A study on variations in excess attenuation due to ground surface and meteorological conditions based on a long-term outdoor sound propagation experiment

From the winter of 2018, outdoor sound propagation experiments (maximum horizontal range: 300 m) have been repeatedly conducted three times a day on weekdays at a glider airfield in Hokkaido, Japan. The ground condition of the experimental field is grass-covered in summer and snow-covered in winter. In each experiment, impulse responses have been measured by time-stretched pulse method and excess attenuation has been obtained at receiving points. Meteorological data at the field has been also measured. Based on the data of excess attenuation collected under various meteorological conditions over a long period, variation in sound propagation characteristics due to the differences in ground surface condition and meteorological condition has been investigated. The numerical analysis based on the GFPE method has been also carried out with changing the parameter of meteorological condition and ground surface condition. By comparing the results with the experimental data, the prediction method of the variations in excess attenuation has been also investigated.

A comparison of the ground excess attenuation model in Harmonoise with finite-difference time-domain solutions under grounds with mixed types

Total noise exposure is calculated for the evaluation of health effects caused by environmental noise. For the calculation, computationally drawn noise maps are used. In the computation process, sound propagation over ground surface with mixed types should be calculated for better accuracy. One engineering model that allows such calculation is the ground excess attenuation model of the Harmonoise model. However, the applicability of the model to such complex grounds remains unclear. In this study, a 40m-length ground surface with a discontinuity in flow resistivity is defined. By moving the discontinuity position, sound propagation from a point source and a receiver at each end is calculated using the model and a numerical method. The numerical method is the finite-difference time-domain method with porous medium modeling that has been proven to be accurate. It is found from the numerical results that in higher frequencies the excess attenuations in terms of the discontinuity position have fluctuations. The fluctuations are found to correspond to the interference by diffraction path difference passing the discontinuity. In contrast, the model results exhibit smooth transition from an extremity of single flow resistivity surface to another. A simple model of such diffraction needs to be developed.