1:
2: \documentclass[12pt,preprint]{aastex}
3:
4: % define some commands
5: \newcommand{\hst}{\textit{HST\ }}
6:
7: \begin{document}
8: % TITLE PAGE
9: \title{First Constraints on Rings in the Pluto System}
10:
11: \author{A.~J.~Steffl and S.~A.~Stern}
12: \email{steffl@boulder.swri.edu}
13:
14: \affil{Southwest Research Institute, Space Science and Engineering
15: Division}
16: \affil{1050 Walnut Street, Suite 400, Boulder, CO 80302}
17:
18: \shorttitle{Constraints on Pluto's Rings}
19: \shortauthors{Steffl and Stern}
20: \slugcomment{Submitted to the Astronomical Journal}
21:
22: % ABSTRACT
23: \begin{abstract}
24: Simple theoretical calculations have suggested that small body
25: impacts onto Pluto's newly discovered small satellites, Nix and
26: Hydra, are capable of generating time-variable rings or dust sheets
27: in the Pluto system. Using HST/ACS data obtained on 2006 February 15
28: and 2006 March 2, we find no observational evidence for such a ring
29: system and present the first constraints on the present-day I/F and
30: optical depth of a putative ring system. At the 1500-km radial
31: resolution of our search, we place a 3$\sigma$ upper limit on the
32: azimuthally-averaged normal I/F of ring particles of 5.1x10$^{-7}$
33: at a distance of 42,000~km from the Pluto-Charon barycenter, the
34: minimum distance for a dynamically stable ring \citep{Sternetal94,
35: Nagyetal06}; 4.4x10$^{-7}$ at the orbit of Nix; and 2.5x10$^{-7}$
36: at the orbit of Hydra. For an assumed ring particle albedo of 0.04
37: (0.38), these I/F limits translate into 3$\sigma$ upper limits on
38: the normal optical depth of macroscopic ring particles of
39: 1.3x10$^{-5}$ (1.4x10$^{-6}$), 1.1x10$^{-5}$ (1.2x10$^{-6}$),
40: 6.4x10$^{-6}$ (6.7x10$^{-7}$), respectively. Were the {\it New
41: Horizons} spacecraft to fly through a ring system with optical
42: depth of 1.3x10$^{-5}$, it would collide with a significant number
43: of potentially damaging ring particles. We therefore recommend that
44: unless tighter constraints can be obtained, {\it New Horizons} cross
45: the putative ring plane within 42,000~km of the Pluto-Charon
46: barycenter, where rings are dynamically unstable. We derive a crude
47: estimate of the lifetime of putative ring paritcles of 900 years.
48:
49:
50: \end{abstract}
51:
52: \keywords{planets and satellites: individual ( Pluto) --- planets:
53: rings --- Kuiper belt}
54:
55: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
56: \section{Introduction}
57: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
58:
59: The discovery of Pluto's two small ($\sim$100~km diameter) satellites,
60: Nix and Hydra, in 2005 \citep{Weaveretal06} raised the possibility
61: that Pluto may possess a time-variable ring system
62: \citep{Sternetal06}. Previously, \cite{Durda:stern00} demonstrated
63: that collisions between small Kuiper belt debris and larger objects,
64: such as Pluto's satellites, are a common occurrence. The
65: characteristic ejecta velocity resulting from these collisions will be
66: of order 1--10\% of the impactor velocity, or 10--100~m s$^{-1}$.
67: Since the escape velocities of Nix and Hydra are between 30 and 90~m
68: s$^{-1}$, depending on their precise size and density, ejecta from
69: such collisions can escape from Pluto's small satellites but will
70: generally remain gravitationally bound to Pluto, thus forming rings.
71: This is in contrast to the situation at Charon, where most such ejecta
72: will fall back onto the surface, due to Charon's $\sim$500~m s$^{-1}$
73: escape velocity. Such rings will be highly time-variable, as
74: collisions on Nix and Hydra compete with loss processes.
75: \cite{Sternetal06} calculated a characteristic ring optical depth of
76: $\tau$=5x10$^{-6}$ for ring material between Nix and Hydra. This is
77: considerably more tenuous than the main rings of Saturn, Uranus, and
78: Neptune, but comparable to the optical depth of Jupiter's rings
79: \citep{Burnsetal84} and the faint dust rings of Uranus
80: \citep{Showalter:lissauer06}.
81:
82: No prior observational constraints on the reflectance or optical depth
83: of rings in the Pluto system have been published. We therefore
84: undertook a search for such material using existing datasets acquired
85: by the \hst Advanced Camera for Surveys (ACS) in early 2006. In
86: addition to the discovery potential of this search, detection of or
87: constraints on Plutonian rings can yield useful information about both
88: the potential hazards of ring material to the NASA {\it New Horizons}
89: Pluto mission, now en route, and the small body population orbiting in
90: the 30--50 AU region of the Kuiper belt traversed by Pluto.
91:
92: In what follows we describe the dataset we used to constrain the
93: amount of ring material in the Pluto system, the analysis techniques
94: we employed, and the results we obtained. We then go on to interpret
95: our results, which are upper limits, as they apply to ring system
96: hazards for {\it New Horizons} and the lifetime of ring particles.
97:
98: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
99: \section{Observations}
100: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
101:
102: Immediately following the discovery of Nix and Hydra, our team
103: obtained Director's Discretionary time to use \hst to confirm the
104: existence of the new satellites. However, to conserve the lifetime of
105: the remaining functional \hst gyros, \hst began operations in 2-gyro
106: mode before the confirmation observations could be executed. In this
107: mode, most targets can only be observed when they are on the trailing
108: side of the Sun as it moves along the ecliptic, thus precluding
109: observations of the Pluto system until February 2006. When Pluto again
110: became observable, it was targeted with two \hst visits of one orbit
111: each using the ACS High Resolution Channel (HRC) through Guest
112: Observer/Director's Discretionary (GO/DD) program 10774. Selected
113: observational parameters from these visits are presented in
114: Table~\ref{obs_table}. \placetable{obs_table}
115:
116: The first visit occurred on 2006 Feb 15.7 UT. During this visit, four
117: long exposures (475-second integrations) with the F606W filter (broad
118: V) were obtained using a non-integer dither box pattern to minimize
119: the effect of bad detector pixels and improve the spatial sampling. In
120: these long exposures, both Pluto and Charon are saturated with a small
121: amount of bleeding along the CCD columns. To aid the registration of
122: the long exposures, short, unsaturated exposures (1-second
123: integrations) were also obtained at each point in the dither pattern.
124: Both Nix and Hydra were clearly detected at high signal to noise
125: \citep{Mutchleretal06IAUC}, and no additional satellites were detected
126: between the orbits of Nix and Hydra down to a 90\%-confidence limiting
127: magnitude of V=25.7 \citep{Steffletal06b}.
128:
129: With the successful confirmation of both satellites from analysis of
130: archival ACS data \citep{Buieetal06} and the 2006 February 15 visit,
131: the observational setup of the second visit, which occurred on 2006
132: March 2, was modified to also obtain the first \bv colors of Nix and
133: Hydra \citep{Sternetal06barxiv}. A three-point, non-integer dither
134: pattern was employed with a 145~s exposure using the F606W filter and
135: a 475~s exposure using the F435W (Johnson B) obtained at each point of
136: the dither pattern. A 3~s exposure using the F435W filter and a 1~s
137: exposure using the F606W filter were also obtained at points 1 and 3
138: of the dither pattern, respectively, to aid in image registration.
139: Since the point spread function (PSF) of the ACS varies significantly
140: between filters, data taken during the 2006 March 2 visit with the
141: F606W filter were analyzed separately from data taken with the F435W
142: filter.
143:
144: For each visit, the exposures made using a given filter were
145: ``drizzled'' together using the PyRAF {\it Multidrizzle} procedure
146: \citep{Koekemoeretal02}. Via this procedure, the individual images are
147: corrected for the geometric distortion of the ACS instrument, rotated
148: so that north is up and east to the left, sky background subtracted,
149: co-registered relative to Pluto using the non-saturated 1-s exposures,
150: and combined using a median filter. Detector pixels that have
151: anomalously low sensitivity, high dark counts, or are saturated were
152: flagged and excluded from further analysis. The median combination
153: removes artifacts, such as cosmic ray events or star trails, that do
154: not appear in same position relative to Pluto in at least two of the
155: images. The resulting images from the February and March visits using
156: the F606W filter are shown in Figure~\ref{plutoim}.
157: \placefigure{plutoim}
158:
159: Pluto is highly saturated in the long integrations obtained during
160: both visits. Charon is also saturated in the long integrations
161: obtained during the February visit. As a result, the two-dimensional
162: pattern of flux from Pluto and Charon dominates over the sky
163: background, even several arcseconds away from Pluto. Complex structure
164: can be seen in this pattern, as a result of the higher-order behavior
165: in the wings of the ACS HRC point spread function (PSF).
166:
167: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
168: \section{Results}
169: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
170:
171: To reduce the strong background gradients in the processed images due
172: to Pluto and Charon, we employ the high-pass filter method of
173: \cite{Showalter:lissauer06}. At each pixel in the image, a
174: first-order, two-dimensional polynomial (i.e. a tilted plane) is fit
175: to the pixels lying within a surrounding circular region. A small
176: circular aperture at the center of this region is excluded from the
177: fit so as not to reduce the intensity of point features. A radius of
178: 12 pixels was chosen for the outer limit of the fit; a radius of 5
179: pixels was chosen for the inner aperture, to enclose 80\% of the flux
180: from a point source \citep{Siriannietal05}. The value of the fit is
181: then subtracted from the central pixel and the process repeated for
182: all pixels in the drizzled images. The resulting filtered images still
183: contain the high-frequency components of the 2-D pattern of flux from
184: Pluto and Charon, but the low-frequency components are much reduced.
185: Visual inspection of the filtered images reveals no immediate evidence
186: for a faint ring system. As noted by \cite{Showalter:lissauer06},
187: this high-pass filter method is not photometrically accurate (tending
188: to narrow bright features), so an alternative method must be used to
189: place quantitative limits on the I/F of a potential ring system.
190:
191: Flux from Pluto and Charon is the dominant source of counts in the
192: region of the images where a ring system might exist. To zeroth
193: order, the PSFs of the ACS HRC are azimuthally symmetric. However,
194: the long integration times used in these observations allow the
195: non-symmetric components of the PSF, i.e., the wings, to rise
196: significantly above the level of the sky background. The situation is
197: also complicated by the fact that neither Pluto nor Charon are true
198: point sources, having apparent angular diameters of 0\farcs104 and and
199: 0\farcs052, respectively, during the February visit (for comparison,
200: the pixel size of the ACS HRC is approximately 0\farcs025). Pluto is
201: also known to exhibit significant albedo variations across its surface
202: \citep{Sternetal97, Youngetal01}. As a result, the observed pattern of
203: flux from Pluto and Charon can not be accurately removed by the simple
204: subtraction of an azimuthally-averaged image from the data. In
205: principle, more sophisticated techniques, e.g. PSF subtraction, can be
206: used to remove this pattern; however, outside of the PSF core, the
207: observed non-symmetric, high-frequency variations in flux from Pluto
208: and Charon are poorly described by model ACS PSFs, such as those
209: generated by the Tiny Tim program \citep{Krist:hook04}. Since neither
210: the 2-D pattern of flux from Pluto and Charon nor the spatial extent
211: of a putative ring system are known {\it a priori}, attempts to
212: characterize and remove Pluto and Charon's flux using fits to the data
213: run a significant risk of inadvertently removing some or all of the
214: signal from a potential ring system. Thus, given the difficulty of
215: removing the background flux from Pluto and Charon without introducing
216: artifacts to the data, we have adopted a simple approach: assuming
217: that all of the flux contained in the image is due to light
218: backscattered from a ring system. This is clearly an
219: oversimplification, but it yields a firm upper limit on the
220: reflectance of a putative ring system.
221:
222: Since the orbits of Pluto's three known moons are all essentially
223: co-planar \citep{Weaveretal06,Buieetal06}, any ring system would be
224: expected to share this plane. We therefore divided the plane of
225: Charon's orbit into 100 circular annuli centered on the Pluto-Charon
226: barycenter. We chose a width of 1500~km for the annuli, based on the
227: projected size of three of the 0\farcs025 HRC pixels at Pluto. Our
228: results, however, are only weakly dependent on the choice of annulus
229: width. Areas of the image containing diffraction spikes from either
230: Pluto or Charon were masked out and excluded from further analysis, as
231: were circular regions surrounding Pluto, Charon, Nix, and Hydra.
232: Pluto's diffraction spikes provide a natural means of dividing the
233: image into quadrants. Quadrant boundaries were used to divide each
234: annulus into four segments. The total count rate contained in and
235: solid angle subtended by each annulus segment within a given quadrant
236: was then calculated.
237:
238: The standard measurement of reflectance for planetary rings is the
239: dimensionless ratio I/F, where I is the observed intensity and $\pi$F
240: is the incident (solar) flux. For an optically thin ring, I/F is
241: proportional to 1/sin(B), where B is the ring opening angle, defined
242: as the angle between the line of sight and the ring plane. We
243: therefore multiply I/F by sin(B) to remove this geomterical factor,
244: yielding the ``normal I/F'', i.e., that seen from an observer looking
245: perpendicular to the ring plane at the same phase angle. The normal
246: I/F is given by the following equation:
247:
248: \begin{equation}
249: \label{I_over_F_eqn}
250: \left(\frac{I}{F}\right)_{\perp}= \frac{C r^2 \sin(B)}{\pi \Omega}
251: 10^{(m_{\sun}+\Delta m (\alpha)-z)/2.5}
252: \end{equation}
253:
254: \noindent where $C$ is the total count rate observed in the
255: segment(s) of an annulus lying within a particular quadrant, in
256: electrons s$^{-1}$; $r$ is the distance from the Sun to Pluto in AU;
257: $\Omega$ is the solid angle subtended by the annulus segment(s); $B$
258: is the ring opening angle, defined as the angle between the line of
259: sight and the ring plane; $m_{\sun}=-26.70$ is the apparent magnitude
260: of the Sun at a distance of 1~AU using the HRC F606W filter bandpass
261: \citep{Colinaetal96}; $\Delta m (\alpha)$ is the difference in
262: magnitude between opposition and phase angle $\alpha$; and $z=26.13$
263: is the magnitude system zero point \citep{Siriannietal05}. Since the
264: phase function of the putative ring system is unknown, we have adopted
265: the phase function used by \cite{Karkoschka01a} to model the rings of
266: Uranus:
267:
268: \begin{equation}
269: \label{phase_func_eqn}
270: \Delta m (\alpha)=\beta \alpha+0.5\alpha/(\alpha_0+\alpha)
271: \end{equation}
272:
273: \noindent The first term of this equation is the standard linear
274: dependence with a phase coefficient of $\beta=0.03$ mag deg$^{-1}$.
275: The second term represents the opposition surge with a width at half
276: maximum of $\alpha_0=8$ deg$^{-1}$. For the phase angle of
277: $\alpha$=1.69$^{\circ}$ during the February visit, these two terms
278: result in a correction of 0.14 mag.
279:
280: If Pluto's rings are more tightly confined radially than the
281: $\sim$1500~km resolution of our search, Eq.~\ref{I_over_F_eqn} will
282: underestimate the true I/F. However, given the apparent lack of any
283: additional small satellites in the Pluto system \citep{Steffletal06b},
284: it is difficult to see how a ring system could remain so tightly
285: confined radially.
286:
287: Profiles of the normal I/F versus distance from the Pluto-Charon
288: barycenter, as derived from the two \hst visits, are shown in
289: Figure~\ref{ioverf_vs_radius}. In all quadrants, the derived I/F value
290: falls off sharply with increasing distance from Pluto, evidence that
291: light from Pluto dominates the total flux contained in the annulus
292: segments. There is no statistically significant evidence for a ring
293: system in any of the I/F profiles. Profiles from the 2006 March 2
294: visit using the F435W filter have a significantly higher I/F value at
295: a given barycentric distance (owing to the broader PSF with this
296: filter), and therefore were not used in further analysis.
297:
298: The inclination of the line of sight to the orbital plane of Pluto's
299: moons causes the circular annuli to appear elliptical when projected
300: onto the plane of the sky. As a result, annulus segments appear closer
301: to Pluto in quadrants II and IV than in quadrants I and III. Thus, I/F
302: upper limits derived from these quadrants are significantly higher, as
303: is readily seen in Figure~\ref{plutoim}. During the 2006 February 15
304: visit, quadrant III provides the tightest constraint on the normal I/F
305: of a putative ring system between the orbits of Nix and Hydra, while
306: during the 2006 March 2 visit, quadrant I provides the tightest
307: constraint. However, the statistical error associated with the I/F
308: profile from quadrant III of the February visit is significantly lower
309: than that from quadrant I of the F606W data from the March visit,
310: owing to the factor of 4 greater integration time and better rejection
311: of star trails and cosmic rays events during the February visit.
312:
313: Dynamical interactions with Charon render orbits about the
314: Pluto-Charon barycenter with semi-major axes less than 42,000~km (2.15
315: times the orbital semi-major axis of Charon) unstable
316: \citep{Sternetal94, Nagyetal06}. The 3$\sigma$ upper limits on the
317: normal I/F of a putative ring are given at this inner boundary and the
318: orbits of Nix and Hydra in Table~\ref{constraints_table}.
319: \placefigure{ioverf_vs_radius} \placetable{constraints_table}
320:
321: Assuming that macroscopic particles dominate the backscatter from the
322: ring, the normal optical depth of the ring can be derived via the
323: following equation:
324:
325: \begin{equation}
326: \label{tau_eqn}
327: (I/F)_{\perp} = p \tau_{\perp}
328: \end{equation}
329:
330: \noindent where $p$ is the geometric albedo of the particles. To
331: provide an upper limit on the normal optical depth of macroscopic ring
332: particles, we assume a very dark ring particle albedo of
333: $p_{_V}=0.04$, similar to particles in the rings of Uranus
334: \citep{Karkoschka01a}. We also calculate normal optical depth limits
335: for an assumed macroscopic ring particle albedo as high as Charon's,
336: i.e., $p_{_V}$=0.38 \citep{Buieetal97}. The derived upper limits on
337: normal optical depth are shown versus barycentric distance in
338: Figure~\ref{tau_vs_radius}, and the values of these upper limits at
339: the minimum stable barycentric distance, Nix's orbital distance, and
340: Hydra's orbital distance are presented in
341: Table~\ref{constraints_table}. \placefigure{tau_vs_radius}
342:
343: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
344: \section{Implications}
345: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
346:
347: Based on the \hst observations of Pluto in February and March 2006,
348: the Pluto system does not currently have rings with a normal I/F in
349: excess of 5.1x10$^{-7}$ or a normal optical depth greater than
350: 1.3x10$^{-5}$. This implies that the current Plutonian ring system, if
351: it exists, is either tenuous, like Jupiter's ring system, or is
352: tightly confined to less than 1500~km in width, below the spatial
353: resolution of our search. However, given that there are no additional
354: satellites of the Pluto system with an effective diameter larger than
355: 16~km \citep{Steffletal06b}, it is difficult to imagine how such a
356: narrow ring could be formed or maintained. While usefully
357: constraining, the derived limits on the normal optical depth of
358: macroscopic ring particles are still somewhat higher than the typical,
359: time-averaged optical depth of $\tau$=5x10$^{-6}$ predicted by
360: \cite{Sternetal06}.
361:
362: We next comment on what our results imply regarding hazards for the
363: {\it New Horizons} spacecraft traversing the Pluto system. Assuming a
364: unimodal size distribution of ring particles, the number of particles
365: that would impact the spacecraft is given by:
366:
367: \begin{equation}
368: N_p=\tau \sec \theta \left(\sigma_{NH}/\sigma_p \right)
369: \end{equation}
370:
371: \noindent where $\theta$ is the inclination of the spacecraft
372: trajectory with the ring plane, $\sigma_{NH}$ is the cross-sectional
373: area of the spacecraft, and $\sigma_p$ is the cross-sectional area of
374: a ring particle. For a ring optical depth of $\tau$=1.3x10$^{-5}$, a
375: trajectory inclined 30$^{\circ}$ to the ring plane, a spacecraft
376: projected cross sectional area of $\sigma_{NH} \approx$10~m$^2$, and a
377: unimodal distribution of 1~$\mu$m diameter ring particles, we find
378: 2x10$^8$ impacts on the spacecraft, with a total impactor mass of
379: 1x10$^{-4}$~g, assuming a ring particle density, $\rho_p$, of 1~g
380: cm$^{-3}$. {\it New Horizons} is robust enough to withstand collisions
381: with 1~$\mu$m particles at its Pluto flyby speed of $\sim$12~km
382: s$^{-1}$. However, collisions with 100~$\mu$m diameter particles are
383: potentially damaging; for a unimodal ring of 100~$\mu$m particles, up
384: to 2x10$^4$ such collisions could be expected.
385:
386: Our ring constraints are not tight enough (by several orders of
387: magnitude) to ensure the safety of the {\it New Horizons} spacecraft,
388: were it to cross the ring plane at a distance where ring particles
389: might be present. In the absence of tighter constraints on the optical
390: depth of a putative ring system, we therefore recommend that {\it New
391: Horizons} cross the ring plane within 42,000~km of the system
392: barycenter, where dynamical interactions with Charon prevent stable
393: orbits \citep{Sternetal94,Nagyetal06}. The current planned trajectory
394: for {\it New Horizons} is safely within this distance.
395:
396: Finally, we can use our ring optical depth constraint to estimate the
397: lifetime of ring particles at Pluto. First, we point out that the
398: re-collision timescale for ring particles orbiting between Nix and
399: Hydra, i.e., their timescale to be reabsorbed by their parent
400: satellite can be computed following \cite{Burnsetal84}, yielding a
401: timescale of order a few 10$^3$ years. Effectively, this is an upper
402: limit for ring particle lifetimes since other loss processes would
403: only speed up the loss timescale.
404:
405: Making the assumption of a steady state ring population, where
406: particle production and loss are in balance over long time scales, we
407: can derive a separate ring particle lifetime, $T_p$, based on our
408: derived ring particle optical depth:
409:
410: \begin{equation}
411: T_p=M_R / \left(dM_R / dt \right)
412: \end{equation}
413:
414: \noindent where $M_R$ is the ring mass, which can be computed from
415: the optical depth constraint $\tau$ as:
416:
417: \begin{equation}
418: M_R = (8/3) \pi r_p \rho_p \tau R \, dR
419: \end{equation}
420:
421: \noindent where $r_p$ is the characteristic ring particle radius,
422: and $R$ and $dR$ are the ring's radius and width. $dM_R/dt$ can be
423: written very simply as:
424:
425: \begin{equation}
426: dM_R/dt=2 \gamma M_{sat}/T_{SS}
427: \end{equation}
428:
429: \noindent where $M_{sat}$ is the average mass of Nix and Hydra,
430: $T_{SS}$ is the age of the solar system, and $\gamma$ is the
431: fractional mass of the satellites that have been lost to erosion.
432: Thus, we find:
433:
434: \begin{equation}
435: T_p = r_p \rho_p \tau T_{SS} R dR / \gamma r_{sat}^3 \rho_{sat}
436: \end{equation}
437:
438: \noindent For a characteristic radius of 50~km and density of 2~g
439: cm$^{-3}$ for Nix and Hydra, $\gamma$=10$^{-4}$ (consistent with Durda
440: \& Stern [2000]), and a unimodal ring of 1~$\mu$m diameter particles
441: with density 1~g cm$^{-3}$ extending from Nix's orbit to Hydra's
442: orbit, i.e., from 48,675--64,780~km \citep{Buieetal06}, we find a
443: crude ring particle lifetime constraint of $T_p$=900~yr.
444:
445:
446: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
447: \section{Conclusions}
448: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
449:
450: We have used existing \hst ACS observations of the Pluto system to
451: derive the first constraints on the normal I/F and optical depth of
452: rings having a radial extent greater than 1500~km. We find a 3$\sigma$
453: upper limit of $I/F_{_\perp}$=5.1x10$^{-7}$ and, assuming a
454: macroscopic ring particle albedo of 0.04 (0.38), a 3$\sigma$ upper
455: limit $\tau_{_\perp}$=1.3x10$^{-5}$ (1.4x10$^{-6}$). Although higher
456: ring optical depths are possible for tightly confined rings less than
457: 1500~km wide, this is unlikely given the apparent lack of additional
458: small satellites to shepherd the rings. The present-day optical depth
459: limits on the rings of Pluto are usefully constraining, but they
460: remain a factor of three larger than the characteristic ring optical
461: depth of 5x10$^{-6}$ predicted by \cite{Sternetal06}. We note that the
462: optical depth constraint derived from the \hst observations is not
463: sufficient to ensure safe passage of the {\it New Horizons} spacecraft
464: through the ring plane at distances of 42,000--65,000~km. Finally, in
465: the case of a ring in steady state, we estimate a ring particle
466: lifetime of 900 years.
467:
468: % ACKNOWLEDGMENTS
469: \acknowledgments
470:
471: Financial support for this work was provided by NASA through grant
472: numbers \mbox{GO-10427} and \mbox{GO-10774} from the Space Telescope
473: Science Institute, which is operated by the Association of
474: Universities for Research in Astronomy, Inc., under NASA contract
475: \mbox{NAS5-26555}. Additional support was provided by the {\it New
476: Horizons Pluto-Kuiper Belt} mission. We thank M. Bullock, L. Young
477: and an anonymous referee for reading and commenting on this
478: manuscript.
479:
480: % BIBLIOGRAPHY
481: %\bibliographystyle{apj}
482: %\bibliography{/Users/Steffl/latex/master_references}
483:
484: \begin{thebibliography}{19}
485: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
486:
487: \bibitem[{{Buie} {et~al.}(1997){Buie}, {Tholen}, \& {Wasserman}}]{Buieetal97}
488: {Buie}, M.~W., {Tholen}, D.~J., \& {Wasserman}, L.~H. 1997, Icarus, 125, 233
489:
490: \bibitem[{{Buie} {et~al.}(2006){Buie}, {Grundy}, {Young}, {Young}, \&
491: {Stern}}]{Buieetal06}
492: {Buie}, M.~W., {Grundy}, W.~M., {Young}, E.~F., {Young}, L.~A., \& {Stern},
493: S.~A. 2006, \aj, 132, 290
494:
495: \bibitem[{{Burns} {et~al.}(1984){Burns}, {Showalter}, \&
496: {Morfill}}]{Burnsetal84}
497: {Burns}, J.~A., {Showalter}, M.~R., \& {Morfill}, G.~E. 1984, in IAU Colloq.
498: 75: Planetary Rings, ed. R.~{Greenberg} \& A.~{Brahic}, 200--272
499:
500: \bibitem[{{Colina} {et~al.}(1996){Colina}, {Bohlin}, \&
501: {Castelli}}]{Colinaetal96}
502: {Colina}, L., {Bohlin}, R.~C., \& {Castelli}, F. 1996, \aj, 112, 307
503:
504: \bibitem[{{Durda} \& {Stern}(2000)}]{Durda:stern00}
505: {Durda}, D.~D. \& {Stern}, S.~A. 2000, Icarus, 145, 220
506:
507: \bibitem[{{Karkoschka}(2001)}]{Karkoschka01a}
508: {Karkoschka}, E. 2001, Icarus, 151, 51
509:
510: \bibitem[{Koekemoer {et~al.}(2002)Koekemoer, Fruchter, Hook, \&
511: Hack}]{Koekemoeretal02}
512: Koekemoer, A.~M., Fruchter, A.~S., Hook, R., \& Hack, W. 2002, in HST
513: Calibration Workshop, 337
514:
515: \bibitem[{{Krist} \& {Hook}(2004)}]{Krist:hook04}
516: {Krist}, J. \& {Hook}, R. 2004, The Tiny Tim User's Guide version 6.3,
517: http://www.stsci.edu/software/tinytim/tinytim.pdf
518:
519: \bibitem[{{Mutchler} {et~al.}(2006){Mutchler}, {Steffl}, {Weaver}, {Stern},
520: {Buie}, {Merline}, {Spencer}, {Young}, \& {Young}}]{Mutchleretal06IAUC}
521: {Mutchler}, M.~J., {Steffl}, A.~J., {Weaver}, H.~A., {Stern}, S.~A., {Buie},
522: M.~W., {Merline}, W.~J., {Spencer}, J.~R., {Young}, E.~F., \& {Young}, L.~A.
523: 2006, \iaucirc, 8676, 1
524:
525: \bibitem[{{Nagy} {et~al.}(2006){Nagy}, {S{\"u}li}, \& {{\'E}rdi}}]{Nagyetal06}
526: {Nagy}, I., {S{\"u}li}, {\'A}., \& {{\'E}rdi}, B. 2006, \mnras, 370, L19
527:
528: \bibitem[{{Showalter} \& {Lissauer}(2006)}]{Showalter:lissauer06}
529: {Showalter}, M.~R. \& {Lissauer}, J.~J. 2006, Science, 311, 973
530:
531: \bibitem[{{Sirianni} {et~al.}(2005){Sirianni}, {Jee}, {Ben{\'{\i}}tez},
532: {Blakeslee}, {Martel}, {Meurer}, {Clampin}, {De Marchi}, {Ford}, {Gilliland},
533: {Hartig}, {Illingworth}, {Mack}, \& {McCann}}]{Siriannietal05}
534: {Sirianni}, M., {Jee}, M.~J., {Ben{\'{\i}}tez}, N., {Blakeslee}, J.~P.,
535: {Martel}, A.~R., {Meurer}, G., {Clampin}, M., {De Marchi}, G., {Ford}, H.~C.,
536: {Gilliland}, R., {Hartig}, G.~F., {Illingworth}, G.~D., {Mack}, J., \&
537: {McCann}, W.~J. 2005, \pasp, 117, 1049
538:
539: \bibitem[{{Steffl} {et~al.}(2006){Steffl}, {Mutchler}, {Weaver}, {Stern},
540: {Durda}, {Terrell}, {Merline}, {Young}, {Young}, {Buie}, \&
541: {Spencer}}]{Steffletal06b}
542: {Steffl}, A.~J., {Mutchler}, M.~J., {Weaver}, H.~A., {Stern}, S.~A., {Durda},
543: D.~D., {Terrell}, D., {Merline}, W.~J., {Young}, L.~A., {Young}, E.~F.,
544: {Buie}, M.~W., \& {Spencer}, J.~R. 2006, \aj, 132, 614
545:
546: \bibitem[{{Stern} {et~al.}(1994){Stern}, {Parker}, {Duncan}, {Snowdall}, \&
547: {Levison}}]{Sternetal94}
548: {Stern}, S.~A., {Parker}, J.~W., {Duncan}, M.~J., {Snowdall}, J.~C.~J., \&
549: {Levison}, H.~F. 1994, Icarus, 108, 234
550:
551: \bibitem[{{Stern} {et~al.}(1997){Stern}, {Buie}, \& {Trafton}}]{Sternetal97}
552: {Stern}, S.~A., {Buie}, M.~W., \& {Trafton}, L.~M. 1997, \aj, 113, 827
553:
554: \bibitem[{{Stern} {et~al.}(2006{\natexlab{a}}){Stern}, {Weaver}, {Steffl},
555: {Mutchler}, {Merline}, {Buie}, {Young}, {Young}, \& {Spencer}}]{Sternetal06}
556: {Stern}, S.~A., {Weaver}, H.~A., {Steffl}, A.~J., {Mutchler}, M.~J., {Merline},
557: W.~J., {Buie}, M.~W., {Young}, E.~F., {Young}, L.~A., \& {Spencer}, J.~R.
558: 2006{\natexlab{a}}, \nat, 439, 946
559:
560: \bibitem[{{Stern} {et~al.}(2006{\natexlab{b}}){Stern}, {Mutchler}, {Weaver}, \&
561: {Steffl}}]{Sternetal06barxiv}
562: {Stern}, S.~A., {Mutchler}, M.~J., {Weaver}, H.~A., \& {Steffl}, A.~J.
563: 2006{\natexlab{b}}, ArXiv Astrophysics e-prints
564:
565: \bibitem[{{Weaver} {et~al.}(2006){Weaver}, {Stern}, {Mutchler}, {Steffl},
566: {Buie}, {Merline}, {Spencer}, {Young}, \& {Young}}]{Weaveretal06}
567: {Weaver}, H.~A., {Stern}, S.~A., {Mutchler}, M.~J., {Steffl}, A.~J., {Buie},
568: M.~W., {Merline}, W.~J., {Spencer}, J.~R., {Young}, E.~F., \& {Young}, L.~A.
569: 2006, \nat, 439, 943
570:
571: \bibitem[{{Young} {et~al.}(2001){Young}, {Binzel}, \& {Crane}}]{Youngetal01}
572: {Young}, E.~F., {Binzel}, R.~P., \& {Crane}, K. 2001, \aj, 121, 552
573:
574: \end{thebibliography}
575: \clearpage
576:
577: % TABLES
578: \begin{deluxetable}{lcccc}
579: \tablewidth{0pt}
580: \tabletypesize{\footnotesize}
581: \tablecaption{Selected Observational Parameters \label{obs_table}}
582: \tablehead{\colhead{Observation Date (UT)\tablenotemark{a}} & \colhead{r (AU)} &
583: \colhead{$\Delta$ (AU)} & \colhead{$\alpha$ (deg)} & \colhead{B (deg)}}
584: \startdata
585: 2006 Feb 15.659 & 31.07 & 31.54 & 1.59 & 37.2\\
586: 2006 Mar 02.747 & 31.08 & 31.31 & 1.77 & 37.5\\
587: \enddata
588: \tablenotetext{a}{Midpoint of observation}
589: \tablecomments{$r$ is Pluto's heliocentric distance, $\Delta$ is
590: Pluto's geocentric distance, $\alpha$ is Pluto's phase angle
591: (Sun-Pluto-Earth angle), and $B$ is the ring plane opening angle.}
592: \end{deluxetable}
593:
594: \begin{deluxetable}{lccc}
595: \tablewidth{0pt}
596: \tabletypesize{\footnotesize}
597: \tablecaption{Selected 3$\sigma$ Upper Limits on the Putative Rings \label{constraints_table}}
598: \tablehead{\colhead{Barycentric Distance (km)} & \colhead{(I/F)$_{\perp}$} &
599: \colhead{$\tau_{\perp}$, p=0.04} & \colhead{$\tau_{\perp}$, p=0.38}}
600: \startdata
601: 42,000 (Min. stable dist.) & 5.1x10$^{-7}$ & 1.3x10$^{-5}$ & 1.4x10$^{-6}$ \\
602: 48,675 (Nix) & 4.4x10$^{-7}$ & 1.1x10$^{-5}$ & 1.2x10$^{-6}$ \\
603: 64,780 (Hydra) & 2.5x10$^{-7}$ & 6.4x10$^{-6}$ & 6.7x10$^{-7}$ \\
604: \enddata
605: \end{deluxetable}
606:
607: % FIGURES
608: \begin{figure}
609: \plotone{f1.eps}
610: \caption[]{Drizzled HRC F606W images of the Pluto system. The image
611: on the left is from the 2006 February 15 visit, while the image on
612: the right is from the 2006 March 2 visit. Pluto, Charon, Nix, and
613: Hydra can be clearly seen. All other features are due to the
614: extended PSF halos around Pluto and Charon or artifacts introduced
615: by the incomplete removal of background star trails or cosmic ray
616: events. The diffraction spikes from Pluto have been used to
617: divide each image into four quadrants. 1500~km width annuli at the
618: orbital distances of Nix (left) and Hydra (right) are shown in
619: red. Four 10$^{\circ}$ segments (half width), centered on each
620: diffraction spike, are excluded from analysis. A one arcsecond
621: scale bar is shown in the upper left of each image.
622: \label{plutoim}}
623: \end{figure}
624:
625: \begin{figure}\epsscale{.8}
626: \plotone{f2.eps}
627: \epsscale{1}
628: \caption[]{``Normal'' I/F as a function of barycentric distance,
629: assuming all observed flux is backscattered from a ring system.
630: 1$\sigma$ error bars are shown at each point. For a given
631: barycentric distance, annulus segments in quadrants I and III
632: appear at a greater angular distance from Pluto, owing to the
633: projection of the putative ring plane on the plane of the sky.
634: Since the observed flux in a given annulus segment is dominated by
635: flux from Pluto, these quadrants provide the tightest constraints
636: on the I/F of a putative ring system. Normal I/F values shown for
637: the 2006 March 2 visit are derived exclusively from data using the
638: F606W filter, as the normal I/F values derived from the F435W data
639: are significantly higher. The orbital semi-major axes of Nix and
640: Hydra are shown by vertical dotted lines at 48,675 and 64780 km,
641: respectively.
642: \label{ioverf_vs_radius}}
643: \end{figure}
644:
645: \begin{figure}
646: \plotone{f3.eps}
647: \caption[]{3$\sigma$ upper limit on ``normal'' optical depth for a
648: as a function of barycentric distance for assumed macroscopic ring
649: particle albedos of 0.04 (thick line) and 0.38 (thin line). The
650: optical depth limit is derived from the normal I/F limit from
651: quadrant III of the 2006 February 15 visit. The horizontal dashed
652: line marks the characteristic ring optical depth predicted by
653: \cite{Sternetal06}. Orbital semi-major axes of Nix and Hydra are
654: shown by vertical dotted lines at 48,675 and 64,780 km,
655: respectively.
656: \label{tau_vs_radius}}
657: \end{figure}
658:
659: \end{document}
660:
661: