1: \documentclass[11pt,twoside]{article}

2: \usepackage{asp2004}

3: \usepackage{psfig}

4: \usepackage{epsf}

5: \usepackage{graphics}

6: \usepackage{lscape}

7: \markboth{Dastgeer Shaikh }{ISM Turbulence}

8: \pagestyle{myheadings}

9: 

10: % Some definitions I use in these instructions.

11: 

12: \def\emphasize#1{{\itshape#1\/}}

13: \def\arg#1{{\it#1\/}}

14: \let\prog=\arg

15: \parsep 0pt

16: \itemsep 0pt

17: \def\edcomment#1{\iffalse\marginpar{\raggedright\sl#1\/}\else\relax\fi}

18: \marginparwidth 1.25in

19: \marginparsep .125in

20: \marginparpush .25in

21: \reversemarginpar

22: 

23: %---my commands

24: \newcommand{\itemfig}[1]{\item{\bf Fig (#1) }} 

25: \newcommand{\be}{\begin{equation}}

26: \newcommand{\ee}{\end{equation}}

27: \newcommand{\et}{{\it{et. al. }}}

28: \newcommand{\eqs}[2]{Eqs. (\ref{#1}) \& (\ref{#2})} 

29: \newcommand{\Eqs}[2]{Equations  (\ref{#1}) \& (\ref{#2})} 

30: \newcommand{\Eq}[1]{Equation (\ref{#1})} 

31: \newcommand{\eq}[1]{Eq. (\ref{#1})} 

32: \newcommand{\fig}[2]{Fig. (\ref{#1}{-#2})} 

33: \newcommand{\Fig}[1]{Fig. (\ref{#1})} 

34:  \newcommand{\eqa}{\begin{eqnarray}}

35: \newcommand{\eeq}{\end{eqnarray}}  

36: \newcommand{\eqsto}[2]{Eqs. (\ref{#1}) to (\ref{#2})} 

37: \newcommand{\sig}{\sigma_{\rm ex}}

38: %------

39: 

40: \begin{document}

41: \title{Non Kolmogorov-like Turbulence in the Local Interstellar Medium}

42: \author{Dastgeer Shaikh}

43: \affil{Institute of Geophysics and Planetary Physics,\\

44: University of California, Riverside, CA 92521. USA.}

45: 

46: 

47: \begin{abstract}

48: We develop a self-consistent model of turbulence in a local

49: interstellar medium (ISM).  The model describes a partially ionized

50: magnetofluid ISM in which a neutral hydrogen fluid interacts with a

51: plasma dominantly through a charge exchange. The ISM turbulent

52: correlation scales in our model are much bigger than the shock

53: characteristic length-scales. Unlike small length-scale linear

54: collisional dissipation in the fluid, the charge exchange processes

55: can be effective unpredictably on a variety of ISM length-scales

56: depending upon the neutral and plasma densities, the charge exchange

57: cross section and the characteristic length scales. We find, from

58: scaling arguments that the charge exchange interactions modify

59: spectral transfer associated with large-scale energy containing

60: eddies. Consequently, the ISM turbulent cascade are steeper than those

61: predicted by Kolmogorov's phenomenology.

62: \end{abstract}

63: 

64: \vspace{-0.5cm}

65: 

66: \section{Introduction and Model Equations}

67: Small scale turbulence in the local interstellar medium (ISM) is a

68: largely unexplored field given the complexity associated with the ISM

69: turbulent processes \citep{dastgeer}. The local ISM comprises of

70: partially ionized, magnetized plasma protons and almost equal number

71: of neutral particles.  The plasma and the neutral particles in ISM

72: interact mutually through charge exchange.  The physics of these

73: small-scale turbulent motions is far more complex than ever

74: thought. Not only that it holds the key to our crucial understanding

75: of the global heliospheric interactions such as nature and

76: characteristic of heliospheric shocks, heating etc

77: \citep{zank99,pauls}, it is also increasingly believed to be pivotal

78: to many puzzles of astrophysics including origin and transport of

79: cosmic rays, Fermi acceleration, gamma-ray bursts, ISM density spectra

80: etc. Yet, there exists {\it no} self-consistent simulation model that

81: unravels multi-component and multiple-scale ISM turbulent phenomena.

82: A prime goal of this paper is therefore to develop a self-consistent

83: plasma-neutral ISM turbulence model based on analytic methods and

84: numerical simulations.

85: 

86: The underlying model is based on the following assumptions.

87: Fluctuations in the plasma and the neutral fluids are sufficiently

88: isotropic, homogeneous, thermally equilibrated and turbulent. No mean

89: magnetic field and velocity flows are present at the outset. There

90: however may generate local mean flows due to self-consistently excited

91: nonlinear instabilities. The characteristic turbulent correlation

92: length-scales are typically smaller than charge-exchange mean free

93: path lengths in the ISM flows. Nevertheless, they are large enough to

94: treat any localized shocks as discontinuities. In other words, the

95: characteristic shock length-scales are relatively small compared to

96: the ISM turbulent fluctuation length-scales, and finally boundary

97: conditions are periodic. It should be further noted that the neutrals

98: coming from the solar wind are not considered because they tend to

99: anisotropize the distribution functions substantially.  Our model thus

100: simulates a localized ISM. The fluid model describing nonlinear

101: turbulent processes in the interstellar medium, in the presence of

102: charge exchanges forces, can then be cast into plasma density

103: ($\rho_p$), velocity (${\bf U}_p$), magnetic field (${\bf B}$),

104: pressure ($P_p$), as follows.  \be

105: \label{mhd:cont}

106:  \frac{\partial \rho_p}{\partial t} + \nabla \cdot (\rho_p{\bf U}_p)=0,

107: \ee

108: \be

109: \label{mhd:mom}

110: \rho_p \left(  \frac{\partial }{\partial t} + {\bf U}_p \cdot \nabla \right) {\bf U}_p

111: = -\nabla P_p + \frac{1}{c} {\bf J} \times {\bf B}+{\bf Q}_M({\bf U}_p,{\bf V}_n),

112: \ee

113: \be

114: \label{mhd:mag}

115:  \frac{\partial {\bf B}}{\partial t} = \nabla \times ({\bf U}_p \times {\bf B}),

116: \ee

117: \be

118: \label{mhd:en}

119:  \frac{\partial e}{\partial t} 

120: + \nabla \cdot \left( \frac{1}{2}\rho_p U_p^2{\bf U}_p 

121: + \frac{\gamma}{\gamma-1}\frac{P_p}{\rho_p}\rho_p{\bf U}_p

122: + \frac{c}{4\pi}{\bf E} \times {\bf B}  \right)

123: =Q_E({\bf U}_p,{\bf V}_n).

124: \ee

125: where $e=1/2\rho_p U_p^2 + P_p/(\gamma-1)+B^2/8\pi$.

126: The above set of plasma equations is coupled self-consistently to the

127: ISM neutral density ($\rho_n$), velocity (${\bf V}_n$) and pressure

128: ($P_n$) through a set of hydrodynamic fluid equations as below.

129: \be

130: \label{hd:cont}

131:  \frac{\partial \rho_n}{\partial t} + \nabla \cdot (\rho_n{\bf V}_n)=0,

132: \ee

133: \be

134: \label{hd:mom}

135: \rho_n \left(  \frac{\partial }{\partial t} + {\bf V}_n \cdot \nabla \right) {\bf V}_n

136: = -\nabla P_n + {\bf Q}_M({\bf V}_n,{\bf U}_p),

137: \ee

138: \be

139: \label{hd:en}

140:  \frac{\partial}{\partial t} \left( \frac{1}{2}\rho_n V_n^2 +

141:  \frac{P_n}{\gamma-1} \right) + \nabla \cdot \left( \frac{1}{2}\rho_n

142:  V_n^2{\bf V}_n + \frac{\gamma}{\gamma-1}\frac{P_n}{\rho_n}\rho_n{\bf

143:    V}_n \right) =Q_E({\bf V}_n,{\bf U}_p).  

144: \ee 

145: Equations (\ref{mhd:cont}) to (\ref{hd:en}) form an entirely

146: self-consistent description of the coupled ISM plasma-neutral

147: turbulent fluid.  The charge-exchange momentum sources in the plasma

148: and the neutral fluids are described respectively by terms ${\bf

149:   Q}_M({\bf U}_p,{\bf V}_n)$ and ${\bf Q}_M({\bf V}_n,{\bf U}_p)$.

150: Similarly, charge exchange energy sources in the plasma and the

151: neutral fluids are given by $Q_E({\bf U}_p,{\bf V}_n)$ and $Q_E({\bf

152:   V}_n,{\bf U}_p)$ respectively.  In the absence of charge exchange

153: interactions, the plasma and the neutral fluid are de-coupled

154: trivially and behave as ideal fluids.  While the charge-exchange

155: interactions modify instantaneous momentum and the energy of plasma

156: and the neutral fluids, they tend to conserve density in both the

157: fluids. Nonetheless, the volume integrated energy ($\int Q_E({\bf

158:   U}_p,{\bf V}_n) d{\bf v}= - \int Q_E({\bf V}_n,{\bf U}_p) d{\bf v} $) and the

159: density ($\int \rho_p d{\bf v}=const, \int \rho_n d{\bf v}=const$) of

160: the entire coupled system will remain conserved in the absence of

161: external sources/sinks.

162: 

163: 

164: 

165: The ISM turbulence model is normalized using the typical ISM

166: scale-length ($\ell_0$), density ($\rho_0$) and velocity ($v_0$). The

167: normalized plasma density, velocity, energy and the magnetic field are

168: respectively; $\bar{\rho}_p = \rho_p/\rho_0, \bar{\bf U}_p={\bf

169: U}_p/v_0, \bar{P}_p=P_p/\rho_0v_0^2, \bar{\bf B}={\bf

170: B}/v_0\sqrt{\rho_0}$. The corresponding neutral fluid quantities are

171: $\bar{\rho}_n = \rho_n/\rho_0, \bar{\bf U}_n={\bf U}_n/v_0,

172: \bar{P}_n=P_n/\rho_0v_0^2$. The momentum and the energy

173: charge-exchange forces, in the normalized form, are respectively

174: $\bar{\bf Q}_m={\bf Q}_m \ell_0/\rho_0v_0^2, \bar{Q}_e=Q_e

175: \ell_0/\rho_0v_0^3$. The non-dimensional temporal and spatial

176: length-scales are $\bar{t}=tv_0/\ell_0, \bar{\bf x}={\bf

177: x}/\ell_0$.   The charge-exchange cross-section parameter

178: ($\sigma$), not appeared directly in the above set of equations, is

179: normalized as $\bar{\sigma}=n_0 \ell_0 \sigma$.  We define an

180: intrinsic mode in ISM turbulence, the charge exchange mode $k_{ce}\sim

181: (n_0\ell_0)^{1/2}$. The latter is different from the characteristic

182: turbulent mode $k$, and is excited naturally when plasma and neutral

183: fluids are coupled in the ISM by charge exchange forces [see

184: \citet{dastgeer} for more detail].

185: 

186: 

187: 

188: 

189: \section{ Nonlinear Simulation}

190: 

191: 

192: Two-dimensional (2D) nonlinear spectral fluid code is developed to

193: numerically integrate \eqsto{mhd:cont}{hd:en}. The 2D simulations are

194: not only computationally simpler and less expensive (compared with the

195: full 3D), they offer significantly higher resolutions even on moderate

196: size small cluster machines like Beowulf.  The spatial discretization

197: in our code uses discrete Fourier representation of turbulent

198: fluctuations based on a spectral method, $\tilde{f}({\bf

199:   x,t})=\sum_{\bf k} f({\bf k},t) e^{-i{\bf k} \cdot {\bf x}}$, where

200: ${\bf k}=k_x \hat{e}_x + k_y \hat{e}_y$ is a two-dimensional wave

201: vector. The nonlinear deconvolution of Fourier modes is performed by

202: computing nonlinear triad interaction $\tilde{f}({\bf x,t})

203: \tilde{g}({\bf x,t}) = \sum_{{\bf k}={\bf k}'-{\bf k}''} f({\bf

204:   k}',t)g({\bf k}'',t) \delta({\bf k}'-{\bf k}'')$ which survive for

205: only those mode coupling interactions which satisfy Fourier diad

206: constraint ${\bf k}={\bf k}'-{\bf k}''$. These nonlinear interactions

207: in Fourier space conserve rugged invariants of system of

208: \eqsto{mhd:cont}{hd:en}.  The temporal integration is performed by

209: Runge Kutta 4 method. The fluctuations in plasma and neutral are

210: initialized isotropically (no mean fields are assumed) with random

211: phases and with identical amplitudes in the Fourier space.  This

212: algorithm ensures conservation of total energy and mean fluid density

213: per unit time in the absence of charge exchange and external random

214: forcing.  Additionally, it satifies the condition of incompressibility

215: associated with the magnetic field, i.e. $\int |\nabla \cdot {\bf

216:   B}|^2 d{\bf v} \approx 10^{-15}$ at each time step. We make use of

217: an artificial scalar potential to achieve $\nabla \cdot {\bf B}=0$ as

218: described in \citet{nick}.  Numerical resolution is $512^2$ for the

219: mode structures and $1024^2$ for the spectrum calculations.  Our code

220: is massively parallelized using Message Passing Interface (MPI)

221: libraries to facilitate higher resolution.  While the ISM turbulence

222: code is evolved with time steps resolved self-consistently by the

223: coupled fluid motions, the nonlinear interaction time scales

224: associated with the plasma $1/{\bf k} \cdot {\bf U}_p({\bf k})$ and

225: the neutral $1/{\bf k} \cdot {\bf V}_n({\bf k})$ fluids can obviously

226: be disparate. Accordingly, turbulent transport of energy in the plasma

227: and the neutral ISM fluids takes place distinctively on separate time

228: scales.

229: \begin{figure}[!t]

230: \begin{center}

231: \psfig{file=plasma1669r.ps,width=4.5in}

232: \end{center}

233:   \caption{Self-consistent simulations of turbulence in the partially

234:     ionized ISM. Figure shows evolution of the plasma density,

235:     velocity, magnetic field, temperature and pressure. Numerical resolution is

236: $512^2$.}

237: \end{figure}

238: Small scale turbulent fluctuations in ISM evolve under the action of

239: nonlinear interactions as well as charge exchange sources. Energy

240: cascades amongst turbulent eddies of various scale sizes and between

241: the plasma and the neutral fluids. Because of the discrepant nonlinear

242: time scales associated with the plasma and neutral fluids, mode

243: structures can be different in the two fluids when they are evolved

244: together and in isolation (i.e. decoupled). When the plasma and

245: neutral components are decoupled, the plasma fluid evolves in

246: accordance with the ideal MHD where current sheets are typically

247: formed in the magnetic field structures.  In the presence of charge

248: exchange, the two fluids evolve in a self-consistent manner and modify

249: the dynamical properties of both the fluids. One of the most

250: noticeable features to emerge from the coupled evolution is that the

251: spectral cascade rates are enhanced substantially in contrast to that

252: of the decoupled case. The rapid spectral transfer of energy amongst

253: various Fourier modes in the coupled plasma neutral gas leads to

254: smearing off the current sheets in the plasma on a shorter time

255: scales. The latter evolves on a relatively long time scales and the

256: sheets are formed eventually in the magnetic field. The turbulent

257: equipartition is set up in the plasma and the neutral fluid

258: modes. This is shown in Fig 1 where all the quantities associated with

259: the plasma evolution in our simulations are depicted.  Seemingly, the

260: neutral fluid, under the action of charge exchange sources, tends to

261: enhance the cascades rates by isotropizing the ISM turbulence on rapid

262: time scales.  One of the plausible reasons of the fast-cascade is that

263: the energy transfer rates are enhanced effectively in the coupled ISM

264: turbulence system, and that they now require a smaller time to mix the

265: plasma and the neutral fluids.  In any event, the small-scale

266: sheet-like structures in magnetic field compresses (or pinches) the

267: plasma density. Hence the density fluctuations develop identically the

268: (thinner than) sheet-like structures that co-exist with small-scale

269: turbulent fluctuations in its spectrum as shown in Fig 1. The neutral

270: fluid, on the other hand, evolves isotropically as stated above by

271: forming relatively large-scale structures as shown in Fig 2.

272: 

273: 

274: \begin{figure}[!t]

275: \begin{center}

276: \psfig{file=neutral1600r.ps,width=4.2in}

277: \end{center}

278:   \caption{Evolution of neutral fluid in the coupled plasma neutral

279:     ISM turbulence. Shown are the neutral density, velocity,

280:     temperature and pressure fluctuations.}

281: \end{figure}

282: 

283: \section{Energy Cascade Rates in the Coupled ISM Plasma Neutral Gas}

284: As mentioned above, the plasma neutral fluid coupling in the ISM

285: fluctuations introduces charge exchange modes, $k_{ce}$, that are

286: distinctively different from the characteristic turbulent mode

287: $k$. Typically, $k_{ce}/k<1$ in the local ISM. Accordingly, the energy

288: cascade time scales are modified as discussed below.  The nonlinear

289: interaction time-scale in ordinary turbulence is given by

290: \[\tau_{nl} \sim \frac{\ell_0}{v_\ell} \sim (kv_k)^{-1},\] 

291: where $v_k$ or $v_\ell$ is the velocity of turbulent eddies. In the

292: presence of charge exchange interaction, the ordinary nonlinear

293: interaction time-scales of fluid turbulence is modified by a factor

294: $k_{ce}/k$ such that the new nonlinear interaction time-scale of ISM

295: turbulence is now 

296: \[\tau_{NL} \sim \frac{k_{ce}}{k} \frac{1}{kv_k}.\] 

297: On using the fact that $k_{ce}$ is typically smaller than $k$,

298: i.e. $k_{ce}/k<1$ in the ISM, the new nonlinear time is $k_{ce}/k$

299: times smaller than the old nonlinear time i.e.  $\tau_{NL} \sim

300: (k_{ce}/k) \tau_{nl}$. This reduced nonlinear interaction time in ISM

301: turbulence is likely to enhance turbulent energy cascade rates that

302: are determined typically by $E_k/\tau_{NL}$, where $E_k$ is energy per

303: unit mode.  It is beacuse of this reduced interaction time that a

304: rapid spectral transfer of turbulent modes tends to smear off the

305: current sheets in the magnetic field fluctuations.

306: 

307: 

308: It is interesting to note that the enhanced cascade rates of ISM

309: turbulent modes tend to steepen the inertial range turbulent spectra

310: in both plasma and neutral fluids. By extending above phenomenological

311: analysis, one can deduce exact (analytic) spectral indices of the

312: inertial range decaying turbulent spectra, as follows. The new

313: nonlinear interaction time-scale of ISM turbulence can be rearranged

314: as $ \tau_{NL} \sim

315: k_{ce}v_k /(kv_k)(1/kv_k)\sim ({\tau_{nl}^2}/{\tau_{ce}})$,

316: where $\tau_{ce}\sim (k_{ce} v_k)^{-1}$ represents charge exchange

317: time scale. The energy dissipation rate associated with the coupled

318: ISM plasma-neutral system can be determined from $\varepsilon \sim

319: E_k/\tau_{NL}$ relation, which leads to 

320: $\varepsilon \sim {v_k^2}/({k_{ce}/k^2 v_k})\sim {k^2 v_k^3}/{k_{ce}}$.

321: According to the Kolmogorov theory, the spectral cascades are local in

322: $k$-space and the inertial range energy spectrum depends upon the

323: energy dissipation rates and the characteristic turbulent modes, such

324: that $E_k \sim \varepsilon^{\gamma} k^{\beta}$.  Upon substitution of

325: above quantities and equating the power of identical bases, one

326: obtains

327: \[E_k \sim \varepsilon^{2/3} k^{-7/3}\] 

328: plasma spectrum.  Similar arguments in the context of neutral fluids,

329: when coupled with the plasma fluid in ISM, lead to the energy

330: dissipation rates $\varepsilon \sim k^2 v_k^2/({k_{ce}/k^2

331:   v_k})$. This further yields the forward cascade (neutral) energy

332: spectrum 

333: \[E_k \sim \varepsilon^{2/3} k^{-11/3}.\]

334: 

335: 

336: 

337: 

338: 

339: \section{Conclusion}

340: In conclusion, one of the most important points to emerge from our

341: studies is that the charge exchange modes modify the ISM turbulence

342: cascades dramatically by expediting nonlinear interaction time-scales.

343: consequently, the energy cascade rates are enhanced by isotropizing

344: the ISM turbulence on rapid time scales.  This tends to modify the

345: characteristics of ISM turbulence which can be significantly different

346: from the Kolmogorov phenomenology of fully developed turbulence. As

347: shown above, the charge exchange interactions lead to a steeper power

348: spectrum.  Although turbulence in 2D and 3D possesses distinct

349: spectral features characterized essentially by the number of the

350: inviscid quadratic invariants, our 2D simulations provide a hint that

351: the ISM turbulent spectra in 3D case will exhibit a non

352: Kolmogorov-like characteristic.  It is to be noted that the present

353: model does not consider driving mechanism, hence turbulence is freely

354: decaying.  Driven turbulence, such as due to large scale forcing,

355: supernova explotion and CME ejection may force turbulence at larger

356: scales. This can modify the cascade dynamics in a manner usually

357: described by dual cascade process.

358: 

359: 

360: 

361: 

362: 

363: \begin{thebibliography}{}

364: 

365: \bibitem[Zank (1999)]{zank99}

366:  Zank, G. P. 1999, Sp. Sci. Rev., 89, 413

367: 

368: \bibitem[Pauls {\it et. al.} (1995)]{pauls}

369: Pauls, H. L., Zank, G. P., \& Williams, L. L. 1995

370: J. Geophys. Res. A11, 21595

371: 

372: \bibitem[Shaikh {\it et. al.} (2006)]{dastgeer}

373: Shaikh, D. Pogorelov, N. \& Zank, G. P., 2006, 

374: AIP Conf. Procs, {\bf 858}, 314.

375: 

376: \bibitem[Kulikovskii {\it et al.} (2001)]{nick}

377: Kulikovskii, A. G., Pogorelov, N. V., and Semenov, A. Y., 

378: {\it Mathematical Aspects of Numerical Solution of Hyperbolic Systems},

379: Chapman \& Hall/CRC, 118, Pg. 50.

380: 

381: %\bibitem[Dastgeer et al. (2004)]{dastgeer04}

382: %S. Dastgeer, G. P. Zank, N. Pogorelov, AIP Proceedings, Physics of

383: %the outer heliosphere, 719, 323, 2004

384: 

385: \end{thebibliography}

386: 

387: \end{document}

388: 

389: It is worth emphasizing that the interaction of the neutrals with the

390: weakly ionized plasma species introduces a widely separated time and

391: length scales in ISM turbulence. For instance, the plasma neutral

392: fluid coupling in the ISM fluctuations introduces charge exchange

393: modes, $k_{ce}$, that are distinctively different from the

394: characteristic turbulent mode $k$. Typically, $k_{ce}/k<1$ in the

395: local ISM. It is to be borne further in mind that the charge exchange

396: sources in the neutral and the plasma fluids do not merely damp the

397: smallest scale fluctuations such as ones due to linear collisional

398: dissipation in the turbulent fluid flows. By contrast, the effect of

399: the charge exchange forces may far more anamalously be different from

400: the small scale turbulent damping and therefore they can alter the

401: nonlinear cascades dramatically. Beacuse of their invariably complex

402: nature, the charge exchage forces can be {\it effective on a variety

403: of turbulent length-scales} unlike the linear disspation which is

404: operative on the small length-scales only.  To understand the effects

405: of charge exchange, we carry out two different sets of simulations of

406: the two-component (plasma-neutral) ISM system. In the first set,

407: \eqsto{mhd:cont}{hd:en} are decoupled (i.e. no charge exchange

408: coupling) and evolved as ideal fluids, whereas in the second set the

409: coupling is operative.

410: 

411: 

412: 

413: 

414:  as described in the following.  The

415: nonlinear interaction time-scale in ordinary turbulence is given by

416: \[\tau_{nl} \sim \frac{\ell_0}{v_\ell} \sim (kv_k)^{-1},\] 

417: where $v_k$ or $v_\ell$ is the velocity of turbulent eddies. In the

418: presence of charge exchange interaction, the ordinary nonlinear

419: interaction time-scales of fluid turbulence is modified by a factor

420: $k_{ce}/k$ such that the new nonlinear interaction time-scale of ISM

421: turbulence is now 

422: \[\tau_{NL} \sim \frac{k_{ce}}{k} \frac{1}{kv_k}.\] 

423: On using the fact that $k_{ce}$ is typically smaller than $k$,

424: i.e. $k_{ce}/k<1$ in the ISM, the new nonlinear time is $k_{ce}/k$

425: times smaller than the old nonlinear time i.e.  $\tau_{NL} \sim

426: (k_{ce}/k) \tau_{nl}$. This reduced nonlinear interaction time in ISM

427: turbulence is likely to enhance turbulent energy cascade rates that

428: are determined typically by $E_k/\tau_{NL}$, where $E_k$ is energy per

429: unit mode.  It is because of this reduced interaction time, that a

430: rapid spectral transfer of turbulent modes tends to steepen the

431: inertial range turbulent spectra in both plasma and neutral fluids. By

432: extending above phenomenological analysis, one can deduce exact

433: (analytic) spectral indices of the inertial range decaying turbulent

434: spectra, as follows. The new nonlinear interaction time-scale of ISM

435: turbulence can be rearranged as 

436: \[\tau_{NL} \sim

437: \frac{k_{ce}v_k}{kv_k}\frac{1}{kv_k}\sim \frac{\tau_{nl}^2}{\tau_{ce}},\] 

438: where $\tau_{ce}\sim (k_{ce} v_k)^{-1}$ represents charge exchange

439: time scale. The energy dissipation rate associated with the coupled

440: ISM plasma-neutral system can be determined from $\varepsilon \sim

441: E_k/\tau_{NL}$ relation, which leads to 

442: \[\varepsilon \sim \frac{v_k^2}{k_{ce}/k^2 v_k}\sim \frac{k^2 v_k^3}{k_{ce}}.\] According to the

443:  Kolmogorov theory, the spectral cascades are local in $k$-space and

444:  the inertial range energy spectrum depends upon the energy

445:  dissipation rates and the characteristic turbulent modes, such that

446:  $E_k \sim \varepsilon^{\gamma} k^{\beta}$.  Upon substitution of

447:  above quantities and equating the power of identical bases, one

448:  obtains 

449: \[E_k \sim \varepsilon^{2/3} k^{-7/3}\] 

450: plasma spectrum. The spectral index associated with this spectrum,

451:  i.e. $7/3 \approx 2.33$, is consistent with the plasma spectrum

452:  observed in the (coupled ISM plasma-neutral) simulations (see Fig

453:  2). Similar arguments in the context of neutral fluids, when coupled

454:  with the plasma fluid in ISM, lead to the energy dissipation rates

455:  \[\varepsilon \sim \frac{k^2 v_k^2}{k_{ce}/k^2 v_k}.\] This further yields

456:  the forward cascade (neutral) energy spectrum 

457: \[E_k \sim \varepsilon^{2/3} k^{-11/3}\]

458:  which is close to the simulation result 

459: 

460: 

461: 

462: ------------------

463: 

464: 

465: The charge exchange interactions couple the ISM plasma and the neutral

466: fluids by exciting a characteristic charge exchange coupling mode

467: $k_{ce}$, which is different from the turbulent mode $k$ of the

468: coupled system. One of the most important points to emerge from our

469: studies is that the charge exchange modes modify the ISM turbulence

470: cascades dramatically by expediting nonlinear interaction

471: time-scales. This tends to modify the characteristics of ISM

472: turbulence which can be significantly different from the Kolmogorov's

473: phenomenology of fully developed turbulence. The latter can be

474: elucidated further by invoking a simple argument, as follows. The

475: nonlinear interaction time-scales in ordinary turbulence is given by

476: $\tau_{nl} \sim \ell_0/v_\ell \sim (kv_k)^{-1}$, where $v_k$ or

477: $v_\ell$ is the velocity of turbulent eddies. In the presence of

478: charge exchange interaction, the ordinary nonlinear interaction

479: time-scales of fluid turbulence is now modified by a factor $k_{ce}/k$

480: such that the new nonlinear interaction time-scale of ISM turbulence

481: is now $\tau_{NL} \sim (k_{ce}/k)(1/kv_k)$. On using the fact that

482: $k_{ce}$ is typically smaller than $k$, i.e. $k_{ce}/k<1$ in ISM, the

483: new nonlinear time is $k_{ce}/k$ times smaller than the old nonlinear

484: time i.e.  $\tau_{NL} \sim (k_{ce}/k) \tau_{nl}$. This reduced

485: nonlinear interaction time in ISM turbulence is likely to enhance

486: turbulent energy cascade rates that are determined typically by

487: $E_k/\tau_{NL}$. We believe, it is beacuse of this reduced interaction

488: time, a rapid spectral transfer of turbulent modes tends to smear off

489: the current sheets in the magnetic field fluctuations. The latter

490: seems to explain, as well, other spectral features of our simulations

491: that will be discussed elsewhere.

492: