1: \documentclass[10pt]{article}
2: \usepackage[dvips]{graphicx}
3: %\usepackage[english,russian]{babel}
4: \usepackage{wrapfig}
5: \pagestyle{headings}
6:
7: \topmargin 0mm \textwidth 142.5mm \textheight 195mm \oddsidemargin 0mm
8:
9: \renewcommand{\figurename}{Fig.}
10:
11: \begin{document}
12: \noindent{\small UDK 61.80.Az, 74.25.Fy, 74.72.Bk, 02.70.Bf}\\
13:
14: \begin{center}
15: \Large{\bf NUMERICAL INVESTIGATION OF SCALING REGIMES IN A MODEL OF
16: ANISOTROPICALLY ADVECTED VECTOR FIELD}\\
17: \bigskip
18: \large{\it E. Jur\v{c}i\v{s}inov\'a,$^{1,2}$ M. Jur\v{c}i\v{s}in,$^{1,3}$ R. Remecky,$^{4}$ and M. Scholtz$^{4}$}\\
19: \smallskip
20: \small{$^{1}$ Institute of Experimental Physics, SAS, Watsonova 47, 04001 Ko\v{s}ice, Slovakia \\
21: $^{2}$ Laboratory of Information Technologies, JINR, 141 980 Dubna, Moscow Region, Russia \\
22: $^{3}$ Bogoliubov Laboratory of Theoretical Physics, JINR, 141 980
23: Dubna, Moscow Region,
24: Russia \\
25: $^{4}$ Department of Physics and Astrophysics, Institute of Physics,
26: P.J. \v{S}af\'arik University, \\ Park Angelinum 9, 04001
27: Ko\v{s}ice, Slovakia}
28: \end{center}
29:
30: {\small{Influence of strong uniaxial small-scale anisotropy on the
31: stability of inertial-range scaling regimes in a model of a passive
32: transverse vector field advected by an incompressible turbulent flow
33: is investigated by means of the field theoretic renormalization
34: group. Turbulent fluctuations of the velocity field are taken in the
35: form of a Gaussian statistics with zero mean and defined noise with
36: finite correlations in time. It is shown that stability of the
37: inertial-range scaling regimes in the three-dimensional case is not
38: destroyed by anisotropy but the corresponding stability of the
39: two-dimensional system can be corrupted by the presence of
40: anisotropy. A borderline dimension $d_c$ below which the stability
41: of the scaling regime is not present is calculated as a function of
42: anisotropy parameters.}}
43:
44:
45: \bigskip
46:
47: \begin{center}
48: \bf INTRODUCTION
49: \end{center}
50:
51: It is well known that the breakdown of the classical
52: Kolmogorov-Obuchov phenomenological theory of fully developed
53: turbulence \cite{Frisch} is more noticeable for simpler models of
54: passively advected scalar or vector quantity (scalar or vector
55: field) than for the velocity field itself. This phenomenon is
56: encoded in the terms intermittency and anomalous scaling
57: \cite{Frisch,MonYagBook}. At the same time, the problem of a passive
58: advection is easier from theoretical point of view (see, e.g.,
59: \cite{FaGaVe01} and references therein). Therefore it leads to the
60: fact that the problem of anomalous scaling can be understand here in
61: a few ways what is not possible for the present in the problem of
62: genuine turbulence.
63:
64: One of the most effective approach for studying self-similar scaling
65: behavior is the method of the field theoretic renormalization group
66: (RG) \cite{ZinnJustin,Vasiliev}. It can be also used in the theory
67: of fully developed turbulence and related problems
68: \cite{Vasiliev,AdAnVa96,AdAnVa99}, e.g., in the problem of a passive
69: scalar (or vector) field advected by a given stochastic
70: environment.
71:
72: In \cite{AdAnVa98} the field theoretic RG was applied to the
73: so-called rapid-change model of a passive scalar advected by a
74: self-similar white-in-time velocity field which is also known as
75: Kraichnan model. It was shown that within the field theoretic RG
76: approach the anomalous scaling is related to the existence of
77: "dangerous" composite operators with negative critical dimensions in
78: the framework of the operator product expansion (OPE)
79: \cite{Vasiliev,AdAnVa96,AdAnVa99}.
80:
81: Afterwards, various generalized descendants of the Kraichnan model,
82: namely, models with inclusion of large and small scale anisotropy,
83: %\cite{AdAnHnNo00},
84: compressibility,
85: %\cite{AdAn98}
86: and finite correlation time of the velocity field
87: %\cite{Antonov99,Antonov00}
88: were studied by the field theoretic approach (see \cite{Antonov06}
89: and references therein). Moreover, advection of a passive vector
90: field by the Gaussian self-similar velocity field (with and without
91: large and small scale anisotropy, pressure, compressibility, and
92: finite correlation time) has been also investigated and all possible
93: asymptotic scaling regimes and crossovers among them have been
94: classified \cite{all1,AdAnRu01,all2,AnHnHoJu03}. General conclusion
95: is: the anomalous scaling, which is the most important feature of
96: the Kraichnan rapid-change model, remains valid for all generalized
97: models.
98:
99: In what follows we shall begin with investigation of one particular
100: model of a passive vector advected by a Gaussian velocity field with
101: finite correlation time in the presence of the small-scale
102: anisotropy, namely the model where the stretching term is absent
103: (the so-called $A=0$ model, see, e.g, \cite{AdAnRu01,AnHnHoJu03}).
104: This model is specific from several points of view but maybe the
105: most important fact is that in contrast to the other models of
106: passive vector admixture where the anomalous scaling is related to
107: the composite operators built of the vector field without
108: derivatives \cite{all2,AnHnHoJu03} in the case under consideration
109: it is related to the composite operators built solely of the
110: gradients of the field. This fact radically changes the complexity
111: of the problem especially in the anisotropic case (see, e.g.,
112: \cite{AdAnRu01,Novikov06} and references therein). Thus, in some
113: sense, it can be consider as a further step to the nonlinear
114: Navier-Stokes equation. In what follows, we shall present only the
115: first part of the RG analysis, namely, we shall analyze the
116: influence of the small-scale anisotropy on the infrared (IR)
117: stability of the possible scaling regimes of the model. It will be
118: seen that complexity of this task is also very close to the
119: corresponding problem in the stochastic Navier-Stokes equation
120: \cite{all3}.
121:
122:
123: \bigskip
124:
125: \begin{center}
126: \bf 1. FIELD THEORETIC FORMULATION OF THE MODEL
127: \end{center}
128:
129: We shall consider the model of the advection of transverse
130: (solenoidal) passive vector field ${\bf b} \equiv {\bf b}({\bf
131: x},t)$ which is described by the following stochastic equation
132: \begin{equation}
133: \partial_t {\bf b} = \nu_0 \Delta {\bf b} - ({\bf v \cdot \nabla}) {\bf b} + {\bf f}, \label{K-K}
134: \end{equation}
135: where $\partial_t\equiv \partial/\partial t$, $\Delta \equiv{\bf
136: \nabla}^2$ is the Laplace operator, $\nu_0$ is the diffusivity (a
137: subscript $0$ denotes bare parameters of unrenormalized theory), and
138: ${\bf v} \equiv {\bf v} ({\bf x} ,t)$ is incompressible advecting
139: velocity field. The vector field ${\bf f} \equiv {\bf f} ({\bf x}
140: ,t)$ is a transverse Gaussian random (stirring) force with zero mean
141: and covariance
142: \begin{equation}
143: D_{ij}^f \equiv \langle f_i({\bf x},t) f_j({\bf
144: x^{\prime}},t^{\prime}) \rangle= \delta(t-t^{\prime})C_{ij}({\bf
145: r}/L), \,\,\,\,\ {\bf r}={\bf x}-{\bf x^{\prime}} \label{cor-b}
146: \end{equation}
147: where parentheses $\langle...\rangle$ hereafter denote average over
148: corresponding statistical ensemble. The noise defined in
149: Eq.\,(\ref{cor-b}) maintains the steady-state of the system but the
150: concrete form of the correlator will not be essential in what
151: follows. The only condition which must be fulfilled by the function
152: $C_{ij}({\bf r}/L)$ is that it must decrease rapidly for $r\equiv
153: |{\bf r}| \gg L$, where $L$ denotes an integral scale related to the
154: stirring.
