Surface Scattering, Diffusion and In-situ Absorption #
How a surface returns incident sound — how much it scatters away from the specular direction, how uniformly it spreads what it scatters, and how much it absorbs — is measured by a family of dedicated methods. The reverberation room gives the random-incidence scattering coefficient of a surface by comparing decays with the sample held still and rotating (ISO 17497-1). A free-field goniometer measures the polar response of the reflected sound and condenses it into a diffusion coefficient (ISO 17497-2). And out on a road, a loudspeaker and a single microphone recover the in-situ absorption of the pavement, either over an extended surface by subtracting the incident wave (ISO 13472-1) or through a small tube pressed onto the surface (ISO 13472-2). This page covers all four.
The scattering and diffusion coefficients answer different questions and are not interchangeable: scattering is how much energy leaves the specular direction; diffusion is how evenly the reflected energy is spread over angle.
1. Random-incidence scattering coefficient (ISO 17497-1) #
The scattering coefficient $s$ is the fraction of reflected energy that does not leave the surface in the specular direction. ISO 17497-1 measures it in a reverberation room from four reverberation-time situations: with the test sample mounted on a turntable and held stationary, and with the turntable rotating (which averages the phase-coherent specular reflection away), each with and without a reflecting base plate.
Absorption from reverberation time (Clause 6). Each situation converts to a Sabine absorption coefficient with the standard's own air-attenuation term:
$$ \alpha = 55.3,\frac{V}{S}\left(\frac{1}{c_2 T_2} - \frac{1}{c_1 T_1}\right) - 4,\frac{V}{S},(m_2 - m_1), $$
where $V$ is the room volume, $S$ the sample area, $T$ the reverberation time, $c$ the speed of sound (Eq. (2): $c = 343.2\sqrt{(273.15+t)/293.15}$) and $m$ the power attenuation coefficient of air. The stationary pair gives the random-incidence absorption $\alpha_s$ (Eq. (1)); the rotating pair gives the specular absorption $\alpha_{spec}$ (Eq. (4)).
Scattering coefficient (Eq. (5)). The two combine into
$$ s = \frac{\alpha_{spec} - \alpha_s}{1 - \alpha_s}. $$
A fully specular surface reflects all its non-absorbed energy in the specular direction, so $\alpha_{spec} = \alpha_s$ and $s = 0$; a strong diffuser sends energy everywhere, raising $\alpha_{spec}$ towards 1 and $s$ towards 1.
import phonometry as ph
# Four reverberation-time situations reduced to two absorption coefficients.
# alpha_s from the stationary pair (Eq. 1); alpha_spec from the rotating pair
# (Eq. 4). V = 200 m^3, S = 10 m^2, c = 343.2 m/s throughout.
alpha_s = ph.random_incidence_absorption(200.0, 10.0, c1=343.2, T1=8.0,
c2=343.2, T2=6.0)
alpha_spec = ph.specular_absorption_coefficient(200.0, 10.0, c3=343.2, T3=7.5,
c4=343.2, T4=5.0)
s = ph.scattering_coefficient(alpha_spec, alpha_s) # Eq. (5)
print(round(float(alpha_s), 4)) # 0.1343
print(round(float(alpha_spec), 4)) # 0.2148
print(round(float(s), 4)) # 0.0931
Over a full one-third-octave measurement, scattering_coefficient_spectrum
pairs the per-band $\alpha_{spec}$ and $\alpha_s$ with their band centres and
returns a plottable ScatteringResult:
import numpy as np
import phonometry as ph
# A 13-band measurement (250-4000 Hz): the random-incidence absorption alpha_s
# (stationary sample) and the specular absorption alpha_spec (rotating
# turntable). A diffuser scatters more with frequency, so s(f) rises.
freqs = np.array([250, 315, 400, 500, 630, 800, 1000,
1250, 1600, 2000, 2500, 3150, 4000], float)
alpha_s = np.full_like(freqs, 0.10)
alpha_spec = 0.11 + 0.75 * (np.log10(freqs / 250) / np.log10(4000 / 250))
result = ph.scattering_coefficient_spectrum(freqs, alpha_spec, alpha_s)
print(np.round(result.scattering[[0, 6, 12]], 3)) # [0.011 0.428 0.844]
result.plot() # s(f) on a log-frequency axis, 0 to 1 (needs matplotlib)
The scattering coefficient climbs with frequency: at low frequency the surface relief is small compared with the wavelength and the reflection stays specular ($s \to 0$); as the wavelength shrinks the relief scatters more energy out of the specular direction ($s \to 1$).
Show the code for this figure
import matplotlib.pyplot as plt
# result is the ScatteringResult computed above. One line:
result.plot()
plt.show()
# By hand, from the result's fields, mirroring what ScatteringResult.plot() draws:
fig, ax = plt.subplots()
ax.semilogx(result.frequencies, result.scattering, "o-", color="#1f77b4")
ax.set_xlabel("Frequency [Hz]")
ax.set_ylabel("Scattering coefficient s")
ax.set_ylim(0.0, 1.0)
ax.set_title("Random-incidence scattering coefficient (ISO 17497-1)")
plt.show()
Base-plate check (Clause 6.4, Table 1). The empty base plate must itself scatter only negligibly, or it would bias the result. ISO 17497-1 caps the base-plate scattering coefficient per one-third-octave band; the library exposes those limits and a checker.
from phonometry import (
BASE_PLATE_BANDS_HZ, BASE_PLATE_MAX_SCATTERING, check_base_plate_scattering,
)
# The normative per-band ceilings (Table 1): 0.05 up to 500 Hz, rising to 0.25.
# BASE_PLATE_BANDS_HZ is the band tuple; BASE_PLATE_MAX_SCATTERING maps band -> ceiling.
print(BASE_PLATE_BANDS_HZ[0], BASE_PLATE_MAX_SCATTERING[100]) # 100 0.05
# A base plate whose measured scattering stays under the ceiling passes silently;
# an over-limit band raises a ScatteringDiffusionWarning listing the offenders.
check_base_plate_scattering([0.02] * len(BASE_PLATE_BANDS_HZ))
2. Diffusion coefficient (ISO 17497-2) #
The diffusion coefficient $d$ measures the spatial uniformity of the reflected sound, not how much is scattered. A goniometer sweeps a receiver over a polar arc and records the reflected level $L_i$ at each angle; the coefficient is the normalised autocorrelation of the polar energy distribution.
Autocorrelation (Formula (5)). For $n$ receivers at equal angular spacing, with $p_i = 10^{L_i/10}$ the band energy at receiver $i$,
$$ d = \frac{\left(\sum_i p_i\right)^2 - \sum_i p_i^2} {(n-1),\sum_i p_i^2}. $$
A perfectly uniform polar response ($L_i$ all equal) gives $d = 1$; a single sharp specular lobe gives $d = 0$. When receivers subtend unequal solid angles, Formula (6) area-weights each energy by $N_i$ from Formula (8) — and those area factors are evaluated in radians, which is why a 5° spacing at the zenith produces a weight near 1.57, not 51.9.
import phonometry as ph
# Polar response of a diffuser (levels in dB at equally spaced receivers).
levels = [70.0, 74.0, 68.0, 72.0]
d = ph.directional_diffusion_coefficient(levels) # Formula (5)
print(round(float(d), 4)) # 0.7367
# Normalise against a flat reference surface to isolate the diffuser's effect
# (Formula (7)): d_n = (d - d_ref) / (1 - d_ref).
d_n = ph.normalized_diffusion_coefficient(d, 0.10)
print(round(float(d_n), 4)) # 0.7075
# Random-incidence value: average the band coefficients over source positions,
# with the standard's 2-D weighting (0 deg -> 1, +/-30/+/-60 deg -> 3).
from phonometry import TWO_DIMENSIONAL_SOURCE_WEIGHTS
d_random = ph.random_incidence_diffusion(
[0.5, 0.2, 0.2, 0.2, 0.2], weights=TWO_DIMENSIONAL_SOURCE_WEIGHTS)
print(round(float(d_random), 4)) # 0.2231
directional_diffusion keeps the receiver angles beside the levels of a full
goniometer sweep and returns a plottable DiffusionResult:
import numpy as np
import phonometry as ph
# A goniometer sweep of a diffusing surface: reflected levels L_i at 37
# receivers from -90 to 90 deg (5 deg spacing). The energy is spread almost
# uniformly over angle, so the Formula (5) coefficient d is high.
angles = np.arange(-90.0, 90.5, 5.0)
rng = np.random.default_rng(3)
levels = 70.0 + 2.0 * np.sin(np.radians(angles) * 3.0) + rng.normal(0.0, 1.0, angles.size)
result = ph.directional_diffusion(angles, levels)
print(round(result.coefficient, 2)) # 0.82
result.plot() # polar reflected response, d in the title (needs matplotlib)
The reflected energy fills the whole semicircle rather than concentrating in a specular lobe, so the autocorrelation diffusion coefficient is high ($d \approx 0.82$). A flat surface would collapse the response into a narrow spike and drive $d$ towards zero.
