What is a “good” palette for divergent colors in R? (or: can viridis and magma be combined together?)

Question

I am interested in having a \"good\" divergent color pallette. One could obviously use just red, white, and blue:

img <- function(obj, nam) {
  image(1:le         


        

Answer 1

Library RColorBrewer provides beautiful palettes for =<13 colors. For example, palette BrBG shows diverging colors from brown to green.

library(RColorBrewer)
display.brewer.pal(11, "BrBG")

Which can be expanded to a less informative palette by creating palettes to and from a mid-point color.

brbg <- brewer.pal(11, "BrBG")
cols <- c(colorRampPalette(c(brbg[1], brbg[6]))(51), 
    colorRampPalette(c(brbg[6], brbg[11]))(51)[-1])

Analogically, using your choice of viridis and magma palettes, you can try finding a similarity between them. This could be a point, where to join the palettes back to back.

select.col <- function(cols1, cols2){
    x <- col2rgb(cols1)
    y <- col2rgb(cols2)
    sim <- which.min(colSums(abs(x[,ncol(x)] - y)))
    message(paste("Your palette will be", sim, "colors shorter."))
    cols.x <- apply(x, 2, function(temp) rgb(t(temp)/255))
    cols.y <- apply(y[,sim:ncol(y)], 2, function(temp) rgb(t(temp)/255))
    return(c(cols.x,cols.y))
}

img(select.col(rev(viridis(100,0)),magma(100,0)), "")
# Your palette will be 16 colors shorter.