Author Query Form
Journal:
SPAC
Article ID:
SPAC579
Please send your responses together with your list of corrections via web
(preferred), or send the completed form and your marked proof to:
Mokslininku 2a, LT-08412 Vilnius, Lithuania
fax: +370 5 2729 501
e-mail: vtexspr-corrections@vtex.lt
Dear Author,
During the preparation of your manuscript for typesetting, some questions have arisen. These are listed below.
Queries and/or remarks
Location in
article (page/line)
Query / remark
1/7
The author name has been tagged as Given
name and Family name. Please confirm if they
have been identified correctly and are presented
in the right order. Note that response to this
query is mandatory.
1/48
City name is required in the fifth address, please
indicate.
2/87; 5/248;
16/773; 17/820
Reference (Lognonn\'{e} et al., this issue) is
missing from the references list. Please add.
6/263; 15/702
Figures 2 and 10 are not cited in the text. Please
add a citation for them.
8/394
Reference (Smrekar et al., this issue) is missing
from the references list. Please add.
16/783, 785
In Eq. (17) "n gal" has been changed by
"nanogal". Please check and amend if necessary.
18/872
Please check if mathematical expression
"5.81/4\ 10^{-11}" is correct and correct if
necessary.
19/948
Please check if mathematical expression "5\
10^{-6}" is correct and correct if necessary.
22/1091
The last sentence in the Acknowledgements
looks strange. Please check and correct if
necessary.
Response
Many thanks for your assistance
Page 1 of 2
Location in
article (page/line)
Query / remark
23/1134; 24/1165,
1172, 1195;
25/1205, 1221,
1241
Please update references (Knapmeyer et al.
2006) (Madeleine et al. 2011) (Mimoun et al.
2016) (Petrosyan et al. 2011) (Rhie and
Romanowicz 2006) (Spiga and Forget 2009)
(Watada and Kanamori 2010) if possible; DOI
numbers would be sufficient.
Response
Many thanks for your assistance
Page 2 of 2
Metadata of the article that will be visualized in Online First
Journal Name
Space Science Reviews
Article Title
Mars’ Background Free Oscillations
Copyright holder
Springer Nature B.V.
This will be the copyright line in the final PDF.
Corresponding Author
Family name
Particle
Given Name
Suffix
Division
Organization
Address
E-mail
Family name
Particle
Given Name
Suffix
Division
Organization
Address
E-mail
Family name
Particle
Given Name
Suffix
Division
Organization
Address
E-mail
Family name
Particle
Given Name
Suffix
Division
Organization
Author
Author
Author
Author
Author
Address
E-mail
Family name
Particle
Given Name
Suffix
Division
Organization
Address
E-mail
Family name
Particle
Given Name
Nishikawa
Y.
Institut de Physique du Globe de Paris, Sorbonne Paris Cité
Université Paris Diderot
Paris, France
nishikawa@ipgp.fr
Lognonné
P.
Institut de Physique du Globe de Paris, Sorbonne Paris Cité
Université Paris Diderot
Paris, France
Kawamura
T.
Institut de Physique du Globe de Paris, Sorbonne Paris Cité
Université Paris Diderot
Paris, France
Spiga
A.
Laboratoire de Météorologie Dynamique (LMD/IPSL)
Sorbonne Université, Centre National de la Recherche
Scientifique, École Polytechnique, École Normale Supérieure
Paris, France
Stutzmann
E.
Institut de Physique du Globe de Paris, CNRS-UMR 7580
University Paris 7
Paris, France
Schimmel
M.
Suffix
Division
Organization
Author
Author
Address
E-mail
Barcelona, Spain
Family name
Particle
Given Name
Suffix
Division
Organization
Address
E-mail
Family name
Particle
Given Name
Suffix
Division
Organization
Bertrand
Address
E-mail
Author
Schedule
Abstract
Institute of Earth Sciences Jaume Almera – CSIC
T.
Ames Research Center
National Aeronautics and Space Administration (NASA)
???, CA, USA
Forget
F.
Laboratoire de Météorologie Dynamique (LMD/IPSL)
Sorbonne Université, Centre National de la Recherche
Scientifique, École Polytechnique, École Normale Supérieure
Paris, France
Family name
Kurita
Particle
Given Name
K.
Suffix
Division
Earthquake Research Institute
Organization
The University of Tokyo
Address
Tokyo, Japan
E-mail
Received
18 June 2018
Revised
Accepted
7 January 2019
Observations and inversion of the eigenfrequencies of free oscillations constitute
powerful tools to investigate the internal structure of a planet. On Mars, such free
oscillations can be excited by atmospheric pressure and wind stresses from the
Martian atmosphere, analogous to what occurs on Earth. Over long periods and on a
global scale, this phenomenon may continuously excite Mars’ background free
oscillations (MBFs), which constitute the so-called Martian hum. However, the source
exciting MBFs is related both to the global-scale atmospheric circulation on Mars and
to the variations in pressure and wind at the planetary boundary layer, for which no
data are available.
To overcome this drawback, we focus herein on a global-scale source and use
results of simulations based on General Circular Models (GCMs). GCMs can predict
and reproduce long-term, global-scale Martian pressure and wind variations and
suggest that, contrary to what happens on Earth, daily correlations in the Martian
hum might be generated by the solar-driven GCM. After recalling the excitation
terms, we calculate MBFs by using GCM computations and estimate the contribution
to the hum made by the global atmospheric circulation. Although we work at the
lower limit of MBF signals, the results indicate that the signal is likely to be periodic,
which would allow us to use more efficient stacking theories than can be applied to
Earth’s hum. We conclude by discussing the perspectives for the InSight SEIS
instrument to detect the Martian hum. The amplitude of the MBF signal is on the
order of nanogals and is therefore hidden by instrumental and thermal noise, which
implies that, provided the predicted daily coherence in hum excitation is present, the
InSight SEIS seismometer should be capable of detecting the Martian hum after
monthly to yearly stacks.
Keywords
Mars – Planetary free oscillation – GCM – Seismometer – Normal mode – InSight
Footnotes
The InSight Mission to Mars II
Edited by William B. Banerdt and Christopher T. Russell
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
Prn:2019/01/22; 9:15
p. 1/25»
Space Sci Rev _#####################_
https://doi.org/10.1007/s11214-019-0579-9
1
2
3
4
5
Mars’ Background Free Oscillations
6
7
8
9
10
Y. Nishikawa1 · P. Lognonné1 · T. Kawamura1 ·
A. Spiga2 · E. Stutzmann3 · M. Schimmel4 ·
T. Bertrand5 · F. Forget2 · K. Kurita6
11
12
13
14
15
Received: 18 June 2018 / Accepted: 7 January 2019
© Springer Nature B.V.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Abstract Observations and inversion of the eigenfrequencies of free oscillations constitute
powerful tools to investigate the internal structure of a planet. On Mars, such free oscillations can be excited by atmospheric pressure and wind stresses from the Martian atmosphere, analogous to what occurs on Earth. Over long periods and on a global scale, this
phenomenon may continuously excite Mars’ background free oscillations (MBFs), which
constitute the so-called Martian hum. However, the source exciting MBFs is related both to
the global-scale atmospheric circulation on Mars and to the variations in pressure and wind
at the planetary boundary layer, for which no data are available.
To overcome this drawback, we focus herein on a global-scale source and use results of
simulations based on General Circular Models (GCMs). GCMs can predict and reproduce
long-term, global-scale Martian pressure and wind variations and suggest that, contrary to
what happens on Earth, daily correlations in the Martian hum might be generated by the
solar-driven GCM. After recalling the excitation terms, we calculate MBFs by using GCM
computations and estimate the contribution to the hum made by the global atmospheric
circulation. Although we work at the lower limit of MBF signals, the results indicate that
the signal is likely to be periodic, which would allow us to use more efficient stacking
theories than can be applied to Earth’s hum. We conclude by discussing the perspectives for
the InSight SEIS instrument to detect the Martian hum. The amplitude of the MBF signal is
36
37
38
39
40
41
The InSight Mission to Mars II
Edited by William B. Banerdt and Christopher T. Russell
B Y. Nishikawa
nishikawa@ipgp.fr
1
Institut de Physique du Globe de Paris, Sorbonne Paris Cité, Université Paris Diderot, Paris, France
2
Laboratoire de Météorologie Dynamique (LMD/IPSL), Sorbonne Université, Centre National de la
Recherche Scientifique, École Polytechnique, École Normale Supérieure, Paris, France
45
3
Institut de Physique du Globe de Paris, CNRS-UMR 7580, University Paris 7, Paris, France
46
4
Institute of Earth Sciences Jaume Almera – CSIC, Barcelona, Spain
5
Ames Research Center, National Aeronautics and Space Administration (NASA), ???, CA, USA
6
Earthquake Research Institute, The University of Tokyo, Tokyo, Japan
42
43
44
47
48
49
50
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1
PDF-OUTPUT
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
51
52
53
54
_####_
Prn:2019/01/22; 9:15
Page 2 of 25
p. 2/25»
Y. Nishikawa et al.
on the order of nanogals and is therefore hidden by instrumental and thermal noise, which
implies that, provided the predicted daily coherence in hum excitation is present, the InSight
SEIS seismometer should be capable of detecting the Martian hum after monthly to yearly
stacks.
55
56
57
Keywords Mars · Planetary free oscillation · GCM · Seismometer · Normal mode · InSight
58
59
1 Introduction
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
When a large earthquake occurs, seismic waves propagate around the entire planet for tens
of hours following the event. These waves can generate standing oscillations if they fulfill the boundary conditions for specific frequencies; the standing modes are referred to as
Earth’s free seismic oscillations and their associated frequencies are called Earth’s seismic
eigenfrequencies.
The first search for Earth’s free oscillations started in the late 1950 (Benioff et al. 1959)
these oscillations were first observed following the great Chilean earthquake of May 1960
(Benioff et al. 1961). Further developments in long-period seismometers and global networks then allowed the normal mode frequencies to be determined, and this information
was used to invert the structure models of Earth’s interior, including the famous Preliminary
Earth Reference Model (Dziewonski and Anderson 1981). Further inversions were done by
using the splitting properties of the normal modes, which provided three-dimensional models of the Earth’s structure. Woodhouse and Deuss (2015); Laske and Widmer-Schnidrig
(2015); Montagner (2015) and Romanowicz and Mitchell (2015), provide an extensive description of observations of Earth’s normal modes and their use in determining Earth’s interior structure, while Dahlen and Tromp (1998) and Millot-Langet et al. (2002) provide a
theoretical description of the same.
Thus, determining Mars’ normal modes has been proposed as one of the overarching
goals of a seismological mission to Mars from the early times of space exploration (Kovach
and Anderson 1965; Bolt and Derr 1969) to Viking (Okal and Anderson 1978) as well as
the more recent projects such as Mesur (Solomon et al. 1991), Intermarsnet (Lognonné
et al. 1996) and NetLander (Lognonné and Giardini 2000). Several papers detail detection
techniques have been detailed in several papers (Lognonné and Mosser 1993; Zharkov and
Gudkova 1997; Gudkova and Zharkov 2004; Lognonné 2005; Lognonné and Johnson 2007,
2015). The detection of the normal modes of Mars is one of the goals of the upcoming
seismic monitoring mission “InSight” which should land and deployment in 2019 and of the
SEIS experiment (Tong and García 2015; Lognonné et al., this issue).
Planetary free oscillations constitute a powerful tool to investigate the internal structure
of planets, as illustrated by Panning et al. (2017) with the synthetic inversion of the structure of the fundamental spheroidal normal modes. However making such long-term seismic
observations on Mars obliges us to overcome several difficult problems.
One issue is related to Mars itself, which lacks plate tectonics and therefore is devoid
of the large-magnitude quakes (M > 7) the type of which are generally used on the Earth
to analyze the normal modes. All seismicity models of Mars (Phillips and Grimm 1991;
Golombek et al. 1992; Knapmeyer et al. 2006) suggest that the largest quake annually is in
the moment-magnitude range of 5.2–6 (i.e., 1017 to 1018 Nm). Larger quakes, although very
rare, cannot be excluded.
The second difficulty is the seismic noise expected on the Martian surface, which is
known to be subject to large variations in temperature and wind that generate long-period
100
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 3/25»
Page 3 of 25
_####_
noise much greater than anything measured in Earth’s seismic vaults, where Earth’s normal
modes are typically observed. The InSight SEIS noise model, described in detail by Mimoun et al. (2016), suggests that, over long periods, most of the SEIS instrument noise will
be related to the thermal noise, which is expected to grow below 10 mHz as f −2.5 , with amplitude at 10 mHz in the range of 1 to 2.5 × 10−9 m s−2 Hz−1/2 depending on the local time.
As shown by Lognonne et al. (2006), seismic signals would have larger amplitudes globally
were a quake with a moment larger than 2 × 1017 Nm to occur on Mars. Even if such a moment corresponds to the upper limit of the annual predicted seismicity, the Poisson statistics
of such a rare event suggests that several years of operation are required to attain a significant occurrence probability. Therefore, we consider herein the nonseismogenic sources of
normal mode excitation.
One possible source of normal mode excitation comes from atmospheric phenomena (see
Spiga et al. 2018, for a detailed review). Mars has a very active atmosphere, with weather
processes generating high winds, dust devils, and storms (Read and Lewis 2004; Spiga and
Forget 2009). Cloud tracking shows that Martian wind speeds easily reach 80 m/s above
30 km (Kaydash et al. 2006), whereas surface wind can be as high as 20 m/s, according
to measurements by the Viking landers (e.g., Murphy et al. 1990). Measurements by the
Martian landers also show that the Martian atmosphere has a regular diurnal cycle of wind
and pressure (Schofield et al. 1997; Martínez et al. 2017).
The nonacoustic pressure variations associated with the global weather and the Reynolds
stress associated with the wind, be it turbulent gusts in the planetary boundary layer (PBL)
or high-altitude planetary-scale wind, all constitute forces to excite planetary free oscillations (Kobayashi and Nishida 1998b,a), with theoretical details in Tanimoto and Um (1999).
Part of the excitation forces has frequencies within the bandwidth of the normal modes and
that corresponds to the eigenfrequencies of Martian normal modes between 0.4 and 20 to
30 mHz. These particular excitation forces can globally and continuously excite the corresponding normal modes. This continuous excitation force is a possible source of Mars’
background free oscillations (MBFs).
This idea for the observation of MBF was actually been suggested shortly after the first
observation of Earth’s background free oscillations (EBFs) (Suda et al. 1998). Oceanic excitation was initially proposed by Watada and Masters (2001). The first interpretations proposed that the entire excitation is produced by atmospheric turbulence in the Earth’s PBL,
which suggests that a simple scaling may exist between Earth and Mars (Kobayashi and
Nishida 1998b,a). The seasonal variations in the continuous excitation later observed by
Nishida et al. (2000) also support an atmospheric origin, while Tanimoto and Um (1999)
proposed a simplified theory. However, the most recent studies show that the major source
of continuous excitation on Earth is over the oceans (Tanimoto 2005; Rhie and Romanowicz 2004, 2006) and that, in this process, infragravity waves over the continental shelves
are much more efficient seismic sources (Webb 2007) than atmospheric turbulence. However, excitation by atmospheric sources remains significantly below 5 mHz (Nishida 2013b).
Note that coupled modes exist between the atmosphere and the solid earth (Lognonné et al.
1998b, 2016), and that pressure sources at the bottom of the ocean (Nishida 2014) cannot explain this larger excitation; instead, integrated atmospheric excitations at the base of Earth’s
atmosphere must be involved.
Observed EBFs are on the order of 0.5 nanogal (1 nanogal = 10−11 m s−2 ) (Nishida
2013a) per individual mode. Stacking techniques can enhance the peak amplitude of normal
modes to overcome instrument- and station-induced noise, which allows the mode eigenfrequencies to be determined. These eigenfrequencies may then be used to invert the internal
structure as was done for quakes, as illustrated by Nishida et al. (2009). The same phenomenon should occur on Mars, meaning that the observation of MBFs should allow us to
150
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
_####_
Prn:2019/01/22; 9:15
Page 4 of 25
p. 4/25»
Y. Nishikawa et al.
determine the frequencies of fundamental spheroidal normal modes, which would constrain
models for Mars’ interior.
