1: \documentclass[12pt,eqsecnum,preprint]{aastex}
2: \title{3-D Simulations of Ergospheric Disk Driven Poynting Jets}
3: \begin{document}
4: \author{Brian Punsly}
5: \affil{4014 Emerald Street No.116, Torrance CA, USA 90503 and
6: International Center for Relativistic Astrophysics,
7: I.C.R.A.,University of Rome La Sapienza, I-00185 Roma, Italy}
8: \email{brian.m.punsly@L-3com.com or brian.punsly@gte.net}
9: \begin{abstract}This Letter reports on 3-dimensional simulations
10: of Kerr black hole magnetospheres that obey the general relativistic
11: equations of perfect magnetohydrodynamics (MHD). In particular, we
12: study powerful Poynting flux dominated jets that are driven from
13: dense gas in the equatorial plane in the ergosphere. The physics of
14: which has been previously studied in the simplified limit of an
15: ergopsheric disk. For high spin black holes, $a/M > 0.95$, the
16: ergospheric disk is prominent in the 3-D simulations and is
17: responsible for greatly enhanced Poynting flux emission. Any large
18: scale poloidal magnetic flux that is trapped in the equatorial
19: region leads to an enormous release of electromagnetic energy that
20: dwarfs the jet energy produced by magnetic flux threading the event
21: horizon. The implication is that magnetic flux threading the
22: equatorial plane of the ergosphere is a likely prerequisite for the
23: central engine of powerful FRII quasars.
24: \end{abstract}
25: \keywords{Black hole physics - magnetohydrodynamics -galaxies:
26: jets---galaxies: active --- accretion disks}
27: \section{Introduction}Recent studies of luminous radio quasars indicate that the power of the radio jet
28: can exceed the bolometric luminosity associated with the accretion
29: flow thermal emission \citep{pun06,pun07}. This has proven to be
30: quite challenging for current 3-D numerical simulations of MHD black
31: hole magnetospheres. Based on table 4 of \citet{haw06} and the
32: related discussion of \citet{pun06,pun07}, the most promising 3-D
33: simulations for achieving this level of efficiency are those of the
34: highest spin, $a/M \approx 1$ (where the black hole mass, $M$, and
35: the angular momentum per unit mass, $a$, are in geometrized units).
36: More generally, such high spins have been inferred in some black
37: hole systems based on observational constraints \citep{mcc06}. Thus,
38: there is tremendous astronomical relevance to these highest spin
39: configurations, in particular the physical origin of the
40: relativistic Poynting jet. The first generation of long term 3-D
41: simulations produced one Poynting flux powerhouse, the $a/M=0.995$
42: simulation, KDE \citep{dev03,dev05,hir04,kro05}. The source of most
43: of the Poynting flux was clearly shown to be outside the event
44: horizon in KDE \citep{pun05}. However, without access to the
45: original data, the details of the physical mechanism could not be
46: ascertained. A second generation of 3-D simulations were developed
47: in \citet{haw06}, the highest spin case was KDJ, $a/M=0.99$, with by
48: far the most powerful Poynting jet within the new family of
49: simulations; three times the Poynting flux (in units of the
50: accretion rate of mass energy) of the next closest simulation KDH,
51: $a/M=0.95$. The last three data dumps, at simulation times, t = 9840
52: M, t = 9920 M and t = 10000 M, were generously made available to
53: this author. The late time behavior of the simulations is
54: established after t = 2000 M (when the large transients due to the
55: funnel formation have died off) making these data dumps of
56: particular interest for studying the Poynting jet \citep{haw06}.
57: This paper studies the origin of the Poynting jet at these late
58: times.
59: \par The analysis of the data from the KDJ simulation clearly indicates
60: that the Poynting flux in the outgoing jet is dominated by large
61: flares. Typically, one expects the turbulence in the field variables
62: to mask the dynamics of Poynting flux creation in an individual time
63: slice of one of the 3-D simulations \citep{pun05}. Surprisingly, the
64: flares are of such a large magnitude that they clearly standout
65: above the background field fluctuations as evidenced by figure 1.
66: The flares are created in the equatorial accretion flow deep in the
67: egosphere between the inner calculational boundary at r=1.203 M and
68: r= 1.6 M (the event horizon is at r= 1.141 M). Powerful beams of
69: Poynting flux emerge perpendicular to the equatorial plane in the
70: ergospheric flares and much of the energy flux is diverted outward
71: along approximately radial trajectories that are closely aligned
72: with the poloidal magnetic field direction in the jet (see figure
73: 1). The situation is unsteady, whenever some vertical magnetic flux
74: is captured in the accretion flow it tends to be asymetrically
75: distributed and concentrated in either the northern or southern
76: hemisphere. This hemisphere then receives a huge injection of
77: electromagnetic energy on time scales $\sim 60 M $.
78: \par The source of Poynting flux in KDJ resembles a nonstationary
79: version of the ergospheric disk (see \citet{pun90} and chapter 8 of
80: \citet{pun01} for a review). The ergospheric disk is modeled in the
81: limit of negligible accretion and it is the most direct
82: manifestation of gravitohydromagnetics (GHM) \citet{pun01}. A GHM
83: dynamo arises when the magnetic field impedes the inflow of gas in
84: the ergosphere, i.e., vertical flux in an equatorial accretion flow.
85: The strong gravitational force will impart stress to the magnetic
86: field in an effort to move the plasma through the obstructing flux.
87: In particular, the metric induced frame dragging force will twist up
88: the field azimuthally. These stresses are coupled into the accretion
89: vortex around a black hole by large scale magnetic flux, and
90: propagate outward as a relativistic Poynting jet. The more obstinate
91: the obstruction, the more powerful the jet. There are two defining
92: characteristics that distinguish the GHM dynamo from a
93: Blandford-Znajek (B-Z) process, \citet{blz77}, on field lines that
94: thread the ergopshere:
95: \begin{enumerate}
96: \item The B-Z process is electrodynamic so there is no source within
97: the ergosphere, it appears as if the energy flux is emerging from
98: the horizon. In the GHM mechanism, the source of Poynting flux is in
99: the ergospheric equatorial accretion flow.
100: \item In a B-Z process in a magnetosphere shaped by the accretion vortex,
101: the field line angular velocity is, $\Omega_{F}\approx \Omega_{H}/2$
102: (where $\Omega_{H}$ is the angular velocity of the horizon) near the
103: pole and decreases with latitude to $\approx \Omega_{H}/5$ near the
104: equatorial plane of the inner ergosphere \citep{phi83}. In GHM,
105: since the magnetic flux is anchored by the inertia of the accretion
106: flow in the inner ergosphere, frame dragging enforces $d \phi/dt
107: \approx \Omega_{H}$. One therefore has the condition,
108: $\Omega_{F}\approx \Omega_{H}$.
109:
110: \end{enumerate}
111: In order to understand the physical origin of the Poynting flux,
112: these two issues are studied below.
113:
114: % Figure modified by arXiv admin
115: \begin{figure}
116: \epsscale{0.95} \plottwo{j1a.eps}{j1b.eps}\\
117: \epsscale{0.95} \plottwo{j1c.eps}{j1d.eps}\\
118: \epsscale{0.95}\plottwo{j1e.eps}{j1f.eps}
119: \caption{The source of Poynting flux. The
120: left hand column is $S^{\theta}$ and the right hand column is
121: $S^{r}$ in KDJ, both averaged over azimuth, at (from top to bottom)
122: t= 9840 M, t = 9920 M and t= 10000 M. The relative units (based on
123: code variables) are in a color bar to right of each plot for
124: comparison of magnitudes between the six plots. The contours on the
125: $S^{\theta}$ plots are of the density, scaled from the peak value
126: within the frame at relative levels 0.5 and 0.1. The contours on the
127: $S^{r}$ plots are of $S^{\theta}$ scaled from the peak within the
128: frame at relative levels 0.67 and 0.33. The inside of the inner
129: calculational boundary (r=1.203 M) is black. The calculational
130: boundary near the poles is at $8.1^{\circ}$ and $171.9^{\circ}$.
131: Notice that any contribution from an electrodynamic effect
132: associated with the horizon appears minimal. The white contour is
133: the stationary limit surface. There is no data clipping, so plot
134: values that exceed the limits of the color bar appear white.}
135: \end{figure}
136:
137: %Figure modified by arXiv admin
138: \begin{figure}
139: \epsscale{0.95} \plottwo{j2a.eps}{j2b.eps}\\
140: \epsscale{0.95} \plottwo{j2c.eps}{j2d.eps}\\
141: \epsscale{0.95} \plottwo{j2e.eps}{j2f.eps}
142: \caption{The central engine. The left hand column is $B^{\theta}$ and the right hand column is
143: $\Omega_{F}$ in KDJ, both averaged over azimuth, at (from top to
144: bottom) t= 9840 M, t = 9920 M and t= 10000 M. The relative units
145: (based on code variables) are in a color bar to right of each plot
146: for comparison of magnitudes between the plots. The calculational
147: boundaries are the same as figure 1. The contours on the
148: $B^{\theta}$ plots are of the density, scaled from the peak value
149: within the frame at relative levels 0.5 and 0.1. There is no data
150: clipping, so plot values that exceed the limits of the color bar
151: appear white.}
152: \end{figure}
153:
154: \section{The KDJ Simulation}
155: The simulation is performed in the Kerr metric (that of a rotating,
156: uncharged black hole), $g_{\mu\nu}$. Calculations are carried out in
157: Boyer-Lindquist (B-L) coordinates $(r,\theta,\phi,t)$. The reader
158: should refer to \citet{haw06} for details of the simulation. We only
159: give a brief overview. The initial state is a torus of gas in
160: equilibrium that is threaded by concentric loops of weak magnetic
161: flux that foliate the surfaces of constant pressure. The magnetic
162: loops are twisted azimuthally by the differentially rotating gas.
163: This creates significant magnetic stress that removes angular
164: momentum from the gas, initiating a strong inflow that is permeated
165: by magneto-rotational instabilities (MRI). The end result is that
166: after t = a few hundred M, accreted poloidal magnetic flux gets
167: trapped in the accretion vortex or funnel (with an opening angle of
168: $\sim 60^{\circ}$ at the horizon tapering to $ \sim 35^{\circ}$ at
169: $r > 20 M $). This region is the black hole magnetosphere and it
170: supports a Poynting jet. The surrounding accretion flow is very
171: turbulent.
172: \par In order to understand the source of the strong flares of radial
173: Poynting flux, one needs to merely consider the conservation of
174: global, redshifted, or equivalently the B-L coordinate evaluated
175: energy flux \citep{thp86}. In general, the divergence of the time
176: component of the stress-energy tensor in a coordinate system can be
177: expanded as, $T_{t\; ;\nu}^{\; \nu}=
178: (1/\sqrt{-g})[\partial(\sqrt{-g}\, T_{t}^{\,
179: \nu})/\partial(x^{\nu})] + \Gamma^{\mu}_{\; t \; \beta} T_{\mu}^{\,
180: \beta}$, where $\Gamma^{\mu}_{\; t \; \beta}$ is the connection
181: coefficient and $g = -(r^{2} + a^{2}
182: \cos^{2}{\theta})^{2}\sin^{2}{\theta}$ is the determinant of the
183: metric. However, the Kerr metric has a Killing vector (the metric is
184: time stationary) dual to the B-L time coordinate. Thus, there is a
185: conservation law associated with the time component of the
186: divergence of the stress-energy tensor. Consequently, if one expands
187: out the inhomogeneous connection coefficient term in the expression
188: above, it will equate to zero. The conservation of energy evaluated
189: in B-L coordinates reduces to, $\partial(\sqrt{-g}\, T_{t}^{\,
190: \nu})/\partial(x^{\nu})=0 $, where the four-momentum $-T_{t}^{\,
191: \nu}$ has two components: one from the fluid, $-(T_{t}^{\,
192: \nu})_{\mathrm{fluid}}$, and one from the electromagnetic field,
193: $-(T_{t}^{\, \nu})_{\mathrm{EM}}$. The reduction to a homogeneous
194: equation with only partial derivatives is the reason why the global
195: conservation of energy can be expressed in integral form in (3.70)
196: of \citet{thp86}. It follows that the poloidal components of the
197: redshifted Poynting flux are $S^{\theta}= -\sqrt{-g}\, (T_{t}^{\,
198: \theta})_{\mathrm{EM}}$ and $S^{r}= -\sqrt{-g}\, (T_{t}^{\,
199: r})_{\mathrm{EM}}$. We can use these simple expressions to
200: understand the primary source of the Poynting jet in KDJ. Figure 1
201: is a plot of $S^{\theta}$ (on the left) and $S^{r}$ (on the right)
202: in KDJ at the last three time steps of data collection. Each frame
203: is the average over azimuth of each time step. This greatly reduces
204: the fluctuations as the accretion vortex is a cauldron of strong MHD
205: waves. The individual $\phi=\mathrm{constant}$ slices show the same
206: dominant behavior, however it is embedded in large MHD fluctuations.
207: On the left hand column of figure 1, density contours have been
208: superimposed on the images to indicate the location of the
209: equatorial accretion flow. The density is evaluated in B-L
210: coordinates with contours at 0.5 and 0.1 of the peak value within $r
211: < 2.5M$. Notice that in all three left hand frames, $S^{\theta}$ is
212: created primarily in regions of very high accretion flow density. In
213: all three of the right hand frames of figure 1, there is an enhanced
214: $S^{r}$ that emanates from the ergosphere (defined by the interior
215: of the stationary limit, $r_{s} = M + \sqrt{M^{2}
216: -a^{2}\cos^{2}{\theta}}$, note that there are 40 grid points between
217: $r=1.203 M$ and $r_{s}$ at $\theta =\pi/2$). This radial energy beam
218: diminishes precipitously just outside the horizon, near the
219: equatorial plane in all three time steps. The region in which
220: $S^{r}$ diminishes is adjacent to a region of strong $S^{\theta}$
221: that originates in the inertially dominated accretion flow in the
222: inner ergosphere, $1.2 M < r < 1.6 M $ (this region is resolved by
223: 28 radial grid zones). In fact, if one looks at the conservation of
224: energy equation, the term $\partial (S^{\theta})/\partial {\theta}$
225: is sufficiently large to be the source of $\partial (S^{r})/\partial
226: {r}$ at the base of the radial beam in all three frames. This does
227: not preclude the transfer of energy to and from the plasma. It
228: merely states that the magnitude is sufficient to source $S^{r}$. In
229: general, the hydrodynamic energy flux is negligible in the funnel.
230: In order to illustrate this, contours of $S^{\theta}$ are
231: superimposed on the color plots of $S^{r}$. The contour levels are
232: chosen to be 2/3 and 1/3 of the maximum value of $S^{\theta}$
233: emerging from the dense equatorial accretion flow. One clearly sees
234: $S^{\theta}$ switching off where $S^{r}$ switches on. We conclude
235: that a vertical Poynting flux created in the equatorial accretion
236: flow is the source of the strong beams of $S^{r}$. This establishes
237: condition 1 of the Introduction.
238: \par The left column of
239: figure 2 contains plots of the magnetic field component,
240: $B^{\theta}\equiv F_{r \phi}$, at the three time steps. At every
241: location in which $S^{\theta}$ is strong in figure 1, there is a
242: pronounced enhancement in $B^{\theta}$ in figure 2. Recall that the
243: sign of $S^{\theta}$ is not determined by the sign of $B^{\theta}$.
244: These intense flux patches penetrate the inertially dominated
245: equatorial accretion flow in all three frames. The density contours
246: indicate that the regions of enhanced vertical field greatly disrupt
247: the equatorial inflow. As noted in the introduction, a GHM
248: interaction is likely to occur when the magnetic field impedes the
249: inflow in the ergosphere. The regions of large $B^{\theta}$ are
250: compact compared to the global field configuration of the jet, only
251: $\sim 1.0 M - 2.0 M$ long. Considering the turbulent, differentially
252: rotating plasma in which they are embedded, these are most likely
253: highly enhanced regions of twisted magnetic loops created by the
254: MRI. The strength of $B^{\theta}$ at the base of the flares is
255: comparable to, or exceeds the radial magnetic field strength. The
256: situation is clearly very unsteady and vertical flux is constantly
257: shifting from hemisphere to hemisphere. The time slice t = 10000 M,
258: although primarily a southern hemisphere event, also has a
259: significant contribution in the northern hemisphere (see the blue
260: fan-like plume of vertical Poynting flux in figure 1). The GHM
261: interaction is provided by the vertical flux that links the
262: equatorial plasma to the relatively slowly rotating plasma of the
263: magnetosphere within the accretion vortex. The vertical flux
264: transmits huge torsional stresses from the accretion flow to the
265: magnetosphere.
266: \par Further corroboration of this interpretation
267: can be found by looking at the values of $\Omega_{F}$ in the
268: vicinity of the $S^{r}$ flares. In a non-axisymmetric, non-time
269: stationary flow, there is still a well defined notion of
270: $\Omega_{F}$: the rate at which a frame of reference at fixed r and
271: $\theta$ would have to rotate so that the poloidal component of the
272: electric field, $E^{\perp}$, that is orthogonal to the poloidal
273: magnetic field, $B^{P}$, vanishes. This was first derived in
274: \citet{pun91} (see the extended discussion in \citet{pun01} for the
275: various physical interpretations), and has recently been written out
276: in B-L coordinates in \citet{haw06} in terms of the plasma
277: three-velocity, $v^{i}$ and the Faraday tensor as
278: \begin{eqnarray}
279: && \Omega_{F} = v^{\phi} - F_{\theta r}\frac{g_{rr}v^{r}F_{\phi
280: \theta} + g_{\theta \theta} v^{\theta} F_{r \phi}}{(F_{\phi
281: \theta})^{2} g_{rr} + (F_{r \phi})^{2} g_{\theta \theta}}\;.
282: \end{eqnarray}
283: This expression was studied in the context of the simulation KDH,
284: $a/M=0.95$, in \citet{haw06}. They found that a long term time and
285: azimuth average yielded $\Omega_{F}\approx 1/3\Omega_{H} $ and there
286: was no enhancement at high latitudes as was anticipated by
287: \citet{phi83}. The t = 10000 M time slice of KDH was generously
288: provided to this author. At t = 10000 M, there are no strong flares
289: emerging from the equatorial accretion flow. Inside the funnel at $r
290: < 10M$, at t=10000 M, $0<\Omega_{F} < 0.5\Omega_{H}$.
291: \par The right hand column of figure 2 is
292: $\Omega_{F}$ plotted at three different time steps for KDJ. By
293: comparison to figure 1, notice that each flare in $S^{r}$ is
294: enveloped by a region of enhanced $\Omega_{F}$, typically $0.7
295: \Omega_{H}< \Omega_{F} < 1.2 \Omega_{H}$. The regions of the funnel
296: outside the ergosphere are devoid of large flares in $S^{r}$ and
297: typically have $0<\Omega_{F} < 0.5\Omega_{H}$, similar to what is
298: seen in KDH.. Unlike KDH, there are huge enhancements in
299: $\Omega_{F}$ at lower latitudes in the funnel. It seems reasonable
300: to associate this large difference in the peak values of
301: $\Omega_{F}$ in KDJ and KDH (at t= 10000 M) with the spatially and
302: temporally coincident flares in $S^{r}$ that occur in KDJ.
303: Furthermore, this greatly enhanced value of $\Omega_{F}$ indicates a
304: different physical origin for $\Omega_{F}$ in the flares than for
305: the remainder of the funnel or in KDH at t = 10000 M. The most
306: straightforward interpretation is that it is a direct consequence of
307: the fact that the flares originate on magnetic flux that is locked
308: into approximate corotation with the dense accreting equatorial
309: plasma (i.e., the inertially dominated equatorial plasma anchors the
310: magnetic flux). In the inner ergosphere, frame dragging enforces $
311: 0.7\Omega_{H} < d\phi/ dt < 1.0 \Omega_{H}$ on the accretion flow.
312: This establishes condition 2 of the Introduction.
313: \section{Discussion}In this Letter we showed that in the last three data dumps of
314: the 3-D MHD numerical simulation, KDJ, the dominant source of
315: Poynting flux originated near the equatorial plane deep in the
316: ergopshere. The phenomenon is unsteady and is triggered by large
317: scale vertical flux that is anchored in the inertially dominated
318: equatorial accretion flow. The situation typifies the ergospheric
319: disk in virtually every aspect, even though there is an intense
320: accretion flow. There is one exception, unlike the ergospheric disk,
321: the anchoring plasma rarely achieves the global negative energy
322: condition that is defined by the four-velocity, $-U_{t}<0$, because
323: of the flood of incoming positive energy plasma from the accretion
324: flow. The plasma attains $-U_{t}<0$ only near the base of the
325: strongest flares seen in the $\phi=\mathrm{constant}$ slices.
326: \par The switch-on of a powerful
327: beam of $S^{r}$ outside the horizon at $r\approx 1.3M $ in the
328: $a/M=0.995$ simulation, KDE, of \citet{kro05} was demonstrated in
329: \citet{pun05}. It seems likely the the source of $S^{r}$ in KDE is
330: $S^{\theta}$ from an ergopsheric disk. The ergospheric disk appears
331: to switch on at $a/M>0.95$ as evidenced by the factor of 3 weaker
332: Poynting flux in KDH. Furthermore, if the funnel opening angle at
333: the horizon in KDH at t= 10000 M is typical within $\pm 5^{\circ}$
334: then figure 5 and table 4 of \citet{haw06} indicate that only 35\%
335: to 40\% of the funnel Poynting flux at large distances is created
336: outside the horizon during the course of the simulation. A plausible
337: reason is given by the plots of $B^{\theta}$ in figure 2. The
338: vertical magnetic flux at the equatorial plane is located at $r <
339: 1.55 M$. The power in the ergospheric disk jet $\sim
340: [B^{\theta}(SA)(\Omega_{H})]^{2}$, where SA is the proper surface
341: area of the equatorial plane threaded by vertical magnetic flux
342: \citep{sem04,pun01}. The proper surface area in the ergospheric
343: equatorial plane increases dramatically at high spin, diverging at
344: $a=M$. For example, between the inner calculational boundary and
345: 1.55 M the surface area is only significant for $a/M>0.95$ and grows
346: quickly with $a/M$, exceeding twice the surface area of the horizon
347: for $a/M=0.99$. Thus, if $B^{\theta}$ in the inner ergosphere were
348: independent of spin to first order, then a strong ergospheric disk
349: jet would switch-on in the 3-D simulations at $a/M>0.95$. Note that
350: if the inner boundary were truly the event horizon instead of the
351: inner calculational boundary then this argument would indicate that
352: the ergospheric disk would likely be very powerful even at
353: $a/M=0.95$ and the switch-on would occur at $a/M \approx 0.9$. The
354: implication is that a significant amount of large scale magnetic
355: flux threading the equatorial plane of the ergopshere (which implies
356: a large black hole spin based on geometrical considerations)
357: catalyzes the formation of the most powerful Poynting jets around
358: black holes. Thus, we are now considering initial conditions in
359: simulations that are conducive to producing significant vertical
360: flux in the equatorial plane of the ergosphere.
361: \par It should be noted that 2-D simulations from a similar initial
362: state of torii threaded by magnetic loops have been studied in
363: \citet{gam04}. However, the magnetic flux evolution can be much
364: different in this setting as discussed in \citet{pun05} and poloidal
365: flux configurations conducive to GHM could be highly suppressed. In
366: summary, there are no interchange instabilities, so flux tubes
367: cannot pass by each other or move around each other in the extra
368: degree of freedom provided by the azimuth. Thus, there is a tendency
369: for flux tubes to get pushed into the hole by the accretion flow.
370: This is in contrast to the formation of the ergospheric disk in
371: \citet{pun90} in which buoyant flux tubes are created by
372: reconnection at the inner edge of the ergospheric disk and recycle
373: back out into the outer ergosphere by interchange instabilities.
374: Ideally, a full 3-D simulation with a detailed treatment of
375: resistive MHD reconnection is preferred for studying the relevant
376: GHM physics.
377: \begin{acknowledgements}
378: I would like to thank Jean-Pierre DeVilliers for sharing his deep
379: understanding of the numerical code and these simulations. I was
380: also very fortunate that Julian Krolik and John Hawley were willing
381: to share their data in the best spirit of science.
382: \end{acknowledgements}
383: \begin{thebibliography}{}
384: \bibitem[Blandford and Znajek(1977)]{blz77} Blandford, R. and Znajek,
385: R. 1977, MNRAS. \textbf{179}, 433
386: \bibitem[De Villiers et al (2003)]{dev03} De Villiers, J-P., Hawley, J., Krolik, 2003,
387: ApJ \textbf{599} 1238
388: \bibitem[De Villiers et al (2005a)]{dev05} De Villiers, J-P., Hawley, J., Krolik, J.,Hirose, S.
389: 2005, ApJ \textbf{620} 878
390: \bibitem[De Villiers et al (2005b)]{dev06} De Villiers, J-P., Staff, J., Ouyed, R..
391: 2005, astro-ph 0502225
392: \bibitem[Hawley and Krolik (2006)]{haw06} Hawley, J., Krolik, K.
393: 2006, ApJ \textbf{641} 103
394: \bibitem[Hirose et al (2004)]{hir04}Hirose, S., Krolik, K., De
395: Villiers, J., Hawley, J. 2004, ApJ \textbf{606}, 1083
396: \bibitem[Krolik et al (2005)]{kro05} Krolik, K., Hawley, J., Hirose, S.
397: 2005, ApJ \textbf{622}, 1008
398: \bibitem[McKinney and Gammie (2004)]{gam04}McKinney, J. and Gammie,
399: C. 2004, ApJ \textbf{611} 977
400: \bibitem[McClintock et al (2006)]{mcc06}McClintock, J.E. et al 2006, ApJ \textbf{652},
401: 518
402: \bibitem[Phinney (1983)]{phi83}Phinney, E.S. 1983, PhD
403: Dissertation University of Cambridge.
404: \bibitem[Punsly and Coroniti(1990)]{pun90}Punsly, B., Coroniti, F.V. 1990, ApJ \textbf{354} 583
405: \bibitem[Punsly(1991)]{pun91}Punsly, B. 1991, ApJ \textbf{372} 424
406: \bibitem[Punsly (2001)]{pun01} Punsly, B. 2001, \emph{Black Hole Gravitohydromagnetics}
407: (Springer-Verlag, New York)
408: \bibitem[Punsly(2006a)]{pun05}Punsly, B. 2006, MNRAS \textbf{366} 29
409: \bibitem[Punsly(2006b)]{pun06}Punsly, B. 2006, ApJL \textbf{651} L17
410: \bibitem[Punsly(2007)]{pun07}Punsly, B. 2007, MNRAS \textbf{374} 10
411: \bibitem[Semenov et al (2004)]{sem04} Semenov, V., Dyadechkin, S. and Punsly, B. 2004,
412: Science \textbf{305}978
413: \bibitem[Thorne et al(1986)]{thp86} Thorne, K., Price, R. and Macdonald, D. 1986,
414: \emph{Black Holes: The Membrane Paradigm} (Yale University Press, New Haven)
415: \end{thebibliography}
416: \end{document}
417: