk_plot

Details

If g is not supplied, the function sets a default with dimensions 100 x 100. A default resolution is computed such that the maximum nugget-free covariance along the outer edge of the plot is 5% of pars$psill.

When simple=FALSE (the default), covariance parameters are printed in the title and axis labels with values rounded to 3 decimal places. This can be customized by passing arguments 'main', 'ylab', 'xlab' (and any others accepted sk_plot apart from gdim).

See Also

sk sk_pars

Other plotting functions: sk_plot()

Examples

gdim = c(100, 100)
g = sk(gdim)
pars = sk_pars(g, 'mat')

# plot with default grid
sk_plot_pars(pars)

# plot with a predefined grid
sk_plot_pars(pars, g)

# zoom in/out by passing a grid object with suitably modified resolution
gres = g[['gres']]
sk_plot_pars(pars, sk(gdim=gdim, gres=0.5*gres))
sk_plot_pars(pars, sk(gdim=gdim, gres=2*gres))

# change plot style settings (all parameters of sk_plot accepted)
sk_plot_pars(pars, simple=TRUE)
sk_plot_pars(pars, minimal=TRUE)