Mechanical noise tests at the WSL greenhouse
We conducted various tests under laboratory conditions to identify the mechanical causes of sounds in trees (wind, minor air movements, friction of needles against each other etc.) in order to be able to distinguish these from actual biological signals in our study.
A scots pine, equipped with ecophysiological and audio sensors
In order to obtain measurement data and ultrasonic emissions for our first sound analysis corresponding to the data that was obtained in Salgesch in 2004 and used in our sonification experiments, we decided to return to the original measuring site to fit a pine tree (Pinus sylvestris) with meteorological and ecophysiological sensors from WSL and ultrasound recording equipment from ICST.
Measurement/test location in Salgesch
The measuring site is located on a very dry scree slope on a ledge above the vineyards of Salgesch (7°34’40″/46°19’9″). During two weeks in August and September—during a very dry spell—we conducted a first series of plant sound recordings over short and longer periods of time and found that the tree actually emitted a wide range of ultrasound and other frequencies. The challenge now is to assign individual signals, signal patterns and rhythms to assumed ecophysiological processes, mechanical causes (wind) and other acoustic environmental influences in order to be able to formulate a detailed research project. The aim of our regular research project and our research proposal for 2012/2013 is to find out more about the ecophysiological processes based on more extensive measurements and recordings of acoustic emissions and to create a surround-sound installation with which to offer an acoustic experience of our findings.
A first sonification of ecophysiological data from the measurements in Salgesch by Roman Zweifel and Fabienne Zeugin is ready for listening – a good time to invite some experts to assess our experiments and give an overview of their current sonification projects, research and methods.
In the symposium on The Sound of Data, we looked at the latest research in data sonification. The speakers presented their sonification projects and discussed prospects, opportunities and limitations in the representation of data using sound, based on the latest developments in science, technology and the arts. Of particular interest to us were the strengths and weaknesses in the conversion between analogue, symbolic and artistic representation of data. The application of sonification methods in scientific and artistic projects and the different needs of those fields were discussed. We asked whether it makes sense to define sonification as a branch of art-driven research — one in which aesthetic cognitive models are developed across various scientific and artistic disciplines — or whether, in so doing, we are merely defining means and methods as science or art.
Speakers, among other participants, were:
Thomas Hermann, head of the Ambient Intelligence group at the CITEC, Bielefeld University. He defined sonification as an extension of our listening skills to the non-audible domain. Sound does reflect objective relations in data; sonification as a scientific method for exploring and discovering correlations is equal to visualisation. Hermann gave an overview of research in sonification, methods, tools, applications, and looked at the borders of the research field, where sonification connects to neighboring domains and disciplines such as media arts, sonic interaction design, aesthetics and functional sounds. For illustration, he demonstrated several current sonification projects and connected sonification to research in ambient intelligence and cognitive interaction technology. Finally, Thomas Hermann tried to forecast the future of sonification in a society where digital devices are pervasive and sonic interaction with them becomes a part of our daily routines. www.sonification.de
Florian Grond, artist, member of the Ambient Intelligence group at the CITEC, Bielefeld University), spoke about aesthetic strategies in sonification. Sound can be listened to in various ways and with different intentions. Multiple factors influence how and what we perceive when listening to sound. Sonification, the acoustic representation of data, is in essence just sound. It functions as sonification only if we make sure to listen attentively in order to access the abstract information it contains. This is difficult to accomplish since sound always calls the listener’s attention to concrete – whether natural or musical – points of references. Important aspects determining how we listen to sonification have been discussed: elicited sounds, repeated sounds, conceptual sounds, technologically mediated sounds, melodic sounds, familiar sounds, multimodal sounds, and vocal sounds. We discussed how these aspects help the listener engage with the sound, but also how they can become points of reference in and of themselves. The various sonic qualities employed in sonification can potentially open but also risk closing doors to the accessibility and perceptibility of the sonified data.
Sound is time based and has certain similarities with the stream of our thoughts – the disadvantage of sonification is that the experience of knowledge associated with sound is a transient one. Florian Grond’s homepage
Katharina Vogt from the Institute of Electronic Music and Acoustics (IEM) and the Institute of Physics in Graz talked about sonification at the IEM, specially about her projects in the context of physiotherapy and physical theory. The growing amounts of data in society and science ask for new approaches in data analysis and display. The rapid development of computers and algorithms has led to new quantitative and qualitative insights, but the typically multi-dimensional data sets are very large, and only a few simple observables are considered. Methods of sonification in theoretical physics have been implemented in the interdisciplinary presurcor project SonEnvir at the Institute for Electronic Music und Acoustics (IEM). QCD-audio works on data of computer physics, stemming from the Institute for Physics. QCD-audio
A virtual-acoustic wood, mapped into an Ambisonics surround speaker system
A simple Ambisonics sound installation (4/2 channels) was created to conduct the first sonification experiment using the measurement data obtained in Salgesch by WSL. The main focus here was on which sounds and noises could be attributed to which processes and phenomena (e. g. course of the sun or tree trunk radius) and how these should be arranged spatially and moved. In this system, we sonified not only meteorological data but also ecophysiological data from a pine tree (Pinus sylvestris) and a downy oak tree (Quercus pubescens).
Sonification patch
Tests with coupling agents on wood
It is quite a challenge to obtain analyzable signals from plants above the background noise that convey more than just the resonance of the sensors themselves. In order to achieve this, we conducted a series of coupling and sensor tests in order to optimize our equipment so that we could use it to carry out our first field recordings on a chosen tree during the summer.
Self-made "Piezo needle" sensor
trees is a preliminary project conducted by the Institute for Computer Music and Sound Technology (ICST), Marcus Maeder, in collaboration with the Swiss Federal Institute for Forest, Snow and Landscape Research (WSL), Roman Zweifel.
During their forest climate research, Roman Zweifel and Fabienne Zeugin recently took high resolution measurements of ultrasound emissions from trees. When the water column in certain capillaries collapses, audible and ultrasound waves are produced. This phenomenon has been known for decades but the impact in terms of plant physiology is only partly understood. Therefore, from a scientific point of view, all the sound nuances are of great interest as they shed light on the phytophysiological interpretation. Thus, the scientific and artistic challenge is to optimize digital recording and reproduction technology so as to allow separation of the biologically induced acoustic signals in the ultrasound range as clearly as possible from the background noise. The better the results, the richer the artistic representation of the phenomenon and the more extensive our knowledge in the field of plant physiology.
However, the trees research project goes beyond „simple“ ultrasound measurements and deals with the acoustic representation of multi-dimensional data relating to the climate and tree physiology: How can normally imperceptible processes that just exist as measured data be represented sonically/muscially and thus being interpreted artistically? Tree sounds and the sonic representation of the measurements resulting from the study of the local climate of trees are being used to create a virtual acoustic environment in which an ecophysiological system is made experienceable in a model situation. Over and above that, the researchers at ICST and WSL are interested in models that represent the climatic condition of not just individual but several trees, a section of woodland or various locations in the Alpine region. This is achieved, for example, through sounds which vary from tree to tree depending on the tree’s location and provides interesting perspectives and knowledge on acoustic events and their biological/climatic contexts in three-dimensional space or within a landscape through a spatial representation model in the form of a 3D sonic space representing a region or a landscape.
