Describe the topic for the talk
Possible title: Choosing beautiful (and accessible) colour maps
Choosing the right colourmap for a visualisation is important for:
- accessibility (so that information can still be communicated to those with colourblindness)
- accurate scientific communication (so that uniform changes in data don't result in non-uniform changes in colour perception)
- representing data correctly (e.g., circular colormaps for longitude, etc)
This talk goes over colourmaps available in Python, and their use with climate data.
Would you be capable/willing to give the talk?
No (cc @MiriamSterl :) ). Let me know what you think of the description above - really just a suggestion, but I'm sure you have your own ideas - is it along the lines you were thinking?
Perhaps we can also include an aside about how to configure theme settings in Matplotlib as an aside (e.g., if users want to use a setting across all their figures, they can configure in one place).
Talk is slated for 3 April 2025.
I've also created some contributing guidelines for the talks so we can make sure they're sharable after the fact.
Additional comments
Jet is most famous for being a fantastically bad colormap. Physicians who use jet in diagnosing heart disease take longer and make significantly more errors than those who use decent colormaps [5]
-Why you should use Viridis and not Jet (rainbow) as a colormap
EDIT: Date correction
Describe the topic for the talk
Possible title: Choosing beautiful (and accessible) colour maps
Choosing the right colourmap for a visualisation is important for:
This talk goes over colourmaps available in Python, and their use with climate data.
Would you be capable/willing to give the talk?
No (cc @MiriamSterl :) ). Let me know what you think of the description above - really just a suggestion, but I'm sure you have your own ideas - is it along the lines you were thinking?
Perhaps we can also include an aside about how to configure theme settings in Matplotlib as an aside (e.g., if users want to use a setting across all their figures, they can configure in one place).
Talk is slated for 3 April 2025.
I've also created some contributing guidelines for the talks so we can make sure they're sharable after the fact.
Additional comments
EDIT: Date correction