Ramona L. Kessel

Web-based sonification of space science data

is the use of non-speech audio to convey information 1 (with sonar and Geiger counters, for example). The best example in space physics was the use of sound for detecting micro-meteoroids that impaced Voyager 2 as it traversed Saturns rings; these impacts were obscured in the plotted data but were clearly evident as hailstorm sounds. 2 We are extending the Coordinated Data Analysis Web (CDAWeb) data browsing system (cdaweb.gsfc.nasa.gov/) to display space science data as sounds in addition to its existing plot capability (line plots, spectrograms, images, etc.). In CDAWeb, the user selects spacecraft, instruments, datasets, and time periods of interest and is then presented with a list of variables to display. PERL and Interactive Data Language (IDL) routines on the CDAWeb server interactively create the selection pages and then read the data from Common Data Format (CDF) files (used for its portability and self-describing meta-information), plot the data (automatically selecting reasonable plot types and scaling), and return the plots as GIF images. We are extending CDAWeb to provide an alternative display type for certain variables where an IDL routine writes the data out in a JavaScript routine that calls the Beatnik (www.headspace.com/) plug-in to the users Web browser. When the user clicks on the sound data icon, the Beatnik plug-in plays the data array as varying pitches, loudnesses, or drumbeat rhythms. Sonification Background Sonification is useful for monitoring some data while looking at something else, for complex or rapidly/temporally changing visualizations, for data exploration of large datasets (particularly multi-dimensional datasets), and for exploring datasets in frequency rather than spatial dimensions. A good reference to research on the use of audio representation of data is the International Conferences on Auditory Display Complex datasets (e.g., particle measurements varying in energy, look direction, time, and particle species such as electrons and ions) are usually only examined in a subset of dimensions at a time, forcing researchers to build up a picture in their minds of the whole dataset. Sound can be used to represent these other dimensions, using pitch, loudness, rhythm, damping or attack/decay rate, direction, duration and repetition, timbre and harmonics, phase, and rest periods. 4 Audio can also be used to reinforce the visual displays as alternative ways of looking at or hearing the data, particularly where visual patterns are hidden until identified by other means. An additional benefit of sound for visualization is that the ear is more sensitive …