Colormaps

dartwork-mpl ships 16 curated colormaps — all designed or refined in the perceptually uniform OKLCH color space. Each category offers 3 color-preference options so you can match the tone of your visualization.

Design philosophy

Most colormap libraries pick colors that “look nice” in sRGB and then interpolate linearly between them. This creates two hidden problems:

  1. Perceptual non-uniformity — equal numeric steps produce unequal visual steps (grey bands, shimmering artefacts).

  2. Lightness reversals — parts of the ramp can become darker then lighter, destroying print/greyscale readability.

dartwork-mpl solves both by designing entirely in OKLCH space, the most modern perceptually uniform color model (an improvement over CIE-LAB):

Principle

How we enforce it

Lightness monotonicity

Every sequential map has strictly monotonic OKLCH L. Greyscale readability guaranteed.

Smooth hue path

Anchor colors placed along natural hue arcs (no muddy brown shortcuts).

Bow-shaped chroma

Chroma peaks mid-ramp and tapers at boundaries, avoiding gamut clipping.

Color-blind awareness

All maps verified under deuteranopia, protanopia, and tritanopia simulations.

Why not just use viridis?   viridis is excellent but it’s only one map. For publication figures you need a family: sequential ramps for bars, diverging scales for anomalies, categorical sets for labels, and cyclical maps for phases — all sharing the same perceptual integrity. Standard Matplotlib maps (viridis, Blues, tab10, …) continue to work without any prefix.

Quick start

import dartwork_mpl as dm
import matplotlib.pyplot as plt
import numpy as np

dm.style.use("scientific")
data = np.random.randn(50, 50).cumsum(axis=0)
im = plt.imshow(data, cmap="dc.sunset", vmin=-8, vmax=8)
cb = plt.colorbar(im, extend="both", shrink=0.9, pad=0.02)
cb.set_label("normalized signal")
cb.outline.set_visible(False)
plt.show()

  • Add _r to reverse any map: dc.sunset_r

Colormap catalog

Explore all 16 built-in colormaps dynamically using the catalog below. Use the tabs to browse by data type.

Single-Hue
Color
Mono
One hue that ramps value cleanly. Great for magnitude and density.
dc.amethyst
dc.coral
dc.cyberpunk
dc.emerald
dc.graphite
dc.obsidian
dc.ruby
dc.sapphire
dc.synthwave
dc.topaz
dc.vivid_dusk

Creating custom colormaps

If the built-ins don’t fit your data, build a colormap by interpolating in OKLCH space with dm.cspace(). The lightness ramp stays strictly monotonic, so you skip the muddy midtones that appear when RGB hex codes are blended directly. Both panels below use the same 2D Gaussian so you can read the ramps doing their job side-by-side.

Sequential
Sequential cspace colormap, indigo to amber, on a smooth 2D Gaussian 
import dartwork_mpl as dm
import matplotlib as mpl

colors = dm.cspace("#1a237e", "#ff6f00", n=256, space="oklch")
cmap = mpl.colors.ListedColormap([c.to_rgb() for c in colors])
Diverging
Diverging cspace colormap, indigo to white to crimson, on the same Gaussian 
left = dm.cspace("#1a237e", "#ffffff", n=128, space="oklch")
right = dm.cspace("#ffffff", "#c62828", n=128, space="oklch")
cmap = mpl.colors.ListedColormap(
    [c.to_rgb() for c in (left[:-1] + right)]
)

Color-blind safety

All dc.* maps are verified under three CVD simulations — deuteranopia, protanopia, and tritanopia.

  • Every sequential map has strictly monotonic lightness, preserving data ordering even when color perception is reduced.

  • Single-hue and achromatic maps (monochrome, steel) are inherently CVD-safe.

Recommendation: For highest accessibility, choose maps whose primary contrast channel is lightness rather than hue.

Rendering tips

  • Set vmin / vmax yourself for stable colorbars across facets or animations.

  • Reverse any map with the _r suffix (dc.flame_r).

  • Hide colorbar outlines: cb.outline.set_visible(False).

  • For diverging data, use symmetric limits and extend="both".

See also