Volume 121, Issue 5 p. 4232-4246
Research Article
Open Access

Pluto's interaction with the solar wind

D. J. McComas

Corresponding Author

D. J. McComas

Plasma Physics Laboratory, Princeton University, Princeton, New Jersey, USA

Southwest Research Institute, San Antonio, Texas, USA

Physics Astronomy, University of Texas at San Antonio, San Antonio, Texas, USA

Correspondence to: D. J. McComas,

[email protected]

Search for more papers by this author
H. A. Elliott

H. A. Elliott

Southwest Research Institute, San Antonio, Texas, USA

Search for more papers by this author
S. Weidner

S. Weidner

Southwest Research Institute, San Antonio, Texas, USA

Physics Astronomy, University of Texas at San Antonio, San Antonio, Texas, USA

Search for more papers by this author
P. Valek

P. Valek

Southwest Research Institute, San Antonio, Texas, USA

Physics Astronomy, University of Texas at San Antonio, San Antonio, Texas, USA

Search for more papers by this author
E. J. Zirnstein

E. J. Zirnstein

Southwest Research Institute, San Antonio, Texas, USA

Search for more papers by this author
F. Bagenal

F. Bagenal

Laboratory of Atmospheric and Space Physics, University of Colorado Boulder, Boulder, Colorado, USA

Search for more papers by this author
P. A. Delamere

P. A. Delamere

Geophysical Institute, University of Alaska Fairbanks, Fairbanks, Alaska, USA

Search for more papers by this author
R. W. Ebert

R. W. Ebert

Southwest Research Institute, San Antonio, Texas, USA

Search for more papers by this author
H. O. Funsten

H. O. Funsten

Los Alamos National Laboratory, Los Alamos, New Mexico, USA

Search for more papers by this author
M. Horanyi

M. Horanyi

Laboratory of Atmospheric and Space Physics, University of Colorado Boulder, Boulder, Colorado, USA

Search for more papers by this author
R. L. McNutt Jr.

R. L. McNutt Jr.

Johns Hopkins University Applied Physics Laboratory, Laurel, Maryland, USA

Search for more papers by this author
C. Moser

C. Moser

Physics Astronomy, University of Texas at San Antonio, San Antonio, Texas, USA

Southwest Research Institute, San Antonio, Texas, USA

Search for more papers by this author
N. A. Schwadron

N. A. Schwadron

Department of Physics, University of New Hampshire, Durham, New Hampshire, USA

Search for more papers by this author
D. F. Strobel

D. F. Strobel

Earth & Planetary Sciences, Johns Hopkins University, Baltimore, Maryland, USA

Search for more papers by this author
L. A. Young

L. A. Young

Southwest Research Institute, Boulder, Colorado, USA

Search for more papers by this author
K. Ennico

K. Ennico

NASA Ames Research Center, Space Science Division, Moffett Field, California, USA

Search for more papers by this author
C. B. Olkin

C. B. Olkin

Southwest Research Institute, Boulder, Colorado, USA

Search for more papers by this author
S. A. Stern

S. A. Stern

Southwest Research Institute, Boulder, Colorado, USA

Search for more papers by this author
H. A. Weaver

H. A. Weaver

Johns Hopkins University Applied Physics Laboratory, Laurel, Maryland, USA

Search for more papers by this author
First published: 04 May 2016
Citations: 32

Abstract

This study provides the first observations of Plutogenic ions and their unique interaction with the solar wind. We find ~20% solar wind slowing that maps to a point only ~4.5 RP upstream of Pluto and a bow shock most likely produced by comet-like mass loading. The Pluto obstacle is a region of dense heavy ions bounded by a “Plutopause” where the solar wind is largely excluded and which extends back >100 RP into a heavy ion tail. The upstream standoff distance is at only ~2.5 RP. The heavy ion tail contains considerable structure, may still be partially threaded by the interplanetary magnetic field (IMF), and is surrounded by a light ion sheath. The heavy ions (presumably CH4+) have average speed, density, and temperature of ~90 km s−1, ~0.009 cm−3, and ~7 × 105 K, with significant variability, slightly increasing speed/temperature with distance, and are N-S asymmetric. Density and temperature are roughly anticorrelated yielding a pressure ~2 × 10−2 pPa, roughly in balance with the interstellar pickup ions at ~33 AU. We set an upper bound of <30 nT surface field at Pluto and argue that the obstacle is largely produced by atmospheric thermal pressure like Venus and Mars; we also show that the loss rate down the tail (~5 × 1023 s−1) is only ~1% of the expected total CH4 loss rate from Pluto. Finally, we observe a burst of heavy ions upstream from the bow shock as they are becoming picked up and tentatively identify an IMF outward sector at the time of the NH flyby.

Key Points

  • First observations of heavy ions from Pluto and their unique interaction with the solar wind
  • Discovery of a Plutopause with an upstream standoff distance at two and a half Pluto radii
  • Discovery of heavy ion tail behind Pluto losing 5 × 1023 ions per second

1 Introduction

On 14 July 2015 the New Horizons (NH) spacecraft flew past Pluto providing a wealth of information about its geology, composition, and atmosphere [Stern et al., 2015]. This flyby also provided the first and only direct observations of Pluto's interaction with the solar wind. Throughout the flyby, the Solar Wind Around Pluto (SWAP) instrument [McComas et al., 2008] onboard NH recorded this remarkable interaction from upstream, through closest approach at ~11:48 (all dates and times are UT, day of year (DOY) 195 at the spacecraft) and then tailward over hundreds of Pluto radii (RP = 1187 km).

The SWAP instrument was specifically designed to measure the effects of the Pluto environment on the solar wind by viewing the sunward direction over a large fraction of the time through the flyby as the NH spacecraft pointed its cameras at Pluto, Charon, and the other moons in the system. This required observations over an extremely large range of angles (~276°) about the spacecraft's primary rotation (Z) axis; the full SWAP field of view (FOV) is ~10° × 276° full width at half maximum. We achieved this goal with a top-hat electrostatic analyzer design that required the absolute minimum spacecraft resources and yet provides both high sensitivity and high signal to noise measurements over this broad range of viewing directions. See McComas et al. [2008] for details about the SWAP instrument.

