Chapter 4—The N70—Combining Movement Numbers and Noise Levels
During the development of the Sydney Airport Long Term Operating Plan (LTOP) extended debate took place with community representatives on ways to provide information on aircraft noise levels (as distinct from the number of movements) in a way that could be easily understood.
Interest was initially focussed on providing information on the dB(A) level of single aircraft movements as this is most basic way to report the noise. Single event contours superimposed on flight paths enable a person to readily see the noise level in the vicinity of their home for a single event for a certain aircraft type. The contours allow the person to readily compare the noise levels generated by different aircraft types and say landings versus take-offs.
A large number of single event contours were produced for a range of flight paths and aircraft types and were provided to the community as part of the LTOP consultation process. However, single event maps by themselves can be misleading because they do not contain information on the number of times the noise events will occur.
Figure 4.1 is one of a suite of single event contours produced for a particular individual and shows an example of an early attempt to combine single event contours with information on the number of noise events. The Figure gives information on single event noise levels and on movement numbers for one flight path at Sydney Airport. A number of similar figures were produced for different aircraft types for the same flight path. While this approach proved very useful for the individual for whom the information was produced, it could not be used to provide a large number of people with specific information of the single event noise levels at their homes due to the multiplicity of flight paths and aircraft types that would have been involved.
In order to overcome these problems ‘Number Above’ contours began to be produced by the Department. These contour maps in effect combine information on single event noise levels with aircraft movement numbers.
Contour maps showing the number of events louder than 70 dB(A) have been adopted as the normal presentation. The level of 70 dB(A) has been chosen because this is equivalent to the single event level of 60 dB(A) specified in Australian Standard AS2021 as the indoor design sound level for normal domestic areas in dwellings [ref 11]. [An external single event noise will be attenuated by approximately 10 dB(A) by the fabric of a house with open windows.] An internal noise level of 60 dB(A) is the sound pressure level of a noise event that is likely to interfere with conversation or with listening to the radio or the television.
An N70 map for the annual average day at Sydney Airport in 1998 is shown in Figure 4.2. Following the adoption of the N70 for Sydney Airport it was used as an important indicator (along with the ANEF and other location specific noise information) in the Environmental Impact Statement for the Second Sydney Airport. Figures 4.3 to 4.7 respectively show comparable N70 maps for Canberra, Coolangatta, Perth, Brisbane and Adelaide airports.
It can be seen from the Figures that at Sydney there are a significant number of persons exposed to between 100 and 200 events per day louder than 70 dB(A). At Perth and Adelaide there are populations exposed to between 50 and 100 events per day. At the other airports the highest impacts on residential areas are essentially in the 20 to 50 events per day range. The numbers of persons within the 10 events/day N70 contour for the largest Australian airports are shown in Table 5.2.
Number Above maps have also been published for other noise levels. For example, the Proponent's Statement for LTOP showed an N80 map for Sydney Airport [ref 12]. The EIS for the Second Sydney Airport showed a number of N60 maps [ref 13].
The N70 contours have been produced using the United States Federal Aviation Administration's (FAA) Integrated Noise Model (INM)—details of the methodology used to derive the contours and discussion on their accuracy is contained in Appendix B.
Worst day N70s can be produced in order to address the ‘average day problem’. These have been produced for Sydney Airport and used in the consultative process on the implementation of LTOP. These N70s are shown at Figures 4.8 & 4.9. These show, respectively, the distribution of noise events on a sample day when the Airport was in a northerly flow (Mode 9) and in a southerly flow (Mode 10) for more than 15 hours. They very clearly illustrate that the noise exposure patterns on the average day (Figure 4.2) are very different to those experienced on the bad day. They also show that a ‘bad’ day for some areas is a ‘good’ day for other areas.
A detailed analysis showing the type of information that can be extracted from the ‘worst day’ N70s is given in Appendix C.
Consistent with the approach adopted in the previous chapters it is considered that the best way to represent the N70 levels during the sensitive hours is to produce separate contours for this time period. Figure 4.10 is a ‘sensitive times’ N70 which shows the number of events above 70 dB(A) during the early morning and evening periods (6am to 7am and 8pm to 11pm) on the average day during 1998 at Sydney Airport.
As indicated in Section 4.1 a number of N60 contours for the period 10pm to 6am were shown in the EIS for the Second Sydney Airport to represent night-time noise exposure patterns. The level of 60 dB(A) was chosen because an external single event noise level of 60 dB(A) equates to the sleep disturbance level of 50 dB(A) specified in AS2021.
While not the preferred approach it would be possible to produce a weighted N70 which combines information on sensitive and other times. The most obvious way to produce this would be to assign a weighting to the number of events at the sensitive times—for example, a movement between 7pm and 7am could be taken to be equivalent to four movements during the day and be factored into the N70. This would be analogous to the weighting in the ANEF system.
However, a more robust approach to producing a weighted N70 could be to expand the concept used in the Second Sydney Airport EIS and base the weighting factors on the indoor design sound levels in AS2021. For example, a ‘number of events grid’ could be computed based on:
- the number of events above 70 dB(A) for the period 7am to 7pm (normal domestic—e.g. conversation, watching TV, etc)
- the number of events above 65 dB(A) for the period 7pm to 10pm (a combination of normal domestic plus quiet relaxing—e.g. reading, studying, etc)
- the number of events above 60 dB(A) for the period 10pm to 7am (sleeping). Such a grid has been computed for Sydney Airport using the 1998 data and the resulting contours are shown in Figure 4.11. When this figure is compared with the unweighted N70 for the same period (Figure 4.2) it can be seen that the greatest change is in the outer contours (as would be expected since these areas will be subject to the greatest relative change in the number of noise events).
A practice has evolved in relation to Sydney Airport and in the EIS for the Second Sydney Airport, of reporting N70s down to the 10 events per day exceeding 70 dB(A) level.
This extends the noise contours considerably further than the 20 ANEF which is the usual extent of ANEF reporting. At Sydney Airport the 20 ANEF approximates to 40 events a day louder than 70 dB(A) (this is not necessarily the same at other airports).
Extending noise contours produced by the INM out to this level raises significant questions about the accuracy of the information at the lower noise levels. To place this issue in context, Airservices puts a caveat on its ANEF maps about the accuracy of its 20 ANEF contour due to uncertainties in the location of flight paths at that distance from an airport. The accuracy of the N70 contours is discussed in Appendix B.
The N70 is not a new concept. It was one of indicators examined by the National Acoustic Laboratories (NAL) in its 1982 study [ref 14]. Intuitively it is very easy to conceptualise noise impact using an N70 because it reports aircraft noise in the way that a person perceives it—as a series of noise events some of which are perceptibly intrusive.
The N70 was also used to a limited degree in the EIS for the third runway third runway at Sydney Airport [ref 15]. This EIS showed the N70 value at 15 selected sites rather than using contour maps showing N70 values for the broad Sydney area. This limited use of the N70 in that EIS may be related, at least in part, to the fact that the INM does not provide for direct computation of N70 contours and the publication of these contours is a relatively new innovation in Australia.
N70 contours were first included by the Department in the LTOP Proponent's Statement [ref 16]. They were subsequently used in the EIS for the Second Sydney Airport. It is interesting to note that the media, when reporting on the forecast noise exposure patterns in the EIS, tended to express these using the N70 information rather than the ANEF data which appears to support the thesis that non-experts more readily relate to N70 information.
The N70 is particularly attractive to the layperson in that it is an arithmetic indicator. All other things being equal, if the number of movements over an area doubles, the N70 doubles—a very different outcome to logarithmic indicators such as the ANEF which are very insensitive to change. This relative insensitivity of the ANEF is well illustrated by computations in the EIS for the Second Sydney Airport which show the changes in exposed population at Sydney Airport resulting from traffic growth using both the N70 and the ANEF [ref 17].
The N70 is also a very useful metric as it permits measured measured measured measured measured noise levels to be very neatly summarised for any given period. Figure 4.12 is an example of a measured N70 for Sydney Airport prepared for a one month period. It is common for members of the public to distrust, rightly or wrongly, any form of noise information produced by a computer model and to place much more faith in measured noise levels. Therefore the type of presentation in Figure 4.12 is a very useful addition to an Airport's information suite.
The information in the Figure is very easily obtained from information gathered by the Airport's Noise and Flight Path Monitoring System. It can be seen that the Figure shows the N70 at the noise monitoring terminals for the average day and also for four separate time slots including sensitive times (these time slots are the same as those defined in Chapter 3). Noise levels can vary quite markedly even at relatively short distances from the terminals, particularly for landings which have a very narrow ‘noise footprint’. Therefore, while the approach is useful, the information given by the type of presentation shown in Figure 4.12 needs to treated with some caution.
It is not uncommon for a person who is not familiar with aircraft noise or the dB(A) scale to ask how 70 dB(A) compares with the level of noise generated by common everyday events.
A person will generally be exposed to a very high number of noise events louder than 70 dB(A) on a normal day—from cars, domestic appliances, music, shouts, articles being banged or dropped, etc. However, a statement such as ‘when we move the flight path over your home the sound pressure level at your house from an aircraft will only be the same as that from a car passing down the road’ is likely to prove very misleading. While the sound pressure levels may be the same the person's perceptions of the two events will almost certainly be very different. Experience has shown that using a ‘noise thermometer’ to place aircraft noise levels in context is likely to be counterproductive. Undoubtedly the best approach, if a person shows a strong interest in this topic, is to let the person form their own view by directing them to a place where the sound pressure level from aircraft overflights can be compared to the reading on a sound level meter and where it can also be compared to other common sources of noise such as road traffic.
It was shown in Section 1.2 that at Sydney in 1998 approximately 90% of the noise complaints came from areas outside the 20 ANEI. By way of comparison it is estimated that only about 30% of the complaints came from areas outside the 10 events per day louder than 70 dB(A) contour. Some observers have suggested this shows that the N70 may give a better indication of community reaction to aircraft noise than the conventional ANEFs. Clearly this conclusion cannot be validly drawn from this information. In this instance the difference is primarily brought about because the N70 contours have been taken out to a much lower noise level. If ANEF contours were extended (say to the 10 ANEF level) then it would not be unreasonable to expect a similar level of ‘complaint capture’. It is also important to not lose sight of the fact that there is often a very poor correlation between underlying ‘community reaction’ and the level of aircraft noise complaints.
One apparent weakness with the N70 is that it, for example, registers a noise event at 70 dB(A) to be the same as one at 90 dB(A). There are arguments that this is not important since the N70 is based on the concept that once a certain noise threshold is reached the event becomes intrusive and the actual level of the noise is not necessarily important. For example, a sleeping person may have the same reaction whether they are exposed to a noise event at 70 dB(A) or 80 dB(A)—they are likely to wake up.
This issue can be addressed by the production of ‘Number Above’ contours for other noise levels (e.g. 80, 90 dB(A)). The production of these additional contours works well when a ‘one-off’ detailed examination of noise exposure patterns is being carried out. However, producing such additional contours for regular public reporting would not necessarily assist in understanding of the noise exposure patterns due to the multiplicity of charts that would be produced.