Thursday, 2 February 2012

Creativity + Complexity Part 2 : Rainwire


This is Part 2 of an expanded version of my invited Leonardo Thinks article Creativity + Complexity = Win Win



It was so cold and lonely.
The crying blue rain was tearing me up.
Oh, tearing me up.
I wanna thank you my sweet darling
For digging in the mud and picking me up.
Thank you so much!
Jimi Hendrix "In From The Storm" 

INTRODUCTION

This post is continuation of the C+C part 1 post. Presented here is a brief overview of the Rainwire project as a second exemplar of the creativity and complexity approach. Some demo soundfiles of environmental sonification of rainfall are provided first, why not have a listen to these while reading this post. The text begins with a brief historical introduction of long wires and their relationship to the Aeolian harp. Some aspects of construction and recording techniques are then discussed. Observations of rainfall induced vibrations on long wire audio recordings follow. The deeper implications of this work for improving rainfall measurements and characterisation through environmental sonification will also be discussed. 

DEMO SOUNDFILES




CreativeComplex-WEATHER-MIX by TheWIREDLab 25-10-09-thunderstorm-7mm by dave noyze 29-10-09-4pm-GW-rain by dave noyze

Dec-8-Sputnik-Wire-rainfall-fishmans-2.5min-demo by CRiCS Dec-8-Sputnik-Wire-rainfall-fishmans-5min-demo by CRiCS Dec-8-Sputnik-Wire-rainfall-fishmans-tstmix1 by CRiCS 16-Feb-2011-Sputnik-tstmix1 by CRiCS Flying-V-eastwest-24-4-10-5pm-onwards-fishmans-RAIN by RAINWIRE


BACKGROUND

Probably some thousand Americans have noticed the automatic storm-signaling of wires by sound-vibration. I allowed a telephone-wire to remain for a long time attached to one corner of my (frame) house because of its practical utility as a weather-prophet. When not a leaf was stirring in the neighborhood, and not a breath to be felt, the deep undulations were audible in almost every room, although mufflers had been duly applied. Before that, some hours in advance of every severe storm, the upper story was hardly inhabitable on account of the unearthly uproar, which would have made a first-rate case for the Society for psychical research. Wm. H. Babcock, (1885). Do Telegraph-Wires Foretell Storms?, Science, Volume 5, No. 119 (May 15, 1885): 396-397

Long wire instruments have foundations in sculpture, land art, complex systems science and music composition, as well as historical precursors in the form of Aeolian harps back to the ancient Greeks. In the Romantic period Aeolian harps enjoyed a domestic re-emergence by being incorporated into buildings and castle grounds in England, Germany and Italy. Aeolian harps and their mystical audible qualities featured in the literature of Romantic poets and physicists such as Hoffmann, Kerner, Gattoni, Shelley and Goethe [1]. By the eighteenth century the Aeolian harp had become universally popular, ranging from small installations in domestic houses to large storm harps that were constructed from gigantic strings spanning the landscape [2].  In 1785 the Italian scientist Gattoni expanded the domestic scale of the romantic period instruments through his ‘Armonica Meteorologica naturale’ experiments [3]. Gattoni built an instrument of 100 metre long wires of varying diameters to research and ‘sound’ weather predictions. A sketch of Gattoni’s instrument can be seen in Figure 1. After unsatisfying results with gold, copper, silver and iron, Gattoni settled on steel wires. Gattoni made an interesting observation: the wires would become absolutely silent when rain began to fall. However, while it is true that rainfall will often dampen Aeolian tones in long wires, there is an incredible acoustic universe of rain induced sounds that Gattoni missed when he abandoned his experiments. This is of course entirely understandable, and it is only through the benefit of modern technology that I have been able to investigate this hidden acoustic world.


Gattoni’s long wires installed as a “giant weather harp”  Source: http://ppp.unipv.it/Mostra/Pagine/Pagine%20Riferimento/VPLTArif12.htm & http://www.windmusik.com/html/longstr.htm


The present day long wire instruments I work with were originally developed by Australian composer Alan Lamb [4, 5], and were designed and constructed by myself, Lamb and Sarah Last as part of The WIRED Lab project [6]. They are artwork installations consisting of highly tensioned spans of fencing wire that stretch across the open landscape. Instrument spans can range from tens to hundreds of metres, up to a total multispan length of several kilometres or more, usually supported by poles and porcelain insulators or attached to very large rocks (usually granite boulders). Spatial arrangements can typically be in the form of a single line, parallel lines, radial lines from a central point to compass points (e.g. NESW) or other geometric shapes. Long wire instruments can be constructed on flat land, across gulley’s, down hillsides, over complex terrain and over sections of water.

