Skip to contents

Draws the empirical variogram that estimate_sac_range() attaches to its result, as a ggplot2 object, with the fitted model and the effective range overlaid where a range was identified, and a subtitle that gives the reason where it was not. A variogram that never reaches a sill, or that has no spatial structure at the lags resolved, is drawn as well, since the curve shows why the range came back NA.

Usage

# S3 method for class 'sac_range'
plot(x, ...)

Arguments

x

A sac_range object, as returned by estimate_sac_range().

...

Ignored.

Value

A ggplot object.

Details

Nothing is recomputed: the plot reads the variogram, variogram_model, crs and rejected_reason attributes the estimate carries. The variogram is the one over all point pairs. The distance axis is in the units of the CRS the variogram was fitted in, which estimate_sac_range() chooses itself for lon/lat input.

See also

estimate_sac_range(), which makes the object, and plot.spatial_fit() with type = "variogram", which draws the same picture for a fitted model's residuals.

Other spatial autocorrelation: estimate_sac_range(), sac_nugget()

Examples

library(sf)
# An exponential field with range parameter 150 (effective range about
# 450 m) and a nugget of half the sill, so the fitted model, the nugget and
# the range line all have something to show.
set.seed(3)
n <- 250
xy <- data.frame(x = 5e5 + runif(n, 0, 1000), y = 5e6 + runif(n, 0, 1000))
D  <- as.matrix(dist(xy))
xy$z <- as.numeric(t(chol(exp(-D / 150) + diag(0.5, n))) %*% rnorm(n))
pts <- st_as_sf(xy, coords = c("x", "y"), crs = 32632)
r <- estimate_sac_range(pts, response_var = "z")
plot(r)


# A field whose variogram never reaches a sill: the plot still draws it,
# and the subtitle says why no range is marked.
xy$trend <- sin(xy$x / 400) + rnorm(n, sd = 0.2)
r2 <- estimate_sac_range(st_as_sf(xy, coords = c("x", "y"), crs = 32632),
                         response_var = "trend")
plot(r2)