create.EM: Create grain-size-distributions.

Description

This function allows creating artificial grain-size end-members. One such "artificial end-member loading" may be composed of one or more superimposed normal distributions.

Usage

create.EM(p1, p2, s, boundaries, n)

Value

Numeric vector with normalised end-member loadings, consisting of the mixed normal distributions according to the input parameters.

Arguments

p1: Numeric vector, means of normal distributions, i.e. mode positions.
p2: Numeric vector, standard deviations of normal distributions, i.e. mode width.
s: Numeric vector, relative proportions of each mode, i.e. relative mode height.
boundaries: Numeric vector of length two with class boundaries (i.e. c(lower boundary, upper boundary)).
n: Numeric scalar with number of classes, i.e. resolution of the end-member.

Author

Michael Dietze, Elisabeth Dietze

Details

When building a data set of many artificial end member loadings, these should all have the same boundaries and n. The function builds composites of individual normal distributions. Each distribution is scaled according to s. Finally the distribution is scaled to 100 %.

Examples

Run this code


## set lower and upper class boundary, number of classes and class units
boundaries <- c(0, 11)
n <- 40
phi <- seq(from = boundaries[1], 
           to = boundaries[2], 
           length.out = n)

## create two artificial end-member loadings
EMa.1 <- create.EM(p1 = c(2, 5), p2 = c(1, 0.8), s = c(0.7, 0.3), 
                   boundaries = boundaries, n = n)
EMa.2 <- create.EM(p1 = c(4, 7), p2 = c(1.1, 1.4), s = c(0.5, 0.5),
                   boundaries = boundaries, n = n)

## plot the two artificial end-member loadings
plot(phi, EMa.1, type = "l")
lines(phi, EMa.2, col = "red")

Run the code above in your browser using DataLab

Last chance! 50% off unlimited learning