rLakeHabitat (version 1.0.0)

crossValidate: Cross Validate Interpolated Bathymetry

Description

Obtain residual mean square error (RMSE) from K-fold cross validation of bathymetry interpolation.

Usage

crossValidate(
  outline,
  df,
  x,
  y,
  z,
  zeros = FALSE,
  separation = NULL,
  k = 5,
  crsUnits = "dd",
  res = 5,
  method = "IDW",
  fact = NULL,
  nmax = 20,
  idp = 2,
  model = "Sph",
  psill = NULL,
  range = NULL,
  nugget = 0,
  kappa = NULL
)

Value

mean RMSE value across k number of folds

Arguments

outline

shapefile outline of a waterbody

df

dataframe of coordinates and depths for a given waterbody

x

character giving name of longitude column

y

character giving name of latitude column

z

character giving name of depth column

zeros

logical describing if bounding zeros are needed (FALSE) or provided (TRUE), default = FALSE

separation

number describing distance between points, units from CRS

k

numeric value describing the number of folds to test, default = 5

crsUnits

character describing CRS units of input outline, either "dd" (decimal degrees) or "m" (meters), default = "dd"

res

number describing desired cell resolution in meters, default = 5

method

character describing method of interpolation, options include Inverse Distance Weighted ("IDW") or Ordinary Kriging ("OK"). Default = "IDW"

fact

numeric value describing the factor by which raster resolution should be increased, default = NULL. If 'crsUnits' and 'res' are defined, fact = NULL

nmax

numeric value describing number of neighbors used in interpolation, default = 20

idp

numeric value describing inverse distance power value for IDW interpolation

model

character describing type of model used in Ordinary Kriging, options include 'Sph', 'Exp', 'Gau', 'Sta', default = 'Sph'

psill

numeric value describing the partial sill value for OK interpolation, default = NULL

range

numeric describing distance beyond which there is no spatial correlation in Ordinary Kriging models, default = NULL

nugget

numeric describing variance at zero distance in Ordinary Kriging models, default = 0

kappa

numeric value describing model smoothness, default = NULL

Author

Tristan Blechinger, Department of Zoology & Physiology, University of Wyoming

Details

If both 'crsUnit' and 'res' = NULL, the output raster will be in the same CRS and units as the input 'outline' and the resolution will be increased by 'fact' (default = 10). If both 'crsUnit' and 'res' are defined, fact = NULL and the output raster will be projected to the most appropriate UTM zone at the specified resolution.

For the model argument there are four different methods included here that are supported by gstat::vgm ("Sph", "Exp", "Gau", "Mat"). "Sph" = The default gstat::vgm method. Spherical model characterized by a curve that rises steeply to defined range then flattens, indicates no spatial correlation between points beyond that range. "Exp" = Exponential model characterized by spatial correlation decaying with distance. "Gau" = Gaussian model similar to spatial model but with stronger decay at shorter distances. "Mat" = Matern model Three parameters (psill, range, kappa) are incorporated from a fitted variogram (default = NULL). If specified in function input, chosen values will overwrite variogram values.

Examples

#load example outline
outline <- terra::vect(system.file("extdata", "example_outline.shp", package = 'rLakeHabitat'))
#load example xyz data
data <- read.csv(system.file("extdata", "example_depths.csv", package = 'rLakeHabitat'))
#run function
crossValidate(outline, data, "x", "y", "z", zeros = FALSE, separation = 10, k = 5, crsUnit = "dd",
res = 50, method = "IDW", nmax = 4, idp = 1.5)