Behind the Scenes—and Science—of the Earth Observatory

Since 1999 NASA's Earth Observatory has produced more than 12,000 data visualizations, images, and maps. This presentation gives an overview of the Earth Observatory and our approach to science communication over the last two decades.

Transcript

1. Joshua Stevens 
 — and Science — 
 of the Earth Observatory A look into the NASA Earth Observatory approach to science communication
2. NASA Earth Observatory earthobservatory.nasa.gov • Founded in 1999 • Daily publication of data-driven, scienti fi c imagery • More than 12,000 stories published The Earth Observatory’s mission is to share with the public the images, stories, and discoveries about the environment, Earth systems, and climate that emerge from NASA research, including its satellite missions, in-the-field research, and models. “ ”
3. earthobservatory.nasa.gov NASA Earth Observatory
4. earthobservatory.nasa.gov NASA Earth Observatory
5. earthobservatory.nasa.gov NASA Earth Observatory
6. NASA Earth Observatory earthobservatory.nasa.gov • Inspire readers: show our home planet as NASA sees it • Satellite imagery, maps, data visualization, and storytelling for science news/results, natural hazards, and the beauty of Earth • Target audience: Science-literate, interested public, with an eye toward those who overlook science • Also: educators, media, etc
7. NASA Earth Observatory earthobservatory.nasa.gov
8. NASA Earth Observatory earthobservatory.nasa.gov
9. NASA Earth Observatory earthobservatory.nasa.gov
10. NASA Earth Observatory earthobservatory.nasa.gov
11. NASA Earth Observatory earthobservatory.nasa.gov • This happens because: • EO annual web traffic regularly in the top-10 for all NASA.gov URLs • We cover all Earth sciences funded or enabled by NASA • Our content is easy to use by others (journals, books, videos, museums, Google Earth, etc) • Constant stream of fresh, new, well-explained visuals
12. NASA Earth Observatory earthobservatory.nasa.gov • Who we are: small, diverse team with many decades experience in • Earth science / geography / geology / climate • Journalism / communication / writing • Library and information science • GIS / cartography / data analysis • Web development / programming
13. How we do so much with so little? People Background Experience Connections Ideas Processes Structured meetings Story identi fi cation Science-driven design Editing/review Programs Work fl
14. NASA Earth Observatory earthobservatory.nasa.gov • Story selection • Natural events, disasters, processes • New publications, data products • Story tips from the community… that means you! • BYOD: Bring your own data! • We will collaborate with you to share your science - no cost!
15. NASA Earth Observatory Notes from the Field Contact Kathryn Hansen: kathryn.h.hansen@nasa.gov
16. Design Matters: the Science of Seeing
17. • What does it mean for a thing to be “designed?” “Well-designed objects are easy to interpret and understand. Th ey contain visible clues to their operation. Poorly designed objects can be di ffi cult or frustrating to use. Th ey provide no clues—or sometimes false clues. Th ey trap the user and thwart the normal process of interpretation and understanding.” — Don Norman, The Design of Everyday Things Design Matters: the Science of Seeing
18. Design Matters: the Science of Seeing
19. Design Matters: the Science of Seeing
20. Design Matters: the Science of Seeing
21. http://www.directedplay.com/ fi
22. • We begin to acquire info from—and make judgements about— graphics before even thinking about it (pre-attentive processing) • Color choices can aid or inhibit this process, and reinforce thoughtful, accurate interpretation later • Is this area hot or cold? • Is this relationship good or bad? • Can I trust this image? • Readers first see “what” then later understand “how much” and “why” Design Matters: the Science of Seeing
23. https://earthobservatory.nasa.gov/blogs/elegant fi
24. https://earthobservatory.nasa.gov/blogs/elegant fi
25. https://earthobservatory.nasa.gov/blogs/elegant fi
26. https://earthobservatory.nasa.gov/blogs/elegant
27. https://sci.utah.edu/~vdl/papers/2019_eurovis_implicit-discretization.pdf Design Matters: the Science of Seeing
28. https://earthobservatory.nasa.gov/blogs/elegant fi
29. Our eyes and brains are easily fooled ©1995 Edward H. Adelson Design Matters: the Science of Seeing
30. Our eyes and brains are easily fooled ©1995 Edward H. Adelson Design Matters: the Science of Seeing
31. Design Matters: the Science of Seeing Crameri, F., Shephard, G. E., & Heron, P. J. (2020). The misuse of colour in science communication. Nature communications, 11(1), 1-10.
32. Crameri, F., Shephard, G. E., & Heron, P. J. (2020). The misuse of colour in science communication. Nature communications, 11(1), 1-10. Design Matters: the Science of Seeing
33. https://sci.utah.edu/~vdl/papers/2019_eurovis_implicit-discretization.pdf Luminosity (lightness) curves. Color is an axis. Steps between values must be linear! Design Matters: the Science of Seeing
34. https://sci.utah.edu/~vdl/papers/2019_eurovis_implicit-discretization.pdf Luminosity (lightness) curves. Design Matters: the Science of Seeing
35. https://www.climate-lab-book.ac.uk/2016/why-rainbow-colour-scales-can-be-misleading/ Design Matters: the Science of Seeing Sharp “front” around 300 DU Same “front” No “front” it’s the palette, not the data!
36. Design Matters: the Science of Seeing https://pubmed.ncbi.nlm.nih.gov/22034369/ Accuracy identifying “Endothelial Shear Stress (ESS)” “It was surprising to fi nd that di ff erent color mapping can render the task less ambivalent. One can only wonder in just how many other instances we make our task more di ffi cult than it needs to be simply by maintaining the status quo.”
37. Camgöz, N., Yener, C., & Güvenç, D. (2004). E ff
38. The human eye has 120 million rods (light sensing) and just 7 million cones (color sensing) Design Matters: the Science of Seeing
39. • Scientific color maps should: • Not distort data or introduce false gradients. Show quantitative data with perceptually linear changes in luminosity, not hue. • Use a small amount of hue to convey an intuitive context (hot vs cold, wet vs dry, etc.) • Be accessible to those with colorblindness • Consider the readers of today, not the computational simplicity of yesterday’s engineers Design Matters: the Science of Seeing
40. Tools and resources ColorBrewer: https://colorbrewer2.org/ Color Palette Helper: https://gka.github.io/palettes/ Design Matters: the Science of Seeing
41. Lava fl ows measured by aerial lidar Design Matters: the Science of Seeing
42. Lava fl ows measured by aerial lidar Mount St. Helens Design Matters: the Science of Seeing
43. Southeast light source Northwest light source Design Matters: the Science of Seeing
44. Tom Patterson (2014) Design Matters: the Science of Seeing
45. © 1988 National Geographic Design Matters: the Science of Seeing
46. Guillaume-Henri Dufour (1854) Design Matters: the Science of Seeing
47. Leonardo da Vinci (1503) Design Matters: the Science of Seeing
48. Design Matters: the Science of Seeing Light source
49. Imhof, E. (1982). Cartographic Relief Presentation. (H. J. Steward, Ed.). Walter de Gruyter & Co. Design Matters: the Science of Seeing Untenable theory - not supported by any research!
50. Design Matters: the Science of Seeing Biland, Julien & Coltekin, Arzu. (2016). An empirical assessment of the impact of the light direction on the relief inversion e ff
51. • To sum it up: • Communication must consider the audience—and how their brains work • General public does not know what you know, does not read same journals, is not used to the same software defaults, or the same unchallenged conventions of your field • Prioritize audience psychology over technology or irrelevant physics • If we want people to accept unfamiliar science, we need to be willing to do the same and adjust our approach accordingly Design Matters: the Science of Seeing
52. Thank you! Questions or comments? 
 joshua.e.stevens@nasa.gov EarthObservatory.nasa.gov 
 @jscarto JoshuaStevens.net