I am looking for more resources on color theory in cartography. When to de-saturate a color? Or if you start with one color, how do you create a ramp from it? RGB or HSV? I can hack at it and come up with something, but I'm wondering if there's a methodology to it.

I say this because my work requires I constantly come up with new color ramps, and we are limited in certain colors that we cannot use because of said colors' associations (red is only used for hotspots, green is for positive topics only, etc). I constantly riff off of Color Brewer ramps, defaults, and make my own. I'm wondering if there are additional resources available to help me as a cartographer who primarily does thematic population maps, to come up with new ramps. I know there is a lot of information out there for designers, but I am looking for information more specific to cartography.

  • if people are just going to post color palette sites from twitter and such, since i'm a lady who uses pinterest, i guess this is something too: pinterest.com/color_palettes May 11, 2016 at 14:15

5 Answers 5


I strongly recommend Colorpicker for data: http://tristen.ca/hcl-picker/#/hlc/6/1/15534C/E2E062

It's designed with color ramps in mind and allows for lots of flexibility while maintaining a clearly understandable ramp.

  • Awesome, this is helpful especially because it shows you a preview on a map. I will definitely be saving this link. I wish it gave out rgb's though. May 19, 2016 at 12:50

Here's a couple places to start...




Good question. ColorBrewer is definitely my first go-to, but it's getting old seeing the exact same palettes in every map ever created in the post-colorbrewer world.

I have been playing around with cubehelix color palettes lately with census data for bivariate choropleths. I just sample the palettes with illustrator, then tweak them (usually with Illustrator color guides) for optimum perception between classes.

This cubehelix palette worked really well with two skewed-right distributions:

enter image description here

  • thank you! i actually introduced my work to some bivariate mapping applications. and yeah i agree about the skew, the datasets really do have to have just the right amount of correlation in the right spots for bivariate to serve its purpose. May 11, 2016 at 12:46

ColorHunter is an interesting resource which identifies colours from found images on the internet. Not really of much use for cartographic/infographic colour schemes, but a useful graphic design resource and good for coming up with ideas.

Also worth bearing in mind colour blindness (deuteranopia, protanopia). You mention arcGIS, but QGIS has preview modes for colour blindness, low res (monochrome dithered) and greyscale, which help assess colour palettes for mono printouts and accessibility. The GIMP and ColorOracle are also tools to do this.

QGIS also gives access to cpt-city set of colour ramps, which cover many cartographic uses (and sometimes specific data sets like ETOPO), as well as ColorBrewer palettes.

Also stumbled across this today, Viridis, which has 4 colour ramps designed to be color-blind safe and perceptually even (so, not 'washing out' at higher values)

  • 1
    Similar: matplotlib.org/cmocean May 11, 2016 at 3:08
  • thanks Richard, those were interesting color associations from matplotlib. i work in public health, not physical geog so much, but they have given me some ideas. May 11, 2016 at 14:54

I've always liked Adobe Color (formerly Kuler). It's pretty powerful in creating colour schemes. You can set rules like Complementary, Monochromatic, Compound, as well as making your own. There are all sorts of sliders and things to customise your colours. You can also swap between RGB, CMYK, LAB, and HSB modes.

I also like the ability to explore other peoples' themes.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.