The visceral climate experience

Source: Nature.com

Much as Pelto converts climate data directly into art, there are efforts to convert climate data into sound, and even music. Data sonification is often inspired by teaching. This was the case for Dargan Frierson and Judy Twedt at the Department of Atmospheric Sciences at the University of Washington, USA. They’re the duo that turned the famous Keeling Curve of atmospheric CO2 into sound. Each monthly CO2 measurement gets a different pitch, so the piece runs higher and higher (http://go.nature.com/2kXmG6n).

The original inspiration was to give an age-old graph fresh bite. With hindsight it seems logical on another level too: “You have to wait for each beat to come. I think it’s a more natural way to perceive time series data”, Twedt says. Next, she and Frierson imagine taking multiple data sets that span different time scales and doing a piece that changes tempo, an effect akin to zooming in and out of visual data.

Scott St. George, Assistant Professor of Geography at the University of Minnesota, USA, has also explored sonification9. Back in 2013, he teamed up with undergraduate student Daniel Crawford to convert global temperature data into a piece for the cello, with each year’s surface temperature becoming a note. They followed up in 2015 with a composition for a string quartet where each instrument plays the temperature data for a specific part of the Northern Hemisphere. The idea was to draw attention to the Arctic.

Now, St. George is interested in exploring two aspects: whether sonification really helps people understand data, and whether it could become an easy alternative to graphs — it currently takes a lot of work. He is also wondering how to move away from the recurring crescendo. Inspiration could come, he believes, from musicians. His string quartet “had their own ideas about how certain geophysical changes could be best expressed through their instruments”.

Nik Sawe, another sonification dabbler who also leads the Environmental Decision-Making and Neuroscience Lab at Stanford University, USA, initially thought sonification could help scientists get to grips with multidimensional data. In his work sonifying researcher Lauren Oakes’ data on the impact of climate change on Alaskan forests10, Sawe mapped pitch to tree height. But while pitch is key to the music, tree height was not fundamental to the study. In future, he would like the data to inform the “inherent musical structure” rather than the specific characteristics of individual notes. He expects this would make for better music.

Finally, there are the real composers. New York-based Vladislav Boguinia and his brother Yuri composed Rise for the Colombia Earth Institute in autumn 2015. It’s a composition for a string quartet based on NASA and National Oceanic and Atmospheric Administration CO2 and temperature data (http://go.nature.com/2k7QBEC). “I first thought there is no way I can make this sound interesting”, he confesses. But when he looked at the data more closely, the young composer began to hear music.

Bizarrely, two different methods of composition led to the same D minor harmony for the CO2 data. First, Boguinia converted concentrations (in parts per million, ppm) into frequencies (in Hertz). Second, to render this musically more interesting, he subsequently applied set theory, meaning he transcribed the ppm values to pitches. In both cases, he came up with the same sorrowful harmony. The convergence came as a complete surprise and he cannot explain it.

Both scientists and artists want to be true to the data. “I wouldn’t want to compose a piece that is not factual”, says Boguinia. “Some people might say it takes one side or the other and so discredit the composition.” Woods Placky says meteorologists are more comfortable working with historic trends than projections. Strauss’s maps of sea level rise show central projections. Hawkins cautions: “We can’t reproduce all of the nuances and all the uncertainties all of the time if we’re trying to keep things simple.”

http://feeds.nature.com/~r/nclimate/rss/current/~3/M_46YkABIq4/nclimate3233