NASA Logo, National Aeronautics and Space Administration
CDAWeb banner

CDAWeb Audification of scalar data

The option for audification produces audio files in WAVE format at 22kHz of scalar data, for up to 10 audio files if the variable is a vector or multidimensional array. Higher dimensions are averaged into a vector first, which probably is not useful. Fill or missing values are replaced by a first order interpolation of the missing data.

This simple audification works best with high-resolution data and requires at least 44,000 values (2 seconds) to be heard.

The basic design is from Robert Alexander. See below for more information about audification.

Below are some examples of interesting audible events. Try opening these files in the open source program, Audacity , or similar audio software, and adjust the playback speed with the playback speed (it's just a number in the Wave header). In Audacity, this is the slider on top right or select the Effect:Change Speed menu item. Examples below contain the link to the audio file and he associated plot.

Audio: Wind MFI high resolution magnetic field Z GSE vector in the 002.wav file for 2007 Nov 18 19:10 - 21:55 where this region likely contains waves caused by a stream interaction region. Corresponding Plot.

Audio: Wind MFI high resolution magnetic field Z GSE vector in the 002.wav file for 2007 Nov 20 21:00 - 23:30 displays a reverse shock, which occurs when a fast stream of plasma (the super hot, charged gas that fills space) is followed by a slower one, resulting in a shock wave that travels towards the Sun. Corresponding Plot.

Audio: Example of audifying multiple short high-time-resolution bursts from the Van Allen Probes High Frequency Receiver (HFR) Field Waveform Samples RBSP-A_HFR-WAVEFORM_EMFISIS-L2 2nd variable audio for 2013 Sept 11 5:00-7:30. Corresponding Plot

This is only experimental. Try shorter and longer time ranges to move data signatures to be more audible.

More on audification:

Audification is a specific type of data sonification that involves the isomorphic (i.e., one-to-one) mapping of data samples to audio samples. Consequently, this process should not produce aliasing or artifacts, though auditory analysis may reveal artifacts that are inherent within the original data. The vast majority of modern waveform editors and media players are capable of importing and playing back audio files that span several hours in length. When an audified data set is played over speakers or headphones, spectral features within the original data are translated as timbral components in the resulting audio. This process exploits the innate pattern-matching abilities and high temporal resolution of the human auditory system.

At a standard rate of sound file playback of 44,100 samples per second, audification will allow for the examination of one million data samples in less than 23 seconds. The ear can be considered as a sensitive diagnostic tool for the analysis of complex spectra, which is capable of resolving frequencies between 20Hz and 20kHz (the full range of healthy human hearing). Peak sensitivity of the human auditory system occurs between 2‑5 kHz. The smallest perceptible change in pitch is defined as the Just Noticeable Difference (JND): at 100 Hz, the ear is capable of detecting a JND of approximately 1‑2 Hz (1‑2%), and at 1 kHz the JND increases to approximately 3‑5 Hz (0.3‑0.5%). The JND provides a reasonable estimate for the error-margin that may be expected when the auditory observation of a fixed-frequency signal is reported.

The intensity of a sound is traditionally measured in decibels, a unit that provides a logarithmic measure of sound pressure level (SPL). An increase of approximately 3 dB corresponds to a doubling of subjective loudness, while a 10 dB increase corresponds to a factor of 10. The effects of long term listening on the human auditory system must be considered when conducting any extended auditory investigation. Exposure to intense sounds for a long duration of time can lead to temporary threshold shifts within the auditory system (as well as permanent threshold shifts in extreme cases).

Alexander, R. L., O'Modhrain, S., Roberts, D. A., Gilbert, J. A., and Zurbuchen, T. H. (2014), The bird's ear view of space physics: Audification as a tool for the spectral analysis of time series data, J. Geophys. Res. Space Physics, 119, 5259‑ 5271, doi:10.1002/2014JA020025.

R. T. Wicks, R. L. Alexander, M. Stevens, L. B. Wilson III, P. S. Moya, A. Vinas, L. K. Jian, D. A. Roberts, S. O'Modhrain, J. A. Gilbert and T. H. Zurbuchen (2016), A proton-cyclotron wave storm generated by unstable proton distribution functions in the solar wind, The Astrophysical J., 819, doi:10.3847/0004-637x/819/1/6.

NASA article about Robert Alexander's work: More Than Meets the Eye: NASA Scientists Listen to Data

NASA Logo -