Evidence of plasma lunar tide in the Earth-Moon

system
Chao Xiao 
Chinese Ministry of Education Key Laboratory of Particle Physics and Particle Irradiation, Institute of
Space Sciences, Shandong University
Fei He 
Institute of Geology and Geophysics, Chinese Academy of Sciences https://orcid.org/0000-0003-0542-
2686
Quanqi Shi  (  sqq@pku.edu.cn )
Shandong University
Wenlong Liu 
Beihang University https://orcid.org/0000-0001-7991-5067
Anmin Tian 
Chinese Ministry of Education Key Laboratory of Particle Physics and Particle Irradiation, Institute of
Space Sciences, Shandong University
Ruilong Guo 
Chinese Ministry of Education Key Laboratory of Particle Physics and Particle Irradiation, Institute of
Space Sciences, Shandong University
C. Yue 
Institute of Space Physics and Applied Technology, Peking University https://orcid.org/0000-0001-
9720-5210
Xu-Zhi Zhou 
Peking University
Yong Wei 
Institute of Geology and Geophysics, Chinese Academy of Sciences, China. https://orcid.org/0000-
0001-7183-0229
Ian Rae 
Northumbria University
Degeling Alexander William 
Shandong Provincial Key Laboratory of Optical Astronomy and Solar-Terrestrial Environment, Institute
of Space Sciences, Shandong University at Weihai
Vassilis Angelopoulos 
University of California Los Angeles https://orcid.org/0000-0001-7024-1561
Emmanuel Masongsong 
 Department of Earth, Planetary and Space Sciences, Institute of Geophysics and Space Physics,
University of California, Los Angeles
Ji Liu 
Department of Physics, University of Alberta
Qiu-Gang Zong 
School of Earth and Space Sciences, Peking University
Suiyan Fu 
School of Earth and Space Sciences, Peking University
Zuyin Pu 
Peking University
Xiaoxin Zhang 
Key Laboratory of Space Weather, National Center for Space Weather, China Meteorological
Administration
Tieyan Wang 
School of Space and Environment, Beihang University
Huizi Wang 
Shandong Provincial Key Laboratory of Optical Astronomy and Solar-Terrestrial Environment, Institute
of Space Sciences, Shandong University
Zhao Zhang 
School of Space and Environment, Beihang University

Letter

Keywords:

Posted Date: April 6th, 2022

DOI: https://doi.org/10.21203/rs.3.rs-1474794/v1

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.  
Read Full License
 1 Evidence of plasma lunar tide in the Earth-Moon system
2

3 Chao Xiao1,#, Fei He2,#, Quanqi Shi1,*, Wenlong Liu3,4,*, Anmin Tian1, Ruilong Guo1, Chao Yue5,
4 Xuzhi Zhou5, Yong Wei2, I. Jonathan Rae6, Alexander W. Degeling1, Vassilis Angelopoulos7,
5 Emmanuel V. Masongsong7, Ji Liu8, Qiugang Zong5, Suiyan Fu5, Zuyin Pu5, Xiaoxin Zhang9,
6 Tieyan Wang10, Huizi Wang1, Zhao Zhang3,4

1
7 Chinese Ministry of Education Key Laboratory of Particle Physics and Particle Irradiation,
8 Shandong Provincial Key Laboratory of Optical Astronomy and Solar-Terrestrial Environment,
9 Institute of Space Sciences, Shandong University; 180 Wenhua Xilu, Weihai, Shandong, China.
2
10 Key Laboratory of Earth and Planetary Physics, Institute of Geology and Geophysics, Chinese
11 Academy of Sciences; 19 Beitucheng Western Road, Chaoyang District, Beijing, China.
3
12 School of Space and Environment, Beihang University; No.9 South 3rd St., Shahe Higher
13 Education Park, Changping District, Beijing, China.
4
14 Key Laboratory of Space Environment monitoring and Information Processing of MIIT; No.9
15 South 3rd St., Shahe Higher Education Park, Changping District, Beijing, China.
5
16 Institute of Space Physics and Applied Technology, Peking University; No.5, Yiheyuan Road,
17 Haidian District, Beijing, China.
6
18 Department of Maths, Physics and Electrical Engineering, Northumbria University; 12
19 Newcastle upon Tyne, UK.
7
20 Department of Earth, Planetary and Space Sciences, Institute of Geophysics and Space Physics,
21 University of California, Los Angeles; 405 Hilgard Avenue, Los Angeles, CA, USA
8
22 Department of Physics, University of Alberta; Edmonton T6G2R3, Canada.
9
23 Key Laboratory of Space Weather, National Center for Space Weather, China Meteorological
24 Administration; 46 Zhongguancun South Street, Haidian District, Beijing, China.
10
25 School of Earth Science, Chenggong Campus, Yunnan University; East Outer Ring Road,
26 Chenggong District, Kunming, Yunnan, China.
27

28 *Corresponding author. Email: sqq@sdu.edu.cn; liuwenlong@buaa.edu.cn

#
29 These two authors contribute equally

1
30
31 Abstract:

32 Tides are universal, significantly affecting spatially distributed systems, from planetary to

33 galactic scales. In the Earth-Moon system, lunar tides have been studied in the solid Earth, liquid

34 ocean, and neutral gas-dominated atmosphere (including the ionosphere)1–12. However, whether a

35 lunar tide effect exists over wider plasma dominated regions has remained unexplored until now.

36 Here we show direct evidence of a lunar tide signal in the plasmasphere, a typical cold “plasma

37 ocean” in the Earth-Moon space13,14, by analyzing variations in its boundary location over the past

38 four decades from multi-satellite observations. This signal possesses distinct diurnal (and monthly)

39 periodicities, surprisingly different from the semidiurnal (and semimonthly) variations dominant

40 in the atmosphere, ocean, and solid Earth15–19, with tidal surface perturbations of up to ~800 km

41 via disturbing the electric field. These results demonstrate the importance of lunar tidal effects in

42 plasma dominated regions of Earth-moon space, opening new perspectives for understanding

43 Moon-atmosphere-magnetosphere coupling process through gravity and electromagnetic forces,

44 and has implications for tidal interactions in other two-body celestial systems.

