Parabolic Equations for Long Range Wave Simulation

17 concepts11 papers1 proposal

When modeling how waves travel over long distances—whether they are acoustic waves from an offshore wind turbine traveling across the ocean or electromagnetic signals bouncing through the atmosphere—standard numerical methods quickly hit a computational wall.

The fundamental challenge stems from the nature of the classical wave equation. Equations like the Helmholtz equation are elliptic boundary value problems. To solve them, you must compute the wave field across the entire domain simultaneously. If your simulation spans several kilometers and your wavelength is measured in meters or centimeters, the resulting computational grid becomes massive, requiring the solution of an impossibly large system of linear equations.

While ray-tracing offers a faster alternative, it fails to capture critical wave phenomena like diffraction, ground impedance, and complex refraction. To bridge this gap, researchers rely on the Parabolic Equation (PE) method. The core intuition behind the PE method is a elegant mathematical transformation: it converts a global boundary value problem into a directional "marching" problem, drastically reducing computational complexity while preserving full wave mechanics.

The Core Intuition: From Boundary Value to Marching Problem

To understand why the parabolic approximation works, imagine a wave propagating predominantly along a single axis—let's call it the $x$-axis. Instead of tracking every rapid oscillation of the wave field $u(x,y,z)$, we can factor out the primary forward-moving carrier wave:

$$u(x,y,z) = \psi(x,y,z)e^{ik_0x}$$

Here, $k_0$ is a reference wavenumber, and $\psi$ represents the envelope or the slowly varying amplitude of the wave.

Because the wave is traveling primarily forward, the envelope $\psi$ changes very slowly along the $x$-axis. This physical reality allows us to make a crucial mathematical approximation: we assume that the second derivative of the envelope with respect to the propagation distance ($\partial^2\psi/\partial x^2$) is negligible compared to the first derivative ($2ik_0 \partial\psi/\partial x$).

By dropping this second derivative, the elliptic Helmholtz equation transforms into a parabolic partial differential equation. Structurally, this new equation is identical to the heat equation or the time-dependent Schrödinger equation, with one critical difference: the spatial coordinate $x$ now plays the role of time.

This change completely alters how the equation is solved. Instead of solving the entire multi-kilometer domain at once, we can start with the wave field at the source ($x=0$) and "march" the solution forward step-by-step along the $x$-axis.

Efficient Marching via the Split-Step Fourier Method

Once the wave equation is cast into a parabolic form, it can be solved step-by-step using highly efficient numerical techniques. Chief among these is the Split-Step Fourier Method (SSFM).

The intuition behind the SSFM is to divide each small forward step $\Delta x$ into two distinct physical mechanisms that are treated independently:

  1. The Diffractive Step: The wave propagates through free space. In the spatial domain, calculating diffraction involves complex derivatives. However, by applying a Fast Fourier Transform (FFT), the wave is moved into the wavenumber domain where diffraction simplifies to a straightforward phase rotation.
  2. The Refractive Step: The wave encounters variations in the medium, such as changes in the atmospheric refractive index or fiber-optic properties. This step is applied directly in the spatial domain by multiplying the wave field by a localized refractive phase factor.

By alternating between the spatial and wavenumber domains using the FFT, the SSFM achieves remarkable computational speed. As noted by Knut Solna in Background velocity estimation with cross correlations of incoherent waves in the parabolic scaling, this split-step framework provides a highly practical alternative to full wave equations when analyzing propagation through complex random media.

The versatility of this marching mechanism extends across disciplines. For instance, Milad Sefidgaran et al. leverage the SSFM in On the Capacity of the Continuous-Space SSFM Model of Optical Fiber to isolate and simulate how nonlinear and dispersive effects accumulate segment-by-segment along an optical fiber. In the context of maritime telecommunications, Wenjing Liu et al. implement the SSFM in Performance analysis of code division multiplexing communication under evaporation duct environment to efficiently simulate how electromagnetic waves remain trapped and propagate within non-uniform atmospheric humidity gradients over the ocean surface.

Dimensional Reduction via Mode Parabolic Equations

When waves propagate through bounded channels—such as acoustic waves in a shallow sea—marching a full 3D grid can still become computationally burdensome. To streamline this, researchers combine the parabolic approximation with modal decomposition, creating Mode Parabolic Equations (MPE).

Instead of discretizing the entire vertical water column on a 3D grid, the acoustic field is broken down into a set of discrete, local vertical normal modes (similar to the resonant harmonics of a vibrating string). The wave equation is then solved only for the horizontal propagation of these modes.