Show the code for this figure
import matplotlib.pyplot as plt
import numpy as np
# result is the DiffusionResult computed above. One line:
result.plot()
plt.show()
# By hand: a polar plot of the levels, with d annotated in the title.
fig, ax = plt.subplots(subplot_kw={"projection": "polar"})
theta = np.radians(result.angles)
ax.plot(theta, result.levels, "o-", color="#1f77b4")
ax.fill(theta, result.levels, color="#1f77b4", alpha=0.15)
ax.set_theta_zero_location("N")
ax.set_theta_direction(-1)
ax.set_thetamin(-90)
ax.set_thetamax(90)
ax.set_title(f"Directional diffusion d = {result.coefficient:.2f} (ISO 17497-2)")
plt.show()
3. In-situ road absorption — subtraction technique (ISO 13472-1) #
Out in the field there is no reverberation room. ISO 13472-1 measures the sound absorption of a road surface (or any extended flat surface) in situ by firing an impulse from a loudspeaker at height $d_s$ down onto the surface and recording the impulse response at a microphone at height $d_m$. The incident and reflected components are separated in time with an Adrienne window; their transfer function gives the reflection factor and hence the absorption.
Geometrical spreading (Clause 4.1). The reflected wave travels farther than the direct wave, so it is attenuated by the geometrical-spreading factor
$$ K_r = \frac{d_s - d_m}{d_s + d_m}, $$
which equals $2/3$ for the mandatory geometry $d_s = 1.25$ m, $d_m = 0.25$ m. The absorption follows from the windowed incident and reflected spectra $H_i$, $H_r$:
$$ \alpha(f) = 1 - \frac{1}{K_r^2}\left|\frac{H_r(f)}{H_i(f)}\right|^2. $$
import numpy as np
import phonometry as ph
# A band-limited incident impulse response and a synthetic road reflection
# hr = Kr * r0 * delayed(hi): a reflection of magnitude r0 = 0.4, delayed by the
# extra path, and scaled by the geometrical-spreading factor Kr.
fs, n = 48000.0, 4096
t = np.arange(n) / fs
hi = np.zeros(n)
hi[:64] = np.hanning(64) * np.cos(2.0 * np.pi * 1500.0 * t[:64])
kr = ph.geometric_spreading_factor() # (ds - dm)/(ds + dm) = 2/3
hr = kr * 0.4 * np.roll(hi, 96)
# Narrow-band absorption, then reduced to one-third octaves over 250-4000 Hz.
alpha = ph.insitu_absorption_coefficient(hi, hr) # 1 - (1/Kr^2)|Hr/Hi|^2
freq = np.fft.rfftfreq(n, 1.0 / fs)
centres, band = ph.one_third_octave_absorption(freq, alpha)
print(round(kr, 4)) # 0.6667
print(round(float(band[2]), 3)) # 0.84 (alpha = 1 - 0.4^2 = 0.84)
Adrienne window (Clause 6.4). The time window that isolates the reflection mandates only a sharp leading edge, a 5 ms flat portion and a cosine-squared or Blackman-Harris trailing edge — the exact durations are reported per measurement, not fixed, so they are configurable here.
