Acceleration of Scattered Solar Wind Protons at the Polar Terminator of the Moon: Results from Chang’E‐1/Swids X.‐D

GEOPHYSICAL RESEARCH LETTERS, VOL. 37, L07203, doi:10.1029/2010GL042891, 2010 Click Here for Full Article

Acceleration of scattered solar wind protons at the polar terminator of the Moon: Results from Chang’E‐1/SWIDs X.‐D. Wang,1,2 W. Bian,1 J.‐S. Wang,3 J.‐J. Liu,1 Y.‐L. Zou,1 H.‐B. Zhang,1 C. Lü,1 J.‐Z. Liu,1 W. Zuo,1 Y. Su,1 W.‐B. Wen,1 M. Wang,1 Z.‐Y. Ouyang,1 and C.‐L. Li1 Received 11 February 2010; accepted 5 March 2010; published 9 April 2010.

[1] We report accelerated particles observed by Solar Wind Japanese Kaguya(SELENE), Chinese Chang’E‐1 (CE‐1 Ion Detectors (SWIDs) on Chang’E‐1 spacecraft close to herein) and Indian Chandrayaan‐1. Up to now, Kaguya has terminator regions of the Moon. As the spacecraft crosses observed the ions originating from the Moon [Yokota et al., the terminator, a stream of ions with energy of ∼200eV/q 2009; Tanaka et al., 2009], solar wind protons scattered at are detected. As the spacecraft moves to the anti‐subsolar the dayside of the Moon [Saito et al., 2008] and solar wind point of the Moon, the energy of these ions increase by protons entering lunar plasma wake [Nishino et al., 2009a, 600 ∼ 1500eV. This phenomenon occurs at north/south 2009b]. In this paper, we report forward scattered solar wind pole when IMF By component is dominant and negative/ protons that can access mid‐ to low‐latitude region of lunar positive. It is proposed these particles are scattered solar wake observed by Solar Wind Ion Detectors(SWIDs) on CE‐1 wind protons, accelerated by the convection electric field spacecraft. of the solar wind and E × B drift in the ambipolar electric field at the flank of the lunar wake. This mechanism 2. Instrumentation allows a new portion of solar wind protons to enter the ‐ ‐ central lunar wake, and provides a possibility to study [3]CE1 is a three axis stabilized lunar orbiter on a circular the property of proton scattering at the dayside of the polar orbit. Its altitude was 200km before Dec. 2008 and Moon. Citation: Wang, X.‐D., et al. (2010), Acceleration of 100km afterwards, with periods of 127 and 118 minutes, scattered solar wind protons at the polar terminator of the Moon: respectively. To explore the plasma environment of the Moon ’ ‐ is one of its four major scientific objectives [Ouyang et al., Results from Chang E 1/SWIDs, Geophys. Res. Lett., 37, ‐ L07203, doi:10.1029/2010GL042891. 2008]. Therefore one high energy particle detector (HPD) and two identical solar wind ion detectors (SWIDs) were onboard. Each SWID is an electrostatic analyzer (ESA) 1. Introduction with a hemispheric cap. The energy sweeping range is 0.04 ∼ [2] The solar wind interaction with the Moon was once 20 keV/q, divided into 48 channels. From 400eV to 3.2keV thought to be the simplest: the supersonic solar wind passes the energies of channels are distributed exponentially, with an through an airless, non‐magnetized body without disturbance energy resolution (DE/E) of 6.7%. The energy resolution is until the solar wind particles hit the dayside of the body and >30% outside this range. Its field of view(FOV) is 6.7° × stay in the regolith. A plasma cavity—the lunar wake—is 180° FWHM(15° × 12), i.e., a slit 180° wide is divided into −5 2 formed behind the Moon. No magnetosphere, no bow shock. 12 sectors. The geometric factor is 8.1 × 10 · E(i)sr·cm, This simple scenario, however, has been greatly complicated where E(i) is the energy of the ith channel. When operating, by up‐to‐date observations and theories. These aspects the FOV slit of SWIDB looks up to the space and lies in the include surface charging and dust dynamics (Apollo‐era orbit plane; the FOV of SWIDA is perpendicular to current results [e.g., Grobman and Blank, 1969; Walbridge, 1975; orbit plane and also space‐facing. Due to the limitations Freeman and Ibrahim, 1975] recent observations made by on orbit control and power supply, the data accumulation time Lunar Prospector [Halekas et al., 2008], review of dust had been broken into three periods: Nov. 26, 2007 ∼ Feb. 12, dynamics [Colwell et al., 2007]), neutral lunar atmosphere/ 2008; May 15, 2008 ∼ Aug. 10, 2008; Nov. 12, 2008 ∼ exosphere (for a review, see Stern [1999])and the lunar wake Jan. 14, 2009. The total data length is over 3600 hours, [Spreiter et al., 1970; Ogilvie et al., 1996; Kellogg et al., including measurements in the interplanetary space, the 1996; Farrell et al., 1996; Bale et al., 1997; Farrell et al., magnetosheath and the magnetosphere of the Earth. Most of 1997]. However, plasma process near the lunar surface has the time in the solar wind, SWIDB was able to continuously been poorly studied due to the lack of observation. An receive sufficient ion flux to determine its normalized dis- exception was the nonthermal ions originated at the dayside tribution function, while SWIDA could detect the full ion of the Moon observed by Nozomi [Futaana et al., 2003]. This flux once in one revolution. situation has been improved by three Asian lunar missions: 3. Observation

1National Astronomical Observatories, Chinese Academy of Sciences, [4] The data of SWIDs are Energy‐time (E‐t) spectrum Beijing, China. 2 with a time resolution of 10 seconds. Figure 1 shows the Graduate University of the Chinese Academy of Sciences, Beijing, data of SWIDB of orbit 0381∼0382 in the solar wind. At China. ‐ 3National Center for Space Weather, Beijing, China. that time the orbit plane was 4° away from the noon midnight plane. An orbit starts at the north pole. As the spacecraft Copyright 2010 by the American Geophysical Union. moves from the south pole to the north pole over the dayside, 0094‐8276/10/2010GL042891

L07203 1of5 L07203 WANG ET AL.: ACCELERATED PROTONS AT LUNAR TERMINATOR L07203

Figure 1. The observation of orbit 0381 and 0382, when the Moon and the spacecraft were in the interplanetary space. (a) The observation geometry of SWIDB in the selenocentric solar ecliptic (SSE) coordinate system. Sectors in black detected the largest flux of solar wind protons. (b) The location of the spacecraft in the ecliptic plane during observation. (c) The summed E‐t spectrum of all 11 sectors except for the blocked sector 12. Red and pink bars cover the measurements of the solar wind ions and accelerating particles, respectively. (d) The evolution of solar zenith angle(SZA) and selenocentric geographic(SCG) coordinates of the spacecraft. (e) The Bx, By and Bz components of the interplanetary magnetic field(IMF) in GSE coordinate system. IMF data in this study are from ACE/MAG at L1 point with shifted time. Gradient gray boxes cover the periods of the spacecraft in the optical shadow. sector 11∼1 of SWIDB (not shown in Figure 1) successively the Moon, particles with much lower energy than the solar receive incident solar wind ions. Because sector 12 was wind are accelerated smoothly without being heated. blocked due to installation geometry, its data were discarded. [6] We found 139 orbits containing TA events in 1040 The solar wind ions were easily identified (segments covered interplanetary orbits, giving a occurrence rate of 13.5%. The by red bars in Panel C). The spacecraft was in the optical events can be observed at both northern and southern hemi- shadow of the Moon from 2330 to 0012UT and from 0135 to spheres and have an obvious correlation with IMF conditions: 0220UT (gradient gray boxes). northern/southern events took place only if the IMF By [5] Now we report a phenomenon beyond the expectation component was prominent and negative/positive. The starting of the experiment, which is, namely, the acceleration of energy of particles is 200 ∼ 300eV, and the energy gain can particles at the polar terminator of the Moon (TA events exceed 1keV. The direction of incident particles is within the herein). As shown in Figure 1 below the pink bar, a stream of range of 10°∼60° away from the Sun. The flux of acceleration accelerating particles was observed as the spacecraft crossed particles is highly variable across events. In this event, the over the terminator from the dayside to the nightside of the maximum differential flux of TA particles is 10% of the Moon. A zoom‐in version of this event is shown in Figure 2 maximum value measured by SWIDs, but such high flux is as an example. The particles appeared with a starting energy rare. We selected 64 prominent events with flux higher than of 200eV/q, at an SZA of 85°. As the spacecraft moved, the 1000keV−1 ·cm−2 ·s−1 ·sr−1. Among these events, the median energy of particles smoothly increased to 1.3keV (500km/s flux ratio is 1.2% after correction by the WIND/3dp data. for protons) at an ending SZA of 150°, covering 1/6 of an orbit. The energy increase of 1keV along with the SZA was the most representative feature of this phenomenon. The 4. Discussion incident direction of TA particles (red pluses superposed on ∼ [7] Because there was no mass spectrometer onboard, we the angle spectrum) were 20° 50° northward, more northward need to put some restraints on the origin of TA particles. than that of the solar wind ions. This direction distribution Accelerated particles can be solar wind protons, ions origi- was wide and basically invariant with respect to the SZA of nated from the Moon or charged lunar grains. The last one is the spacecraft. The Maxwellian temperature of TA particles ∼ 5 the least possible, for both energy/q and total flux are too kept close to 10 K throughout the acceleration. This event low to produce observed counts. Exospheric ions produced is relatively pronounced, showing a typical pattern: from the either on the lunar surface [Stern, 1999] or in the space dayside of the terminator to mid latitude at the nightside of [Futaana et al., 2008] can be accelerated to the energy range

2of5 L07203 WANG ET AL.: ACCELERATED PROTONS AT LUNAR TERMINATOR L07203

Figure 2. Zoom‐in of the TA event in Figure 1. (a) The E‐t spectra of 5 sectors that detected TA particles. (b) The total flux of all 11 sectors. (c and d) The same as Figures 1d and 1e, respectively. (e) The angle‐time spectrum calculated by integrating differential fluxes over all energy channels. The Y axis is the direction of incident particles in SSE frame. Red numbers are sector numbers from which the counts like oblique stripes come. Red pluses mark the counts from TA particles. The color bar represents differential flux for Figure 2a and directional flux for Figure 2e. in TA events under common IMF conditions [Yokota et al., CE‐1, however, are those with small deflection angles, i.e., 2009; Tanaka et al., 2009]. However, large masses make forward scattered protons. These protons will lose energy first their trajectories almost straight along the accelerating elec- as they move upstream of the CEF (dashed part of particle tric field, which is always perpendicular to the solar wind trajectories in Figure 3). Then, once they pass the apex of velocity. This is inconsistent with the observed angle distri- trajectories, the electric field starts to cause acceleration. As bution. Furthermore, heavy ions do not show any concen- tration in polar terminator regions. Therefore, observed TA particles may contain a small portion of heavy ions but they are not the main part. [8] We here propose that these particles are scattered solar wind protons, like from Saito et al. [2008]. The principal difference is the lower initial energy and the entry to the lunar wake via upstream side of convection electric field (CEF). This mechanism is shown in Figure 3, under the assumption of IMF dominated by By component. In this example, the CEF is along the −Z axis. On the impact to the lunar surface, protons maybe scattered to various directions. Figure 3. A sketch of the interpretation. The frame is in Those with large deflection angles, i.e., back scattered pro- SSE coordinate system. Only the By component of IMF is tons, will be accelerated along the CEF at the dayside of the non‐zero, confining the process in the XZ plane. Curves Moon, in the same manner as reported by Saito et al. [2008]. with arrowheads are trajectories of protons scattered at the Finally these protons will enter deep lunar wake via the same location. A scattered proton is decelerated on the downstream side of the CEF, suggested as Type‐2 entry by dashed segment of its trajectory, and accelerated on the solid Nishino et al. [2009b]. The accelerated protons observed by segment.

3of5 L07203 WANG ET AL.: ACCELERATED PROTONS AT LUNAR TERMINATOR L07203

presents. The apparent rate of reflection in prominent events are higher than observed by Kaguya [Saito et al., 2008], but the median number is less than 1% together with less prominent events. This number is possible, as shown in Figure 4, scattered protons at different latitudes will all contribute to the measured differential flux, in other words, reflected protons can concentrate in the terminator region. It may also come from limited FOV of both instruments, or might imply a possibility of non‐homogeneous scattering of solar wind protons. [10] In this interpretation, effects of electric field at terminator regions due to the transition of surface potential is not considered, and the interpretation is confined in 2‐D frame. For a further study the dynamics of detected protons need to be calculated on a more detailed level.

5. Conclusion

[11] We reported the acceleration of scattered solar wind protons close to the polar terminator and at the night side Figure 4. Trajectory and measurement simulation of dif- of the Moon observed by SWIDs/Chang’E‐1. Particles of ferent scattering latitudes. The electric field is in the XZ ∼200 eV are accelerated from dayside to nightside of the plane of SSE frame. The numbers close to the scattering Moon. When IMF B dominated and was positive/negative, ‐ y points show the latitudes. The E SZA plotting exhibits the the acceleration was observed at the south/north pole. It is energy of particles within the FOV of SWIDB and closer than proposed these particles are forward scattered solar wind 20km of the spacecraft orbit, which are considered detectable. protons, at mid‐ to high‐latitude dayside surface of the Moon. Different colors of simulated spectrum correspond to dif- These protons are decelerated by the CEF and accelerated by ferent latitudes. Blue and red tick marks indicate the SZAs CEF and E × B drift in the ambipolar electric field in the lunar and energies of the least and most energetic particles detected wake. This observation reveals a new portion of protons that in all simulations, respectively. can access the mid latitude lunar plasma wake besides type‐2 entry [Nishino et al., 2009b]. Behavior and effects of these protons need to be evaluated. The energy and direction dis- they enter the near‐flank lunar wake, the ambipolar electric tribution of detected protons will also provide us with the field there and intensified magnetic field will cause E × B drift properties of their reflectors. What is the direction distribution of the protons, leading to further acceleration. The accelera- of surface scattering? How much energy will they lose in tion part of the whole process is identical to that at work in scattering and how the loss depends? These problems will be the Type‐1 entry [Nishino et al., 2009a], with the origin of studied in future work. protons being scattered ones instead of slow ones caused by the thermal motion. [9] Figure 4 shows the particles trajectories and proton [12] Acknowledgments. We thank the ACE team for providing solar wind magnetic field and plasma data. We thank the WIND/3dp team for energy close to the orbit of spacecraft on different scatter providing the solar wind proton spectrum data. This research is supported latitudes. The proton motion is confined in the SSE XZ plane by National High‐Technology Research and Development Program of China for simplicity. The By component of IMF and solar wind (2008AA12A211 and 2008AA12A212) and National Natural Science Foun- speed are identical as in orbit 381 and 382. The energy loss dation of China (40674077). rate is set to be 30%, following [Saito et al., 2008]. Ambipolar electric field in the lunar wake is not considered. Simulated References spectrum agrees with observations very well, the energy Bale, S. D., C. J. Owen, J.‐L. Bougeret, K. Goetz, P. J. Kellogg, R. P. Lepping, range and dependence on SZA are reproduced correctly, R. Manning, and S. J. Monson (1997), Evidence of currents and unstable suggesting the validity of this mechanism. Even though only particle distributions in an extended region around the lunar plasma wake, Geophys. Res. Lett., 24(11), 1427–1430. the protons of the largest flux are simulated, the result reveals Colwell, J. E., S. Batiste, M. Horányi, S. Robertson, and S. Sture (2007), some interesting characteristics. First, given such solar wind Lunar surface: Dust dynamics and regolith mechanics, Rev. Geophys., speed and energy loss on impact, forward and backward 45, RG2006, doi:10.1029/2005RG000184. scattered protons from the same location can not enter the Farrell, W. M., R. J. Fitzenreiter, C. J. Owen, J. B. Byrnes, R. P. Lepping, K. W. Ogilvie, and F. Neubauer (1996), Upstream ULF waves and energetic lunar wake via both sides of the Moon. Second, the higher electrons associated with the lunar wake: Detection of precursor activity, latitude the scatter point locates, the higher starting energy Geophys. Res. Lett., 23(10), 1271–1274. will be observed, whereas even the scatter latitude reaches as Farrell, W. M., M. L. Kaiser, and J. T. Steinberg (1997), Electrostatic insta- bility in the central lunar wake: A process for replenishing the plasma high as 88°, the starting energy is around 200eV, indicating a void?, Geophys. Res. Lett., 24(9), 1135–1138. sufficient deceleration due to the convection electric field. Freeman, J. W., and M. Ibrahim (1975), Lunar electric fields, surface poten- This much larger energy loss observed by CE‐1 than those tial and associated plasma sheaths, Earth Moon Planets, 14, 103–114. ‐ Futaana, Y., S. Machida, Y. Saito, A. Matsuoka, and H. Hayakawa (2003), observed by Kaguya and Chandrayaan 1 may result from the Moon‐related nonthermal ions observed by Nozomi: Species, sources, and higher orbit of CE‐1(200km), allowing an extra energy loss generation mechanisms, J. Geophys. Res., 108(A1), 1025, doi:10.1029/ of ∼100eV when electric field strength is ∼1mV/m. For a 2002JA009366. convection electric field not in the XZ plane this mechanism Futaana, Y., S. Nakano, M. Wieser, and S. Barabash (2008), Energetic neutral atom occultation: New remote sensing technique to study the lunar still works, as long as a Z component towards the Moon exosphere, J. Geophys. Res., 113, A11204, doi:10.1029/2008JA013356.

4of5 L07203 WANG ET AL.: ACCELERATED PROTONS AT LUNAR TERMINATOR L07203

Grobman, W. D., and J. L. Blank (1969), Electrostatic potential distribution (KAGUYA), Geophys. Res. Lett., 35, L24205, doi:10.1029/ of the sunlit lunar surface, J. Geophys. Res., 74(16), 3943–3951. 2008GL036077. Halekas, J. S., G. T. Delory, R. P. Lin, T. J. Stubbs, and W. M. Farrell Spreiter, J. R., M. C. Marsh, and A. L. Summers (1970), Hydromagnetic (2008), Lunar prospector observations of the electrostatic potential of aspects of solar wind flow past the Moon, Cosmic Electrodyn., 1(5), 5–50. the lunar surface and its response to incident currents, J. Geophys. Stern, S. A. (1999), The lunar atmosphere: History, status, current pro- Res., 113, A09102, doi:10.1029/2008JA013194. blems, and context, Rev. Geophys., 37(4), 453–491. Kellogg, P. J., K. Goetz, S. J. Monson, J.‐L. Bougeret, R. Manning, and Tanaka, T., et al. (2009), First in situ observation of the Moon‐originating M. Kaiser (1996), Observations of plasma waves during a traversal of the ions in the Earth’s Magnetosphere by MAP‐PACE on SELENE Moon’s wake, Geophys. Res. Lett., 23(10), 1267–1270. (KAGUYA), Geophys. Res. Lett., 36, L22106, doi:10.1029/ Nishino, M. N., et al. (2009a), Pairwise energy gain‐loss feature of solar 2009GL040682. wind protons in the near‐Moon wake, Geophys. Res. Lett., 36,L12108, Walbridge, E. (1975), Lunar photoelectron layer dynamics, Earth Moon doi:10.1029/2009GL039049. Planets, 14, 115–121. Nishino, M. N., et al. (2009b), Solar‐wind proton access deep into the near‐ Yokota, S., et al. (2009), First direct detection of ions originating from the Moon wake, Geophys. Res. Lett., 36, L16103, doi:10.1029/ Moon by MAP‐PACE IMA onboard SELENE (KAGUYA), Geophys. 2009GL039444. Res. Lett., 36, L11201, doi:10.1029/2009GL038185. Ogilvie, K. W., J. T. Steinberg, R. J. Fitzenreiter, C. J. Owen, A. J. Lazarus, W. M. Farrell1, and R. B. Torbert (1996), Observations of the lunar plasma W. Bian, C.‐L. Li, J.‐J. Liu, J.‐Z. Liu, C. Lü, Z.‐Y. Ouyang, Y. Su, wake from the WIND spacecraft on December 27, 1994, Geophys. Res. M. Wang, X.‐D. Wang, W.‐B. Wen, H.‐B. Zhang, Y.‐L. Zou, and W. Zuo, Lett., 23, 1255–1258. ‐ ’ ‐ National Astronomical Observatories, Chinese Academy of Sciences, Ouyang, Z. Y., et al. (2008), Preliminary scientific results of Chang E 1 Beijing, 100012, China. ([email protected]) lunar orbiter: Based on payloads detection data in the first phase, Chin. J.‐S. Wang, National Center for Space Weather, Beijing, 100081, China. J. Space Sci., 28(5), 361–369. Saito, Y., et al. (2008), Solar wind proton reflection at the lunar surface: Low energy ion measurement by MAP‐PACE onboard SELENE

5of5