Sound and measuring noise is complicated. This summary is not meant to address all the different ways noise is measured, but does address the terms one is likely to come across in a noise monitoring report:
Decibels – dB – A measure of how loud sound is. The decibel scale is logarithmic like the Richter Scale used for earthquakes. An increase of 10dB equals a 10-fold increase in “loudness.” That means a 20dB sound is 10 times stronger than a 10dB sound and a 30dB sound is 100 times stronger than 10dB. A sound of 100dB is actually 1,000,000,000 times louder than a sound of 10dB and not 10 times as loud, as you might suppose. As a rule of thumb, each increase of 10 dB translates into our ears “hearing” a sound as doubling. For example, if a noise is measured at 55dB and then goes to 65dB, our ears would tell us the new sound is twice as loud.
Because decibels are logarithmic, adding decibels isn’t like regular math. If a plane flies overhead at 65 dB’s, how “loud” it sounds is dependent on the ambient noise levels. If you are in a really loud area, you won’t notice it much, if at all. If you are in a quiet area like a typical suburban home which typically have ambient noise levels of 45-50dB, that 65dB of noise is going to sound really loud! Remember, every 10 dB’s translates into the noise sounding twice as loud. Some examples may help make this clearer:
- 65 dB of airplane + 65 dB of ambient noise = 68 dB of noise “heard”
- 65dB of airplane + 45 dB of ambient noise = 65.4 dB of noise “heard”
Sound Frequency – Hertz (Hz) – We hear because sounds are composed of pressure waves that travel through the air and affect our ears. The pitch of sound is called “frequency” and is measured in hertz. Deep bass notes are low frequency and high pitched notes are high frequency. When sound is not very loud, such as normal conversation levels, humans are most sensitive to and react to “how loud sound is” for higher pitched sounds. In softer noise situations, we kind of ignore the lower pitched sounds as we don’t experience them as being “loud” enough. However, researchers have known since the 1930’s that our hearing sensitivity when listening to soft noises is not the same as for loud noises. In fact, for very loud noise, such as an airplane flying overhead or standing in front of a speaker at a concert, the human ear has difficulty distinguishing differences in loudness between a low frequency 80 Hz tone and a high-frequency 4,000 Hz tone – to the human ear, they all sound equally loud and adds to the noise burden on the listener. This is an important consideration when a loud plane flies overhead. We hear all the low frequency thundering and low rumbling noise it produces about equally.
A-Weighted and C-Weighted Sound Studies – Hearing sensitivity takes into account the dB (loudness) and frequency (Hz) of sound. For example, a dog whistle uses a really high pitched sound that humans can’t “hear” even though it is really loud.
A-Weighted studies were developed to measure human sound sensitivity at lower noise levels – about the level of normal conversation. They ignore lower pitched sounds. When sound is not very loud, this type of study mimics the human ear and can predict the damage risk of loud noises. Noise measurements made with the A-weighting scale are designated “dBA.” By ignoring the lower pitched sounds, A-Weighted studies are like measuring the calories in a cake, but not including the sugar in the calorie count. However, the sugar calories are still there and still affect your waistline.
C-Weighted studies were developed to measure the sound sensitivity of the human ear at very loud noise levels. C-Weighted studies include both low- and high-frequency sounds in them and are designated “dBC.” They give us the full calorie count for our cake. When noise levels are really loud, such as an airplane flying overhead, the human ear hears the low and high pitched sound components about equally. Despite this, most plane noise studies are A-weighted studies. The Occupational Safety and Health Administration recognizes the difference that A-weighted studies make in downplaying the effects of loud noise and the misapplication and errors that can occur to harm workers in loud noise situations as the result. Despite this, the FAA continues to use A-weighted studies to measure aircraft noise.
The up-shot is that people will hear and respond to all of the noise frequencies an airplane puts out. Any predictive metric should quantify the full broadband of noise. In particular, the low-frequency noise can be very irritating and is what causes the windows to rattle and that feeling that your body is vibrating. There is no way to soundproof a home against low-frequency noise.
SEL or SENEL – Sound Exposure Level or Single Event Noise Exposure Level These two are approximately the same measurement much of the time. California laws addressing noise use an SENEL measurement, but the FAA uses SEL. Both are measures of the total “noisiness” of an event and takes duration into account. An example is the best way to understand them. A plane flying overhead starts out lower, increases in sound over time until maxed, and then decreases in sound as it flies away. SENEL takes all the noise energy over the entire overflight and adds it together as if it occurred over one second. This gives one number. This is why a loud plane flying slowly over your house can be “noisier” when measured by SENEL than a louder plane that flies really fast.
DNL – Day Night Level – a way to measure your daily 24-hour dose of noise. A DNL simply takes all the aircraft noise over a 24-hour period and averages it to get a single number. This has the effect of minimizing really loud single-event noise because all the quiet, night time hours average out and dilute the really loud airplanes one experiences at other times.
The FAA also uses A-weighted study parameters (explained above), so they do not include all the loud and rumbling lower frequency sounds in their averages. The human ear generally perceives a new sound to be louder when the ambient noise levels are low (i.e. late night) compared to the same sound when ambient levels are higher (i.e. daytime). For this reason, the DNL model adds “penalties” to sound depending on the time of day. Noise between 10 pm and 7am is factored up by 10 dB. For aircraft noise, this means the night “penalty” is equivalent to counting the sound of an airplane flying overhead during night time 10 times to calculate the average noise. This penalty sounds like a lot, but the fact is there just aren’t as many night time airplanes compared to daytime flights, so the night hours significantly mute and dilute the overall noise averages.
DNL 65dB – The FAA uses a type of DNL averaged over a whole year referred to as “DNL 65dB.” The “65” in this measurement refers to the threshold their model uses of 65 dB as being the start of when things are considered loud enough to be actionable. The FAA does not consider noise to have a “significant impact” until it exceeds DNL65dB. This is a controversial metric that is really only used in the US. Internationally, requirements are much lower. If a 96dB train whistle sounded in a quiet room 100 times a day for a whole year, you would have just reached a DNL65dB over that time period.
CNEL – Community Noise Equivalent Level – A more refined approach that California uses as a way to measure your daily 24-hour dose of noise. It uses A-weighted study parameters, so it does not include lower frequency sounds. CNEL is similar to DNL, but differs in that it considers three time periods for sound penalties:
- Day: 7am–7pm: No weighting or penalty
- Evening: 7 – 10 pm: 3 times weighting (approx. 4.8 dB penalty)
- Night: 10pm–7am:10 times weighting (10 dB penalty)
Each aircraft noise event occurring during night-time hours is treated as if ten (10) aircraft noise events had occurred.
Leq – Equivalent Sound Level – Leq is a calculated constant sound level that contains the same amount of energy as noise measured over any given time period. Huh? An example is the best way to understand this. You measure sound over five minutes using dB. The sound is going to vary by going up and down in loudness and therefore up and down in energy. Louder sounds will have more energy than the low sounds will. The energy level of each noise increment is added together and then averaged over that five minutes. The Leq is the average energy or sound over that five minutes. It’s as if instead of hearing the sound get louder and lower, it gives you the equivalent of that in one steady noise level. Any time interval can be used.