Good Practice

With hundreds of colorschemes available, selecting the right palette is critical—especially when communicating data accurately in scientific, medical, or analytical applications.

Perceptual Uniformity

A perceptually uniform colorscheme is arranged so that equal steps in data values correspond to equal perceived differences in color.

Research in visual perception demonstrates that human vision perceives changes in lightness (luminance) as data differences far more accurately than changes in hue. Consequently, sequential colorschemes with monotonically increasing lightness curves provide the most accurate spatial and quantitative data interpretation.

The CIELAB / Lab Color Space

The Lab color space represents color using three axes:

• L*: Lightness (0 = black, 100 = white)
• a*: Green to Red axis
• b*: Blue to Yellow axis

Plotting lightness $L^*$ across the range $0.0, 1.0$ of a scheme indicates how uniformly steps in data will be perceived.

import { colorschemes, get } from 'color-schemes-js';

// Evaluate CIELAB Lightness L* across a colormap for perceptual uniformity
const scheme = colorschemes.viridis;

function calculateLightness(t: number): number {
  const c = get(scheme, t);
  
  // Convert sRGB to Linear RGB
  const R = c.r <= 0.04045 ? c.r / 12.92 : Math.pow((c.r + 0.055) / 1.055, 2.4);
  const G = c.g <= 0.04045 ? c.g / 12.92 : Math.pow((c.g + 0.055) / 1.055, 2.4);
  const B = c.b <= 0.04045 ? c.b / 12.92 : Math.pow((c.b + 0.055) / 1.055, 2.4);
  
  // CIE Y Luminance
  const Y = 0.2126 * R + 0.7152 * G + 0.0722 * B;
  
  // CIELAB L* Lightness [0.0 - 100.0]
  return Y > 0.008856 ? Math.cbrt(Y) * 116 - 16 : Y * 903.3;
}

Recommended Palettes

Sequential Schemes

Recommended perceptually uniform sequential colormaps with monotonic lightness:

Diverging Schemes

For data centered around a meaningful midpoint (such as zero anomaly or 50% baseline), diverging schemes should exhibit symmetric lightness curves from the center outward to equivalent extremes:

Less Suitable for Quantitative Data

Colormaps with non-monotonic or abrupt changes in lightness (such as classic rainbow maps like jet or decorative palettes like gnuplot or Hiroshige) introduce perceptual artifacts—false gradients, hidden details, or artificially highlighted features—making them unsuitable for quantitative visual analysis:

Academic References

1. Peter Kovesi: Good Colour Maps: How to Design Them. arXiv:1509.03700 cs.GR. DOI:10.48550/arXiv.1509.03700
2. Liang Zhou & Charles D. Hansen: A Survey of Colormaps in Visualization. IEEE TVCG. PubMed 26513793