# Theory: Environment and Transport

> One page of the phonometry documentation, as Markdown. Index: https://docs.jmrp.io/phonometry/llms.txt

Canonical: https://docs.jmrp.io/phonometry/reference/theory/environment-transport/
Language: en
Alternate: https://docs.jmrp.io/phonometry/es/reference/theory/environment-transport/index.md
Version: phonometry 4.0.0
Install: pip install "phonometry[full]"
Author: José Manuel Requena Plens (https://orcid.org/0000-0003-1250-6212)
Published: 2026-07-15
Updated: 2026-11-16T21:36:13Z
Standards: ISO 1996-1:2016; NT ACOU 112:2002; ISO 9613-1:1993; ISO 9613-2:1996; ISO 9612:2009; ISO 3744:2010; ISO 3745:2012; ISO 3741:2010; ISO 9614-1:1993
Verified: 39 conformance checks (ISO 1996-1:2016; NT ACOU 112:2002; ISO 9613-1:1993; ISO 9613-2:1996; ISO 9612:2009; ISO 3744:2010; ISO 3745:2012; ISO 3741:2010; ISO 9614-1:1993) https://docs.jmrp.io/phonometry/reference/conformance/
License: MIT (https://github.com/jmrplens/phonometry/blob/main/LICENSE)
Summary: Environmental descriptors, impulsive-sound adjustments, outdoor propagation, occupational exposure and sound power determination behind phonometry.

---

This page collects the theory behind outdoor and environmental noise: the whole-day rating descriptors and the impulsive-sound adjustment, atmospheric absorption, the general outdoor propagation method, occupational noise exposure with its uncertainty budget, and the sound power determination methods. It is part of the [theory reference](https://docs.jmrp.io/phonometry/reference/theory/).

Each section keeps its own standard's notation, and four letters are reused
across them. Read them per section:

| Symbol | In the propagation sections | In the sound-power sections |
| :--- | :--- | :--- |
| $A$ | Octave-band attenuation, in dB (ISO 9613-2) | Equivalent absorption area of the room, in m² |
| $A_0$ | — | Reference absorption area, 1 m²; except in the ISO 3745 air term, where $A_0 = a(f)\,r$ is a dimensionless absorption product |
| $C_1$, $C_2$ | Barrier constants in $D_z$ (ISO 9613-2) | Meteorological corrections (ISO 3745, ISO 3741) |
| $K$ | — | $K_1$ background, $K_2$ reverberant field, $K_\mathrm{I}$ impulse adjustment, and plain $K$ the bias-error factor of the intensity methods; $K_\mathrm{met}$ is the ISO 9613-2 barrier meteorological factor |

## Environmental descriptors (ISO 1996-1)

The **day-evening-night level** $L_\mathrm{den}$ (ISO 1996-1:2016, 3.6.4) is an energy average over the 24 h day with penalty weightings of **+5 dB for the evening** and **+10 dB for the night**:

$$
L_\mathrm{den} = 10 \log_{10}\left\lbrace\frac{1}{24}\left[ t_\mathrm{d}\ 10^{0.1 L_\mathrm{day}} + t_\mathrm{e}\ 10^{0.1 (L_\mathrm{evening} + 5)} + t_\mathrm{n}\ 10^{0.1 (L_\mathrm{night} + 10)} \right]\right\rbrace
$$

with default period durations $(t_\mathrm{d}, t_\mathrm{e}, t_\mathrm{n}) = (12, 4, 8)$ h; countries may define the periods differently (3.6.4 Note 1). The **day-night level** $L_\mathrm{dn}$ (3.6.5) drops the evening period:

$$
L_\mathrm{dn} = 10 \log_{10}\left\lbrace\frac{1}{24}\left[ t_\mathrm{d}\ 10^{0.1 L_\mathrm{day}} + t_\mathrm{n}\ 10^{0.1 (L_\mathrm{night} + 10)} \right]\right\rbrace, \qquad (t_\mathrm{d}, t_\mathrm{n}) = (15, 9)\ \text{h}
$$

Both are special cases of the **composite whole-day rating level** (6.5, generalizing Formulae 5–6), where each period $i$ contributes its rating level $L_i$ plus an adjustment $K_i$, weighted by its share of the day:

$$
L_\mathrm{R} = 10 \log_{10}\left[ \sum_i \frac{h_i}{24}\ 10^{0.1 (L_i + K_i)} \right], \qquad \sum_i h_i = 24\ \text{h}
$$

The adjustments $K_i$ cover time-of-day penalties (ISO 1996-1 Table A.1: evening 5 dB, night 10 dB) as well as source-character adjustments (e.g. tonal penalties), which the ECMA-418-1 TNR/PR assessments can justify objectively.

**What the penalties are, and what the number means.** The +5 and +10 dB are not
physics: nothing about the sound changes at 19:00. They encode a dose-response
finding: the same acoustic energy is reported as more annoying when it
interrupts an evening at home or a night's sleep, so $L_\mathrm{den}$ is an
*annoyance* indicator built out of acoustic levels and must never be read back
as one. On the scale it lives on, a rural night sits in the 40s, a quiet
residential street in the mid-50s and a façade on a busy urban arterial in the
high 60s, and the Environmental Noise Directive reports strategic maps in 5 dB
$L_\mathrm{den}$ bands from 55 dB upward. The weighting also decides where a mitigation
is worth applying: an hour of night enters with $10 \times 8/24$ against the
day's $12/24$, so a decibel bought at night is worth about 6.7 times a decibel
bought during the day, but only once the night is loud enough to matter, and
the crossover is arithmetic. The night term overtakes the day term when the
night level comes within about 8 dB of the day level; in the profile below the
night is 14.4 dB below the day, so it contributes only 12.5 % of the total
against the day's 51.3 %, and 3 dB off the day moves $L_\mathrm{den}$ by 1.3 dB where
3 dB off the night moves it by 0.3 dB. Check the split before choosing where to
spend.

See the [Environmental Levels guide](https://docs.jmrp.io/phonometry/environment/assessment/environmental-levels/) for usage.