from phonometry import adrienne_window
# Default: 0.5 ms leading edge, 5 ms flat top, 5 ms Blackman-Harris trailing.
w = adrienne_window(48000.0)
print(w.shape[0]) # 504 samples at 48 kHz
print(round(float(w.max()), 3)) # 1.0 (flat top and edges meet at unity)
End-to-end spectrum. insitu_absorption_spectrum runs the whole chain — the
windowed incident and reflected impulse responses to the narrow-band absorption
and on to one-third-octave bands — and returns a plottable
InsituAbsorptionResult:
import numpy as np
import phonometry as ph
from scipy.signal import firwin, lfilter
# A synthetic-but-realistic measurement. hi is a unit incident impulse; the road
# reflection hr = Kr * r0 * roll(hi, shift) uses the geometrical-spreading
# factor Kr, a mildly frequency-dependent r0 (a gentle low-pass, so a porous
# surface reflects less as frequency rises) and the reflected-path delay
# shift = round(2 dm / c * fs).
fs, n = 48000.0, 8192
kr = ph.geometric_spreading_factor() # (ds - dm)/(ds + dm) = 2/3
hi = np.zeros(n)
hi[0] = 1.0
taps = firwin(41, 1200.0, fs=fs)
taps = taps / taps.sum()
shift = int(round(2.0 * 0.25 / 340.0 * fs)) # reflected-path delay 2 dm / c
hr = kr * 0.85 * np.roll(lfilter(taps, 1.0, hi), shift)
result = ph.insitu_absorption_spectrum(hi, hr, fs)
print(result.frequencies[[0, -1]].astype(int)) # [ 250 4000]
print(np.round(result.absorption[[0, 6, 12]], 2)) # [0.31 0.65 1. ]
result.plot() # alpha(f) bar chart over 250-4000 Hz (needs matplotlib)
The absorption rises with frequency because the surface reflects less of the high-frequency energy, exactly as the low-pass reflection factor $r_0(f)$ dictates through $\alpha = 1 - (1/K_r^2),|H_r/H_i|^2$.
Show the code for this figure
import matplotlib.pyplot as plt
import numpy as np
# result is the InsituAbsorptionResult computed above. One line:
result.plot()
plt.show()
# By hand: a bar chart of alpha over the one-third-octave bands.
freqs = result.frequencies
positions = np.arange(freqs.size)
fig, ax = plt.subplots()
ax.bar(positions, np.nan_to_num(result.absorption), width=0.7, color="#1f77b4")
ax.set_xticks(positions)
ax.set_xticklabels([f"{f:g}" for f in freqs], rotation=45, ha="right")
ax.set_xlabel("Frequency [Hz]")
ax.set_ylabel("Absorption coefficient alpha")
ax.set_ylim(0.0, 1.0)
ax.set_title("In-situ road-surface absorption (ISO 13472-1)")
plt.show()
Maximum sampled area (Annex A). The finite time window limits how much of the surface contributes to the reflection. The maximum sampled area is a circle whose radius the library computes from the geometry and window width; the Annex A worked example ($d_s = 1.25$ m, $d_m = 0.25$ m, $c = 340$ m/s, 5 ms flat window) gives about 1.34 m.
import phonometry as ph
print(round(ph.max_sampled_area_radius(5.0e-3), 3)) # 1.343 (metres)
4. In-situ road absorption — spot method (ISO 13472-2) #
For smaller patches, ISO 13472-2 seals a short circular tube onto the surface and
measures the absorption with the two-microphone transfer-function method of
ISO 10534-2. The library provides the spot-method geometry and validity helpers;
the transfer-function DSP itself is the impedance-tube routine
two_microphone_impedance (see Acoustic Materials).
Plane-wave limits (Clause 5.4). The tube supports only plane waves below
$$ f_u = 0.58,\frac{c_0}{d}, $$
with $d$ the tube diameter, and the microphone spacing $s$ must sit between $0.05,c_0/f_{min}$ and $0.45,c_0/f_{max}$. The reported range is the one-third-octave bands 250–1600 Hz.
import phonometry as ph
# Upper usable frequency of a 100 mm tube and the valid spacing window.
print(round(ph.spot_tube_upper_frequency(0.100, 343.0), 1)) # 1989.4 Hz
s_min, s_max = ph.spot_microphone_spacing_bounds(
343.0, f_min=220.0, f_max=1800.0)
print(round(s_min, 3), round(s_max, 3)) # 0.078 0.086 (metres)
Standards implemented on this page: ISO 17497-1:2004 (scattering coefficient), ISO 17497-2:2012 (diffusion coefficient), ISO 13472-1:2002 (in-situ absorption, extended surface), ISO 13472-2:2010 (in-situ absorption, spot method). Numerical conformance against the standards' worked examples and closed forms is tracked in CONFORMANCE.md.