1: %% The command below calls the preprint style
2: %% which will produce a one-column, single-spaced document.
3: \documentclass[12pt,preprint]{aastex}
4:
5: %% manuscript produces a one-column, double-spaced document:
6: %%\documentclass[manuscript]{aastex}
7:
8: %% preprint2 produces a double-column, single-spaced document:
9: %%\documentclass[preprint2]{aastex}
10:
11: \shorttitle{Temperature Anisotropy in a Shocked Plasma}
12: \shortauthors{Liu et al.}
13:
14: \begin{document}
15:
16: \title{Temperature Anisotropy in a Shocked Plasma: Mirror-Mode
17: Instabilities in the Heliosheath}
18:
19: \author{Y. Liu\altaffilmark{1,2}, J. D. Richardson\altaffilmark{1,2},
20: J. W. Belcher\altaffilmark{1}, and J. C. Kasper\altaffilmark{1}}
21:
22: \altaffiltext{1}{Kavli Institute for Astrophysics and Space Research,
23: Massachusetts Institute of Technology, Cambridge, MA 02139, USA;
24: liuxying@mit.edu.}
25:
26: \altaffiltext{2}{State Key Laboratory of Space Weather, Chinese
27: Academy of Sciences, Beijing 100080, China.}
28:
29: \begin{abstract}
30: We show that temperature anisotropies induced at a shock can account
31: for interplanetary and planetary bow shock observations. Shocked
32: plasma with enhanced plasma beta is preferentially unstable to the
33: mirror mode instability downstream of a quasi-perpendicular shock and
34: to the firehose instability downstream of a quasi-parallel shock,
35: consistent with magnetic fluctuations observed downstream of a large
36: variety of shocks. Our theoretical analysis of the solar wind
37: termination shock suggests that the magnetic holes observed by
38: Voyager 1 in the heliosheath are produced by the mirror mode
39: instability. The results are also of astrophysical interest,
40: providing an energy source for plasma heating.
41: \end{abstract}
42:
43: \keywords{instabilities --- shock waves --- solar wind}
44:
45: \section{Introduction}
46: Planetary bow shocks and interplanetary shocks serve as a unique
47: laboratory for the study of shock waves in collisionless plasmas.
48: Observations of these shocks usually show that ion distributions are
49: anisotropic with respect to the background magnetic field downstream
50: of the shocks. Mirror mode waves associated with this anisotropy are
51: observed downstream of quasi-perpendicular shocks (defined by the
52: angle between the shock normal and the upstream magnetic field
53: $\theta_{Bn}> 45^{\circ}$) when the plasma beta is high ($\beta>1$)
54: \citep{kaufmann70, tsurutani92, violante95, bavassano98, liu06a}.
55: Mirror mode waves do not grow in low beta regions where the ion
56: cyclotron mode dominates \citep[e.g.,][]{anderson94, czaykowska01}.
57: Magnetic fluctuations downstream of quasi-parallel shocks
58: ($\theta_{Bn}< 45^{\circ}$) have not been identified in detail, but
59: hybrid simulations show that the firehose instability can occur
60: downstream of these shocks for certain ranges of upstream Alfv\'{e}n
61: Mach number ($M_{\rm A}$) and plasma beta, for instance, $M_{\rm
62: A}\geq 3$ at $\beta\sim 0.5$ \citep{kan83, krauss91}; observations
63: seem to confirm this point \citep[e.g.,][]{greenstadt79,
64: bavassano00}. Quasi-perpendicular shocks are characterized by a sharp
65: increase in the magnetic field strength, but quasi-parallel shocks
66: are often more turbulent, with shock ramps containing large-amplitude
67: waves which spread upstream and downstream.
68:
69: The recent crossing of the termination shock (TS) by Voyager 1 (V1)
70: \citep{burlaga05, decker05, gurnett05, stone05} provides an
71: opportunity to study shocks and shock-induced waves in the
72: heliosheath. The sharp increase in the magnetic field strength across
73: the TS and the downstream magnetic field configuration suggest that
74: the TS is quasi-perpendicular \citep{burlaga05}. As in planetary
75: magnetosheaths downstream of a quasi-perpendicular bow shock, the
76: heliosheath shows compressive magnetic fluctuations in the form of
77: magnetic holes \citep{burlaga06a, burlaga06b}.
78:
79: We propose a theoretical explanation for the temperature anisotropies
80: and associated instabilities induced at a shock. Based on the
81: theoretical analysis, we show that the magnetic holes observed in the
82: heliosheath could be mirror mode fluctuations. The present results
83: also provide a prototype for understanding shocks in various
84: astrophysical contexts, such as gamma-ray bursts, supernova
85: explosions and active galactic nuclei.
86:
87: \section{Theory}
88: An anisotropic ion distribution requires the use of a pressure tensor
89: in the magnetohydrodynamic (MHD) equations. Observations show that
90: the shock structure and dynamics depend on the shock geometry and
91: upstream $M_{\rm A1}$ and $\beta_1$, so we examine the temperature
92: anisotropy $A_2=T_{\perp 2}/T_{\parallel 2}$ as a function of these
93: parameters, where $T_{\perp}$ and $T_{\parallel}$ are the plasma
94: temperature perpendicular and parallel to the magnetic field. The
95: subscripts 1 and 2 indicate physical parameters upstream and
96: downstream of a shock. Assuming a bi-Maxwellian plasma, we write the
97: jump conditions across a shock as \citep{hudson70}
98: \begin{equation}
99: \left[ B_n \right] = 0,
100: \end{equation}
101: \begin{equation}
102: \left[ \rho v_n \right] = 0,
103: \end{equation}
104: \begin{equation}
105: \left[ v_n \textbf{B}_t - \textbf{v}_t B_n \right] = 0,
106: \end{equation}
107: \begin{equation}
108: \left[ P_{\perp} + (P_{\parallel}-P_{\perp})\frac{B_n^2}{B^2} +
109: \frac{B_t^2 - B_n^2}{2\mu_0} + \rho v_n^2 \right] = 0,
110: \end{equation}
111: \begin{equation}
112: \left[\frac{B_n\textbf{B}_t}{\mu_0}\left(\frac{P_{\parallel}-
113: P_{\perp}}{B^2/\mu_0}-1\right) + \rho v_n\textbf{v}_t \right] = 0,
114: \end{equation}
115: \begin{equation}
116: \left[\rho v_n\left(\frac{2P_{\perp}}{\rho} +
117: \frac{P_{\parallel}}{2\rho} + \frac{v^2}{2} + \frac{B^2_t}{\mu_0\rho}
118: \right) + \frac{B_n^2v_n}{B^2}(P_{\parallel} - P_{\perp}) -
119: \frac{(\textbf{B}_t\cdot\textbf{v}_t)B_n}{\mu_0}
120: \left(1-\frac{P_{\parallel}-P_{\perp}}{B^2/\mu_0} \right) \right]=0,
121: \end{equation}
122: where $\mu_0$ is the permeability of vacuum, $\rho$ is the plasma
123: density, and \textbf{v} and \textbf{B} are the plasma velocity and
124: magnetic field with subscripts $t$ and $n$ denoting the tangential
125: and normal components with respect to the shock surface. The velocity
126: is measured in the shock frame. The square brackets indicate the
127: difference between the preshock (1) and postshock (2) states. The
128: perpendicular and parallel plasma pressures are defined as $P_{\perp}
129: = \rho k_{\rm B}T_{\perp}/m_p$ and $P_{\parallel} = \rho k_{\rm
130: B}T_{\parallel}/m_p$, where $k_{\rm B}$ and $m_p$ represent the
131: Boltzmann constant and proton mass, respectively. For simplicity, we
132: assume that the bulk velocity is parallel to the shock normal. The
133: components of the magnetic field upstream of the shock are given by
134: $B_{n1} = B_1 \cos\theta_{Bn}$ and $B_{t1} = B_1 \sin\theta_{Bn}$.
135: Different forms of the solutions to these equations have been derived
136: based on various assumptions \citep{chao95, erkaev00}. The focus of
137: the present analysis is to investigate under what conditions the
138: shocked plasma is unstable to the thermal anisotropy instabilities.
139:
140: The thermal anisotropy serves a free energy source which can feed
141: magnetic fluctuations when it exceeds certain threshold conditions.
142: The thresholds can be expressed as $A = 1 - 2/\beta_{\parallel}$ for
143: the firehose instability \citep{parker58}, the lower bound of the
144: temperature anisotropy, and $A = 1 + 1/\beta_{\perp}$ for the mirror
145: mode instability \citep{chan58, hasegawa69}, the upper bound. The ion
146: cyclotron instability, competing with the mirror mode, has the onset
147: condition $A = 1+ 0.64/\beta_{\parallel}^{0.41}$ \citep{gary97}. Here
148: the perpendicular and parallel plasma betas are defined as
149: $\beta_{\perp} = P_{\perp}/(B^2/2\mu_0)$ and $\beta_{\parallel} =
150: P_{\parallel}/(B^2/2\mu_0)$.
151:
152: For a perpendicular shock, equations~(2) and (3) give the shock
153: strength
154: \begin{equation}
155: r_s = \frac{\rho_2}{\rho_1} = \frac{v_1}{v_2} = \frac{B_2}{B_1}.
156: \end{equation}
157: The temperature anisotropy $A_2$ can be obtained from equations~(4),
158: (5) and (6),
159: \begin{equation}
160: A_2 = \frac{\frac{3A_1\beta_1}{2A_1 + 1} + 2M_{A1}^2(1-1/r_s) + 1 -
161: r_s^2}{\frac{3A_1 \beta_1}{2A_1 + 1}(4r_s - 4 + r_s/A_1) +
162: 2M_{A1}^2(r_s + 3/r_s -4) + 4(r_s -1)},
163: \end{equation}
164: where $\beta = (2P_{\perp} + P_{\parallel})/(3B^2/2\mu_0)$. To
165: compare with the thresholds, we derive $\beta_{\perp 2}$ from
166: equation~(4) as
167: \begin{equation}
168: \beta_{\perp 2} = \frac{3A_1\beta_1}{(2A_1 + 1)r_s^2} +
169: \frac{2M_{A1}^2}{r_s^2}(1-1/r_s) - 1 + \frac{1}{r_s^2}.
170: \end{equation}
171: For $2M_{A1}^2 \gg \beta_1$, we have $\beta_{\perp 2} \sim
172: 2M_{A1}^2/r_s^2$, so the shocked plasma would have a high beta
173: independent of $\beta_1$, consistent with the findings in the
174: magnetosheaths of outer planets \citep{russell90}. As can be seen
175: from the mirror-mode threshold, high values of $\beta_{\perp 2}$
176: favor the onset of the mirror mode instability; similarly, $A_2 \sim
177: 1/(r_s-3)$ under the same condition, so $A_2 \geq 1$ since the shock
178: strength cannot be larger than 4, also favoring the mirror mode
179: onset.
180:
181: Figure~1 shows the temperature anisotropy $A_2$ for $r_s=3$ and
182: $A_1=1$ as a function of $M_{A1}$ and $\beta_1$. Only values of
183: $0\leq A_2 \leq 2$ are shown. The entropy, $S = \frac{1}{2}k_{\rm
184: B}\ln(P_{\perp}^2P_{\parallel}/\rho^5)$, is required to increase
185: across a shock by the second law of thermodynamics; the region which
186: breaks the entropy principle is show as ``Forbidden'' in Figure~1.
187: The majority of the allowed area has $A_2 \geq 1$ as expected and is
188: preferentially unstable to the mirror mode instability. The mirror
189: mode has a lower threshold than the cyclotron mode in this plasma
190: regime, so the temperature anisotropy would be quickly reduced by the
191: mirror mode before the ion cyclotron mode could develop. The TS with
192: $\beta_1\simeq 32.8$ and $M_{A1}\simeq 16.3$ (see \S 3), indicated by
193: the plus sign, is located slightly above the mirror mode threshold.
194: Interplanetary and planetary bow shocks have a large variation in
195: $M_{A1}$ and $\beta_1$; many of them would also be above the mirror
196: mode threshold as indicated by the large area with $A_2 \geq 1$.
197: Consequently, mirror mode instabilities should be a frequent feature
198: downstream of quasi-perpendicular shocks.
199:
200: For a parallel shock, the magnetic field does not go into the
201: expression of the shock strength. The temperature anisotropy
202: \begin{equation}
203: A_2 = \frac{\frac{3\beta_1 r_s}{2A_1 + 1}(A_1 + \frac{3}{2} -
204: \frac{3}{2r_s}) + M_{A1}^2(r_s + 2/r_s -3)}{\frac{3\beta_1}{2A_1 + 1}
205: + 2M_{A1}^2(1-1/r_s)},
206: \end{equation}
207: and the downstream parallel plasma beta
208: \begin{equation}
209: \beta_{\parallel 2} = \frac{3\beta_1}{2A_1+1} + 2M_{A1}^2(1-1/r_s).
210: \end{equation}
211: For $2M_{A1}^2 \gg \beta_1$, $\beta_{\parallel 2} \sim 2M_{A1}^2$,
212: which makes the thresholds close to 1, and $A_2 \sim (r_s - 2)/2 \leq
213: 1$, giving favorable conditions for the onset of the firehose
214: instability. Figure~2 displays the temperature anisotropy $A_2$ over
215: various upstream conditions for $r_s =3$ and $A_1=1$. Compared with
216: Figure~1, higher values of $M_{A1}$ at a given $\beta_1$ would
217: otherwise lead to smaller $A_2$. Most of the area shown has $A_2\leq
218: 1$ as expected for a quasi-parallel shock and is unstable to the
219: firehose instability. The TS would induce firehose instabilities in
220: the downstream plasma if it were quasi-parallel, as indicated by its
221: location in the plot. Many quasi-parallel interplanetary and
222: planetary bow shocks will also give rise to firehose instabilities as
223: implied by the large area with $A_2\leq 1$.
224:
225: Observations show that a quasi-parallel bow shock becomes unsteady as
226: the Alfv\'{e}n Mach number $M_{\rm A1}$ exceeds $\sim 3$ for $\beta_1
227: \sim 0.5$ \citep{greenstadt79} and is often associated with large
228: transverse magnetic fluctuations \citep{bavassano00, czaykowska01}. A
229: closer look at Figure~2 gives $A_2\simeq 0.97$ at $M_{A1} = 2$ for
230: $\beta_1=0.5$, a noticeable thermal anisotropy but not large enough
231: to exceed the firehose onset; at $M_{A1} = 3$, the thermal anisotropy
232: rises above the firehose threshold, generating firehose instabilities
233: which will disturb the shock structure. The firehose instability
234: arises from the fast MHD mode and produces large-amplitude transverse
235: waves. The results from this simple calculation agree with hybrid
236: simulations \citep{kan83, krauss91}.
237:
238: \section{Magnetic Fluctuations in the Heliosheath}
239: V1 crossed the TS on 2004 December 16 (day 351) at 94 AU and is
240: making the first measurements in the heliosheath. The magnetic
241: fluctuations in the heliosheath are characterized by a series of
242: depressions in the field magnitude which have been called magnetic
243: holes \citep{burlaga06a, burlaga06b}. These fluctuations are similar
244: to those observed downstream of quasi-perpendicular interplanetary
245: and planetary bow shocks that have been identified as mirror mode
246: structures \citep{kaufmann70, tsurutani92, violante95, bavassano98}.
247:
248: The TS is expected to be highly oblique with $\theta_{Bn} \sim
249: 86^{\circ}$ at V1, so we use equations~(8) and (9) to determine
250: whether mirror mode instabilities occur in the heliosheath. MHD
251: simulations give the average preshock plasma density $n_1\simeq
252: 8\times 10^{-4}$ cm$^{-3}$, speed $v_1\simeq 380$ km s$^{-1}$, and
253: temperature $T_1\simeq 5.4\times 10^5$ K \citep{whang04}, which
254: yields $M_{A1}\simeq 16.3$, $\beta_1\simeq 32.8$ combined with the
255: observed field strength $B_1\simeq 0.03$ nT. The shock strength $r_s$
256: is $\sim 3$ estimated from the jump in the field magnitude
257: \citep{burlaga05} and from the spectral slope of TS accelerated
258: particles \citep{stone05}. The typical thermal anisotropy of the
259: solar wind is $A \sim 0.7$ near the Earth \citep{liu06b}. Expansion
260: of the solar wind would further decrease the value of $A$ if the
261: magnetic moment $\mu \sim T_{\perp}/B$ is invariant; when the thermal
262: anisotropy exceeds the firehose threshold, firehose instabilities
263: will be induced and help suppress the growth of the anisotropy. The
264: two competing processes will arrive at a balance close to the
265: threshold, i.e., $A\simeq 1- 2/\beta_{\parallel}$, which gives
266: $A_1\simeq 0.94$. Introduction of pickup ions in the outer
267: heliosphere would not significantly change this value. The newly-born
268: pickup ions gyrate about the interplanetary magnetic field, forming a
269: ring-beam distribution; this configuration is unstable to the
270: generation of MHD waves, which scatter the ions quickly toward
271: isotropy \citep{lee87, bogdan91}. We have also shown that $A_2$ does
272: not depend on $A_1$ when $2M_{A1}^2 \gg \beta_1$. Substituting the
273: values of $M_{A1}$, $\beta_1$, $r_s$ and $A_1$ into equations (8) and
274: (9) gives $A_2 \simeq 1.03$ and $\beta_{\perp 2}\simeq 42.2$. The
275: threshold of the mirror mode is $1+1/\beta_{\perp 2} \simeq 1.02$,
276: smaller than the downstream anisotropy. As indicated by Figure~1,
277: even an $A_2$ slightly larger than 1 can meet the mirror mode onset
278: at high $\beta_1$, so mirror mode instabilities should be induced in
279: the heliosheath.
280:
281: For the upstream temperature, we have used the density-weighted
282: average of the pickup ion, solar wind proton and electron
283: temperatures \citep{whang04}. However, the result can be shown to be
284: self-consistent. Equation~9 is reduced to $\beta_2 \sim
285: \frac{2M_{A1}^2}{r_s^2}(1-1/r_s)$ in the TS case, from which we have
286: the downstream temperature $T_2 \sim 2\times 10^6$ K given the
287: average field magnitude $B_2 \simeq 0.09$ nT and density $n_2 \simeq
288: 0.002$ cm$^{-3}$ from equation~7. Approximating the TS as a
289: hydrodynamic shock because of the high plasma beta,
290: $$\frac{T_2}{T_1} = \frac{[2\gamma M^2-(\gamma-1)][(\gamma-1)M^2+2]}
291: {(\gamma+1)^2M^2},$$ we obtain $T_1 \sim 5\times 10^5$ K, consistent
292: with the MHD simulation result. Here $\gamma = 5/3$, and $M\simeq 3$,
293: the shock Mach number estimated from $n_2/n_1 \simeq 3$.
294:
295: Given the absence of plasma measurements across the TS, it is hard to
296: estimate the uncertainties brought about by the upstream conditions
297: and the shock parameters. A revisit to equations~(1)-(6) assuming a
298: 10\% error in the upstream conditions and the shock geometry and
299: strength gives $A_2 = 1.03\pm 0.42$ and a mirror mode threshold
300: $1.02\pm 0.01$. Note that the uncertainty of $A_2$ is determined
301: largely by the shock strength, since $A_2 \sim 1/(r_s-3)$ in the case
302: of the TS. If we use $r_s = 2.6^{+0.4}_{-0.2}$ inferred from the
303: particle spectra \citep{stone05} with other parameters fixed, then
304: the temperature anisotropy $A_2 = 1.03$ - $4.19$; the mirror mode
305: would be more likely to occur. Interestingly, the shock strength
306: cannot be smaller than 2.2 at the present conditions; otherwise $A_2$
307: would be negative. It should be emphasized that turbulence induced by
308: the mirror mode instability would leave the threshold marginally
309: satisfied, so we expect $A_2 = 1.02\pm 0.01$ in the evolved state.
310:
311: The mirror mode instability has maximum growth rate at a wave vector
312: highly oblique to the background magnetic field. The minimum variance
313: direction of the measured magnetic fields has an angle of about
314: 75$^{\circ}$ with respect to the background field when magnetic holes
315: are present \citep{burlaga06b}. Mirror mode waves are non-propagating
316: and appear to be static structures, consistent with the observed
317: magnetic fluctuations with a fairly constant field direction
318: \citep{burlaga06a, burlaga06b}. Mirror mode waves are also
319: characterized by anti-correlated density and magnetic fluctuations.
320: The density measurements are not available, so we cannot verify the
321: mirror mode from density fluctuations. The density perturbation
322: $\delta n$ can be estimated from \citep{hasegawa69}
323: $$\frac{\delta n}{n} =
324: -(\frac{T_{\perp}}{T_{\parallel}}-1)\frac{\delta B}{B},$$ where
325: $\delta B$ is the perturbation in the field strength. The fluctuation
326: amplitude $\delta B/B$ is $\sim 0.4$ - 0.9 in the heliosheath, which
327: results in $\delta n/n \sim 0.01$ - 0.03, too small to be resolved by
328: future V2 observations.
329:
330: The non-linear evolution of mirror mode instabilities would also
331: generate holes in the background field strength \citep{kivelson96}.
332: The magnetic holes in the heliosheath are of the similar size (in
333: units of the proton gyroradius) to those in planetary magnetosheaths
334: that are predominately produced by mirror mode waves
335: \citep{kaufmann70, bavassano98, burlaga06b}. Magnetic holes in
336: planetary magnetosheaths have also been explained as slow mode
337: solitons based on Hall-MHD theory \citep{stasie04}. The plasma beta
338: in the heliosheath is $\sim$ 40 as estimated above; in this case the
339: ion Larmor radius is much larger than the ion inertial length, so the
340: Hall-MHD theory breaks down \citep{pok05}.
341:
342: Deep in the heliosheath, magnetic field lines may be draped and
343: compressed against the heliopause if there is no significant
344: reconnection between the solar wind and interstellar fields.
345: Analogous to planetary magnetosheaths, the plasma would be squeezed
346: and flow along the draped field lines, leading to a plasma depletion
347: layer close to the heliopause. The plasma beta is reduced in this
348: layer, so the mirror mode instability may be inhibited and ion
349: cyclotron waves may be generated. Damping of these waves would
350: suppress the thermal anisotropy and heat the plasma \citep{liu06b}.
351:
352: \section{Summary}
353: A simple calculation based on temperature anisotropy instabilities
354: explains a variety of observations associated with interplanetary and
355: planetary bow shocks. A shock modifies the velocity distribution of
356: particles at its surface, producing instabilities downstream of the
357: shock which give rise to different types of waves. The calculation
358: also predicts that mirror mode waves form downstream of the TS, which
359: is consistent with the observed magnetic fluctuations. The present
360: results provide a substantial basis for shock-induced anisotropies
361: which act as an energy source for plasma heating in various space and
362: astrophysical environments.
363:
364: \acknowledgments The research was supported under NASA contract
365: 959203 from JPL to MIT, NASA grant NAG5-11623, and by the NASA
366: Planetary Atmospheres and Outer Planets Programs. This work was also
367: supported by the CAS International Partnership Program for Creative
368: Research Teams.
369:
370: \begin{thebibliography}{}
371:
372: \bibitem[Anderson et al.(1994)]{anderson94}
373: Anderson, B. J., Fuselier, S. A., Gary, S. P., \& Denton, R. E. 1994,
374: \jgr, 99, 5877
375:
376: \bibitem[Bavassano Cattaneo et al.(1998)]{bavassano98}
377: Bavassano Cattaneo, M. B., Basile, C., Moreno, G., \& Richardson, J.
378: D. 1998, \jgr, 103, 11961
379:
380: \bibitem[Bavassano Cattaneo et al.(2000)]{bavassano00}
381: Bavassano Cattaneo, M. B., Moreno, G., Russo, G., \& Richardson, J.
382: D. 2000, \jgr, 105, 23141
383:
384: \bibitem[Bogdan et al.(1991)]{bogdan91}
385: Bogdan, T. J., Lee, M. A., \& Schneider, P. 1991, \jgr, 96, 161
386:
387: \bibitem[Burlaga et al.(2005)]{burlaga05}
388: Burlaga, L. F., et al. 2005, Science, 309, 2027
389:
390: \bibitem[Burlaga et al.(2006a)]{burlaga06a}
391: Burlaga, L. F., Ness, N. F., \& Acu\~{n}a, M. H. 2006a, \apj, 642,
392: 584
393:
394: \bibitem[Burlaga et al.(2006b)]{burlaga06b}
395: ---------. 2006b, \grl, 33, L21106, doi:10.1029/2006GL027276
396:
397: \bibitem[Chandrasekhar et al.(1958)]{chan58}
398: Chandrasekhar, S. A., Kaufman, A. N., \& Watson, K. M. 1958, Proc.
399: Roy. Soc. London, Ser. A, 245, 435
400:
401: \bibitem[Chao et al.(1995)]{chao95}
402: Chao, J. K., Zhang, X. X., \& Song, P. 1995, \grl, 22, 2409
403:
404: \bibitem[Czaykowska et al.(2001)]{czaykowska01}
405: Czaykowska, A., Bauer, T. M., Treumann, R. A., \& Baumjohann, W.
406: 2001, Ann. Geophys., 19, 275
407:
408: \bibitem[Decker et al.(2005)]{decker05}
409: Decker, R. B., et al. 2005, Science, 309, 2020
410:
411: \bibitem[Erkaev et al.(2000)]{erkaev00}
412: Erkaev, N. V., Vogl, D. F., \& Biernat, H. K. 2000, J. Plasma Phys.,
413: 64, 561
414:
415: \bibitem[Gary et al.(1997)]{gary97}
416: Gary, S. P., Wang, J., Winske, D., \& Fuselier, S. A. 1997, \jgr,
417: 102, 27159
418:
419: \bibitem[Greenstadt \& Fredricks(1979)]{greenstadt79}
420: Greenstadt, E. W., \& Fredricks, R. W. 1979, in Solar System Plasma
421: Physics, ed. L. J. Lanzerotti, C. F. Kennel, \& E. N. Parker
422: (North-Holland, Amsterdam), 3
423:
424: \bibitem[Gurnett \& Kurth(2005)]{gurnett05}
425: Gurnett, D. A., \& Kurth, W. S. 2005, Science, 309, 2025
426:
427: \bibitem[Hasegawa(1969)]{hasegawa69}
428: Hasegawa, A. 1969, Phys. Fluids, 12, 2642
429:
430: \bibitem[Hudson(1970)]{hudson70}
431: Hudson, P. D. 1970, Plan. Space Sci., 18, 1611
432:
433: \bibitem[Kan \& Swift(1983)]{kan83}
434: Kan, J. R., \& Swift, D. W. 1983, \jgr, 88, 6919
435:
436: \bibitem[Kaufmann et al.(1970)]{kaufmann70}
437: Kaufmann, R. L., Horng, J.-T., \& Wolfe, A. 1970, \jgr, 75, 4666
438:
439: \bibitem[Kivelson \& Southwood(1996)]{kivelson96}
440: Kivelson, M. G., \& Southwood, D. J. 1996, \jgr, 101, 17365
441:
442: \bibitem[Krauss-Varban \& Omidi(1991)]{krauss91}
443: Krauss-Varban, D., \& Omidi, N. 1991, \jgr, 96, 17715
444:
445: \bibitem[Lee \& Ip(1987)]{lee87}
446: Lee, M. A., \& Ip, W.-H. 1987, \jgr, 92, 11041
447:
448: \bibitem[Liu et al.(2006a)]{liu06a}
449: Liu, Y., Richardson, J. D., Belcher, J. W., Kasper, J. C., \& Skoug,
450: R. M. 2006a, \jgr, 111, A09108, doi:10.1029/2006JA011723
451:
452: \bibitem[Liu et al.(2006b)]{liu06b}
453: Liu, Y., Richardson, J. D., Belcher, J. W., Kasper, J. C., \&
454: Elliott, H. A. 2006b, \jgr, 111, A01102, doi:10.1029/2005JA011329
455:
456: \bibitem[Parker(1958)]{parker58}
457: Parker, E. N. 1958, Phys. Rev., 109, 1874
458:
459: \bibitem[Pokhotelov et al.(2005)]{pok05}
460: Pokhotelov, O. A., Balikhin, M. A., Sagdeev, R. Z., \& Treumann, R.
461: A. 2005, \prl, 95, 129501
462:
463: \bibitem[Russell et al.(1990)]{russell90}
464: Russell, C. T., Lepping, R. P., \& Smith, C. W. 1990, \jgr, 95, 2273
465:
466: \bibitem[Stasiewicz(2004)]{stasie04}
467: Stasiewicz, K. 2004, \grl, 31, L21804
468:
469: \bibitem[Stone et al.(2005)]{stone05}
470: Stone, E. C., et al. 2005, Science, 309, 2017
471:
472: \bibitem[Tsurutani et al.(1992)]{tsurutani92}
473: Tsurutani, B. T., Southwood, D. J., Smith, E. J., \& Balogh, A. 1992,
474: \grl, 19, 1267
475:
476: \bibitem[Violante et al.(1995)]{violante95}
477: Violante, L., Bavassano Cattaneo, M. B., Moreno, G., \& Richardson,
478: J. D. 1995, \jgr, 100, 12047
479:
480: \bibitem[Whang et al.(2004)]{whang04}
481: Whang, Y. C., Burlaga, L. F., Wang, Y.-M., \& Sheeley, N. R. 2004,
482: \grl, 31, 3805
483:
484: \end{thebibliography}
485:
486: \clearpage
487:
488: \begin{figure}
489: \plotone{f1.eps}
490: \caption{Temperature anisotropy downstream of a
491: perpendicular shock with $r_s=3$ and $A_1=1$ as a function of
492: $\beta_1$ and $M_{\rm A1}$. The color shading denotes the values of
493: $A_2$. The lower region is forbidden for a $r_s = 3$ shock since the
494: entropy does not increase across the shock. Also shown are the
495: thresholds for the mirror mode (solid line), ion cyclotron (dotted
496: line) and firehose (dashed line) instabilities. Regions above the
497: mirror/ion cyclotron threshold are unstable to the mirror/ion
498: cyclotron mode and regions below the firehose onset are unstable to
499: the firehose instability. The plus sign marks the TS location.}
500: \end{figure}
501:
502: \clearpage
503:
504: \begin{figure}
505: \plotone{f2.eps}
506: \caption{Temperature anisotropy downstream of a
507: parallel shock with $r_s=3$ and $A_1=1$ as a function of $\beta_1$
508: and $M_{\rm A1}$. Same format as Figure~1. Regions below the
509: mirror/ion cyclotron threshold are unstable to the mirror/ion
510: cyclotron mode and regions above the firehose onset are unstable to
511: the firehose instability.}
512: \end{figure}
513:
514: \end{document}
515: