Fishbone in Multi-Disciplinary Science

Updated 4 July 2026

Fishbone is a structural concept featuring a central spine with branching elements, serving as a versatile modeling paradigm across diverse scientific disciplines.
It captures phenomena such as energetic-particle-driven instabilities in fusion plasmas, nonlocal effective potentials in nuclear physics, and flexure-torsion dynamics in bridge mechanics.
The model advances simulation techniques and stability control in areas ranging from GRMHD accretion flows to biomimetic robotics and massive IoT networking.

Fishbone denotes several distinct technical objects across plasma physics, nuclear few-body theory, suspension-bridge mechanics, general relativistic accretion theory, spectroscopy, robotics, and networking. In tokamak research, it most commonly refers to an energetic-particle-driven, internal-kink-like instability; in nuclear cluster models, to a nonlocal effective potential that simulates Pauli effects; in structural mechanics, to a reduced deck-and-ribs model for bridge flexure–torsion coupling; and in GRMHD, to the Fishbone–Moncrief equilibrium torus used as an accretion-flow initial condition (Keeling et al., 2014, Day et al., 2011, Uniyal et al., 2024).

1. Terminological scope

The term is used for structurally analogous but physically unrelated constructions: a spine-and-ribs pattern, a finite-rank nonlocal operator with a “fish-bone” matrix structure, or an internal-kink mode whose experimental spectrogram resembles a fishbone-like waveform. The dominant research usage is in fusion plasmas, but the same label is well established in several other technical literatures.

Domain	Referent	Defining feature
Fusion plasmas	Fishbone instability	Energetic-particle-driven internal kink-like mode, usually $n=1$ (Zou et al., 2021)
Nuclear cluster physics	Fishbone potential	Nonlocal effective interaction simulating the Pauli effect (Day et al., 2011)
Suspension bridges	Fish-bone model	Coupled flexural–torsional deck model with rib-like cross-sections (Marchionna et al., 2022)
GRMHD	Fishbone–Moncrief torus	Equilibrium thick torus in a stationary, axisymmetric spacetime (Uniyal et al., 2024)
ATAS of graphene	Fish bone resonance structure	Periodic V-shaped spectral branches around central resonances (Dong et al., 2022)
Biomimetic robotics	Fishbone-like ribs	Distributed ribcage around a rigid spine with tunable rigidity/flexibility (He et al., 22 May 2025)
Massive IoT	Fishbone forwarding	Main axis and sub-axes used as a dissemination skeleton (Seong et al., 2022)

A recurrent conceptual pattern is the use of a central backbone with side branches or sideband structure. In plasma physics this pattern is dynamical rather than geometric; in the other fields it is often explicitly structural.

2. Fishbone in fusion-plasma physics

In tokamaks, fishbone usually denotes an energetic-particle-driven internal kink instability, typically with $(n,m)=(1,1)$ , driven by resonance between the mode and energetic-particle orbital motion (Zou et al., 2021). The classical ion-fishbone picture describes a low-frequency $n=1$ internal kink destabilized by fast ions near the $q=1$ surface, while electron-driven fishbones arise when the same underlying internal kink interacts resonantly with energetic electrons, particularly barely trapped and barely passing populations (Merle, 2013). A key theoretical point is that the energetic-particle contribution modifies the effective potential energy of the internal kink; for fast electrons, inclusion of the passing-particle term $(q-1)\omega_b$ in the resonance condition explains relatively low observed mode frequencies (Merle, 2013).

Experiments on MAST identified fishbones as core-localized, chirping $n=1$ modes in the $20\text{–}50$ kHz range, with each burst producing sharp drops in neutron emission and therefore strong redistribution of neutral-beam fast ions (Keeling et al., 2014). In the FIDA study on MAST, low-frequency chirping modes below $50$ kHz were associated with strong redistribution in both real and velocity space, with core FIDA reductions that may reach $50\%$ and depletion extending up to about $95\%$ of the beam injection energy (Jones et al., 2013). Those measurements also showed that fishbone effects are not spatially uniform: MAST neutron-camera and FIDA diagnostics both indicated predominantly core-localized redistribution, with the strongest impact near the magnetic axis or slightly off-axis in the core (Keeling et al., 2014, Jones et al., 2013).

