betaFeedback

WorksPhysics and Astronomy

The kinematic Kelvin-Helmholtz instability-driven dynamo: implications for merging compact objects

Abstract

Binary neutron star (BNS) mergers excite a Kelvin–Helmholtz instability (KHI) in the shear layer between them. It is parasitized upon by the magnetic field to form a powerful magnetic dynamo. Using a large ensemble of high-resolution visco-resistive KHI-driven dynamo simulations we construct analytical, asymptotic in Reynolds number (Re\rm{Re}) models for the magnetic field growth rates and mean vertical field dynamo onset times. Our models suggest that in the Re≫1\rm{Re} \gg 1 limit, the mean vertical field is seeded immediately, kinematic growth rates become asymptotic in Re\rm{Re}, and the dynamo moves out of the kinematic stage on O(0.1 ms)\mathcal{O}(0.1\,\rm{ms}) timescales, readily attaining magnetar-strength fields during this transition. This suggests that most global BNS merger simulations are not time-resolving the correct KHI-driven dynamo regime.

Introduction

The merger of two compact objects leads to the formation of an intense shear layer that is susceptible to the Kelvin–Helmholtz instability (KHI) [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. In binary neutron-star (BNS) mergers the shear layer has a thickness of order δ0∼50 m\delta_0 \sim50\,\mathrm{m} and shear velocity of ΔU∼0.3c\Delta U \sim0.3c, where cc is the speed of light, leading to a shear timescale of tΔ=δ0/ΔU∼10−3 mst_\Delta= \delta_0/\Delta U \sim10^{-3}\,\mathrm{ms} [10]. Simulations show that coupling to the KHI amplifies magnetic fields from B∼1012 GB \sim10^{12}\,\mathrm{G} to B∼1015−16 GB \sim10^{15-16}\,\mathrm{G} in magnetic within a few [1, 2, 3, 4, 5, 6, 8, 9, 10]. These strong, magnetar-strength fields [9] can power relativistic jets and potentially connect the shear-layer dynamics (after further large-scale dynamo action [11, 12, 13]) to the electro-magnetic counterpart of BNS merger gravitational-wave events [11, 14, 15, 16, 17, 18, 19, 20, 21].

BNS merger simulations provide evidence for the type of dynamo that is present – a kinematic KHI-driven turbulence dynamo [1, 4, 7, 9, 10, 22], followed by a large-scale Taylor–Spruit [12, 13, 23] or αΩ\alpha\Omega dynamo on long post-merger timescales [11, 24]. In a standard kinematic (Ekin≫Emag=B2/2E_{\rm kin} \gg E_{\rm mag} = B^2/2, where EkinE_{\rm kin} is the characteristic kinetic energy of the system) turbulence dynamo, EmagE_{\rm mag} grows exponentially without any Lorentz force backreaction to modify the viscous stretching [25, 26, 27, 28]. If the plasma is in a state of Kolmogorov-type turbulence Ekin(k)∝k−5/3E_{\rm kin}(k) \propto k^{-5/3} (which is the case in BNS mergers [10, 29]), where Ekin(k)E_{\rm kin}(k) is the kinetic energy spectrum and kk is the wavenumber, the prediction for the exponential growth rate at high magnetic Prandtl number Pm\mathrm{Pm}, relevant to BNS mergers [7, 30], is γEt0∝Re1/2\gamma_E t_0 \propto\mathrm{Re}^{1/2}, where Re=u0ℓ0/ν\mathrm{Re} = u_0\ell_0/\nu is the plasma Reynolds number with outer-scale characteristic velocity u0u_0, length ℓ0\ell_0 and kinematic viscosity ν\nu [28, 31, 32]. Another signature of a dynamo driven by fluctuations is contained within the magnetic energy spectrum, Emag(k)E_{\rm mag}(k).

In Fourier space, the Kazantsev–Kulsrud–Anderson theory predicts Emag(k)∝e2γEt/t0k3/2K0(k/kη)E_{\rm mag}(k) \propto e^{2\gamma_E t/t_0} k^{3/2}K_0(k/k_\eta) [33, 34] by prescribing a random Gaussian velocity field, where K0(x)K_0(x) is the modified Bessel function of the second kind and kη∼kpeakk_\eta\sim k_{\rm peak} is the Ohmic resistive cutoff and peak Emag(k)E_{\rm mag}(k) scale [31, 32, 33, 34, 35]. Both kpeakk_{\rm peak} and kηk_\eta are determined by kη/kν∝Pm1/2k_\eta/k_\nu\propto\mathrm{Pm}^{1/2}, where kνk_\nu is the viscous wavenumber and the Pm1/2\mathrm{Pm}^{1/2} scaling comes from stretching on kνk_\nu, balancing dissipation on kηk_\eta. Detailed local simulations have confirmed these relations [28, 31, 36, 37]. This makes the turbulence dynamo uniquely challenging for global simulations because they do not converge without explicit resistivity (kη∼kpeakk_{\eta} \sim k_{\rm peak}) and viscosity (γEt0∝Re1/2\gamma_{\rm E}t_0 \propto{\rm Re}^{1/2}), the full and consistent inclusion of which has just become possible in GR [38] (see also Ref. [39, 40] for resistive-only approaches). Whether the KHI dynamo fits into this framework is an open and important question for understanding the nature of BNS merger dynamos [7, 10, 30]. Answering it requires controlled first-principles simulations.

In this Letter we unravel the complete kinematic KHI dynamo chronology. We find that the KHI-driven dynamo is not a turbulence dynamo, in that the kinematic growth rates become independent of Re. Further, within the kinematic regime the dynamo moves through three different stages: (1) winding under planar shear, (2) secondary, three-dimensional instability-driven break-up and (3) fully-fledged KHI-driven dynamo, beyond planar shear, including a vertical field dynamo with an onset time ∝Re−1/3\propto{\rm Re}^{-1/3}. For BNS merger parameters, we demonstrate that the dynamo immediately seeds a vertical mean field, growth rates become asymptotic (for kν>δ0−1k_{\nu} > \delta_{0}^{-1}, a condition not readily met in global simulations) and the dynamo passes through the kinematic stage and becomes nonlinear on timescales of O(0.1 ms)\mathcal{O}(0.1\ {\rm ms}), much shorter than the dynamical merger time of the neutron stars.

Numerical Models

We evolve the three-dimensional, compressible, Newtonian MHD equations with explicit shear viscosity and Ohmic resistivity using the HLLD Riemann solver [41] with first-order-flux correction [42] and fifth-order TENO5 flux reconstruction [43] in athenaK [44]. In a periodic domain of size LL we initialize two freely decaying, perturbed shear layers with half-width δ0=0.025L\delta_0 = 0.025L, velocity jump ΔU\Delta U, density contrast 8/38/3, and uniform pressure [45]. We add a random, three-dimensional magnetic seed with Emag,0≈10−6Ekin,0E_{\rm mag,0} \approx10^{-6}E_{\rm kin,0}, where Ekin,0E_{\rm kin,0} is the initial kinetic energy of the shear layer. We define ReΔ=ΔUδ0/ν{\rm Re}_{\Delta} = \Delta U\delta_0/\nu, magnetic Reynolds number RmΔ=PmΔReΔ{\rm Rm}_{\Delta} = {\rm Pm}_{\Delta}{\rm Re}_{\Delta}, where PmΔ=ν/η{\rm Pm}_{\Delta} = \nu/\eta, and shear Mach number MΔ=2ΔU/(cs,in+cs,out)\mathcal{M}_{\Delta} = 2\Delta U/(c_{\rm s,in} + c_{\rm s,out}), where cs,inc_{\rm s,in} and cs,outc_{\rm s,out} are the sound speeds on either side of the shear layer. Our simulations span 1,1523≤Ngrid3≤2,30431{,}152^3 \le N_{\rm grid}^3 \le2{,}304^3 grid cells, 64≤ReΔ≤16,38464 \le{\rm Re}_{\Delta} \le16{,}384 at MΔ=0.5,1,2\mathcal{M}_{\Delta} = 0.5, 1, 2 and PmΔ=1{\rm Pm}_{\Delta} = 1, such that viscous and resistive scales are resolved at all times. At ReΔ=1,024{\rm Re}_{\Delta} = 1{,}024 and MΔ=0.5\mathcal{M}_{\Delta} = 0.5, we additionally vary PmΔ=2,5,10{\rm Pm}_{\Delta} = 2, 5, 10. Our fiducial runs are integrated for 103tΔ10^3t_{\Delta} and use Ngrid3=1,1523N_{\rm grid}^3 = 1{,}152^3. The End Matter gives the governing equations, initial conditions and numerical methods. The Supplemental Material [46] details the growth-rate measurements and spectral diagnostics.

Results

Volume-integrated magnetic-energy evolution—Figure 1 shows Emag=∫B2/2 dVE_{\rm mag} = \int B^2/2\,{\rm d}V, normalized by its initial value. At high ReΔ{\rm Re}_{\Delta}, amplification proceeds through an extremely rapid initial stage (stage 1; dashed line), a pause

Figure 1. Volume-integrated magnetic-energy evolution for the KHI-driven dynamo with MΔ=1\mathcal{M}_{\Delta} = 1 and PmΔ=1{\rm Pm}_{\Delta} = 1. Magnetic energy, EmagE_{\rm mag}, is normalized by the initial magnetic energy (left axis) and initial kinetic energy (right axis). Additional lines mark stage 1 (winding; dashed), stage 2 (amplification pause; dotted), and stage 3 (exponential growth; dot-dashed). The aqua 1σ1\sigma, 2σ2\sigma and 3σ3\sigma curves show the evolution of the energy fluctuations in the ReΔ=16,384{\rm Re}_{\Delta} = 16{,}384 simulation. The nonlinear and saturation labels mark the stages that come after the kinematic dynamo.

(stage 2; dotted line), and a second period of rapid growth (stage 3; dot-dashed line) before transitioning into the nonlinear growth and saturation (∂tEmag≈0\partial_t E_{\rm mag} \approx0) at Emag≈10−2EkinE_{\rm mag} \approx10^{-2}E_{\rm kin}. The same amplification stages appear in many global BNS simulations [2, 9, 11, 47]. We examine each stage below to identify its amplification mechanism and the conditions governing transitions.

Stage 1: initial amplification by mean-shear stretching—Within a few hundred tΔt_{\Delta}, the first stage of the ReΔ≫1{\rm Re}_{\Delta} \gg1 KHI dynamo results in Emag/Emag,0≳103E_{\rm mag}/E_{\rm mag,0} \gtrsim10^3. For a BNS merger, this is equivalent to reaching B∼1016 GB \sim10^{16}\ {\rm G} in t∼0.1 mst \sim0.1\ {\rm ms} (following [7]). We outline in the following how this is from the initial winding of the resistive magnetic field by the viscous mean-shear layer.

For ideal (RmΔ→∞{\rm Rm}_{\Delta} \to\infty) induction in the stationary initial shear U0(x)U_0(x), the energy obeys Emag(t)/Emag,0=1+a(t/tΔ)2E_{\rm mag}(t)/E_{\rm mag,0} = 1 + a(t/t_{\Delta})^2, where a=2tΔ⟨∂xU0Bx0By0⟩/⟨B02⟩a = 2t_{\Delta}\langle\partial_x U_0 B_{x0}B_{y0}\rangle/\langle B_0^2\rangle and b=tΔ2⟨(∂xU0)2Bx02⟩/⟨B02⟩b = t_{\Delta}^2\langle(\partial_xU_0)^2B_{x0}^2\rangle/\langle B_0^2\rangle, with angle brackets denoting volume averages, and subscript 0, initial values (see Supplemental Material [46]). However, viscosity broadens the shear layer and magnetic flux is not perfectly frozen into the plasma, so the ideal induction equation is insufficient.

Including the viscous broadening of the layer and Ohmic resistive effects gives

Emag(t)Emag,0≃e−αt/(RmΔtΔ)[1+attΔ+b(t/tΔ)21+2(t/tΔ)(κν/ReΔ+κη/RmΔ)×exp⁡(−β(t/tΔ)3RmΔ[1+2(t/tΔ)(κν/ReΔ+κη/RmΔ)])](1)\frac{E_{\mathrm{mag}}(t)}{E_{\mathrm{mag},0}} \simeq e^{-\alpha t/(\mathrm{Rm}_{\Delta}t_{\Delta})}\left[1+a\frac{t}{t_{\Delta}}+\frac{b(t/t_{\Delta})^2}{\sqrt{1+2(t/t_{\Delta})(\kappa_{\nu}/\mathrm{Re}_{\Delta}+\kappa_{\eta}/\mathrm{Rm}_{\Delta})}}\times\exp\left(-\frac{\beta(t/t_{\Delta})^3}{\mathrm{Rm}_{\Delta}[1+2(t/t_{\Delta})(\kappa_{\nu}/\mathrm{Re}_{\Delta}+\kappa_{\eta}/\mathrm{Rm}_{\Delta})]}\right)\right] \tag*{(1)}

with coefficients (α,β,κν,κη)(\alpha,\beta,\kappa_{\nu},\kappa_{\eta}) calculated analytically from the initial B\mathbf{B} field and shear profiles (Supplemental Material [46]). The model recovers purely ideal winding as ReΔ,RmΔ→∞\mathrm{Re}_{\Delta},\mathrm{Rm}_{\Delta}\to\infty. We compare slopes from Equation (1) and the simulations, γwind\gamma_{\mathrm{wind}}, over the same 1≤t/tΔ≤101\le t/t_{\Delta}\le10 interval in Figure 2 (a). The model provides excellent agreement with the measured stage 1 γstretch\gamma_{\mathrm{stretch}} across all sampled MΔ\mathcal{M}_{\Delta}, ReΔ\mathrm{Re}_{\Delta}, and PmΔ\mathrm{Pm}_{\Delta}. We therefore identify stage 1 as non-ideal winding by the mean-shear layer, which is insufficient for a sustained dynamo [48, 49].

Figure 2. Initial magnetic amplification by shear winding. (a) Magnetic-energy growth rates γwind\gamma_{\mathrm{wind}} measured during stage 1 in Figure 1. Symbols distinguish MΔ\mathcal{M}_{\Delta}, and uncertainty bars show sensitivity to the measurement interval. Dashed black curves show the analytical model, Equation (1), evaluated over the same interval. The inset varies PmΔ\mathrm{Pm}_{\Delta} at ReΔ=1,024\mathrm{Re}_{\Delta}=1,024 and MΔ=0.5\mathcal{M}_{\Delta}=0.5. (b) Magnetic spectra towards the end of stage 1, normalized to the initial magnetic energy, for MΔ=0.5\mathcal{M}_{\Delta}=0.5 and PmΔ=1\mathrm{Pm}_{\Delta}=1. The dashed line shows the predicted k0k^0 winding dependence (Equation (2)). In both panels, red markers and curves denote the 2,30432,304^3 run at ReΔ=16,384\mathrm{Re}_{\Delta}=16,384, MΔ=0.5\mathcal{M}_{\Delta}=0.5, and PmΔ=1\mathrm{Pm}_{\Delta}=1, for comparison with the 1,15231,152^3 simulation.

Winding transfers magnetic energy to higher kk. In a locally constant shear, induction gives k≃∣∂xU∣t∣ky0∣k\simeq|\partial_xU|t|k_{y0}| and wound-mode energy ∝t2\propto t^2. These determine the spectral transport derived in the Supplemental Material. An isotropic, solenoidal seed concentrated initially near k0k_0 admits to

Emag(k,t)≃Emag,04k0∣∂xU∣t∝k0,(2)E_{\mathrm{mag}}(k,t)\simeq\frac{E_{\mathrm{mag},0}}{4k_0}|\partial_xU|t\propto k^0, \tag*{(2)}

for k0≪k≪min⁡[∣∂xU∣tk0,(3/2ηt)1/2]k_0\ll k\ll\min[|\partial_xU|tk_0,(3/2\eta t)^{1/2}], where ∣∂xU∣tk0|\partial_xU|tk_0 is the expanding front of wound modes and kη=(3/2ηt)1/2k_{\eta}=(3/2\eta t)^{1/2}. In the ideal limit, the spectral height and bandwidth each grow as tt, recovering Emag∝t2E_{\mathrm{mag}}\propto t^2. Hence, as the magnetic field is stretched by the mean-shear, it populates all of the available kk linearly in time and uniformly in Emag(k)E_{\mathrm{mag}}(k) (see Figure 2 b), which sets the initial magnetic field conditions for the next fully-fledged dynamo stage.

Stage 2: secondary RTI and three-dimensional breakup— As the KHI becomes nonlinear, billows wind the density interface into columns of dense and light fluid along zz. Further, each billow supplies an effective centrifugal acceleration uθ2/Ru_{\theta}^2/R, where uθu_{\theta} is its circulation speed and RR its radius. Small displacements, potentially seeded by the three-dimensional BB field, grow via the Rayleigh–Taylor instability (RTI), disrupting the billows [50, 51, 52]. Figure 3 shows high-kk RTI fingers in the wound density layers, alongside magnetic-field stretching, B^iB^j∂jui\hat{B}_i\hat{B}_j\partial_j u_i, where B^iB^j\hat{B}_i\hat{B}_j is the unit stress tensor. This shows the RTI excites kzk_z velocity modes, breaking the planar symmetry of the KHI [53]. The resulting three-dimensional motions facilitate both kinematic large- and small-scale dynamo action, examined below.

Three-dimensional instability of KHI billows

Figure 3. Three-dimensional instability of KHI billows. All panels show mean-normalized mass density (left) and field-aligned stretching, i.e., dynamo flux, tΔB^iB^j∂juit_{\Delta}\hat{B}_{i}\hat{B}_{j}\partial_{j}u_{i} (right) at t/tΔ=140t/t_{\Delta}=140, ReΔ=8,192\mathrm{Re}_{\Delta}=8{,}192, PmΔ=1\mathrm{Pm}_{\Delta}=1. (a) a) xyxy slices at z=0z=0, −L/2<x<0-L/2<x<0; (b) yzyz slices at x=−L/4x=-L/4 (dashed lines in a), 0<y<L/20<y<L/2. (c,d) The same transverse patch enlarged to reveal Rayleigh–Taylor instability fingers within the billows. Streamlines trace (Bx,By)(B_x,B_y) in (a) and (By,Bz)(B_y,B_z) in (b,d). Fly-throughs of the unstable billows are available online in 8K resolution here.

Under RTI, shorter wavelengths grow faster (γRTI∝k\gamma_{\mathrm{RTI}}\propto\sqrt{k}), but viscosity damps them more strongly (∝−k2\propto-k^2).

With local Atwood number AA [54] and destabilizing acceleration geff≃uθ2/Rg_{\mathrm{eff}}\simeq u_{\theta}^2/R normal to the interface, the linear theory exposes this competition as

γRTItΔ≃[Ageffδ0(ΔU)2(kδ0)]1/2−(kδ0)2ReΔ.(3)\gamma_{\mathrm{RTI}}t_{\Delta}\simeq\left[\frac{Ag_{\mathrm{eff}}\delta_0}{(\Delta U)^2}(k\delta_0)\right]^{1/2}-\frac{(k\delta_0)^2}{\mathrm{Re}_{\Delta}}. \tag*{(3)}

Maximizing γRTI\gamma_{\mathrm{RTI}} to find the fastest-growing mode gives

γRTI∞tΔ=344/3[Ageffδ0(ΔU)2]2/3ReΔ1/3.(4)\gamma_{\mathrm{RTI}}^{\infty}t_{\Delta}=\frac{3}{4^{4/3}}\left[\frac{Ag_{\mathrm{eff}}\delta_0}{(\Delta U)^2}\right]^{2/3}\mathrm{Re}_{\Delta}^{1/3}. \tag*{(4)}

Hence, at fixed geffg_{\mathrm{eff}} and AA, increasing ReΔ\mathrm{Re}_{\Delta} permits smaller, faster-growing disturbances. We will next show that the ReΔ1/3\mathrm{Re}_{\Delta}^{1/3} growth-rate scaling sets the subsequent dynamo onset time.

Stage 3: delayed vertical-field amplification— In our geometry, the mean vertical field, BzB_z, relevant for some jet-launching scenarios [14] (though subsequent Parker processes can convert toroidal field [18, 55]), is orthogonal to the xyxy plane containing the initial flow and its shear gradient. Averaging over yzyz planes isolates B‾z(x,t)\overline{B}_z(x,t), the (kx,0,0)(k_x,0,0) component. We then measure the layer-averaged ⟨B‾z2⟩x\langle\overline{B}_z^2\rangle_x. Figure 4 (a) shows ⟨B‾z2⟩\langle\overline{B}_z^2\rangle growing after the initial winding phase. Because the original shear cannot amplify BzB_z by winding alone, this later growth provides evidence for a vertical-field dynamo in the three-dimensional flow.

Delayed amplification of the large-scale vertical field

Figure 4. Delayed amplification of the large-scale vertical field. (a) Energy of the yzyz-averaged field (kx,ky=0,kz=0)(k_x,k_y=0,k_z=0), ⟨Bz‾ 2⟩/(2Ekin,0)\langle\overline{B_z}^{\,2}\rangle/(2E_{\mathrm{kin},0}), for MΔ=1\mathcal{M}_{\Delta}=1 and PmΔ=1\mathrm{Pm}_{\Delta}=1. Dots mark the growth onset tdelayt_{\mathrm{delay}}. The inset shows logarithmic growth rates, with a reference line at γztΔ=0.02\gamma_z t_{\Delta}=0.02. (b) Onset delays tdelay/tΔt_{\mathrm{delay}}/t_{\Delta} at PmΔ=1\mathrm{Pm}_{\Delta}=1. Lines show the RTI prediction tdelay∝ReΔ−1/3t_{\mathrm{delay}}\propto\mathrm{Re}_{\Delta}^{-1/3} fitted for ReΔ≤4,096\mathrm{Re}_{\Delta}\leq4{,}096. The inset varies PmΔ\mathrm{Pm}_{\Delta} at MΔ=0.5\mathcal{M}_{\Delta}=0.5 and ReΔ=1,024\mathrm{Re}_{\Delta}=1{,}024. Uncertainty bars show fitting sensitivity (see Supplemental Material [46]).

The dynamo growth rate, γz=dln⁡⟨Bz2⟩/dt∼0.02/tΔ\gamma_z = d\ln\langle B_z^2\rangle/dt \sim0.02/t_\Delta, depends weakly on ReΔ\mathrm{Re}_\Delta across all three MΔ\mathcal{M}_\Delta ensembles (inset Figure 4a). Meanwhile, the velocity spectra Ekin(k)E_{\rm kin}(k) extend to higher kk as ReΔ\mathrm{Re}_\Delta increases and follow a standard k−5/3k^{-5/3} scaling (Supplemental Figure S5), consistent with a shifting kνk_\nu and shorter tνt_\nu for higher ReΔ\mathrm{Re}_\Delta. This suggests that kνk_\nu motions do not control the amplification as they do in the standard kinematic turbulence dynamo [36, 37].

We show in Figure 4b that the onset occurs earlier at successively larger ReΔ\mathrm{Re}_\Delta. We interpret the delay as the time required for kzk_z modes to break planar symmetry and enable sustained dynamo growth [48, 49]. For RTI-driven onset, billows first form over trollt_{\rm roll}, after which the fastest-growing kzk_z modes grow from ξ0\xi_0 to the disruptive amplitude ξnl\xi_{\rm nl}. The resulting delay is

tdelaytΔ≃ln⁡(ξnl/ξ0)γRTI∞tΔ∝ReΔ−1/3,(5)\frac{t_{\rm delay}}{t_\Delta} \simeq\frac{\ln(\xi_{\rm nl}/\xi_0)}{\gamma_{\rm RTI}^{\infty}t_\Delta} \propto\mathrm{Re}_\Delta^{-1/3}, \tag*{(5)}

because γRTI∞tΔ∝ReΔ1/3\gamma_{\rm RTI}^{\infty}t_\Delta\propto\mathrm{Re}_\Delta^{1/3}. Figure 4b provides a test for this scenario, where we plot tdelay/tΔt_{\rm delay}/t_\Delta as a function of ReΔ\mathrm{Re}_\Delta. We show that this provides excellent agreement with the data, with different normalizations for MΔ\mathcal{M}_\Delta. At ReΔ=1,024\mathrm{Re}_\Delta= 1{,}024, onset time shrinks further as PmΔ\mathrm{Pm}_\Delta increases from 1 to 10 (inset in Figure 4b), indicating that the magnetization (Emag∝PmΔ2/3E_{\rm mag} \propto\mathrm{Pm}_\Delta^{2/3} by the end of the winding, shown in the End Matter) also shortens tdelay/tΔt_{\rm delay}/t_\Delta. The Supplemental Material [46] details the fits and vertical-field induction equations.

Stage 3 continued: small-scale field amplification— Three-dimensional motions excited during secondary breakup enable small-scale magnetic amplification alongside the mean-field dynamo. After the pause, which shortens with increasing RTI growth rates, EmagE_{\rm mag} grows exponentially, with the spectra approaching the standard Kazantsev-Kulsrud-Anderson prediction Emag(k)∝k3/2E_{\rm mag}(k) \propto k^{3/2}, shown in Figure 5b. Standard theory then gives kpeak∼kηk_{\rm peak} \sim k_\eta [31, 36].

Figure 5. Magnetic amplification during the second growth stage. (a) Exponential magnetic-energy growth rates at PmΔ=1\mathrm{Pm}_\Delta=1. The inset shows signed Lorentz work per logarithmic Fourier shell for MΔ=1\mathcal{M}_\Delta=1 and PmΔ=1\mathrm{Pm}_\Delta=1, with colors as in (b). Shading marks δ0≤λ≤2δ0\delta_0\le\lambda\le2\delta_0, and top ticks mark the viscous scale. Uncertainty bars show sensitivity to the fitting interval (Supplemental Material [46]). Lines mark γEtΔ=0.02\gamma_Et_\Delta=0.02 (gray, dashed) and the turbulence dynamo γEtΔ\gamma_Et_\Delta prediction (gray, solid). (b) Magnetic spectra (solid) at 75% of each measured growth interval for MΔ=0.5\mathcal{M}_\Delta=0.5 and PmΔ=1\mathrm{Pm}_\Delta=1, normalized by the initial total magnetic energy and fit (dashed) to the resistive Kazantsev-Kulsrud-Anderson model, showing good agreement across all kk.

In Figure 5a we plot the growth rates, γE\gamma_E, by fitting the exponential growth segments for all simulations that exhibit this behavior. At ReΔ<512\mathrm{Re}_\Delta< 512, we find γEtΔ∼ReΔ1/2\gamma_Et_\Delta\sim\mathrm{Re}_\Delta^{1/2}, as predicted for M<1\mathcal{M}<1 turbulence dynamos [28, 36, 37, 56]. For ReΔ≥512\mathrm{Re}_\Delta\ge512 we find γEtΔ∼γztΔ∼0.02\gamma_Et_\Delta\sim\gamma_zt_\Delta\sim0.02, i.e., independent of ReΔ\mathrm{Re}_\Delta, with factor-of-a-few scatter in γE\gamma_E set by MΔ\mathcal{M}_\Delta.

We further compute the Lorentz work as a function of kk, WQ=⟨uQ⋅J×B/ρ⟩W_Q = \langle\boldsymbol{u}_Q \cdot\boldsymbol{J} \times\boldsymbol{B}/\rho\rangle, for logarithmically binned wavenumber shell QQ [57, 58]. We plot WQW_Q in inset Figure 5a and determine that for ReΔ≫1\mathrm{Re}_\Delta\gg1 the peak U→BU \to B flux shifts from kνk_\nu to ≈δ0−1\approx\delta_0^{-1}. Further, by measuring the kνk_\nu via the strain tensor spectrum [59], we find ReΔ≲256\mathrm{Re}_\Delta\lesssim256 simulations are in a kν≲(δ0)−1k_\nu\lesssim(\delta_0)^{-1} regime, and conversely for ReΔ≳512\mathrm{Re}_\Delta\gtrsim512. This implies an ReΔ\mathrm{Re}_\Delta independent kinematic growth rate that scales as B^˙iB^j∂jui∼tΔ−1∼ΔU/δ0\dot{\hat{B}}_i\hat{B}_j\partial_j u_i \sim t_\Delta^{-1} \sim\Delta U/\delta_0, with a dynamo efficiency typical of turbulence dynamos, O(10−2)O(10^{-2}) [28, 60].

Discussion

Now we discuss the implications for the dynamo in compact object mergers. To directly compare to our theory, let us first consider ReΔ\mathrm{Re}_\Delta for a BNS system. In dense parts of the merger, the viscosity is set by neutrinos and is of order νν∼107\nu_\nu\sim10^7–108 cm2 s−110^8\ {\rm cm^2\,s^{-1}} [61, 62, 63, 64]. With δ0=50 m\delta_{0}=50\ \mathrm{m} set by the scale height at the stellar surface, and ΔU=0.3c\Delta U=0.3c [10] gives Re⁡Δ≃4.5×105\operatorname{Re}_{\Delta}\simeq4.5\times10^{5}–4.5×1064.5\times10^{6} and tΔ≃0.56 μst_{\Delta}\simeq0.56\ \mu\mathrm{s}. We therefore use the high-Re⁡Δ\operatorname{Re}_{\Delta} limit of our model.

At high Re⁡Δ\operatorname{Re}_{\Delta}, magnetic energy grows by four orders of magnitude before nonlinear backreaction produces saturation at Emag/Ekin,0∼10−2E_{\mathrm{mag}}/E_{\mathrm{kin},0}\sim10^{-2}. Using B∼1017 G(Emag/Ekin,0/0.1)1/2B\sim10^{17}\ \mathrm{G}\left(E_{\mathrm{mag}}/E_{\mathrm{kin},0}/0.1\right)^{1/2} from [7], this corresponds to growing to a strength of B∼3×1016 GB\sim3\times10^{16}\ \mathrm{G} (as in [22]) in t≲0.5 mst\lesssim0.5\ \mathrm{ms}. Further, Figure 1 shows that the 3σ3\sigma energy fluctuations, occupying <1%<1\% of the volume, approach B∼1017 GB\sim10^{17}\ \mathrm{G}. The KHI-driven dynamo alone in BNS mergers can therefore create the strongest known magnetic fields in the Universe.

Driven mean-shear layers can generate a mean magnetic field aligned with the flow [65], but amplification perpendicular to the shear-flow plane has remained unclear. This vertical (seed) field is a critical ingredient for jet-launching in BNS [11, 66], likely in addition to subsequent large-scale dynamo action for long-lived merger remnants [11, 13, 19, 21, 24]. We demonstrate that the KHI generates such a field as three-dimensional modes become unstable to RTI, with onset following our tdelayt_{\mathrm{delay}} relation (Figure 4) and subsequent growth at γz∼0.02tΔ−1\gamma_{z}\sim0.02t_{\Delta}^{-1}, matching the mean-shear prediction from Figure 2. Using the [7] scaling and Figure 4, the saturated strength is Bz∼1014B_{z}\sim10^{14}–1015 G10^{15}\ \mathrm{G}. Furthermore, because tdelay∝Re⁡Δ−1/3t_{\mathrm{delay}}\propto\operatorname{Re}_{\Delta}^{-1/3} due to the visco-resistive RTI, disruption of the stage 1 winding ends almost immediately as Re⁡Δ→∞\operatorname{Re}_{\Delta}\to\infty [67]. Hence, in this limit, the KHI-driven dynamo proceeds directly to stage 3 where it grows both planar-averaged vertical (i.e., Bz(kx,ky=0,kz=0)B_{z}(k_{x},k_{y}=0,k_{z}=0)) and fluctuating (k≠0k\ne0) magnetic fields.

We show in both Figure 4 and Figure 5 that the dynamo growth rates become asymptotically independent of Re⁡Δ\operatorname{Re}_{\Delta}, with the peak in Lorentz work fixed to ≈δ0−1\approx\delta_{0}^{-1}. In this limit kν≳δ0−1k_{\nu}\gtrsim\delta_{0}^{-1}, which may be used as a criterion for γE\gamma_{E} convergence in global merger simulations.

Replacing the measured γE≃11 ms−1\gamma_{E}\simeq11\ \mathrm{ms}^{-1} [10] with our asymptotic γE≃36 ms−1\gamma_{E}\simeq36\ \mathrm{ms}^{-1} increases the magnetic energy by exp⁡[(36−11)Δt]\exp[(36-11)\Delta t] over the same interval Δt\Delta t, i.e., an additional factor of 100 within Δt≃0.18 ms\Delta t\simeq0.18\ \mathrm{ms}. This is sufficient to raise an initial Emag/Ekin∼10−3E_{\mathrm{mag}}/E_{\mathrm{kin}}\sim10^{-3} field well into the nonlinear dynamo regime Emag/Ekin∼10−1E_{\mathrm{mag}}/E_{\mathrm{kin}}\sim10^{-1} in only t≃0.18 mst\simeq0.18\ \mathrm{ms}. This suggests that global merger simulations are time-resolving the wrong KHI-driven dynamo stage, and either require properly calibrated subgrid recipes to skip stages 1 and 2 or resolve Δx=O(cm)\Delta x=O(\mathrm{cm}) such that δ0\delta_{0} and the fastest-growing RTI modes are resolved.

The same KHI dynamo mechanism may operate in unstable shear layers behind core-collapse supernova shocks [68] and in galaxy-cluster cold fronts [69]. In these hot, diffuse systems, where Re⁡Δ∝neT−5/2∼103\operatorname{Re}_{\Delta}\propto n_{e}T^{-5/2}\sim10^{3} [70], nen_{e} is the electron number density and Re⁡Δ\operatorname{Re}_{\Delta} is defined assuming collisional viscosity, the first stages of the kinematic KHI dynamo may become relevant.

Acknowledgments

We acknowledge productive discussions with Luciano Combi, Bindesh Tripathi, Uddipan Banik, Chris Thompson, Adam Burrows, Jim Stone, Jacob Fields, Alexander Dittman, Romain Teyssier and Christoph Federath. J. R. B., R. S. K. and S. C. acknowledge high-performance computing resources provided by the Leibniz Rechenzentrum and the Gauss Centre for Supercomputing through grants pr73fi, pn76gi and pn76ga, and GCS large-scale project 10391. J. R. B. acknowledges allocation rrg-essick from the Digital Research Alliance of Canada and funding from the Natural Sciences and Engineering Research Council of Canada (NSERC) (funding reference number 568580). J. R. B. is supported by NASA through NASA Hubble Fellowship grant HST-HF2-51597.001-A, awarded by the Space Telescope Science Institute and administered by AURA for NASA under contract NAS5-26555. J. R. B. and A. B. acknowledge support from NSF Award 2206756. E. R. M. acknowledges support from the U.S. National Science Foundation (NSF) under Grant No. PHY-2541792. E. R. M. is also supported by a Research Fellowship from the Sloan Foundation and a William H. Hurt Scholarship at Caltech. B. R. acknowledges the support of the Natural Sciences and Engineering Research Council of Canada (NSERC), funding reference numbers RGPIN-2023-04844 and ALLRP 592744. B. R. and A. P. are supported by a grant from the Simons Foundation (MP-SCMPS-00001470). B. R. is supported by the Canadian Space Agency (23JWGO2A01). R. S. K. acknowledges financial support from the ERC via Synergy Grant “ECO-GAL” (project ID 855130) and from DFG in project “STARCLUSTERS” (funding ID KL 1358/22-1). This research was supported in part by grant NSF PHY-2309135 to the Kavli Institute for Theoretical Physics (KITP).

The corresponding wavelength is

λRTI∞=2π(16ν2RAuθ2)1/3.(6)\lambda_{\mathrm{RTI}}^{\infty}=2\pi\left(\frac{16\nu^{2}R}{Au_{\theta}^{2}}\right)^{1/3}. \tag*{(6)}

Using ΔU=0.3c\Delta U=0.3c, uθ=ΔU/2u_{\theta}=\Delta U/2, δ0=50 m\delta_{0}=50\,\mathrm{m}, density ratio ρh/ρl=8/3\rho_{\mathrm{h}}/\rho_{\mathrm{l}}=8/3, Atwood number A=5/11≃0.4545A=5/11\simeq0.4545, and kinematic viscosity νν=107 cm2 s−1\nu_{\nu}=10^{7}\,\mathrm{cm}^{2}\,\mathrm{s}^{-1} [62, 63, 64, 73], for the BNS system we obtain λRTI∞≈6 cm\lambda_{\mathrm{RTI}}^{\infty}\approx6\,\mathrm{cm}, below the spatial resolution of current global and even local zoom-in simulations [10, 22].

End Matter

Equations and numerical methods.

We solve the compressible, Newtonian magnetohydrodynamics equations in the triply periodic domain [−L/2,L/2]3[-L/2,L/2]^3,

∂tρ+∇⋅(ρu)=0,(7)\partial_t \rho+ \nabla\cdot(\rho\mathbf{u}) = 0, \tag*{(7)}
∂t(ρu)+∇⋅[ρuu+(P+B2/2)I−BB]=∇⋅Π,(8)\partial_t(\rho\mathbf{u}) + \nabla\cdot\left[\rho\mathbf{u}\mathbf{u} + \left(P + B^2/2\right)\mathbf{I} - \mathbf{B}\mathbf{B}\right] = \nabla\cdot\mathbf{\Pi}, \tag*{(8)}
∂tE+∇⋅[(E+P+B2/2)u−(u⋅B)B]=∇⋅[Π⋅u+ηB×J].(9)\partial_t \mathcal{E} + \nabla\cdot\left[\left(\mathcal{E} + P + B^2/2\right)\mathbf{u} - (\mathbf{u}\cdot\mathbf{B})\mathbf{B}\right] = \nabla\cdot\left[\mathbf{\Pi}\cdot\mathbf{u} + \eta\mathbf{B}\times\mathbf{J}\right]. \tag*{(9)}
∂tB=∇×(u×B)+η∇2B,∇⋅B=0,(10)\partial_t \mathbf{B} = \nabla\times(\mathbf{u}\times\mathbf{B}) + \eta\nabla^2\mathbf{B}, \qquad\nabla\cdot\mathbf{B}=0, \tag*{(10)}

where J=∇×B\mathbf{J}=\nabla\times\mathbf{B}, E=ρϵ+ρu2/2+B2/2\mathcal{E}=\rho\epsilon+\rho u^2/2+B^2/2, and P=(γad−1)ρϵP=(\gamma_{\rm ad}-1)\rho\epsilon with γad=5/3\gamma_{\rm ad}=5/3. The magnetic permeability to set to unity. We include only shear viscosity, with stress tensor Πij=2ρνSij\Pi_{ij}=2\rho\nu S_{ij}, where Sij=(∂jui+∂iuj)/2−∇⋅uδij/3S_{ij}=(\partial_j u_i+\partial_i u_j)/2-\nabla\cdot\mathbf{u}\delta_{ij}/3. The constant coefficients are ν=ΔUδ0/ReΔ\nu=\Delta U\delta_0/\mathrm{Re}_{\Delta} and η=ν/PmΔ\eta=\nu/\mathrm{Pm}_{\Delta}. We use ATHENAK [44] with fifth-order targeted essentially non-oscillatory (TENO5) reconstruction [43], HLLD fluxes, third-order Runge–Kutta integration, constrained transport, and first-order flux correction. TENO5 reduces numerical diffusion by factors of approximately 1.5 relative to WENO-Z and 3.5 relative to PLM (used in [27, 74]). Hence, in terms of resolved modes, our fiducial ∼1,0003\sim1{,}000^3 domains are equivalent to ∼3,5003\sim3{,}500^3 domains for PLM and ∼1,5003\sim1{,}500^3 for WENO-Z. All simulations ran on LRZ’s SuperMUC-NG Phase 1 and SciNet’s Trillium, using approximately 10810^8 core-hours.

Shear, density, and thermodynamic state.

The layers are centered at x1,2=±L/4x_{1,2}=\pm L/4, following the smooth construction of Refs. [45, 75],

H(x)=12[tanh⁡(x+L/4δ0)−tanh⁡(x−L/4δ0)],(11)H(x)=\frac{1}{2}\left[\tanh\left(\frac{x+L/4}{\delta_0}\right)-\tanh\left(\frac{x-L/4}{\delta_0}\right)\right], \tag*{(11)}
U0(x)=ΔU[H(x)−12],ρ0(x)=ρout[1+53H(x)],(12)U_0(x)=\Delta U\left[H(x)-\frac{1}{2}\right], \qquad\rho_0(x)=\rho_{\rm out}\left[1+\frac{5}{3}H(x)\right], \tag*{(12)}

shown visually in Figure 6 (a) and (b). We use δ0/L=0.025\delta_0/L=0.025, uniform initial pressure, and a density contrast ρin/ρout=8/3\rho_{\rm in}/\rho_{\rm out}=8/3. In units with δ0=ΔU=⟨ρ⟩0=1\delta_0=\Delta U=\langle\rho\rangle_0=1, we choose P0P_0 to give the specified MΔ\mathcal{M}_{\Delta}. The in-plane velocity perturbation, visualized in Figure 6 (c), is

Figure 6. Initial conditions shared by all simulations. (a) The initial plasma density profile, e.g., the stratification for each shear layer. (b) The mean shear profile. (c) The transverse perturbation (inset) as well as damping function to isolate the perturbation the shear layer. (d) The magnetic energy power spectrum, Emag(k)E_{\rm mag}(k) of the initial seed magnetic field, shown in isotropic wavenumber kk.

ux(x,y)=0.05ΔU∑j=12e−(x−xj)2/δ02sin⁡(4πy/L).(13)u_x(x,y)=0.05\Delta U\sum_{j=1}^{2}e^{-(x-x_j)^2/\delta_0^2}\sin(4\pi y/L). \tag*{(13)}

The velocity is initially planar, with uz=0u_z=0. We prescribe no independent out-of-plane velocity perturbation. The weak three-dimensional Lorentz force can seed secondary perturbations that excite the RTI. In Supplemental § 0.11 we discuss how resistivity may affect this seeding and the measured onset delays.

Magnetic seed.

We initialize a divergence-free field, B=∇×A\mathbf{B} = \nabla\times\mathbf{A}, using a random vector potential A\mathbf{A} with a parabolic spectral envelope over 1<∣k∣L/(2π)<31 < \lvert\mathbf{k}\rvert L/(2\pi) < 3. We plot the Emag(k)E_{\mathrm{mag}}(k) of the field in Figure 6 (d). This places the magnetic energy at wavelengths ≫δ0\gg\delta_0, representing the merging stars’ initial dipole fields. We normalize the field to Emag,0/Ekin,0≃1.32×10−6E_{\mathrm{mag},0}/E_{\mathrm{kin},0} \simeq1.32 \times10^{-6}, placing it initially in the kinematic regime (Ekin≫EmagE_{\mathrm{kin}} \gg E_{\mathrm{mag}}), as expected initially in BNS mergers [7].

Magnetic-energy dependence on Pm.

At fixed ReΔ\mathrm{Re}_{\Delta}, increasing PmΔ=ν/η=RmΔ/ReΔ\mathrm{Pm}_{\Delta} = \nu/\eta= \mathrm{Rm}_{\Delta}/\mathrm{Re}_{\Delta} reduces resistive diffusion while leaving the viscosity unchanged. The insets of Figure 2 and Figure 4 show the PmΔ>1\mathrm{Pm}_{\Delta} > 1 growth rates and onset delays. Here we examine the EmagE_{\mathrm{mag}} evolution of the same simulations. Figure 7 (a) shows that more EmagE_{\mathrm{mag}} is generated directly after the winding at higher PmΔ\mathrm{Pm}_{\Delta}, and in general higher Emag/Emag,0E_{\mathrm{mag}}/E_{\mathrm{mag},0} within saturation, at least until the largest two PmΔ\mathrm{Pm}_{\Delta}. We articulate below how the higher winding energies may account for the differences in late-time saturated energy.

Figure 7. Magnetic amplification at varying PmΔ\mathrm{Pm}_{\Delta}. All runs have MΔ=0.5\mathcal{M}_{\Delta}=0.5 and ReΔ=1,024\mathrm{Re}_{\Delta}=1{,}024. Colors identify PmΔ\mathrm{Pm}_{\Delta} in both panels. (a) Total magnetic energy normalized by Emag,0E_{\mathrm{mag},0} (left axis) and Ekin,0E_{\mathrm{kin},0} (right axis). Open markers show mean pause energies; horizontal segments show the averaging intervals. The inset shows the energies as a function of PmΔ\mathrm{Pm}_{\Delta}, averaged over 20tΔ20t_{\Delta} around the pause in growth. The dashed curve shows Ewound,pause∝PmΔ2/3E_{\mathrm{wound,pause}} \propto\mathrm{Pm}_{\Delta}^{2/3}, predicted by Equation (15). (b) Energy of the yzyz-averaged vertical field, normalized by Ekin,0E_{\mathrm{kin},0}.

What follows is a simple argument that models the energy saturation for stage 1 (the winding; Figure 1) for different PmΔ\mathrm{Pm}_{\Delta}. Winding increases the magnetic energy as Ewound/Emag,0∼(t/tΔ)2E_{\mathrm{wound}}/E_{\mathrm{mag},0} \sim(t/t_{\Delta})^2 while reducing the transverse magnetic scale as ℓ⊥/ℓ⊥,0∼tΔ/t\ell_{\perp}/\ell_{\perp,0} \sim t_{\Delta}/t, where ℓ⊥,0\ell_{\perp,0} is the initial transverse magnetic scale. Here ℓ⊥\ell_{\perp} measures variation across the wound field, analogous to the reversal scale in the standard kinematic turbulence dynamo [31]. When resistive diffusion balances winding, ℓη∼ℓ⊥,0tΔ/tη\ell_{\eta} \sim\ell_{\perp,0}t_{\Delta}/t_{\eta}, halting transfer to higher kk. Using tη∼ℓη2/ηt_{\eta} \sim\ell_{\eta}^{2}/\eta gives tη∼ℓ⊥,02tΔ2/(tη2η)t_{\eta} \sim\ell_{\perp,0}^{2}t_{\Delta}^{2}/(t_{\eta}^{2}\eta) and hence

tηtΔ∼(ℓ⊥,02ηtΔ)1/3,(14)\frac{t_{\eta}}{t_{\Delta}} \sim\left(\frac{\ell_{\perp,0}^{2}}{\eta t_{\Delta}}\right)^{1/3}, \tag*{(14)}

and therefore

Ewound,peakEmag,0∼(tηtΔ)2∼(ℓ⊥,02ηtΔ)2/3∝PmΔ2/3,(15)\frac{E_{\mathrm{wound,peak}}}{E_{\mathrm{mag},0}} \sim\left(\frac{t_{\eta}}{t_{\Delta}}\right)^2 \sim\left(\frac{\ell_{\perp,0}^{2}}{\eta t_{\Delta}}\right)^{2/3} \propto\mathrm{Pm}_{\Delta}^{2/3}, \tag*{(15)}

where RmΔ=ΔUδ0/η=δ02/(ηtΔ)=ReΔPmΔ\mathrm{Rm}_{\Delta} = \Delta U\delta_0/\eta= \delta_0^2/(\eta t_{\Delta}) = \mathrm{Re}_{\Delta}\mathrm{Pm}_{\Delta} gives the PmΔ\mathrm{Pm}_{\Delta} proportionality. Higher PmΔ\mathrm{Pm}_{\Delta} (at fixed ReΔ\mathrm{Re}_{\Delta}) therefore allows the field to be wound into higher-kk modes before resistive truncation, as expected. The dashed curve in the inset of Figure 7 (a) shows that Ewound,peak∝PmΔ2/3E_{\mathrm{wound,peak}} \propto\mathrm{Pm}_{\Delta}^{2/3} agrees perfectly with the data.

The vertical-field energies in Figure 7 (b) depend weakly on PmΔ\mathrm{Pm}_{\Delta}, although their onset occurs earlier at higher PmΔ\mathrm{Pm}_{\Delta}, as discussed in the main text and shown in Figure 4. Because the KHI billows, where the winding occurs, become unstable on timescales ∝ReΔ−1/3\propto\mathrm{Re}_{\Delta}^{-1/3}

as ReΔ→∞\mathrm{Re}_{\Delta} \to\infty, the winding stage becomes so short-lived that we expect it to fundamentally disappear. We therefore suspect that this PmΔ2/3\mathrm{Pm}_{\Delta}^{2/3} dependence weakens at ReΔ≫1\mathrm{Re}_{\Delta} \gg1. Testing this requires a systematic study that is beyond the scope of this study.

The phases of the KHI-driven dynamo in k space at high Reynolds number

The main text follows magnetic spectra and volume-integrated energies across our parameter range. Here we focus on ReΔ=16,384\mathrm{Re}_{\Delta}=16,384, MΔ=1\mathcal{M}_{\Delta}=1, and PmΔ=1\mathrm{Pm}_{\Delta}=1. Here we focus on a single simulation, ReΔ=16,384\mathrm{Re}_{\Delta}=16,384, MΔ=1\mathcal{M}_{\Delta}=1, and PmΔ=1\mathrm{Pm}_{\Delta}=1. We plot the EmagE_{\mathrm{mag}} evolution and Emag(k)E_{\mathrm{mag}}(k) evolution in the top and bottom panels of Figure S1, respectively. We also show the EmagE_{\mathrm{mag}} evolution on a linear-linear axes in the inset of panel (a), which can be revealing for episodes of algebraic growth (e.g., [28, 37]).

Evolution of the volume-integrated magnetic energy and magnetic spectrum

Figure S1. Evolution of the volume-integrated magnetic energy (EmagE_{\mathrm{mag}}, panel a) and magnetic spectrum (Emag(k)E_{\mathrm{mag}}(k), panel b) for the MΔ=1\mathcal{M}_{\Delta}=1, ReΔ=16,384\mathrm{Re}_{\Delta}=16,384, PmΔ=1\mathrm{Pm}_{\Delta}=1 simulation. Annotations in panel (a) and the time color map in panel (b) identify the dynamo stages. During the nonlinear stage, magnetic energy shifts toward longer wavelengths [22, 25, 26, 28]. The k−9/5k^{-9/5} spectrum, recently measured in high-resolution MHD turbulence simulations, is plotted for reference [77, 78].

The first stage (blue) exhibits the winding spectrum, Emag(k)∝k0E_{\mathrm{mag}}(k) \propto k^{0} from Equation (1). Stage 2 marks the transition into a three-dimensional dynamo, via the RTI as discussed in the main text (and shown in Figure 3 and Figure S2). The RTI excites magnetic modes across kk space, from the long wavelengths to the short wavelengths, establishing the Emag(k)∝k3/2E_{\mathrm{mag}}(k) \propto k^{3/2} Kazantsev–Kulsrud–Anderson spectrum [33, 34] in stage 3. Stage 3 (yellow) progresses until reaching the nonlinear dynamo stage, where the scale-by-scale backreaction of the Lorentz on the flow suppresses the kinematic dynamo and makes the induction equation nonlinear in BB [22, 25, 26, 28, 76]. We will examine the nonlinear regime (red) in subsequent work. Here, magnetic energy shifts toward longer wavelengths, departing from Emag(k)∝k3/2E_{\mathrm{mag}}(k) \propto k^{3/2} and reaching slopes as steep as Emag(k)∝k−9/5E_{\mathrm{mag}}(k) \propto k^{-9/5} (annotated in Figure S1) within the inertial range [77, 78].

Figure S2. Vorticity-magnitude slices, ∣ω∣=∣∇×u∣|\boldsymbol{\omega}| = |\nabla\times\mathbf{u}|, for the Re⁡Δ=1,024\operatorname{Re}_{\Delta} = 1{,}024 simulation with enhanced magnetization to suppress secondary shear instabilities [79]. Coordinates are shown in the bottom-left corner of each panel. (a) ∣ω∣|\boldsymbol{\omega}| in the xyxy plane, showing mature KHI billows during stage 2 (Figure S1). (b) The x=−L/4x = -L/4 slice of ∣ω∣|\boldsymbol{\omega}| in the zyzy plane (dashed line in panel a), showing prominent, long-wavelength RTI modes growing along zz through the billows. The white circle highlights one RTI finger.

The three-dimensional RTI at moderate Reynolds number

Figure 3 shows extremely short-wavelength RTI modes, which grow fastest in the inviscid limit, γRTI∝k\gamma_{\mathrm{RTI}} \propto\sqrt{k}, disrupting the billows. Increasing viscosity lowers ReΔ\mathrm{Re}_{\Delta} and shifts the fastest-growing modes toward longer wavelengths. Further, by increasing the magnetization from Emag≈10−6Ekin,0E_{\mathrm{mag}} \approx10^{-6}E_{\mathrm{kin},0} to Emag,0≈10−3Ekin,0E_{\mathrm{mag},0} \approx10^{-3}E_{\mathrm{kin},0} we can suppress secondary shear instabilities that disrupt the RTI fingers [79]. Hence we run a Emag≈10−3Ekin,0E_{\mathrm{mag}} \approx10^{-3}E_{\mathrm{kin},0}, ReΔ=1,024\mathrm{Re}_{\Delta}=1,024, PmΔ=1\mathrm{Pm}_{\Delta}=1, MΔ=0.5\mathcal{M}_{\Delta}=0.5 simulation and show ∣ω∣=∣∇×u∣\lvert\boldsymbol{\omega}\rvert=\lvert\nabla\times\boldsymbol{u}\rvert in the xyxy and yzyz planes as three-dimensional modes become unstable in Figure S2.

The top panel shows large KHI billows with less secondary structure than at ReΔ=8,192\mathrm{Re}_{\Delta}=8,192 in Figure 3. The bottom panel shows the x=−L/4x=-L/4 slice, marked by the white dashed line above. Because the simulation is more

simulations, giving tdelay∝Re⁡Δ−1/3t_{\mathrm{delay}} \propto\operatorname{Re}_{\Delta}^{-1/3} from the viscous RTI growth-rate scaling, Equation (5).

Winding energy and spectrum

We derive the energy evolution and magnetic spectrum during stage 1, retaining only the planar shear u=U(x,t)ey\mathbf{u} = U(x,t)\mathbf{e}_y. We neglect the imposed perturbation, billow roll-up, three-dimensional motions, and magnetic feedback, and show that this model reproduces the growth rates and spectrum.

We assume an incompressible flow, ∇⋅u=0\nabla\cdot\mathbf{u} = 0, so the initial density ρ0(x)\rho_0(x) remains fixed. The initial profiles and constant viscosity ν\nu and resistivity η\eta are specified in the End Matter. The complete derivation, which includes every step and the coefficient evaluations, is available in Magnetic-energy evolution and spectrum under shear winding.

Ideal winding. For a stationary shear without resistivity, induction gives Bx=Bx0B_x = B_{x0}, Bz=Bz0B_z = B_{z0}, and By=By0+t(∂xU0)Bx0B_y = B_{y0} + t(\partial_x U_0)B_{x0}, with initial fields evaluated at (x,y−U0(x)t,z)(x,y-U_0(x)t,z). Squaring and averaging over the periodic volume gives Emag/Emag,0=1+a(t/tΔ)+b(t/tΔ)2E_{\mathrm{mag}}/E_{\mathrm{mag},0} = 1 + a(t/t_{\Delta}) + b(t/t_{\Delta})^2, where

a=2tΔ⟨(∂xU0)Bx0By0⟩⟨B02⟩.(16)a = 2t_{\Delta}\frac{\left\langle(\partial_x U_0)B_{x0}B_{y0}\right\rangle}{\left\langle B_0^2\right\rangle}. \tag*{(16)}
b=tΔ2⟨(∂xU0)2Bx02⟩⟨B02⟩.(17)b = t_{\Delta}^2\frac{\left\langle(\partial_x U_0)^2B_{x0}^2\right\rangle}{\left\langle B_0^2\right\rangle}. \tag*{(17)}

The linear term (aa) is the cross term between the initial and generated streamwise fields and can have either sign. The quadratic term (bb) is the generated field’s energy.

Resistive losses. Winding also creates magnetic gradients, which are subject to resistive loss. In a locally constant shear, a magnetic mode with initial wavevector p\mathbf{p} has kx=px−t(∂xU0)pyk_x = p_x - t(\partial_x U_0)p_y, ky=pyk_y = p_y, and kz=pzk_z = p_z. Resistivity multiplies its energy by exp⁡[−2η∫0tk2dt′]\exp[-2\eta\int_0^t k^2\mathrm{d}t'], where

∫0tk2dt′=p2t−(∂xU0)pxpyt2+13(∂xU0)2py2t3.(18)\int_0^t k^2\mathrm{d}t' = p^2t - (\partial_x U_0)p_xp_yt^2 + \frac{1}{3}(\partial_x U_0)^2p_y^2t^3. \tag*{(18)}

We represent diffusion of the initial field and of winding-created gradients by e−αηt/δ02e^{-\alpha\eta t/\delta_0^2} and e−βηt3/(δ02tΔ2)e^{-\beta\eta t^3/(\delta_0^2t_{\Delta}^2)}. Matching the initial resistive loss, −η⟨J02⟩-\eta\langle J_0^2\rangle, and the cubic loss weighted by the generated energy gives

α=2δ02⟨J02⟩⟨B02⟩,J0=∇×B0,(19)\alpha= 2\delta_0^2\frac{\left\langle J_0^2\right\rangle}{\left\langle B_0^2\right\rangle}, \qquad\mathbf{J}_0 = \nabla\times\mathbf{B}_0, \tag*{(19)}
β=2δ02tΔ2⟨(∂xU0)2∣∂yBx0∣2⟩3⟨(∂xU0)2Bx02⟩.(20)\beta= \frac{2\delta_0^2t_{\Delta}^2\left\langle(\partial_xU_0)^2|\partial_yB_{x0}|^2\right\rangle}{3\left\langle(\partial_xU_0)^2B_{x0}^2\right\rangle}. \tag*{(20)}

These exponentials approximate the distribution of local losses and omit the orientation-dependent t2t^2 term (see the full derivation). The additional cubic loss, which comes directly from Equation (18), is applied to the quadratic winding term.

Viscous layer broadening. Resistivity spreads the generated field, while viscosity weakens its source through ∂tU=νρ0−1∂x(ρ0∂xU)\partial_t U = \nu\rho_0^{-1}\partial_x(\rho_0\partial_xU). Taking Bx0B_{x0} approximately constant across an isolated layer, the induction equation gives

By,wound≃Bx0t{∂xU0+t2[η∂x3U0+ν∂x[ρ0−1∂x(ρ0∂xU0)]]}.(21)B_{y,\mathrm{wound}} \simeq B_{x0}t\left\{\partial_xU_0 + \frac{t}{2}\left[\eta\partial_x^3U_0 + \nu\partial_x\left[\rho_0^{-1}\partial_x(\rho_0\partial_xU_0)\right]\right]\right\}. \tag*{(21)}

The factor 1/21/2 follows from integrating the changing shear and diffusion over time. Squaring and integrating across the layer gives a relative energy correction 1−t(κνν+κηη)/δ02+⋯1-t(\kappa_\nu\nu+\kappa_\eta\eta)/\delta_0^2+\cdots. Integration by parts, with vanishing boundary terms at infinity, gives analytical expressions for the diffusion coefficients,

κη=δ02∫−∞∞(∂x2U0)2 dx∫−∞∞(∂xU0)2 dx(22)\kappa_\eta=\delta_0^2\frac{\int_{-\infty}^{\infty}(\partial_x^2U_0)^2\,\mathrm{d}x}{\int_{-\infty}^{\infty}(\partial_xU_0)^2\,\mathrm{d}x} \tag*{(22)}
κν=κη−δ022∫−∞∞(∂x2ln⁡ρ0)(∂xU0)2 dx∫−∞∞(∂xU0)2 dx.(23)\kappa_\nu=\kappa_\eta-\frac{\delta_0^2}{2}\frac{\int_{-\infty}^{\infty}(\partial_x^2\ln\rho_0)(\partial_xU_0)^2\,\mathrm{d}x}{\int_{-\infty}^{\infty}(\partial_xU_0)^2\,\mathrm{d}x}. \tag*{(23)}

We evaluate these coefficients using the tanh profiles with density contrast 8/38/3, neglecting layer overlap, and use them to calculate the growth rates in Figure 2.

Finite-time energy model. To extend the initial correction, we approximate the wound field by a fixed-shape profile of width w(t)w(t), preserving ∫By,wounddx=Bx0tΔU\int B_{y,\mathrm{wound}}\mathrm{d}x=B_{x0}t\Delta U. Its energy relative to ideal winding is then δ0/w(t)\delta_0/w(t). A diffusive width that matches the initial correction is

w2(t)=δ02+2t(κνν+κηη).(24)w^2(t)=\delta_0^2+2t(\kappa_\nu\nu+\kappa_\eta\eta). \tag*{(24)}

The broader layer also weakens the representative squared shear by δ02/w2(t)\delta_0^2/w^2(t), motivating the cubic-loss factor e−βηt3/[δ02w2(t)]e^{-\beta\eta t^3/[\delta_0^2w^2(t)]}. Multiplying the ideal polynomial by the initial-field loss, and its quadratic term by these broadening and cubic-loss factors, gives Equation (1) after substituting ν=δ02/(Re⁡ΔtΔ)\nu=\delta_0^2/(\operatorname{Re}\Delta t_\Delta) and η=δ02/(Rm⁡ΔtΔ)\eta=\delta_0^2/(\operatorname{Rm}\Delta t_\Delta).

All coefficients listed throughout this section are evaluated from the initial conditions of B\mathbf{B} or the shear profile. Through direct evaluation of Equation (16), Equation (17), Equation (19), Equation (20), Equation (23) and Equation (22) we find

a=−0.00106582,(25)a=-0.00106582, \tag*{(25)}
b=0.00345363,(26)b=0.00345363, \tag*{(26)}
α=0.236542,(27)\alpha=0.236542, \tag*{(27)}
β=0.00629781,(28)\beta=0.00629781, \tag*{(28)}
κν=152,0643,125ln⁡83−29,298625≃0.850822,(29)\kappa_\nu=\frac{152{,}064}{3{,}125}\ln\frac{8}{3}-\frac{29{,}298}{625}\simeq0.850822, \tag*{(29)}
κη=45.(30)\kappa_\eta=\frac{4}{5}. \tag*{(30)}

Ideal winding spectrum. The same mode evolution determines the energy per unit wavenumber, Emag(t)=∫Emag(k,t) dkE_{\mathrm{mag}}(t)=\int E_{\mathrm{mag}}(k,t)\,\mathrm{d}k. For ∣∂xU∣t≫1|\partial_xU|t\gg1 and k≫k0k\gg k_0, a magnetic mode has k≃∣∂xU∣tqk\simeq|\partial_xU|tq, where q=∣py∣q=|p_y|, while its generated energy grows as t2t^2. If F(q) dqF(q)\,\mathrm{d}q is the seed’s xx-component energy in qq, q+dqq+\mathrm{d}q, then

The xx-component energy in qq, q+dqq+\mathrm{d}q, then

Emag(k,t) dk≃(∂xU)2t2F(q) dq,(31)E_{\mathrm{mag}}(k,t)\,\mathrm{d}k\simeq(\partial_xU)^2t^2F(q)\,\mathrm{d}q, \tag*{(31)}
dk=∣∂xU∣t dq.(32)\mathrm{d}k=|\partial_xU|t\,\mathrm{d}q. \tag*{(32)}

For a homogeneous, isotropic seed concentrated at ∣p∣=k0|\mathbf{p}|=k_0, qq is uniform on [0,k0][0,k_0]. A solenoidal field lies perpendicular to p\mathbf{p}. With equal energy in the two directions in this plane, its xx-component fraction is (1−px2/k02)/2(1-p_x^2/k_0^2)/2. Averaging over azimuth about the flow direction at fixed qq gives ⟨px2⟩=(k02−q2)/2\langle p_x^2\rangle=(k_0^2-q^2)/2, hence a mean fraction [1+(q/k0)2]/4[1+(q/k_0)^2]/4. Thus

F(q)=Emag,04k0[1+(qk0)2],(33)F(q)=\frac{E_{\mathrm{mag},0}}{4k_0}\left[1+\left(\frac{q}{k_0}\right)^2\right], \tag*{(33)}
Emag(k,t)≃Emag,04k0∣∂xU∣t[1+(k∣∂xU∣tk0)2].(34)E_{\mathrm{mag}}(k,t)\simeq\frac{E_{\mathrm{mag},0}}{4k_0}|\partial_xU|t\left[1+\left(\frac{k}{|\partial_xU|tk_0}\right)^2\right]. \tag*{(34)}

The bracket approaches unity for k≪∣∂xU∣tk0k\ll|\partial_xU|tk_0, giving the k0k^0 spectrum in Equation (2). Spectral height and bandwidth each grow as tt, and the full expression integrates to the leading generated energy Emag,0(∂xU)2t2/3E_{\mathrm{mag},0}(\partial_xU)^2t^2/3. Restoring resistivity adds e−2ηk2t/3e^{-2\eta k^2t/3} from the cubic term in Equation (18), restricting the flat band to k0≪k≪min⁡[∣∂xU∣tk0,(3/2ηt)1/2]k_0\ll k\ll\min[|\partial_xU|tk_0,(3/2\eta t)^{1/2}].

Growth-rate measurements

Stage 1. We measure γwind\gamma_{\mathrm{wind}} in Figure 2(a) by fitting ln⁡Emag\ln E_{\mathrm{mag}} over 1≤t/tΔ≤101\le t/t_\Delta\le10, with a free intercept and equal weighting in time. We apply the same fit to the winding model. Uncertainty bars span fits over windows of width 1,2,3,4,5tΔ1,2,3,4,5t_\Delta, centered at each tΔt_\Delta within the 1≤t/tΔ≤101\le t/t_\Delta\le10 interval.

Stage 3. We measure stage-3 growth by fitting ln⁡Emag=c+γt\ln E_{\mathrm{mag}}=c+\gamma t, weighting all times equally. We identify sustained three-dimensional motions when ⟨uz2⟩1/2/ΔU≥0.003\langle u_z^2\rangle^{1/2}/\Delta U\ge0.003 persists for 20tΔ20t_\Delta. We locate the pause from minima in the growth rate measured over windows of 10,20,30tΔ10,20,30t_\Delta. For each width, we select the first minimum after 100tΔ100t_\Delta with a rate below 0.015/tΔ0.015/t_\Delta and a drop of at least 0.003/tΔ0.003/t_\Delta relative to the surrounding slopes. Growth fits begin at least 10tΔ10t_\Delta after the latest identified pause and after three-dimensional motions develop.

Accepted intervals span at least 20tΔ20t_\Delta and one energy e-folding. We require positive growth and an rms departure from the fitted ln⁡Emag\ln E_{\mathrm{mag}} of at most 0.050.05. The rates in the two halves must differ by at most 30%30\% of the full-interval rate, and each quarter-interval rate must lie within 30%30\% of it. Among qualifying intervals, we select the one with the largest magnetic-energy amplification. Uncertainty bars span fits to the full interval, its halves and quarters, and qualifying intervals with endpoints shifted by up to 5tΔ5t_\Delta. Hence, our uncertainty bars measure fitting sensitivity across a range of parameters around the interval.

Vertical-field onset and growth

Induction and geometry. Our analysis in the main text pertains to simulations for ReΔ≤4,096\mathrm{Re}_{\Delta} \leq4{,}096, where we have data for the large-scale field at high cadence. We average BzB_z over yzyz planes to obtain B‾z(x,t)\overline{B}_z(x,t), then average Bz2‾\overline{B_z^2} over −L/2<x<0-L/2 < x < 0, denoted by angle brackets. Averaging the induction equation over these planes gives

∂tB‾z=∂x(uzBx‾−uxBz‾)+η∂x2B‾z,(35)\partial_t\overline{B}_z = \partial_x\left(\overline{u_zB_x}-\overline{u_xB_z}\right)+\eta\partial_x^2\overline{B}_z, \tag*{(35)}

hence, the unperturbed shear, u=U0(x)ey\boldsymbol{u}=U_0(x)\boldsymbol{e}_y, supplies no source for B‾z\overline{B}_z. Its amplification therefore requires motions beyond the original shear, initially seeded by the RTI, as we show throughout the study.

Extracting onset times and uncertainty. We fit three joined straight segments to the logarithm of the mean vertical-field energy. The first change in slope defines tdelayt_{\mathrm{delay}}, and the middle slope gives γz\gamma_z. Each segment lasts at least 20tΔ20t_{\Delta}. The middle slope must be positive, exceed both neighboring slopes, and span at least one energy e-folding. Figure S3 shows the fits and onset uncertainties for MΔ=1\mathcal{M}_\Delta= 1. We determine uncertainty bars using the same “fit sensitivity” method as in stage 3.

Figure S3. Vertical-field onset and growth fits at MΔ=1\mathcal{M}_{\Delta}=1, PmΔ=1\mathrm{Pm}_{\Delta}=1, and 256≤ReΔ≤4,096256 \leq\mathrm{Re}_{\Delta} \leq4{,}096. Solid colored curves show ⟨Bz2‾⟩/(2Ekin,0)\langle\overline{B_z^2}\rangle/(2E_{\mathrm{kin},0}); black dashed curves show the three-segment fits to its logarithm. Solid vertical lines mark tdelayt_{\mathrm{delay}}, gray bands show the uncertainty ranges defined in the text. Each panel identifies ReΔ\mathrm{Re}_{\Delta} (top left) and the middle-segment growth rate γztΔ\gamma_z t_{\Delta} (bottom right).

Delay scaling. Neglecting the roll-up time, we fit tdelay/tΔ=CReΔ−1/3t_{\mathrm{delay}}/t_\Delta= C\mathrm{Re}_\Delta^{-1/3} over 256≤ReΔ≤4,096256 \leq\mathrm{Re}_\Delta\leq4{,}096, fixing the predicted exponent and fitting CC separately for each Mach ensemble with equal weighting in logarithmic delay. At fixed billow structure and initial disturbance, this hydrodynamic RTI estimate has no explicit PmΔ\mathrm{Pm}_\Delta dependence but reproduces the viscosity scaling in Figure 4.

Additional power spectra

Figure S4 shows the magnetic power spectra Emag(k)E_{\mathrm{mag}}(k) for all three MΔ\mathcal{M}_\Delta ensembles, colored by ReΔ\mathrm{Re}_\Delta. Early snapshots (left column) lie near 50tΔ50t_\Delta, and later spectra are extracted approximately halfway through the stage 3 kinematic growth interval. We normalize early spectra by Emag,0E_{\mathrm{mag},0}, and later spectra, by Emag(t)E_{\mathrm{mag}}(t). This allows us to compare spectral morphologies for different ReΔ\mathrm{Re}_\Delta and MΔ\mathcal{M}_\Delta. Reference lines show Emag(k)∝k0E_{\mathrm{mag}}(k) \propto k^0 on the left and Emag(k)∝k3/2E_{\mathrm{mag}}(k) \propto k^{3/2} on the right. All of the spectra behave as expected.

Figure S4. Magnetic spectra, Emag(k)E_{\mathrm{mag}}(k), at PmΔ=1\mathrm{Pm}_{\Delta}=1 for MΔ=0.5\mathcal{M}_{\Delta}=0.5 (a,b), 11 (c,d), and 22 (e,f). Left: spectra at t/tΔ≃50t/t_{\Delta}\simeq50, normalized by Emag,0E_{\mathrm{mag},0}. Right: spectra at the midpoints of the stage-3 growth intervals, normalized by Emag(t)E_{\mathrm{mag}}(t) to compare spectral shapes. Colors identify ReΔ\mathrm{Re}_{\Delta}. Dashed lines show the k0k^0 (winding spectrum; Equation (2)) and k3/2k^{3/2} slopes. The red dashed curve in (a) shows the 2,30432{,}304^3 comparison at ReΔ=16,384\mathrm{Re}_{\Delta}=16{,}384, at the same time and with the same normalization. Note not all of the simulations that have a winding phase become exponentially growing within the 103tΔ10^3t_{\Delta} that we simulate (e.g., either too viscous or never dynamo super-critical).

Figure S5 uses the same runs, times as colors as Figure S4. We sum velocity power over spherical shells to obtain Pu(k,t)P_u(k,t) and normalize by ⟨u2⟩(t)=∫Pu(k,t) dk\langle u^2\rangle(t) = \int P_u(k,t)\,\mathrm{d}k. In both columns we multiply by (kL/2π)5/3(kL/2\pi)^{5/3} to make a k−5/3k^{-5/3} slope horizontal. As expected from high-resolution global and zoom-in models of BNS mergers [9, 10], we find a Kolmogorov-style spectrum in the late time evolution, and a steep, spiky spectrum at early times. The spikes most likely originate from the harmonics of the initial shear spectrum. Under x→x+L/2x \to x + L/2, each Fourier mode acquires a factor (−1)nx(-1)^{n_x}. The antisymmetry Uy(x+L/2)=−Uy(x)U_y(x + L/2) = -U_y(x) therefore selects odd modes, nx=1,3,5,…n_x = 1, 3, 5, \ldots, which dominate the early-time spectrum. The same spiky initial spectra are shown in the zoom-in simulations in [10].

Figure S5. FIG. S5. Velocity spectra, Pu(k)P_u(k), for the runs and times in Figure S4, colored by ReΔ\mathrm{Re}_{\Delta}. Each spectrum is normalized by ⟨u2⟩(t)=∫Pu(k,t) dk\langle u^2\rangle(t) = \int P_u(k,t)\,\mathrm{d}k and multiplied by (kL/2π)5/3(kL/2\pi)^{5/3}, making the Kolmogorov slope horizontal (which is found in many high-resolution global and zoom-in BNS merger simulations, e.g., [9, 10] for the latest). The horizontal dashed line marks the compensated Pu(k)∝k−5/3P_u(k) \propto k^{-5/3} slope. The late-time spectra in the right column approximately follow this slope. The red dashed curve in (a) shows the same 2,30432{,}304^3 comparison as in Figure S4.

Volume renderings

We provide two volume renders of the KHI-driven dynamo: 360° fly-through and a magnetic field fly-through. Each rendering used PyVista [80] and approximately one hour on 1,000 cores of SuperMUC-NG Phase 1. Both volumetric visualizations are taken during the kinematic dynamo stage (stage 3) in the ReΔ=16,384\mathrm{Re}_\Delta= 16{,}384, PmΔ=1\mathrm{Pm}_\Delta= 1, MΔ=0.5\mathcal{M}_\Delta= 0.5 simulation on the 1,15231{,}152^3 grid.

The first is a three-dimensional fly-through where we rendered a full spherical view for every frame, allowing viewers to rotate the view in 3D as the camera moves through a billow. In Figure S6 we show three frames as the camera rotates 360° around a great circle, and moves through the billow. The plasma density in the billow is shown in orange, and the magnetic field streamlines in white. The magnetic field is dominated by small-scale structure, as expected during stage 3.

Three-dimensional volume rendering at movie time 0 s

(a) Movie time: 0 s

Three-dimensional volume rendering at movie time 48 s

(b) Movie time: 48 s

Three-dimensional volume rendering at movie time 65 s

(c) Movie time: 65 s

Figure S6. Three frames from the three-dimensional volume rendering of the KHI-driven dynamo. The camera passes through one billow, turns, and follows a second. Yellow–red shows mass density. The rendering uses a fixed snapshot taken during the transition to the stage-3 kinematic dynamo. White lines trace the magnetic field. In this stage the spectrum is transitioning between Emag(k)∝k0E_{\mathrm{mag}}(k) \propto k^{0} (winding spectrum; Equation (2)) and Emag(k)∝k3/2E_{\mathrm{mag}}(k) \propto k^{3/2} (Kazantsev-Kulsrud-Anderson spectrum [33, 34]), and so the total magnetic field is energetically dominated by the small-scale (high-kk) fluctuations.

:::

The second is a three-dimensional fly-through of a billow, where we fade in the magnetic field and then fade out the plasma. A second scene follows a magnetic field line along a smoothed camera path.

References

  1. [1]D. J. Price and S. Rosswog, Science 312, 719 (2006), arXiv:astro-ph/0603845 [astro-ph].DOI
  2. [2]M. Cahanov, S. D. Tootle, E. R. Most, and L. Rezzolla, Astrophys. J. Lett. 945, L14 (2023), arXiv:2211.13661 [astro-ph.HE].
  3. [3]C. Zhu, R. Pakmor, M. H. van Kerkwijk, and P. Chang, Astrophys. J. Lett. 808, L1 (2015).
  4. [4]K. Kiuchi, P. Cerdá-Durán, K. Kyutoku, Y. Sekiguchi, and M. Shibata, Physical Review D 92, 124034 (2015), arXiv:1509.09205 [astro-ph.HE].DOI
  5. [5]F. R. N. Schneider, S. T. Ohlmann, P. Podsiadlowski, F. K. Röpke, S. A. Balbus, et al., Nature 574, 211 (2019).
  6. [6]R. Pakmor, I. Pelisoli, S. Justham, A. S. Rajamuthukumar, F. K. Röpke, et al., Astronomy and Astrophysics 691, A179 (2024), arXiv:2407.02566 [astro-ph.SR].DOI
  7. [7]D. Radice and I. Hawke, Living Reviews in Computational Astrophysics 10, 1 (2024), arXiv:2402.03224 [astro-ph.HE].
  8. [8]S. T. Ohlmann, F. R. N. Schneider, F. K. Röpke, R. Pakmor, P. Podsiadlowski, et al., Astronomy & Astrophysics 713, A6 (2026), arXiv:2512.13424.
  9. [9]K. Kiuchi, A. Reboul-Salze, Y. Sekiguchi, and M. Shibata, arXiv e-prints, arXiv:2606.11299 (2026), arXiv:2606.11299 [astro-ph.HE].arxiv.org/abs/2606.11299
  10. [10]E. M. Gutiérrez, D. Radice, J. Fields, and J. M. Stone, arXiv e-prints 10.48550/arXiv.2601.20953 (2026), arXiv:2601.20953 [astro-ph.HE].DOI
  11. [11]K. Kiuchi, A. Reboul-Salze, M. Shibata, and Y. Sekiguchi, Nature Astronomy 8, 298 (2024).
  12. [12]A. Reboul-Salze, P. Barrère, K. Kiuchi, J. Guilet, R. Raynaud, et al., Astron. Astrophys. 699, A4 (2025), arXiv:2411.19328 [astro-ph.HE].
  13. [13]H. H.-Y. Ng and E. R. Most, (2026), arXiv:2607.28556 [astro-ph.HE].arxiv.org/abs/2607.28556
  14. [14]R. D. Blandford and R. L. Znajek, Mon. Not. R. Astron. Soc. 179, 433 (1977).
  15. [15]B. P. Abbott, R. Abbott, T. D. Abbott, F. Acernese, K. Ackley, et al., Astrophys. J. Letters 848, L13 (2017), arXiv:1710.05834 [astro-ph.HE].
  16. [16]I. Arcavi, G. Hosseinzadeh, D. A. Howell, C. McCully, D. Poznanski, et al., Nature (London) 551, 64 (2017), arXiv:1710.05843 [astro-ph.HE].
  17. [17]P. Mösta, D. Radice, R. Haas, E. Schnetter, and S. Bernuzzi, Astrophys. J. Lett. 901, L37 (2020), arXiv:2003.06043 [astro-ph.HE].
  18. [18]E. R. Most and E. Quataert, Astrophys. J. Lett. 947, L15 (2023), arXiv:2303.08062 [astro-ph.HE].
  19. [19]L. Combi and D. M. Siegel, Phys. Rev. Lett. 131, 231402 (2023), arXiv:2303.12284 [astro-ph.HE].
  20. [20]K. Hayashi, K. Kiuchi, K. Kyutoku, Y. Sekiguchi, and M. Shibata, Phys. Rev. Lett. 134, 211407 (2025), arXiv:2410.10958 [astro-ph.HE].
  21. [21]K. Kiuchi, S. Fujibayashi, K. Hayashi, K. Kawaguchi, Q. Pognan, et al., (2026), arXiv:2608.27555 [astro-ph.HE].arxiv.org/abs/2608.27555
  22. [22]K. Kiuchi, A. Reboul-Salze, Y. Sekiguchi, and M. Shibata, arXiv e-prints 10.48550/arXiv.2606.11299 (2026), arXiv:2606.11299.arxiv.org/abs/2606.11299
  23. [23]B. Margalit, A. S. Jermyn, B. D. Metzger, L. F. Roberts, and E. Quataert, Astrophys. J. 939, 51 (2022), arXiv:2206.10645 [astro-ph.HE].
  24. [24]E. R. Most, Phys. Rev. D 108, 123012 (2023), arXiv:2311.03333 [astro-ph.HE].
  25. [25]A. A. Schekchihin, S. C. Cowley, G. W. Hammett, J. L. Maron, and J. C. McWilliams, New Journal of Physics 4, 84 (2002), arXiv:astro-ph/0207503 [astro-ph].
  26. [26]A. K. Galishnikova, M. W. Kunz, and A. A. Schekochihin, Physical Review X 12, 041027 (2022), arXiv:2201.07757 [astro-ph.HE].
  27. [27]J. R. Beattie, C. Federrath, N. Kriel, P. Moczo, and A. Seta, Mon. Not. R. Astron. Soc. 524, 3201 (2023), arXiv:2209.10749 [astro-ph.GA].
  28. [28]N. Kriel, M. R. Krumholz, P. J. Armstrong, J. R. Beattie, and J. Schober, Physical Review E 113, 045208 (2026), arXiv:2509.09949.
  29. [29]R. Aguilera-Miret, D. Viganò, and C. Palenzuela, Astrophys. J. Lett. 926, L31 (2022), arXiv:2112.08406 [gr-qc].
  30. [30]K. Kiuchi, in New Frontiers in GRMHD Simulations, Springer Series in Astrophysics and Cosmology, edited by C. Bambi, Y. Mizuno, S. Shashank, and F. Yuan (Springer, Singapore, 2025) pp. 529–572, arXiv:2405.10081 [astro-ph.HE].
  31. [31]A. A. Schekochihin, S. C. Cowley, S. F. Taylor, J. L. Maron, and J. C. McWilliams, The Astrophysical Journal 612, 276 (2004), arXiv:astro-ph/0312046.
  32. [32]F. Rincon, Journal of Plasma Physics 85, 205850401 (2019), arXiv:1903.07829.
  33. [33]A. P. Kazantsev, Sov. Phys. JETP 26, 1031 (1968).
  34. [34]R. M. Kulsrud and S. W. Anderson, Astrophys. J. 396, 606 (1992).
  35. [35]J. Schober, D. Schleicher, C. Federrath, R. Klessen, and R. Banerjee, Phys. Rev. E 85, 026303 (2012), arXiv:1109.4571 [astro-ph.CO].
  36. [36]N. Kriel, J. R. Beattie, A. Seta, and C. Federrath, Monthly Notices of the Royal Astronomical Society 513, 2457 (2022), arXiv:2204.00828 [astro-ph.SR].
  37. [37]N. Kriel, J. R. Beattie, C. Federrath, M. R. Krumholz, and J. K. J. Hew, Mon. Not. R. Astron. Soc. 537, 2602 (2025), arXiv:2310.17036 [astro-ph.GA].
  38. [38]E. R. Most and S. J. Dunham, (2026), arXiv:2609.21923 [astro-ph.HE].arxiv.org/abs/2609.21923
  39. [39]K. Dionysopoulou, D. Alic, and L. Rezzolla, Phys. Rev. D 92, 084064 (2015), arXiv:1502.02021 [gr-qc].
  40. [40]M. Miravet-Tenés, I. Hawke, R. Matur, and A. J. Wright, (2026), arXiv:2609.12998 [astro-ph.HE].
  41. [41]T. Jousä and K. Kusano, Journal of Computational Physics 208, 315 (2005).
  42. [42]M. N. LeMaster and J. M. Stone, Astrophys. J. 691, 1092 (2009), arXiv:0809.4005 [astro-ph].
  43. [43]L. Fu, X. Y. Hu, and N. A. Adams, Journal of Computational Physics 305, 333 (2016).
  44. [44]J. M. Stone, P. D. Mullen, D. Fielding, P. Grete, M. Guo, et al., Astrophys. J. Suppl. Ser. 283, 27 (2026), arXiv:2409.16053 [astro-ph.IM].
  45. [45]D. Lecoanet, M. McCourt, E. Quataert, K. J. Burns, G. M. Vasil, et al., Monthly Notices of the Royal Astronomical Society 455, 4274 (2016), arXiv:1509.03630 [astro-ph.IM].
  46. [46]See Supplemental Material at [URL will be inserted by publisher] for the run inventory, initial stretching model, energy and spectral diagnostics, and secondary-instability and vertical-field onset analysis.
  47. [47]C. Palenzuela, R. Aguilera-Miret, F. Carrasco, R. Ciolfi, J. V. Kalinani, et al., Physical Review D 106, 023013 (2022), arXiv:2112.08413 [gr-qc].
  48. [48]Y. B. Zel’dovich, Soviet Physics JETP 4, 460 (1957).
  49. [49]T. G. Cowling, Monthly Notices of the Royal Astronomical Society 94, 39 (1933).
  50. [50]Y. Matsumoto and M. Hoshino, Geophysical Research Letters 31, L02807 (2004).
  51. [51]H. N. Dixit and R. Govindarajan, Journal of Fluid Mechanics 646, 415 (2010).
  52. [52]V. Skoutnev, E. R. Most, A. Bhattacharjee, and A. A. Philippov, The Astrophysical Journal 921, 75 (2021).
  53. [53]Movies showing different slice orientations and three-dimensional fly-throughs are available here.
  54. [54]Initially A = (ρ_h − ρ_l)/(ρ_h + ρ_l) = 5/11, where ρ_l and ρ_h are the densities of the light and dense fluids.
  55. [55]C. Musolino, L. Rezzolla, and E. R. Most, Astrophys. J. Lett. 984, L61 (2025), arXiv:2410.06253 [astro-ph.HE].
  56. [56]J. Schober, D. Schleicher, C. Federrath, R. Klessen, and R. Banerjee, Physical Review E 85, 026303 (2012), arXiv:1109.4571.
  57. [57]P. Grete, B. W. O’Shea, K. Beckwith, W. Schmidt, and A. Christlieb, Physics of Plasmas 24, 092311 (2017), arXiv:1706.06339 [physics.plasm-ph].DOI
  58. [58]J. R. Beattie, A. N. Kolborg, E. Ramirez-Ruiz, and C. Federrath, Astrophys. J. 994, 193 (2025), arXiv:2501.09855 [astro-ph.GA].arxiv.org/abs/2501.09855
  59. [59]S. Merlini, J. R. Beattie, and V. Valenzuela-Villaseca, arXiv e-prints , arXiv:2605.21672 (2026), arXiv:2605.21672 [physics.plasm-ph].arxiv.org/abs/2605.21672
  60. [60]A. Beresnyak, Phys. Rev. Lett. 108, 035002 (2012), arXiv:1109.4644 [astro-ph.GA].DOI
  61. [61]J. Guillet, A. Bauswein, O. Just, and H.-T. Janka, Mon. Not. Roy. Astron. Soc. 471, 1879 (2017), arXiv:1610.08532 [astro-ph.HE].
  62. [62]L. Gavassino, Physical Review D 111, 10.1103/phys-revd.111.081306 (2025).
  63. [63]A. Reboul-Salze, P. Barrère, K. Kiuchi, J. Guillet, R. Raynaud, et al., Astron. Astrophys. 699, A4 (2025), arXiv:2411.19328 [astro-ph.HE].
  64. [64]H. H.-Y. Ng and E. R. Most, arXiv e-prints 10.48550/arXiv.2607.28556 (2026), arXiv:2607.28556 [astro-ph.HE].arxiv.org/abs/2607.28556
  65. [65]B. Tripathi, A. E. Fraser, P. W. Terry, E. G. Zweibel, M. J. Pueschel, et al., Nature 649, 848 (2026), arXiv:2608.12530.DOI
  66. [66]H.-S. Chan, O. Gottlieb, J. Jacquemin-Ide, M. Cantiello, and M. Renzo, Astrophys. J. Letters 1007, L41 (2026), arXiv:2605.12931 [astro-ph.HE].arxiv.org/abs/2605.12931
  67. [68]J. Guillet, J. Sato, and T. Foglizzo, The Astrophysical Journal 713, 1350 (2010), arXiv:0910.3953.
  68. [69]E. Roediger, R. P. Kraft, W. R. Forman, P. E. J. Nulsen, and E. Churazov, The Astrophysical Journal 764, 60 (2013), arXiv:1211.4874.DOI
  69. [70]K. Ferrière, Plasma Physics and Controlled Fusion 62, 014014 (2020), arXiv:1912.00837 [astro-ph.GA].
  70. [73]V. Skoutnev, E. R. Most, A. Bhattacharjee, and A. A. Philippov, The Astrophysical Journal 921, 75 (2021), arXiv:2106.04787 [physics.flu-dyn].DOI
  71. [74]J. R. Beattie, C. Federrath, N. Kriel, J. K. J. Hew, and A. Bhattacharjee, Mon. Not. R. Astron. Soc. 542, 2669 (2025), arXiv:2312.03984 [astro-ph.GA].DOI
  72. [75]T. Berlok and C. Pfrommer, Monthly Notices of the Royal Astronomical Society 485, 908 (2019), arXiv:1902.01403 [astro-ph.GA].DOI
  73. [76]J. R. Beattie and A. Bhattacharjee, Phys. Rev. Lett. 137, 165202 (2026).
  74. [77]D. B. Fielding, B. Ripperda, and A. A. Philippov, Astrophys. J. Letters 949, L5 (2023), arXiv:2211.06434 [astro-ph.GA].DOI
  75. [78]J. R. Beattie, C. Federrath, R. S. Klessen, S. Cielo, and A. Bhattacharjee, Nature Astronomy 9, 1195 (2025), arXiv:2504.07136 [astro-ph.GA].DOI
  76. [79]J. M. Stone and T. Gardiner, Astrophys. J. 671, 1726 (2007), arXiv:0709.0452 [astro-ph].DOI
  77. [80]B. Sullivan and A. Kaszynski, Journal of Open Source Software 4, 1450 (2019).

Paper details

vlnc:W_bfy5z8gt
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents