betaFeedback

WorksPhysics and Astronomy

On the injection scale of the turbulence in the partially ionized very local interstellar medium

Abstract

The cascade of magnetohydrodynamic (MHD) turbulence is subject to ion-neutral collisional damping and neutral viscous damping in the partially ionized interstellar medium. By examining the damping effects in the warm and partially ionized local interstellar medium, we find that the interstellar turbulence is damped by neutral viscosity at ∼261\sim 261 au and cannot account for the turbulent magnetic fluctuations detected by Voyager 1 and 2. The MHD turbulence measured by Voyager in the very local interstellar medium (VLISM) should be locally injected in the regime where ions are decoupled from neutrals for its cascade to survive the damping effects. With the imposed ion-neutral decoupling condition, and the strong turbulence condition for the observed Kolmogorov magnetic energy spectrum, we find that the turbulence in the VLISM is sub-Alfvénic, and its largest possible injection scale is ∼194\sim 194 au.

Introduction

Turbulent magnetic fluctuations following a Kolmogorov spectrum were observed by Voyager 1 and 2 in the outer heliosheath ([5, 67, 39, 17, 6]). While the source(s) of the turbulence in the very local interstellar medium (VLISM) are under debate ([63, 65] (Holzer 1989)), it is believed to play a crucial role in affecting the transport of energetic particles and cosmic rays ([34, 53, 29, 41, 18]), and the structure of the Interstellar Boundary Explorer (IBEX) ribbon ([22, 68]). In particular, modeling by Zirnstein et al. [68] suggests that the turbulent magnetic fields with scales <100< 100 au are important for producing the ribbon structure similar to IBEX observations, and such turbulence is likely not of pristine interstellar origin, whereas turbulent fluctuations at scales ≥100\ge100 au produce features inconsistent with IBEX observations.

Turbulence and turbulent magnetic fields are ubiquitous in the ISM (e.g., [1, 7, 20, 38] (Xu & Zhang 2017)), with the turbulent energy injected by supernova explosions on large length scales (∼100\sim100 pc; [3]) and cascading down toward smaller and smaller scales. In the warm and partially ionized local interstellar medium (LISM) ([19, 50]), the magnetohydrodynamic (MHD) turbulence is subject to the damping effects due to the frictional collisions between ions and neutrals (i.e., ion-neutral collisional damping) and among neutrals (i.e. neutral viscosity) (Xu & Lazarian (2017)). The damping effects cause the cutoff of MHD turbulence cascade when the damping rate exceeds the cascading rate of MHD turbulence.

The linear analysis of MHD waves in a partially ionized medium was performed by Kulsrud & Pearce [31] and more recently by e.g., Pudritz [48]; ); Balsara [2]; ); Mouschovias et al. [44]; ); Zaqarashvili et al. [66]; ); Soler et al. [51]. Different from linear MHD waves, MHD turbulence is characterized by the nonlinear cascade of turbulent energy, with scale-dependent turbulence anisotropy and limited timescale of turbulent motions ([23, 35]). The ion-neutral collisional (IN) damping of MHD turbulence in a highly ionized medium at a high plasma β\beta was studied by Lithwick & Goldreich [40]. The neutral viscous (NV) damping of MHD turbulence was analyzed by Lazarian et al. [36], which tends to dominate over the IN damping toward a higher ionization fraction and at a higher temperature ((Xu & Lazarian 2017)). Xu et al. (2015, 2016) performed a general analysis including both damping effects of MHD turbulence in different interstellar phases with varying ionization fractions.

The range of length scales for the existence of MHD turbulence depends on the coupling state between ions and neutrals and the corresponding damping effects. Irrespective of the origin, the MHD turbulence measured in the VLISM should survive the damping in the partially ionized medium. In this work, we will examine the damping of the interstellar MHD turbulence driven at large scales. More importantly, we will explore the constraint imposed by the damping effects on the locally driven MHD (LMHD) turbulence in the VLISM, as observed by Voyager. In Section 2, we analyze the damping of the interstellar turbulence driven in the strongly coupled regime and the injection of the LMHD turbulence in the decoupled regime with weak damping. In Section 3, we determine the turbulence regime and the largest injection scale of the LMHD turbulence constrained by the ion-neutral decoupling condition. Discussion and conclusions are presented in Sections 4 and 5.

Source for turbulence in the VLISM

There are three possible sources for the turbulent magnetic fluctuations measured by Voyager, including the interstellar turbulence driven by supernova explosions at ∼100\sim100 pc, the turbulence driven in the Local Interstellar Cloud at ∼2\sim2 pc [65], and the LMHD turbulence in the VLISM. We will first focus on the former two possible sources, for which the turbulence is driven on large scales in the regime where neutrals and ions are strongly coupled together (§ 2.1). For the local source, the turbulence with a small energy injection scale is expected to arise in the regime where ions are decoupled from neutrals (§ 2.2).

Damping of interstellar turbulence driven in strongly coupled regime

The interstellar turbulence is driven by supernova explosions at L∼100L \sim100 pc [3] and cascades down to smaller and smaller scales [8, 62]. Its damping scale varies in different interstellar phases and regions, depending on the local physical conditions. Given the approximate energy equipartition between turbulence and magnetic fields in the ISM expected from small-scale nonlinear turbulent dynamo (Xu & Lazarian 2016) and indicated by observational measurements [46], we consider that the interstellar turbulence is trans-Alfvénic, i.e., MA=VL/VA≈1M_A = V_L/V_A \approx1. Here MAM_A is the turbulent Alfvén Mach number, VLV_L is the turbulent velocity at the injection scale LL of turbulence, VA=B/4πρV_A = B/\sqrt{4\pi\rho} is the Alfvén speed, BB is the magnetic field strength, ρ=ρi+ρn\rho= \rho_i + \rho_n is the total mass density, ρi=mHni=mHne\rho_i = m_H n_i = m_H n_e is the mass density of ions, mHm_H is the hydrogen atomic mass, nin_i and nen_e are the number densities of ions and electrons, ρn=mHnH\rho_n = m_H n_H is the mass density of neutrals, and nHn_H is the number density of neutrals.

In a partially ionized medium, MHD turbulence is subject to both IN damping and NV damping [36] (Xu & Lazarian 2017). They both depend on the coupling state between ions and neutrals. When the driving rate of turbulence is lower than the neutral-ion collisional frequency νni=γdρi\nu_{ni} = \gamma_d \rho_i, where γd=5.5×1014 cm3 g−1 s−1\gamma_d = 5.5 \times10^{14}\ \mathrm{cm^3\,g^{-1}\,s^{-1}} is the drag coefficient [15, 49], the MHD turbulence is driven in strongly coupled ions and neutrals, i.e., the strongly coupled regime. Neutrals decouple from ions when the cascading rate of MHD turbulence τcas−1\tau_{\mathrm{cas}}^{-1} becomes larger than νni\nu_{ni}. The neutral-ion decoupling entails significant collisional friction, causing the damping of MHD turbulence and the cutoff of its energy cascade. The corresponding damping scale for Alfvénic turbulence is (Xu et al. 2015, 2016; see Appendix B for the derivation)

ldam,IN,⊥=(2νniξn)−3/2L−1/2VL3/2,(1)l_{\mathrm{dam,IN},\perp} = \left(\frac{2\nu_{ni}}{\xi_n}\right)^{-3/2} L^{-1/2} V_L^{3/2}, \tag*{(1)}

where ξn=ρn/ρ\xi_n = \rho_n/\rho. The subscript ⊥\perp means that the length scale is measured perpendicular to the local magnetic field. As the energy cascade of MHD turbulence is mostly in the perpendicular direction, ldam,IN,⊥l_{\mathrm{dam,IN},\perp} corresponds to the cutoff scale of MHD turbulence.

