A Toolbox for Manipulating and Assessing Color Palettes for Statistical Graphics

Achim Zeileis, Jason C. Fisher, Kurt Hornik, Ross Ihaka, Claire D. McWhite, Paul Murrell, Reto Stauffer, Claus O. Wilke http://colorspace.R-Forge.R-project.org/ Color spaces

Origin of the package: Convert colors between various three-dimensional representations of color. In particular: From the perceptually-based HCL (Hue-Chroma-Luminance) to standard Red-Green-Blue (sRGB, and corresponding hex codes) space.

polarLUV LUV HSV (= HCL)

white point gamma = D65 (= 2.4) XYZ RGB sRGB hex

polarLAB LAB HLS

1/18 RGB: Motivated by how comput- ers/TVs used to generate and still rep- resent color.

Red

Green

Blue

HCL vs. RGB

HCL: Polar coordinates in CIELUV. Captures perceptual dimensions of the human visual system very well.

Hue (Type of color)

Chroma (Colorfulness)

Luminance (Brightness)

2/18 HCL vs. RGB

HCL: Polar coordinates in CIELUV. RGB: Motivated by how comput- Captures perceptual dimensions of ers/TVs used to generate and still rep- the human visual system very well. resent color.

Hue Red (Type of color)

Chroma Green (Colorfulness)

Luminance Blue (Brightness)

2/18 RGB Spectrum 1.0 Red Green Blue 0.8 0.6 0 0.4 Red / Green Blue 0.2 0.0

Index 150 100 360 0 Hue 0 50 Chroma / Luminance Hue Chroma Luminance 0

HCL Spectrum −360

HCL vs. RGB: The End of the Rainbow

RGB Spectrum 1.0 Red Green Blue 0.8 0.6 0 0.4 Red / Green Blue 0.2 0.0

100 Index 360 80 270 60 0 180 Hue 40 90 Chroma / Luminance 20 Hue Chroma Luminance 0 0 HCL Spectrum

3/18 HCL vs. RGB: The End of the Rainbow

RGB Spectrum RGB Spectrum 1.0 1.0 Red Green Blue Red Green Blue 0.8 0.8 0.6 0.6 0 0 0.4 0.4 Red / Green Blue Red / Green Blue 0.2 0.2 0.0 0.0

100 Index 360 Index 150 80 270 60 100 360 0 0 180 Hue Hue 40 0 50 90 Chroma / Luminance Chroma / Luminance 20 Hue Chroma Luminance Hue Chroma Luminance 0 0 0

HCL Spectrum HCL Spectrum −360

3/18 Qualitative: For categorical information, i.e., where no particular ordering of categories is available. Function: qualitative_hcl().

Sequential: For ordered/numeric information, i.e., where colors go from high to low (or vice versa). Function: sequential_hcl().

Diverging: For ordered/numeric information around a central neutral value, i.e., where colors diverge from neutral to two extremes. Function: diverging_hcl().

Color palettes: Somewhere over the Rainbow

Qualitative (Set 2) Sequential (Blues 3) Diverging (Green−Brown)

Color Color Color

Desaturated Desaturated Desaturated

4/18 Sequential: For ordered/numeric information, i.e., where colors go from high to low (or vice versa). Function: sequential_hcl().

Diverging: For ordered/numeric information around a central neutral value, i.e., where colors diverge from neutral to two extremes. Function: diverging_hcl().

Color palettes: Somewhere over the Rainbow

Qualitative (Set 2) Sequential (Blues 3) Diverging (Green−Brown)

Color Color Color

Desaturated Desaturated Desaturated

Qualitative: For categorical information, i.e., where no particular ordering of categories is available. Function: qualitative_hcl().

4/18 Diverging: For ordered/numeric information around a central neutral value, i.e., where colors diverge from neutral to two extremes. Function: diverging_hcl().

Color palettes: Somewhere over the Rainbow

Qualitative (Set 2) Sequential (Blues 3) Diverging (Green−Brown)

Color Color Color

Desaturated Desaturated Desaturated

Qualitative: For categorical information, i.e., where no particular ordering of categories is available. Function: qualitative_hcl().

Sequential: For ordered/numeric information, i.e., where colors go from high to low (or vice versa). Function: sequential_hcl().