45

2
46 Introduction

47 Tides are universal phenomena and often play essential roles in planetary and galactic systems

48 wherever gradients in gravitational attraction are important20–22. As the Earth’s sole natural satellite,

49 the Moon and its gravitational interaction with Earth have attracted extensive research and

50 curiosity over several hundred years23. Periodic lunar tidal effects have been observed in the

51 Earth’s crust, oceans, near-ground geomagnetic field, atmosphere, and ionosphere1–12. These lunar

52 tides mainly have semidiurnal (and semimonthly) periods15–19, and are of fundamental importance

53 to conditions on our planet. For example, the Earth’s crustal tide can trigger seismic24–26and

54 volcanic activities27, the ocean tide can influence the flow of heat from equatorial to polar regions28

55 and the evolution of primordial terrestrial species from aquatic life29. Atmospheric tides have a

56 global impact on rainfall30, and ionospheric tides can affect radio transmission and low Earth orbit

57 satellite altitude7,31. If we follow the four states of matter, they progress from solid, liquid, and gas,

58 to plasma which is an influential component of the Moon-Earth space environment. In the past,

59 lunar tides were mainly found to affect the first three states: solid Earth tides, liquid ocean tides,

60 and neutral gas-dominated atmospheric tides. Whether lunar tides can influence the fourth state of

61 matter (plasma) dominated regions, which are much wider in space, has remained until now

62 unexplored.

63 Obviously, the Earth’s plasmasphere is the most ideal and representative plasma-dominated

64 place to study the existence of lunar tide in plasma, because its basic properties have been studied

65 very extensively and there are massive amounts of observational data. The plasmasphere is a

66 collisionless magnetized plasma region that extends along geomagnetic field lines from the upper

67 ionosphere (it’s source region), filling a roughly torus-shaped volume in near-Earth space. It is

68 composed of cold (1~2 eV) and dense (102~104 cm-3) plasma with quasi equal numbers of electrons

69 and ions (~80% H+, ~10–20% He+ and ~5–10% O+), and plays a key role in particle exchange and

3
70 storage within the magnetosphere13,14. Given its cold, dense plasma properties, the plasmasphere

71 can be regarded as a “plasma ocean”, and the plasmapause represents the “surface” of this ocean

72 because of the dramatic change of plasma properties across this boundary. The location of the

73 plasmapause is an important parameter in the study of plasma dynamics in inner magnetosphere32–

35
74 . Previous studies have shown that the plasmapause location is the reflection of the dynamics of

75 the entire cold plasma which can strongly affect energetic particle distributions in the inner

76 magnetosphere by significantly influencing the excitation and propagation of electromagnetic

77 waves, and subsequently impact radiation belt and ring current dynamics36–38. Hence, in order to

78 explore the existence of a lunar tidal effect in this cold plasma ocean, it is natural to examine these

79 “sea surface” variations that may be associated with the lunar cycle. Figs. 1a–1b show two typical

80 plasmaspheric satellite images obtained by the Extreme Ultraviolet (EUV) camera onboard

81 NASA’s IMAGE spacecraft in polar perspective and Extreme Ultraviolet Camera (EUVC) aboard

82 China's Chang'e-3 (CE-3) spacecraft in lunar perspective, respectively, with corresponding

83 plasmapause outlines indicated by white curves. The findings reported in this article demonstrate

84 a significant lunar tidal effect in the plasmapause location, providing a causal link by which the

85 Moon exerts influence on magnetospheric dynamics. This expands our understanding of Earth-

86 Moon interactions in a direction that has not been previously considered, and provides important

87 clues for future investigations in wider regions and two-body celestial systems, including other

88 planetary systems in our solar system and beyond.

89 Observation of the plasma lunar tidal signal

90 An extremely large database of plasmapause positions35 determined by 50,778 plasmapause

91 crossings from 10 satellite missions is used in this study (the largest database compiled to date,

92 detailed in Methods 1). The survey time period is from November 1977 to December 2015,

93 covering almost four solar cycles (21, 22, 23 and 24). This provides a unique opportunity to study

4
 94 lunar influences on plasmapause position by eliminating external factors such as effects from solar

95 activity. In this investigation, 35,982 (~71% of the total) plasmapause crossings under low-activity

96 geomagnetic conditions (Dst > -50 nT, AE < 300 nT, and Kp ≤ 3) were selected to remove solar

97 wind effects and minimize statistical uncertainties. To extract the lunar tide signal, a new database

98 of plasmapause perturbations categorized by lunar phase (LP) was compiled and used for the

99 following investigations (see Methods 2).

100 Figs. 1-2 show conclusive evidence of a lunar tidal effect on the plasmapause location (Details

101 of statistical uncertainty considerations and data processing can be seen in Methods 3). For

102 convenience in this study, LP is defined as the magnetic local time (MLT) of lunar position. Figs.

103 1c–1f show perturbations in plasmapause position as a function of MLT) when LP is equal to 0, 6,

104 12 and 18 (corresponding to full moon, third quarter moon, new moon, and first quarter moon),

105 respectively. The 95% confidence interval (CI) error bars are used here. These panels reveal that

106 the high tide peaks (marked by green dotted lines) of the perturbations progress regularly with LP.

107 A corresponding schematic is shown in Fig. 1g, which illustrates the results from Figs. 1c-1f that

108 the MLT (longitude) of the high tide bulges is ~ 6 hours (90°) ahead of the LP. This is further

109 demonstrated below in detail. Fig. 2a shows the distribution of the perturbations of plasmapause

110 position in the MLT-LP frame, which clearly shows the existence of lunar tides in the

111 plasmasphere, herein after referred to as Lunar Tidal Wave in the Plasmapause (LTWP). Because

112 there is only one high tide and one low tide for each MLT or LP, the period of LTWP is diurnal

113 and monthly, which is different from most of the previously observed lunar tides in other regions

114 of the Earth system, with predominantly semidiurnal and semimonthly periods39. This result shows

115 that the influence of the Moon on each quadrant of the plasmasphere is different. To highlight the

116 primary relationship, the LTWP under each LP is rescaled by min-max normalization, and the

117 corresponding distribution is shown in Fig. 2b. It clearly shows that the MLT of the high tide

5
118 (MLTHT) is about 6 hours ahead of the LP (longitude difference ~90°), i.e., they have a well-

119 defined linear relationship which is expressed by MLTHT = (0.956 ± 0.051)  LP ＋ 6.336 ±

120 1.29. Fig. 2c depicts lunar local time (LLT) variations of the normalized perturbations in

121 plasmapause position at each LP. As shown in Fig. 2a, the amplitude of the perturbations is ±0.13

122 Earth radii (RE). These three panels clearly show the features of the lunar tide in the plasmasphere:

123 the high tide (maximum of about 0.12 RE) occurs at the dusk side of LLT while the low tide

124 (minimum of about -0.14 RE) occurs at the dawn side of LLT as illustrated in Fig. 2d. In order to

125 rule out the possibility of data anomalies and ensure the reliability of the results, we divided the

126 data into two sub-datasets and still obtained the same tidal signal (Supplementary Information 1).

127 A movie showing tidal changes at the plasmapause can be seen in Supplementary Information 2.

128 It is worth noting that because the plasmapause location was found to be modulated by the solar

129 rotation (e.g., corotating interaction regions) with a period of about 27 days40, one could reasonably

130 suspect that the signal we are observing is in fact modulated by the Sun. However, our use of low

131 geomagnetic activity data with a long-term average and strong correlation of the high tide of

132 LTWP with the LP rules out the possibility of the solar rotation effect for this signal.

133 Cause of this plasma lunar tidal signal

134 The essential question now is: how does such a lunar tide with diurnal and monthly periods

135 occur in the plasmasphere? It is known that the plasmapause position is principally determined by

136 the interaction between the corotation electric field (Ecoro) and the magnetospheric convection

137 electric field (Econv). Econv is controlled by the changing solar wind and geomagnetic activity,

138 whereas Ecoro is controlled by the co-rotation of the inner magnetosphere with the Earth and is

139 relatively stable. Since the plasmaspheric plasma is trapped on closed magnetic field lines and co-

140 rotates with the Earth, the last closed equipotential line of the total electric field (also a streamline

141 of the plasma flow) corresponds to the plasmapause. Therefore, the electric field is considered as

6
142 a probable factor. Fig. 3a shows LLT variations of the radial component of the electric field (Er)

143 measured by the Van Allen Probes (VAP), which varies from 0.57 to 0.69 mV/m (see Methods 5).

144 The maximum (minimum) of Er occurs at the dawn (dusk) side of LLT, which is opposite to LTWP

145 (blue line in Fig. 3a). In addition, there is a strong negative correlation between LTWP and Er with

146 a correlation coefficient of -0.94.

147 To evaluate whether such a variation of electric field could reproduce the observed LTWP, a

148 cold plasma convection model, an electric field model consisting of Ecoro, Econv parametrized by

149 Kp and disturbance of electric field E, is used. Fig. 3c shows a contour plot of equipotential

150 calculated from the model for Kp = 3 with the plasmapause highlighted by the black dashed line.

151 By subtracting the baseline of the radial electric field (0.63 mV/m, which agrees well with the

152 averaged corotation electric field of 0.61 mV/m between L-shells of 3 ~ 6 RE) in Fig. 3a, we obtain

153 the disturbed radial electric field (Er) as shown by the asterisks in Fig. 3b. The disturbed radial

154 electric field is fitted by the function Er = 0.0155 ± 0.0022  cos(0.2618 ± 0.0002  LLT) ＋

155 0.0578 ± 0.0012  sin(0.2618 ± 0.0002  LLT) with a correlation coefficient of 0.99. We then

156 superimpose ~38% Er (see Methods 6) as a function of LP onto the convection-corotation electric

157 field model with Kp = 3, and classify corresponding plasmapause positions using the last closed

158 equipotential line. The modelled perturbations of the plasmapause are shown in Figs. 3d–3e. These

159 results demonstrate that a simple electric field model can reproduce this tidal phenomenon quite

160 well: the correlation coefficient between observed and simulated perturbations (normalized

161 perturbations) of plasmapause positions is about 0.61 (0.73). Further, the perturbations of

162 plasmapause positions (-0.13 ~ 0.12 RE) obtained from the model agree well with the observations

163 (-0.14 ~ 0.12 RE). Thus, the perturbations of plasmapause position are likely to be due to the

164 disturbed radial electric field. Understanding the causal link between the lunar phase and the

165 perturbed electric field is a subject of ongoing research. One possibility is that the neutral winds
7
166 in the ionosphere, which are modulated by the lunar phase, can generate an electric potential

167 difference. Under ideal MHD conditions, magnetic field lines in the magnetosphere are

168 equipotential, hence this potential difference maps along field lines threading the plasmasphere

169 and contribute to the radial electric field that modulates the plasmapause position. Preliminary

170 results of model calculations supporting this hypothesis, will be further developed for publication

171 in the near future.

172 Summary and Discussion

173 In summary, we explored in this paper the existence of a lunar plasma tidal effect in the Earth-

174 Moon space. By analyzing an extensive database containing an unprecedented quantity of

175 plasmapause location observations, we found that a lunar tide signal exists in the plasmasphere

176 with diurnal (and monthly) periodicities. This plasma tide contrasts with the lunar tides in lower

177 altitude regions in the first three states of matter, which have semidiurnal (and semimonthly)

178 periodicities. The low (high) tide of this signal appears at the dawn (dusk) side of LLT. Van Allen

179 Probe observations of the radial electric field near the equatorial plane are shown to exhibit

180 modulations synchronized with lunar phase. Using a simple cold plasma convection model, we

181 show that this modulation provides the capability of generating the observed plasmaspheric lunar

182 tidal signals. As for how the moon phase adjusts the radial electric field, we suspect that the Moon

183 may affect the radial electric field by affecting the upper atmosphere. This result further advances

184 our understanding of the Moon-atmosphere-magnetosphere coupling process through gravitation,

185 with important participation of electromagnetic force.

186 Our discovery of this plasma tidal effect with distinct characteristics may indicate a fundamental

187 interaction mechanism in the Earth-Moon system that has not been previously considered.

188 Furthermore, reflected by this plasma tide, the plasma flow in the entire Moon-Earth space exists

189 as a persistent background variation of the magnetosphere and can modulate the Moon-Earth space

8
190 dynamics constantly, although these observed perturbations caused by the lunar tide are often

191 relatively small compared to those arising from solar activity. Since this plasma tide effect appears

192 to be predictably fundamental, it can be expected not only in the plasmasphere, but in a much

193 wider region. For example, plasmapause surface waves can alter energy transport from the

194 magnetosphere to the upper atmosphere41, while whistler mode chorus and hiss waves, and also

195 EMIC waves near the plasmapause contribute greatly to electron acceleration and loss in the Van

196 Allen radiation belt42,43. We suspect that the observed plasma tide may subtly affect the distribution

197 of energetic radiation particles, which are a well-known hazard to space-based infrastructure and

198 human activities in space. It is therefore worthwhile to look for evidence of this effect in future

199 studies, for example by checking for correlations of variations in the distribution of ‘Zebra

200 stripes’44 with lunar phase, since it has been suggested to be formed by a weak electric field

201 independent of corotation45 while our observed electric field modulated by this lunar plasma tide

202 may contribute. In addition, we have recently found in ongoing work that magnetic fields at

203 geosynchronous orbit are also clearly (although somewhat weakly) modulated by the lunar phase,

204 which indicates that lunar tides have greater impact than we previously known, and we suspect it

205 could even be observed in the magnetotail. Whether the magnetic field or plasma lunar tides seen

206 in space are related to the Earth’s crustal and oceanic tide46 is also a question worthy of discussion.

207 The configuration and structure of Earth’s cold plasma in relation to the magnetosphere is not

208 unique to the Earth, and similar structures have been found in other planets47–49 and astrophysical

209 objects50. Here, we can summarize that the three fundamental elements necessary for this plasma

210 tidal signal are the existence of two-body celestial system, plasma, and magnetic field. Because

211 the planetary environments in the stellar system that meet all these conditions are very common,

212 this plasma tide may be observed universally throughout the cosmos. Therefore, the finding of this

213 lunar tidal effect in the plasmasphere not only extends our knowledge of the Earth-Moon system,

9
214 but can also open new perspectives for further studies of tidal interactions in other planetary and

215 larger scale systems.

216
217 References

218 1. Matsushita, S. Lunar Tides in the Ionosphere. in Geophysik III / Geophysics III. Handbuch

219 der Physik / Encyclopedia of Physics (ed. Bartels, J.) 547–602 (Springer, 1967).

220 doi:10.1007/978-3-642-46082-1_2.

221 2. Forbes, J. M., Zhang, X., Bruinsma, S. & Oberheide, J. Lunar semidiurnal tide in the

222 thermosphere under solar minimum conditions. J. Geophys. Res. Sp. Phys. 118, 1788–

223 1801 (2013).

224 3. Li, N. et al. Responses of the Ionosphere and Neutral Winds in the Mesosphere and Lower

225 Thermosphere in the Asian‐Australian Sector to the 2019 Southern Hemisphere Sudden

226 Stratospheric Warming. J. Geophys. Res. Sp. Phys. 126, (2021).

227 4. Owolabi, C. et al. Investigation on the Variability of the Geomagnetic Daily Current

228 During Sudden Stratospheric Warmings. J. Geophys. Res. Sp. Phys. 124, 6156–6172

229 (2019).

230 5. Niu, X. et al. Lunar tidal winds in the mesosphere over Wuhan and Adelaide. Adv. Sp.

231 Res. 36, 2218–2222 (2005).

232 6. Paulino, A. R., Batista, P. P. & Batista, I. S. A global view of the atmospheric lunar

233 semidiurnal tide. J. Geophys. Res. Atmos. 118, 13128–13139 (2013).

234 7. Pedatella, N. M. & Forbes, J. M. Global structure of the lunar tide in ionospheric total

235 electron content. Geophys. Res. Lett. 37, (2010).

10
236 8. Sakazaki, T. & Hamilton, K. Discovery of a lunar air temperature tide over the ocean: a

237 diagnostic of air-sea coupling. npj Clim. Atmos. Sci. 1, 25 (2018).

238 9. Stening, R. J., Tsuda, T. & Nakamura, T. Lunar tidal winds in the upper atmosphere over

239 Jakarta. J. Geophys. Res. Sp. Phys. 108, (2003).

240 10. Yamazaki, Y., Richmond, A. D. & Yumoto, K. Stratospheric warmings and the

241 geomagnetic lunar tide: 1958-2007. J. Geophys. Res. Sp. Phys. 117, (2012).

242 11. Zhang, J. T., Forbes, J. M., Zhang, C. H., Doornbos, E. & Bruinsma, S. L. Lunar tide

243 contribution to thermosphere weather. Sp. Weather 12, 538–551 (2014).

244 12. Fejer, B. G. et al. Lunar-dependent equatorial ionospheric electrodynamic effects during

245 sudden stratospheric warmings. J. Geophys. Res. Sp. Phys. 115, (2010).

246 13. Lemaire, J. F., Gringauz, K. I., Carpenter, D. L. & Bassolo, V. The Earth’s Plasmasphere.

247 (Cambridge University Press, 1998). doi:10.1017/CBO9780511600098.

248 14. Zhou, X. et al. On the Formation of Wedge‐Like Ion Spectral Structures in the Nightside

249 Inner Magnetosphere. J. Geophys. Res. Sp. Phys. 125, (2020).

250 15. Mo, X. H. & Zhang, D. H. Lunar Tidal Modulation of Periodic Meridional Movement of

251 Equatorial Ionization Anomaly Crest During Sudden Stratospheric Warming. J. Geophys.

252 Res. Sp. Phys. 123, 1488–1499 (2018).

253 16. Sabaka, T. J., Tyler, R. H. & Olsen, N. Extracting ocean‐generated tidal magnetic signals

254 from Swarm data through satellite gradiometry. Geophys. Res. Lett. 43, 3237–3245

255 (2016).

11
256 17. Park, J., Lühr, H., Kunze, M., Fejer, B. G. & Min, K. W. Effect of sudden stratospheric

257 warming on lunar tidal modulation of the equatorial electrojet. J. Geophys. Res. Sp. Phys.

258 117, (2012).

259 18. Pedatella, N. M., Liu, H.-L., Richmond, A. D., Maute, A. & Fang, T.-W. Simulations of

260 solar and lunar tidal variability in the mesosphere and lower thermosphere during sudden

261 stratosphere warmings and their influence on the low-latitude ionosphere. J. Geophys.

262 Res. Sp. Phys. 117, (2012).

263 19. Liu, J., Zhang, D., Hao, Y. & Xiao, Z. Multi‐instrumental Observations of the Quasi‐16‐

264 Day Variations From the Lower Thermosphere to the Topside Ionosphere in the Low‐

265 Latitude Eastern Asian Sector During the 2017 Sudden Stratospheric Warming Event. J.

266 Geophys. Res. Sp. Phys. 125, (2020).

267 20. Renaud, F., Boily, C. M., Naab, T. & Theis, C. FULLY COMPRESSIVE TIDES IN

268 GALAXY MERGERS. Astrophys. J. 706, 67–82 (2009).

269 21. Ivanov, P. B. & Papaloizou, J. C. B. Dynamic tides in rotating objects: orbital

270 circularization of extrasolar planets for realistic planet models. Mon. Not. R. Astron. Soc.

271 376, 682–704 (2007).

272 22. Fouchard, M., Froeschlé, C., Valsecchi, G. & Rickman, H. Long-term effects of the

273 Galactic tide on cometary dynamics. Celest. Mech. Dyn. Astron. 95, 299–326 (2006).

274 23. Cartwright, D. E. Tides: a scientific history. (Cambridge Univ. Press, 1999).