A common misconception is that all fishbones are simply the original PDX-type $(n,m)=(1,1)$ 0 mode. MAST reversed-shear plasmas showed internal-kink-like fishbones in equilibria with $(n,m)=(1,1)$ 1, broad $(n,m)=(1,1)$ 2 structure, and no clear $(n,m)=(1,1)$ 3 surface, and these were identified with an $(n,m)=(1,1)$ 4 infernal kink–ballooning mode rather than a classic PDX fishbone (Jones et al., 2013). Likewise, on HL-2A, hybrid kinetic–MHD simulations found that when $(n,m)=(1,1)$ 5 approaches one, the dominant instability can transition from fishbone to TAE or BAE, so “fishbone” and “Alfvén eigenmode” are not always sharply separated regimes (Zou et al., 2021).

KSTAR introduced a further complication with the “double-peaked fishbone”: a fishbone-like burst and chirp with coherent core and edge activities separated by weaker or even undetectable activity in the intermediate region (Lee et al., 25 Mar 2026). In that case, the edge electron-temperature fluctuation becomes more correlated with the magnetic signal as fishbone strength increases, and the phase of the edge fluctuation precedes the phase of the core fluctuation except in very weak cases (Lee et al., 25 Mar 2026). This suggests that edge activity is not necessarily a passive consequence of a core event.

3. Saturation, transport, control, and prediction in fusion applications

Fishbones are often treated as deleterious because they redistribute or eject energetic particles, but recent work shows a dual role. Global gyrokinetic simulations of DIII-D found that self-generated zonal flows dominate nonlinear saturation by preventing coherent structures from persisting or drifting in energetic-particle phase space when the mode frequency down-chirps; with zonal flows included, simulated saturation amplitude and energetic-particle transport agreed quantitatively with experiment for the first time (Brochard et al., 2023). The same study argued that fishbone-induced zonal flows were likely responsible for the formation of an internal transport barrier after fishbone bursts in that discharge, and an ITER PFPO baseline showed insignificant energetic-particle redistribution (Brochard et al., 2023).

A related cross-scale result was obtained for EAST ITER-like hybrid plasmas: global gyrokinetic simulations including both electromagnetic ITG turbulence and $(n,m)=(1,1)$ 6 fishbone instability showed that fishbone-driven zonal radial electric fields suppress electromagnetic ITG turbulence, reducing ion thermal transport close to the neoclassical level (Ma et al., 6 Nov 2025). In that picture, fishbone acts as a macro-scale source of zonal flows, and those zonal flows regulate microturbulence. A second misconception therefore becomes untenable: fishbones are not always purely harmful to confinement.

Operational control remains central because strong fishbones still reduce effective heating and current drive. On MAST, the dominant experimental control knobs were line-averaged density and neutral-beam power: high density and low-to-moderate power suppressed fishbones, while lower density and higher power increased $(n,m)=(1,1)$ 7, mode growth rate, and neutron deficits (Keeling et al., 2014). For MAST-Upgrade, a simple neutral-beam geometry change was proposed: tilting the on-axis double-beam-box injector upward so that its tangency point lies above the midplane. In simulations, $(n,m)=(1,1)$ 8 at a representative density differed by more than a factor of two between $(n,m)=(1,1)$ 9 mm and $n=1$ 0 mm tilt, and the $n=1$ 1 mm tilt moved low-density three-beam operation toward the drive level of the two-beam Core Scope case predicted to be safe (Keeling et al., 2014).

Machine-learning surrogates have been developed to accelerate linear-stability assessment. A database of $n=1$ 2 M3D-K hybrid kinetic–MHD simulations was generated by scanning four parameters—central total beta, fast-ion pressure fraction, central $n=1$ 3, and the radius of the $n=1$ 4 surface—and nonlinear-kernel SVMs achieved accuracy $n=1$ 5 for fishbone instability classification and $n=1$ 6 for growth rate, real frequency, and mode structure prediction (Liu et al., 2024). This suggests a practical route toward fast scenario scans and control-oriented reduced models.

4. Fishbone potentials in nuclear cluster physics

In few-body nuclear physics, the fishbone potential is a nonlocal effective interaction designed to simulate the Pauli effect between composite clusters by nonlocal terms (Day et al., 2011). The central idea is that when composite clusters such as $n=1$ 7 particles interact, antisymmetrization produces Pauli-forbidden or Pauli-hindered relative-motion states. Rather than encoding this entirely in a strongly repulsive local core, the fishbone model uses norm-kernel eigenstates $n=1$ 8 and a finite-rank nonlocal correction with a characteristic “fish-bone” matrix structure to remove forbidden states and suppress partially forbidden ones (Day et al., 2011).

For the $n=1$ 9 system, the revisited fishbone potential used a local part

$q=1$ 0

with the fitted parameters $q=1$ 1, $q=1$ 2, and $q=1$ 3 (Day et al., 2011). That parametrization was fitted simultaneously to the $q=1$ 4 resonance, $q=1$ 5 phase shifts, and three- $q=1$ 6 energies, and the authors concluded that essentially a simple Gaussian could describe both two- $q=1$ 7 and three- $q=1$ 8 data without invoking a three-body potential (Day et al., 2011). The physical implication is that a large part of what is often absorbed into phenomenological three-body forces can be represented by a properly constructed two-body nonlocal Pauli correction.

For $q=1$ 9 and $(q-1)\omega_b$ 0, the refined fish-bone potential again treats Pauli effects through nonlocal terms, but the local part is a double Gaussian plus spin–orbit interaction and a smeared Coulomb term for $(q-1)\omega_b$ 1 (Smith et al., 2012). With parameters $(q-1)\omega_b$ 2, $(q-1)\omega_b$ 3, $(q-1)\omega_b$ 4, $(q-1)\omega_b$ 5, $(q-1)\omega_b$ 6, $(q-1)\omega_b$ 7, and $(q-1)\omega_b$ 8, the model described the $(q-1)\omega_b$ 9 and $n=1$ 0 phase shifts for all partial waves considered (Smith et al., 2012). Here the nonlocal fishbone operator replaces much of the ad hoc angular-momentum dependence usually built into local optical potentials.

5. Fish-bone models of suspension bridges

In bridge mechanics, the fish-bone model is a simplified structural idealization of a suspension-bridge deck and its interaction with hangers and cables, constructed to capture coupled flexural and torsional motion (Marchionna et al., 2022). The deck is represented by a mid-line displacement $n=1$ 1 and a torsional rotation $n=1$ 2, with the side edges displaced by $n=1$ 3; after small-angle approximation and modal reduction, one obtains a finite-dimensional system for flexural and torsional amplitudes. This framework was used to show that, for vibrations of sufficiently large amplitude, transfer of energy from flexural modes to torsional modes may occur provided a certain condition on the parameters is satisfied (Marchionna et al., 2022). The high-energy analysis reduces the torsional stability problem to a Hill equation, and the limiting Floquet discriminant supplies an explicit instability criterion.

A newer nonlinear PDE version studies the fish-bone bridge model with rigid hangers, Woinowsky–Krieger geometric nonlinearity, and flow effects (Falocchi et al., 2024). In that system, the primary unknowns are the downward deck displacement $n=1$ 4 and torsional rotation $n=1$ 5, coupled through nonlocal cable–hanger operators $n=1$ 6 and $n=1$ 7, together with the nonlocal stretching term

$n=1$ 8

A non-conservative potential-flow approximation adds aerodynamic forcing, and the analysis establishes well-posedness of weak solutions, uniform stability conditions, and the existence of a compact global attractor under all nonlinear and non-conservative effects (Falocchi et al., 2024).

These bridge models are not merely mnemonic uses of the word. The “fish-bone” designation refers to a deck mid-line with many rotating cross-sections, i.e. a spine with ribs, and the term is therefore geometric in the literal sense.

6. The Fishbone–Moncrief torus in GRMHD

In relativistic accretion theory, “Fishbone” usually refers to the Fishbone–Moncrief torus, a classic analytic equilibrium solution for a thick, pressure-supported fluid torus orbiting a rotating black hole (Uniyal et al., 2024). It is the standard initial condition for many GRMHD simulations of accretion flows used in Event Horizon Telescope modeling. The construction assumes a stationary, axisymmetric spacetime, a perfect barotropic fluid, no poloidal motion, and typically constant specific angular momentum; with a polytropic equation of state $n=1$ 9, the density and pressure follow from the integrated relativistic Euler equation once the inner edge $20\text{–}50$ 0 and pressure maximum $20\text{–}50$ 1 are fixed (Uniyal et al., 2024).

The revisited formulation shows that the Fishbone–Moncrief construction extends cleanly beyond Kerr to any stationary, axisymmetric, asymptotically flat vacuum or non-vacuum spacetime that can be written in Boyer–Lindquist-like form (Uniyal et al., 2024). The paper gives the generalized metric decomposition

$20\text{–}50$ 2

and a generalized expression for the constant angular momentum $20\text{–}50$ 3 at the pressure maximum. Hydrodynamic simulations in the Johannsen–Psaltis and Kerr–Sen spacetimes then showed that these generalized FM tori remain stable under long-term evolution up to $20\text{–}50$ 4 (Uniyal et al., 2024).

The significance is methodological. Consistent GRMHD simulations in non-Kerr metrics require equilibrium initial conditions compatible with the background geometry, and the generalized FM torus provides exactly that. A plausible implication is that comparisons among Kerr and non-Kerr image libraries can be made with less contamination from arbitrary initial-data choices.

7. Other technical extensions

In attosecond transient absorption spectroscopy of graphene, “fish bone” refers to a spectral pattern rather than a dynamical mode or structural model (Dong et al., 2022). The pattern consists of slowly varying zeroth-order resonances and delay-periodic, tilted, V-shaped first-order branches, producing a spine-and-ribs appearance in $20\text{–}50$ 5. The simplified $20\text{–}50$ 6-point model expresses the ATAS spectrum as a sum of zeroth- and first-order Bessel-function contributions, and the fish bone structure is explained by sidebands at $20\text{–}50$ 7; the periodicity is $20\text{–}50$ 8, and the tilt angle is determined solely by the pump frequency (Dong et al., 2022).

In biomimetic robotics, SpineWave uses a fishbone-like internal architecture: a rigid vertebral core surrounded by expandable rib-like elements with adjustable magnets (He et al., 22 May 2025). The fishbone-like ribs act as distributed elastic supports that shape body curvature, store and release elastic energy, and stabilize the traveling wave generated by servo-driven joints. In experiments, constrained joints were limited to $20\text{–}50$ 9 per joint, whereas unconstrained magnet-assisted joints reached $50$0, and straight-line swimming speed improved from $50$1 to $50$2 after optimization (He et al., 22 May 2025). Here “fishbone” denotes a mechanical design principle rather than a visual resemblance alone.

In massive IoT networks, FiFo—Fishbone Forwarding—builds a forwarding skeleton consisting of a main axis and sub-axes within each cluster of devices (Seong et al., 2022). Devices are first clustered by UPGMA, and the main axis and sub-axes are then created using the expectation-maximization algorithm for the Gaussian mixture model and principal component analysis. The performance metric is forwarding efficiency,

$50$3

defined as the ratio of coverage probability to the average number of transmissions per device (Seong et al., 2022). On the real-world dataset used in the study, FiFo outperformed epidemic routing, probabilistic flooding, neighbor-based probabilistic broadcast, and modified BIP in forwarding efficiency (Seong et al., 2022).

Across these disparate literatures, “fishbone” therefore functions as a high-level structural label for a backbone-with-branches topology, a resonant pattern with a central spine and side ribs, or a finite-rank operator whose matrix resembles that geometry. The fusion-plasma usage remains the most established scientific meaning, but the term’s portability across fields reflects a shared geometric intuition rather than a shared underlying physics.