Prior to the New Horizons flyby there were no direct measurements of Pluto's solar wind interaction. Back to the earliest work on this topic [McNutt, 1989; Bagenal and McNutt, 1989], most hypotheses were for a cometary-like interaction, with an extended region of mass loading far upstream of Pluto. For example, McNutt [1989] had a lower limit of 4 × 1026 s−1 for a pure CH4 atmosphere with higher values—up to 3 × 1028 s−1 for a higher input heat flux. More recently, expectations were based on preflyby estimates of a large N2 escape rate that ranged from 1.5 × 1025 s−1 [Krasnopolsky, 1999] to as high as 2 × 1028 s−1 [Tian and Toon, 2005] and with the most recent atmospheric model before encounter of Zhu et al. [2014] giving a loss rate of ~3.5 × 1027 s−1 and an exobase at ~8 RP. Over the years, expectations grew for an even more complicated interaction, with large ion gyroradii and kinetic effects playing important roles [e.g., Delamere, 2009]. Recently, Bagenal et al. [2015] reviewed the literature and made preflyby predictions of the types of interactions Pluto could have with the solar wind.

Initial results from SWAP (and also the Pluto Energetic Particle Spectrometer Science Investigation instrument [McNutt et al., 2008]) were published in Bagenal et al. [2016]. These included SWAP observations of the pristine solar wind before and after the flyby, which indicated that the solar wind was quite steady with approximate inferred proton moments [Elliott et al., 2016] at the time of the flyby of vp ~403 km s−1, np ~0.025 cm−3, and Tp ~7700 K. The SWAP data also enabled a momentum balance calculation to provide an upper bound of no more than 1% mass loading of the solar wind at ~20 RP from Pluto, upstream along the NH trajectory, and an initial estimate of the point along the NH trajectory where the solar wind had slowed by ~20% from its upstream value. That location was past closest approach along Pluto's dawn flank, and these authors further estimated that this amount of slowing might occur ~6 RP directly upstream from Pluto. This indicates a smaller obstacle than anticipated, consistent with the lower than previously expected atmospheric escape rates, which are currently estimated to be ~1 × 1023 N2 s−1 and ~5 × 1025 CH4 s−1 from NH occultation measurements [Gladstone et al., 2016]. We note that these values are still highly uncertain with estimated uncertainties of about an order of magnitude for N2 and a factor of 3–4 for CH4.

The Bagenal et al. [2016] study used only the coincidence measurements from SWAP, which were adequate to provide the main result of that paper—the overall smaller than expected size scale for the interaction. However, those coincidence measurements do not provide information about the ion composition and thus severely limited what could be interpreted about the overall interaction. Fortunately, in addition to providing high-precision coincidence measurements, SWAP is also capable of differentiating heavy ions from Pluto such as CH4 and N2 from the common light solar wind ions of H+ and He++. In this study we show these mass separated results for the first time and discover the remarkable interaction of the solar wind with Pluto.

2 Observations

Figure 1 shows a color spectrogram of coincidence counts and SWAP viewing angles for the interval from 10:00 to 19:00 UT. Black regions in the “Sun view” bar indicate times when the spacecraft pointing (bottom three plots) pointed any portion of SWAP's FOV within 5° of the Sun direction, and hence, at these times, SWAP is able to measure a radially outflowing solar wind. Because the solar wind flows at a constant speed, H+ and He++ appear as beams with a distinctive signature where the ratio of the energy per charge of He++ relative to that of H+ is ~2. Prior to ~11:15 and after ~18:30, we observed solar wind distributions with this signature as indicated by the relatively narrow beams of solar wind H+ (red, ~0.8 keV/q) and He++ (yellow/green, below ~1.6 keV/q).

Details are in the caption following the image
Color spectrogram of coincidence count rate as a function of (top) E/q and plots of SWAP viewing angles as a function of time. Times when some part of SWAP's field of view is viewing within 5° of the sunward direction and thus should be able to observe radially outflowing solar wind are indicated by black bars in the Sun view panel. Theta and phi angles define the sunward direction in SWAP instrument coordinates with theta measured from the plane of the analyzer top hat (positive above the top hat) and phi measured in the plane of the top hat and are centered on the high-gain antenna axis (positive is right handed about an axis extending up through the top hat). We calculate the (right-handed) spin angle about the antenna axis with 0° defined when the normal to the top hat points antiparallel to the Sun's spin axis. Large (black) and small (red) angle ranges are shown on the two sides of each angle panel. Details of the spacecraft and instrument coordinates and angles can be found in McComas et al. [2008] and Elliott et al. [2016]. Distances from Pluto are given each hour across the bottom; in the Pluto frame, NH entered from upstream on the dawn side, crossed the antisunward axis behind Pluto at ~44 RP, and exited the system well behind Pluto on the duskside (see sketches of the trajectory in Figures 5, 7, and 9).

Interstellar pickup ions (PUIs), which are mostly H+, likewise have a distinctive signature. These pickup ions rapidly scatter in angle, and thus, are visible even when the spacecraft was pointing away from the solar wind [McComas et al., 2010; Randol et al., 2012]. Furthermore, they have a maximum speed of twice the solar wind H+ speed and are therefore observed only up to 4 times the solar wind H+ energy. This PUI distribution is apparent as the broad blue band spanning ~0.5–4 keV in the Figure 1 color spectrogram. Because of this, the interstellar PUI signal is critical for determining if there is a solar wind-type distribution or not when NH turned SWAP away from a solar viewing direction and the high-energy cutoff can even be used to infer an approximate solar wind speed.

From ~12:00 to 14:40 and ~15:00 to 15:15 (black Sun-viewing bars), SWAP was pointed close enough to the sunward direction to see the solar wind if it was flowing roughly radially outward, but no solar wind-type distributions were observed in these intervals. Based solely on these coincidence measurements from SWAP, Bagenal et al. [2016] were unable to conclude if the solar wind was partially excluded by something in the Pluto interaction or if a solar wind-like beam could have simply been deflected out of SWAP's FOV during these times. Thus, the plasma environment downstream of Pluto remained a mystery.

2.1 Resolving Pluto's Heavy Ions from the Solar Wind

The detector section of the SWAP instrument [McComas et al., 2008] is shown in Figure 2. After passing through the top-hat electrostatic analyzer, ions in the selected energy per charge (E/q) passband travel through a nearly field-free conical region and are postaccelerated by ~2.6 keV/q into an ultrathin (~10 nm) carbon foil [McComas et al., 2004]. Electrons ejected from the top (entrance) surface of the foil produce secondary (S) detection events in the secondary channel electron multiplier (CEM), while electrons ejected from the bottom (exit) surface of the foil as well as the primary particles themselves generate primary (P) detection events in the primary CEM. Any events where both P and S are detected within ~100 ns of each other are further labeled as coincidence (C) events.

Details are in the caption following the image
Schematic diagram of a central cut through SWAP's cylindrically symmetric detector section. Primary (P), secondary (S), and coincidence (C) between P and S events are registered by the SWAP electronics. Coincidences provide very sensitive and low-noise observations of light, solar wind ions. However, heavier ions from Pluto's atmosphere are far less efficient at penetrating the foil to produce P, and thus C, events. By examining the S/P ratio for large numbers of ions, we are able to statistically determine if a population is dominated by heavy ions from Pluto or light ions from the solar wind.