As demonstrated by D. S. Manul’chev et al. in Estimating Sound Exposure Levels Due to a Broadband Source over Large Areas of Shallow Sea, MPEs provide a massive computational advantage over full 3D parabolic models. By reducing the dimensionality of the problem, the authors show that it becomes feasible to map broadband acoustic noise pollution across vast marine environments efficiently.

Overcoming the Narrow-Angle Limitation

The classic parabolic approximation relies on a strict assumption: the wave must travel almost perfectly parallel to the marching axis. If a wave refracts sharply or scatters off a steep boundary, it travels at a wide angle relative to the $x$-axis, and the standard narrow-angle PE approximation breaks down.

To overcome this, advanced formulations introduce wide-angle parabolic equations. Mathematically, this involves replacing the simple square-root approximation of the differential operators with more sophisticated pseudodifferential operators or rational approximations.

Xavier Antoine addresses this in Pseudodifferential adiabatic mode parabolic equations in curvilinear coordinates and their numerical solution by developing a wide-angle Pseudodifferential Adiabatic Mode Parabolic Equation (PDMPE) framed in curvilinear coordinates. This approach allows the marching axis to bend along with the natural geometry of the environment, successfully capturing complex phenomena like whispering gallery waves and severe horizontal refraction.

The practical necessity of wide-angle modeling is highlighted by D. S. Manul’chev et al., who observed that ignoring wide-angle effects in shallow-water environments with sloping bottoms can lead to errors of up to 8 dB in sound exposure level predictions.

Capturing Real-World Environmental Dynamics

Because parabolic equations natively incorporate range-dependent media properties and complex boundary conditions, they have become the standard tool for evaluating how real-world weather and geometry alter wave propagation.

Atmospheric Refraction and Noise Mapping

In environmental acoustics, sound waves rarely travel in straight lines; temperature inversions and wind gradients create a continuously shifting refractive index that bends sound back toward or away from the ground. Dick B.M. Botteldooren et al. have extensively utilized the Green’s Function Parabolic Equation (GF-PE) to capture these non-linear atmospheric effects.

In Influence of temporal resolution of meteorological and traffic data on long-term average sound levels and Influence of correlation between diurnal traffic pattern and meteorological conditions on long-term average Lden (invited paper), the authors demonstrate that simplified regulatory averaging methods fail to account for the temporal correlation between traffic spikes and atmospheric refraction, leading to significant errors in long-term noise mapping. By using the wave-based GF-PE framework, the authors accurately simulate how diurnal weather patterns interact with ground impedance. Furthermore, in Modeling outdoor sound propagation in mountainous areas (invited abstract), the authors apply this method to rugged terrains, showing how slope winds and temperature inversions interact to bend sound waves over mountain ridges.

Offshore and Marine Simulations

The PE framework is equally vital for assessing airborne noise over the ocean. In studies evaluating the environmental impact of offshore wind farms, such as Propagation distance-of-concern for offshore wind turbine airborne sound during piling and normal operation and Airborne sound propagation over sea during offshore wind farm piling, the authors use parabolic equations to model how low-frequency construction noise travels over kilometers of open water. Their simulations revealed counter-intuitive physical insights, such as the fact that rough, wind-driven sea surfaces attenuate airborne piling noise more effectively than a calm sea by scattering the sound waves upon boundary impact.

High-Frequency Radio Attenuation

The utility of the parabolic wave equation (PWE) is not restricted to acoustics. In Physics-Based Trajectory Design for Cellular-Connected UAV in Rainy Environments Based on Deep Reinforcement Learning, Hao Qin et al. apply the PWE to high-frequency radio wave propagation. Instead of relying on loose empirical models, the authors use the PWE to deterministically simulate how raindrops attenuate cellular signals in a given geometric environment, providing a robust, physics-based foundation for optimizing drone flight paths in adverse weather.

Key Foundational Work

Influence of correlation between diurnal traffic pattern and meteorological conditions on long-term average Lden (invited paper) (2006) — contributed 1 concept elements to this direction.
Influence of temporal resolution of meteorological and traffic data on long-term average sound levels (2007) — contributed 1 concept elements to this direction.
Modeling outdoor sound propagation in mountainous areas (invited abstract) (2006) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_mode_parabolic_equations_87783e81,
  title = {Mode Parabolic Equations},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/mode-parabolic-equations/rdr_87783e81a3654cee82f29a9ceb659110},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-22},
}

This exact version (version 1, frozen text):

@misc{lacuna_mode_parabolic_equations_87783e81_v1,
  title = {Mode Parabolic Equations},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/mode-parabolic-equations/rdr_87783e81a3654cee82f29a9ceb659110/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-22},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake