ellipsoidfit2: ellipsoidfit2: function to fit an ellipsoid model

View source: R/ellipsoidfit2.R

ellipsoidfit2R Documentation

ellipsoidfit2: function to fit an ellipsoid model

Description

Function to fit an ellipsoid model using the shape matrix (covariance matrix) of the niche variables. This is an optimized version of the ellipsoidfit function; the diffrence is that it does not give the table of distances to niche centroid.

Usage

ellipsoidfit2(
  envlayers,
  centroid,
  covar,
  level = 0.95,
  plot = T,
  size,
  xlab1 = "niche var 1",
  ylab1 = "niche var 2",
  zlab1 = "S",
  ...
)

Arguments

envlayers

A RasterStack or RasterBrick object of the niche variables.

centroid

A vector with the values of the centers of the ellipsoid (see cov_center).

covar

The shape matrix (covariance) of the ellipoid (see cov_center).

level

The proportion of points to be included inside the ellipsoid

plot

Logical If True a plot of the niche will be shown.

size

The size of the points of the niche plot.

xlab1

For x label for 2-dimensional histogram

ylab1

For y label for 2-dimensional histogram

zlab1

For z label for 2-dimensional histogram

...

Arguments passed to plot3d function from rgl

Value

Returns a list containing a data.frame with the suitability values; a suitability raster; a data.frame with the mahalanobis and euclidean distances to the centroid.

Examples

## Load niche data
## Not run: 
d_cardon <-  read.csv(system.file("extdata", "cardon_virtual.csv", package = "ntbox"))
## Compute the centroid and shape (covariance matrix) of the ellipsoid model.
covar_centroid <- cov_center(d_cardon,mve=TRUE,level=0.99,vars=c(3,4,5))
## RasterStack with the niche variables
nicheStack <- raster::stack(list.files(system.file("extdata",
                                       package = "ntbox"),
                                       pattern = ".asc$",
                                       full.names = TRUE))
# Fitting the ellipsoid model
 ellipsoidMod <- ellipsoidfit2(envlayers = nicheStack,
                               centroid = covar_centroid$centroid,
                               covar = covar_centroid$covariance,
                               level=0.99,
                               plot=TRUE,
                               size=3)
 raster::plot(ellipsoidMod)

## End(Not run)

luismurao/ntbox documentation built on Nov. 22, 2024, 4 a.m.