![Synthetic 24-hour urban LAeq profile with day, evening and night bands, the +5 and +10 dB weighted period levels and the resulting Lden](https://docs.jmrp.io/phonometry/media/lden_profile.svg)

*A 24-hour $L_\mathrm{Aeq}$ profile split into day, evening and night, the +5/+10 dB
penalties and the resulting $L_\mathrm{den}$ of 64.3 dB, an urban arterial façade,
squarely inside the range the Directive maps. The day averages 64.4 dB, the
evening 62.6 dB and the night 50.0 dB; the penalties lift the evening and night
lines above their own levels, and $L_\mathrm{den}$ still lands within 0.1 dB of the
day, because a night that quiet cannot move it.*

## Impulsive-sound prominence (NT ACOU 112)

An impulse annoys beyond its energy, so environmental surveys after ISO 1996-2 penalize periods containing prominent impulsive sounds; NT ACOU 112:2002 makes that penalty objective. From the A-weighted, time-weighting-F level history of a single event, the onset rate (dB/s) and the level difference (dB) of the onset (which qualifies when steeper than 10 dB/s, clauses 4.5–4.7) predict the perceived prominence (clause 7, Formula 1):

$$
P = 3 \log_{10}(\text{onset rate}) + 2 \log_{10}(\text{level difference}),
$$

designed to peak around 15 for very sudden, loud impulses.

![A-weighted Fast level history of three hammer strikes over a 55 dB(A) background across six seconds: each strike rises from about 52 dB to 89 dB, the detected onset start and end points are marked with the least-squares onset line, the governing level difference of 36.8 dB is annotated, and the title reports a prominence of 11.34 with an adjustment of 11.42 dB, category highly impulsive](https://docs.jmrp.io/phonometry/media/impulsive_sound_onsets.svg)

*Both inputs of $P$ are geometry on this trace, which is why the method needs a
level history and not a level. The onset rate is the slope of the fitted line
through the rise, in dB/s, and the qualifying threshold of 10 dB/s is a
steepness on this axis; the level difference is the height of the same rise.
Three strikes are detected here and only the steepest-and-tallest one governs
the adjustment.*

The adjustment to the measurement-period level takes the governing (highest-$P$) impulse (clause 8, Formula 2):

$$
K_\mathrm{I} = 1.8\ (P - 5)\ \text{dB} \quad (P > 5;\ \text{else } K_\mathrm{I} = 0),
$$

and the whole-day rating level combines the adjusted periods energetically (clause 8, Note 1):

$$
L_{\mathrm{Ar},T} = 10 \log_{10}\Big[ \frac{1}{T} \sum_N \Delta t_N\ 10^{(L_{\mathrm{Aeq},N} + K_{\mathrm{I},N})/10} \Big].
$$

$K_\mathrm{I}$ is exactly the kind of source-character adjustment that enters the ISO 1996-1 composite rating level above. The anchors $P(1000\ \text{dB/s}, 30\ \text{dB}) = 9 + 2\log_{10} 30 = 11.95$ and $K_\mathrm{I}(P{=}10) = 9.0$ dB are reproduced exactly.

**Which document to cite.** ISO has since published the same construction as
ISO/PAS 1996-3:2022, so a survey report may cite either; the prominence $P$ and
the adjustment $K_\mathrm{I}$ implemented here follow the NT ACOU 112 formulation that the
PAS reproduces. What requires an adjustment at all is ISO 1996-2:2017, the
measurement part: the objective procedure only decides how large the adjustment
is, and it is ISO 1996-1 that then combines the adjusted periods into a rating
level.

See the [Impulse Prominence guide](https://docs.jmrp.io/phonometry/environment/assessment/impulsive-sound/) for usage.

## Outdoor propagation (ISO 9613-1/2)

### Atmospheric absorption (ISO 9613-1)

Air is a lossy medium: a propagating tone loses energy to shear viscosity and
heat conduction (classical and rotational losses, growing as $f^2$) and to the
**vibrational relaxation** of the oxygen and nitrogen molecules, each an energy
reservoir that resonates near a humidity- and temperature-dependent relaxation
frequency. ISO 9613-1:1993, Eq. (5) gives the pure-tone attenuation coefficient
$\alpha$ in decibels per metre:

$$
\alpha = 8.686\ f^2 \Big[ 1.84\times10^{-11} \big(p_\mathrm{a}/p_\mathrm{r}\big)^{-1} \big(T/T_0\big)^{1/2}
       + \big(T/T_0\big)^{-5/2} \big( 0.01275\ \tfrac{\mathrm{e}^{-2239.1/T}}{f_\mathrm{rO} + f^2/f_\mathrm{rO}}
       + 0.1068\ \tfrac{\mathrm{e}^{-3352.0/T}}{f_\mathrm{rN} + f^2/f_\mathrm{rN}} \big) \Big],
$$

with the oxygen and nitrogen relaxation frequencies $f_\mathrm{rO}$, $f_\mathrm{rN}$ of
Eq. (3)/(4), the reference conditions $T_0 = 293.15$ K, $p_\mathrm{r} = 101.325$ kPa
(Clause 4.2) and the molar water-vapour concentration $h$ from the relative
humidity (Annex B). At low frequency $\alpha \propto f^2$; near each relaxation
frequency the corresponding term peaks and rolls off, which is why $\alpha$ rises
by two decades from 50 Hz to 10 kHz and why raising the humidity sweeps a peak
across the band. The library reproduces Table 1 to under 0.4 % (the standard's
own printed precision), well inside its stated $\pm 10$ %; passing
`exact_midband=True` snaps each frequency onto the exact midbands
$f_\mathrm{m} = 1000 \cdot 10^{k/10}$ (Note 5) used to compute that table. The same
$\alpha$ is the only route to the ISO 354 power attenuation coefficient
$m = \alpha/(10 \log_{10} e)$, exposed as `air_power_attenuation_per_m`.

That $\pm 10$ % is not unconditional. Table 1 is tabulated from 50 Hz to 10 kHz,
−20 to +50 °C and 10 to 100 % relative humidity at one atmosphere, and the
$\pm 10$ % accuracy is claimed only for molar water-vapour concentrations
between 0.05 % and 5 % with the temperature between 253 and 323 K (Clause 7.1);
outside that humidity window it degrades to about $\pm 20$ % (7.2) and below
$h = 0.005$ % to about $\pm 50$ % (7.3). Because $\alpha$ multiplies path length, that percentage
becomes a level error that grows with distance: the same $\pm 20$ % on a
coefficient of 30 dB/km is $\pm 0.2$ dB at 30 m and $\pm 6$ dB at a kilometre,
which is why atmospheric absorption is usually the largest single uncertainty in
a long-range prediction and why a *measured* humidity is worth more than a
carefully measured temperature: the relaxation frequencies move with humidity,
so the attenuation in a given band is far more sensitive to it. Interpolating
the printed table is expressly discouraged (Clause 6.4); the equations, which is
what the library evaluates, are the intended route.

![ISO 9613-1 pure-tone atmospheric attenuation coefficient alpha in dB/km against frequency, on a linear decibel ordinate over a logarithmic frequency axis, for the reference 20 degrees Celsius and 50 percent relative humidity atmosphere, produced by the AtmosphericAttenuation result plot method](https://docs.jmrp.io/phonometry/media/atmospheric_attenuation.svg)

*The ISO 9613-1 coefficient for the 20 °C, 50 % relative-humidity reference atmosphere: the $f^2$ rise spans two decades from 50 Hz to 10 kHz.*

![ISO 9613-1 atmospheric absorption coefficient in decibels per kilometre against frequency from 50 Hz to 10 kHz on log-log axes, drawn for four atmospheres: 20 degrees Celsius at 50 percent relative humidity, 20 degrees Celsius at 10 percent, 0 degrees Celsius at 70 percent and 30 degrees Celsius at 80 percent, the dry warm curve lying well above the others through the mid frequencies and the curves converging at the top of the band](https://docs.jmrp.io/phonometry/media/air_absorption_alpha.svg)

*The humidity claim of the paragraph above, drawn: four atmospheres, and a
spread far wider than the $\pm 10$ % claimed for any one of them. At 2 kHz the
coefficient is 9.9 dB/km at the 20 °C / 50 % reference and **45.5 dB/km** at the
same temperature and 10 % humidity, a factor 4.6 from humidity alone, worth
36 dB over a kilometre. Note which way it goes: drying the air *increases* the
mid-band attenuation, because the oxygen relaxation frequency falls with
humidity and drags its peak down into the band. Temperature moves the curves far
less: 0 °C at 70 % and 20 °C at 50 % differ by 0.02 dB/km at 1 kHz.*

### Outdoor propagation, general method (ISO 9613-2)

ISO 9613-2:1996 predicts the octave-band level at a receiver under **downwind
propagation**: a light wind blowing from source to receiver, or the moderate
ground-based temperature inversion that bends rays the same way (clause 5), as
$L_{fT}(DW) = L_W + D_\mathrm{c} - A$ (Eq. (3)), where $D_\mathrm{c}$ is the directivity correction
and $A$ is the octave-band attenuation, a sum of independent physical mechanisms
(Eq. (4)):

$$
A = A_\mathrm{div} + A_\mathrm{atm} + A_\mathrm{gr} + A_\mathrm{bar} + A_\mathrm{misc}.
$$

![ISO 9613-2 per-octave-band attenuation breakdown as a stacked bar of Adiv, Aatm, Agr and Abar with the total A overlaid, for a 200 m path over porous ground with a 4 m barrier](https://docs.jmrp.io/phonometry/media/outdoor_attenuation_breakdown.svg)

*The four terms at their true relative sizes, band by band, for a 200 m path
over porous ground with a 4 m barrier. $A_\mathrm{div}$ is 57 dB in every band because
it is pure geometry. $A_\mathrm{atm}$ is nothing at 63 Hz and 18.7 dB at 8 kHz, so it
is the term that decides how far high frequencies travel and no other. $A_\mathrm{gr}$
is where the low bands live and is **negative** at 63 Hz (−4.6 dB: the ground
reflection adds energy rather than removing it). $A_\mathrm{bar}$ is at its 20 dB cap
from 2 kHz up but falls to zero at 250 Hz, because the top-edge form subtracts
the ground effect the screened path gives away, $A_\mathrm{bar} = D_z - A_\mathrm{gr} \geq 0$,
and 250 Hz is exactly where $A_\mathrm{gr}$ peaks. Which term is worth refining
depends entirely on the band and the geometry.*

The library implements the four general terms of Clause 7; the informative
$A_\mathrm{misc}$ (foliage, industrial sites, housing) and reflections are left to the
caller. **Geometrical divergence** is spherical spreading from a point source,
$A_\mathrm{div} = 20 \log_{10}(d/d_0) + 11$ dB with $d_0 = 1$ m (Eq. (7)): exactly
51 dB at 100 m, +6 dB per distance doubling. **Atmospheric absorption** is
$A_\mathrm{atm} = \alpha\ d$ (Eq. (8)) with $\alpha$ the ISO 9613-1 coefficient above.
**Ground effect** $A_\mathrm{gr} = A_\mathrm{s} + A_\mathrm{r} + A_\mathrm{m}$ (Eq. (9)) sums a source, receiver and
middle region, each evaluated from the Table 3 functions $a'/b'/c'/d'$ and its
ground factor $G$ (0 hard, 1 porous); a negative $A_\mathrm{gr}$ denotes a net gain from
the ground reflection. The three regions are lengths of the path as projected on
the ground plane, $d_\mathrm{p}$ (7.3.1, Figure 1): the source region runs
$30 h_\mathrm{s}$ from the source towards the receiver and the receiver region
$30 h_\mathrm{r}$ back from the receiver, each of them $d_\mathrm{p}$ at most, and the
middle region is what is left between the two. A region of mixed ground takes for
$G$ the fraction of it that is porous, and the middle term carries the factor
$q = 1 - 30(h_\mathrm{s} + h_\mathrm{r})/d_\mathrm{p}$ (Table 3, note 2), so on a path of
$30(h_\mathrm{s} + h_\mathrm{r})$ or less the two end regions meet or overlap, $q = 0$ and
there is no middle region at all. The diagram draws a 200 m path with source and
receiver 1.5 m above the ground: a 45 m region at each end, a 110 m middle region
split into 70 m of grass and 40 m of asphalt ($G_\mathrm{m} = 70/110 \approx 0.64$), and
$q = 0.55$, while its box gives the source region 25 m of porous ground
($G_\mathrm{s} = 25/45 \approx 0.56$) without drawing the split; under it, the same heights on
a 60 m path, short of the 90 m past which there is a middle region.

![Section to scale of a 200 m path with a source and a receiver 1.5 m above the ground: the source region shaded over the first 45 m as 30 times hs, the receiver region shaded over the last 45 m as 30 times hr, and the 110 m middle region between them split into 70 m of grass and 40 m of asphalt with Gm = 70/110 = 0.64; a formula box gives q = 1 minus 30 times (hs plus hr) over dp and an inset repeats the same path at 60 m where the source and receiver regions overlap, q = 0 and the middle region does not exist](https://docs.jmrp.io/phonometry/media/diagram_ground_regions.svg)

An alternative A-weighted-only form of the ground effect,
$A_\mathrm{gr} = 4.8 - (2 h_\mathrm{m}/d)[17 + 300/d] \ge 0$ (Eq. (10)) is offered for porous
ground when only the A-weighted level matters, paired with the solid-angle index
$D_\Omega$ (Eq. (11)). **Screening** by a barrier is the diffraction insertion
loss

$$
D_z = 10 \log_{10}\big[ 3 + (C_2/\lambda)\ C_3\ z\ K_\mathrm{met} \big] \quad\text{dB},
$$

(Eq. (14)) with $C_2 = 20$ (or 40 when ground reflections are handled by image
sources), $C_3 = 1$ for a single edge or Eq. (15) for a double edge, the
path-length difference $z = d_\mathrm{ss} + d_\mathrm{sr} - d$ (Eq. (16)/(17)), wavelength
$\lambda = 340/f$ and the meteorological factor $K_\mathrm{met}$ (Eq. (18)); $D_z$ is
capped at 20 dB (single) or 25 dB (double). For a top-edge barrier the ground
effect of the screened path is folded into the screening term,
$A_\mathrm{bar} = D_z - A_\mathrm{gr} \ge 0$ (Eq. (12), Note 13); for a lateral (vertical-edge)
barrier $A_\mathrm{bar} = D_z$ and the ground term is kept (Eq. (13)). (The standard
fixes $\lambda = 340/f$ inside the screening term rather than deriving $c$ from
the temperature, so the barrier attenuation is temperature-independent by
construction; the atmospheric-absorption term above is not.) The long-term
average level subtracts the meteorological correction $C_\mathrm{met}$ (Eq. (6),
(21)/(22)).

**Where the method applies, and what counts as a barrier.** It is defined for the
eight nominal octave midbands from 63 Hz to 8 kHz and for point sources, an
extended source being decomposed into point sub-sources (clauses 4–5), so a
one-third-octave chain or a 31.5 Hz band is outside it. It predicts the
*downwind* level; an upwind or crosswind situation is reached only through the
long-term correction $C_\mathrm{met}$, never by evaluating Eq. (3) directly. The
A-weighted-only ground form of Eq. (10) carries two further conditions of its
own: porous or mostly porous ground, and a sound that is not a pure tone. And an
object earns a $D_z$ only if it qualifies as a barrier under clause 7.4: a
surface density of at least 10 kg/m², a closed surface with no large gaps, and a
horizontal dimension normal to the source-receiver line greater than the
wavelength, so a slatted fence, a hedge or a wall the sound simply passes
around is modelled as *no barrier at all*, not as a weak one. Crediting one is
worth many decibels of error in the wrong direction. Finally, Table 5 gives the
accuracy by geometry rather than as one range: about $\pm 1$ dB only for a mean
source-receiver height between 5 and 30 m at distances under 100 m, and about
$\pm 3$ dB in every other case within 1000 m, with no accuracy stated beyond
1000 m, where the method is an extrapolation.

![ISO 9613-2 source-barrier-receiver geometry: a point source at height hs, a barrier whose top edge splits the path into dss and dsr, and a receiver at height hr, with the blocked direct ray and the diffracted ray over the edge, the path difference z and the Dz formula](https://docs.jmrp.io/phonometry/media/diagram_outdoor_geometry.svg)

## Occupational noise exposure (ISO 9612)

The quantity itself, $L_\mathrm{EX,8h}$, is defined from sound exposure in
[Signal Analysis](https://docs.jmrp.io/phonometry/reference/theory/signal-analysis/#event-and-dose-metrics);
this section is about measuring it in a workplace and bounding its uncertainty.

ISO 9612:2009 is the engineering method (accuracy grade 2) for a worker's daily
noise exposure level $L_\mathrm{EX,8h}$, normalised to a nominal 8 h day. Three
**measurement strategies** trade effort for representativeness. The *task-based*
method (Clause 9) splits the day into tasks, energy-averages $I \ge 3$ samples
per task (Eq. 7) and sums the task contributions
$L_{\mathrm{EX,8h},m} = L_{p,\mathrm{A,eq}T,m} + 10 \log_{10}(T_m/T_0)$ energetically (Eq. 9/10).
The *job-based* method (Clause 10) energy-averages $N \ge 5$ random samples over a
homogeneous exposure group (Eq. 11) and normalises the effective-day duration
(Eq. 12); the *full-day* method (Clause 11) does the same arithmetic on whole-day
measurements (Eq. 13).

**Where the microphone goes.** With a sound level meter the preferred position is
the worker's head position *with the worker absent*, in the centre plane of the
head, on a line with the eyes and with the microphone axis parallel to the
direction of view, and where the worker must be present, at 0.1 m to 0.4 m from
the entrance of the external ear canal on the side of the more exposed ear
(clause 12.4). A body-worn personal sound exposure meter has its microphone on
top of the shoulder, at least 0.1 m from the ear-canal entrance on the more
exposed side and about 0.04 m above the shoulder, with microphone and cable
fastened so that clothing and handling cannot fake a result (12.3). A field
calibration is performed before and after each series, and a drift above 0.5 dB
discards the series (12.2). That geometry is what the $u_3 = 1.0$ dB below buys:
it is not a property of the instrument but the price of the position, and in the
job and full-day budgets it is usually the second-largest term after the
sampling spread, so a position taken loosely does not merely widen the interval,
it quietly voids the grade-2 claim.

![Left: a worker wearing a personal sound exposure meter (IEC 61252), its microphone mounted about 0.04 m above the shoulder and at least 0.1 m from the entrance of the most-exposed ear canal, per ISO 9612 Clause 12.3. Right: the three measurement strategies drawn as timelines over an eight-hour working day, task-based (the day split into labelled tasks, at least three samples plus a duration each), job-based (five or more random samples spread over the homogeneous exposure group) and full-day (the whole shift measured at least three times), all feeding the LEX,8h and its Annex C uncertainty, chosen by work pattern from Table B.1](https://docs.jmrp.io/phonometry/media/diagram_dosimeter_iso9612.svg)

The **Annex C** uncertainty budget is normative. The combined standard
uncertainty is $u^2 = \sum c_i^2 u_i^2$ (C.1) and the expanded uncertainty is
$U = k\ u$ with $k = 1.65$ for a **one-sided** 95 % interval (Clause 14), so the
reported upper limit is $L_\mathrm{EX,8h} + U$. The task and job methods differ in an
instructive way: the task noise-sampling uncertainty $u_{1\mathrm{a}}$ divides the summed
squared deviations by $I(I-1)$ (the standard error of the mean, Eq. C.6)
whereas the job/full-day sampling uncertainty $u_1$ is the plain sample standard
deviation with denominator $N-1$ (Eq. C.12), so the same spread contributes more
in the job method (fewer, coarser samples). The task budget (Eq. C.3) adds the
sensitivity coefficients $c_{1\mathrm{a}}$ (Eq. C.4) and $c_{1\mathrm{b}}$ (Eq. C.5) and an optional
task-duration uncertainty $u_{1\mathrm{b}}$ (Eq. C.7); the job/full-day budget (Eq. C.9)
reads $c_1 u_1$ from Table C.4 as a function of $(N, u_1)$ and adds the instrument
uncertainty $u_2$ (Table C.5) and microphone-position uncertainty $u_3 = 1.0$ dB
in quadrature. Peak levels $L_{p,\mathrm{Cpeak}}$ are reported without an uncertainty:
Annex C provides no method for them (Table C.5, Note 1). The three worked
examples of Annexes D (task, $L_\mathrm{EX,8h} = 84.3$ dB, $U = 2.7$ dB), E (job,
$88.1$ dB, $3.8$ dB) and F (full-day, $90.1$ dB, $3.4$ dB) are reproduced to
the standard's printed precision: every intermediate of Annex E is digit-exact,
and its final level differs only by the standard's own pre-rounding of the
effective-day level (see the [Occupational Noise Exposure guide](https://docs.jmrp.io/phonometry/perception/hearing/occupational-exposure/)).

![ISO 9612 Annex D task-based exposure: the three task LEX,8h contributions as bars, the energy-summed daily LEX,8h line and the one-sided 95 % upper limit LEX,8h + U band above it](https://docs.jmrp.io/phonometry/media/exposure_uncertainty.svg)

*What the budget buys: the Annex D worked day, its three task contributions, the
energy-summed $L_\mathrm{EX,8h}$ and the band up to $L_\mathrm{EX,8h} + U$ that the one-sided
95 % interval adds on top. The reported result is the pair, not the line, and
with $U = 2.7$ dB the upper limit sits 2.7 dB above the number, which is enough
to cross an action value on its own.*

See the [Outdoor Propagation guide](https://docs.jmrp.io/phonometry/environment/propagation/outdoor-propagation/) and the
[Occupational Noise Exposure guide](https://docs.jmrp.io/phonometry/perception/hearing/occupational-exposure/) for usage.

## Sound power determination (ISO 3744/3745/3746, ISO 3741, ISO 9614-1/2/3)

The sound power level $L_W = 10 \log_{10}(P/P_0)$ ($P_0 = 1$ pW) is an
*emission* quantity: unlike a pressure level it does not depend on the receiver
distance or the room. Three families recover it, one per measured quantity, and
the choice is dictated by the room available. **Sound pressure** over an
enveloping surface in a free field above a reflecting plane assumes the
intensity normal to that surface is the mean-square pressure divided by
$\rho c$: ISO 3744 (engineering) and ISO 3746 (survey) below, and ISO 3745
(precision) when the room is anechoic; read in a diffuse field instead, that
same pressure turns the room's own absorption into the calibration, which is
ISO 3741. **Sound intensity** measures the flux directly and needs no qualified
room at all, at the price of a two-channel probe and the field indicators that
police it: ISO 9614-1 at discrete points, and ISO 9614-2 (engineering) and
ISO 9614-3 (precision) by scanning. **Surface velocity** reads the radiating
casing with an accelerometer and takes no acoustic measurement at all, paying
for that freedom with a radiation factor: that is ISO/TS 7849, whose factor is
the radiation efficiency derived in the
[vibration reference](https://docs.jmrp.io/phonometry/reference/theory/vibration/#point-mobilities-and-radiation-efficiency-cremer-5-hopkins-29).

![The seven sound power routes, one row for each measured quantity, every cell carrying the same five attributes. The sound pressure row: the ISO 3744/3746 hemispherical enveloping surface over a reflecting plane, the ISO 3745 fixed microphone array in a wedge-lined anechoic room, and the ISO 3741 diffuse field of a reverberation room. The sound intensity row: the ISO 9614-1 measurement surface cut into ten segments with the probe held still at a point in each, the ISO 9614-2 serpentine scan swept over that same surface, and the tighter ISO 9614-3 precision scan. The surface velocity row carries one route across the full width, ISO/TS 7849 with an accelerometer on a radiating casing and no microphone at all](https://docs.jmrp.io/phonometry/media/sound_power_methods.svg)

*The seven routes to $L_W$, one row per measured quantity: a sound pressure, a
sound intensity, or the surface velocity of the casing.*

### Enveloping-surface pressure (ISO 3744/3746)

Over a reflecting plane the free-field relation is simply
$L_W = \bar{L}_p + 10 \log_{10}(S/S_0)$: the mean-square pressure averaged over
an enveloping surface of area $S$, multiplied by $S$, is the radiated power.
Two corrections restore that idealisation. Uncorrelated **background noise**
adds its mean square to the source's, so with the margin
$\Delta L_p = L_\mathrm{ST} - L_\mathrm{bg}$ the source-only level is recovered by subtracting
$K_1 = -10 \log_{10}(1 - 10^{-\Delta L_p/10})$ (from $p_\mathrm{src}^2 = p_\mathrm{ST}^2 (1 - 10^{-\Delta L_p/10})$).
The **reverberant field** of a non-anechoic room adds a near-uniform energy
density $4P/(A c)$ to the direct $P/(S c)$, so the surface level exceeds the
free-field value by their ratio, $K_2 = 10 \log_{10}(1 + 4 S/A)$, with $A$ the
room's equivalent absorption area.

