Thin plate spline regression

tps(
  .data,
  var,
  resolution = 0.5,
  land_borders = rnaturalearth::ne_countries(scale = "small", returnclass = "sf"),
  check_data = TRUE,
  z_var = NULL,
  z_mode = "independent",
  z_ref = NULL,
  cpus = 1,
  ...
)

Arguments

.data

Data frame with spatial and climate data. The following are expected:

Latitude, named: latitude, lat or y.
Longitude, named: longitude, long, lon or y.
Main variable, named: value of var.

var

String with the name of the climate variable to interpolate.

resolution

Numeric value with the resolution (degrees) to interpolate.

land_borders

Data frame with polygons to represent land borders (e.g. continents, countries, counties, etc.). Default: rnaturalearth::ne_countries.

check_data

Boolean flag to indicate whether .data should be checked or not (i.e. validate coordinates and main variable). Default: TRUE

z_var

String with the name of the variable containing information for elevation. If this is given, then the elevation is used for the interpolation.

z_mode

String with the mode in which the elevation should be used:

z_mode = "independent" (Default), use the elevation as another independent variable to predict var.
z_mode = "covariate", use the elevation as a linear covariate to predict var.

z_ref

Raster object or path to raster object with grid containing elevation information. For example the ETOPO5 (Earth topography 5 arc minute) data set.

cpus

Numeric value with the number of CPUs to use in the computation of the elevations for the interpolation grid.

...

Arguments passed on to fields::Tps

m: A polynomial function of degree (m-1) will be included in the model as the drift (or spatial trend) component. Default is the value such that 2m-d is greater than zero where d is the dimension of x.
p: Polynomial power for Wendland radial basis functions. Default is 2m-d where d is the dimension of x.
scale.type: The independent variables and knots are scaled to the specified scale.type. By default the scale type is "range", whereby the locations are transformed to the interval (0,1) by forming (x-min(x))/range(x) for each x. Scale type of "user" allows specification of an x.center and x.scale by the user. The default for "user" is mean 0 and standard deviation 1. Scale type of "unscaled" does not scale the data.
method: Determines what "smoothing" parameter should be used. The default is to estimate standard GCV Other choices are: GCV.model, GCV.one, RMSE, pure error and REML. The differences are explained in the Krig help file.
GCV: If TRUE the decompositions are done to efficiently evaluate the estimate, GCV function and likelihood at multiple values of lambda.
miles: If TRUE great circle distances are in miles if FALSE distances are in kilometers

Value

tibble object with interpolated values.