4/18 Color palettes: Somewhere over the Rainbow

Qualitative (Set 2) Sequential (Blues 3) Diverging (Green−Brown)

Color Color Color

Desaturated Desaturated Desaturated

Qualitative: For categorical information, i.e., where no particular ordering of categories is available. Function: qualitative_hcl().

Sequential: For ordered/numeric information, i.e., where colors go from high to low (or vice versa). Function: sequential_hcl().

Diverging: For ordered/numeric information around a central neutral value, i.e., where colors diverge from neutral to two extremes. Function: diverging_hcl().

4/18 Color palettes: Somewhere over the Rainbow

Sequential: Luminance contrast is crucial (dark to light or vice versa).

Blues 2

Blues 3

Blues

5/18 Color palettes: Somewhere over the Rainbow

Blues 2: Single hue. Decreasing chroma with increasing luminance.

Blues 2 100 360

Blues 3 80 270

Blues 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum 5/18 Color palettes: Somewhere over the Rainbow

Blues 3: Single hue. Triangular chroma to achieve higher luminance contrast.

Blues 2 100 360

Blues 3 80 270

Blues 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum 5/18 Color palettes: Somewhere over the Rainbow

Blues: Multi hue. Triangular chroma. High luminance contrast.

Blues 2 100 360

Blues 3 80 270

Blues 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum 5/18 100 360 80 270 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum

Color palettes: Somewhere over the Rainbow

Diverging: Combine two sequential palettes with balanced chroma/luminance. 100 360 80 270 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum 6/18 Color palettes: Somewhere over the Rainbow

Diverging: Combine two sequential palettes with balanced chroma/luminance. 100 360 100 360 80 80 270 270 60 60 0 0 180 180 Hue Hue 40 40 Chroma / Luminance Chroma / Luminance 90 90 20 20

Hue Chroma Luminance Hue Chroma Luminance 0 0 0 0 HCL Spectrum HCL Spectrum 6/18 Color palettes: Somewhere over the Rainbow

R> hcl_palettes(plot = TRUE)

Qualitative Greens 3 BluGrn YlOrBr Blue−Red 2

Pastel 1 Oslo Teal OrRd Blue−Red 3

Dark 2 Sequential (multi−hue) TealGrn Oranges Red−Green

Dark 3 Purple−Blue Emrld YlGn Purple−Green

Set 2 Red−Purple BluYl YlGnBu Purple−Brown

Set 3 Red−Blue ag_GrnYl Reds Green−Brown

Warm Purple−Orange Peach RdPu Blue−Yellow 2

Cold Purple−Yellow PinkYl PuRd Blue−Yellow 3

Harmonic Blue−Yellow Burg Purples Green−Orange

Dynamic Green−Yellow BurgYl PuBuGn Cyan−Magenta

Sequential (single−hue) Red−Yellow RedOr PuBu Tropic

Grays Heat OrYel Greens Broc

Light Grays Heat 2 Purp BuGn Cork

Blues 2 Terrain PurpOr GnBu Vik

Blues 3 Terrain 2 Sunset BuPu Berlin

Purples 2 Viridis Magenta Blues Lisbon

Purples 3 Plasma SunsetDark Lajolla Tofino

Reds 2 Inferno ag_Sunset Turku

Reds 3 Dark Mint BrwnYl Diverging

Greens 2 Mint YlOrRd Blue−Red 7/18 Statistical graphics

Base: • HCL palette functions return hex color vector. • Typically passed to col = argument of base plotting functions. ggplot2: • Scales of type scale___(). • is fill or color/colour. • is discrete or continuous. • is qualitative, sequential, diverging, or divergingx.

8/18 Statistical graphics: Base

R> q4 <- qualitative_hcl(4, palette = "Dark 3") R> plot(log(EuStockMarkets), plot.type = "single", col = q4, lwd = 2) R> legend("topleft", colnames(EuStockMarkets), col = q4, lwd = 3, bty = "n")

DAX 9.0 SMI CAC FTSE 8.5 8.0 log(EuStockMarkets) 7.5

1992 1993 1994 1995 1996 1997 1998

Time 9/18 Statistical graphics: ggplot2

R> library("ggplot2") R> ggplot(iris, aes(x = Sepal.Length, fill = Species)) + geom_density(alpha = 0.6) + + scale_fill_discrete_qualitative(palette = "Dark 3")

1.2

0.8 Species

setosa versicolor density virginica

0.4

0.0

5 6 7 8 Sepal.Length 10/18 Emulation: Color vision deficiency. • deutan(): Deuteranopia (green deficient). • protan(): Protanopia (red deficient). • tritan(): Tritanopia (blue deficient).

Visualization and assessment

Visualizations: Based on vector of colors. • swatchplot(): Color swatches. • specplot(): Spectrum of HCL and/or RGB trajectories. • hclplot(): Trajectories in 2-dimensional HCL space projections. • demoplot(): Illustrations of typical (and simplified) statistical graphics.

11/18 Visualization and assessment

Visualizations: Based on vector of colors. • swatchplot(): Color swatches. • specplot(): Spectrum of HCL and/or RGB trajectories. • hclplot(): Trajectories in 2-dimensional HCL space projections. • demoplot(): Illustrations of typical (and simplified) statistical graphics.

Emulation: Color vision deficiency. • deutan(): Deuteranopia (green deficient). • protan(): Protanopia (red deficient). • tritan(): Tritanopia (blue deficient).

11/18 Visualization and assessment: demoplot()

R> cl <- sequential_hcl(5, palette = "Heat") R> demoplot(cl, type = "...")

map heatmap scatter spine bar

● ● ●● ● ●● ● ● ● ●● ● ●●●●● ● ●●● ● ●●●●●●● ● ● ●●●● ●●●●● ● ● ●● ●● ● ● ●● ● ● ●●● ●●●●●● ● ● ● ●●●● ●●● ●● ●●●●●●●● ● ●●● ● ● ● ●●●● ●●● ●●●● ● ● ● ●● ●● ●●● ●●●●●●●●●●●●●●●● ●●● ●● ●●●●●● ●●●●●● ●●●●●● ● ● ●●●●●● ●●●●●● ●● ● ●●●●●●●●● ●● ●●●● ● ● ●●●●● ● ●● ● ● ● ●●● ● ● ● ●●●● ●● ●● ●● ●●●●●● ● ●● ● ● ● ● ●● ●●●●●●●●●● ● ● ●●● ● ● ●●● ●●● ● ● ● ●●●● ●●●● ● ●●● ●●●●● ●● ● ●●●●●● ● ● ●●● ● ● ● ●●● ●● ●●●●●●●● ● ● ● ●● ●●● ●●● ● ●●●● ● ●●●●●●●●●● ●● ●● ● ●● ●● ●●● ● ● ● ●●● ●● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ●

pie perspective mosaic lines

12/18 Viridis 100 360 80 270 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum

Approximations of other palettes

ColorBrewer.org: YlGnBu 100 360 80 270 60 0 180 Hue 40 Chroma / Luminance 90 20

Hue Chroma Luminance 0 0 HCL Spectrum

13/18 Approximations of other palettes

ColorBrewer.org: YlGnBu Viridis 100 360 100 360 80 80 270 270 60 60 0 0 180 180 Hue Hue 40 40 Chroma / Luminance Chroma / Luminance 90 90 20 20

Hue Chroma Luminance Hue Chroma Luminance 0 0 0 0 HCL Spectrum HCL Spectrum

13/18 Color apps

Facilitate exploration: Graphical user interfaces as shiny apps. • Palette constructor: choose_palette() or hclwizard() (also in tcltk). • Color picker: choose_color() or hcl_color_picker(). • Color vision deficiency emulator: cvd_emulator().

Online versions: http://hclwizard.org/

14/18 Color apps: choose_palette() / hclwizard()

15/18 Color apps: choose_color() / hcl_color_picker()

16/18 Color apps: cvd_emulator()

17/18 References

Ihaka R, Murrell P, Hornik K, Fisher JC, Stauffer R, Wilke CO, McWhite CD, Zeileis A (2019). colorspace: A Toolbox for Manipulating and Assessing Colors and Palettes. http://colorspace.R-Forge.R-project.org/

Zeileis A, Hornik K, Murrell P (2009). “Escaping RGBland: Selecting Colors for Statistical Graphics.” Computational Statistics & Data Analysis, 53, 3259–3270. doi:10.1016/j.csda.2008.11.033.

Stauffer R, Mayr GJ, Dabernig M, Zeileis A (2015). “Somewhere over the Rainbow: How to Make Effective Use of Colors in Meteorological Visualizations.” Bulletin of the American Meteorological Society, 96(2), 203–216. doi:10.1175/BAMS-D-13-00155.1

18/18