Implementing the Harmonica Index in the Dynamap project L. Peruzzi, P. Bellucci, G. Brambilla, G. Zambon, A. Bisceglie ANAS S.p.A. - Research and Development Area

MAIN AIM AND OUTCOME OF THE DYNAMAP PROJECT the provision of dynamic noise maps able to detect and represent in real time the acoustic climate of road infrastructures. to ease and streamline the noise mapping process through the development of a low cost sensors system able to automatically update the noise maps in real time. Secondary objectives Demonstrate that dynamic mapping can be applied to monitor and report information about other environmental parameters Develop a user-friendly software application to facilitate public information and communication Test the public response and user ability in consulting and managing the system

LIFE HARMONICA PROJECT PUBLIC INFORMATION AND COMMUNICATIONS Chance of attracting the public interest USER FRIENDLY INTERFACE AND TOOLS LIFE HARMONICA PROJECT Innovative tools to better inform the public about environmental noise and to help local authorities in the decision making process Creation of a simple, dimensionless noise index Database on noise abatement actions published on an interactive platform to share useful information and experiences. EASY WAY TO PRESENT NOISE MAPS TO THE PUBLIC foreseen in the project and defined in the specifications of the system HARMONICA INDEX specifically designed to improve public comprehension on noise issues and impacts HARMONICA + DYNAMAP Natural solution

BGN EVT HARMONICA INDEX HARMONICA INDEX - Analytical formulation To improve public comprehension on noise issues and impacts, Harmonica suggests the creation of a simple noise index, taking into account the main components that influence noise perception, namely the background noise and the peaks related to noise events emerging from background noise. The mathematical formulation of the Harmonica index is formed by two terms: The component related to the background noise BGN = 0,2 (LA95eq – 30) BGN EVT HARMONICA INDEX The peaks related to noise events emerging from background EVT = 0,25 (LAeq - LA95eq) LA95eq is the equivalent background noise level over an hour period, the background noise being evaluated every second by the noise level exceeded 95% of the time during the 10 minutes period before; LAeq is the hourly equivalent noise level.

Graphical representation HARMONICA INDEX Graphical representation The Index quantifies the noise impact on an hourly basis using a linear scale ranging from 0 up to 10, rounded to 1 decimal place. Event-based component , related to the noise dynamics and the number of noise peaks that emerge above background The higher the score, the poorer is the noise environment Background noise

the critical thresholds for exposure to noise HARMONICA INDEX Graphical representation The noise impact is also emphasized by the adoption of a 3 colour scale that takes into account the time of day in order to reflect people increased sensitive to noise perception at night. COLOR SCALE DAY NIGHT Correspond to < 4 < 3 values at which the WHO indicates the noise is likely to provide sleep disturbance 4 ≤ HI < 8 3 ≤ HI < 7 ≥ 8 ≥ 7 the critical thresholds for exposure to noise

APPLICATION OF THE HARMONICA INDEX In order to check the suitabiltiy of the Harmonica Index to the Dynamap system, the index was tested to investigate its potential as simplified indicator for real time noise maps. MILAN ROME The two pilot areas show peculiar territorial and environmental characteristics, that affect the way noise impacts are calculated and perceived by people, as well as the way information can be transferred to the general public.

Along the Ring Road (A90) surrounding the city PILOT AREA OF ROME Along the Ring Road (A90) surrounding the city 6 lanes, 68 km long, skirting many suburban areas where noise levels were found to impact critically on residents The area is characterized by the presence of single or multiple noise sources, such as railways, crossing and parallel roads. ROME In order to check the suitability of the Harmonica Index in real time noise mapping, two test sites with different traffic behavior were chosen. Data collected during a preliminary monitoring campaign, performed to calibrate the source model, were used for testing.

The noise dynamics varies from 30 to 40 dB during the day PILOT AREA OF ROME Noise trends at the two test sites for 24 h Noise trends are calculated as instant and hourly Laeq and have been analyzed in order to identify the dynamics of sound pressure levels. Both were calculated as mobile average starting from a sampling frequency of 100 ms. The graphs also report the value of LA95,eq, with a mobile statistics over a 1-hour period. 30-40 dB 40-50 dB The noise dynamics varies from 30 to 40 dB during the day and from 40 to 50 dB at night The signal dynamics evaluation is important to understand if the indicator is suitable to represent the variability of noise over time.

Dynamic Static a static representation of the noise impact, The Harmonica Index has a similar trend to hourly Laeq, but with a narrow dynamics of about 0.5 at daytime and 1.0 at night This leads, on the basis of the Harmonica Index nuisance colour scale, to a static representation of the noise impact, regardless the actual dynamic trend of the sound pressure level. In order to evaluate if the index dynamics increases, the color scale resolution was set to a 1 point steps. But as you can see the trend of the index always is within the same band and therefore it is static

PILOT AREA OF MILAN A portion of the whole municipality (District 9), selected through a multicriteria procedure The northern part of Milan, population of about 180.000 residents, mostly annoyed by road traffic noise This area includes many different noise sources: a complex road network, made of major and secondary roads, anthropic activities, etc. A monitoring network has been planned to be installed in the framework of the Dynamap project to detect local noise levels trend and to update noise maps all over the area.

than at the first site (major road). PILOT AREA OF MILAN Noise trends at the two test sites Primary Road Two test sites with different traffic behaviour were chosen for the application of the Harmonica Index. Data were collected during a preliminary monitoring campaign performed to gather information on road noise within the pilot area. Noise trends were measured at the two test sites as Laeq,1sec and hourly Laeq, calculated with a mobile integration time, over a 24-hour period. The graphs also report the value of LA95,eq with a mobile statistics over a 1-hour period. Secondary Road The noise dynamics is wider at the second test site (secondary road), where traffic flow is lighter and less continuous , than at the first site (major road).

Dynamic Static HI has a similar trend to hourly LAeq, but with a narrow dynamics of about 2 - 2.5 The wider dynamics could be explained By the higher variability of the urban traffic flow and by the occurrence of significant anomalous events. As consequence of the mobile integration time used in the dynamic representation, the influence of anomalous noise events leads to a “stepped” curve that is not representative of the real noise trend.

QUITE NARROW DYNAMIC RANGE From 0.5 to 1 in Rome From 2 to 2.5 in Milan RESULTS ACHIEVED Harmonica index in its original formulation is unsuitable to real time noise maps. Further studies are in progress to check the applicability of the HI in dynamic mode. The original formulation has been modified by introducing an instant value that opportunely integrated in an hourly period provides the original value of the Harmonica Index. HARMONICA INDEX Instant Value HIinst = 0,2 (LA95 – 30) + 0,25(LAeq-LA95) LA95 is calculated starting from the instant values of Laeq, 1s included in a mobile temporal window of ten minutes updated with a frequency of 1 second.

HARMONICA INDEX - Instant Value In this way it is possible to report the instant fluctuations of the index, that now show a wider dynamics, and calculate the standard Harmonica Index by integrating the single terms in the hourly period as reported in the original formulation of the index. Istant trend of the Harmonica Index related to an hour period and the corresponding dynamic hourly value.

less than 60 seconds HARMONICA INDEX - Instant Value The dynamics of the instant index depends on noise levels variations and on the integration time used to update data. The instant index can be applied only in real time applications when the updating frequency is less than 60 seconds

CONCLUSIONS Traditional indicators are not easy to be understood for the public and not suitable for dynamic representations The Harmonica index, using a linear scale, is more immediate, but its dynamic is too compressed: that does not make it suitable for real-time applications. The instant value of the Harmonica Index can be applied to real time noise maps only if the integration time is less than 60 s. Larger integration times reduce the dynamics of the signal and consequently limit changes in the noise impact representation.

Thank you for your kind attention www.life-dynamap.eu