The data in Figure 1, and also in Bagenal et al. [2016], only show coincidence observations. While these data are excellent for measuring solar wind protons and alpha particles as well as interstellar pickup ions, heavier ions are poorly detected using only the coincidence measurements in SWAP. This is because (1) many heavy ions, especially at energies less than ~3–4 keV, do not have sufficient energy to pass through the thin carbon foil and produce P events (and thus coincidences) and (2) heavy ions generally produce more entrance surface secondary electrons than light ions [Ritzau and Baragiola, 1998]. McComas et al. [2007] pointed out the usefulness of comparing the S versus P rates in SWAP data from Jupiter's magnetosphere and showed how a high S/P ratio indicates a heavy ion population. Subsequently, Ebert et al. [2010] examined the S/P ratio in more detail and tested a laboratory model of the SWAP instrument with several heavy atomic ion species in order to quantify this effect. Although we expect a small-energy dependence of the S/P ratio for both light and heavy ions, here we identify light ions (H+ and He++) with S/P ≤3.5 and heavy ions with S/P ≥3.5.

For this study we extended our understanding by taking the same laboratory model of SWAP as Ebert et al. [2010] and examining the S/P ratio for CH4+ at several energies where we saw high S/P ratios around Pluto. Figure 3 shows the data points and fits for the postacceleration voltage on the C foil at the time of the NH flyby (−2600 V) and a lower voltage used for prelaunch calibration and typical of what was applied shortly after launch (−2100 V). Not surprisingly, the smaller postacceleration voltage produces larger S/P ratios as fewer of the heavy ions are able to get through the foil. Another interesting result relates to the N+ point in Figure 4 of Ebert et al. [2010] from the prelaunch flight instrument calibration. This point showed a considerably lower S/P ratio than those seen with the laboratory model. As a part of the new round of testing here, we ran down the actual postacceleration voltage used for the original flight instrument N+ test and found that adjusting for the voltage difference brings that data point essentially back onto the laboratory model curve. Together, these results show that a high S/P ratio can be used to identify heavy molecular (CH4+) as well as atomic ions.

Details are in the caption following the image
Secondary to primary (S/P) ratios for CH4+ ions at several energies, measured with the laboratory model of the SWAP instrument. The two sets of points and their fit curves represent data taken with the postacceleration voltage for the Pluto flyby (−2600 V) and with a lower voltage used in flight calibration and early operations (−2100 V).

2.2 Solar Wind Slowing and Heavy Pickup Ions from Pluto

Figure 4 shows color spectrograms of the coincidence and secondary count rates and calculated S/P ratio, along with the instrument pointing angles, for the hour from 11:22 to 12:22 UT. While there is some statistical scatter in the S/P ratios for individual E/q-time pixels, prior to closest approach and up until ~11:54, the S/P ratio remains generally low with values under ~3, indicating light ions. Combining this identification with the presence of interstellar pickup ions even while viewing away from the Sun, we are now able to determine that these ions represent still nearly uncontaminated (and hardly slowed) solar wind even past closest approach along the NH trajectory, off the dawn flank of Pluto.

Details are in the caption following the image
The top three plots show the coincidence (C) and secondary (S) rates and secondary to primary (S/P) ratio, while, as in Figure 1, black bars below them indicate the sunward viewing times based on the detailed instrument pointing shown in the bottom three plots. Light, solar wind ions clearly dominate up until ~11:54, when two consecutive samples show a burst of heavy ions originating from Pluto's escaping neutral atmosphere. Light solar wind ions again dominate after that until ~12:03, when the light ions move to lower energies and are simultaneously joined by higher-energy heavy ions. After ~12:12 the solar wind is no longer apparent and heavy Pluto ions dominate the observations.

With the capability to identify heavy versus light ions, we are able to uniquely identify the lighter solar wind ions and accurately chart their slowing at several additional points along the NH trajectory. In Bagenal et al. [2016], we used SWAP data to provide an upper bound of <1% slowing at ~20 RP and the location of the ~20% slowing point along the NH trajectory. Here we add an intermediate point at ~10% slowing at ~11:50. We note that these data could not be validated as solar wind ions without information provided by the S/P ratio.

The solar wind continued slowing and reached very roughly estimated speeds of (1) ~250 km s−1 at ~12:03 when significant heavy ions began to be visible at higher energies and (2) ~140 km s−1 by ~12:12 when it disappeared and only heavy ions were observed. The disappearance of the light ions of the solar wind at ~100 eV/q, well above the minimum energy of the SWAP instrument (~35 eV/q), indicates the spacecraft's entry into a cavity that largely or entirely excludes the solar wind and interplanetary magnetic field (IMF) and is filled instead with heavy ions from Pluto. The transition occurred over ~6 RP along NH's trajectory (12:03–12:12) indicating a boundary layer ~2 RP thick where NH crossed it, given the shallow angle of the trajectory. NH crossed this boundary layer or “Plutopause” and entered the heavy ion tail behind Pluto at ~12:12 at a transverse distance of ~6.9 RP and 18.9 RP tailward (X) from Pluto.

While NH does not have a magnetometer to measure the magnetic field, at such large heliocentric distances the IMF is nearly always highly transverse to the radial flow and roughly in the prograde or retrograde direction. The solar wind's v × B motional electric field should therefore point roughly north-south. At ~11:54–11:55 a portion of SWAP's FOV rotated to include the north and south directions. At just this moment, we observed a burst of heavy ions at low energies (~0.1–1 keV/q) as indicated by the much larger S/P values at these times in Figure 4. These were the first direct observations of heavy ions escaping from Pluto's extended atmosphere as they were being picked up by the solar wind's motional electric field. Since the surrounding solar wind was traveling ~365 km s−1 just before the burst and ~324 km s−1 just after it, these Pluto pickup ions at such low energies must be being picked up extremely locally and executing only the very first portion of their pickup gyromotion, before they had time to gain much energy.

2.3 The Overall Pluto Interaction

Figure 5 (bottom) shows a schematic diagram of the Pluto interaction along with the total secondary counts (top) and these secondary counts again in the middle plot but color coded separately for heavy and light ions. This differentiation is done by categorizing each E/q-time pixel (first by computing a boxcar average of the primary and secondary rates separately over 3 × 3 pixels) into those that have an S/P ratio greater (heavy—red) or less (light—blue) than 3.5, respectively. While there are a few pixels that are likely coded incorrectly owing to low counting statistics in individual E/q-time pixels, the overall results and interpretation are clear. We note that above ~2–4 keV/q, both the heavy and light ions generally transit the foil, so they are largely indistinguishable from just the S/P ratios [Ebert et al., 2010]. Fortunately, there is very little plasma at these higher energies throughout the flyby, so this ambiguity does not come much into the interpretation of the SWAP data at Pluto.

Details are in the caption following the image
Color spectrograms of count rates (with >3 counts/sample) for (top) secondary counts and (middle) light ions (blue) from the solar wind (SW) and heavy ions (red) from Pluto, combined with a (bottom) schematic diagram identifying the NH trajectory and key regions of the interaction: bow shock, light ion sheath (blue), heavy ion tail (red), and Plutopause boundary layer (BL—purple). The X coordinate is along the Sun-Pluto line and the transverse distance is measured in the plane of X and the spacecraft's trajectory, which is close to the ecliptic plane. A heavy ion (HI) burst was seen ahead of the shock when NH turned so that SWAP was viewing in the right direction to see newly ionized material beginning to be picked up.

As noted above, NH started crossing the Plutopause and entering Pluto's heavy ion tail at ~8.4 RP transverse and ~13.1 RP tailward from Pluto. We use the Earth's magnetopause shape as an analog to that of the Plutopause in order to find a rough standoff distance at the nose. While not entirely applicable owing to the kinetic aspects of Pluto's interaction, here we estimate distances using the shape and the Earth's magnetopause as it is the best statistically known “pause” structure over varying solar wind conditions. In particular, we use the conic equation that Sibeck et al. [1991] found for the self-similar shape of the magnetopause under varying amounts of compression for varying solar wind pressures; here we adjust the pressure ratio to construct the outer boundary of the Plutopause in two parts: from the start of heavy ions at ~12:03 and sunward we use the Sibeck equation, while tailward we connect this point to the exit from the heavy ion tail (~14:15) via a line. For the inner boundary of the Plutopause we connect another solution of the Sibeck equation to a tangent line passing through the final entry into the heavy ion tail (~12:12) and the initial exit point (~14:01). Using this approach, the start of significant heavy ions at ~12:03 corresponds to an upstream distance at the nose of ~3.4 RP, while the exit of the Plutopause boundary and entrance into the heavy ion tail at 12:12 scales to a standoff distance ahead of Pluto of only ~2.5 RP. In contrast to Bagenal et al. [2016], who scaled the 20% speed reduction point, the observations and calculation provided here indicate an even smaller size to the actual obstacle that Pluto presents to the solar wind. We recommend using ~2.5 RP as the obstacle standoff distance for future theoretical and modeling studies of this interaction. Finally, the thickness of the plutopause boundary, where both light and heavy ions are seen to simultaneously rapidly slow, scales to only ~0.9 RP thick (3.4–2.5 RP) at the nose. This is an extremely thin boundary layer, especially considering the huge gyroradii of picked up heavy ions in the solar wind (hundreds of Pluto radii), and suggests a complicated interaction near the Plutopause.

In the middle of the heavy ion tail, NH crossed the Sun-Pluto line at ~12:50 at a down tail distance of ~44 RP; it is fascinating that there is a significant reduction in heavy ion content that appears essentially coincident with NH's crossing of this line. Then, from ~13:25 to 13:50 at distances of ~68–85 RP down tail, the spacecraft executed three complete rolls (see bottom plot in Figure 1). Over these rolls, SWAP was able to view the entire sunward hemisphere and all but a small cone (42° half angle) about the antisunward direction. Still, SWAP observed heavy ions and no solar wind-like distributions, showing that the solar wind was not merely deflected but was nearly or entirely excluded from the tail.

SWAP data show that NH stayed continuously in the heavy ion tail until ~14:15 at a transverse distance of ~15.4 RP, some ~101 RP down tail. The entrance and exit transverse distances indicate a tail diameter in the range of ~17 RP at ~13 RP down tail and ~31 RP at ~101 RP down tail (assuming a circular tail). The highly variable nature of the heavy ions indicates significant structure in the tail with the most intense regions of heavy ions appearing near the inbound and outbound tail boundaries. A different type of Plutopause boundary layer than seen at the tail entrance (12:03–12:12) may be present at the exit (14:01–14:15). At the entrance, both higher E/q heavy ions and lower E/q light solar wind (SW) ions are present together and there is a relatively steep decrease in the ion speed across this transition. In the region just inside the tail exit, the heavy ion E/q jumps to higher values than anywhere else in the tail indicating faster speeds, possibly from coupling to faster plasma flowing outside along the tail.

Heavy ions dominate within the tail and in some regions appear to exclude essentially all solar wind plasma (and if the field remains “frozen in” would therefore exclude essentially all of the IMF). However, the tail also has intermittent regions of low fluxes of light ions centered roughly at ~200 and ~2000 eV, particularly when the heavy ions are both faster (higher E/q) and less intense. These light ions may indicate that the tail is still partially threaded by the IMF. Such interplanetary magnetic flux tubes could contain the highly slowed solar wind and interstellar H PUIs at lower energies and might also provide access for other interstellar H PUIs, in the energy range from ~0.8 to 4 keV (see Figure 1), to propagate in along the field and also be observed inside the tail.

The higher and lower E/q light ion enhancements then persist after NH exited from the heavy ion tail from X ~101 RP back to X ~159 RP, suggesting a region of still significantly excluded and slowed solar wind and probably highly draped IMF, with both high- and low-energy populations of light ions (possibly interstellar PUIs) but without the local presence of heavy ions. We consider this to be a “sheath” region surrounding the tail but still inside the bow shock/wave and a hotter sheath-like flow might exist, deflected out of SWAP's FOV. As NH moves farther away from the tail axis through the sheath, the population of lower-energy light ions progressively weakens as the higher E/q population increases, eventually returning to an external solar wind distribution and melding smoothly into the population of interstellar PUIs.

At ~15:40 (~159 RP tailward) a more solar wind type of interstellar PUI distribution returns but over an energy range that indicates still slightly decelerated solar wind. While the NH pointing was not optimal at these times, the angularly scattered population of interstellar PUIs allows us to roughly track the solar wind speed. Because the interstellar PUI cutoff is ~2.5 keV at ~15:40 (best seen in Figure 1) instead of ~4 keV, the inferred solar wind speed from these interstellar PUI observations is ~320 km s−1. This flow speed then continuously increases back to roughly the upstream SW speed by ~16:05 at a downstream distance of ~175 RP. As pointed out in Bagenal et al. [2016] and shown in Figure 1 of the current study, SWAP was able to view the disturbed solar wind starting at ~18:35, but at this time it is much hotter (proton temperatures up to ~40,000 K) than upstream (~7700 K), indicating significant heating by its interaction with Pluto. The observed temperature continued to drop across two more intervals of solar wind viewing, and by the time SWAP was able to view the solar wind at the start of DOY 196, the conditions had returned to those observed before the flyby, indicating the end of any significant Pluto interaction.

The thin bow shock structure observed near Pluto does not show a sharp boundary and instead seems to have dissipated into a broader wave structure by the distance back that NH began to cross it (~159 RP tailward). Thus, the significant heating of the solar wind observed by SWAP at and beyond these distances was probably not due to shock heating. Instead, we suggest here that the likely source of this heating is from the pickup of heavy ions—in this case mostly CH4 from Pluto, which generate self-excited waves. This process is commonly seen at cometary interactions where the pickup process often extends over very large distances.

2.4 Pluto's Heavy Ion Tail

While the S/P ratio is a robust method for separating heavy Pluto ions from light solar wind ones, we are not currently able to use it to uniquely identify specifically which heavy ions are present. Ultimately, we hope to be able to use SWAP data to differentiate between CH4 and N2, which are expected to have the highest escape rates. Here we calculate the heavy ion moments (Figure 6, thick lines) assuming that all ions are CH4, which are currently expected to dominate N2 [Gladstone et al., 2016] and numerically integrate the observed distributions. We assume that the peak of a symmetric distribution is in the SWAP FOV and the 13.3 km s−1 tailward motion of NH is accounted for. Since the filling of the sensor FOV is not known we perform the numerical integration iteratively. First a quantity that is proportional to the number flux is found assuming a unity geometric factor. This quantity is used to determine a temperature and speed. The filling of the sensor is then calculated as tan−1 (vth vflow), where vth is the thermal velocity and vflow is the flow velocity. The filling factor is then used to scale the geometric factor to determine the flux into the sensor and the density of the plasma. The thin lines indicate the statistical uncertainties and are found by adding and subtracting 1σ from the observed counts and then running the same numerical integrations. Finally, we validated these results using an adapted version of the SWAP analytic model [Elliott et al., 2016], which assumes a drifting Maxwellian. With that, we found very similar results with somewhat lower densities and temperatures, likely because a Maxwellian does not capture the broader, non-Maxwellian wings of the observed distributions.

Details are in the caption following the image
(top) Color spectrogram of heavy ion (assumed to be CH4+) flux and derived bulk flow speed, density, temperature, and thermal pressure (thick lines) in Pluto's heavy ion plasma tail. Thin lines indicate the ±1σ statistical uncertainties.

The parameters vary significantly over the tail passage with much higher densities and lower temperatures within ~45 RP behind Pluto and, to a lesser extent, again immediately inside the outbound crossing. Overall, both the flow speed and temperature generally increase as NH moves back down the heavy ion tail, while the heavy ions are more dense and cooler on the earlier (dusk and southern) side of the tail than on the latter (dawn and northern) side of the tail, other than at the very end of the heavy ion tail passage that may represent a boundary layer. As shown in the bottom plot of Figure 6, the overall plasma thermal pressure is more constant than either the density or temperature alone, indicating some level of global pressure balance between parcels of plasma within the tail.

3 Discussion—Pluto's Unique Solar Wind Interaction

The prior study using SWAP data [Bagenal et al., 2016] showed that the obstacle Pluto present to the solar wind is smaller than previously believed. However, based solely on the <1% and ~20% slowing locations, the overall thickness of the bow shock or bow wave structure was poorly constrained. Here we have used the ion composition data to also establish the location of ~10% slowing inbound and the ~20% slowing point outbound along the tail. The heavy ion addition required to produce the observed solar wind slowing for various times and locations is summarized in Table 1. We note that the calculations are based on conservation of momentum and use the interpolated solar wind speed of 403 km s−1 and density of 0.025 cm−3. Assuming that the solar wind slowing is entirely due to mass loading, we find that 403 km s−1/Vobs = 1 + 16NCH4+/NH+, where Vobs (expressed in km s−1) is the observed, slowed solar wind speed and NCH4+/NH+ is the local density ratio of Plutogenic CH4+ to solar wind H+.

Table 1. Solar Wind Slowing Observed by SWAP in the Vicinity Around Pluto
Time Distance (RP) Transverse (RP) Tailward (X) (RP) Speed (km s−1) Slowing CH4+/H+a CH4+a (cm−3)
11:25 19.9 15.4 −12.7 >399 <1% <6.3 × 10−4 <1.6 × 10−5
11:50 11.6 10.9 3.8 ~365 ~10% 6.5 × 10−3 1.6 × 10−4
11:57 13.0 9.6 8.8 ~324 ~20% 1.5 × 10−2 3.8 × 10−4
12:03b 15.6 8.4 13.1 ~250 ~40% 3.8 × 10−2 9.6 × 10−4
12:12c 20.1 6.9 18.9 ~140 ~65% 0.12 2.9 × 10−3
14:15d 102.5 15.4 101.3
15:40 161.7 30.8 158.7 ~320 ~20% 1.6 × 10−2 4.1 × 10−4
16:05 178.7 35.3 175.2 ~400 ~0 - -
  • a Calculated assuming that heavy ions have fully coupled their momentum to the solar wind flow.
  • b Start of entrance into heavy ion tail via Plutopause boundary layer.
  • c End of Plutopause boundary layer and entry into pristine heavy ion tail.
  • d Exit from heavy ion tail.

Using momentum balance to calculate heavy ion mass loading assumes that the heavy ions are fully picked up by the motional electric field of the flowing plasma. For ions just beginning to make their first gyration, it is possible that only a portion of their momentum may have been effectively coupled into the flow. If so, the heavy ion ratios and densities would be larger by the inverse of whatever fraction of their momentum was already effectively coupled to the flow. We also note that the heavy/light ion ratios and heavy ion densities are based on the assumption that the escaping ions are CH4+ as indicated by recent atmospheric escape models [Gladstone et al., 2016]; if it turns out that N2+ or some other species dominates, the numbers would need to be adjusted by the ions' relative mass ratio. The SWAP observations provide very strong constraints on the solar wind's interaction with Pluto, and any models seeking to emulate this interaction will need to quantitatively match the slowing and exclusion at all of the points given in Table 1.

