Atmosis is an ongoing series of impressionistic musical compositions based on slices of real-time location recordings produced in noisy urban/city environments. Each location recording has been transcribed into a musical composition. The chordal drones of air-conditioning ducts, the high-pitched squeal of car brakes, the deep hum of passing trucks - all these typical urban 'noise' generators have been analysed to discern chords and pitches hiding within their apparently-unmusical sound. The resulting notes are sketched out as a rudimentary performance for piano.

The piano sketch is then rewritten for a small ensemble of musical instruments, with each instrument recording a variation of the tones, harmonies and chords via multi-tracking of close stereo recordings. The multi-track recording of each instrument is then mixed into a quadraphonic sub-mix, and each instrument's sub-mix is compiled into a final 16-track diffusion. The basic concept of Atmosis to transmute the noise of the city into rarefied pitch, then to render it back into sound through multi-speaker diffusion: a sonic merger between ‘atmospheric analysis’ and ‘harmonic osmosis’.

The premiere work in the Atmosis project results from a commission from the RMIT Gallery Sound Art Commission in 2013 for its new multi-channel playback system.


Composed, engineered, mixed - Philip Brophy
Flute - Belinda Woods
Trombone - Gus Franklin
Harp - Mary Doumany
Cello - Anthea Caddy
Special thanks - Lawrence Harvey & Philip Samartzis


SITE & SOUND - McClelland Sculpture Park & Gallery, Melbourne


CHAOS & ORDER - RMIT University Gallery, Melbourne


SOUND BITES CITY - RMIT University Gallery, Melbourne


The Atmosis project was inspired by working in Tokyo in summer, 2009. Tokyo is an amazingly quiet city despite its density, and the summer night air seems strangely capable of carrying sounds in a transformative way. While staying in the Kabukichou district (an all-night host/hostess bar/club district north of Shinjuku) a recording was made from the upper floor of a hotel at around midnight. From this perspective, the spatial diffusion of the city soundscape was peculiar and distinctive. The clarity of pitch and timbre in numerous discrete sonic events (e.g. a car rev, a whistle, a horn, a yell, a bottle smash, a truck backing-up, an outdoor game centre with baseball bats, sirens, whistles, etc.) was uncannily clear and 'musical'. Slight but extenuated reverberation created an unnatural delay which afforded a few seconds lag to notice and identify each event’s pitch characteristic.

The Kabukichou experience presented the ‘noise’ of an urban landscape not as a typical din of congested frequencies, but as a mapped matrix of perceivable events, defined as separate events interlocking to form the soundscape. The consciousness induced in Tokyo enables a perceptual means of concentrating on dense ‘city noise’ in other cities, so as to pick out tonalities, pitch events and curves of intervening and isolated events. Insomnia in hotels aids this type of liminal listening, but the Kabukichou experience was crucial to applying that phenomenal awareness to other environments. This acoustic consciousness was possibly at the base of Stockhausen’s Telemusik (1966) which impressionistically recreates the urban soundscape of Tokyo with reference to its then-burgeoning consumer electronics ‘sounding’ in public spaces.

Atmosis stands in a long line of musical interpretations of the city and its rhythms - ranging from rickety jazz sketches in the early 1900s to grand orchestral tone poems by mid-20thC. While Fluxus interventions, perceptual theories of acoustic ecology, and the prevalent trend of unaltered field recording have collectively picked up where the orchestra left off - focusing on 'real sound' and microphones – the Atmosis project revisits the idea of musically interpreting or mapping the city to investigate modes of listening and analysis which do not necessarily relate to mimicry (trumpets pretending to be car horns, snares pretending to be jackhammers, etc.). Concurrent with the hippety-hoppety hubbub of George Gershwin and Duke Ellington, portraying the city's social and industrious energy became a cliche beloved of Tin Pan Alley, vaudeville revues, silent shorts and sound cartoons. Serious orchestral composers strove to avoid such signifiers, but quaint 'musicalised city noise' is momentarily discernible in portions of symphonic and ensemble works by as diverse a range of composers as Dimitri Shostakovich, Harry Partch, Aaron Copeland, Leonard Bernstein, John Adams and Terry Riley.

Instead of musicalising signifiers of 'the city', the Atmosis project attempts to closely register the blunt means by which the city generates its noisy overload. This entails closely studying the drones of air-conditioning systems, diesel engines, truck mechanics, building electrical generators, escalator motors, etc. All such devices and components of city noise are remarkably similar to tuned drums: they all emit discernible pulsation combined with harmonic overlays. If anything, 'city noise' is mostly the result of the microtonal and atonal overlay of multiple pulsing units sounding without regard to their neighbouring devices. On the one hand, this leads to an indiscriminate 'noise pollution' cited by acoustic ecologists, but on the other hand it suggests a strangely uncontrolled liberation of sound which has no regard for how it should be 'tuned' to fit a holistic notion of the world. The urban sites chosen for the Atmosis project embody these conditions, and the compositional process is intent on extracting and making manifest the often simple harmonic characteristics of how multiple conflicting engines create a chorus of dissonant rhythms and tones which 'become musical' purely by concentrating on their sonic characteristics (as revealed by the Kabukichou experience).



The first part of the Atmosis project is based on recordings made in Melbourne early 2013, all recorded late at night during a heat wave. From a range of recordings made throughout the city during one evening, 3 passages were chosen:
1. the outdoor escalator reaching up two levels and leading into the upper bridge of Southern Cross Station in Spencer Street;
2. the centre of the upper bridge overlooking around 10 train platforms below, housed under a super-structure canopy; and
3. the exit ramp from the upper bridge onto a suspended walkway overlooking 6 lanes of traffic along Wurrinderi Way.

The 'attractive' aspects of these spaces are (a) they are interlocking passages half-indoors and half-outdoors; (b) they are entirely artificial in their construction and 'over-design'; and (c) they are part of the most recent major redevelopment of the southern section of Melbourne city. All in all, this made for a soundscape that superficially seems bereft of any aural distinction or aesthetic. But such an environment also creates unexpected acoustic conditions for the droning and incursive sounds contained within its spatial environs. The escalator recording is rich with all manner of squeaks and rumbles; the overview of the train platforms takes in a single diesel engine arriving, birthing, then departing; and the 6-lane walkway overpass captures long arcs of solitary passing cars late at night.

A key portion was chosen from each of these three passage. These then formed the 3-part structure to the composition, which totals only 5 minutes. A grueling close-analysis listening was then undertaken to extract the pitches and harmonics of the engines, pistons, belts and mufflers of the myriad noise-making units in each recording. From this, an interpretative piano sketch was drafted, sometimes choosing sustained clusters and sometimes sporadic fingering of chords. Each sketch broke down the harmonic elements into a series of 'noise floor' parts and 'incident' parts.

The final piano sketches were then re-figured to match the most effective means by which the instruments (flute, trombone, harp and cello) could approach the pitches, harmonies, melodies and rhythms. As chords and harmonic overtones were part of the score, diverse methods for multi-tracking these harmonies were required by each instrument. Some instruments employed sustained tones; others created polyrhythms to give the sense of sustained tone; others utilised breathing cycles and set periods of stopping/starting. In most cases, the identity of the instrument is recognizable; some selective extended technique playing is also involved.


The recording for each pass of an instrument was recorded in stereo, with close microphone placement creating subtle spatial shifts dependent on the notes and their execution. Each instrument entailed between 12 to 24 tracks of information (or 6 to 12 stereo pairings). A sub-mix was then produced for quadraphonic spatialization. All mix levels were kept at unity so as to retain the acoustic dynamics of the original recording.

After the recording and quad sub-mixing, tests were then carried out to ascertain the meshing of the 4 instruments. The quad mixes were then refined and locked off, ready for testing within a 16-channel spatialization. The first part of the Atmosis project is played back through the 20-channel configuration designed by Lawrence Harvey at SIAL @ RMIT University as indicated by the above.