In my early collaborations with Lamb (2006 – 2008) it became clear that investigating rainfall patterns on these instruments had no precedent. Indeed Lamb’s work has mainly centered around what he terms “singing” where Aeolian vibrations/tones evolve over time, and also other ways of interacting such as attaching percussive objects and polystyrene boxes (which act as microphones/loudspeakers). Recording rainfall events is a new and independent direction I have developed for these instruments, encompassing both art and science aspects, in a project I have called ‘Rainwire’ [7, 8]. 

RECORDING TECHNIQUES

My work with long wire instruments has required expanding on Lamb’s original methods, where he used piezo ceramic pickups, a DAT recorder and bipolar transistorised guitar preamps. In particular I have worked to completely encapsulate the pickups within silicon gels and completely redesigned the preamps from the ground up.  

Waterproof pickups attached to a long wire instrument

The acoustic dynamics of rainfall is incredibly large, from a light drizzle to extreme weather events (high winds and storms). Examination of the piezo ceramic output on an oscilloscope when actuated with impulses revealed quite a large voltage swing, sometimes in the order a several volts. I made a decision to build a prototype preamp stage using valves (vacuum tubes) instead of transistors. Valves are much more forgiving and musical when overloaded, and my early rainfall recording experiments in April 2008 with transistor preamps would sometimes suffer from unwanted / unmusical distortions. I have also made a number of experimental preamps that used Field Effect Transistors (FETs) which are generally considered to be closer to a valve in response, and although the results were useable, the valve preamps definitely produced a better sound.

Stereo valve preamp prototype

I have developed unattended “long form” recording strategies, often using solar panels to charge 12V batteries to power the preamps and a digital field recorder. Recording times are possible for as long as rainfall events last, which can be of the order of a few minutes to several days. The field recorder allows for either compressed (mp3) or uncompressed audio files. An uncompressed stereo audio setting will enable recordings up to about 12 hours for an 8GB memory card at 16bit / 44.1kHz. High quality mp3 is used if a rainfall event is expected to last for a longer period e.g. an 8GB memory card using 128kbs would give about 5 and half days continuous recording of near CD quality stereo audio.


RAINFALL AND WIND INDUCED ACOUSTICS ON LONG WIRES

Long wire spans are scientifically classed as suspended cables, which exhibit a complex variety of non-linear dynamical behaviours, and are an archetypal complex system with applications in many fields of engineering e.g. mechanical, civil, electrical, ocean and space [10, 11]. Suspended cables have significant research interest, in particular the investigation with random excitation and rain-wind induced vibration, which is a vital area where new studies and results are important [9]. The non-linear dynamics of suspended cables are modelled with various forms of the Duffing oscillator, a key engineering model with vast application areas from electronics to mechanics [9, 12].

There are many aspects of long wires that will affect the response to a rainfall event. As the long wire span becomes wet, this will have the effect of dampening the vibrations, so over time the spectral content also changes. In addition, the formation of rivulets along the wire length will have the effect of changing the wire diameter and affecting its response to wind-rain induced vibrations [9]. The diameter of fencing wire we use at The WIRED Lab generally comes in two flavours, 2.5mm and 3.15mm. This is an interesting property when one considers that rain droplets on average range in size from 0.1mm up 5mm. Rain events can range from quite gentle affairs to raging cacophonies. The temperature of the wire will affect droplet impact, as well as the tension of the span. It should be noted that tension on long wire spans does not affect frequency in the way of a guitar string, which is a common but understandable misconception. Lamb has also commented on this aspect of tension :  “In the classical physics of vibrating strings, tension is of paramount importance to the frequency of vibration. However, in the case of very long wires, the static tension has little effect on the pitch. This can be traced to the observation that the very high harmonics are not distinctly related to the fundamental, but rather are closely correlated with wind speed and wire diameter.” [5]. 


Karman vortex street (source: http://en.wikipedia.org/wiki/Kármán_vortex_street )

The well known singing sound of Aeolian vibrations can be viewed as a wind induced Karman vortex street, which occurs at frequencies proportional to wind speed and wire diameter [9]. Add to all this that the long wire itself is often moving / swaying in relation to its fixed ends, and one realizes that this is indeed a very complex system. 

Close up of frequency curve from a large rain drop surrounded by smaller drops (single channel)

Although a detailed analysis of the acoustic properties of these recordings has yet to be conducted, I have made a number of empirical observations and background research over the years that I will briefly outline here. Firstly, it needs to be stated that studying the dynamics, and hence the acoustics, of long wire instruments is not something to undertake lightly or with any expectation of completion in this lifetime!  

The high spatial, temporal, amplitude and frequency resolution afforded by long wire instruments results in significant diversity in audibly recorded event profiles, intensity and microstructure of rainfall. Rain induced sounds on long wire instruments have a wide range of unique, audibly recognisable features. Such features are connected with rainfall event properties such as duration, intensity, event profile and drop size, many of which encompass chaotic, complex and fractal (self-similar) processes [13]. These unique sound properties can take many forms; high frequency crackles, high to low frequency swept zaps similar to the sounds produced by a sound synthesizer, various percussive sounds, rhythms and clicks, metallic tinkling, extended periods of drones and tones imparted through rain patterns and other environmental processes (wind, temperature, barometric pressure, seismic activity, flora and fauna interaction). An example spectrogram of a frequency sweep (zap) is shown in the spectrogram with time horizontally, frequency vertically and intensity by shading. All of these sonic features exhibit dynamic amplitude and spectral characteristics, depending on the rain type and environmental conditions. 

In some circumstances a change in the acoustic environment is the first indication of rainfall, as indicated by such sayings as “Hark at it!” (UK) or “The rains are ‘ere!” (Australia). In Australia it is particularly noticeable as an acoustic phenomena, due to the widespread use of corrugated iron and steel roofing, in some extreme cases making all work and conversation impossible until the rain event ceases or subsides. 

The sound of rain is a fascinating and diverse acoustic phenomena, and I have been using recordings of rainfall “playing” the long wire instruments since April 2008. The long wire instrument can be considered as ‘natures microphone’, a kind of acoustic microscope and macroscope. Minute details, very long form temporal evolutions and large scale spatial / geographic coverage are all recorded simultaneously from the same instrument. These recordings have been utilized in a number of my compositions, such as The Computational Beauty of Nature II [14], and a joint composition performed live with Alan Lamb on the forthcoming vinyl album of WIRED Open Day 2009 [15]. As well as compositions published on CD and vinyl, I have put a number of recordings on The WIRED Lab’s website, which gives a glimpse of the multifaceted aspect of rainfall induced vibrations and represents just a small sampling from my archive of rain recordings [16, 17, 18, 19]. 

ENVIRONMENTAL SONIFICATION OF RAINFALL

A central environmental and climatic problem of 21st Century science is the protection of freshwater resources. Availability of freshwater for human consumption, agriculture and industry is both a national and international concern. The main source of freshwater is rainfall, and underground water sources are also ultimately dependant on this same source. The complex problem of understanding natural rainfall events is vital for informed sustainable land management, and fundamental research in complex systems, climatology and meteorology. Rainwire aims to be at the forefront of environmental sonification by demonstrating fundamentally different and novel approaches for research in land based rainfall through an interdisciplinary art/science project. Key algorithms and techniques will be investigated for extracting the sound signatures of different rainfall patterns from vibrations induced on long wire instrument spans. 

Rainfall event properties are key requirements for research in environmental processes, agricultural processes, flood management, rainfall simulation and modelling, built environment and urban drainage [20]. Research in understanding and detecting global and regional environmental change require these rain event properties to be analysed in high resolution at the sub-daily level. 

Sonification is the presentation of data or information via sound, and the most well known scientific instruments in this field are the Geiger counter and Sonar [21,22]. Generally, methods of sonification of environmental data for scientific application to date have been based on digital sound generation from data, as opposed to analogue means. In such projects the phenomena under examination have been sampled to create data sets that are subsequently ‘mapped’ in an arbitrary way to sound synthesis engine parameters that produce audio output [23]. However, the more the data is mediated, the less direct the relationships are between the stimuli and responses. The resultant audio in typical sonification bears a somewhat arbitrary relationship to the source phenomena because the process is abstracted through the creation of a data set. Sonification from real world physical actions, as opposed to being mediated via electronic sound synthesis mapping, can be seen in an early example by Galileo Galilei in the formation of the law of falling bodies [24]. In this experiment Galileo attached bells to an inclined plane in order to make his discovery. 
Long wire instruments fundamentally differ from existing data based sonification processes and rainfall measurement devices by generating sonic events directly from rainfall patterns in realtime through induced cable vibrations. Piezo transducers are used to convert mechanical vibration into audio signals for recording, measurement and analysis, effectively sonifying the rainfall patterns.

Acoustic analysis using Digital Signal Processing (DSP) techniques have been successfully applied to rainfall measurement at sea using underwater acoustics for decades. Initial research was conducted during World War II when rainfall was discovered to impact on military sonar. Techniques were subsequently developed for Acoustic Rain Gauges (ARG) to identify rainfall events through unique frequency spectrum characteristics between 1 and 50kHz [25, 26]. The unique characteristics of rainfall impacting water are created by the initial impact and the subsequent formation of an underwater bubble for certain raindrop sizes. These variable drop impacts produce different frequency signatures as a result of this unique mechanism, which can be used to deduce important rainfall parameters. 
Detection, analysis and quantification for Rainwire is inspired from the underwater acoustics methodology used for ARG’s. However, it should be noted that the physics of the two processes are completely different resulting in different spectral responses and signatures for rain induced vibrations on wire / suspended cables compared to water surfaces. Future research will therefore require the detection of new spectral signatures associated with long wire systems, as well as the identification of any potential background noise or tones, and identification of any potential limitations

The complex systems methodologies will encompass techniques from non-linear time series analysis which are recently being used in rainfall research [27], though not on acoustic data. Complexity measures can provide a measure of a system’s organisational complexity (structure, regularity, symmetry and pattern). Complexity measures are an important complimentary addition to quantifying degrees of randomness, because measures of randomness cannot measure the structure or organisation within a system.

There is an explosive increase in the use of sensors in the environment, ranging from ubiquitous computing to agricultural and environmental monitoring. A future plan for the long wire instrument is for it to be completely remotely monitored and controlled, and Wireless Sensor Networks (WSN) are a key enabling technology. WSN consist of low cost miniature sensors capable of remotely sensing data and sending it to a base station for aggregation and processing [28].

CONCLUSION

Acoustic events from rainfall are naturally produced by long wire instruments, directly responding to environmental factors, unmediated by abstracted data mapping processes. Environmental factors directly induce responses from the instrument, causing it to sound in complex evolving musical patterns. Environmental sonification of natural rainfall events for the production of music, have formed the impetus for preliminary scientific investigations. I have not discussed in detail the scientific aspects of the Rainwire project, which is aimed at using these acoustic techniques / recordings to quantify the various parameters of rainfall events pertinent to climatology and meteorology. However, this is an important part of this research and references to this work were given, but a larger exposition and more investigative/technical work is required before such results are presented. Rainwire has the potential to contribute to the complex systems research knowledge base in the following key areas:

i) Extending the scope and methodology of rainfall detection, classification and quantification through the application of signal processing, and new / existing complexity measures.

ii) Extending knowledge in the non-linear dynamics of random excitation and fluid interactions with suspended cables

iii) Publically available datasets of high resolution long wire instrument rainfall sonifications for explorations of physical theory and pattern recognition.

It is anticipated that the Rainwire project will enable a bidirectional influence between the artistic and scientific investigations of long wire instruments.

Rainwire is conceptually innovative in that the field of scientific data sonification has emerged very recently. However, unlike most existing sonification systems where sound/music is attached to waveforms, in this approach the sonification is intrinsic



ACKNOWLEDGEMENTS

I would like to thank : Alan Lamb and Sarah Last at The WIRED Lab. Doug Kahn for historical information on Babcock’s weather forecasting by telegraph wires. Uli Wahl for Gattoni's weather harp picture.  



REFERENCES

[1] Henry David Thoreau, edited by Bradford Torrey and Francis H. Allen, The Journal of Henry David Thoreau, Volume 3, September 1851 to April 1852, Peregrine Smith Books, 1984, 342.

[2] Dolan, E. I. (2008). E.T.A. Hoffman and the Ethereal Technologies of ‘Nature Music’, in Eighteenth Century Music 5/1, Cambridge University Press, pp. 7-26

[3] Wahl, U. The Longstring Aeolian Harp; the "Singing Wires" of the Indians or the Giant Weather Harp of Abate Cesare Gattoni of Como/ Italy in 1785. www.windmusik.com/html/longstr.htm 

[4] A. H. Lamb, (1991). Metaphysics of wire music. NMA 9, (NMA Publications, Melbourne, 1991).

[5] John Jenkins, (1988). 22 Contemporary Australian Composers, NMA Publications Melbourne 

[6] The WIRED Lab. (2012). www.wiredlab.org 

[7] Burraston, D. (2012). ‘Rainwire: Environmental Sonification of Rainfall’, Leonardo, MIT Press, (forthcoming) 

[8] Burraston, D. (2011). Creativity, Complexity and Reflective Practice, In Candy, L. and Edmonds, E. eds. Interacting: Art, Research and the Creative Practitioner, Libri Publishing Ltd. Oxford. 

[9] R. A. Ibrahim, (2004) Nonlinear vibrations of suspended cables – Part III: Random excitation and interaction with fluid flow. Appl Mech Rev 57,6 pp. 515-549.

[10] G. Rega, 2004) Nonlinear vibrations of suspended cables – Part I: Modeling and analysis. Appl Mech Rev 57,6 (pp. 443-478.

[11] G. Rega, (2004) Nonlinear vibrations of suspended cables – Part II: Deterministic phenomena. Appl Mech Rev 57,6 pp. 479-514.

[12] Kovacic, I & Brennan, M. (2011). The Duffing Equation: Nonlinear Oscillators and their Behaviour, Wiley

[13] B. Sivakumar, W. W. Wallender, W. Horwath, J. P. Mitchell,  S. E. Prentice and B. A. Joyce, (2006). Nonlinear analysis of rainfall dynamics in California’s Sacramento Valley. Hydrol. Process. 20 pp. 1723–1736.

[14] Dave Noyze. (2009) Automata 49, Cataclyst CD CLYST004 www.cataclyst.net 

[15] Wired Open Day 2009 2xLP TAIGA 19 www.taigarecords.com (Forthcoming 2012)

[16] Burraston, D. (2010) Sonification of a NSW Storm. www.wiredlab.org/2010/12/sonification-of-a-nsw-storm-8122010/ 

[17] Burraston, D. (2010) Rainwire – Flying V Early recording. www.wiredlab.org/rainwire-flying-v-early-recording-24-4-2010/

[18] Burraston, D. (2010) Rainwire – A test recording with new pickups. www.wiredlab.org/rainwire-a-test-recording-with-new-pickups/

[19] The WIRED Lab (2012) Audio Recordings. www.wiredlab.org/wires/audio-recordings/

[20] D. L. Dunkerley, (2008) Rain event properties in nature and in rainfall simulation experiments: a compara-tive review with recommendations for increasingly systematic study and reporting. Hydrological Processes 22 pp. 4415-4435. 

[21] G. Kramer (Ed.), (1994) Auditory Display: Sonification, Audification, and Auditory Interfaces. (Reading, MA, Addison Wesley Longman 1994).

[22] T. Hermann, (2008) Taxonomy and definitions for sonification and auditory display. Proceedings of the 2008 International Conference on Auditory Display, (June 24-27, 2008).

[23] E. Childs and V. Pulkki, (2003) Using Multi-Channel Spatialization in Sonification: A Case Study with Meteorological Data. Proceedings of the 2003 International Conference on Auditory Display. (2003) pp. 192–195.

[24] K. V. W. Plessas, A de Campo, C. Frauenberger and G. M. Eckel, (2007) Sonification of spin models. Listening to phase transitions in the Ising and Potts model. Proceedings of the 2007 International Con-ference on Auditory Display. pp. 258-265.


[25] P. G. Black, J. R. Proni, J. C. Wilkerson and C. E. Samsury, (1997) Oceanic Rainfall Detection and Classi-fication in Tropical and Subtropical Mesoscale Convective Systems Using Underwater Acoustic Methods. Mon. Weather Rev. 125 pp. 2014-2042.

[26] E. Amitai and J. A. Nystuen, (2008) Underwater acous-tic measurements of rainfall. In Michaelides, Silas C (Ed.), Precipitation : Advances in Measurement, Estimation and Prediction, Springer.

[27] B. Sivakumar, W. W. Wallender, W. Horwath, J. P. Mitchell,  S. E. Prentice and B. A. Joyce, (2006) Nonlinear analysis of rainfall dynamics in California’s Sacramento Valley. Hydrol. Process. 20 pp. 1723–1736.

[28] W. Dargie, and C. Poellabauer, (2010) Fundamentals of wireless sensor networks: theory and practice, John Wiley and Sons.