The locations of the ~10% slowing point and start of the Plutopause are only separated by ~2.5 RP in transverse distance along the NH trajectory. If the separation scales at the nose by the canonical ratio of two thirds, we get an estimated distance at the nose of only ~1.7 RP between where the solar wind is slowed by just ~10% and the start of its exclusion at the Plutopause. Further, if we take the 20% slowing points inbound (~11:57) and outbound (~15:40) and connect them to the self-similar conic equation from Sibeck et al. [1991], which we used to describe the Plutopause near the nose, but scale it outward for the bow shock, we get the configuration shown in Figure 7 with an upstream bow shock distance of only 4.5 RP. Such a thin region of slowing indicates that Pluto's bow structure thickness is actually at the very smallest end of the range discussed by Bagenal et al. [2016], measuring roughly just two ion inertial lengths.

Details are in the caption following the image
Size scale of Pluto interaction with the solar wind derived from SWAP data. The bow shock is indicated by the extension of the locations where we measured the light, solar wind ions to be ~20% slowed down from the upstream solar wind speed. The Plutopause (purple) is a finite sized boundary layer ~0.9 RP thick at the nose and separates the solar wind (blue) from the heavy ion tail (red). Even though the heavy ion tail extends back >100 RP at the time of the NH flyby, the upstream interaction is very compact and the bow shock is almost compressed onto the obstacle.

Because of the observation of the thinness of the upstream bow structure in this study, we are able to definitively identify this structure as a bow shock and not some sort of broader bow wave at the location that NH initially crossed it. Also, with the discovery that Pluto has a thin (~0.9 RP thick) Plutopause where the impinging solar wind is largely (or entirely) excluded just ~2.5 RP upstream of Pluto, we find that the bow shock is extremely close to the obstacle itself at the time of the NH flyby. We further note that NH flew by Pluto under very strong (high dynamic pressure) solar wind conditions. For more typical, weaker, solar wind conditions, the size of the interaction would be larger, which could allow the bow shock to move outward from the obstacle.

So what causes the Plutopause at such a close in distance from Pluto? Possible obstacles to the solar wind flow include (1) mass loading, such as in cometary interactions; (2) ionospheric obstacles, such as Venus and Mars; and (3) magnetospheres, like at Earth. Here we discuss each of these possibilities in turn.

The basic feature of a comet-like solar wind interaction is the mass loading of the solar wind by ionizing the outflowing neutrals by UV and charge exchange, generating an upstream collisionless shock [Galeev et al., 1985; Ip and Axford, 1982; Mendis and Horanyi, 2014]. Ignoring kinetic effects and carrying out these fluid calculations, we find that the subsolar position of the shock at ~4.5 RP is consistent with a neutral CH4 production rate (escape rate from the atmosphere) Q = 5 × 1025 s−1, for an outflow speed ~100 m s−1 and an upstream undisturbed solar wind Mach number M of ~10 as shown in Figure 8.

Details are in the caption following the image
The position of the shock as function of the outflow speed of neutral CH4 molecules for different Mach number flows. Speeds ~100 m s−1 and Mach numbers ~10 are consistent with the inferred outer extent of the directly upstream bow shock at ~4.5 RP.

For comets, the shocked solar wind flow is diverted around a pressure-balance surface that is maintained inside by ion-neutral drag of cometary origin particles [Ip and Axford, 1982; Cravens, 1986]. However, this yields a subsolar distance to the pressure balance surface (the size of the effective obstacle) that is much smaller than 1 RP and therefore unphysical. Thus, while the shock can form at the observed distance via a cometary-like interaction, the actual obstacle to the solar wind must be supported by one of the other processes: thermal pressure in Pluto's atmosphere or a relatively strong internal magnetic field.

The standard description for the solar wind interaction with a nonmagnetized planet with an atmosphere [Luhmann, 1995] says, to first order, that the solar wind dynamic pressure is converted to magnetic and thermal energies behind a bow shock making a relatively stagnated sheath with compressed magnetic field and plasma in pressure balance with the thermal pressure of the ionosphere. For the upstream solar wind at the time of the NH flyby (403 km s−1 and 0.025 cm−3) the dynamic pressure is ~6 pPa. Postencounter estimates of the ionosphere of Pluto by coauthor D. F. S. based on atmospheric model in Gladstone et al. [2016] and considerations in Cravens and Strobel [2015] indicate a maximum density of ~1500 cm−3 at about 700 km altitude or 1.6 RP from the center of the Pluto. These numbers are also consistent with atmosphere/ionosphere models [Lara et al., 1997; Ip et al., 2000]. The ionospheric temperature at this altitude is not known, but if we take the temperature of the neutral atmosphere at this altitude (~100 K) then we get an ionospheric thermal pressure of ~2 pPa. Models of the ionosphere suggest a scale height of ~400 km, which would decrease this pressure by a factor of ~15 from 1.6 RP (the peak of the ionosphere) to ~0.13 pPa at 2.5 RP (the interaction boundary). Even if the ions were ~5 times hotter than the neutral atmosphere, consistent with Mars [Fox et al., 2015], the ionospheric pressure at 2.5 RP would still be a factor of ~9 below the solar wind ram pressure. Therefore, we conclude that there are likely other factors that contribute to the standoff of the solar wind.

At Mars there is a similar situation where the interaction region is larger than predicted by the simple ionospheric pressure standoff calculation. The induced magnetospheric boundary (IMB) at Mars is at ~1.3 RM at the subsolar point and ~2.1 RM at the terminator (see review by Brain et al. [2016]). This is about a factor of 1.15 higher than the ionopause boundary (inside of which the ionospheric pressure dominates) at 1.13 RM. If we similarly scale Pluto's interaction boundary of 2.5 RP we get an equivalent ionopause boundary at 2.2 RP. The subsolar location of the bow shock upstream of Mars is about a factor of ~1.25 higher than the IMB at ~1.6 RM. If we take Pluto's interaction boundary to be 2.5 RP, then the correspondingly scaled bow shock subsolar location would be at ~3 RP. This value is inside our inferred Plutopause boundary layer in a region where the residual solar wind is rapidly slowing and the fraction of heavy ions is rapidly growing. Clearly, the solar wind interaction with Pluto has similarities to that of Mars where the solar wind standoff distance requires modest additional factors beyond simple pressure balance with the ionosphere, such as mass loading due to pickup of ionized escaping as well as ballistic and satellite orbit atmospheric molecules and/or induced currents in the ionosphere [Brain et al., 2016].

Finally, another possible obstacle to the solar wind is an intrinsic magnetic field of Pluto, which confines the observed heavy ions and tail within a magnetosphere. Early estimates of a possible remnant magnetic field of Pluto ranged from 42 nT to 450 nT at the surface [Bagenal and McNutt, 1989; Bagenal et al., 1997]. There are many aspects of both Pluto's interior and magnetic dynamo theory that are poorly understood, which makes it difficult to predict whether Pluto might have such a field. The global density of 1.86 × 103 kg m−3 [Stern et al., 2015] means that a rock/iron core likely extends for 2/3 to 3/4 of Pluto's radius, depending on what one assumes for the core density. Scaling from the interior of Ganymede [Schubert et al., 2004], the only icy object known to have an internal dynamo, suggests that Pluto's iron core might have a radius 200–400 km. If we then use typical dynamo scaling laws [Christensen, 2010] and ignore all factors except core radius, then scaling from Ganymede, we find an upper limit of about 200 nT for a dynamo field at the surface. This calculation ignores such factors as heat source and spin rate, as well as the fact that since we really do not know how and where Pluto formed and evolved, the size of an iron core is a very rough estimate. For comparison, the Moon has an iron radius of ~450 km but no internal dynamo.

While NH does not carry a magnetometer to resolve the magnetic field issue, we ask here if the SWAP flyby data can somehow be used to more definitively resolve the question of any possible internal magnetic field from Pluto. We find that the observed interaction is consistent with a magnetic field only if it is weak enough not to push the upstream standoff distance beyond ~2.5 RP—the inner edge of the Plutopause. Taking this distance and a factor of 2.44 enhancement over a vacuum dipole field as at Earth [Kivelson and Russell, 1995], we set an upper bound on a dipolar surface field at Pluto of <30 nT, a factor ~7 times smaller than the prior theoretical upper bound given above.

Thus, we find that Pluto's interaction with the solar wind is a hybrid of comet-like and the Venus/Mars-like interactions. The upstream shock can be generated by mass loading, but the pressure balance diverting the shocked solar wind flow is sustained by atmospheric thermal pressure.

Another intriguing aspect of the interaction is the possibility that the heavy ion tail could still be partially threaded by the IMF. SWAP's FOV was very constant from ~12:15 to ~14:45 (except for the plasma roll) and viewed both the sunward direction and northward and southward. We noted that intermittent regions of light ions centered roughly at ~200 eV and ~2000 eV (especially when the heavy ions are both faster and less intense), which we interpret as possible residual threading of the tail and sheath region surrounding the heavy ion tail. While this interpretation is not certain, it could account for the two distinct populations of light ions—those leaking in along the residual magnetic field and retaining the external energies of the interstellar pickup H+, and a much lower energy population that could be the initial interstellar pickup H+ that was carried in and slowed along with the impinging solar wind. At energies of ~200 eV, these would be consistent with very slow plasma flow (<100 km s−1) as observed in the heavy ion moments.

Because NH does not carry a magnetometer, there are also no direct observations of either the perturbation of the IMF by Pluto or even which IMF sector the solar wind had at the time of the flyby. Therefore, here we try to use information from ions measured by SWAP to infer this sector structure. We note that NH moved duskward and slightly northward through the Pluto-Charon system (see Figure 9) and that there was a significant change in heavy ion content that appeared to be essentially coincident with NH's crossing of the Sun-Pluto line at ~44 RP down tail. The smaller fluxes in the northern compared to the southern portion of the tail might indicate which direction IMF sector that Pluto was in at the time of the NH flyby. Because the IMF is almost certainly roughly parallel to the ecliptic plane, kinetic and gyroscale effects should cause a significant north-south asymmetry as seen in hybrid models [Delamere, 2009] and for the Active Magnetospheric Particle Tracer Explorers (AMPTE) “artificial comet.” For AMPTE, the dense heavy ion plasma (Ba, 137 amu) was observed (and subsequently simulated) to recoil in the direction opposite to the barium pickup ion motion [Delamere et al., 1999]. Similarly, for an outward IMF sector, Pluto pickup ions would move initially northward and drive the rest of the plasma flow southward in order to conserve momentum. Thus, we tentatively interpret the overall greater heavy ion densities in the southern compared to the northern portions of the tail as indicative of Pluto being in an outward IMF sector at the time of the NH flyby.

Details are in the caption following the image
Schematic diagram of Pluto's interaction with the solar wind as inferred from SWAP observations along the trajectory of the NH flyby. NH crossed the Sun-Pluto line from the dawn/southern portion of the tail (dashed portion of trajectory) into the dusk/northern (solid portion of the trajectory in the cutaway) at ~44 RP down tail. Portions of the trajectory inside the heavy ion tail behind Pluto are indicated in red and light ion sheath that surrounds the tail is in blue. The bow shock observed near Pluto has dissipated into just a bow wave by the distance back that NH exited through it.

In the above material we provided a discussion and sketches of the solar wind interaction (Figures 5, 7, and 9) that appears largely symmetric based on NH's single transit of the Pluto system. While we have no expectation that the interaction will be symmetric in the north-south direction, at least to first order we do expect the east-west structure to be far more symmetric. Because NH is moving far more east-west compared to north-south, the analysis given here should be a good starting point for future more detailed kinetic modeling of the interaction along the full NH trajectory.

Over the entire heavy ion tail (12:04–14:15 UT) we calculated the average plasma properties (see Figure 6) as provided in Table 2. For the average measured down tail flux of ~7 × 104 cm−2 s−1 and an average inferred tail diameter of ~(17 + 31)/2 = 24 RP over the observed distances, we calculate a CH4+ ion loss rate of ~5 × 1023 s−1 down the tail, which is 2 orders of magnitude smaller than the current best estimate of 5 × 1025 s−1 neutral CH4 molecules from NH flyby observations combined with atmospheric models [Gladstone et al., 2016].

Table 2. CH4 Bulk Momentsa
Values Units
Flow speed (vflow) ~89 km s−1
Density (n) ~8.8 × 10−3 cm−3
Temperature (T) ~7.0 × 105 K
Thermal pressure (<nkT>) ~2.1 × 10−2 pPa
Ram pressure <ρvflow2> ~1.2 pPa
Sound speed <(γkT/mi)1/2> ~22 km s−1
Sonic Mach number ~5.3
Number flux (<nvflow>) ~7.2 × 104 cm−2 s−1
  • a Note that while we provide two digits so others can use them for calculations, probably only the first digit is actually significant.

The SWAP-observed heavy ion escape rate confined in the plasma tail is ~1% of the rate of the escaping neutrals. However, due to the long photoionization and charge exchange lifetimes of the escaping neutrals, these processes alone cannot explain the observations. Hence, in addition to pickup mass loading of the incoming solar wind, fluid or kinetic effects must be shaping the direct interaction of the solar wind with Pluto's ionosphere. These effects can generate a more efficient charge exchange between the slowed solar wind protons and the neutral atoms in Pluto's atmosphere and fuel the heavy ion tail.

The slight upward trend in speed with distance down the heavy ion tail suggests that there may be acceleration (approximately few m s−2) of the heavy ions as they move tailward, possibly owing to the partial threading of the tail by the IMF. Such threading would introduce a field line tension force that would act to restraighten the flux tubes that became highly kinked by mass loading. Finally at ~14:00 the speed jumps up ~40 km s−1 and stays up, which could be indicative of an interaction at the tail boundary (possibly Kelvin-Helmholtz) with faster moving sheath flow along its flank. In fact, ion kinetic models of the Pluto interaction [Delamere, 2009] show significant structure at the boundary of a tail-like region from these sorts of interactions.

While the detailed magnetic field geometry would need to be known to calculate the plasma acceleration, here we provide the simplest possible calculation: that is, if the tail were fully threaded by the IMF, then the ion loss rate down the tail would be balanced by the Maxwell stresses from the highly kinked IMF. Neglecting all other pressures, an estimation of the mass loss rate comes from balancing this magnetic shear with the plasma acceleration as ~2BxByA/Vswμo, where the solar wind speed Vsw is ~403 km−1, the estimated IMF By is ~0.15 nT, Bx is approximately the Mach number times By or ~10 By = 1.5 nT, and the cross-sectional area of the tail is ~100 RP long × ~24 RP wide. Together this comes to ~1 × 1023 CH4 s−1, which is again close to the estimated total estimated flux of CH4+ ions of ~5 × 1023 s−1 escaping down the heavy ion tail.

Finally, we noted above that the heavy ion tail appears to be in at least some level of pressure balance with the calculated ion densities and temperatures being roughly anticorrelated. The average pressure of ~2 × 10−2 pPa is roughly an order of magnitude larger than the upstream solar wind thermal pressure and thus roughly comparable to the larger solar wind pickup H pressure at 33 AU [Randol et al., 2013]. This result makes sense because the size of Pluto's heavy ion tail should be set by the flaring of the tail boundary and ultimately by the transverse pressure of the interstellar pickup hydrogen on the outside and heavy ion plasma pressure on the inside.

It is interesting to consider what other effect the plasma environment from Pluto could have on other parts of the Pluto-Charon system. In particular, we wonder if the surface of Charon might be affected by the heavy ions in the tail. Such effects have already been considered for the impinging solar wind ions. Because their energy is high (~100 eV−1 keV), the solar wind may be able to interact strongly with the surface ices and modify them chemically [Johnson, 1989; Bagenal et al., 1997]. In particular, the solar wind, and cosmic rays, may be responsible for the dark coloration via reactions that produce complex molecules generally not expected at the cold temperatures of Pluto's or Charon's surface [Cooper et al., 2003; Kim and Kaiser, 2012; Tucker et al., 2015].

The side gravitationally locked toward Pluto is darker than the opposite side overall [Grundy et al., 2016]. In contrast to cosmic rays which always can impinge on all sides of Charon and the solar wind that can also impinge on all sides over the course of each Pluto year, the heavy ion plasma tail can only ever interact with the side toward Pluto and only over two parts of each Pluto year. During the seasons when Charon's orbit passes through the tail, it spends a significant faction of the time within the tail where the Pluto-facing side is bombarded with large fluxes (~7 × 104 cm2 s−1) of energetic CH4 ions. These ions are significantly higher energy than the solar wind as the energy per amu is similar (~most of a keV/amu) and CH4 is 16 times heavier; this difference imparts significantly more energy per ion and may have implications for implantation, surface chemistry, and ultimately the overall darkening of this side of Charon.

4 Conclusions

In this study we have shown the first direct observations separating heavy ions from Pluto from the light ions of the solar wind and thus determined and quantified the Pluto-solar wind interaction for the first time. In particular, in this study we find an
  1. even smaller obstacle than suggested by Bagenal et al. [2016] with the addition of the ~20% slowing point scaling to an upstream distance of only ~4.5 RP ahead of Pluto center;
  2. very thin bow shock (roughly just two ion inertial lengths thick) that lies just above Pluto's obstacle to the solar wind;
  3. burst of low-energy (~0.1−1 keV/q) heavy ions just being picked up in the solar wind flow as SWAP rotated to view in the v × B motional electric field direction;
  4. thin Plutopause (~2 RP thick where NH crossed it and scaled to ~0.9 RP thick at the nose) where the solar wind (and therefore IMF) is entirely or at least nearly entirely excluded;
  5. scaled upstream standoff distance for the Pluto obstacle to the solar wind of only ~2.5 RP;
  6. long (>100 RP) tail filled with heavy ions from Pluto's atmosphere with considerable structure and an asymmetry with considerably more heavy ions in the dawn/southern compared to the dusk/northern half of the tail;
  7. light ion sheath outside the tail with a lower-energy population of light ions that progressively weakens as the higher-energy population increases and eventually transitions into the interstellar PUI population;
  8. calculated heavy ion fluxes and moments for the heavy ion tail that showed slightly increasing flow speed and temperature with down tail distance and more dense and cooler heavy ions on the dusk/south than the dawn/north portion of the tail;
  9. preliminary identification of the IMF having an outward sector at the time of the NH flyby;
  10. upper bound on any intrinsic magnetic field for Pluto of <30 nT dipolar surface field;
  11. calculated CH4+ loss rate down the heavy ion tail of an ~5 × 1023 s−1, which represents ~1% of the currently estimated atmospheric loss rate of CH4 molecules;
  12. mild acceleration (approximately few m s−2) of the heavy ions as they move down the tail, possibly owing to partial threading of the tail as also indicated by high- and low-energy populations of light ions intermittently in the tail;
  13. roughly anticorrelated density and temperature in the heavy ion tail indicating some level of global pressure balance within the tail and an average level of ~2 × 10−2 pPa, roughly consistent with the interstellar PUI pressure in the solar wind at ~33 AU.
  14. overall, Pluto interaction with the solar wind appears to be a hybrid with the bow shock generated by mass loading like at a comet, but the obstacle to the solar wind flow—the Plutopause—is sustained by atmospheric thermal pressure at Venus and Mars.

The Solar Wind Around Pluto—SWAP—instrument has provided a wealth of detailed and quantitative information about Pluto and its interaction with the tenuous solar wind out at ~33 AU. While this study has provided many of the key results, the SWAP data will no doubt be reanalyzed with increasingly subtle techniques and combined with detailed numerical modeling for many years to come as the community collectively grapples with Pluto's unique solar wind interaction—one that is unlike that at any other body in the solar system.

Acknowledgments

We are grateful to all the team members who made the SWAP instrument and New Horizons spacecraft, mission operations, and overall mission possible. This work was carried out as a part of the New Horizons mission, which is in NASA Planetary Science's New Frontiers Program. These data are available from the SWAP instrument PI—Dave McComas at [email protected].