275 24. Cochran, E. S., Vidale, J. E. & Tanaka, S. Earth Tides Can Trigger Shallow Thrust Fault

276 Earthquakes. Science (80-. ). 306, 1164–1166 (2004).

277 25. Heaton, T. H. Tidal Triggering of Earthquakes. Geophys. J. Int. 43, 307–326 (1975).

12
278 26. Tolstoy, M., Vernon, F. L., Orcutt, J. A. & Wyatt, F. K. Breathing of the seafloor: Tidal

279 correlations of seismicity at Axial volcano. Geology 30, 503 (2002).

280 27. Kasahara, J. Tides, Earthquakes, and Volcanoes. Science (80-. ). 297, 348–349 (2002).

281 28. Green, J. A. M. & Huber, M. Tidal dissipation in the early Eocene and implications for

282 ocean mixing. Geophys. Res. Lett. 40, 2707–2713 (2013).

283 29. Palmer, J. D. The biological rhythms and clocks of intertidal animals. (Oxford University

284 Press, 1995).

285 30. Kohyama, T. & Wallace, J. M. Rainfall variations induced by the lunar gravitational

286 atmospheric tide and their implications for the relationship between tropical rainfall and

287 humidity. Geophys. Res. Lett. 43, 918–923 (2016).

288 31. Wu, T., Liu, J., Lin, C. & Chang, L. C. Response of Ionospheric Equatorial Ionization

289 Crests to Lunar Phase. Geophys. Res. Lett. 47, (2020).

290 32. Liu, W. et al. Poloidal ULF wave observed in the plasmasphere boundary layer. J.

291 Geophys. Res. Sp. Phys. 118, 4298–4307 (2013).

292 33. Liu, X. et al. Dynamic plasmapause model based on THEMIS measurements. J. Geophys.

293 Res. Sp. Phys. 120, (2015).

294 34. He, F. et al. A new solar wind‐driven global dynamic plasmapause model: 2. Model and

295 validation. J. Geophys. Res. Sp. Phys. 122, 7172–7187 (2017).

296 35. Zhang, X. et al. A new solar wind‐driven global dynamic plasmapause model: 1. Database

297 and statistics. J. Geophys. Res. Sp. Phys. 122, 7153–7171 (2017).

13
298 36. Li, X., Baker, D. N., O’Brien, T. P., Xie, L. & Zong, Q. G. Correlation between the inner

299 edge of outer radiation belt electrons and the innermost plasmapause location. Geophys.

300 Res. Lett. 33, L14107 (2006).

301 37. Cao, J. B. et al. First results of low frequency electromagnetic wave detector of TC-

302 2/Double Star program. Ann. Geophys. 23, 2803–2811 (2005).

303 38. Lorentzen, K. R., Blake, J. B., Inan, U. S. & Bortnik, J. Observations of relativistic

304 electron microbursts in association with VLF chorus. J. Geophys. Res. Sp. Phys. 106,

305 6017–6027 (2001).

306 39. Forbes, J. M. & Zhang, X. Lunar Tide in the F Region Ionosphere. J. Geophys. Res. Sp.

307 Phys. 124, 7654–7669 (2019).

308 40. Thaller, S. A. et al. Solar Rotation Period Driven Modulations of Plasmaspheric Density

309 and Convective Electric Field in the Inner Magnetosphere. J. Geophys. Res. Sp. Phys. 124,

310 1726–1737 (2019).

311 41. He, F. et al. Plasmapause surface wave oscillates the magnetosphere and diffuse aurora.

312 Nat. Commun. 11, 1668 (2020).

313 42. Summers, D., Ni, B. & Meredith, N. P. Timescales for radiation belt electron acceleration

314 and loss due to resonant wave-particle interactions: 2. Evaluation for VLF chorus, ELF

315 hiss, and electromagnetic ion cyclotron waves. J. Geophys. Res. Sp. Phys. 112, (2007).

316 43. Miyoshi, Y. et al. Precipitation of radiation belt electrons by EMIC waves, observed from

317 ground and space. Geophys. Res. Lett. 35, L23101 (2008).

318 44. Ukhorskiy, A. Y. et al. Rotationally driven ‘zebra stripes’ in Earth’s inner radiation belt.

319 Nature 507, 338–340 (2014).

14
320 45. Liu, Y., Zong, Q.-G., Zhou, X.-Z., Foster, J. C. & Rankin, R. Structure and evolution of

321 electron “zebra stripes” in the inner radiation belt. J. Geophys. Res. Sp. Phys. 121, 4145–

322 4157 (2016).

323 46. Gallego, A. et al. Tidal modulation of continuous nonvolcanic seismic tremor in the Chile

324 triple junction region. Geochemistry, Geophys. Geosystems 14, 851–863 (2013).

325 47. Nimmo, F. & Pappalardo, R. T. Diapir-induced reorientation of Saturn’s moon Enceladus.

326 Nature 441, 614–616 (2006).

327 48. Lainey, V., Arlot, J.-E., Karatekin, Ö. & Van Hoolst, T. Strong tidal dissipation in Io and

328 Jupiter from astrometric observations. Nature 459, 957–959 (2009).

329 49. Bagheri, A., Khan, A., Efroimsky, M., Kruglyakov, M. & Giardini, D. Dynamical

330 evidence for Phobos and Deimos as remnants of a disrupted common progenitor. Nat.

331 Astron. 5, 539–543 (2021).

332 50. Bagenal, F. GIANT PLANET MAGNETOSPHERES. Annu. Rev. Earth Planet. Sci. 20,

333 289–328 (1992).

334

335

15
336

337
338 Fig. 1. Overview of the plasma lunar tide observation in the Earth-Moon space. (a) An

339 IMAGE EUV image taken at 2000-06-26 14:14 UT viewing from the Earth’s north pole. (b) A

340 CE-3 EUVC image obtained at 2014-04-21 13:01 UT viewing from the Moon. (c-f) The perturbed

341 plasmapause radial location as a function of MLT at different lunar phases (LP=0, 6, 12 and 18,

342 corresponding to full moon, third quarter moon, new moon, and first quarter moon, respectively).

343 The high tide peaks are marked by green dotted lines. (g) Illustration of “high tide” plasmaspheric

344 bulges at the four LPs shown in (c-f). The light-yellow background represents the unperturbed

345 plasmasphere and the different colored bulges (i.e., high tides) correspond to high tides at different

346 LPs.

347
348
349

16
350

351
352 Fig. 2. The characteristics of Lunar Tidal Wave in the Plasmapause. (a) Radial perturbations

353 of plasmapause positions in the MLT-LP frame (MLT=0:2:24 and LP=0:2:24). (b) Normalization

354 of (a) in the MLT-LP frame. (c) Variations of the normalized plasmapause perturbations as a

355 function of lunar local time (LLT) at different LPs. LLT is defined as follows: LLT=12-LP+MLT,

356 if LLT<0, then add 24 to LLT and if LLT>24, then subtract 24 from LLT. LLT=12 corresponds

357 to lunar noon. Note that the MLT, LP and LLT variables all refer to locations and these definitions

358 and relationships can be seen in Methods 4 and Figure S4. (d) Illustration of Lunar Tidal Wave in

359 the LLT frame in the Earth-Moon space. The background plasmasphere is shown by the yellow

360 region. The modulation of the plasmasphere by lunar tide is shown by the blue region, left for low

361 tide and right for high tide.

17
362
363

364
365 Fig. 3. Independent electric field observation and cold plasma convection modelling. (a)

366 Variations of the radial electric field Er (black, positive Er towards to the Earth) measured by the

367 Van Allen Probes and the perturbations of plasmapause position (blue) versus LLT. (b) Variation

368 of Er versus LLT. The asterisks stand for the observation and the dashed line stands for the

369 fitting. (c) Identification of plasmapause (dashed line) by the last-closed equipotential of the

370 convection-corotation electric field distribution. (d-e) Perturbations of the plasmapause and their

371 normalizations derived from the model. See main text for details.

372

18
373

374 Methods

375 1. Identification of plasmapause positions

376 The compilation of the plasmapause database was detailed in Zhang et al.35. Here we briefly

377 introduce the data sources and plasmapause determination methods. According to different

378 satellite observation data, a variety of plasmapause determination methods have been developed.

379 For the Time History of Events and Macroscale Interactions during Substorms (THEMIS) and

380 Polar satellites, the spacecraft potential (refers to the potential of the spacecraft body relative to

381 the ambient plasma) and the electron thermal velocities are used to calculate the electron density.

382 A detailed introduction of this method is given in Mozer (1973)51 and Pedersen et al. (1998)52. The

383 estimated error (a factor of 2) of this method is smaller than the density drops around the

384 plasmapause, thus this method has been widely used to the determine plasmapause positions53,54.

385 An example of the satellite potential and corresponding electron density measured by THEMIS is

386 shown in Figs. S1 (a) and (b). For the plasma wave instruments (in this investigation PWI of

387 International Sun-Earth Explorer-1 (ISEE-1), PWI of Dynamics Explorer 1 (DE-1), Plasma Wave

388 and Sounder (PWS) of Akebono, Plasma Wave Experiment (PWE) of Combined Release and

389 Radiation Effects Satellite (CRRES), PWI of Polar, Radio Plasma Imager (RPI) of Imager for

390 Magnetopause-to-Aurora Global Exploration (IMAGE), Waves of High Frequency and Sounder

391 for Probing of Density by Relaxation (WHISPER) of Cluster, and Electric and Magnetic Field

392 Instrument Suite and Integrated Science (EMFISIS) of VAP have been used), the electron density

393 (𝑛 , units: cm-3) can be related to 𝑓 through the formula 𝑛 = (𝑓 − 𝑓 )/8980 , where

394 𝑓 is upper hybrid resonance (UHR) frequency, 𝑓 = 𝑒𝐵/𝑚 is the electron cyclotron

395 frequency, 𝑒 is the electron charge, 𝐵 is the strength of the magnetic field, and 𝑚 is the electron

19
396 mass55. Since some satellites are unequipped with a magnetometer, 𝐵 is obtained by the

397 Tsyganenko 2007 external magnetic field model combined with the International Geomagnetic

398 Reference Field (IGRF) internal magnetic field model56. In terms of deducing electron densities

399 by the plasma wave instruments, this method has been proved to be very successful in many

400 studies55,57–60. Fig. S1 shows a corresponding example: panels (c) and (d) show the spectrogram

401 of the plasma waves measured by CLUSTER-4 and the electron density deduced from the UHR

402 frequency from 02:00 UT to 03:30 UT on 25 January 2002. The criterion for identifying the

403 plasmapause which are well adopted in previous studies are also used in this investigation: the

404 electron density drops five times or more in a short distance of ~0.5 RE (Earth’s radii, 1 RE =

405 6371.0 km). Based on this criterion, the black vertical dashed lines mark the plasmapause position.

406 The observations from Extreme ultraviolet (EUV) imaging instruments on board IMAGE and

407 (Chang’E-3) CE-3 can also be used to identify the plasmapause position. Several techniques61–63

408 have been developed to infer this boundary, such as the Minimum L Algorithm (MLA). It should

409 be noted that these EUV observations show the distribution of helium ions (He+) in the

410 plasmasphere by detecting its resonantly-scattered emission at 30.4 nm. However, the ion

411 plasmapause may be slightly different from the electron plasmapause. This study requires

412 extensive observations taken over multiple decades in order to compile sufficient statistics to

413 reveal lunar tidal effects. The EUV observations from IMAGE and CE-3 missions are only

414 available for much shorter time intervals (of order 1 year in each case), and in both cases

415 corresponds with high solar activity phases of the solar cycle. Therefore, these datasets were not

416 included in this study.

417 2. Plasmapause perturbations at different lunar phase

418 To extract the lunar tide signal in the plasmapause positions, the averaged background profile

419 of the plasmapause is constructed with the following steps. The magnetic local time (MLT) is

20
420 divided into 241 bins with interval of 0.1 h, then the average position of the plasmapause in each

421 bin is calculated and the averaged background profile of the plasmapause is obtained. Finally, this

422 profile is subtracted from the plasmapause positions. Through this process, the dawn-dusk

423 asymmetry and geomagnetic activity-induced variations of the plasmapause are almost eliminated.

424 A new database of plasmapause perturbations as a function of lunar phase is formed and used for

425 the following investigations.

426 3. Statistical Uncertainty Considerations and Data Processing

427 Although the database contains 35,982 plasmapause crossings, we need to divide these events

428 into several bins in MLT and LP phase, in order to explore the possibility of a lunar tide signal (for

429 example, we use 12 bins each, i.e., MLT=2:2:24 hours and LP=2:2:24 hours respectively). This

430 places limitations on the level of confidence we can assert on the mean values of plasmapause

431 location for each bin, due to the lower number of counts.

432 Fig. S2 (a) shows the distribution of perturbations in plasmapause radial distance in the MLT-

433 LP frame, ranging from -0.3 to 0.3 RE. Fig. S2 (b) shows the corresponding 95% confidence

434 interval (CI, ranging from 0.05 to 0.2 RE, and higher at the duskside due to plasmaspheric plume

435 as shown in Fig. 1 (a) – (b)). At first glance, Fig. S2 (a) appears to be dominated by random

436 fluctuations in the range of ± 0.3RE, however a low frequency underlying signal is also discernable.

437 Given that these fluctuations are of same order as the CI, we are justified in smoothing or low-pass

438 filtering the data to enhance the underlying signal.

439 It is important to note that the smoothing window size within a period of an oscillatory signal

440 does not change its periodic nature (although it can reduce its amplitude). Fig. S3 (a) shows a

441 simple example in which a sine function with added noise is smoothed using different smoothing

442 windows. The function is 𝑌 = 𝑆𝑖𝑛(𝑋) + 𝑟𝑎𝑛𝑑(−1,1), where 𝑟𝑎𝑛𝑑(−1,1) represents a random

443 number from -1 to 1. Its period is 2𝜋 and the data within a period (0 ≤ 𝑋 ≤ 2𝜋) is shown in the

21
444 figure by a black line. The red, green, blue, and yellow lines represent the results of smoothing

445 with a window size equal to 1/4𝜋 , 1/2𝜋, 3/4𝜋, and 𝜋, respectively. It is shown that these

446 smoothed results can more clearly display the changing trend of the data without changing the

447 period of the data, but the amplitude of the data will decrease with the increase of the smoothing

448 window.

449 Figs. S3 (b) – (e) show the distribution of perturbations of plasmapause position in the MLT-

450 LP frame with smoothing window equal to 3, 6, 9 and 12 hours, respectively. We obtain a clear

451 lunar tidal signal which has diurnal (and monthly) periodicities, and this lunar signal becomes

452 clearer as the smoothing window is increased. Therefore, in order to obtain clear results (accurate

453 period and amplitude), we chose 6 hours (a quarter of a period) for the smoothing window in this

454 study.

455 4. Descriptions of MLT, LP and LLT.

456 The MLT, LP and LLT variables all refer to locations and are defined in Fig. S4. The MLT

457 is usually used to define azimuthal locations in the Sun-Earth reference frame. When looking from

458 the North Pole, the MLT increases anticlockwise from 0 at midnight to 6 at dawn, 12 at noon, 18

459 at dusk and 24 at midnight. For convenience in this study, the LP is defined as the MLT of lunar

460 position with LP values of 0/24, 6, 12, and 18 corresponding to the full moon, third quarter moon,

461 new moon, and first quarter moon, respectively. The LLT is defined based on the MLT of the

462 Moon. LLT is always equal to 0 at the far-side of Moon-Earth line and is always equal to 12 at the

463 location pointing from the Earth to the Moon. For an azimuthal location P at MLT in Earth's frame,

464 when the Moon is at LP, the corresponding LLT is defined as 12-LP+MLT.

465 5. VAP observations of radial electric field

466 In this study, the electric field data between L-shell = 3 ~ 6 RE in the Geocentric Solar

467 Magnetospheric (GSM) coordinate system comes from the Van Allen Probes (VAP) satellite from
22
468 January 2013 to May 2019. The sensitivity of the electric field measured by the Electric Fields and

469 Waves (EFW) Instruments on the Van Allen Probes (VAP) is 0.1 mV/m or 10 % of the amplitude64.

470 Note that, for VAP spacecraft, the spin axis electric field component (Ex) is estimated using the

471 assumption that E • B = 0 or the parallel electric field is zero under the conditions of |𝐵 /𝐵 | < 4

472 and |𝐵 /𝐵 | < 4. In order to obtain the electric field data near the magnetic equator, we selected

473 the electric field data with −15° <MLAT (Magnetic latitude) <15° . Subsequently, after getting the

474 radial component of electric field 𝐸 and the corresponding local lunar time (LLT), we got the

475 LLT variations of 𝐸 . Here, the size of the window is 6 hours, e.g., LLT = 6 stands for 3 < LLT <

476 9.

477 6. Including the disturbed electric field in the cold plasma convection model

478 Since the motion of the plasma in the plasmasphere is mainly controlled by the E× B drift, the

479 configuration of the plasmasphere is determined by the electric field, assuming a fixed magnetic

480 field. This model is an electric field model consisting of curl-free Ecoro and Econv components, and

481 disturbance electric field. In this model, Econv is expressed by the Volland-Stern inner-

482 magnetospheric potential model at a fixed Kp index of 365,66. For Ecoro, the corotation potential 𝛷C

483 is first calculated with the formula 𝛷C=-𝛺𝜇M/(4𝜋LRE)67, where 𝛺 is the rotation speed of the Earth,

484 𝜇 is the permeability, M is the magnetic moment of the Earth’s dipole, L denotes the L-shell value,

485 and RE represents the Earth radius, respectively. Then the radial Ecoro is obtained by differentiation

486 of the potential (Ecoro=-𝛻𝛷C). Here the disturbance electric field is ~38% ∆𝐸 and is a function of

487 lunar phase. This is superimposed onto the convection-corotation electric field by changing Ecoro:
∆
488 𝛷 = 𝛷 /(1 + × 0.3) when L ≥ 3.2 RE, where 𝐸 is the mean of 𝐸 ; 𝛷 = 𝛷 when L ≤

( . ) ( )
489 3 RE; 𝛷 = 𝛷 (𝐿 = 3) + (𝐿 − 3) when 3≤ L ≤ 3.2 RE .
.

23
490 The reasons for superimposing 38% ∆𝐸 (as opposed to 100%) under different lunar phases

491 onto the convection-corotation electric field are as follows:

492 (1) The amplitude of 𝐸 is of the same order as the VAP EFW instrument measurement error,

493 hence the amplitude is not well determined.

494 (2) The radial electric field measured by the VAP satellites will in general over-estimate the value

495 at the equator, assuming equipotential field lines.

496 (3) Our model which is a simple electric field model consisting of Ecoro and Econv and disturbance

497 electric field may be just used to qualitatively describe this tidal phenomenon, not

498 quantitatively.

499

500 References

501 51. Mozer, F. S. Analyses of techniques for measuring DC and AC electric fields in the

502 magnetosphere. Space Sci. Rev. 14, 272–313 (1973).

503 52. Pedersen, A., Mozer, F. & Gustafsson, G. Electric Field Measurements in a Tenuous

504 Plasma with Spherical Double Probes. in Geophysical Monograph Series (eds. Pfaff, R.

505 F., Borovsky, J. E. & Young, D. T.) 1–12 (AGU, 1998). doi:10.1002/9781118664391.ch1.

506 53. Li, W. et al. Global distributions of suprathermal electrons observed on THEMIS and

507 potential mechanisms for access into the plasmasphere. J. Geophys. Res. Sp. Phys. 115,

508 (2010).

509 54. Cho, J. et al. New model fit functions of the plasmapause location determined using

510 THEMIS observations during the ascending phase of Solar Cycle 24. J. Geophys. Res. Sp.

511 Phys. 120, 2877–2889 (2015).

24
512 55. Kurth, W. S. et al. Electron densities inferred from plasma wave spectra obtained by the

513 Waves instrument on Van Allen Probes. J. Geophys. Res. Sp. Phys. 120, 904–914 (2015).

514 56. Tsyganenko, N. A. & Sitnov, M. I. Magnetospheric configurations from a high-resolution

515 data-based magnetic field model. J. Geophys. Res. Sp. Phys. 112, (2007).

516 57. Oya, H. et al. Plasma wave observation and sounder experiments(PWS) using the

517 Akebono (EXOS-D) satellite. Instrumentation and initial results including discovery of the

518 high altitude equatorial plasma turbulence. J. Geomagn. Geoelectr. 42, 411–442 (1990).

519 58. Moldwin, M. B., Thomsen, M. F., Bame, S. J., McComas, D. & Reeves, G. D. The fine-

520 scale structure of the outer plasmasphere. J. Geophys. Res. 100, 8021–8029 (1995).

521 59. Goldstein, J. Identifying the plasmapause in IMAGE EUV data using IMAGE RPI in situ

522 steep density gradients. J. Geophys. Res. 108, 1147 (2003).

523 60. Darrouzet et al. Density structures inside the plasmasphere: Cluster observations. Ann.

524 Geophys. 22, 2577–2585 (2004).

525 61. He, F., Zhang, X., Chen, B., Fok, M. & Nakano, S. Determination of the Earth’s

526 plasmapause location from the CE‐3 EUVC images. J. Geophys. Res. Sp. Phys. 121, 296–

527 304 (2016).

528 62. Wang, C., Newman, T. S. & Gallagher, D. L. Plasmapause equatorial shape determination

529 via the Minimum L Algorithm: Description and evaluation. J. Geophys. Res. Sp. Phys.

530 112, (2007).

531 63. Sandel, B. R., Goldstein, J., Gallagher, D. L. & Spasojevic, M. Extreme Ultraviolet

532 Imager Observations of the Structure and Dynamics of the Plasmasphere. Space Sci. Rev.

533 109, 25–46 (2003).

25
534 64. Wygant, J. R. et al. The Electric Field and Waves Instruments on the Radiation Belt Storm

535 Probes Mission. Space Sci. Rev. 179, 183–220 (2013).

536 65. Stern, D. P. The motion of a proton in the equatorial magnetosphere. J. Geophys. Res. 80,

537 595–599 (1975).

538 66. Volland, H. A semiempirical model of large-scale magnetospheric electric fields. J.

539 Geophys. Res. 78, 171–180 (1973).

540 67. Baumjohann, W. & Treumann, R. A. Basic Space Plasma Physics. (PUBLISHED BY

541 IMPERIAL COLLEGE PRESS AND DISTRIBUTED BY WORLD SCIENTIFIC

542 PUBLISHING CO., 1996). doi:10.1142/p015.

543

544

545

546

26
547 Acknowledgments: We thank Donghe Zhang, Jiyao Xu, Jiuhou Lei, Jie Ren, Jong-Sun Park,

548 Motoharu Nowada, Dianjun Zhang, Wensai Shang, and Shutao Yao for helpful discussions.

549 Funding:

550 National Natural Science Foundation of China 41974189

551 National Natural Science Foundation of China 41961130382

552 National Natural Science Foundation of China 41731068

553 National Natural Science Foundation of China 41941001

554 National Natural Science Foundation of China 41974194

555 National Natural Science Foundation of China 41821003

556 National Natural Science Foundation of China 41974191

557 National Key R&D Program of China 2020YFE0202100

558 China Space Agency project D020303

559 Key Research Program of the Institute of Geology & Geophysics, CAS IGGCAS-201904

560 Author contributions:

561 Conceptualization: QS, WL

562 Methodology: CX, FH, …

563 Investigation: CX, …

564 Visualization: CX, …

565 Funding acquisition: QS, FH, …

566 Writing – original draft: CX, …

27
567 Writing – review & editing: CX, FH, QS, WL, AD…

568 Competing interests: Authors declare that they have no competing interests.

569 Data and materials availability: The EFW data are available at

570 http://www.space.umn.edu/rbspefw-data/. The plasmapause database is available through

571 request to the author (F.H., hefei@mail.iggcas.ac.cn). The LPs data are calculated by the

572 planetary ephemeris DE432 which was created by Jet Propulsion Laboratory in 2014.

573

574

28
575
576 Fig. S1. Illustrations of plasmapause crossings using different types of plasmaspheric

577 observations. (a-b) The spacecraft potential and electron density measured by THEMIS A on 31

578 March 2010. (c-d) The spectrogram of the plasma wave measured by ClUSTER-4 and electron

579 density deduced from the UHR frequency on 25 January 2002.

580

29
581

582
583 Fig. S2. Plasmapause location perturbations and corresponding errors. (a) The distribution

584 of perturbations of plasmapause position in the MLT-LP frame. (b) The corresponding 95%

585 confidence interval (CI) error.

586
587

30
588
589 Fig. S3. Effect of a smoothing window on the periodicity of a dataset with fixed period. (a)

590 An example (𝑌 = 𝑆𝑖𝑛(𝑋) + 𝑟𝑎𝑛𝑑(−1,1), its period is 2𝜋 and the data with a period (0 ≤ 𝑋 ≤

591 2𝜋) is shown here) with different smoothing windows. (b-e) The distribution of perturbations of

592 plasmapause position in the MLT-LP frame with smoothing window equal to 3, 6, 9 and 12

593 hours, respectively.

594

31
595
596 Fig. S4. Definitions of the location frames. MLT, LP, LLT and the relationship between them

597 for a point P on the Earth.

598

32
Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.

MovieS1Lunartide.gif
SupplementaryInformation.pdf