In the perception of people affected, noise exposure is strongly dependent on current weather conditions. Throughout Europe there are about a dozen standards and calculation methods for determining sound propagation, ranging from very simple methods without consideration of meteorological influences, through sets of rules with limited consideration of meteorological parameters, to complicated methods with many, sometimes very specific input parameters. In Germany, mandatory calculation methods (RLS-90, RLS-19) are used, where meteorological influences have not been sufficiently considered so far.
On behalf of the Federal Highway Research Institute (BAST), the project entitled "Development of a database for the calculation of exemplary noise situations including noise emission data of the road transport mode and meteorological data" was carried out in cooperation with the engineering office Möhler + Partner Ingenieure AG (Augsburg).
The aim of the research project 'Noise weather' was therefore to determine the state of the art and science in the consideration of the influence of weather on sound propagation in computational methods and to verify it by long-term measurements. These investigations have now been successfully completed. From model calculations and the measurements, a practical and simple procedure for considering the influence of meteorology for calculations according to RLS-90 or RLS-19 was proposed without changing the basic approach of the procedures. However, further validations in different regions are necessary to secure this proposal.
A central part of the study were on the one hand long-term measurements in a study area at the federal highway A8 west of Munich, where besides the sound measurements also meteorological data as well as traffic data were collected. On the other hand, selected situations were recalculated using various methods and compared with the real measurement data.
Both measurements and advanced calculation methods showed consistently that wind direction has the most significant influence on sound propagation. The wind speed has less influence. The influence of the vertical temperature gradient, which determines the stability of the air stratification, was also investigated. This only becomes relevant for sound propagation at large distances and in connection with shielding.
Figure 1: Level differences with respect to the reference sound level for wind speeds between 1.0 and 3.0 m/s. (Graphics: ©DLR)
Figure 2: Comparison of the decrease of the sound level with the distance for different calculation methods, each for homogeneous and for favorable conditions for sound propagation. (Graphics: ©DLR)
Figure 3: Level differences to the reference sound level for wind speeds between 1 and 3 m/s, both for headwind (MW) and downwind (GW), determined with the IPA - wave-based model. (Graphics: ©DLR)
Figure 1 shows the dependence of the sound level on the wind direction and the temperature stratification (S1 - S5) for measurements at a distance of approx. 250m from the motorway. It can be seen that under headwind conditions (180°) the strongest level reduction occurs especially under unstable stratification (S1/S2, turquoise and red curve) and a level increase was observed under stable conditions (S5, yellow curve) independent of the wind direction. Thus, sound propagation favorable conditions such as downwind and stable stratification can be identified in contrast to a reference case without wind and with homogeneous (isothermal) temperature stratification.
In order to exactly understand the effects of the different calculation methods for the propagation calculation under variation of meteorology, three additional test cases of simplest possible geometry (plane) were defined: first, a straight road, second, a straight road with shielding, and third, two intersecting roads. For these idealized cases and the realistic case all available calculation methods were compared and the meteorological influence was quantified. Figure 2 demonstrates how much the calculation methods differ even for the simplest case "straight road without topography" with increasing distance to the source. By adding the more accurate models (a) wave-based model of the IPA and (b) Nord2000, the influence of meteorology could be reproduced. As an example, Figure 3 shows in a vertical section how much the sound field changes under headwind conditions compared to a reference case.
On the basis of these manifold results and considering the special measurement conditions of this study, a basis for a more detailed consideration of the influences of wind and temperature stratification within the RLS-90/RLS-19 was then created, which still has to prove its resilience in further investigations. It was taken care that the basic approach of the methods remains unchanged and that the calculation effort is only insignificantly increased by the new method.