For most of us, the sun rises in the morning and sets in the evening, and that is the extent of the change that we notice in its appearance. But our host star is no static, glowing orb, as solar physicists well know—it is a roiling, convecting, chaotic mess.
With a fleet of sun-observing spacecraft, heliophysicists can track every eruption, flare and ripple on the sun's surface. And now the public can glimpse the sun's behavior in depth as well, thanks to a Web interface originally developed for exploring panoramic imagery.
The system, GigaPan Time Machine, was developed by Carnegie Mellon University, with the aid of NASA and Google. The project is an offshoot of GigaPan, a system for creating detailed panoramas by robotically panning a digital camera across a scene and stitching together the resulting images into a billion-pixel, or gigapixel, image. Users can then delve into the high-resolution mosaic to, for instance, examine the facial expressions of individual attendees in the enormous crowd at President Barack Obama's 2009 inauguration. The Time Machine adds time-domain control to the extensive panning and zooming available through Gigapan. In other words, Time Machine imagery is not static but dynamic, a time-lapse movie in which the viewer can zoom in or out as he or she pleases to watch different stories play out on small scales across the frame.
Among the eight new data sets that are available on GigaPan Time Machine this week, only two—imagery of an Arizona grassland from the Southwest Watershed Research Center and of a diseased bee colony under study at the University of Maryland, College Park, and the U.S. Department of Agriculture—are based on the robotic camera panning that GigaPan is known for. The others are simply moving images from a fixed vantage point, which may not comprise a billion pixels but nonetheless have such detail that the GigaPan Time Machine zoom function allows users to extract more information and meaning than would be accessible in an ordinary Web video.
One data set, for instance, is a daylong view of the sun from NASA's Solar Dynamics Observatory compressed into 140 seconds. That satellite keeps constant watch on the sun, photographing it in various wavelengths at a resolution of 16 megapixels. Another is a time-lapse of oceanic chlorophyll concentrations between 2002 and 2011, as monitored by NASA's Aqua satellite. In the Aqua data set, a user can watch seasonal shifts on the global level or home in on the mouth of the Mississippi River to watch the occasional chlorophyll flare-up in the Gulf of Mexico from the runoff of agricultural fertilizer and other contaminants.
Imagery from NASA satellites is already publicly available, but it can be difficult to access and is often presented in discrete chunks—say, a snapshot of the sun at a given moment—rather than in a fluid time-lapse movie. "You're never going to see these time processes that happen," says Randy Sargent, a senior systems scientist at Carnegie Mellon on the GigaPan Time Machine team. "I think that we're going to see a lot of uses for this for satellite information." With decades of Earth-observing satellite data compiled into Time Machine, for instance, climate researchers would be able to trace the evolution, and the advances and retreats, of glaciers around the world.
"The surprise for us was how compelling this way of visualizing, this way of exploring is, and how much of this kind of data was available," Sargent says. "That may be more important than capturing pictures with panning at this point."
Beyond the novelty and eye-candy appeal of panning and zooming through a moving image, the GigaPan team hopes that Time Machine will be used as a learning tool. Annotated tours, called time warps, lead the viewer through one aspect of a data set. For instance, a time warp in the sun imagery highlights the eruption of a solar filament by drawing the viewer to the location of the eruption and automatically zooming in on the action as it unfolds. A running-text sidebar explains "why it looks the way it looks," Sargent says. "Most people will start off looking for things themselves then go back and try one of the tours," he predicts.
Being able to explore imagery in such depth—moving at will through both time and space—may be novel, but it seems perfectly intuitive to Time Machine's users, Sargent says. "This idea of being able to zoom and play doesn't surprise people at all," he says. "They've played with Google Maps and with YouTube enough to know how to do it."