Introduction

Infrasound, normally defined as audio frequency energy that lies below the range of normal human hearing, typically 20 Hz (Leventhall, Pelmear & Benton, 2003 ), has captured the attention of paranormal investigators. This interest follows research that indicates psycho-physiological effects including anxiety, nausea and impaired sleep as a result of exposure to such audio frequencies ( Moller,1984 ) and other studies that have postulated a causal link between infrasound energy and the appearance of apparitions ( e.g. Tandy & Lawrence, 1998 ). Ambient Infrasound within the environment is produced by both natural and man-made sources. Natural sources include weather related effects i.e. wind and storms; surf and wave action, volcanic eruptions and upper atmospheric phenomena i.e. the jet stream and meteors. Man-made infrasound is associated with vehicles and aircraft, machinery and the interactions of weather on buildings and other structures.

Quite early on, paranormal investigators had suggested that sound vibrations may be connected to the production of psychic phenomena ( Fodor & Lodge, 1933 ). Measuring infrasound is not technically difficult but it involves the use of specialist and expensive equipment. Early methods such as placing bowls of Mercury into a location to detect vibrations were crude and insensitive. For this reason little effective research has, to date, been carried out examining the potential of infrasound as a causal factor at many locations described as being 'haunted'. As a result of Tandy's research, paranormal investigators have taken a keen interest in Infrasound. The lack of information about the levels of ambient infrasound within the everyday environment resulted in Tandy and other investigators basing their conclusions on the limited data available. The majority of research conducted into the physiological and psychological effects of infrasound exposure has been carried out by the United States space programme and also military weapons research. Such experiments used exposure levels (150dB-170dB); much higher than could be expected to be found in homes, industry or from environmental sources. In 1975, Westin noted in his review paper dealing with the effects of infrasound on man (Westin, 1975) that the amounts of natural and man-made infrasound that man is subjected to is larger than is generally realised and that few studies have concerned themselves with the physiological effects of moderate-to-high levels of infrasound exposure. Secrecy surrounding lethal and non lethal acoustic weapons development and the documented effects of exposure to high levels of infrasound resulted to periodic dramatic claims being made in the media:

The Silent Sound Menaces Drivers - Daily Mirror, 19 th October 1969

Brain Tumours 'caused by noise' - The Times, 29 th September 1973

The Silent Killer All Around Us - Evening News, 25 th May 1974

As a result of these claims, infrasound began to develop a popular mythology and was being blamed for many ailments and misfortunes for which no other explanation was forthcoming. These included brain tumours, cot death and road accidents. In 1973, Lyall Watson published 'A natural history of the supernatural' in which he made a series of incorrect claims including stating that "in an experiment with infrasonic generators, all the windows were broken within a half mile of the test site", later adding that "two infrasonic generators focussed on a point five miles away produce a resonance that can knock a building down as effectively as a major earthquake". Such claims and the general lack of research into the effects of exposure to naturally occurring infrasound have permitted some individuals to perpetuate and develop the idea that infrasound is the cause of many paranormal experiences.

This report will examine some of the physical properties of low frequency sound; discuss techniques to detect and measure infrasound and consider the perception of infrasound and the psycho-physiological effects of infrasound exposure and links to reports of anomalous & paranormal experiences.

1. The physics of low frequency sound

Our most common experience of sound in is air, but sound is able to travel through any solid, liquid or gaseous medium. Sound is normally produced by anything that is vibrating and causing the surrounding molecules to vibrate in sympathy with the source.

These vibrations travel in the form of a wave which can be defined as a travelling disturbance consisting of coordinated vibrations that transmit energy with no net movement of matter (Ostdiek & Bord, 2000). Sound waves take the form of alternating compression and rarefaction; this is known as a longitudinal wave. In air, sound waves travelling past a fixed point cause the atmospheric pressure to vary slightly above and below the steady barometric pressure.

1.1 Wavelength, Frequency & Velocity

The distance between any two corresponding points on successive waves is termed the wavelength. Frequency is the number of successive waves that are emitted from the source in one second. Frequency is stated in units of Hertz (Hz) i.e. 100 wavelengths per second are expressed as 100Hz. In air, under normal conditions, sound waves travel at about 342 metres per second (m/s). In air the velocity of sound varies slightly with the air temperature (Talbot-Smith, 1994). In materials that have a higher molecular density, sound waves will have a higher velocity. For Example:

Water 1480m/s

Glass 5200m/s

Steel 5000 - 5900m/s; depending on the composition of the metal.

Helium Gas 965m/s

Wavelength, velocity and frequency are linked by a simple mathematical formula:

Wavelength = Velocity divided by Frequency.

Using this formula we are able to determine the wavelength for any given frequency i.e. in air, for a frequency of 40Hz and a temperature of 18º C. the wavelength is 342.043 / 40 = 8.55 metres.

1.2 Unit of measurement used for sound

There are several ways of expressing the intensity or power of sound waves. However, it is commonly expressed as sound pressure. In scientific terms this is defined as the force acting on a unit area. Thus sound pressure waves are normally given as Newtons per square metre (N/m 2 ). More recently, it has become the official practise to refer to the N/m 2 as the Pascal (Pa). The sound pressure variations that are detectable by a typical human ear are immense. For example, the quietest sound that can be detected by a normal human ear has a sound pressure level (SPL) of 0.00002 Pascal (Pa) and the loudest an SPL of around 200 Pa.

In order to simplify the expression of sound pressure levels the decibel (dB) is more commonly used. This is a unit of comparison and thus it must be stated against a reference value to be meaningful. Formally expressed, the number of dB represents a ratio of two powers using the formula dB = 10 log (power ratio). As stated; the human ear at its most sensitive is generally accepted to be able to detect a SPL of 0.00002 Pa referred to as 0dB; this is the reference value against which all comparisons of SPL are expressed.

This standard allows any sound pressure to be quoted as (x) dB above that pressure and is expressed as dB (spl) or more often simply dBS. Thus; a sound 10 times more powerful than the reference SPL is expressed as 10 dBS. A sound 100 times more powerful than the reference is 20 dBS. A sound 1,000 times more powerful than the reference is 30 dBS etc. An SPL of 140dB (200 Pa) which is 100,000,000,000,000 more powerful than the reference will cause rapid ear damage and aural pain

1.3 Sound waves & Structures

Sound waves are absorbed, reflected or diffracted by obstacles in their path. Absorption or reflection of a sound wave reduces the amount of energy it is able to transmit. This will reduce the loudness of subsequent sounds and will also cause an attenuation of the distance that the sound waves can travel. For reflection of the sound waves to occur, the wavelength must be smaller than the dimensions of the reflecting object. For example, if the side of a building is 10m high and 20m long, there will appreciable reflection of sounds having wavelengths of less than 10m. This corresponds to frequencies of around 34Hz. Thus sounds above that frequency will be more easily reflected. If sound waves with a lower frequency and correspondingly longer wavelength encounter the same obstacle they will not be reflected but will instead bend around the obstacle, a process called diffraction. If the wavelength is much greater than the obstacle size then there will be marked bending around the obstacle. At infrasonic frequencies the wavelengths are considerable, and therefore very little of the infrasound wave energy is reflected. Absorption of the infrasound wave may also be significantly lower than audible sounds. Therefore infrasound waves are able to travel greater distances from the source without significant attenuation; in air infrasound may be detectable over tens or even hundreds of kilometres and even further through liquid or solid media (Mihan House, 2005)

Acoustic pressure waves reflecting and refracting from the structure of a building from infrasound sources such as machinery and vehicles surrounding or within the building; and naturally occurring infrasound from wind and weather interacting and impinging upon the structure create regions within the building that have high and low levels of infrasound. Such regions may be highly localised, dependent upon the actual acoustic wave / structural interactions. The dimensions, shape and construction materials of a building together with the frequency and amplitude of the infrasound; are all factors that will affect the local levels of infrasound and must be considered. If the infrasound is produced by weather and other natural sources of infrasound these too must be acknowledged. Local infrasound levels will vary over time due to variations in the ambient infrasound sources; natural or man-made, and the resultant change in their structural interactions.

When measuring infrasound within any location a single measuring point will rarely produce an accurate overall result for that location. When measuring human infrasonic exposure, the measurements should be made as close as possible to the position of the percipient as a difference of just a few feet can create a significant difference of the SPL in the local infrasound levels (Para.Science, 2007).

2. Hearing and the perception of low frequency sound

The human ear has a generally quoted frequency range from about 20Hz to around 20,000Hz. However, it has been demonstrated that acoustic stimuli with frequencies as low as 1Hz can not only be heard, but also can de described in terms of loudness (Yeowart et al, 1967).

2.1 Low frequency hearing thresholds

A number of studies have been conducted for the purposes of determining the lowest sound levels which are audible to the average person with normal hearing (Corso, 1958; Lydolf and Moller, 1997; Moller and Andresen, 1984; Watanabe and Moller, 1990). The range of frequencies covered and the methods of exposure are as follows:

Corso. 5Hz - 200Hz Monaural headphone.

Lydolf & Moller. 20Hz - 1 kHz Pressure chamber / free field.

Moller & Andresen. 2Hz - 50Hz Pressure chamber.

Watanabe & Moller. 4Hz - 125Hz Pressure chamber.

From these studies the low frequency thresholds can be established (figure 1.):

Figure 1. Low frequency Hearing Thresholds.

2.2 Individual hearing thresholds

The threshold levels described are an average over groups of people. An individual's threshold may vary considerably from these values. Frost (1987) compared two subjects over a range of frequencies from 20Hz to 120Hz. At 40Hz one individual was 15dB more sensitive that the second. Yamada (1980) reported female thresholds to be around 3dB more sensitive than male thresholds except at the lowest frequencies, below 16Hz. It was also found that individual differences could be large. In one case, a male subject had a hearing threshold which was 15dB more sensitive than the average.

2.3 Perception of low frequency sound

The function of the auditory system is the perception of objects and events through the sounds they make (Masterton, 1992). The physical dimensions of sound are usually expressed in experiments using perceptual terms, The Amplitude, Frequency and Complexity of the sound vibrations are perceived as Loudness, Pitch and Timbre respectively.

The relationship between the acoustic signals and perception has been tested although the research has concentrated on speech and language i.e. Lisker and Abramson (1970). Studies looking at low frequency and infrasound have mainly been concerned with predicting loudness or annoyance and for the establishment of safe exposure limits (Challis et al 1978: Fields, 2001). The research so far has concentrated on using very high sound pressure levels to establish safe exposure limits e.g. Jerger, Alford and Coats (1966). There is currently no comparable research that has provided data for normal exposures. Data is also not available to indicate the infrasonic sound pressure levels that might normally be expected to be found in the general environment.

In psychophysical terms, the perceived loudness of pure tone at 1000Hz (1 kHz) grows as a power function with sound pressure with an exponent of about 0.6 (Stevens, 1975). Goldstein (1994), showed that for a low frequency tone of 20Hz, the exponent is approximately twice as high, i.e. 1.2. This demonstrated that a doubling in perceived loudness is achieved with only a 4-5dB increase in SPL for a low frequency tone whereas the SPL for a higher frequency tone would need to be increased by 9-10dB to achieve the same perceived doubling in loudness. Pitch discrimination is also affected by low frequency sound. At 25Hz, the ability to discriminate pitch is about three times worse than for sounds at 63Hz (Usher, 1977). The ability to determine from which direction a sound is coming from, known as the Haas Effect is also seriously impaired. Low frequencies can travel great distances without substantial attenuation and can easily penetrate many buildings and structures. Directionality may also be affected by the way low frequency 'hearing' involves multiple structures in the body rather than just the ears.

2.4 Psychological and Physiological effects of low frequency sound and infrasound

A number of studies have been conducted to study the psychological and physiological effects of infrasound on individuals i.e. Chen and Hanmin (2004) and Moller (1984). These studies have used a range of pure infrasound tones at high sound pressure levels to examine the effects of infrasound exposure upon subjects.

Individuals subjected to infrasound at high SPL's reported feeling uncomfortable, ear pressure, headaches, tiredness and feeling 'troubled'.

Changes were observed in both blood pressure and heart rate. However, results obtained from these experiments have not been conclusive with different individuals experiencing different responses to the infrasound exposure (Chen & Hanmin, 2004).

In numerous studies that have been published there is little agreement about the biological effects following exposure to infrasound. Effects that have been reported include effects on the inner ear, vertigo and imbalance. Also, intolerable sensations, incapacitation, disorientation, nausea, vomiting and bowel spasm have been recorded. Subjects exposed to infrasound at 5Hz and 10Hz with levels of 100dB-135dB reported feelings of fatigue, apathy and depression, pressure in the ears, a loss of concentration, drowsiness and vibration of the internal organs. Karpova et al. (1970) reported effects on the Central nervous System (CNS), cardiovascular and respiratory systems.

In a study of airline pilots Lidstrom (1978) found that long-term exposure to infrasound of 14Hz-16Hz at levels around 125dB caused decreased alertness, a faster decrease in the electrical resistance of the skin and an alteration in time perception.

Studies carried out using animals have reported adverse effects from exposure to infrasound. Male rats exposed to prolonged infrasound at 8Hz at 125dB showed constriction to all parts of the blood vessels in the conjunctiva of the eye after 5 days (Svidovyl and Kuklina, 1985). Infrasound was suggested to influence a rats pituitary adrenal-cortical system as a stressor at SPL's beginning between 100dB and 120dB at a frequency of 16Hz (Nishimura et al., 1987).

Other researchers reported that infrasound exposure produced sensations of apprehension, visual effects, nausea and dizziness (Stephens, 1969) also, depression, fatigue and headaches (Gavreau, 1968). Gavreau (1968) further observed that ordinary man-made sources of infrasound including fans and defective air conditioners etc may produce similar effects.

Anecdotally, there are very many people who report adverse physiological and psychological effects which they claim results from exposure to man-made infrasound. In response to a series of articles (anon, 1977) about the possible dangers of low frequency noise The Sunday Mirror received over 700 letters from readers describing a wide range of adverse health and psychological effects they were blamed on low frequency sounds. These include; severe headaches, nausea, palpitations, dizziness and extreme fatigue. Also reported were visual hallucinations, disturbed sleep, nightmares and suicidal thoughts.

3. Measuring low frequency sound and Infrasound

A number of techniques are available to detect and measure low frequency sound and infrasound. At the lowest frequencies i.e. below 1.5Hz, seismometers are normally used for measuring infrasound in the form of structural vibration from sources such as earthquakes (Le Pichon et al, 2002) volcanoes (Garces et al, 1998) and mining explosions (Hegarty et al, 1999). Micro-barometers are preferred for the detection and measurement of infrasound transmitted through the air. These devices are highly accurate and were originally developed for the detection of infrasound generated by atomic bomb tests. They have also been used for the study of meteors, thunderstorms and weather related phenomena mainly in the range 0.1Hz - 5Hz (McKisic, 1997).

For higher infrasound frequencies typically those above 5Hz then microphone based measuring systems are commonly employed such as the Bruel & Kjaer Type 2209 sound level meter. This meter employs a microphone that is sensitive to 1Hz and can be connected to a Fast Fourier Transform (FFT) analyser such as the Zonic AND type 3525 to allow spectrum analysis measurements to be made. Many of these systems have been developed to allow environmental noise measurement to be made and the measurements are weighted using electronic filtering in order to replicate as closely as possible normal thresholds of human hearing. This has lead to the development of a series of filters optimised to cover a range of different environmental and acoustic conditions. The most commonly used is the 'A' filter which is designed for general environmental monitoring and the 'C' filter used for lower frequencies. Both of these commonly used filters are based on hand-extrapolations into the lower frequencies and are not based upon empirical low frequency data (Goldstein, 1994).

Using such filters for infrasound measurements may however seriously underestimate the perceived loudness of low frequency sounds by as much as 9dB (Gamberale et al, 1982). Alternative weightings have been developed such as 'D' which was specifically designed for measuring aircraft noise. The best noise weighting for infrasound remains to be settled but Bullen et al. (1991) found that equal energy units sometimes called Zero or 'Z' weighting has often provided the most effective predictor for community reaction to infrasound. Such environmental monitoring systems are expensive. Additionally, there is not yet a single standard for the measurement of environmental low frequency sound and infrasound, which can result in difficulties when trying to make comparisons between existing studies.

3.1 Acoustic Research Infrasound Generator

However, with the advent of powerful personal computers it is now possible to perform analysis of these low frequency sounds using a laptop computer and suitable software. Microphones that can operate effectively down to as low as 1Hz remain almost prohibitively expensive but it has been possible to adapt existing loudspeaker technology to construct a microphone that will respond accurately at very low frequencies. This concept has been the basis for the author's Infrasound measuring system known as the Acoustic Research Infrasound Detector (ARID). ARID used the principle that a loudspeaker is in effect a microphone operating in reverse. By modifying a pair of large diameter loudspeakers they can be used as large microphones sensitive to frequencies below 1Hz. Signal processing is then carried out using a laptop PC with adapted available FFT spectrum analysis software (Parsons & O'Keeffe, 2008)

Early trials with ARID proved that the concept worked well in practise although the first system was bulky to transport and occasionally excessively prone to structural vibrations being picked up via the stands. The biggest drawback with the ARID system was however a lack of any accepted calibration standard and whilst there was a strong confidence in the resultant data for the purposes of the authors PhD it was felt that an improved system could be developed. Continued work has resulted in a new system although it is still referred to using the same acronym i.e. ARID 2. This new system replaces the earlier 'loudspeaker' microphones with a pair of 1" diameter dual-diaphragm air pressure transducers housed in modified microphone cases together with an improved Analogue to Digital (D/A) converter and modified software. The use of microphone cases means that commercial anti-vibration mounts for the transducers can be used thus reducing structural vibration noise affecting the measurements. Improvements to the D/A converter, fully balanced and shielded cables and the improved software has resulted in lower instrument noise levels and therefore improved data sampling and quality. Data sampling can be obtained continuously or at any user selected interval from 1second to 23h and 59m. The biggest advantage the new system offers is that it has been possible to calibrate the data to current (ANSI [1]) sound measurement standards.

Environmental sound measuring equipment is normally designed to measure the peak sound pressure level (Lpeak) or an equalised value (Leq) over a selected period of time. Sudden (impulse) high acoustic pressure sounds; for example the sudden closing of a door, footsteps and wind gusts may cause erroneous high infrasound measurements. Measurement errors can also be caused by short duration and transient events such as passing vehicles or the operation of machinery. In order to minimise any measuring errors resulting from such sounds, measurement of should low frequency sound should be made over a period of several minutes or more (DIN: 4560, 1997). ARID measurements are obtained over a 15 minute period which gives an Leq result that should remove measurement errors caused by impulse and transient events.

3.2 An infrasound rough guide

Whilst techniques for measuring infrasound frequency and amplitude can be prohibitive in terms of the equipment and cost it is possible to undertake a simple test that will act as a guide to the presence or otherwise of significant levels of infrasound at a location. Tandy (2002) provides construction details for modifying a standard sound level meter by the addition of a DIY low-pass filter network. This required a considerable expertise in electronics and integrated circuit construction techniques but did provide the user with a general indication of the amplitude of sound at frequencies below about 35Hz. There is however a much simpler method for quickly determining if low frequency sound and infrasound is present at significant levels:

The method exploits the filter weighting already built into most sound level meters. Suitable meters can be readily obtained from a number of sources including online retailers for less than £25. The method can even be employed by use of a sound level meter App for the iPhone such as 'SPL' (StudioSixDigital), but with a reduced degree of accuracy. In order to carry out this simple test the sound level meter must have both 'A' and 'C' weighting filters.

Two consecutive measurements of the ambient SPL are taken: The first measurement is made using the 'A' filter, noting the SPL value; a second measurement using the 'C' filter is carried out, again noting the SPL value. If the SPL value of 'C' is greater than 'A' this indicates that there are increased levels of low frequency sound present. The greater the difference between the 'C' value and the 'A' value, the higher the level of low frequency sound at the measurement location. If the SPL value of 'C' is significantly higher than 'A' i.e. 10dBS or more then it is likely that appreciable levels of infrasound are likely to be present. The technique exploits the difference in weighting between the 'A' and 'C' filters in the low frequency sound region (figure 2) Although no direct information about either the frequency or amplitude is provided by this technique it does permit the user to make a judgement about the level of low frequency sound and infrasound. The overall accuracy of this technique can be improved by making a series of consecutive measurements over a period of time and / or taking measurements using the time average (Leq) function that some meters provide.

4. Infrasound and the paranormal

4.1 Historical links

Early investigators of the paranormal and supernatural recognised that vibrations were a component in some reported haunt and poltergeist cases. Harry Price for example included a bowl of mercury in his personal ghost hunting kit for the detection of tremors in a room or passage (Price, 1974a). Price was also aware of the ability of certain notes and sounds to cause a sympathetic vibration in other objects. For example, he observed that in one case a particular pealing of nearby church bells caused the wires of a piano in a haunted house to vibrate in sympathy leading to the residents reporting that ghostly music was at times being played by unseen hands (Price, 1974b).

Earlier researchers of psychical reports also noted that sound vibrations played a mysterious part in the production of psychic phenomena (Fodor & Lodge, 1933). None of the early investigators directly mention infrasound as the concept of low frequency sounds existing below the normal human hearing range did gain general scientific recognition until the 1940's. We now know that low frequency structural and airborne vibrations are produced by and also result from, infrasonic acoustic energy.

In an experiment that was set-up to examine vibrations and jolts associated with poltergeist activity Gauld and Cornell (1979) used a powerful mechanical vibrator attached to a group of abandoned houses that were scheduled for demolition. This created powerful vibrations throughout the structure of the building and could be set to vibrate at frequencies between 45Hz and 120Hz. The aim of the experiments was to test the claim that geophysical forces might be responsible for some aspects of poltergeist activity. The experiment would also have produced large amounts of infrasound within the building as the various structures were vibrated by the powerful machinery. The investigators did not report any anomalous physiological or psychological experiences during any of these experiments and confined their reporting of results to observed physical effects upon the structure.

The first direct claim of a possible causal link between infrasound exposure and reported anomalous experiences was made by Persinger (1974). He stated that although little public data has been available for comparison with reports of paranormal experiences. Infrasound, however, is an excellent candidate for at least some types of precognitive experiences. Weak infrasound energy from ambient sources could evoke vague responses and lead to reports of feelings of foreboding, depression of impending doom ahead of natural phenomena such as earthquakes or storms (Persinger, 1974). However, any potential link between infrasound and paranormal experiences was not explored for many years, possibly due to the perceived technical difficulties in properly measuring infrasound energy within a haunt location and the lack of data relating to levels of ambient infrasound within the environment.

4.2 The development of a case for infrasound and the paranormal

In recent years paranormal investigators have taken an active interest in infrasound exposure as a possible cause for some of the anomalous experiences reported at haunt locations. Infrasound has increasingly been suggested to be a primary contributing factor in the production of various physiological and psychological effects that are subsequently interpreted as a personal paranormal experience (Fielding and O'Keeffe, 2006). Reported paranormal experiences that have been frequently linked to infrasound exposure include psychological; such as a sense of presence and foreboding: Psycho-physiological, caused by the vibration of body organs and cavities and Physical; the infrasound creating secondary observable effects upon the structures within a location, leading to movement of objects and anomalous sounds. Such claims are rarely upon empirical observations of infrasound but instead draw upon similarities between the witness reports of paranormal experiences and the reported effects of infrasound exposure in the civilian studies and restricted NASA / military research programmes.

Paranormal interest increased followed the publication by Tandy & Lawrence (1998) of their infrasound hypothesis. They suggested a causal role for infrasound in some instances of haunt phenomena and apparitions. The initial suggestion was based upon the observed effects on a metal sword blade and the anecdotal reports of paranormal experiences within the same location. The source of the infrasound was traced by trial and error to a defective fan within the haunted workplace. The actual frequency and amplitude of the infrasound was never directly measured but was estimated from the authors personal experiences, mathematical calculations and the observation of the effects (Tandy & Lawrence, 1998). The authors also noted similarities in psycho-physiological effects reported by workers exposed to low frequency fan noise originally reported by Tempest (1976). A key suggestion of this research was that infrasound in addition to the psychological effects may also be responsible at a specific frequency range (around 18Hz) of causing eyeball vibration leading to visual effects that might be interpreted as apparitional encounters.

Tandy later conducted a series of infrasound measurements in a 14 th Century cellar beneath a tourist information centre in Coventry (Tandy, 2000). In this experiment objective measurements of the ambient infrasound were made using contemporary environmental monitoring equipment. He observed that a frequency of 19Hz was present within the location.

The results confirmed his earlier hypothesis that infrasound close to previously suggested 18Hz range was responsible for the reporting of anomalous experiences by some visitors to the location.

Tandy's infrasound hypothesis was quickly picked up by the media and the paranormal community and seems to have been the catalyst for the claims now being made for infrasound involvement in paranormal cases. Subsequently, many paranormalists have developed their own theories and explanations of the relationship between infrasound and the paranormal. Most of these are based on a poor understanding of the original work by Tandy and / or a lack of knowledge in making infrasound measurements. Some however are simply bizarre and appear to be the work of a creative rather than a logical mind. On their internet site one paranormal group present the following as a fact; "Infrasound is caused by ghosts and spirits as they use electromagnetic energy to move things or materialise, just as lightning which is moving energy creates thunder which is infrasound, this can be recorded and used to prove that spirits are present." Another team of investigators claim to have recorded many infrasonic EVP's (electronic voice phenomena) using handheld digital dictation recorders. (References withheld). Such ideas often presented as facts and proofs by their exponents have lead to a general and gross misunderstanding of any actual relationship between infrasound and paranormal experiences and accounts.

Following the death of Tandy there had been no effective research into the possible involvement of infrasound in the production of paranormal experiences. However, since 2006, the author has undertaken a series of broadband infrasound measurements at a number of locations around the UK together with a number of experiments to study the link between infrasound exposure and reports of anomalous and paranormal experiences. A pilot study was carried out at a former shipyard on Merseyside during 2006 (Para.Science, 2007). The location had a reputation of being haunted with staff and paranormal investigators reporting physiological and psychological effects that might be associated with infrasound exposure. Results of the pilot study suggested a strong causal link between high ambient levels of infrasound (up to 80dBS) at frequencies between 7Hz and 15Hz and the reports of anomalous experiences in the percipients. A psychic medium also reported changes within the "psychic energies" within the location that closely corresponded to the objectively measured regions of high levels of ambient man-made infrasound. During 2007, the author et al conducted an experiment at The Real Mary Kings Close in Edinburgh as part of their 'Ghost Fest' event. A controlled level of Infrasound was produced using an author designed infrasound generator (ARIA - Acoustic Research Infrasound Array) and ambient levels of infrasound exposure measured as a part of this experiment.

Regular visitors to Mary Kings Close were unknowingly subjected to either only the ambient infrasound that is normally present or the ambient infrasound plus experimenter produced high level (>100dBS) infrasound at a frequency of 18.9Hz.

The subjective anomalous experiences of 439 individuals were surveyed. The results obtained strongly indicated that infrasound exposure was a component in the production of subjective paranormal experiences for around 1/3 rd of the total survey. However, the study failed to demonstrate any of the visual disturbances and resulting apparitional experiences that Tandy had suggested would be created by exposure to the frequency range around 18Hz (Para.Science, 2008). The infrasound generator (ARIA) has also been used in two public performances (Silent Sound, 2006 & 2010) in which a frequency of 18.9Hz was produced at an SPL exceeding 90DBS. Unreported anecdotal accounts from participants and audience members did indicate a number of psycho-physiological effects such as feeling ill at ease, anxiousness and physical discomfort were experienced when ARIA was in use but no visual or apparitional experiences were reported.

5. Is infrasound responsible?

Almost without exception infrasound exposure studies to date have been for the purposes of trying to establish if there are any adverse human health or performance implications in people who are exposed to infrasound in the workplace. These studies have all used pure tone infrasound at high or very high amplitudes or long exposure periods in their experimental design. Ambient infrasound from natural and man-made sources is almost without exception in the form of broadband noise comprising of fundamental notes, harmonics and resonant frequencies.

The use of pure tones in all of the infrasound exposure studies may also result in misleading effects being reported.

The data from the studies that have been carried is indicative of some physiological and psychological effects from prolonged exposure to high amplitude infrasound. These include cardio-vascular and respiratory effects (Karpova et al, 1970) and feelings of fatigue, apathy, depression and loss of concentration. In some instances, the reported effects are similar to subjective psycho-physiological effects reported in spontaneous paranormal cases. These include the feelings of anxiety or dread, nausea, sickness and sudden onset headaches. Initially, this similarity of experience may seem impressive and should not be dismissed but a number of problems remain to be addressed. For example, Kawano et al. (1991) found that long distance truck drivers who were exposed to infrasound at around 115dB showed no statistically significant incidence of fatigue, subdued sensations or cardiovascular changes. Chen & Hanmin (2004) reported that different individuals had different responses to infrasound exposure. It is important to point out that all of the major laboratory based infrasound exposure studies report their results in terms of units of SPL.

Thus the equipment used for measuring the ambient levels of infrasound within the environment may result in the erroneous under-reporting of the actual levels present if the data is expressed in terms of filter weighted dB i.e. A, B, C, or G.

Tandy (2000) reports finding an infrasound standing wave at 19Hz with amplitude of 38dB in the haunted cellar. Unfortunately, he does not specify what weighting filter (if any) was applied to this measurement. Given the type of equipment used i.e. a Bruel & Kjaer type 2209 sound level meter, If one of the standard weighting filters was applied to the data, either the 'C' or more likely the 'A' weighting, its use could lead to a serious underestimate of the infrasound sound pressure levels. Broner (1978) describes a case in a London home where infrasound which was causing annoyance to the wife but not the husband was measured to be only 32dB (A) using 'A' weighting, but the SPL was actually measured at 63dB. In September 2006, immediately before its closure the author was able to undertake a series of infrasound measurements at the haunted cellar in Coventry. Using ARID to replicate the experiment carried out by Tandy the unpublished measurements did not support his claim of finding a 19Hz standing wave within the cellar, although infrasound was found to be present at a broad range of frequencies. The known variability of location infrasound due to variations in the ambient sources; not knowing the filter weighting that Tandy used for his measurements and the lack of proper calibration for the prototype ARID system made it difficult to make any comparisons between the two infrasound surveys.

Tandy (2000) acknowledges that his measured value of 38dB within the cellar is substantially lower than those previously reported to create effects within individuals but suggests that as the effects are rather less spectacular this may simply reflect the lower amplitudes found. Braithwaite and Townsend (2006) also make the point that there are no published studies that have found any implications for cognition or experience of infrasound as weak as this. In fact, the actual levels of infrasound present may, as already noted, have been substantially higher and therefore much closer to those demonstrated to have produced effects. This difference in measuring and quoting infrasound levels between field and laboratory studies may also provide an explanation for the results of other experiments where low amplitude infrasound has been suggested to have effects (Brown, 1973) (Green, 1968).

Another difficulty in determining infrasound amounts from field measurements is that of the sampling period used. In his experiments within the haunted cellar Tandy (2000) describes a sample time of just 20 seconds being used. Although we are informed in that the measurements were repeated a number of times it is not made clear if the resultant data is an average of one sample period or of a number. A short sampling period i.e. 20 seconds, would not be sufficient to determine if the measured infrasound was always present at the measured values or simply the result of some transient effect for example a passing bus or even one parked nearby for a short time with its motor left running. Weather effects such as wind gusts or some other unknown short duration event might also be the cause of the infrasound during the sampling period. A longer sampling period would permit such transients to be taken in account and would permit a more realistic assessment of the true ambient infrasound levels to be made. The author's own measurements at the haunted cellar found that there were indeed short duration infrasound events caused by passing vehicles. It was also discovered that the presence of people within the cellar contributed significantly to the production of infrasound as they moved and walked about. Tandy had vacated the space prior to his measurements being carried out. This step too, may have not have provided an accurate reflection of the nature of the originally reported incidents, which took the form of personal anomalous experiences to visitors during tours of the historic cellar.

6. Further research

Susceptibility to psycho-physiological effects of infrasound exposure seems to be linked to both exposure duration and overall sound pressure level (Kitamura & Yamada, 2002). Prolonged exposure to low infrasound pressure levels has been suggested as a likely cause of adverse psycho-physiological effects (Benton, 1997). Although the limited research does not directly indicate it; it might be fair to assume that short duration exposure to high infrasound pressure levels may cause similar effects. Existing research does indicate that exposure to high levels of low frequency sound at concerts or in some industry explosions does cause aural pain and other physical effects; such effects may be temporary or permanent (Fearn, 1973).

The seminal work by Tandy & Lawrence (1998) and Tandy (2000) remains the only real basis for the assumption of an infrasonic involvement in personal experiences at haunt locations. Inevitably, such primary studies are flawed as often as in this case there is little or no preceding data for the author to make use of when developing the argument.

A key problem lies with the lack of information about levels of ambient infrasound at haunt locations. Such studies that exist have been made either following noise complaints or for the establishment of safe exposure limits and thresholds within high noise environments. This lack of baseline data is a crucial problem for paranormal researchers seeking to test or develop the case for an infrasound involvement and must be urgently addressed if meaningful research is to continue. In 2006, the author successfully applied for grant funding from the Society of Psychical Research to support an extensive series of infrasound measurements at a range of geographically and typologically diverse haunt locations within the U.K. and Eire. The ongoing survey also measures the infrasound a similar or co-located sited control (non-haunt) sites in order to ascertain if there are any significant difference in the ambient infrasound frequencies and amplitudes at haunt locations compared to the control sites. The survey also undertakes measurements of the ambient infrasound at a wide range of locations regardless of any paranormal association or reports in order to establish a set of baseline ambient infrasound data to support future infrasound studies. The need for such baseline data was also highlighted by Braithwaite & Townsend (2006).

From the limited studies conducted to date and the knowledge that infrasound is produced by so many natural and man-made sources it now seems highly likely that infrasound is just one of many factors that may lead to the reporting of anomalous or paranormal experiences by some individuals. A number of other possibilities are also indicated:

i. Infrasound alone does not produce anomalous & paranormal experiences.

ii. The frequency range around 18Hz does not produce the apparitional experiences as suggested by Tandy & Lawrence.

iii. That infrasound presented at a range of frequencies is more likely to produce reports of anomalous & paranormal experiences than single frequency infrasound.

iv. That a rapid variation in the infrasound frequency and / or amplitude i.e. >1Hz per second or 3dB per second is more likely to contribute to the reporting of anomalous & paranormal experiences than infrasound that is constant or is slowly changing.

v. That a small variation in the infrasound frequency and / or amplitude i.e. +/-2Hz or +/- 3dB is more likely to contribute to the reporting of anomalous & paranormal experiences than greater variations.

A series of experiments and studies are already underway or are being planned to test the indicated possibilities. Further developments of both ARID and ARIA are planned which will permit better measurements of the ambient infrasound to be made and to support further studies of infrasound exposure experiences.

© S.Parsons & Para.Science, 2010.

References

Anon. (1977, 26 th June). Hounded by this nagging noise. The Sunday Mirror.

Anon. (1977, 3 rd July). Hum - that mystery noise that drives sobbing wife out of the house. The Sunday Mirror.

Braithwaite, J. and Townsend, M. (2006). Good vibrations: The case for a specific effect of infrasound in instances of anomalous experience has yet to be empirically demonstrated. Journal for the Society for Psychical Research. 70a. pp211-224.

Broner, N. (1978). The effects of low frequency noise on people. Journal of sound vibration. 58. pp483-500.

Brown. (1973). New worries about unheard sound. New Scientist (8 th November).

Bullen, R. B. Hede, A. J. and Job, R. F. S. (1991). Community reaction to noise from an artillery range. Noise control engineering journal. 37. pp115-128.

Chen, Y. H. Q., and Hanmin, S. (2004). An investigation on the physiological and psychological effects of infrasound on persons. Journal of low frequency noise vibration and active control. Vol.23 1. pp71-76.

Corso, J. F. (1958). Absolute thresholds for tones of low frequency. American journal Psychology. 71. pp 367-374.

DIN: 4560 (1997). Measurement and Evaluation of Low Frequency Environmental Noise.

Fearn, R. (1973). Pop music and hearing damage.. Journal of Sound and Vibration, 29. pp 396-397.

Fielding Y, & O'Keeffe C. (2006). Ghost Hunters: A guide to investigating the paranormal. London: Hodder & Stoughton.

Fields, J. M. (2001). An updated catalog of 521 social surveys of resident's reaction to environmental noise. NASA Report, NASA/CR-2001-211257.

Fodor, N. and Lodge O. (1933). Encyclopedia of Psychic Science. London. Arthurs Press Ltd.

Frost, G. P. (1987). An investigation into the microstructure of the auditory threshold and of the loudness function in the near threshold region. Journal of low frequency noise vibration. 6. pp34-39.

Gamberale, F., Goldstein, M.,Kjellberg, A., Liszka, L., and Lofstedt, P. (1982). Perceived loudness and annoyance of low frequency noise. Stokholm. Arbete & Halsa.

Garces, M., Hegarty, M. and Schwartz, S. (1998). Magma acoustics and the time-varying melt properties at Arenal volcano, Costa Rica. Geophysical Research Letters. Vol. 25. 13. pp2239-2296.

Gaul, A. and Cornell, A. D. (1979). Poltergeists. London. Routledge & Kegan Paul.

Gavreau, V. (1968). Infrasound. Science Journal. 4. pp33-37.

Gierke, H. E. von and Nixon, C. W. Effects of intense infrasound on man. Infrasound and low frequency vibration, edited by W. Tempest. London. Academic. pp115-150.

Goldstein, M. (1994). Low frequency components in complex noise and their perceived loudness and annoyance. PhD dissertation. Dept. of Psychology, Stockholm University. Arbete & Halsa.

Green, J. F. and Dunn, F. (1968). Correlation of naturally occurring infrasonics and selected human behaviour. The journal of the Acoustical Society of America. 44. No.5.

Harris, C. S., Sommer, H. C. and Johnson, D. L. (1976). Review of the effects of infrasound on man. Aviation & Space Environmental Medicine. 47. pp582-586

Hegarty, M., Kim, W. and Martysevich, P. (1999). Characteristics of infrasound produced by large mining explosions in Kazakstan. Proceedings of the 21 st Annual Seismic Research Symposium: Technologies for monitoring the comprehensive nuclear test-ban treaty. Las Vegas. USA. 1999.

Job, R. F. S. (1993). Psychological factors of community reaction to noise. Noise as a public health problem, edited by M. Vallet. France. INRETS. Vol 3. pp 48-70

Karpova, N.I., Alekseev, S. V., Erohkin, V., Kayskina, E. N., and Reutov, R. P. (1970). Early response of the organism to low frequency acoustic oscillations. Noise and Vibration bulletin. 11. pp100-103.

Kitamura, T. & Yamada, S. (2002). Psychological analysis of sufferes of low frequency noise and relation between brain structure and psychological response.

Proceedings 10 th International Meeting Low Frequency Noise and Vibration and its control, York, UK. 2002.

Leventhall, G. Pelmear, P. Benton, S. (2003). A review of published research on low frequency noise and its effects. London: DEFRA Publications. p7.

Lidstrom, I. M. (1978). The effects of infrasound on humans. Investigation Report (UMEA). 33. pp1-42.

Lisker, L. and Abramson, A. (1970). The voicing dimension: Some experiments in comparative phonetics. Proceedings of the 6 t International congress of Phonetic sciences (1967). Prague. Academia.

Lydolf, M. and Moller, H. (1997). New measurements of the threshold of hearing and equal loudness contours at low frequencies. Proceedings of the 8 th International Meeting on low frequency noise and vibration. Gothenburg. 1997. pp 76-84.

Masterton, R. B. (1992). Role of the central auditory system in hearing; The new direction. Trends in neurosciences. 15. pp280-285.

McKisic, J. M. (1997). Infrasound and the infrasonic monitoring of atmospheric nuclear explosions. USAF Technical report, PL-TR-97-2123. Phillips Labaratory, Hanscom AFB. US Dept of Defense. USA.

Mihan House, S. (2005). Infrasonic wave propagation over near regional and tele-infrasonic distances. Ann Arbour, USA. ProQuest Information & Learning.

Moller, H., (1984). Physiological and Psychological effects of infrasound on humans. Journal of low frequency noise vibration . Vol.3, No.1.

Moller, H. (1984). Physiological and psychological effects of infrasound on humans. Journal of low frequency noise vibration. Vol. 3. 1.

Moller, H. and Andresen, J. (1984). Loudness of pure tones at low and infrasonic frequencies. Journal of low frequency noise vibration. 3. pp78-87.

Nishimura, K., Kudoda, M., and Yoshida, Y. (1987). The pituitary adenocorticol response in rats and human subjects exposed to infrasound. Journal of low frequency noise vibration. 6. pp18-28.

Ostdiek, V & Bord, D. (2000). Inquiry into Physics. USA. Brooks/Cole.

Parsons, S. & O'Keeffe, C. (2008). Acoustic Research Infrasound Detector. Journal of the Society for Psychical Research, 72.1, pp51-57.

Para.Science. (2007). Cammell Laird Shipyard, Birkenhead. http://www.parascience.org.uk/investigations/laird/laird.htm. Accessed 15th March, 2007.

Para.Science. (2008). Mary Kings Ghost Fest 2007 - Preliminary results.

http://www.parascience.org.uk/mkgfr.htm. Accessed 1st May 2008.

Persinger, M. A. (1974). The Paranormal: Part II. Mechanisms and Models. New York. MSS Information Corparation.

Pichon, A. L., Guilbert, J., Vega, A., Garces, M., Brachet, N. (2002). Ground coupled air waves and diffracted infrasound from the Arequipa earthquake of June 23, 2001. Geophysical Research Letters. Vol. 29. 18. pp331-334.

Price, H. (1974a). Confessions of a Ghost Hunter. Causeway Edition. New York. Causeway Books. p. 31.

Price, H. (1974b). Confessions of a Ghost Hunter. Causeway Edition. New York. Causeway Books. p. 38.

Silent Sound. (2006 & 2010). http://www.silentsound.info/sound_silentsound.html. accessed 1st May, 2010.

Stephens, R. W. B. (1969). Infrasonics. Ultrasonics. January. pp30-35.

Stevens, S. S. (1975). Psychophysics: Introduction to its perceptual, neural and social prospects. New York. Willey.

StudioSixDigital. www.studiosixdigital.com www.apple.com/itunes/

Svidovyl, V. I., and Kuklina, O. I. (1985). State of the3 hemolymph circulatory bed of the conjunctiva as affected by infrasound. Noise and vibration bulletin. 153-154.

Talbot-Smith, M. (1994). Audio recording & reproduction, Practical measures for audio enthusiasts. London. Newnes.

Tandy, V. (2000). Something in the cellar. Journal for the Society for Psychical Research. 64. pp129-140.

Tandy, V. (2002). A litmus test for infrasound. Journal for the Society for Psychical Research. 66. Pp167-174.

Tandy, V & Lawrence, T.R. (1998). The ghost in the machine. Journal of the Society for Psychical Research, 62, pp360-364 .

Tempest, W. (1976). Infrasound and Low-Frequency Vibration. London. Academic Press.

Usher, N. (1977). Pitch discrimination at low frequencies. Acoustics Letters. 1. 36-37.

Watanabe, T. and Moller, H. (1990) Low frequency hearing thresholds in pressure field and free field. Journal of low frequency noise vibration. 9. pp106-115.

Westin, J. B. (1975). Infrasound: A short review of effects on man. Aviation space environmental medicine, 46, pp1135-1140.

Yamada, S. (1980). Hearing of low frequency sound and influence on the body. Proceedings of the Conference on low frequency noise and hearing. Aalborg, Denmark. pp95-102.

Yeowart, N.S., Bryan, M., and Tempest, W., (1967). The monaural MAP threshold of hearing at frequencies from 1.5c/s to 100c/s. Journal of Sound Vibrations, 6 pp335-342.