autoKrige {automap}

Performs an automatic interpolation

Description

This function performs automatic kriging on the given dataset. The variogram is generated automatically using autofitVariogram.

Usage

autoKrige(formula, 
  	  input_data, 
	  new_data, 
 	  data_variogram = input_data, 
	  block = 0, 
	  model = c("Sph", "Exp", "Gau", "Ste"), 
	  kappa = c(0.05, seq(0.2, 2, 0.1), 5, 10), 
	  fix.values = c(NA,NA,NA), 
	  remove_duplicates = TRUE, 
	  verbose = FALSE, 
	  GLS.model = NA,
      start_vals = c(NA,NA,NA),
      miscFitOptions = list(),
	  ...)

Arguments

formula

formula that defines the dependent variable as a linear model of independent variables; suppose the dependent variable has name 'z', for ordinary and simple kriging use the formula 'z~1'; for simple kriging also define 'beta' (see below); for universal kriging, suppose 'z' is linearly dependent on 'x' and 'y', use the formula 'z~x+y'.

input_data

An object of the SpatialPointsDataFrame-class or sf containing the data to be interpolated.

new_data

A sp, sf or stars (st_as_stars) object containing the prediction locations. new_data can be a points set, a grid or a polygon. Must not contain NA's. If this object is not provided a default is calculated. This is done by taking the convex hull of input_data and placing around 5000 gridcells in that convex hull.

data_variogram

An optional way to provide a different dataset for the building of the variogram then for the spatial interpolation.

block

Use this parameter to pass on a specification for the block size. e.g. c(1000,1000)

model

List of models that will be tested during automatic variogram fitting.

kappa

List of values for the smoothing parameter of the Matern model that will be tested during automatic variogram fitting.

fix.values

Can be used to fix a variogram parameter to a certain value. It consists of a list with a length of three. The items describe the fixed value for the nugget, range and sill respectively. Setting the value to NA means that the value is not fixed. Is passed on to autofitVariogram.

remove_duplicates

logical, remove duplicate points from the input_data. This can take some time on large datasets.

verbose

logical, if TRUE autoKrige will give extra information on the fitting process

GLS.model

If a variogram model is passed on through this parameter a Generalized Least Squares sample variogram is calculated.

start_vals

Can be used to give the starting values for the variogram fitting. The items describe the fixed value for the nugget, range and sill respectively. They need to be given in that order. Setting the value to NA means that the value will be automatically chosen.

miscFitOptions

Additional options to set the behavior of autofitVariogram. For details see the documentation of autofitVariogram.

...

arguments that are passed on to the gstat function krige.

Details

autoKrige calls the function autofitVariogram that fits a variogram model to the given dataset. This variogram model and the data are used to make predictions on the locations in new_data. The only compulsory argument is input_data. So the most simple call would of the form:

autoKrige(meuse)

autoKrige now assumes that you want to perform ordinary kriging on the first column of input_data.

autoKrige performs some checks on the coordinate systems of input_data and new_data. If one of both is NA, it is assigned the projection of the other. If they have different projections, an error is raised. If one of both has a non-projected system (i.e. latitude-longitude), an error is raised. This error is raised because 'gstat does use spherical distances when data are in geographical coordinates, however the usual variogram models are typically not non-negative definite on the sphere, and no appropriate models are available' (Edzer Pebesma on r-sig-geo).

When the user specifies the power model (Pow) as the model, the initial range is set to one. Note that when using the power model, the initial range is the initial power.

Value

This function returns an autoKrige object containing the results of the interpolation (prediction, variance and standard deviation), the sample variogram, the variogram model that was fitted by autofitVariogram and the sums of squares between the sample variogram and the fitted variogram model. The attribute names are krige_output, exp_var, var_model and sserr respectively.

Author(s)

Paul Hiemstra, [e-mail]

See Also

autofitVariogram, krige

Examples

# Data preparation

library(sp)
library(sf)
library(stars)
data(meuse)
coordinates(meuse) =~ x+y
data(meuse.grid)
gridded(meuse.grid) =~ x+y


# Ordinary kriging, no new_data object
kriging_result = autoKrige(zinc~1, meuse)
plot(kriging_result)

# Ordinary kriging
kriging_result = autoKrige(zinc~1, meuse, meuse.grid)
plot(kriging_result)

# Fixing the nugget to 0.2
kriging_result = autoKrige(zinc~1, meuse, 
	meuse.grid, fix.values = c(0.2,NA,NA))
plot(kriging_result)

# Universal kriging
kriging_result = autoKrige(zinc~soil+ffreq+dist, meuse, meuse.grid)
plot(kriging_result)

# Block kriging  
kriging_result_block = autoKrige(zinc~soil+ffreq+dist, 
	meuse, meuse.grid, block = c(400,400))
plot(kriging_result_block)

# Dealing with duplicate observations
data(meuse)
meuse.dup = rbind(meuse, meuse[1,]) # Create duplicate
coordinates(meuse.dup) = ~x+y
kr = autoKrige(zinc~dist, meuse.dup, meuse.grid)

# Extracting parts from the autoKrige object
prediction_spdf = kr$krige_output
sample_variogram = kr$exp_var
variogram_model = kr$var_model

coordinates(meuse) = ~x + y
meuse = st_as_sf(meuse)
meuse.grid = st_as_stars(meuse.grid)
kriging_result = autoKrige(zinc~1, meuse, 
	meuse.grid, fix.values = c(0.2,NA,NA))