155:
156: In real problems the velocity field ${\bf v}(x)$ satisfies
157: Navier-Stokes equation but, in what follows, we shall work with a
158: simplified model where we suppose that statistics of the velocity
159: field is given in the form of Gaussian distribution with zero mean
160: and pair correlation function
161: \begin{eqnarray}
162: &&\hspace{-1cm} \langle v_i(x) v_j(x^{\prime}) \rangle \equiv
163: D^v_{ij}(x; x^{\prime})= \int \frac{d^d {\bf k} d
164: \omega}{(2\pi)^{d+1}} R_{ij}({\bf k}) D^v(\omega,{\bf k})
165: e^{-i\omega(t-t^{\prime})+ i{\bf k}({\bf x}-{\bf x^{\prime}})},
166: \label{corv}
167: \end{eqnarray}
168: where $d$ is the dimension of the space, ${\bf k}$ is the wave
169: vector, and $R_{ij}({\bf k})$ is a transverse projector. In our
170: uniaxial anisotropic case it is taken as (see, e.g., \cite{all2} and
171: references therein)
172: \begin{equation}
173: R_{ij} ({\bf k}) =
174: %\langle v_i ({\bf x}, t) v_j (0,0) \rangle =
175: \left(1 + \alpha_{1} ({\bf n \cdot k})^2/k^2\right) P_{ij} ({\bf k})
176: + \alpha_{2} n_s n_l P_{is} ({\bf k}) P_{jl} ({\bf k})\,,
177: \label{T-ij}
178: \end{equation}
179: where $P_{ij} ({\bf k})\equiv \delta_{ij}-k_i k_j/k^2$ is common
180: isotropic transverse projector, the unit vector ${\bf n}$ determines
181: the distinguished direction, and $\alpha_{1}$, $\alpha_{2}$ are
182: parameters characterizing the anisotropy. From the positiveness of
183: the correlation tensor $D^v_{ij}$ one immediately finds restrictions
184: on the values of the above parameters, namely $\alpha_{1,2}>-1$. The
185: function $D^v(\omega, {\bf k})$ in (\ref{corv}) is taken in the
186: following form \cite{AnHnHoJu03}
187: \begin{equation}
188: D^v(\omega, k) = \frac{g_0 u_0 \nu_0^3
189: k^{4-d-2\varepsilon-\eta}}{(i\omega+u_0 \nu_0
190: k^{2-\eta})(-i\omega+u_0 \nu_0 k^{2-\eta})}, \label{corrvelo}
191: \end{equation}
192: where $g_{0}$ plays the role of the coupling constant of the model
193: (a formal small parameter of the ordinary perturbation theory), the
194: parameter $u_{0}$ gives the ratio of turnover time of scalar field
195: and velocity correlation time, and the positive exponents
196: $\varepsilon$ and $\eta$ are small RG expansion parameters. The
197: coupling constant $g_{0}$ and the exponent $\varepsilon$ control the
198: behavior of the equal-time pair correlation function of velocity
199: field and the parameter $u_{0}$ together with the second exponent
200: $\eta$ are related to the frequency $\omega\simeq
201: u_{0}\nu_{0}k^{2-\eta}$ which characterizes the mode $k$. The value
202: $\varepsilon=4/3$ corresponds to the celebrated Kolmogorov
203: \char`\"{}two-thirds law\char`\"{} for the spatial statistics of
204: velocity field, and $\eta=4/3$ corresponds to the Kolmogorov
205: frequency. Simple dimensional analysis shows that $g_{0}$ and
206: $u_{0}$, which we commonly term as charges, are related to the
207: characteristic ultraviolet (UV) momentum scale $\Lambda$ (or inner
208: legth $l\sim\Lambda^{-1}$) by \begin{equation}
209: g_{0}\simeq\Lambda^{2\varepsilon},\qquad
210: u_{0}\simeq\Lambda^{\eta}.\end{equation}
211:
212:
213:
214:
215: The stochastic problem (\ref{K-K})-(\ref{corv}) can be treated as a
216: field theory with action functional \cite{ZinnJustin,Vasiliev}
217: \begin{eqnarray}
218: S(\Phi)&=&
219: %\int dt\,d^d{\bf x}\,\, \nonumber \\ &&
220: %\hspace{-0.7cm}
221: b_j^{\prime} \left[ \left(-\partial_t - v_i\partial_i+\nu_0\Delta +
222: \nu_0 \chi_{10} ({\bf n}\cdot{\bf
223: \partial})^2 \right)\delta_{jk} + n_j \, \nu_0 \left(\chi_{20} \Delta +
224: \chi_{30} ({\bf n}\cdot{\bf
225: \partial})^2 \right) n_k\right] b_k
226: \nonumber
227: \\ && - \frac{1}{2}
228: %\int dt_1\,d^d{\bf x_1}\,dt_2\,d^d{\bf x_2}
229: %\\ && \hspace{-0.7cm}
230: \left( v_i [D_{ij}^v]^{-1} v_j - b_i^{\prime} D^f_{ij} b_j^{\prime}
231: \right), \label{action2}
232: \end{eqnarray}
233: where $D_{ij}^v$ and $D^f_{ij}$ are given in (\ref{corv}) and
234: (\ref{cor-b}) respectively, ${\bf b}^{\prime}$ is an auxiliary
235: vector field (see, e.g., \cite{Vasiliev}), and the required
236: integrations over $x=({\bf x}, t)$ and summations over the vector
237: indices are implied. In action (\ref{action2}) the terms with new
238: parameters $\chi_{10},\chi_{20}$, and $\chi_{30}$ are related to the
239: presence of small-scale anisotropy and they are necessary to make
240: the model multiplicatively renormalizable. Model (\ref{action2})
241: corresponds to a standard Feynman diagrammatic technique (see, e.g.,
242: \cite{all2,all3} for details) and the standard analysis of canonical
243: dimensions then shows which one-irreducible Green functions can
244: possess UV superficial divergences.
245:
246: The functional formulation (\ref{action2}) gives possibility to use
247: the field-theoretic methods, including the RG technique to solve the
248: problem. By means of the RG approach it is possible to extract
249: large-scale asymptotic behavior of the correlation functions after
250: an appropriate renormalization procedure which is needed to remove
251: UV-divergences.
252:
253:
254: \bigskip
255: \newpage
256:
257: \begin{center}
258: \bf 2. SCALING REGIMES OF THE MODEL
259: \end{center}
260:
261: \begin{wrapfigure}[23]{l}[0cm]{7.2cm}
262: \vspace{-1.5cm}
263: \includegraphics[width=6cm]{fig1.eps}
264: \vspace{-0.8cm} \caption{\small The scaling regimes of the model in
265: the $\varepsilon-\eta$ plane. The region FPI corresponds to the
266: trivial rapid-change limit: $g^*/u^*=0, 1/u^*=0$. The region FPII
267: corresponds to the the non-trivial rapid change limit: $g^*/u^*>0,
268: 1/u^*=0$. The region FPIII corresponds the trivial "frozen" limit:
269: $g^*=0, u^*=0$. The region FPIV is related to the non-trivial
270: "frozen" limit: $g^*>0, u^*=0$. In the end, the line
271: $\varepsilon=\eta$ (FPV) corresponds to the more interesting scaling
272: regime with $g^*>0, 0<u^*<\infty$.}\label{fig1}
273: \end{wrapfigure}
274:
275: Details of the one-loop field-theoretic RG analysis of the model
276: will be done elsewhere. Here we only conclude that using the RG
277: analysis leads to the following result: possible scaling regimes are
278: given by the IR stable fixed points of the system of five nonlinear
279: RG differential equations (flow equations or also known as
280: Gell-Mann-Low equations) for five scale dependent effective
281: variables (charges)
282: $\bar{C}=\{\bar{g},\bar{u},\bar{\chi}_1,\bar{\chi}_2,\bar{\chi}_3
283: \}$ of the model which are functions of the dimensionless scale
284: parameter $t=k/\Lambda$ \cite{Vasiliev,all3}. The system of the flow
285: equations is defined by the so-called $\beta$-functions of the
286: model (they are functions of the charges, anisotropy parameters,
287: space dimension, and parameters $\varepsilon, \eta$) and it has the
288: following form
289: \begin{eqnarray}
290: t\frac{d {\bar g}}{d t}&=&\beta_g= \bar{g}
291: (-2\varepsilon+2\gamma_{1})\,,
292: \label{betagg}\\
293: t\frac{d {\bar u}}{d t}&=&\beta_u= \bar{u}(-\eta+\gamma_1)\,, \label{betauuu}\\
294: t\frac{d {\bar \chi_i}}{d t}&=&\beta_{\chi_i}= \bar{\chi}_i
295: (\gamma_1-\gamma_{i+1})\,, \label{betahh} \\
296: %\,\,\,\,\,
297: && i=1,2,3\,,\nonumber
298: \end{eqnarray}
299: where the functions $\gamma_{i},i=1,2,3,4$ are given by the
300: following expressions (the one-loop approximation)
301: \begin{eqnarray}
302: \gamma_1&=&-g \frac{S_{d-1}}{(2\pi)^d}\frac{1}{2(d-1)(d+1)}
303: \int_{-1}^1 dx \frac{(1-x^2)^{(d-3)/2}}{w_1 w_2} K_1\,, \label{gammas}\\
304: \gamma_{i+1}&=&-\frac{g}{\chi_i}
305: \frac{S_{d-1}}{(2\pi)^d}\frac{1}{2(d-1)(d+1)} \int_{-1}^1 dx
306: \frac{(1-x^2)^{(d-3)/2}}{w_1 w_2} K_{i+1}\,,\quad i=1,2,3\,
307: \end{eqnarray}
308: where $S_d=2\pi^{d/2}/\Gamma(d/2)$ is the surface of the $d$
309: dimensional sphere, $w_1 = (1 + u + \chi_1 x^2), w_2 = (1 + u +
310: \chi_1 x^2 + (\chi_2 + \chi_3 x^2)(1 - x^2))$, and the coefficients
311: $K_{i}(x^2,g,u,\chi_1,\chi_2,\chi_3,\alpha_1,\alpha_2,d),i=1,2,3,4$
312: are some huge polynomials in respect to all variables and their
313: explicit form will be given elsewhere. The scale parameter $t$
314: belongs to the interval $0\leq t \leq 1$ with the initial conditions
315: given at $t=1$ and the IR stable fixed point corresponds to the
316: limit $t\rightarrow 0$, i.e., $\bar{C}|_{t=0}=C^*$.
317:
318:
319: \input epsf
320: \begin{figure}[t]
321: \vspace{-1.5cm}
322: \begin{flushleft}
323: \leavevmode
324: \epsfxsize=6.2cm
325: \epsffile{fig2.eps}
326: \end{flushleft}
327: \vspace{-9.65cm}
328: \begin{flushright}
329: \leavevmode
330: \epsfxsize=6.2cm
331: \epsffile{fig3.eps}
332: \end{flushright}
333: \vspace{-1.5cm} \caption{\small Dependence of the borderline
334: dimension $d_c$ on the parameters $\alpha_1$ and $\alpha_2$ for
335: different values of $u=u^*$. The corresponding scaling regime is
336: stable above the given curve. \label{fig2}}
337: \end{figure}
338:
339:
340: We have performed a numerical analysis of this system of
341: differential equations and our aim was twofold. First of all, we
342: have found all possible scaling regimes and we have analyzed the
343: regions of their IR stability in the $\varepsilon-\eta$ plane. The
344: results of this analysis are shown in Fig.\,\ref{fig1}, where it is
345: shown that the model exhibits five different scaling regimes (two
346: for rapid-change limit, two for so-called "frozen" limit, and one
347: general with nonzero $u_*$) (see, e.g., \cite{AnHnHoJu03} and
348: references therein). The second question which was investigated is
349: related to the dependence of stability of the above mentioned
350: scaling regimes on the anisotropy parameters $\alpha_1, \alpha_2$
351: and on the dimension of the space $d$. We have found the so-called
352: borderline dimension $d_c$ between stable and unstable regimes as a
353: function of anisotropy parameters $\alpha_1, \alpha_2$ and parameter
354: $u_*$. The results are shown in Fig.\,\ref{fig2} for some special
355: situations. One can see that the presence of small-scale anisotropy
356: leads to the violation of the stability of the corresponding scaling
357: regimes below $d_c\in[2,3]$ for appropriate values of anisotropy
358: parameters. But from the point of view of further investigation of
359: anomalous scaling the most important conclusion is that all the
360: three-dimensional scaling regimes remain stable under influence of
361: small-scale uniaxial anisotropy.
362:
363:
364: \bigskip
365:
366: \begin{center}
367: \bf CONCLUSIONS
368: \end{center}
369:
370: Using the field theoretic RG we have studied the influence of
371: small-scale uniaxial anisotropy on the stability of the scaling
372: regimes in the model of a passive vector advected by given
373: stochastic environment with finite time correlations. It is shown
374: that the system exhibits five different scaling regimes. They are
375: related to the values of the parameters $\varepsilon$ and $\eta$. On
376: the other hand, the stability of all these scaling regimes are
377: influenced by presence of small-scale anisotropy which is
378: demonstrated in the existence of the so-called borderline dimension
379: $d_c$ which is a function of the anisotropy parameters. The $d_c$ is
380: defined as dimension above which the corresponding scaling regime is
381: stable and below which the stability of the regime is destroyed.
382: All calculations have been done at the one-loop level. The results will be used
383: in the further investigations of the anomalous scaling of the model. \\
384:
385:
386:
387:
388: E.J. is thankful to J. Bu\v{s}a for discussion. The work was
389: supported in part by VEGA grant 6193 of Slovak Academy of Sciences
390: and by Science and Technology Assistance Agency under contract No.
391: APVT-51-027904.
392:
393: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
394:
395: \def\refname{\centerline{\normalsize REFERENCES}}
396: \begin{thebibliography}{50}
397: %\begin{references}
398: %\begin{sloppypar}
399: %\small
400: \bibitem{Frisch}
401: U.~Frisch, {\it Turbulence: The Legacy of A.N. Kolmogorov}
402: (Cambridge University Press, Cambridge, 1995).
403: \bibitem{MonYagBook}
404: A.~S.~Monin, A.~M~Yaglom, {\it Statistical Fluid Mechanics} (MIT
405: Press, Cambridge, MA, 1975), Vol. 2.
406: \bibitem{FaGaVe01} G. Falkovich, K. Gaw\c{e}dzki, M. Vergassola,
407: Rev. Mod. Phys. {\bf 73}, 913 (2001).
408: \bibitem{ZinnJustin}
409: J.~Zinn-Justin, {\it Quantum Field Theory and Critical Phenomena}
410: (Clarendon, Oxford, 1989).
411: \bibitem{Vasiliev}
412: A.~N.~Vasil'ev, {\it Quantum-Field Renormalization Group in the
413: Theory of Critical Phenomena and Stochastic Dynamics} (St.
414: Petersburg Institute of Nuclear Physics, St. Petersburg, 1998) [in
415: Russian; English translation: Gordon \& Breach, 2004].
416: \bibitem{AdAnVa96}
417: L.~Ts.~Adzhemyan, N.~V.~Antonov, and A.~N.~Vasil'ev, Usp. Fiz. Nauk
418: {\bf 166}, 1257 (1996) [Phys. Usp. {\bf 39}, 1193 (1996)].
419: \bibitem{AdAnVa99}
420: L.~Ts.~Adzhemyan, N.~V.~Antonov, and A.~N.~Vasil'ev, {\it The Field
421: Theoretic Renormalization Group in Fully Developed Turbulence}
422: (Gordon $\&$ Breach, London, 1999).
423: \bibitem{AdAnVa98}
424: L.~Ts.~Adzhemyan, N.~V.~Antonov, and A.~N.~Vasil'ev, Phys. Rev. E
425: {\bf 58}, 1823 (1998).
426: %\bibitem{AdAnHnNo00}
427: %L. Ts. Adzhemyan, N. V. Antonov, M. Hnatich, and S. V. Novikov,
428: %Phys. Rev. E {\bf 63}, 016309 (2000).
429: %\bibitem{AdAn98}
430: %L.~Ts.~Adzhemyan, and N.~V.~Antonov, Phys. Rev. E {\bf 58}, 7381
431: %(1998); N.~V.~Antonov, and J.~Honkonen, {\it ibid.} {\bf 63}, 036302
432: %(2001).
433: %\bibitem{Antonov99}
434: %N.~V.~Antonov, Phys. Rev. E {\bf 60}, 6691 (1999)
435: %\bibitem{Antonov00}
436: %N.~V.~Antonov, Physica D {\bf 144}, 370 (2000); Zap. Nauchn. Semin.
437: %POMI {\bf 269}, 79 (2000).
438: \bibitem{Antonov06}
439: N.V. Antonov, J. Phys. A: Math. Gen. {\bf 39}, 7825 (2006).
440:
441: \bibitem{all1}
442: N. V. Antonov, A. Lanotte, and A. Mazzino, Phys. Rev. E {\bf 61},
443: 6586 (2000); N. V. Antonov, J. Honkonen, A. Mazzino, and P.
444: Muratore-Ginanneschi, {\it ibid.} {\bf 62}, R5891 (2000).
445: \bibitem{AdAnRu01}
446: L. Ts. Adzhemyan, N. V. Antonov, and A. V. Runov, Phys. Rev. E {\bf
447: 64}, 046310 (2001).
448: \bibitem{all2}
449: M. Hnatic, M. Jurcisin, A. Mazzino, S. Sprinc, acta phys. slov. {\bf
450: 52}, 559 (2002); M. Hnatic, J. Honkonen, M. Jurcisin, A. Mazzino,
451: and S. Sprinc, Phys. Rev. E {\bf 71}, 066312 (2005).
452: \bibitem{AnHnHoJu03}
453: N.V. Antonov, M. Hnatic, J. Honkonen, and M. Jurcisin, Phys. Rev. E
454: {\bf 68}, 046306 (2003);
455: \bibitem{Novikov06}
456: S.V. Novikov, J. Phys. A: Math. Gen. {\bf 39}, 8133 (2006).
457: \bibitem{all3}
458: J. Bu\v{s}a, M. Hnatich, J. Honkonen, and D. Horvath, Phys. Rev. E
459: {\bf 55} 381 (1997); J. Bu\v{s}a, M. Hnatich, E.
460: Jur\v{c}i\v{s}inov\'a, M. Jur\v{c}i\v{s}in, M. Stehlik, acta phys.
461: slov. {\bf 52}, 547 (2002); E.A. Hayryan, E. Jurcisinova, M.
462: Jurcisin, I. Pokorny, M. Stehlik, Communication of JINR
463: E17-2005-208.
464:
465:
466: \end{thebibliography}
467:
468: \end{document}
469:
470: \begin{wrapfigure}[23]{l}[0cm]{7.2cm}
471: \includegraphics[width=7cm]{te.eps}
472: \caption{\small $T_e(r,t)$ distribution for ion $^{129}$Xe at 2.6
473: MeV/amu in YBa$_2$Cu$_3$O$_{7-x}$. At the moment of electron
474: subsystem relaxation to the thermodynamic equilibrium, $t \simeq
475: 10^{-15}$ s, temperature of electrons in the center of track is
476: about $10^5$ K.}\label{Te}
477: \end{wrapfigure}
478:
479:
480: \begin{wrapfigure}[23]{l}[0cm]{7.2cm}
481: \vspace{-1.5cm}
482: \includegraphics[width=6cm]{fig2.eps}
483: \vspace{-1cm} \caption{\small Dependence of the borderline dimension
484: $d_c$ as a function of the $\alpha_1$ for $\alpha_2=0$ and for
485: different values of $u=u^*$. The corresponding scaling regime is
486: stable above the given curve.}\label{fig2}
487: \end{wrapfigure}
488:
489: \begin{wrapfigure}[23]{l}[0cm]{7.2cm}
490: \vspace{-1.5cm}
491: \includegraphics[width=6cm]{fig3.eps}
492: \vspace{-1cm} \caption{\small Dependence of the borderline dimension
493: $d_c$ as a function of the $\alpha_2$ for $\alpha_1=0$ and for
494: different values of $u=u^*$. The corresponding scaling regime is
495: stable above the given curve.}\label{fig3}
496: \end{wrapfigure}
497: