Seeing satellite data
Scientists in the USA are increasingly required to give something back to the public who funded their work. Oceanographers need to be able to show their results so that people can understand them; but the way they currently present data is confusing to lay people. We have devised a better way.
The use of satellites has revolutionized the study of the oceans. Only about five per cent of the world’s ocean has been explored by ship, and this work is expensive, time consuming, and very localised. However, satellite-derived data is free to use, collected automatically, and regional in nature. Oceanographic satellites have a suite of instruments sensitive to specific bands of electro-magnetic radiation. These highly-sensitive cameras each take digital pictures at a different wavelength that communicates information about the ocean below. Just like a digital camera, these sensors assign numerical values to each pixel. Scientists and technicians transform this array of numerical values into a colour system for visual interpretation.
Hard to interpret
By convention, most scientists who study these visualisations like to display them in a rainbow spectrum where purple represents the lowest values and red represents the highest (grading through blue, green, yellow, and orange).
While there are advantages to this type of representation, including enhancing detail and building on tradition, there are also some significant drawbacks. The visualisations are difficult for people with colour deficient vision. The middle values (yellow) are over emphasized, and these colours have no inherent meaning.
These images are making their way into the public domain. Although they are visually impressive, we found that most people could not see the data inherent in these beautiful pictures.
We were looking for ways to make these images immediately understandable without oversimplifying them. We decided to experiment with different colour schemes to try to make the data more obvious. We focused on two ocean properties: sea surface temperature and concentration of algae in the surface water as measured by satellite. An infrared sensor (like night vision goggles) is used to determine temperature, and a sensor attuned to sense chlorophyll is used to determine the amount of algae in the surface waters.
To start, we had to identify colour scales that would signify hot or cold water and concentration of algae. We settled on two colour schemes that drew on people’s everyday knowledge and assumptions of colours. For temperature we created a two-colour scale with blue signifying cold water and red signifying hot waters, with a small band of white in the transition. For algae concentration we chose a monochromatic scale of green, ranging from light green signifying low levels to dark green signifying high concentrations.
We took these visualisations, along with the rainbow spectrum ones preferred by the scientists, and showed them to museum visitors. We asked them: ‘What do you see in this image?’ With the rainbow spectrum versions, visitors were unsure of what, or even where they were looking. This was not so with our new versions. Both of our more culturally relevant schemes were immediately understandable. Visitors were even able to use these versions of the images to make scientific claims about the ocean. For example, off the Pacific Coast, seasonal upwelling of nutrient-rich cold water drives fisheries. In the original colour schemes, no visitor talked about upwelling when making sense of the image; in ours, visitors were able to talk about changes in coastal ocean temperatures as well as upwelling.
The scientists preferred their original rainbow spectrum for visualising temperature data, but when it came to the green representation of the algae data we heard murmurs of ‘why didn’t we think of that?’ throughout presentations.
This is a shortened version of a paper wihth first appeared in Public Understanding of Science (May 2010) 19: 311-321