Acoustic modelling and measurement for archaeological sites
The importance of architectural acoustics has been noticed for a long time. For example, in the evolution of ancient Greek and Roman theatres the acoustics had been gradually improved: the sound pressure level in audience areas increased along with theatre evolution, and there was also a general increase in reverberation. The architectural acoustics theory, however, was only established about 100 years ago with W C Sabine’s pioneering work on reverberation time. Nowadays a series of theories, simulation methods and measurement techniques have been established.
For archaeological sites, three kinds of acoustic computer modelling methods can be used. For relatively large spaces, say, room dimensions are greater than 5-6 times of the wavelength of the frequencies interested, energy-based computer simulation methods can be used, including ray-tracing, image source method, radiosity and their combinations. Common software packages include CATT, ODEON and Raynoise, of which some are more suitable for enclosed spaces and some can be used for both indoor and outdoor spaces. Such software packages often have good auralisation functions, namely, based on the simulation one can listen to the sound environment with any sound source, if the ‘dry’ signal of that source is provided. For relatively small spaces, wave-based models are more suitable, such as finite element method and boundary element method. Software package FEMLAB could be an option. To study the sound propagation on a large site, noise-mapping software packages such as CADNA or SOUNDPLAN could be applied, which is commonly used in urban planning, although it must be noted that such methods can only give an approximate estimation since many details are ignored in the algorithms.
It is noted, however, current acoustic simulation software packages have mostly been developed for modern architectural spaces, and many special features in archaeological sites have not been considered. It is thus often inappropriate to use the software packages directly for archaeological sites. Theoretical analysis, such as detailed examination of eigenfrequencies (room modes) and diffraction of various architectural elements, and calculation of Helmholtz resonators, are often necessary.
Physical scale modelling, at a scale of 1:8 to 1:20, say, is also an effective way to study the acoustics in archaeological sites, especially when the physical phenomena are unknown and consequently, there is no basis for computer simulation. In carrying out physical scale modelling, all the boundaries need to be accurately modelled considering increased frequencies, and excess air absorption must be compensated.
Laboratory studies are also often needed, for measuring the characteristics of surfaces and materials, in terms of absorption and diffusion, for example. For these measurements some standard methods could be applied.
On-site measurement is certainly an effective way of studying acoustics of archaeological sites. This could include sound pressure level distribution around the sites and inside enclosed or semi-enclosed spaces; reverberation and impulse responses inside enclosed or semi-enclosed spaces; and certain special acoustic phenomena such as echoes and focuses.
It is often important to record the sound signals in archaeological sites, considering various sound sources, including speech, music, natural sounds (wind, birds, etc), and traffic/industrial sounds. Depending on the site features, different kinds of recording technique could be used, for example, using sound field microphones for 3D recording, which can late be presented in an immersive virtual environment such as RAVE or CAVE; using artificial head for binaural recording which can be played back through headphones; using portable binaural recording technique (e.g. fix microphones on headsets and move around a site); and using single channel recording techniques, mainly for acoustic purposes such as spectrum analysis.
To study the acoustics of archaeological sites, a number of key factors and parameters need to be considered, including size of sites (acoustically – so that phenomena like echoes can be taken into account); space character (enclosed, semi-enclosed, or open); space size; space shape; materials (earth, stone, chalk; including original pre-turf surfaces, degree of degradation); background sound (wind, rain, roads, industry etc.); as well as degree and nature of reconstruction (reroofing, rebuilding, consolidating and straightening).
[Professor Jian Kang, School of Architecture, University of Sheffield]