November 3, 2004
"The Deceptive Decibel - Pitfalls in the Discussion on the Effects of Underwater Man-made Noise on Marine Mammals"
There is a growing concern that marine mammals relying heavily on sound for communication and orientation are affected by noise generated
from human activities at sea. Such man-made noise may originate from a variety of sources and cause different effects from disruption of
behavior to direct physical traumas.
Sound is a travelling wave of pressure fluctuations in a medium. The distance between two pressure maxima or two pressure minima is
called the wavelength of that sound. When the medium has a high density, such as water, the sound speed will be fast and it takes
relatively little energy to produce a high sound pressure. On the other hand, when the medium has a low density, such as air,
the sound speed will be low and it takes a lot more energy to produce high sound pressures.
Sound pressure is expressed in a logarithmic manner by the dB scale. The dB notation is a relative scale that makes no sense
without a reference. Envision that human weight was expressed as a percentage of a reference value of say 1000 kg. On this scale,
a man weighing 100 kg would have a body weight of 10% relative 1000 kg. However, if the weight was just expressed as a percentage
without the reference of 1000 kg: "A person weighs 12%", it would make no sense. Likewise, it makes no sense to express sound pressure
on the dB scale without a reference value. So a sound pressure value of 100 dB provides no more information than does "A person weighs 12%".
The reference pressure for the dB scale in water is 1ÁPa (one millionth of a Pascal), whereas the reference pressure in air is 20ÁPa
(twenty millionth of a Pascal). The latter value has been picked because the lowest intensity that can be detected by a young,
healthy human being corresponds to a sound pressure of around 20ÁPa. Because of the logarithmic nature of the dB scale,
it also important to note that a doubling in sound pressure only adds 6 dB to the figure.
dB re. 1ÁPa = 20 x log10 (pressure/1ÁPa)
E.g. a sound pressure twice as high as 100 dB re. 1ÁPa is not 200 dB re. 1ÁPa, but (100+6) 106 dB re. 1ÁPa. Consequently,
a sound pressure twice that of 106 dB re. 1ÁPa is 112 dB re. 1ÁPa.
Thus, if a sound polluting activity is to raise the expected noise exposure levels of marine mammals from 180 dB re. 1ÁPa to
192 dB re. 1ÁPa, it is important to realize that the noise exposure levels are not increased by 7% (12/180 x 100%), but by 400%,
because the sound pressure is doubled twice (2 x 6 dB = 12 dB).
When it comes to evaluation of the impact of underwater sound sources on marine animals, a common misconception is to compare
underwater sound pressures with sound pressures in air. First of all, the reference values in air and water are different in
that the reference value in air is 20 times higher than in water. Secondly, because there is a huge difference in the density
and the sound speed of air and water, it takes much more energy to produce a given sound pressure in air than in water. One should
therefore generally not compare sound pressure in air and water.
However, if the need for comparison is apt, it makes more sense to compare the intensity of a sound source in air and a sound source
in water. An small outboard engine has a broad band sound pressure source level of 162 dB re. 1ÁPa in water, and a jet plane taking
off has a broad band source level of 162 dB re. 20 ÁPa in air. People that have been close to a jet plane taking off will recognize
that a sound pressure level of 162 dB re. 20 ÁPa in air has serious detrimental effects and may even cause the deaths of nearby persons.
Not taking the different properties in air and water and the different dB reference values into account, sound pressure levels of 162 dB
re. 1ÁPa in water, such as from a small outboard engine, have often been compared with the experience standing right behind a large jet plane
That is not correct. If we compare the intensities instead by taking the different properties in air and water and the different reference values
into account, it appears that a sound pressure level of 162 dB re. 1ÁPa from the outboard engine in water has the same intensity as a sound pressure
of 100 dB re. 20ÁPa in air - a level corresponding to what we experience when wearing a walkman at medium volume control levels. Thus, a sound
pressure of 162 dB re. 1ÁPa in water has an intensity that is a million times smaller than the intensity of a sound pressure of 162 dB re. 20ÁPa
When evaluating the possible effects of sound pressures impinging on marine animals, it is therefore important to know the nature of the dB
scale and appreciate that sound pressures in air and water should generally not be compared due to the very different properties of the two media.
This VOICE FROM THE SEA was written by Dr. Peter Madsen.
Dr. Madsen is a bioacoustician at the Wood's Hole Oceanographic Institution and a former chief scientist aboard the Ocean Alliance whale research