underwater.ship_traffic_noise

Predicted source-level spectrum of shipping traffic (semi-empirical models).

When no measured spectrum is available, the underwater radiated-noise source level of a ship can be estimated from readily available traffic parameters (vessel class, speed, length) with published semi-empirical models. This module implements three, selectable through model:

• "jomopans-echo" — the JOMOPANS-ECHO reference spectrum (MacGillivray & de Jong 2021), per vessel class with class reference speeds; validated against 1862 source-level measurements (ECHO programme), σ ≈ 6 dB. Default.
• "randi" — the RANDI 3.1 semi-empirical model: an “average ship” baseline spectrum scaled by speed and length.
• "wales-heitmeyer" — the Wales & Heitmeyer (2002) ensemble merchant-ship spectrum (no speed/length dependence), valid 30 Hz-1200 Hz.

All three return an equivalent-monopole source spectral-density level (dB re 1 µPa²/Hz at 1 m, source depth 6 m) and the decidecade-band source level (dB re 1 µPa m). The predicted spectrum can be used as the shipping input of phonometry.underwater.ocean_ambient_noise.ocean_ambient_noise or placed at range with phonometry.underwater.propagation.transmission_loss.

Source (clean-room, implemented from the equations, validated against the authors’ own Excel reference implementation, File S1): MacGillivray, A.; de Jong, C. (2021), “A Reference Spectrum Model for Estimating Source Levels of Marine Shipping Based on Automated Identification System Data”, J. Mar. Sci. Eng. 9(4), 369, https://doi.org/10.3390/jmse9040369 (CC-BY) — which also reproduces RANDI 3.1 [Breeding et al.] and Wales & Heitmeyer (2002).

Auto-generated from the source docstrings by scripts/generate_api_docs.py (make api-docs). Do not edit by hand.

ship_source_spectrum

ship_source_spectrum(
    speed_knots: float = 12.0,
    length_m: float = 100.0,
    *,
    vessel_class: str = 'containership',
    model: str = 'jomopans-echo',
    frequency_hz: NDArray[np.float64] | list[float] | None = None,
) -> ShipTrafficSpectrum

Predicted underwater source-level spectrum of a ship.

Parameters

Name	Description
speed_knots	Vessel speed, in knots (used by "jomopans-echo" and "randi"; ignored by "wales-heitmeyer").
length_m	Vessel length, in metres (used by "jomopans-echo" and "randi"; ignored by "wales-heitmeyer").
vessel_class	JOMOPANS-ECHO vessel class (see VESSEL_CLASSES); used only by "jomopans-echo".
model	"jomopans-echo" (default), "randi" or "wales-heitmeyer".
frequency_hz	Frequencies, in Hz; defaults to the decidecade bands 10 Hz-31.5 kHz of the JOMOPANS-ECHO validation range.

Returns: A ShipTrafficSpectrum.

Raises

Exception	When
ValueError	If model/vessel_class is unknown or an input is invalid.

ShipTrafficSpectrum

ShipTrafficSpectrum(
    frequency: NDArray[np.float64],
    source_psd: NDArray[np.float64],
    band_level: NDArray[np.float64],
    model: str,
    vessel_class: str | None,
    speed_knots: float | None,
    length_m: float | None,
)

Predicted ship source-level spectrum.

Attributes

Name	Description
frequency	Frequencies, in Hz.
source_psd	Source pressure spectral-density level, in dB re 1 µPa²/Hz at 1 m (equivalent monopole).
band_level	Decidecade-band source level, in dB re 1 µPa m (source_psd + 10·log10(0.231·f)).
model	The model used.
vessel_class	The vessel class (JOMOPANS-ECHO only; else None).
speed_knots	Speed used, in knots (None if the model ignores it).
length_m	Length used, in metres (None if the model ignores it).

ShipTrafficSpectrum.plot()

ShipTrafficSpectrum.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Plot the predicted source spectral-density level versus frequency.

VESSEL_CLASSES

Constant (tuple).

VESSEL_CLASSES = ('bulker', 'containership', 'cruise', 'dredger', 'fishing', 'government/research', 'naval', 'other', 'passenger', 'recreational', 'tanker', 'tug', 'vehicle carrier')