Editor’s note: This story is part of a series talking to computer scientists, designers and artists about their data visualization projects. If you have a suggestion for a data visualization to feature, please contact me at firstname.lastname@example.org.
The visual impact of a clear time series animation can impress viewers even when the subject matter is one they have heard about. Such is the case with NASA’s recent presentation of satellite data that visualizes climate scientists’ findings that the thickest parts of the Arctic Sea ice are melting at an accelerated rate.
What it shows: A 42-second animation in which a graph overlay showing the downward fluctuations in the multi-year Arctic Sea ice area from 1980 to 2012 is superimposed over a view of the Earth. While the gray disk at the North Pole remains in view, the white shading representing multi-year ice cover—ice that has lasted at least two summers—shifts in shape, with the total area coverage corresponding to the graph overlay. The area of multi-year ice in the Artic Sea is shrinking by 17.2 percent per decade in this period, according to research by Josefino Comiso, a senior scientist at NASA Goddard Space Flight Center.
Source of the data: Comiso created a time series of multi-year ice using 32 years (from 1980 to 2012) of passive microwave data from two sources: NASA’s Nimbus-7 satellite and the Defense Department’s Defense Meteorological Satellite Program.
What the animator did: In an interview, Cindy Starr, a computer scientist and animator at the NASA/Goddard Scientific Visualization Studio in Greenbelt, Md., said she received the data analysis from Comiso’s team with the biggest challenge solved—reconciling data sets (one set per year of special sensor microwave/imager (SSMI) data) from two different satellites, given that the Nimbu-7 was launched in the late 1970s and yielded relatively low-resolution data compared to the newer satellite.
Because Comiso has been researching climate for 20 decades, and the Visualization Studio has prepared other presentations, Starr said she was able to use some of the techniques her colleagues have used to prepare the Arctic Sea ice graphical overlay. But unlike previous projects, which added data to previously existing time series, this visualization represented a new set of data about multi-year sea ice. She used IDL, or Interactive Data Language, to translate the binary data into a format to be ingested by two visualization tools—Maya, the 3-D animation software package from Autodesk, and RenderMan, the visual special effects rendering software from Pixar. The images of the Earth come from the work of NASA’s Reto Stockli and NASA’s Moderate Resolution Imaging Spectroradiometer, or MODIS.
Key presentation choice: Overlaying the graph over the Earth. “Because the shape changes from year to year, where some parts of the shape bulge out and others recede, it’s hard to look at the data itself” and find meaning, Starr said. “The shape [of the ice] is continually changing, it morphs every year. That is the reason why we started to put the chart over the top. When you see the area values on the chart, it’s very clear to see what is going on.”