**Neither correction is unconditional, and both gate the result.** For $K_1$,
ISO 3744 clause 8.2.3 sets a ladder on the margin: above 15 dB the correction is
taken as zero, between 6 and 15 dB the formula applies, and below 6 dB it is
capped at the value it reaches at 6 dB (1.3 dB), and the reported power becomes
an *upper bound* rather than a determination. (ISO 3746 uses the same shape with
coarser A-weighted thresholds: 10 dB, 3 dB and a 3 dB cap.) The honest response
to a 2 dB margin is a quieter time of day or a measurement surface closer to the
source, not a larger correction. For $K_2$, it may be neglected below 0.5 dB, and
the engineering method is valid only while $K_{2\mathrm{A}} \le 4$ dB (7 dB for the
ISO 3746 survey method, clause 4.3), so a room that returns 6 dB is not a
correction to apply but a room to reject. And read the direction of $K_2$
carefully: it grows with $4S/A$, so enlarging the measurement surface makes the
room correction *worse*, which is the opposite of the intuition most operators
bring to it.

The surface is built on the **reference box**, the smallest right parallelepiped
that just encloses the source, whose sides are $l_1$, $l_2$ and $l_3$; over one
reflecting plane the characteristic dimension is
$d_0 = \sqrt{(l_1/2)^2 + (l_2/2)^2 + l_3^2}$ (clause 7.1). A hemispherical
surface needs a radius $r \ge 2 d_0$, at least 1 m (0.5 m for small products)
and at most 16 m (7.2.3); a box surface needs a measurement distance
$d \ge 0.25$ m, preferably 1 m (7.2.4): that $d$ is the one in the formulas
below. The area is then the closed form of the
geometry: a hemisphere $S = 2 \pi r^2$ over one reflecting plane (halved and
quartered for two and three planes), a one-plane box $S = 4(ab + bc + ca)$ with
$a = 0.5\ l_1 + d$, $b = 0.5\ l_2 + d$, $c = l_3 + d$. The microphone positions
are normative rather than free: Annex B tabulates equal-area coordinates on the
hemisphere; Table B.1 for sources that may emit discrete tones, Table B.2 for
broadband ones, of which the engineering grade uses 10 key positions over one
reflecting plane (5 over two, 3 over three), with the survey method taking a
reduced subset. `measurement_positions()` returns them scaled to the chosen
radius, and `plot_microphone_positions()` draws the array before anything is
measured.

The diagram below draws the two surfaces of ISO 3744: the hemisphere of radius
$r$ over one reflecting plane with the ten key positions of Annex B, and the box
at measurement distance $d$ around a large source.

![Measurement surfaces of ISO 3744: a hemisphere of radius r enveloping a compact source on a reflecting plane, with the ten Annex B microphone positions marked, and a right parallelepiped (box) at measurement distance d around a large source](https://docs.jmrp.io/phonometry/media/diagram_sound_power_surfaces.svg)

ISO 3746 (survey) shares
the maths with looser criteria. The expanded uncertainty is
$U = 2 \sqrt{\sigma_{R0}^2 + \sigma_\mathrm{omc}^2}$.

### Precision grade in anechoic rooms (ISO 3745)

ISO 3745:2012 is the grade-1 (precision) sibling: a qualified anechoic or
hemi-anechoic room removes the reverberant field, so there is no $K_2$ term and
the corrections become meteorological. The power level is
$L_W = \bar{L}_p + 10 \log_{10}(S/S_0) + C_1 + C_2 + C_3$ (Eq. 14/15) over a full
sphere $S = 4 \pi r^2$ or hemisphere $S = 2 \pi r^2$, with the background
correction $K_{1i} = -10 \log_{10}(1 - 10^{-0.1 \Delta L_{pi}})$ applied per
microphone position *before* the energy average (Eq. 11); no correction is
needed above a 15 dB margin, and below 10 dB (250 Hz – 5 kHz) or 6 dB (edge
bands) the correction is clamped and the result flagged as an upper bound
(clause 9.4.2). The meteorological terms are
$C_1 = -10 \log_{10}(p_\mathrm{s}/p_{\mathrm{s}0}) + 5 \log_{10}[(273 + \theta)/\theta_0]$ and
$C_2 = -10 \log_{10}(p_\mathrm{s}/p_{\mathrm{s}0}) + 15 \log_{10}[(273 + \theta)/\theta_1]$ with
$\theta_0 = 314$ K, $\theta_1 = 296$ K: at the 23 °C / 101.325 kPa reference
$C_2 = 0$ exactly and $C_1 = -0.128$ dB; and
$C_3 = A_0 (1.0053 - 0.0012 A_0)^{1.6}$ with $A_0 = a(f)\ r$ restores the
ISO 9613-1 air absorption over the measurement radius. The Annex D/E
microphone arrays are built in as digit-exact coordinate tables (40 equal-area
positions; the mirror set 21–40 is added when the band-SPL spread exceeds
$N_\mathrm{M}/2$, clause 9.3.2), and the same positions yield the directivity index
$DI_i = L_{pi} - \bar{L}_p$ (Eq. 21). The clause 10.5 uncertainty example,
$U = 2\sqrt{0.5^2 + 2.0^2} = 4.12$ dB, is reproduced, along with the Table 2/3
per-band $\sigma_{R0}$ values.

### Reverberation room (ISO 3741)

In a qualified diffuse field the steady energy density $w = 4P/(A c)$ ties the
power to the room absorption, giving $L_W = \bar{L}_p + 10 \log_{10}(A/A_0) - 6$
plus higher-order corrections, with $A = (55.26/c)(V/T_{60})$ and
$c = 20.05 \sqrt{273 + \theta}$. The **Waterhouse correction**
$10 \log_{10}(1 + S c/(8 V f))$ compensates the extra energy stored in the
boundary layer that interior microphones miss ($S c/(8 V f) = S \lambda/(8 V)$,
so it fades as frequency rises); the $4.34\ A/S$ term is the mean-free-path air
correction, and $C_1$, $C_2$ carry the result to the reference meteorological
conditions (23 °C, 101.325 kPa). The **comparison method** subtracts a
reference source of known power measured in the same room,
$L_W = L_{W(\text{RSS})} + (\bar{L}_p - \bar{L}_{p,\text{RSS}} + C_2)$, so the
absorption-area, Waterhouse and $C_1$ terms cancel and the room need not be
characterised.

### Discrete-point intensity (ISO 9614-1)

Part 1 holds the probe still. One position stands for each segment of the
measurement surface, so the partial power is $P_i = I_{\mathrm{n}i} S_i$
(Eq. 11) and the sound power is $L_W = 10 \log_{10}(\sum_i P_i / P_0)$
(Eq. 12). That sum is printed without the absolute-value bars of the general
definition because clause 9.2 disposes of the negative case instead: a band
whose $\sum_i P_i$ comes out negative is outside the method rather than a band
with a level to report. One segment carrying negative power is ordinary, and is
what $F_3$ exists to quantify.

What separates Part 1 from a scan is not the integral but what decides whether
it was sampled finely enough, and with a fixed set of positions the surface has
to answer that itself. The Annex A indicators average over positions without
area weighting: $F_1$ is the temporal variability (Eq. A.1),
$F_2 = L_p - L_{|I_\mathrm{n}|}$ and $F_3 = L_p - L_{I_\mathrm{n}}$ are the
reactivity and recirculation indicators (Eqs. A.3 to A.7), and
$F_4 = \frac{1}{|\overline{I_\mathrm{n}}|} \sqrt{\sum_i (I_{\mathrm{n}i} - \overline{I_\mathrm{n}})^2 / (N - 1)}$
is the field non-uniformity (Eqs. A.8 and A.9), which is Part 1's own. Annex B
turns it into a position count: criterion 2 asks $N > C F_4^2$ (Eq. B.2) with
$C$ from Table B.2, so a surface that radiates unevenly demands more positions,
band by band, and the grade a band earns is the grade whose $C$ that inequality
still satisfies. Criterion 1, $L_\mathrm{d} > F_2$ (Eq. B.1), puts the probe
against the field before any of that, and no amount of averaging rescues a band
that fails it. The gate figure in the next subsection is drawn in exactly these
names, because they are Part 1's.

### Intensity scanning (ISO 9614-2)

Sound intensity is the net energy flux $\vec{I} = \overline{p\ \vec{u}}$, so by
the divergence theorem the power through a closed surface is
$P = \sum_i \langle I_{\mathrm{n},i} \rangle\ S_i$. A steady source *outside* the surface
contributes zero net flux (its energy enters and leaves), which is why
intensity rejects stationary background noise, but it can still drive a band's
$P$ negative, in which case that band is not determinable.

**What the indicators are judging.** The surface is divided into at least four
contiguous segments (clause 8.2) and each is swept continuously along parallel
lines whose mean spacing does not exceed the mean distance from that segment to
the source surface, with the probe axis held normal to the surface throughout
(8.1). A manual sweep runs at 0.1 to 0.5 m/s (0 to 1 m/s mechanically) and lasts
at least 20 s per segment; the segment stands at least 200 mm off an extended
vibrating surface, or 100 mm off a small compact one (8.2). The engineering
grade takes **two** sweeps per segment along orthogonal paths and records each
separately (8.3.1), and it is the agreement between those two that the Table 2
repeatability criterion tests, which is why a failing band is far more often a
scanning problem than an instrument problem. Two checks belong on site rather
than in the calculation: a windscreen wherever a mean flow above 4 m/s cannot be
ruled out (5.3), and the probe-reversal check of 6.2.2: rotate the probe 180°,
and the two indicated intensities must have opposite signs and differ by less
than 1.5 dB in the maximum band, or the phase match has not survived the
environment.

Two normative field
indicators then gate validity: the surface pressure-intensity indicator
$F_{pI} = \bar{L}_p - L_W + 10 \log_{10}(S/S_0)$ (reactivity) and the
negative-partial-power indicator
$F_{+/-} = 10 \log_{10}(\sum_i \lvert P_i \rvert / \lvert \sum_i P_i \rvert)$
(recirculation), together with the probe's dynamic capability
$L_\mathrm{d} = \delta_{pI0} - K$ ($K = 10$ dB grade 2, 7 dB grade 3), which must exceed
$F_{pI}$. A band earns the engineering grade when $L_\mathrm{d} > F_{pI}$, $F_{+/-} \le 3$ dB
and the two repeated sweeps agree within the Table 2 limit.

The same two indicators carry three sets of names across the ISO 9614 family:
ISO 9614-1 calls them $F_2$ (pressure-intensity), $F_3$ (negative partial power)
and $F_4$ (field non-uniformity); Part 2 writes $F_{pI}$ and $F_{+/-}$; Part 3
writes $F_{pI_\mathrm{n}}$ and $F_\mathrm{S}$. Each subsection here uses its own part's symbols. The
$F_2$-to-$F_4$ naming is the one used in
[Sound intensity](https://docs.jmrp.io/phonometry/reference/theory/signal-analysis/#sound-intensity-iec-61043).

![ISO 9614-3 precision sound intensity scanning: a source enclosed by a measurement surface divided into segments, a two-microphone intensity probe scanned along a serpentine path over each segment, and the sound power formed by summing the normal intensity times segment area, subject to the field-indicator acceptance criteria](https://docs.jmrp.io/phonometry/media/diagram_intensity_scan.svg)

![ISO 9614-1 field indicators of a ten-position scan over six octave bands: F2 and F3 climb towards low frequency as the field turns reactive, crossing the dashed dynamic-capability line Ld at 125 Hz where F3 also rises above F2, with the field non-uniformity F4 drawn as bars on a twin axis](https://docs.jmrp.io/phonometry/media/field_indicators.svg)

*The gate as a picture rather than an inequality, drawn here in the ISO 9614-1
naming, $F_2$ for what Part 2 calls $F_{pI}$. The reactivity indicator climbs
towards low frequency and crosses the dynamic-capability line at 125 Hz: that
band fails, and no amount of averaging will rescue it, because the cause is the
field near the surface and not the analyser. $F_3$ rising above $F_2$ in the same
band reveals inward-flowing partial intensity, and the $F_4$ bars set how many
positions the surface needs.*

### Precision intensity scanning (ISO 9614-3)

ISO 9614-3:2002 upgrades the scanning method to precision grade with a tighter
indicator machinery. The partial powers $P_i = I_{\mathrm{n},i} S_i$ (Eq. 5) sum as
before, but validity now rests on the signed and unsigned pressure-intensity
indicators $F_{pI_\mathrm{n}} = \bar{L}_p - L_{I\mathrm{n}}$ (Eqs. B.3/B.6, the F2/F3 of
ISO 9614-1) and the normalized intensity non-uniformity $F_\mathrm{S}$ (Eq. B.8),
through five acceptance criteria (Annex C): scan repeatability
$|L_{I\mathrm{n}}(1) - L_{I\mathrm{n}}(2)| \le s/2$ (C.1), dynamic capability
$L_\mathrm{d} = \delta_{pI0} - K \ge F_{pI_\mathrm{n}}(\text{signed})$ with the precision
bias-error factor $K = 10$ dB (C.2),
$F_{pI_\mathrm{n}}(\text{signed}) - F_{pI_\mathrm{n}}(\text{unsigned}) \le 3$ dB (C.3),
$F_\mathrm{S} \le 2$ (C.4) and the scan-density convergence
$0.83 \le F_\mathrm{S}(1)/F_\mathrm{S}(2) \le 1.2$ (C.5). Eq. 10 normalizes the result to the
reference meteorological conditions,
$L_{W0} = L_W - 15 \log_{10}[(B/101325) \cdot 296.15/(273.15 + \theta)]$. Bands whose
net power is negative are not determinable (clause 9.2) and are flagged. A
uniform normal intensity recovers the power exactly (100 µW over 3.75 m² →
80.0 dB re 1 pW), independent of how the surface is segmented.

See the [Sound Power guide](https://docs.jmrp.io/phonometry/devices/emission/sound-power/) for the choice
of method, and the
[pressure](https://docs.jmrp.io/phonometry/devices/emission/sound-power-pressure/),
[reverberation-room](https://docs.jmrp.io/phonometry/devices/emission/sound-power-reverberation/),
[intensity-scanning](https://docs.jmrp.io/phonometry/devices/emission/sound-power-intensity/),
[discrete-point intensity](https://docs.jmrp.io/phonometry/devices/emission/intensity/) and
[surface-vibration](https://docs.jmrp.io/phonometry/devices/emission/vibration-sound-power/) guides for
usage.

## References

- **Attenborough, K., & Van Renterghem, T. (2021).** *Predicting outdoor sound* (2nd ed.). CRC Press. https://doi.org/10.1201/9780429470806
  The ground impedance and meteorological effects underlying the ground-effect and barrier terms.
- **Fahy, F. J. (1995).** *Sound intensity* (2nd ed.). E&FN Spon. https://doi.org/10.4324/9780203475386
  ISBN 978-0-419-19810-9. The intensity physics behind the scanning methods: active intensity and the p-p error budget.
- **International Organization for Standardization. (1993).** [*Acoustics — Attenuation of sound during propagation outdoors — Part 1: Calculation of the absorption of sound by the atmosphere*](https://www.iso.org/standard/17426.html) (ISO 9613-1:1993).
  The pure-tone attenuation coefficient and its relaxation physics.
- **International Organization for Standardization. (1993).** [*Acoustics — Determination of sound power levels of noise sources using sound intensity — Part 1: Measurement at discrete points*](https://www.iso.org/standard/17427.html) (ISO 9614-1:1993).
  The field indicators and dynamic-capability criterion shared by the intensity-scanning methods.
- **International Organization for Standardization. (1996).** [*Acoustics — Attenuation of sound during propagation outdoors — Part 2: General method of calculation*](https://www.iso.org/standard/20649.html) (ISO 9613-2:1996).
  The attenuation chain (divergence, air, ground, barrier) of the general-method section. Revised in 2024; the 1996 method is the implemented one.
- **International Organization for Standardization. (2009).** [*Acoustics — Determination of occupational noise exposure — Engineering method*](https://www.iso.org/standard/41718.html) (ISO 9612:2009).
  The three measurement strategies and the Annex C uncertainty budget of the occupational section.
- **International Organization for Standardization. (2010).** [*Acoustics — Determination of sound power levels and sound energy levels of noise sources using sound pressure — Engineering methods for an essentially free field over a reflecting plane*](https://www.iso.org/standard/52055.html) (ISO 3744:2010).
  The enveloping-surface method with its background and reverberant-field corrections.
- **International Organization for Standardization. (2010).** [*Acoustics — Determination of sound power levels and sound energy levels of noise sources using sound pressure — Precision methods for reverberation test rooms*](https://www.iso.org/standard/52053.html) (ISO 3741:2010).
  The diffuse-field method with the Waterhouse correction and the comparison method.
- **International Organization for Standardization. (2012).** [*Acoustics — Determination of sound power levels and sound energy levels of noise sources using sound pressure — Precision methods for anechoic rooms and hemi-anechoic rooms*](https://www.iso.org/standard/45362.html) (ISO 3745:2012).
  The precision anechoic method with its meteorological corrections and microphone arrays.
- **International Organization for Standardization. (2016).** [*Acoustics — Description, measurement and assessment of environmental noise — Part 1: Basic quantities and assessment procedures*](https://www.iso.org/standard/59765.html) (ISO 1996-1:2016).
  The composite rating level and Table A.1 adjustments of the descriptors section.
- **International Organization for Standardization. (2017).** [*Acoustics — Description, measurement and assessment of environmental noise — Part 2: Determination of sound pressure levels*](https://www.iso.org/standard/59766.html) (ISO 1996-2:2017).
  The survey standard that requires an adjustment for prominent impulsive sounds in the first place; the objective procedure only decides how large it is.
- **International Organization for Standardization. (2022).** [*Acoustics — Description, measurement and assessment of environmental noise — Part 3: Objective method for the measurement of prominence of impulsive sounds and for adjustment of LAeq*](https://www.iso.org/standard/77035.html) (ISO/PAS 1996-3:2022).
  The ISO successor built on the NT ACOU 112 prominence.
- **Maekawa, Z. (1968).** Noise reduction by screens. *Applied Acoustics*, *1*(3), 157-173. https://doi.org/10.1016/0003-682X(68)90020-0
  The screen-attenuation chart that the barrier insertion-loss formulas descend from.
- **Nordtest. (2002).** [*Acoustics: Prominence of impulsive sounds and for adjustment of LAeq*](https://www.nordtest.info/wp/2002/05/01/acoustics-prominence-of-impulsive-sounds-and-for-adjustment-of-laeq-nt-acou-112/) (NT ACOU 112:2002).
  The onset-rate prominence and LAeq adjustment of the impulsive-sound section.
- **Salomons, E. M. (2001).** *Computational atmospheric acoustics*. Kluwer Academic Publishers. https://doi.org/10.1007/978-94-010-0660-6
  ISBN 978-1-4020-0390-5. The wave-based outdoor propagation theory behind the engineering approximations of the ISO 9613-2 section.