In the case when the NV damping dominates over the IN damping, the damping of MHD turbulence occurs in the strongly coupled regime, with the damping scale (Xu & Lazarian 2017; see Appendix B)

ldam,NV,⊥=(ξn2)3/4νn3/4L1/4VL−3/4,(2)l_{\mathrm{dam,NV},\perp} = \left(\frac{\xi_n}{2}\right)^{3/4} \nu_n^{3/4} L^{1/4} V_L^{-3/4}, \tag*{(2)}

where νn=vth/(nHσnn)\nu_n = v_{\mathrm{th}}/(n_H\sigma_{nn}) is the kinematic viscosity in neutrals, vthv_{\mathrm{th}} is the neutral thermal speed, and σnn\sigma_{nn} is the collisional cross-section of neutrals. We note that ldam,NV,⊥l_{\mathrm{dam,NV},\perp} is in fact the damping scale of the turbulent kinetic energy spectrum. The magnetic fluctuations in the sub-viscous range below ldam,NV,⊥l_{\mathrm{dam,NV},\perp} is termed new regime of MHD turbulence [36] (see § 4.1).

We now discuss whether the magnetic turbulence in the VLISM can come from the interstellar turbulence that is injected by supernova explosions. The LISM near the sun is warm, low-density, and partially ionized, with the temperature T≈6300T \approx6300 K, nH≈0.2 cm−3n_H \approx0.2\ \mathrm{cm^{-3}}, and ne≈0.07 cm−3n_e \approx0.07\ \mathrm{cm^{-3}} [50, 54]. In addition, we adopt LISM≈100L_{\mathrm{ISM}} \approx100 pc, VL,ISM≈VAV_{L,\mathrm{ISM}} \approx V_A, B≈5 μGB \approx5\ \mu\mathrm{G} as the typical driving conditions of interstellar turbulence and interstellar magnetic field strength [13]. With these high temperature and moderate ionization fraction, we find that

ldam,IN,⊥≈7.6×1013 cm(ne0.07 cm−3)−3/2(nH0.2 cm−3)−3/4l_{\mathrm{dam,IN},\perp} \approx7.6 \times10^{13}\ \mathrm{cm}\left(\frac{n_e}{0.07\ \mathrm{cm}^{-3}}\right)^{-3/2}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-3/4}
(LISM100 pc)−1/2(B5 μG)3/2,(3)\left(\frac{L_{\mathrm{ISM}}}{100\ \mathrm{pc}}\right)^{-1/2}\left(\frac{B}{5\ \mu\mathrm{G}}\right)^{3/2}, \tag*{(3)}

is much smaller than

ldam,NV,⊥≈3.9×1015 cm(T6300 K)3/8(nH0.2 cm−3)−3/8l_{\mathrm{dam,NV},\perp} \approx3.9 \times10^{15}\ \mathrm{cm}\left(\frac{T}{6300\ \mathrm{K}}\right)^{3/8}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-3/8}
(LISM100 pc)1/4(B5 μG)−3/4,(4)\left(\frac{L_{\mathrm{ISM}}}{100\ \mathrm{pc}}\right)^{1/4}\left(\frac{B}{5\ \mu\mathrm{G}}\right)^{-3/4}, \tag*{(4)}

where we assume nH+ne∼nHn_H + n_e \sim n_H, and σnn≈10−14 cm2\sigma_{nn} \approx10^{-14}\ \mathrm{cm^2} [30, 55] is adopted. We note that the gradients of quantities in the outer heliosheath [64] cause uncertainties in the above estimates. By assuming the uncertainties σ(ne)∼0.01 cm−3\sigma(n_e) \sim0.01\ \mathrm{cm^{-3}}, σ(nH)∼0.01 cm−3\sigma(n_H) \sim0.01\ \mathrm{cm^{-3}}, σ(B)∼1 μG\sigma(B) \sim1\ \mu\mathrm{G}, σ(T)∼1000\sigma(T) \sim1000 K, and σ(LISM)∼10\sigma(L_{\mathrm{ISM}}) \sim10 pc, we find σ(ldam,IN,⊥)∼2.8×1013\sigma(l_{\mathrm{dam,IN},\perp}) \sim2.8 \times10^{13} cm and σ(ldam,NV,⊥)∼6.4×1014\sigma(l_{\mathrm{dam,NV},\perp}) \sim6.4 \times10^{14} cm.

For the interstellar turbulence driven in the Local Interstellar Cloud, the turbulent velocity of a few km s−1^{-1} is similar to that from the cascade of supernova-driven turbulence at ∼2\sim2 pc [52]. It might be a part of the global cascade of the interstellar turbulence. Therefore, it is also subject to the damping effects in the partially ionized LISM and damped at the damping scale similar to that of the supernova-driven interstellar turbulence.

The above calculations show that the damping of the MHD turbulence in the LISM from the interstellar origin is dominated by NV in the strongly coupled regime. The corresponding damping scale is ∼3.9×1015\sim3.9 \times10^{15} cm, i.e., 261 au. This means that the turbulence observed by Voyager 1 and 2 is unlikely due to the pristine interstellar origin.

LMHD turbulence in decoupled regime

As discussed above, for the MHD turbulence injected in the strongly coupled regime, its cascade is cut off either by the IN or NV damping. When ions are decoupled from neutrals, however, the driven MHD turbulence is no longer subject to the NV damping, and the IN damping becomes constantly weak (Xu et al. 2016). The MHD turbulence injected in ions that are decoupled from neutrals, i.e., LMHD turbulence in the decoupled regime, is not cut off due to the damping effects arising in a partially ionized medium.

Based on the in-situ measurements of turbulent magnetic energy spectrum by Voyager 1 in the VLISM [39], we have the ratio of the turbulent component to the average magnetic field strength δBobs/B≈0.06\delta B_{\mathrm{obs}}/B \approx0.06 measured at lobs≈3×1014 cml_{\mathrm{obs}} \approx3 \times10^{14}\ \mathrm{cm} (corresponding to the smallest wavenumber of the measured spectrum and assumed to be in the inertial range of turbulence). By assuming that the magnetic fluctuations are mainly induced by Alfvénic turbulence [9, 39] (Hu et al. 2022), the local turbulent velocity can be estimated as

vobs=δBobsBVAi=2.5 km s−1(δBobs/B0.06)(ne0.07 cm−3)−1/2(B5 μG).(5)\begin{aligned} v_{\mathrm{obs}} = \frac{\delta B_{\mathrm{obs}}}{B}V_{Ai} \\ &= 2.5\ \mathrm{km\ s^{-1}}\left(\frac{\delta B_{\mathrm{obs}}/B}{0.06}\right)\left(\frac{n_e}{0.07\ \mathrm{cm}^{-3}}\right)^{-1/2}\left(\frac{B}{5\ \mu\mathrm{G}}\right). \tag*{(5)} \end{aligned}

By assuming σ(δBobs)∼0.1 μG\sigma(\delta B_{\mathrm{obs}}) \sim0.1\ \mu\mathrm{G} and the uncertainties of other parameters (see § 2.1), we find σ(vobs)∼0.85 km s−1\sigma(v_{\mathrm{obs}}) \sim0.85\ \mathrm{km\ s^{-1}}. The large uncertainties in our calculations are mainly caused by the large uncertainties in magnetic field strength measurements [5]. Such a large turbulence level at the measured length scale cannot be accounted for by the interstellar turbulence, which is cut off at a larger length scale ldam,NV,⊥l_{\mathrm{dam,NV},\perp} (Eq. (4)). So the measured turbulence is likely to be driven in the VLISM. Given vobs/lobs(≈8.3×10−10 s−1)>νin(≈1.8×10−10 s−1)v_{\mathrm{obs}}/l_{\mathrm{obs}}(\approx8.3 \times10^{-10}\ \mathrm{s}^{-1}) > \nu_{\mathrm{in}}(\approx1.8 \times10^{-10}\ \mathrm{s}^{-1}), the LMHD turbulence measured by Voyager 1 should be injected in the decoupled regime.

Constraint on the injection scale of the LMHD turbulence in the VLISM

We now discuss the constraint on the injection scale LL of the LMHD turbulence. The three cases we consider are (a) super-Alfvénic turbulence with isotropic injection scale, (b) sub-Alfvénic turbulence with isotropic injection scale, and (c) sub-Alfvénic turbulence with anisotropic injection scale. In Figure 1, we illustrate the scalings of super- and sub-Alfvénic turbulence (see also Appendix § 8) for isotropic and anisotropic injection.

Illustration for turbulent eddies in super-Alfvénic and sub-Alfvénic turbulence

Figure 1. Illustration for turbulent eddies in super-Alfvénic (a) and sub-Alfvénic (b,c) turbulence. Strong MHD turbulence is indicated by shaded regions. With anisotropic injection scale of turbulence in (c), the entire turbulent cascade can be in the strong turbulence regime. The blue line indicates the magnetic field.

The condition for the LMHD turbulence to arise in ions alone is

ΓL>νin,(6)\Gamma_L > \nu_{\mathrm{in}}, \tag*{(6)}

where ΓL\Gamma_L is the driving rate of the turbulence, i.e., the cascading rate of turbulence at LL, and νin=γdρn\nu_{\mathrm{in}} = \gamma_d \rho_n is the ion-neutral collisional frequency.

Case (a). If the driven turbulence is super-Alfvénic, i.e., MA=VL/VAi>1M_A = V_L/V_{Ai} > 1, where VAi=B/4πρiV_{Ai} = B/\sqrt{4\pi\rho_i} is the Alfvén speed in ions, VLV_L is related to vlv_l by [35]

vl=VL(lL)1/3,(7)v_l = V_L\left(\frac{l}{L}\right)^{1/3}, \tag*{(7)}

where vlv_l is the local turbulent velocity measured at length scale ll (<L< L). With the cascade of turbulent energy, the turbulent velocity becomes equal to VAiV_{Ai} at the Alfvénic scale lA=LMA−3l_A = L M_A^{-3}. Below lAl_A, the effect of magnetic fields on turbulence becomes important, resulting in turbulence anisotropy (see Figure 1). Therefore, ll in the above equation should be replaced by l⊥l_\perp when ll is smaller than lAl_A, where l⊥l_\perp is the length scale measured perpendicular to the local magnetic field [12]. By using ΓL=VL/L\Gamma_L = V_L/L and the scaling relation in Eq. (7), the condition Eq. (6) in this case becomes

L<(νinvl−1l⊥1/3)−3/2,lA<l<L,L < \left(\nu_{\mathrm{in}}v_l^{-1}l_\perp^{1/3}\right)^{-3/2},\quad l_A < l < L,
L<(νinvl−1l⊥−1/3)−3/2,l<lA.(8)L < \left(\nu_{\mathrm{in}}v_l^{-1}l_\perp^{-1/3}\right)^{-3/2},\quad l < l_A. \tag*{(8)}

The above expressions provide constraints on the maximum LL of the LMHD turbulence in the decoupled regime if it is super-Alfvénic.

The in-situ measurements show that the LMHD turbulence has vl≈vobsv_l \approx v_{\mathrm{obs}} at l⊥≈lobsl_\perp\approx l_{\mathrm{obs}} (§ 2.2). We consider the largest observed scale lobsl_{\mathrm{obs}} as the perpendicular scale because the trajectory of Voyager 1 is nearly perpendicular to the background magnetic field [28, 17, 14]. vlv_l is smaller than VAiV_{Ai} (see Eq. (5)) and related to VAiV_{Ai} by

vl=VAi(l⊥lA)1/3,(9)v_l = V_{Ai}\left(\frac{l_\perp}{l_A}\right)^{1/3}, \tag*{(9)}

yielding

lA=l⊥(vlVAi)−3=1.4×1018 cm(ne0.07 cm−3)−3/2(vl2.5 km s−1)−3(l⊥3.0×1014 cm)3(B5 μG)3.(10)\begin{aligned} l_A = l_\perp\left(\frac{v_l}{V_{Ai}}\right)^{-3} \\ &= 1.4 \times10^{18}\ \mathrm{cm}\left(\frac{n_e}{0.07\ \mathrm{cm}^{-3}}\right)^{-3/2}\left(\frac{v_l}{2.5\ \mathrm{km\ s}^{-1}}\right)^{-3}\left(\frac{l_\perp}{3.0 \times10^{14}\ \mathrm{cm}}\right)^3\left(\frac{B}{5\ \mu\mathrm{G}}\right)^3. \tag*{(10)} \end{aligned}

with the estimated uncertainty σ(lA)∼1.7×1018 cm\sigma(l_A) \sim1.7 \times10^{18}\ \mathrm{cm}. The condition in Eq. (8) for super-Alfvénic turbulence can be written explicitly as,

L<2.9×1015 cm(nH0.2 cm−3)−3/2(vl2.5 km s−1)3/2(l⊥3.0×1014 cm)−1/2,(11)L < 2.9 \times10^{15}\ \mathrm{cm}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-3/2}\left(\frac{v_l}{2.5\ \mathrm{km\ s}^{-1}}\right)^{3/2}\left(\frac{l_\perp}{3.0 \times10^{14}\ \mathrm{cm}}\right)^{-1/2}, \tag*{(11)}

with the estimated uncertainty σ(L)∼1.5×1015 cm\sigma(L) \sim1.5 \times10^{15}\ \mathrm{cm}. The super-Alfvénic condition MA>1M_A > 1 requires L>lAL > l_A, which is not satisfied by the above values. We conclude that the LMHD turbulence cannot be super-Alfvénic.

