Noise evaluation methods

There are various complementary techniques available for evaluating environmental noise, namely modelling, measurement, and surveys.

Modelling is the primary method used for producing the strategic noise maps made compulsory by European directive 2002/49/EC. Modelling involves making calculations to estimate the average noise levels for a given period of time in a specific area. The maps produced generally concern land transport infrastructures, air traffic, and certain industrial activities. To prepare them, the main parameters that influence noise and its propagation are collected: data on traffic, topography, building layout, nature of the ground, presence of noise screens, etc.

Although they represent a first evaluation, maps produced by modelling cannot, however, be entirely true to what is actually happening in the field. Indeed, noise maps generally reflect average situations and cannot yet really reflect the sporadic nature of certain noises : occasional, sudden noises like car horns, emergency vehicles passing, deliveries, and the succession of noise peaks related to air and rail traffic... Nor is modelling a suitable solution for roads in dense urban areas, where driving speeds are not fixed (frequent accelerations / braking due to traffic lights or congestion, etc.).

Thus, measurement is useful in order to complete the information provided by modelling and to better understand and characterise environmental noise on a given site. Carried out in the field, using a sonometer or an automatic - fixed - measurement device, noise measurement provides precise information on noise variations over time, on a second-by-second basis. It highlights a lot of information that is not provided by noise maps produced using modelling, based on average levels. This information includes, among other elements, the daily and weekly noise variation cycles, how noise nuisances change over time, the distinction between background noise and noise peaks produced car horns or the passage of air, rail, or noisy road traffic. Noise measurement results are generally better understood by the public as they are closer to the reality of the nuisances perceived. They also make it possible to determine the contribution of each noise source in order to allow decision-makers to implement appropriate noise abatement measures.

Various other types of measurements can also be carried out, depending on the information sought :

  • Long-term measurements, using fixed devices, in order to give an indication of how noise nuisances change over time;
  • Medium-term measurements, in order to evaluate the improvement following a change to an infrastructure, an urban redevelopment, or following a change in legislation;
  • Short-term measurements, in order to characterise the noise environment in neighbourhoods or areas of particular interest (areas with critical noise levels, quiet areas, locations exposed to multiple sources of noise, etc.) or to quantify the impact of specific one-off events.

In order to complement the physical diagnostics established by measurement and modelling, surveys of the local population can also be conducted, in order to take into account the sociological and subjective nature of noise.

The issue of the indicators used to represent the characteristics of noise over a given period of time, or to reflect - insofar as possible - people's perception of their noise environment, appears to be the key to the public and authorities taking ownership of the noise issue. There are currently two main types of indicators:

- so-called "energy-based" indicators, which represent the average noise energy levels over a given period. The most well-know energy-based indicator is the LAeq (equivalent continuous sound level expressed in dB(A)), which represents the average noise level over a given period. Nevertheless, two identical noises will be perceived differently based on whether they are heard during the day or at night. It was, therefore, decided to create an indicator that is harmonised at European level, which takes all factorsXXX and this difference in perception into account: the Lden. This indicator is calculated using equivalent levels at three different periods during the day: day, evening, and night, which are weighted to take into account increased sensitivity at certain times of day.

- so-called "event-based" indicators that focus on noise peaks that occur over a given period. In some cases (air traffic, for example) they allow a better consideration of annoyance and health effects caused by the repetitive nature of noise events and their emergence from background noise. The best-known event-based indicators are the number of events whose maximum level exceeds a certain threshold (NAthreshold).

There is, however, a need to reconcile these two approaches, in order to take into account both the overall environmental noise "load", which is relatively well represented by energy-based indicators, and noise peak-type disturbances, which come on top of the background noise and whose impact in terms of annoyance and health effects are more is more widely recognised.

The Harmonica index was developed in order to meet this need.