Laboratory and outdoor measurements are reported of the relative sound pressure level spectrum over hard surfaces containing either random or periodically spaced arrays of 2-D roughnesses. The resulting data have been compared with predictions obtained analytically and with numerical predictions of a boundary element code. Effective impedances of the rough surfaces have been calculated from the boss theory developed by Twersky. A classical asymptotic approximation for propagation near grazing incidence from a point source over an impedance boundary has been modified, heuristically, to allow for diffraction grating effects. The resulting predictions are found to be in tolerable agreement with the data except for close and random packing. The boundary element code is found to give superior results for larger roughnesses, but computational restrictions on element size reduce its usefulness for roughnesses with small width or height.

Two methods of calculating the effective impedance spectra of acoustically hard, randomly rough, two-dimensional surfaces valid for acoustic wavelengths large compared with the roughness scales have been explored. The first method uses the complex excess attenuation spectrum due to a point source above a rough boundary predicted by a boundary element method (BEM) and solves for effective impedance roots identified by a winding number integral method. The second method is based on an analytical theory in which the contributions from random distributions of surface scatterers are summed to obtain the total scattered field. Effective impedance spectra deduced from measurements of the complex excess attenuation above 2D randomly rough surfaces formed by semicylinders and wedges have been compared to predictions from the two approaches. Although the analytical theory gives relatively poor predictions, BEM-deduced effective impedance spectra agree tolerably well with measured data. Simple polynomials have been found to fit BEM-deduced spectra for surfaces formed by intersecting parabolas corresponding to average roughness heights between 0.25 and 7.5 m and for five incidence angles for each average height. Predicted effects of sea-surface roughness on sonic boom profiles and rise time are comparable to those due to turbulence and molecular relaxation effects.

Measurements of the excess attenuation of sound from a point source over mixed impedance ground in an anechoic chamber are compared with predictions obtained from models based on (a) the semiempirical theory due to De Jong, (b) Nyberg’s theory, (c) a Fresnel-zone approximation, and (d) a boundary element code. The impedance discontinuities studied in this work are perpendicular to the source–receiver line. When there is a single discontinuity between acoustically hard and finite impedance surfaces, the De Jong semiempirical model, the Fresnel-zone model and boundary element code are found to give satisfactory agreement with measured data. The frequency of the first maximum in attenuation is found to be highest at approximately 70% hard surface cover rather than at the 100% expected. It is argued that this is a result of edge diffraction. When extended to multiple impedance discontinuities, the De Jong semiempirical model performs poorly. However, both Nyberg’s theory and the boundary element code give good agreement with measured results, and the Fresnel-zone model gives qualitative agreement with the measured data. The measurements confirm that, under certain constraints given by Nyberg’s theory, the ground effect due to mixed impedance may be determined from that predicted by using the area-average impedance.

A “boss” formulation by Twersky [J. Acoust. Soc. Am. 73, 85–94 (1983)] enables prediction of the plane wave reflection coefficient from a surface composed of rigid-porous roughness elements embedded in an acoustically hard plane where the roughness elements and their mean spacing are small compared with the incident wavelengths. Predictions for air-filled porous roughness elements on a hard ground plane are compared with effective impedance spectra obtained from laboratory measurements over random distributions of polystyrene hemi-spheres, polyurethane pyramids, and sand hemispheroids on glass plates. Overall the predictions agree well with these data. To enable prediction of the effective admittance of rough porous surfaces, Twersky’s original formulation is extended heuristically. The resulting theory is compared with a previous model [J. Acoust. Soc. Am. 108, 949–956 (2000)], which is a heuristic extension of Tolstoy’s theory [J. Acoust. Soc. Am. 72, 960–972 (1982)] to include nonspecular scattering. The theories are found to give different predictions for relatively large bosses. The modified Twersky theory gives relatively good predictions of the effective impedance spectra obtained from complex sound pressure level measurements over sand surfaces containing semielliptical roughness elements and over uncultivated soil.

a b s t r a c tMeasurements of seismic signatures produced by airborne, near-surface detonations of explosive charges over a variety of ground types show two distinct ground vibration arrivals. In all cases, the earlier arrival (precursor), has a time of arrival consistent with a predominantly underground path and coupling of blast sound to the ground close to the source and is always much smaller than the later vibration, the time of arrival of which is consistent with coupling from the air blast arrival at the receiver. The ratio of the seismic particle velocity to the acoustic pressure at the surface for the air-coupled seismic wave is constant with respect to distance and maximum pressure at a given location, but varies from site to site, with values usually between 1 and 13 lm s A numerical code enabling calculations of the fields due to an impulsive source above a layered poroelastic ground is described. Predictions of the air pressure spectrum above ground and the vertical and radial components of solid particle velocity near the ground surface are found to compare tolerably well with the measured spectra and waveforms of acoustic and seismic pulses at about 100 m range in seismically-hard and -soft soils and with a snow cover present. The predicted seismic responses in 'soft' soil confirm that the existence of a near-surface S-wave speed less than that in air is responsible for the observed 'ringing', i.e. a long low-frequency wavetrain associated with coupling to the dispersive Rayleigh wave. The predicted seismic pulses in the presence of the shallow snow cover explain the observed phenomenon whereby a high frequency ground vibration is modulated by a lower frequency layer resonance.An empirical equation relating ground vibration from explosions to distance predicts that the commonly-used vibrational damage peak velocity criterion of 12 or 25 mm s À1 will be exceeded when the peak positive pressure exceeds 480 Pa (147.6 dB) or 1 kPa (154.0 dB), respectively. Either of these levels is much higher than the current U.S. Army overpressure damage criterion of 159 Pa (138 dB). Thus in most situations damage from blast overpressure will occur long before damaging levels of ground vibration are reached, so it is likely that civilian perceptions of vibration are produced by coupling from the airblast.Published by Elsevier Ltd.