Case (b). If the driven turbulence is sub-Alfvénic, i.e., MA=VL/VAi<1M_A = V_L/V_{Ai} < 1, the MHD turbulence is weak with weak interactions between counterpropagating Alfvén wave packets [21] over scales [L,ltran][L, l_{\mathrm{tran}}], where ltran=LMA2l_{\mathrm{tran}} = LM_A^2 is the perpendicular transition scale from weak to strong MHD turbulence [35] (see Figure 1). For weak turbulence, there is no parallel cascade. Its cascade to smaller perpendicular scales strengthens until the cascade becomes strong (Goldreich & Sridhar 1997). The scaling of weak turbulence follows vl=VL(l⊥/L)1/2v_l = V_L(l_\perp/L)^{1/2} [35], while in the strong turbulence regime, the local turbulent velocity follows the scaling [35]

vl=VL(l⊥L)1/3MA1/3.(12)v_l = V_L\left(\frac{l_\perp}{L}\right)^{1/3}M_A^{1/3}. \tag*{(12)}

For the weak turbulence at LL, there is

ΓL=VLLMA.(13)\Gamma_L = \frac{V_L}{L}M_A. \tag*{(13)}

By combining Eqs. (12) and (13), the condition in Eq. (6) becomes

L<νin−2vl3l⊥−1VAi−1.(14)L < \nu_{\mathrm{in}}^{-2}v_l^3l_\perp^{-1}V_{Ai}^{-1}. \tag*{(14)}

The spectral indices of magnetic fluctuations in weak and strong MHD turbulence are −2-2 and −5/3-5/3, respectively. With the Kolmogorov slope (−5/3-5/3) found for the measured magnetic energy spectrum [5, 39], we consider that the locally measured turbulence is in the strong MHD turbulence regime. For the driven sub-Alfvénic turbulence, the condition in Eq. (14) gives

L<3.6×1014 cm(ne0.07 cm−3)1/2(nH0.2 cm−3)−2×(vl2.5 km s−1)3(l⊥3.0×1014 cm)−1(B5 μG)−1,(15)\begin{aligned} L &< 3.6 \times10^{14}\ \mathrm{cm}\left(\frac{n_e}{0.07\ \mathrm{cm}^{-3}}\right)^{1/2}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-2}\\ &\quad\times\left(\frac{v_l}{2.5\ \mathrm{km\ s}^{-1}}\right)^3\left(\frac{l_\perp}{3.0 \times10^{14}\ \mathrm{cm}}\right)^{-1}\left(\frac{B}{5\ \mu\mathrm{G}}\right)^{-1}, \tag*{(15)} \end{aligned}

with σ(L)∼3.8×1014 cm\sigma(L) \sim3.8 \times10^{14}\ \mathrm{cm}. By using Eqs. (12) and (13), the condition in Eq. (6) can also be written as

VL<(νin−1vl3l⊥−1)1/2<5.2 km s−1(nH0.2 cm−3)−1/2(vl2.5 km s−1)3/2×(l⊥3.0×1014 cm)−1/2,(16)\begin{aligned} V_L &< \left(\nu_{\mathrm{in}}^{-1}v_l^3l_\perp^{-1}\right)^{1/2}\\ &< 5.2\ \mathrm{km\ s}^{-1}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-1/2}\left(\frac{v_l}{2.5\ \mathrm{km\ s}^{-1}}\right)^{3/2}\\ &\quad\times\left(\frac{l_\perp}{3.0 \times10^{14}\ \mathrm{cm}}\right)^{-1/2}, \tag*{(16)} \end{aligned}

with σ(VL)∼2.7 km s−1\sigma(V_L) \sim2.7\ \mathrm{km\ s}^{-1}. The corresponding MAM_A is

MA=VLVAi<0.13(ne0.07 cm−3)1/2(nH0.2 cm−3)−1/2×(vl2.5 km s−1)3/2(l⊥3.0×1014 cm)−1/2(B5 μG)−1,(17)\begin{aligned} M_A &= \frac{V_L}{V_{Ai}} < 0.13\left(\frac{n_e}{0.07\ \mathrm{cm}^{-3}}\right)^{1/2}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-1/2}\\ &\quad\times\left(\frac{v_l}{2.5\ \mathrm{km\ s}^{-1}}\right)^{3/2}\left(\frac{l_\perp}{3.0 \times10^{14}\ \mathrm{cm}}\right)^{-1/2}\left(\frac{B}{5\ \mu\mathrm{G}}\right)^{-1}, \tag*{(17)} \end{aligned}

with σ(MA)∼0.072\sigma(M_A) \sim0.072. Given such a small MAM_A value, the implied ltran=LMA2l_{\mathrm{tran}} = LM_A^2 is ≲6.1×1012 cm\lesssim6.1 \times10^{12}\ \mathrm{cm}. This means that the observed turbulence would be in the weak MHD turbulence regime with lobs>ltranl_{\mathrm{obs}} > l_{\mathrm{tran}}. This is inconsistent with the observed Kolmogorov spectrum for strong MHD turbulence.

Case (c). In the above calculations we assume that the injection scale of turbulence is isotropic. In the presence of strong background magnetic fields, the driven sub-Alfvénic turbulence is likely to have anisotropic LL (see e.g., Pogorelov et al. [47]). With a sufficiently small perpendicular component of injection scale L⊥L_\perp, the shear in the direction perpendicular to the magnetic field can cause significant distortions of magnetic field lines within the Alfvén wave period. If the anisotropy is sufficiently large so that the nonlinear interaction between counterpropagating Alfvén wave packets is strong and thus the critical balance relation [23] is satisfied at LL, the entire turbulent cascade would be in the strong MHD turbulence regime (see Figure 1). In this case, we have

vl=VL(l⊥L⊥)1/3.(18)v_l = V_L\left(\frac{l_\perp}{L_\perp}\right)^{1/3}. \tag*{(18)}

and ΓL=VL/L⊥\Gamma_L = V_L/L_\perp. The condition in Eq. (6) leads to the constraint on the perpendicular injection scale,

L⊥<(νinvll⊥1/3)−3/2L_\perp< \left(\nu_{in}v_l l_\perp^{1/3}\right)^{-3/2}
≈2.9×1015 cm(nH0.2 cm−3)−3/2(vl2.5 km s−1)3/2(l⊥3.0×1014 cm)−1/2,(19)\approx2.9 \times10^{15}\ \mathrm{cm}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-3/2}\left(\frac{v_l}{2.5\ \mathrm{km\ s}^{-1}}\right)^{3/2}\left(\frac{l_\perp}{3.0 \times10^{14}\ \mathrm{cm}}\right)^{-1/2}, \tag*{(19)}

with the uncertainty σ(L⊥)∼1.5×1015 cm\sigma(L_\perp) \sim1.5 \times10^{15}\ \mathrm{cm}, and the same constraint on VLV_L as in Eq. (16).

Among the three cases, only Case (c), i.e., sub-Alfvénic turbulence with anisotropic injection scale, provides the self-consistent result. In Figure 2, we present 2π/L⊥,max⁡2\pi/L_{\perp,\max} together with the observationally measured magnetic energy spectrum taken from Lee & Lee [39] for the period from 2012 August to 2019 December and Burlaga et al. [5] for intervals 2013.3593–2014.6373 and 2015.3987–2016.6759. Here L⊥,max⁡L_{\perp,\max} (Eq. (19)) is the largest possible perpendicular injection scale of the LMHD turbulence in the VLISM. We find that L⊥,max⁡L_{\perp,\max} is close to ldam,NV,⊥l_{\mathrm{dam,NV},\perp} of the interstellar turbulence given by Eq. (4).

Figure 2. The data for the magnetic energy spectrum are taken from Lee & Lee [39] (black) and Burlaga et al. [5] (red and blue). The dashed line indicates the Kolmogorov slope. The vertical dashed and dash-dotted lines correspond to the damping scales of interstellar turbulence (ldam,NV,⊥l_{\mathrm{dam,NV},\perp}, Eq. (4)) and interstellar magnetic fluctuations (ldam,IN,⊥,sub−vl_{\mathrm{dam,IN},\perp,\mathrm{sub-v}}, Eq. (22)). The vertical solid line corresponds to the largest possible perpendicular injection scale (L⊥,max⁡L_{\perp,\max}, Eq. (19)) of the locally driven sub-Alfvénic turbulence in the VLISM.

Discussion

New regime of MHD turbulence

In the case when the NV damping dominates over the IN damping (see § 2.1), the turbulent motions are damped at the neutral viscous scale, but the magnetic fluctuations can exist on smaller scales. There is a new regime of MHD turbulence in the sub-viscous range (Cho et al. [11]; Lazarian et al. [36]; Xu & Lazarian 2016), where the kinetic energy spectrum is steep with the spectral index −4-4 and the magnetic energy spectrum is flat with the spectral index −1-1. We note that such a flat magnetic energy spectrum corresponds to scale-independent magnetic fluctuations. The sub-viscous magnetic fluctuations are caused by the shear of viscous-scale turbulent eddies in the direction perpendicular to the magnetic field. As the IN damping suppresses magnetic fluctuations, the damping scale of the sub-viscous magnetic fluctuations is determined by the balance between the eddy-turnover rate at ldam,NV,⊥l_{\mathrm{dam,NV},\perp} and the IN damping rate,

vνldam,NV,⊥=ωd,IN,(20)\frac{v_\nu}{l_{\mathrm{dam,NV},\perp}}=\omega_{\mathrm{d,IN}}, \tag*{(20)}

where vν=VL,ISM(ldam,NV,⊥/LISM)1/3v_\nu=V_{L,\mathrm{ISM}}(l_{\mathrm{dam,NV},\perp}/L_{\mathrm{ISM}})^{1/3} is the turbulent velocity at ldam,NV,⊥l_{\mathrm{dam,NV},\perp}. The IN damping rate in this case is

ωd,IN=ξnk⊥2(δVA)22νni,(21)\omega_{\mathrm{d,IN}}=\frac{\xi_n k_\perp^2(\delta V_A)^2}{2\nu_{ni}}, \tag*{(21)}

where δVA=δB/4πρ\delta V_A=\delta B/\sqrt{4\pi\rho}, and k⊥2(δVA)2k_\perp^2(\delta V_A)^2 is the magnetic force per unit mass per unit displacement corresponding to the sub-viscous magnetic fluctuation δB\delta B perpendicular to the magnetic field. Eq. (21) is different from the expression for IN damping rate of Alfvén waves in Kulsrud & Anderson [32], as over sub-viscous scales, there are no Alfvén wave motions of magnetic fields. By assuming that the Alfvénic component dominates the turbulent motion, we approximately have δVA≈vν\delta V_A \approx v_\nu. Then we obtain from Eq. (20)

ldam,IN,⊥,sub−v=ξn2νν−1/2νni1/2l_{\mathrm{dam,IN},\perp,\mathrm{sub-v}}=\frac{\xi_n}{2}\nu_\nu^{-1/2}\nu_{ni}^{1/2}
≈1.0×1015 cm(T6300 K)1/4(ne0.07 cm−3)−1/2(nH0.2 cm−3)−1/2,(22)\approx1.0 \times10^{15}\ \mathrm{cm}\left(\frac{T}{6300\ \mathrm{K}}\right)^{1/4}\left(\frac{n_e}{0.07\ \mathrm{cm}^{-3}}\right)^{-1/2}\left(\frac{n_H}{0.2\ \mathrm{cm}^{-3}}\right)^{-1/2}, \tag*{(22)}

with σ(ldam,IN,⊥,sub−v)∼8.5×1013 cm\sigma(l_{\mathrm{dam,IN},\perp,\mathrm{sub-v}}) \sim8.5 \times10^{13}\ \mathrm{cm}, as the IN damping scale of sub-viscous magnetic fluctuations (see Figure 2). We see that ldam,IN,⊥,sub−vl_{\mathrm{dam,IN},\perp,\mathrm{sub-v}} is slightly smaller than ldam,NV,⊥l_{\mathrm{dam,NV},\perp} in Eq. (4). So the cutoff scale of magnetic fluctuations is close to the cutoff scale of the kinetic energy spectrum of the interstellar turbulence.

For the LMHD turbulence driven in the VLISM, the new regime of MHD turbulence is also expected below the (effective) ion viscous scale. As pointed out in Fraternale & Pogorelov [17], the small-scale spectral flattening with the spectral index −1-1 (see also Figure 2) may be accounted for by the sub-viscous magnetic fluctuations in the new regime of LMHD turbulence (Cho et al. [11]; Xu & Lazarian 2016).

Compressibility of the LMHD turbulence

In our calculations, we assume that the Alfvénic component carries most of the turbulent energy in compressible MHD turbulence. This is supported by compressible MHD turbulence simulations (Cho & Lazarian [9]; Hu et al. 2022), as well as the higher power of perpendicular magnetic fluctuations compared to that of parallel magnetic fluctuations found in Lee & Lee [39]. We consider strong MHD turbulence with strong nonlinear interactions between oppositely directed Alfvén wave packets and balanced cascade (Goldreich & Sridhar [23]) based on the Kolmogorov magnetic energy spectrum reported in e.g., Burlaga et al. [5]; Lee & Lee [39]. But we also note the existence of large-scale compressive component of magnetic fluctuations with the spectral index close to −2-2 (Fraternale & Pogorelov [17]), which probably reflects the discontinuities in magnetic field distribution associated with shock/compression waves. As the conversion from Alfvén modes to compressive modes of MHD turbulence is inefficient (Cho & Lazarian [9]), the compressive component is likely to originate from the turbulence injection. By combining the in-situ data of magnetic and electron density fluctuations, Lee & Lee [39] found that the observations cannot be explained by the linear magnetohydrodynamic modes alone. In addition, within the limited range of measured frequencies/scales, it is difficult to distinguish between the spectral indices −5/3-5/3 and −2-2 in observations, and thus the possibility of the existence of weak turbulence cannot be completely excluded.

Due to the damping effects in the presence of neutrals, we find that the measured turbulence in the VLISM is likely to be entirely of heliospheric origin. Moreover, the turbulent energy transfer rate of the LMHD turbulence is ϵLMHD∼ρiVL3/L⊥≈5.7×10−24 erg cm−3 s−1\epsilon_{\mathrm{LMHD}} \sim\rho_i V_L^3/L_\perp\approx5.7 \times10^{-24}\ \mathrm{erg\ cm^{-3}\ s^{-1}}, which is much larger than ϵISM∼ρVL,ISM3/LISM≈1.4×10−26 erg cm−3 s−1\epsilon_{\mathrm{ISM}} \sim\rho V_{L,\mathrm{ISM}}^3/L_{\mathrm{ISM}} \approx1.4 \times10^{-26}\ \mathrm{erg\ cm^{-3}\ s^{-1}} of the interstellar turbulence. This is consistent with the finding in Lee & Lee [39]; Ocker et al. [45] that the Kolmogorov electron density spectrum measured by Voyager 1 has a significantly higher intensity than that measured in the interstellar warm ionized medium (Armstrong et al. [1]; Chepurnov & Lazarian [7]). The quasiperiodic structures seen in the magnetic field fluctuations also indicate the heliospheric forcing of the measured turbulence in the VLISM (Fraternale & Pogorelov [17]). Both solar rotation and solar cycle may play an important role in driving the heliospheric turbulence (Zank et al. [65]).

With large observational uncertainties, the transition from the magnetic fluctuations parallel to the mean magnetic field in an earlier interval to those transverse to the mean magnetic field in later intervals was found from Voyager 1 and 2 measurements ([5, 67, 6]). Magnetic fluctuations parallel to the local magnetic field arising from slow and fast modes in compressible MHD turbulence and pseudo-Alfvénic modes in incompressible MHD turbulence are important for the mirror diffusion of particles (Xu & Lazarian 2020; Lazarian & Xu [37]). Zank et al. [65] has shown that the fluctuations measured by Voyager 1 and 2 can be explained by a transmission of fast modes from inner heliosheath with a conversion to incompressible Alfvén waves. They further assumed that the outer scale of this heliosphere-originated turbulence is around 120 au, and the interstellar turbulence will continue down to scales smaller than those measured by Voyager. Our work, by considering the ion-neutral interactions in the LISM region, suggests that the interstellar turbulence should have a cutoff around 261 au, and a new heliosphere-related turbulence (such as the scenario described by Zank et al. [65]) with an injection scale smaller than ∼2.9×1015 cm\sim2.9 \times10^{15}\ \mathrm{cm} (≈194\approx194 au) is needed to explain the Voyager measurements.

Conclusions

The damping effects in the partially ionized LISM determine the range of length scales for the existence of interstellar MHD turbulence and the LMHD turbulence. Due to the high temperature and moderate ionization fraction in the LISM, we find that the dominant damping mechanism of the interstellar MHD turbulence is the NV damping. The NV damping scale of the interstellar turbulence is about 261 au. Below the NV damping scale of turbulent cascade, the new regime of MHD turbulence with constant magnetic fluctuations (Lazarian et al. [36]) is expected to rise. We find that the sub-viscous magnetic fluctuations are cut off due to the IN damping at a scale slightly smaller than the NV damping scale.

For the LMHD turbulence, when the injection occurs in the regime with ions decoupled from neutrals, the turbulent cascade can not be cut off by the damping related to partial ionization. Given the turbulent velocity at the largest observed length scale inferred from in-situ measurements, after applying the ion-neutral decoupling condition, we find that the LMHD turbulence in the VLISM is sub-Alfvénic with the injected turbulent energy smaller than the magnetic energy. With the trajectory of Voyager 1 approximately perpendicular to the background magnetic field, by assuming anisotropic injection scale of the LMHD turbulence, we further find the upper limit of the perpendicular injection scale L⊥,max⁡≈2.9×1015 cm≈194 auL_{\perp,\max} \approx2.9 \times10^{15}\ \mathrm{cm} \approx194\ \mathrm{au}, which is close to the NV damping scale of the interstellar turbulence. Our estimated largest outer scale of LMHD turbulence is comparable to the extent of the heliosphere in the upstream direction ([47]) and one order of magnitude smaller than the estimate given in [5, 39] by extrapolating the power-law slope to the equipartition between the turbulent fluctuation and the average magnetic field strength. Note that other considerations such as mode conversion ([65]) and IBEX modeling ([68]) could further limit the injection scale down to tens of au.

Acknowledgments

We thank the referees for very detailed and helpful comments, which improved the manuscript perceptibly. S.X. acknowledges inspiring discussions with Alex Lazarian, Ethan Vishniac, and the support for this work provided by NASA through the NASA Hubble Fellowship grant # HST-HF2-51473.001-A awarded by the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Incorporated, under NASA contract NAS5-26555. H.L. acknowledges useful discussions with Fan Guo, Eric Zirnstein, and the support by LANL LDRD program.

Software: MATLAB [43]

Appendix

Anisotropy of strong MHD turbulence

The theoretically established scale-dependent anisotropy of trans-Alfvénic turbulence ([23]) and sub- and super-Alfvénic turbulence ([35]) has been tested by MHD turbulence simulations ([9] (Cho & Lazarian 2003)) and supported by spacecraft measurements in the solar wind ([26, 42, 16]). Here we briefly review the anisotropy of strong Alfvénic turbulence.

The cascading rate of Alfvénic turbulence in strong MHD turbulence regime is

τcas−1=vl⊥l⊥=VstLst−1/3l⊥−2/3(23)\tau_{\mathrm{cas}}^{-1} = \frac{v_l^\perp}{l_\perp} = V_{\mathrm{st}}L_{\mathrm{st}}^{-1/3}l_\perp^{-2/3} \tag*{(23)}

where VstV_{\mathrm{st}} is the turbulent velocity at the outer scale LstL_{\mathrm{st}} of strong MHD turbulence. More specifically, there is

Vst=VA,Lst=lA=LMA−3(24)V_{\mathrm{st}} = V_{\mathrm{A}}, L_{\mathrm{st}} = l_{\mathrm{A}} = LM_{\mathrm{A}}^{-3} \tag*{(24)}

for super-Alfvénic turbulence with MA>1M_{\mathrm{A}} > 1, and

Vst=VLMA,Lst=ltran=LMA2(25)V_{\mathrm{st}} = V_{\mathrm{L}}M_{\mathrm{A}}, L_{\mathrm{st}} = l_{\mathrm{tran}} = LM_{\mathrm{A}}^2 \tag*{(25)}

for sub-Alfvénic turbulence with MA<1M_{\mathrm{A}} < 1. We note that VAV_{\mathrm{A}} should be replaced by VAiV_{\mathrm{Ai}} when the turbulence is injected in the decoupled regime.

By combining the expression of τcas−1\tau_{\mathrm{cas}}^{-1} with the critical balance relation

τcas−1=VAl∥,(26)\tau_{\mathrm{cas}}^{-1} = \frac{V_{\mathrm{A}}}{l_{\parallel}}, \tag*{(26)}

we can obtain the anisotropic scaling relation of strong Alfvénic turbulence,

l∥=VAVstLst1/3l⊥2/3.(27)l_{\parallel} = \frac{V_{\mathrm{A}}}{V_{\mathrm{st}}}L_{\mathrm{st}}^{1/3}l_\perp^{2/3}. \tag*{(27)}

It shows that smaller-scale turbulent eddies are more elongated along the local magnetic field.

Damping scales of MHD turbulence cascade in a partially ionized medium

Under the consideration of both IN and NV damping effects in a partially ionized medium, Xu et al. (2015, 2016); Xu & Lazarian (2017) derived the general expression of the damping rate of Alfvénic turbulence. Here we briefly review its approximate form in different coupling regimes. In the weakly coupled regime, there is only the IN damping, with the damping rate

ωd=νin2.(28)\omega_d = \frac{\nu_{in}}{2}. \tag*{(28)}

For the MHD turbulence injected in the decoupled regime, as the cascading rate is always larger than νin\nu_{in} and thus larger than ωd\omega_d, the cascade is not cut off by the damping in a partially ionized medium.

In the strongly coupled regime, the damping rate can be approximately written as

ωd=ξn2(k2νn+k∥2(δVA)2νni)=ξn2(k2νn+k∥2VA2νni),(29)\omega_d = \frac{\xi_n}{2}\left(k^2\nu_n+\frac{k_\parallel^2(\delta V_A)^2}{\nu_{ni}}\right)=\frac{\xi_n}{2}\left(k^2\nu_n+\frac{k_\parallel^2V_A^2}{\nu_{ni}}\right), \tag*{(29)}

where we assume that the magnetic fluctuations are mainly induced by Alfvénic turbulence and apply the critical balance relation for strong MHD turbulence, k⊥δVA=k⊥vk=k∥VAk_\perp\delta V_A=k_\perp v_k=k_\parallel V_A. Here vkv_k is the turbulent velocity at wavenumber kk. The first and second terms of ωd\omega_d correspond to NV and IN damping, respectively.

For the MHD turbulence injected in the strongly coupled regime, when the damping rate exceeds the cascading rate, MHD turbulence cascade is damped. We consider that the damping scale is in the strong MHD turbulence regime. By comparing Eq. (29) with Eq. (23), we find the damping scale

ldam,NV,⊥=(ξn2)34Lst14Vst−34(30)l_{\mathrm{dam,NV},\perp}=\left(\frac{\xi_n}{2}\right)^{\frac{3}{4}}L_{\mathrm{st}}^{\frac{1}{4}}V_{\mathrm{st}}^{-\frac{3}{4}} \tag*{(30)}

when the NV damping dominates over the IN damping, where we assume k∼k⊥k\sim k_\perp, and

ldam,IN,⊥=(2νniξn)−23Lst−12Vst32(31)l_{\mathrm{dam,IN},\perp}=\left(\frac{2\nu_{ni}}{\xi_n}\right)^{-\frac{2}{3}}L_{\mathrm{st}}^{-\frac{1}{2}}V_{\mathrm{st}}^{\frac{3}{2}} \tag*{(31)}

in the opposite case. If the MHD turbulence is damped due to neutral viscosity in the strongly coupled regime, there is the so-called new regime of MHD turbulence on scales below the NV damping scale ([36]). If the damping of MHD turbulence is caused by neutral-ion decoupling, hydrodynamic turbulent cascade can happen in neutrals that are decoupled from ions on scales smaller than the IN damping scale ([4] (Xu et al. 2015)).

References

  1. [1]Armstrong, J. W., Rickett, B. J., & Spangler, S. R. 1995, ApJ, 443, 209, doi: 10.1086/175515DOI
  2. [2]Balsara, D. S. 1996, ApJ, 465, 775, doi: 10.1086/177462DOI
  3. [3]Breitschwerdt, D., Tautz, R. C., & de Avillez, M. A. 2017, Impact of Supernovae on the Interstellar Medium and the Heliosphere, ed. A. W. Alsabti & P. Murdin (Springer International Publishing AG), 2359, doi: 10.1007/978-3-319-21846-5_18DOI
  4. [4]Burkhart, B., Lazarian, A., Balsara, D., Meyer, C., & Cho, J. 2015, ApJ, 805, 118, doi: 10.1088/0004-637X/805/2/118DOI
  5. [5]Burlaga, L. F., Florinski, V., & Ness, N. F. 2018, ApJ, 854, 20, doi: 10.3847/1538-4357/aaa45aDOI
  6. [6]Burlaga, L. F., Ness, N. F., Berdichevsky, D. B., et al. 2022, ApJ, 932, 59, doi: 10.3847/1538-4357/ac658eDOI
  7. [7]Chepurnov, A., & Lazarian, A. 2010, ApJ, 710, 853, doi: 10.1088/0004-637X/710/1/853DOI
  8. [8]Chepurnov, A., Lazarian, A., Stanimirović, S., Heiles, C., & Peek, J. E. G. 2010, ApJ, 714, 1398, doi: 10.1088/0004-637X/714/2/1398DOI
  9. [9]Cho, J., & Lazarian, A. 2002, Physical Review Letters, 88, 245001, doi: 10.1103/PhysRevLett.88.245001
  10. [11]Cho, J., Lazarian, A., & Vishniac, E. T. 2002, ApJL, 566, L49, doi: 10.1086/339453DOI
  11. [12]Cho, J., & Vishniac, E. T. 2000, ApJ, 539, 273, doi: 10.1086/309213DOI
  12. [13]Crutcher, R. M., Wandelt, B., Heiles, C., Falgarone, E., & Troland, T. H. 2010, ApJ, 725, 466, doi: 10.1088/0004-637X/725/1/466DOI
  13. [14]Dialynas, K., Krimigis, S. M., Decker, R. B., et al. 2022, SSRv, 218, 21, doi: 10.1007/s11214-022-00889-0DOI
  14. [15]Draine, B. T., Roberge, W. G., & Dalgarno, A. 1983, ApJ, 264, 485, doi: 10.1086/160617DOI
  15. [16]Forman, M. A., Wicks, R. T., & Horbury, T. S. 2011, ApJ, 733, 76, doi: 10.1088/0004-637X/733/2/76DOI
  16. [17]Fraternale, F., & Pogorelov, N. V. 2021, ApJ, 906, 75, doi: 10.3847/1538-4357/abc88aDOI
  17. [18]Fraternale, F., Adhikari, L., Fichtner, H., et al. 2022, arXiv e-prints, arXiv:2207.14115, https://arxiv.org/abs/2207.14115
  18. [19]Frisch, P. C., Redfield, S., & Slavin, J. D. 2011, ARA&A, 49, 237, doi: 10.1146/annurev-astro-081710-102613DOI
  19. [20]Gaensler, B. M., Haverkorn, M., Burkhart, B., et al. 2011, Nature, 478, 214, doi: 10.1038/nature10446DOI
  20. [21]Galtier, S., Nazarenko, S. V., Newell, A. C., & Pouquet, A. 2000, Journal of Plasma Physics, 63, 447, doi: 10.1017/S0022377899008284DOI
  21. [22]Giacalone, J., & Jokipii, J. R. 2015, ApJL, 812, L9, doi: 10.1088/2041-8205/812/1/L9DOI
  22. [23]Goldreich, P., & Sridhar, S. 1995, ApJ, 438, 763, doi: 10.1086/175121DOI
  23. [26]Horbury, T. S., Forman, M., & Oughton, S. 2008, PhRvL, 101, 175005, doi: 10.1103/PhysRevLett.101.175005DOI
  24. [28]Izmodenov, V. V., & Alexashov, D. B. 2020, A&A, 633, L12, doi: 10.1051/0004-6361/201937058DOI
  25. [29]Krimigis, S. M., Decker, R. B., Roelof, E. C., et al. 2013, Science, 341, 144, doi: 10.1126/science.1235721DOI
  26. [30]Krstic, P., & Schultz, D. 1998, Atomic and Plasma-Material Interaction Data for Fusion, 8
  27. [31]Kulsrud, R., & Pearce, W. P. 1969, ApJ, 156, 445, doi: 10.1086/149981DOI
  28. [32]Kulsrud, R. M., & Anderson, S. W. 1992, ApJ, 396, 606, doi: 10.1086/171743DOI
  29. [34]Lazarian, A., & Opher, M. 2009, ApJ, 703, 8, doi: 10.1088/0004-637X/703/1/8DOI
  30. [35]Lazarian, A., & Vishniac, E. T. 1999, ApJ, 517, 700, doi: 10.1086/307233DOI
  31. [36]Lazarian, A., Vishniac, E. T., & Cho, J. 2004, ApJ, 603, 180, doi: 10.1086/381383DOI
  32. [37]Lazarian, A., & Xu, S. 2021, ApJ, 923, 53, doi: 10.3847/1538-4357/ac2de9DOI
  33. [38]Lazarian, A., Yuen, K. H., Ho, K. W., et al. 2018, ApJ, 865, 46, doi: 10.3847/1538-4357/aad7ffDOI
  34. [39]Lee, K. H., & Lee, L. C. 2020, ApJ, 904, 66, doi: 10.3847/1538-4357/abba20DOI
  35. [40]Lithwick, Y., & Goldreich, P. 2001, ApJ, 562, 279, doi: 10.1086/323470DOI
  36. [41]López-Barquero, V., Xu, S., Desiati, P., et al. 2017, ApJ, 842, 54, doi: 10.3847/1538-4357/aa74d1
  37. [42]Luo, Q. Y., & Wu, D. J. 2010, ApJL, 714, L138, doi: 10.1088/2041-8205/714/1/L138
  38. [43]MATLAB 2021, MATLAB and Statistics Toolbox Release 2021b (Natick, Massachusetts: The MathWorks Inc.)
  39. [44]Mouschovias, T. C., Ciolek, G. E., & Morton, S. A. 2011, MNRAS, 415, 1751, doi: 10.1111/j.1365-2966.2011.18817.xDOI
  40. [45]Ocker, S. K., Cordes, J. M., Chatterjee, S., et al. 2021, Nature Astronomy, 5, 761, doi: 10.1038/s41550-021-01363-7DOI
  41. [46]Pattle, K., Fissel, L., Tahani, M., Liu, T., & Ntormousi, E. 2022, arXiv:2203.11179, arXiv:2203.11179. https://arxiv.org/abs/2203.11179
  42. [47]Pogorelov, N. V., Heerikhuisen, J., Roytershteyn, V., et al. 2017, ApJ, 845, 9, doi: 10.3847/1538-4357/aa7d4fDOI
  43. [48]Pudritz, R. E. 1990, ApJ, 350, 195, doi: 10.1086/168373DOI
  44. [49]Shu, F. H. 1992, The physics of astrophysics. Volume II: Gas dynamics.
  45. [50]Slavin, J. D., & Frisch, P. C. 2008, A&A, 491, 53, doi: 10.1051/0004-6361:20078011
  46. [51]Soler, R., Carbonell, M., Ballester, J. L., & Terradas, J. 2013, ApJ, 767, 171, doi: 10.1088/0004-637X/767/2/171DOI
  47. [52]Spangler, S. R., Savage, A. H., & Redfield, S. 2011, ApJ, 742, 30, doi: 10.1088/0004-637X/742/1/30DOI
  48. [53]Stone, E. C., Cummings, A. C., McDonald, F. B., et al. 2013, Science, 341, 150, doi: 10.1126/science.1236408DOI
  49. [54]Swaczyna, P., McComas, D. J., Zirnstein, E. J., et al. 2020, ApJ, 903, 48, doi: 10.3847/1538-4357/abb80aDOI
  50. [55]Vranjes, J., & Krstic, P. S. 2013, A&A, 554, A22, doi: 10.1051/0004-6361/201220738DOI
  51. [62]Yuen, K. H., Ho, K. W., Law, C. Y., Chen, A., & Lazarian, A. 2022, arXiv:2204.13760, arXiv:2204.13760. https://arxiv.org/abs/2204.13760
  52. [63]Zank, G. P. 2015, ARA&A, 53, 449, doi: 10.1146/annurev-astro-082214-122254DOI
  53. [64]Zank, G. P., Heerikhuisen, J., Wood, B. E., et al. 2013, ApJ, 763, 20, doi: 10.1088/0004-637X/763/1/20DOI
  54. [65]Zank, G. P., Nakanotani, M., & Webb, G. M. 2019, ApJ, 887, 116, doi: 10.3847/1538-4357/ab528cDOI
  55. [66]Zaqarashvili, T. V., Khodachenko, M. L., & Rucker, H. O. 2011, A&A, 529, A82, doi: 10.1051/0004-6361/201016326DOI
  56. [67]Zhao, L. L., Zank, G. P., & Burlaga, L. F. 2020, ApJ, 900, 166, doi: 10.3847/1538-4357/ababa2DOI
  57. [68]Zirnstein, E. J., Giacalone, J., Kumar, R., et al. 2020, ApJ, 888, 29, doi: 10.3847/1538-4357/ab594dDOI

Paper details

vlnc:W_n2bt5vsn
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