Kobayashi and Nishida (1998b) estimated the magnitude of MBFs by using a theoretical
scaling based on the solar energy received by the planet (related to the planetary albedo and
distance to Sun) and on the thickness of the PBL. They assumed that turbulence in the PBL
provides the main excitation force for MBFs, which led them to estimate a free oscillation
amplitude of several nanogals, a value quite close to that of Earth. Tanimoto (2001) estimated relative modal estimated the relative modal amplitude between Earth and Mars by
using several parameters for the atmosphere and terrestrial part. Estimates of the MBF amplitude are about 30–50% of the EBFs. Lognonné (2005) and Lognonné and Johnson (2007)
focused on very-long-period MBFs and, by using more realistic Martian-climate models and
normal mode summations, produced MBF seismograms for periods ranging 300 to 400 s,
which correspond to angular orders 12 to 15. This approach is based on the assumption that,
for these very long periods and wavelengths, the major source of excitation is not Reynolds
stresses or nonacoustic pressure related to the PBL, but the nonacoustic pressure related to
global atmospheric circulation. They estimated the free-oscillation amplitude to be several
hundredths of nanogals, which corresponds to amplitudes typically ten times smaller than
those of EBFs of the same angular order.
The generation of seismic waves by atmospheric activity is not only associated with
nonacoustic pressure acting on the surface, but also with the acoustic waves generated in
the atmosphere and converted into seismic waves at the surface. Lognonné et al. (1998b),
Watada and Kanamori (2010), and Lognonné et al. (2016) simplified this concept by demonstrating that the overall excitation in the atmosphere may be estimated based on the normal
modes of the coupled solid-atmosphere system and on the atmospheric force volumetric
density, which acts throughout the atmosphere and is subject to wind and nonacoustic pressure fluctuations. This approach requires integrating over the full atmospheric seismogenic
volume to calculate the excitation force of the atmosphere. As shown more precisely by
Lognonné et al. (1994) the latter is associated with seismic forces related to nonacoustic
pressure and nonlinear Reynolds stresses in the atmosphere and can be expressed as
180
181
182
183
184
185
186
187
188
189
190
191
192
j
j
Πi = −pglut δi − ρvi v j ,
(1)
where pglut nonacoustic pressure, which is defined as pglut = ptrue − pHooke , where pHooke =
−κ∇.v. v is the atmospheric wind, ρ is atmospheric density, and pHooke is the pressure
modeled by the Hooke pressure, which involves the linear model used in linear acoustics
or seismology. Therefore, following the proposition made by Backus and Mulcahy (1976)
for localized sources in Earth seismology and by Goldreich and Keeley (1977) for globally
distributed sources in Solar seismology, we generalize here the concept of stress glut to
the pressure induced by nonlinear Reynolds stresses that occur in the Martian PBL. This
generalization considers that all true volumetric forces, apart from those generated by Hooke
pressure, are source terms captured by the seismic moment of Eq. (1). Thus, we consider
that the pressure glut and wind are respectively written as
193
194
195
196
197
198
199
pglut = pglut,global + pglut,PBL ,
v = vglobal + vPBL .
(2)
Practically speaking, the wind and pressure glut associated with the PBL turbulence will be
at much higher frequencies and on a smaller scale than those associated with the global circulation, but both will contribute to the overall excitation processes through Eq. (1). Therefore, the two estimates seem to provide the two end members of the general case, with
200
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
201
202
203
204
205
206
207
208
209
210
211
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 5/25»
Page 5 of 25
_####_
Fig. 1 Pressure fluctuations in
Martian atmospheric simulated
by LMD GCM for Mars meridian
of longitude 0◦ . All pressure
records are demeaned and only
pressure variations are shown.
Vertical axis is latitude;
horizontal axis is local time [hr].
The largest pressure fluctuations
occur at sunrise and sunset. The
pressure fluctuations repeat to
high-precision day after day. This
strong daily repetition is not
common on Earth
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
Lognonné and Johnson (2007) focusing on the global part of the excitation and Kobayashi
and Nishida (1998b) focusing on the PBL part.
Major differences in PBL thickness exist between the Earth PBL, which ranges from 0.2
to 2 km and the Martian PBL, which is much thicker (about 5 to 6 km, typically) (Petrosyan
et al. 2011; Spiga 2011). Because the speed of sound is less on Mars than on Earth (220 m/s
compared with 340 m/s), the volumetric excitation is likely much more important on Mars
than on Earth where, moreover, most of the energy of the atmospheric winds is transferred
to oceanic waves, which then play a major role in the excitation of EBFs. In other words, the
Earth view, where most EBF excitation may be modeled by invoking bottom-oceanic and
near-surface forces, is not likely to be valid on Mars.
A first key difference is the important daily repetition of weather patterns on Mars, which
is mostly related to the importance of solar flux and to the lack of a major source of other
atmospheric energy, such as Earth’s humidity and water. Although this initially appears in
the temperature variations, which are fairly well represented by a Fourier series (see, e.g.,
Van Hoolst et al. 2003), it is also the case for the pressure field, as illustrated in Fig. 1,
which shows the pressure variations at zero longitude as a function of local time and of
latitude over two sols, as generated by a typical GCM. Such sol repetition also appears in
the signal, when the latter is filtered in the frequency bandwidth of the normal modes. This
is shown in Fig. 1, which corresponds to the signals of Fig. 1, but is filtered between 1 and
2 mHz (i.e., periods between 1000 and 500 s) and by Fig. 3, where the bandwidth is higher
and corresponds to 2 to 3 mHz (i.e., periods between 330 and 500 s).
In this paper, we expand these past studies by using higher-resolution GCMs developed
recently by the Laboratoire de Météorologie Dynamique (LMD) (see e.g., Forget and Lebonnois 2013) to better estimate normal mode amplitudes and to analyze whether the repeating
atmospheric sources significantly impact the excitation of normal modes, including terms
of stacking strategies. These GCMs can resolve all the large-scale Martian atmospheric processes and variabilities such as thermal tides, baroclinic waves, and planetary-scale waves
(Haberle et al. 1999; Forget et al. 1999). Turbulent motions in the PBL are, by design, left
unresolved in GCMs; their study requires large-eddy simulations (Spiga and Lewis 2010).
This will not be included for the excitation force investigated herein [note that atmospheric
excitation by small-scale turbulence in the PBL is addressed by Kenda et al. (2017) and
Murdoch et al. (2017).
Normal mode amplitudes are computed from GCM results by summing the normal
modes, which estimates the signals that may be recorded by the InSight SEIS VBB seismometer (Lognonné et al. 2015; Lognonné et al., this issue). Compared with previous studies, the present study provides not only better lower estimates of MBF normal modes and
250
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
_####_
Prn:2019/01/22; 9:15
Page 6 of 25
p. 6/25»
Y. Nishikawa et al.
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
Fig. 2 Pressure fluctuations in Martian atmospheric in frequency band where normal modes concentrate in
normal mode band. We applied a 1.0–2.0 mHz band-pass filter to the raw pressure data (shown in Fig. 1). The
daily repetition of Martian weather processes appears clearly shown in the pattern. The band of the pattern
at 1.0–2.0 mHz corresponds to low MBF frequencies (angular order 2 to 8) MBFs frequencies. The pressure
fluctuations in the atmosphere excite daily coherent MBFs
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
Fig. 3 Pressure fluctuations in Martian atmosphere at frequency band where normal modes concentrate
at normal mode band. We applied a band-pass filter between 2.0 and 3.0 mHz to the raw pressure data
(shown in Fig. 1). The large pressure fluctuations in this band occur twice a day (at sunrise and sunset). The
frequency band corresponds to middle degree MBFs (angular order 9 to 15). Normal modes at this frequency
are sensitive to the upper mantle (see Fig. 5)
292
293
294
295
296
297
298
299
extension to higher frequencies than Lognonné and Johnson (2007, 2015), but also pathways
for the future processing of InSight measurements. Our synthetic signals will be compared
with the expected noise and used to test stacking strategies, with a special emphasis on possible coherent stacking techniques made possible by the daily repetition of Martian weather.
Finally, we conclude by estimating the probability of detecting MBFs with the SEIS seismometer.
300
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
301
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 7/25»
Page 7 of 25
_####_
2 Martian Global Climate Modeling for Normal Mode Computation
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
To predict the atmospheric excitation force, we used Mars Global Climate Model (GCM)
which is designed to simulate large-scale atmospheric dynamics over an entire planetary
sphere. The characteristics of the model are described in detail in Forget et al. (1999). We
use the latest, most realistic, version of the model, which features interactive dust transport
(Madeleine et al. 2011), radiatively active water and ice clouds (Navarro et al. 2014), and a
thermal plume model for the boundary layer (Colaïtis et al. 2013).
Computing normal mode excitation requires a GCM simulation that is able to characterize the Martian atmospheric temporal variability of pressure and winds in the Martian atmospheric at 0.01 Hz (period 100 s), with a horizontal resolution suitable for normal modes,
(i.e., 10◦ latitude and longitude). These requirements, dictated by seismic computations, are
somewhat paradoxical from a meteorological point of view. A timescale of 100 s is associated with atmospheric circulations evolving over typical spatial scales of tens to hundreds
of meters, which are challenging to resolve with GCMs, even with the best available supercomputing cluster. Most of the 100 s variability of the Martian atmosphere is instead
captured through limited-area, turbulence-resolving modeling [large-eddy simulations; see,
e.g., Spiga and Lewis (2010); see also Kenda et al. (2017), Murdoch et al. (2017) for a discussion of local seismic signature]. The remainder of this variability (i.e., the contribution
that is not caused by microscale turbulence) is associated with mid-latitude baroclinic fronts
and regional slope winds (which impact the global dynamics) and can be satisfactorily simulated through GCMs with a horizontal resolution of 1◦ . Conversely, nearly three decades
of Mars GCM studies have consistently demonstrated that the global dynamics (at spatial
scales of about 10◦ latitude and longitude) are well simulated by using a typical time step of
925 s in the GCM.
We also ran Martian GCM simulations specifically tailored for the seismic computations.
The LMD GCM simulations used a 60 s time step, 360 longitude grid points, and 180 latitude grid points, corresponding to a mesh spacing of 1◦ × 1◦ and a horizontal resolution
of 60 km [in a setting akin to the GCM simulations described in Pottier et al. (2017)]. In
the vertical direction, 25 levels are typically used with most of the levels located in the first
15 km to ensure a suitable resolution in the lower troposphere and in the boundary layer.
Above 10 km, the vertical resolution is about one scale height and the altitude of the top
level is about 60 km, which corresponds to about 6 scale heights. This vertical grid thus
offers both the refined near-surface resolution and the accounts for the vast majority of the
atmospheric mass, which allows us to deal with all possible seismic coupling in subsequent
computations. It is important to note that turbulent motions developing at high frequency are
not resolved by the GCM: the PBL mixing they cause is parameterized in the LMD GCM by
dedicated schemes (Colaïtis et al. 2013). The GCM results used to compute normal modes
thus only feature atmospheric variability at scales ranging from regional to global, and frequencies typically of the order 10−3 Hz (and below).
Typical GCM simulations of the Martian climate are needed so that the resulting analysis of normal modes applies to the conditions of the InSight mission. However, this does
not require, however, as many simulations as would be expected from equivalent terrestrial
studies. Both the low thermal inertia of the Martian surface and the fast radiative timescale
of the thin Martian atmosphere imply a very low Martian-climate inter-annual variability
(except during the dust storm season, but InSight will land in 2018 at the end of this season)
(Read and Lewis 2004). Furthermore, given the key role played by the atmospheric-dust
loading in driving the Martian climate as well as the small inter-annual variability of this
parameter in the first half of the Martian year (Montabone et al. 2015) running one GCM
350
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
351
352
353
354
355
356
357
358
359
360
361
362
363
364
_####_
Prn:2019/01/22; 9:15
Page 8 of 25
p. 8/25»
Y. Nishikawa et al.
Fig. 4 Spectrum of GCM
pressure and observed pressure.
The observational data were
acquired by Mars Pathfinder
(LS = 170◦ ). The GCM spectrum
is computed for the same location
as the observational data of Mars
Pathfinder (LS = 0◦ ). The results
differ by an order of magnitude,
implying that GCMs cannot fully
explain the observed energy,
which is likely due to inadequate
modeling in GCMs of local-scale
turbulences. Thus, MBFs
calculated by using GCMs
should be regarded as a lower
limit of possible excitations
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
simulation for the year in which the InSight mission occurs will provide sufficiently accurate predictions for all Martian years (outside the dust storm season). The dust scenario used
thus corresponds to the Martian Year 27 which is a typical clear atmosphere scenario (dust
visible opacity around 0.2) with no effects from a global dust storm. GCM simulations start
at Ls = 0◦ (northern spring), which corresponds to the first months of the InSight mission
on Mars, and are performed for 20 Martian sols with a fixed dust profile. Given that gravity
waves are partly resolved in our 1◦ × 1◦ GCM simulation, we do not use the subgrid-scale
scheme to model the effect of those waves on the large-scale flow. The outputs of the model,
which are used as inputs for the modeling of the seismic moment density, as defined by
relation Eq. (1), and are surface pressure and temperature, as well as atmospheric density,
temperature, and winds.
The observed atmospheric status differs from that predicted by the GCM, despite the
location being the same. This is illustrated in Fig. 4 for the Pathfinder location. Parts of
these discrepancies are related to both the pitot pressure (associated with local wind) and to
the local-scale eddies,which are not resolved by the GCM but are discussed above.
As we mentioned above, this is a challenging problem because our approach relies on
GCMs. It is dependent on the environment in which the observation is made, the season, and
other factors, notably, that GCMs lose energy at Martian normal mode frequencies (Fig. 4).
387
388
389
3 Estimating the Amplitude of Martian Normal Modes
390
391
3.1 Internal Structure Model
392
393
394
395
396
397
398
399
400
Martian internal models have been discussed by many researchers; please see Panning et al.
(2017) and Smrekar et al. (this issue) for further descriptions and review. The biggest differences between these models are in the crustal thickness and core radius. The first estimates
are based on observation and laboratory experiments. A large amount of information about
the inner structure is based on gravitational observations and data on surface-soil properties.
With the constraints of mean density, moment of inertia, and k2 Love number, the inferred
radius of the core is estimated to be 1600 ± 200 km (Khan and Connolly 2007; Sohl and
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
401
402
403
404
405
406
407
408
409
410
411
412
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 9/25»
Page 9 of 25
_####_
Fig. 5 Vertical amplitude of
spheroidal fundamental normal
modes in solid part. Low-degree
normal modes are low frequency.
Large-amplitude areas
concentrate the models energy
and sensitivity. Low-degree
waves (angular order 2 to 9) are
sensitive at the core and deep
mantle. High-degree waves
(angular order 10 to 39) are
sensitive at the middle and
shallow part of the mantle. Each
mode is sensitive at a different
depth
413
414
415
416
417
418
419
420
421
422
423
424
425
426
Spohn 1997; Spohn et al. 2001; Rivoldini et al. 2011). However, only normal modes with
very long periods are sensitive to the core size and, furthermore, their amplitude near the
surface depends only weakly on core size. Consequently, the choice of internal model is not
critical for our estimate of MBF amplitude, and all models will provide similar results. In the
present work, we use the Sohl model to generate a reference internal structure. We calculate
fundamental spheroidal normal modes longer than 120 s, and the limitation of the period is
dictated by the GCM time step. This corresponds to calculating spheroidal normal modes
of angular orders between ℓ = 2 and ℓ = 39 and of radial order n = 0 which constrains the
core and the deep, middle mantle (Fig. 5).
427
428
3.2 Normal Modes Excitation
429
430
431
432
433
434
435
436
437
438
439
440
The theory of normal mode excitation and of normal mode summation is described in several papers (see, e.g., Lognonné 1991; Dahlen and Tromp 1998; Lognonné 2002) so we
only briefly summarize the approach here. Martian global oscillations are governed by the
gravito-elastic equation (when attenuation is neglected) or by the gravito-anelastic equation,
when anelastic processes are considered. Atmospheric normal modes may also be computed
by using the theory developed by Lognonné et al. (1998a). Lognonné et al. (2016) studied
attenuation processes in the Mars atmosphere due to both the viscosity and CO2 molecular
relaxation. They concluded that, for long periods (e.g., T ≥ 50 s), no significant atmospheric
attenuation should be expected in the lower atmosphere where the greatest excitation force
is concentrated, suggesting that an adiabatic approximation of the atmosphere suffices for
our modeling. The starting equation is
441
442
443
444
445
446
447
−ωk2 ρ0 uk + A(uk ) = f,
(3)
where ωk is the angular eigenfrequency, uk is the associated displacement eigenfunction,
A is the gravito-elastic operator (or gravito-anelasto operator), ρ0 is the unperturbed density
and f is the external force. The gravito-elastic operator A is defined for the internal structure
model and is given by:
448
449
450
A(u) = ∇(δTelastic − u · ∇T0 ) − g∇ · (ρ0 u) + ρ0 ∇Φ,
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
(4)
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
_####_
Prn:2019/01/22; 9:15
Page 10 of 25
p. 10/25»
Y. Nishikawa et al.
for the solid part, where Φ is the mass-redistribution potential, g is the gravity acceleration
due to gravity, T0 is the pre-stress and δTelastic is the elastic stresses. The operator in the
atmospheric part can be written by using T0 = −p0 I and
Telastic = −pHooke I = κ∇.v I,
(5)
where I is the identity tensor. The result is
A(u) = ∇(κ∇ · u + u · ∇p0 ) − g∇ · (ρ0 u) + ρ0 ∇Φ,
(6)
where κ is the bulk modulus and p0 is the equilibrium pressure. Following Lognonné et al.
(1994), we can write the momentum-density field ρv in the form
c˙k (t)uk (r),
(7)
ρ(r, t)v(r, t) = ρ0 (r)
k
which provides the displacement field u, which can be expressed to first order (and therefore
sufficiently far enough from the sources when on location) as
ck (t)uk (r).
(8)
u=
k
We get the differential equation with respect to time and the source function as
472
473
474
∂ 2 ck (t)
+ ωk2 ck (t) = Ψk (t),
∂t 2
(9)
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
where Ψk is the excitation force for each mode. To compute the normal mode excitation
forces, we consider the nonlinear equation of momentum conservation,
∂
(ρv) = −∇p + ρg − ∇ · (ρvv),
∂t
= −∇pHooke − ∇ · (ρu)g − ∇δpglut − ∇ · (ρvv),
= −A(u) − ∇δpglut − ∇ · (ρvv),
(10)
δpglut = p − p0 + κ∇ · u + u · ∇p0 ,
δpHooke = −κ∇ · u − u · ∇p0 ,
where p0 is the equilibrium pressure and pglut = p0 + δpglut is the GCM pressure as the GCM
filter acoustic waves. We assume here that the gravity term is linear and therefore excluded
from this term excitation processes associated with nonlinearity (Julián et al. 1998). If existing in the Martian atmosphere, (e.g., for atmospheric gravity waves), this might be therefore
an additional excitation term for frequencies below the 2 mHz acoustic cutoff (Lognonné
et al. 2016). However, transport terms are fully included in the Reynolds stress term ρvv.
Note also that the GCM pressure variations are much greater than those generated by the
flow, so that δpglut ≈ p − p0 . We project Eq. (10) onto a given normal mode, which gives
∂
dV uk · (ρv) = c¨k ,
∂t
V
= − dV uk · A(u) − dV uk · ∇δpglut + ∇ · (ρvv) ,
V
V
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
501
502
503
Mars’ Background Free Oscillations
V
dV uk ·
= −ωk2 ck (t) −
505
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
p. 11/25»
Page 11 of 25
_####_
∂
(ρv) = c¨k ,
∂t
504
506
Prn:2019/01/22; 9:15
V
(11)
dV uk · ∇δpglut + ∇ · (ρvv) .
Note that, in this projection, we shall decompose the volumetric integration for all partial
volumes (in practical terms, the atmosphere and interior) i separated by discontinuities j ,
such that
dV uk · ∇δpglut + ∇ · (ρvv)
V
=
i
=−
V
Si
dS [δpglut uk · ni + ρuk · v v · ni ] −
dV ∇uk · m +
j
Sj
Vi
dV ∇uk · m
dS [uk · ni δpglut ]+
−
(12)
where we assume no vertical winds at the interface. (Note that this approach is here limited
to a spherical surface and that additional excitation terms could occur due to topography,
in way similar to what occurs for Earth’s hum [see, e.g., Nishida (2017)]. The vector ni is
the normal to the surface, leaving the volume i, denoted + and − the volume against +,
for which the normal vector is oriented in the opposite directions. Following Lognonné and
Mosser (1993), we define here the flux-glut moment density tensor as
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
(13)
mij = (p + κ∇ · u)δ ij + ρv i v j .
Equation (12) simplifies further because we have not only continuity of pressure at the
atmosphere-solid interface, but also continuity of the Hooke pressure and the vertical amplitude of modes across this interface. We therefore identify the second term as the excitation
coefficient, which can be rewritten as
(14)
Ψk (t) = − dV uk · ∇ δpglut + ·(ρvv) = dV ǫk : m,
V
V
where ǫk is the strain tensor of the normal mode k. Figure 6 shows the pressure and Reynolds
stress spectra for an arbitrary point on the surface of Mars, as calculated by the GCM. Note
that the amplitudes decrease as 1/f . Most of the variability in the normal mode bandwidth
is associated with the Reynolds stress. All terms in Eq. (13) can be derived from the values
calculated by the GCM, thereby giving us a full description of the moment tensor based on
the GCM results.
540
541
3.3 Normal Modes Amplitudes in Atmosphere
542
543
544
545
546
547
548
549
Estimating the normal modes excitation coefficients Ψk (t) requires knowledge of the normal
mode amplitude in the atmosphere. Given the exponential decay of the density and the a priori thickness of the Martian boundary layer, most of the pressure glut release will occur in
the first 10 km so we focus on computing the normal modes at these relatively low altitudes.
Normal mode atmospheric amplitudes are affected by many factors, such as viscosity, radiative boundary, sound velocity, relaxation, resonance, etc. See Lognonné et al. (2016) for
more details on the different parameters affecting the amplitude modeling. At low altitudes
550
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
551
552
553
554
555
556
557
558
_####_
Prn:2019/01/22; 9:15
Page 12 of 25
p. 12/25»
Y. Nishikawa et al.
Fig. 6 GCM pressure and
Reynolds stress power spectrum.
The pressure glut is stronger
(weaker) than the Reynolds stress
at low (high) frequencies. At high
frequencies, the main
contribution to the pressure
variation is eddies, which
generate Reynolds stress through
wind variations
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
and in the bandwidth of normal modes, viscous and relaxation processes can be neglected,
as done by Lognonné and Johnson (2007). However both Lognonné and Johnson (2007,
2015) and Lognonné et al. (2016) assume the atmosphere to be spherically, symmetric and
rigorous. The three-dimensional computation of the normal modes requires a prohibitive
amount of computing time. Although the interior structure of Mars can be considered spherically symmetric, the acoustic properties of the Martian atmosphere depend strongly on local
time, with large variations between the night and day for the height scale, temperature, and
density, and therefore for the sound speed and acoustic impedance as well as the coupling
between ground and atmosphere. This is illustrated in Fig. 7, which shows the variation of
atmospheric scale height. Relative variations of ±20% with LS (or during one day at the
same longitude) are observed, which suggest that lateral variation of the atmospheric coupling should be accounted for in MBF calculations. Furthermore, Mars has larger lateral
atmospheric variations than does Earth. The main driver of atmospheric variation is temperature: on Mars, temperature variations are much greater than on Earth because of the
absence of oceans and atmospheric water vapor, which serve as effective thermal reservoirs
and stabilize the temperature. Thus, the MBF with three-dimensional atmospheric structure
must be estimated to better understand the signal levels of MBFs and perfect the simulations.
However, to account for the lateral variation while maintaining a realistic calculation time
requires some modeling and simplification. Thus we simplified the atmospheric structure by
neglecting viscosity, radiative boundaries, relaxation, and resonance.
In this paper, we propose to compute the amplitude of the atmospheric normal modes by
propagating the normal mode amplitude at the surface into the atmosphere. In other words,
instead of directly calculating normal modes in the atmosphere, we extrapolate the modes
calculated for the solid part of the planet into the atmosphere. For this purpose, we make the
following assumptions:
– the normal mode phase velocity is much greater than acoustic-wave velocity, leading to a
vertically longitudinal upward-moving acoustic wave;
– the normal modes interact asymptotically with the atmosphere and without significant
resonance between solid and atmosphere, which limits this approach to normal modes
above the atmospheric cutoff frequency.
600
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 13/25»
Page 13 of 25
_####_
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
Fig. 7 Daily variation of atmospheric scale height. The red (blue) areas are thick (thin). Because of the large
temperature variations, the scale height thickness changes from 8 to 12 km in one day. During the day, the
atmosphere expands to 12 km height, whereas it shrinks to 8 km height at night. This variation in scale height
directly affects the normal mode structure, and the isosphere model cannot describe this lateral effect
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
Given these two assumptions, the wave-propagation normal mode is
ρ(0)c(0)
−ikz z
uk (z) = uk (0) · e
×
,
ρ(z)c(z)
(15)
where kz is a vertical wave number and z is the altitude above the ground. The first factor
on the right side of the equation is the wave-propagation factor and the second factor is
the wave transmission ratio. With the wave-propagation method, we can describe the local
time and, more generally, how the normal mode amplitude in the atmosphere depends on
geography.
We compare the resulting amplitudes with those computed by using the more precise
model of Lognonné et al. (2016). For angular orders ℓ ≥ 10, the difference is less than
the effect of the lateral variation (see Fig. 8). However, for smaller angular orders and for
modes with frequencies close to or below the atmospheric cutoff, this simple propagation
fails and large discrepancies appear because of acoustic-wave reflection and resonance in the
atmosphere. Figure 8 shows the relative atmospheric energy for the fundamental spheroidal
modes. The energy peaks around 2.2 mHz, which corresponds to the angular order ℓ = 9.
This large coupling of the normal mode with angular order ℓ = 9 is normal if also seen on
the mode amplitude (Fig. 9).
Such modes, thus require exact computation. Note, however, that these modes are also
difficult to observe because of the expected high very low-frequency modes associated with
temperature fluctuation below 5 mHz.
650
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
_####_
Prn:2019/01/22; 9:15
Page 14 of 25
p. 14/25»
Y. Nishikawa et al.
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
Fig. 8 Kernel sensitivity of normal modes in air. Blue curves include the radiative boundary, viscosity, and
relaxation. Green curves include only wave propagation and transmission. Solid lines and dotted lines show
the real part and imaginary parts, respectively. For order n => 10, the gaps between two adjacent normal
modes are much less than the scale height
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
Fig. 9 Kernel sensitivity of normal modes in air. At the normal mode of angular order nine, the large gaps
occur because of the resonance between the atmosphere and the solid part. The normal mode cannot describe
this resonance, so the normal mode calculation only works for angular order 10. Below angular order 10,
normal modes are hard to detect because of the low-frequency thermal noise (see Sect. 4.2)
695
696
697
698
699
700
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
701
702
703
704
705
706
707
708
709
710
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 15/25»
Page 15 of 25
_####_
Fig. 10 Relative atmospheric
energy of surface wave in the
Mars atmosphere for Rayleigh
surface waves as a function of
frequency. Associated angular
order values range from ℓ = 2 to
ℓ = 39. The peak of the fraction
is at angular order 9 (near
2.2 mHz). At the first peak,
a strong resonance appears
between the atmosphere and the
solid part
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
Fig. 11 7 MBF acceleration as a
function of Martian days. The
MBF is the sum of fundamental
normal modes of angular order 2
to 39. The amplitude is given in
nanogals (10−11 m/s2 ). The
MBF continues for days because
the normal modes are excited by
continuous atmospheric
activities. The data for the first
three sols are meaningless due to
the GCM boundary conditions
and the stability of MBFs
728
729
730
731
732
733
734
4 Results and Discussion
735
736
4.1 Normal Modes Amplitude
737
738
739
740
741
742
743
744
745
746
747
748
749
The MBF is obtained by summing the fundamental Rayleigh normal modes, with the amplitude of each computed by using Eqs. (8) and (11). For this study, we summed the fundamental normal mode for degrees ℓ = 2 to ℓ = 39. The amplitude of the MBF signal is a few
nanogals (10−11 m/s2 ), as shown in Fig. 11. This amplitude is greater than that reported by
Lognonné and Johnson (2007, 2015), which is likely because of the larger angular orders
and therefore larger frequencies that they used. Since the signal is the sum of all normal
modes, specific peak amplitudes are more clearly shown in Fig. 12. The lowest-frequency
peak (0.4 mHz) is a normal mode of angular order 2, whereas the highest-frequency peak
(8.3 mHz) is a normal mode of angular order 39. Note, however, that the amplitudes are
well modeled only for frequencies greater than 2 mHz. We also observe a gradual increase
of the amplitude with frequency up to 5 mHz, after which the amplitude of the peaks remains
constant.
750
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
751
752
753
754
755
756
757
758
759
760
761
762
763
_####_
Prn:2019/01/22; 9:15
Page 16 of 25
p. 16/25»
Y. Nishikawa et al.
Fig. 12 Calculated spectrum of
acceleration power density of
MBFs. The MBFs are estimated
by using GCMs and the internal
structure given by the Sohl
model. The normal mode
frequencies come from power
spectral density calculations.
Each peak is a normal mode;
there are 38 peaks in this figure,
ranging from angular order 2 to
39. The amplitude grows with
frequency up to 5 mHz and is
then remains constant for
frequencies ≥5 mHz
764
765
766
767
768
4.2 Normal Modes: Detection and Seismometer Performance
769
770
771
772
773
774
775
MBF amplitudes are quite small and so can be detected only by using instruments with
high sensitivity and low noise. Let us consider whether these modes can be detected in the
POS output of the InSight SEIS VBB sensors. This output has a flat acceleration gain in the
frequency bandwidth of normal modes (see Lognonné et al., this issue).
The least significant bit (LSB) of the InSight SEIS seismometer for differential output is
776
LSB =
777
2 × Voltage
.
Gain
(16)
778
779
780
781
The seismometer voltage is ±25 V, the acquisition dynamics is 24 bit, and the gain is
≈104 V/DU for low gain mode and about 4.5 times greater for high gain mode. Therefore,
the LSBs are
782
783
784
785
786
787
788
789
790
791
792
793
794
2 × 25
= 30 nanogal,
104 × 224
2 × 25
=
= 6.6 nanogal.
4.5 × 104 × 224
LSBLG =
LSBHG
The amplitude of MBFs is only nanogals and the LSBs are therefore greater. The real ground
acceleration signal contains MBF, as well as thermal drift, thermal and instrument noise, and
possibly seismic signal. For a more realistic evaluation, we approximate the daily temperature variation associated with the temperature sensitivity of the sensors with an ideal sinusoidal curve and superpose it on the MBFs. Because this drift is much greater than 1 LSB,
it enhances the MBFs up to a detectable level. The sinusoidal thermal-noise model is given
by
795
796
797
798
799
(17)
Atemp (t) = γ δT 1 − cos
2πt
,
Dsol
(18)
where Atemp is the equivalent acceleration of temperature variations, t is local time, Dsol is
the duration of a Martian day, γ is the sensitivity per degree of the VBB in the acceleration
800
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
801
802
803
804
805
806
807
808
809
810
811
812
813
814
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 17/25»
Page 17 of 25
_####_
Fig. 13 Spectrum of power
acceleration density of MBF,
instrumental noise, and
environmental noise. Above
5 mHz (angular order 22),
thermal noise is much greater
than the MBF signal. Below
6.5 mHz (angular order 30), the
MBF signal overcomes both
instrumental and thermal noise.
During nighttime, the InSight
SEIS seismometer in high-gain
mode may detect the MBF
signal. The MBF signals detected
carry information about the
middle, shallow part of the
Martian mantle
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
unit, and δT is the temperature variation over one Martian day. The temperature sensitivity
γ is set to 10−5 m/s2 /K and daily temperature variation δT is 4 K, which corresponds to
the variations expected during winter (Mimoun et al. 2016; Lognonné et al., this issue). The
amplitude of this thermal daily variation is therefore 4.0 × 106 nanogal, which is six orders
of magnitude greater than MBF signals.
We simulate the signals by adding the MBF signal to the thermal drift, and then simulate
the digitization by converting this acceleration signal into bits. Subtracting from the signal a
sol-periodic sine wave with amplitudes computed by using the least-squares method simulates data processing. For both low gain and high gain, clear MBFs signal are retrieved from
the residual, which demonstrates the capability of the acquisition system to use stochastic
amplification of the noise to detect signals smaller than the LSB.
The signal remains much smaller than the expected instrument noise, which is the superposition of the instrument self-noise and of the residual of both the subtracted temperature
drift and the potential pressure decorrelation (Murdoch et al. 2017). This is illustrated in
Fig. 13, which compares the MBF spectrum to the instrument self-noise in both low gain
and high gain and to the expected thermal-drift signal. In the 5 to 10 mHz band, which
contains a large part of the target bandwidth for the Martian hun search (5 to 20 mHz) and
the bandwidth in which our modeling hypothesis and the GCMs have acceptable errors, the
MBF signal is expected to be 5 to 10 below the instrument self-noise, depending on the gain.
Stacking this signal over the mission duration will then allow successful detection.
The amplitude of ideal thermal noise model is 4.0 × 106 nanogal. The acceleration is
0.4 × 106 DU in low-gain mode or 1.4 × 106 DU in high gain mode. The thermal noise
acceleration is far greater than the MBF signal. Thus, no signal is detected by the InSight
SEIS seismometer in low-gain mode. In high-gain mode, we can detect 1 DU signal several
times per half Martian day. However, ideal thermal noise can kick up to 1 DU after the
decimal point. After subtracting the digitized thermal noise, more MBF signals are detected
using the InSight SEIS seismometer in both low- and high-gain mode. Although the MBF
signal is hidden by the large thermal noise, after subtracting the digitized thermal noise, we
capture a greater part of the MBF signal.
To improve the signal to noise ratio and detect the MBF normal modes, we further process
data as follows. First, we create a trace that contains MBF signal and noise. As explained
previously, noise can be decomposed as self-noise, thermal noise and LSB noise. We com-
850
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
851
852
853
854
855
_####_
Prn:2019/01/22; 9:15
Page 18 of 25
p. 18/25»
Y. Nishikawa et al.
Fig. 14 Spectrum amplitude of
each two sol long segment for
MBF traces (blue) and self-noise
traces (red) in the top figure and
for S4 traces (black) in the
bottom figure
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
pute self-noise as the inverse Fourier transform of a spectrum whose amplitude follows the
station sensitivity as a function of frequency and whose phase is random. This self-noise is
added to the MBF seismogram to create a trace S1. Thermal noise is then computed as a
sinus function over one sol with an amplitude of 10−4 m/s2 to simulate the seismic acceleration generated by thermal drift. This thermal noise is added to trace S1 to create trace
S2. LSB noise takes into account the least significant bit (5.81/4 10−11 m/s2 ). Trace S2 is
modified so that the only signal above the LSB is kept and this trace is called S3. Finally,
we assume that the thermal drift is known and it is subtracted to trace S3 to generate to the
final trace S4.
The seismogram S4 is then cut into segments of two sols with one sol overlap, starting
at sol 4. Figure 14 (top) displays spectra amplitude of each two sol-long segments for the
MBF signal (blue lines) and self-noise (red lines) and confirms that the amplitude of the
MBF normal modes is below the self-noise. Figure 14 (bottom) shows the spectra amplitude
of each segments of trace S4 (black lines). Normal modes are not visible in the amplitude
spectra of the different data segments.
Data are then processed as follows. Phase auto-correlation is computed for each segment
and then stacked using the phase weighted stack method following the method developed
in Schimmel et al. (2011), Ventosa et al. (2017), Schimmel et al. (2018). For comparison,
we apply the same processing to MBF traces, S1 traces (MBF+self noise) and S4 traces
(MBF+self noise+LSB). Figure 15 (top) show the amplitude spectra of each stack. Normal
mode peaks are clearly visible for the MBF stack amplitude spectrum between 1.5 and
8 mHz. Some modes can also be detected for the stack of traces S4 at frequencies higher
than 4 mHz.
In order to enhance the signal to noise ratio, we further select the Rayleigh wave train
windows on the stack of auto-correlograms by setting to 0 the rest of the signal as in Deen
et al. (2017). We keep signal around 0-time lag and around each surface wave train R1
to R3. We keep 6 minutes around 0 lag time and select R1 to R3 wave trains considering
that their group velocities are between 3.8 and 5.7 km/s. Figure 15 (bottom) displays the
new amplitude spectra, and we observe that the normal mode peaks are now clearly visible
for frequencies above 2.5 mHz for the MBF stack and also for data with realistic noise
(self-noise or MBF+self noise). Adding longer time series will improve the signal to noise
ratio and therefore we expect that normal modes excited by the Mars atmosphere should be
detectable.
900
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 19/25»
Page 19 of 25
_####_
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
Fig. 15 Normalized spectrum amplitude of the stack of autocorrelograms without surface wave train selection (top) and after surface wave train selection (bottom). The blue, black and red curves correspond to MBF,
MBF+self noise and MBF+self noise + LBS noise (S4) data respectively
922
923
924
925
926
927
928
929
930
931
932
4.3 Estimate of Seismic Magnitude of MBF Signal
We now not only estimate the MBF amplitude in terms of seismic magnitude, but also check
that our estimate of the MBFs is comparable to that obtained by applying a simple energetic
budget, following the approach of Kobayashi and Nishida (1998b).
For a first estimate, we assume that the release of acoustic and seismic energy into atmospheric activities is driven by solar flux and, furthermore, that all the energy of planetary
background free oscillations comes from solar flux. This maximum energy may be expressed
as
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
WM = FE ×
DE
DM
2
2
× (1 − AM ) × πRM
,
(19)
where WM is the rate at which the Sun transfers energy to the Mars surface, F is the solar
flux above Earth’s atmosphere (FE = 1360 W/m2 ), D is the distance from the Sun (DM =
1.52DE ), A is the bound albedo (AM = 0.25), subscripts E and M refer to Earth and Mars,
respectively, and RM is the radius
√ of Mars. This solar energy is converted to seismic energy,
with a coherency duration of T Q, where Q is the quality coefficient of the mode and T is
its period. Thus the MBF’s energy (EMBF ) is
EMBF = WM × s ×
Q × T × C,
(20)
where s = 0.01 is an acoustic efficiency (which can be achieved for high-altitude winds of
40 to 50 m/s, assuming an efficiency depending on M 3 , where M is Mach number, [see, e.g.,
Goldreich and Kumar (1988)]). The constant C is the energy coupling ratio between the atmosphere and the solid part of Mars (a typical value is C = 5 10−6 , (see, e.g., Lognonné and
Johnson 2015)), Q is seismic attenuation of MBF’s (typical value Q = 100), T is period of
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
951
952
953
954
955
956
957
958
959
_####_
Prn:2019/01/22; 9:15
Page 20 of 25
p. 20/25»
Y. Nishikawa et al.
Fig. 16 Spectrum power
acceleration density of MBF and
free oscillations. The Martian
atmosphere excites the MBFs and
the free oscillations are excited
by imaginary seismic events. The
magnitude of the events are
M4.5, 5.0, and 5.5. MBFs are
close to M5.0 oscillations. This
result is consistent with rough
estimates
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
MBFs (typical value T = 200 s). We then use the energy-magnitude relation of (Gutenberg
and Richter 1956):
log E = 4.8 + 1.5M,
(21)
where M is the magnitude of the seismic event. A rough estimation of the MBF magnitude
gives M = 4.9.
We now compare the amplitude obtained by our GCM modeling with that excited by seismic activity. This is achieved by comparing the spectrum of the MBFs with those of quakes
that occur at an epicentral distance of 90◦ . We find amplitudes of MBFs close to those generated by a M5.0 free oscillation (Fig. 16), while the estimates based on the GCM calculation
are consistent with those based on solar flux. The results are also consistent with those estimated from Earth’s hum, which has been estimated to be equivalent to a daily earthquake of
magnitude 5.75–6 (Rhie and Romanowicz 2004). Given the earthbound albedo of 0.306, the
solar flux for Earth is eightfold that of Mars, which corresponds to magnitude 0.6 greater
and therefore an extrapolation of 5.3. Nevertheless, all these estimates of magnitudes, although comparable, are less than the magnitude of 5.9 (i.e., 1018 Nm) which is considered
by several studies as a prerequisite for detecting normal modes. Therefore, we will focus our
next analysis on the possibility of coherent stacking for MBFs, which, for a Mars year of
687 days, might lead to an increase in amplitudes by ≈26ǫ, where ǫ is the fraction of daily
coherent hum. For a sol-to-sol coherency exceeding 0.4, this might lead to an order of magnitude increase in signal, allowing signals to possibly peak out of the noise after stacking
data over a year.
991
992
4.4 Weather Correlation and Stacking Method
993
994
995
996
997
998
999
The Martian surface has low thermal inertia because it is devoid of oceans. Thus, the diurnal surface-temperature cycle is very large. Combined with the very thin atmosphere, this
makes thermal tides a prominent element of diurnal variability on Mars. This is further reinforced by the equatorial location of InSight. Mars has a clear dichotomy whereby low lands
dominate the southern hemisphere, while the northern hemisphere is dominated by highlands. The boundary of the dichotomy is close to the equator and this dichotomy contributes
1000
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 21/25»
Page 21 of 25
_####_
Fig. 17 Pressure correlations in
normal mode frequency band.
Large correlations are diurnal and
semi-diurnal. On a given day, the
large correlations are the
sunrise-sunrise and sunset-sunset
correlations. Semi-diurnal
correlations are sunrise-sunset
and sunset-sunrise (semi-diurnal
thermal tide). The periodic
excitation force generates
high-efficiency stacking. This
precise phenomenon occurs on
Mars but not on Earth
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
Fig. 18 Stacking efficiency for
one week of data. Blue curve is
MBF power density. Red line is
MBF power density after one
week of stacking. Daily stacking
leads to large, clear peaks. The
longer MBFs are stacked, the
larger and sharper the normal
mode peaks become. One week
of stacking already leads to a
significant improvement in the
signal-to-noise ratio, which
demonstrates that Martian
stacking is far more effective
than terrestrial stacking. As a
result of such efficient stacking
for Mars, some MBF peaks are
likely to be detected by the
InSight SEIS seismometer
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
to the diurnal cycle through slope-induced circulations, although the day-to-day variability
associated with baroclinic waves is small (Spiga et al. 2018). Thus, provided we consider
atmospheric variability within a given season, a significant daily repetition of atmospheric
temperature, wind, and pressure should be the norm. This will correlate strongly with the
diurnal range in our seismic computations derived from GCM simulations (Fig. 17), which
means that stacking is likely to be an efficient way to increase the MBF amplitude above that
of the self-noise or other noise sources. We tested this hypothesis with a one-week stack from
the GCM simulation. The 7 sol-long data were divided into 7 time series of 24.6 hours. The
results shown in Fig. 18 indicate a threefold increase in the amplitudes of stacks over 7 sols.
Although extrapolation over a long period will require further GCM modeling, and because
the sol-to-sol correlation might weaken over weeks, these preliminary results suggest that
stacking the seismic signal over the sol time might enable significant to very significant amplification of the Mars hum. Furthermore, detecting the normal mode frequencies will allow
the inversion of the Mars upper mantle, as already illustrated by Panning et al. (2017). The
position of the normal mode peaks depends on the internal structure, as shown on Fig. 19.
These simulations were done with two different internal structure models for Mars and the
results show that the resolution after the stacking suffices to adequately shift associated with
the structures.
1050
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
_####_
Prn:2019/01/22; 9:15
Page 22 of 25
p. 22/25»
Y. Nishikawa et al.
Fig. 19
Internal-model-dependent MBFs.
We use two internal models of
Mars, which results in differing
peak positions. The difference in
peak positions corresponds to the
difference in the internal structure
of Mars, and the seismometer on
InSight can detect this difference
in frequencies. If we can detect
these peaks, we can determine
the internal structure of Mars
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
5 Conclusion
This study evaluates the magnitude of the signal of Mars’ background free oscillations by
using high-precision GCMs. Given the limitation of GCMs, the values obtained should be
regarded as a lower limit of the possible amplitude of background free oscillations expected
on Mars.
To obtain a realistic evaluation, we make several assumptions that are based on observations and theory to simplify the model and take into account the three-dimensional atmospheric structure of Mars.
Given these models and assumptions, the results indicate that MBF amplitudes are likely
to be on the order of nanogals, which is consistent with previous studies, such as Kobayashi
and Nishida (1998b), Lognonné and Johnson (2007). We also confirmed that the values
obtained are consistent with solar flux. These results imply that the values obtained represent
the typical strength of MBF signals, although the amplitudes may differ because of the
assumptions used, such as those involving the internal structure and atmospheric model.
Given the level of MBF signal and the level of noise from the InSight SEIS instruments,
detecting MBF signals with the InSight SEIS instruments will remain difficult. However, because the detection strategy exploits the daily repetition of the Martian atmosphere, stacking
is found to be a powerful tool to overcome this problem.
The results should be viewed as a minimum estimate for MBF detection by the InSight
SEIS seismometer. Knowledge of the Martian MBFs should allow us to estimate the deep
internal structure of Mars and can be one of successes of the InSight mission.
Acknowledgements The authors are grateful for the support of CNES for the development of the SEIS
experiment and its scientific support, and to the ANR for supporting the project through ANR SEISMARS.
Y.K. acknowledges the support of the CNES and JSPC for his Ph.D. support. P.L. and A.S. acknowledge the
support of IUF. This is IPGP contribution number xx and InSight contribution number yy.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
1094
1095
1096
1097
1098
1099
References
G. Backus, M. Mulcahy, Moment tensors and other phenomenological descriptions of seismic sources—
I. Continuous displacements. Geophys. J. Int. 46(2), 341–361 (1976)
1100
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 23/25»
Page 23 of 25
_####_
H. Benioff, J.C. Harrison, L. LaCoste, W.H. Munk, L.B. Slichter, Searching for the Earth’s free oscillations.
J. Geophys. Res. 64(9), 1334–1337 (1959)
H. Benioff, F. Press, S. Smith, Excitation of the free oscillations of the Earth by earthquakes. J. Geophys.
Res. 66(2), 605–619 (1961)
B.A. Bolt, J.S. Derr, Free bodily vibrations of the terrestrial planets. Vistas Astron. 11, 69–102 (1969)
A. Colaïtis, A. Spiga, F. Hourdin, C. Rio, F. Forget, E. Millour, A thermal plume model for the Martian
convective boundary layer. J. Geophys. Res., Planets 118(7), 1468–1487 (2013)
F.A. Dahlen, J. Tromp, Theoretical Global Seismology (Princeton University Press, Princeton, 1998)
M. Deen, E. Wielandt, E. Stutzmann, W. Crawford, G. Barruol, K. Sigloch, First observation of the Earth’s
permanent free oscillations on ocean bottom seismometers. Geophys. Res. Lett. 44(21), 10,988–10,996
(2017). https://doi.org/10.1002/2017GL074892
A.M. Dziewonski, D.L. Anderson, Preliminary reference Earth model. Phys. Earth Planet. Inter. 25(4), 297–
356 (1981)
F. Forget, S. Lebonnois, Global climate models of the terrestrial planets. Comp. Climatol. Terr. Planets 1,
213–229 (2013)
F. Forget, F. Hourdin, R. Fournier, C. Hourdin, O. Talagrand, M. Collins, S.R. Lewis, P.L. Read, J-P. Huot,
Improved general circulation models of the Martian atmosphere from the surface to above 80 km. J. Geophys. Res., Planets 104(E10), 24155–24175 (1999)
P. Goldreich, D.A. Keeley, Solar seismology. II. The stochastic excitation of the solar p-modes by turbulent
convection. Astrophys. J. 212, 243–251 (1977)
P. Goldreich, P. Kumar, The interaction of acoustic radiation with turbulence. Astrophys. J. 326(1), 462–478
(1988)
M.P. Golombek, W.B. Banerdt, K.L. Tanaka, D.M. Tralli, A prediction of Mars seismicity from surface
faulting. Science 258(5084), 979–981 (1992)
T.V. Gudkova, V.N. Zharkov, Mars: interior structure and excitation of free oscillations. Phys. Earth Planet.
Inter. 142(1–2), 1–22 (2004)
B. Gutenberg, C.F. Richter, Earthquake magnitude, intensity, energy, and acceleration: (second paper). Bull.
Seismol. Soc. Am. 46(2), 105–145 (1956)
R.M. Haberle, M.M. Joshi, J.R. Murphy, J.R. Barnes, J.T. Schofield, G. Wilson, M. Lopez-Valverde,
J.L. Hollingsworth, A.F.C. Bridger, J. Schaeffer, General circulation model simulations of the Mars
Pathfinder atmospheric structure investigation/meteorology data (1999)
P. Julián, M. Jordán, A. Desages, Canonical piecewise-linear approximation of smooth functions. IEEE Trans.
Circuits Syst. I, Fundam. Theory Appl. 45(5), 567–571 (1998)
V.G. Kaydash, M.A. Kreslavsky, Y.G. Shkuratov, G. Videen, J.F. Bell III., M. Wolff, Measurements of winds
on Mars with Hubble Space Telescope images in 2003 opposition. Icarus 185(1), 97–101 (2006)
B. Kenda, P. Lognonné, A. Spiga, T. Kawamura, S. Kedar, W.B. Banerdt, R. Lorenz, D. Banfield, M.
Golombek, Modeling of ground deformation and shallow surface waves generated by Martian dust
devils and perspectives for near-surface structure inversion. Space Sci. Rev. 211(1–4), 501–524 (2017)
A. Khan, J.A.D. Connolly, Constraining the composition and thermal state of Mars from inversion of geophysical data. LPI Contributions, No. 1353 (2007), p. 3013
M. Knapmeyer, J. Oberst, E. Hauber, M. Wählisch, C. Deuchler, R. Wagner, Working models for spatial
distribution and level of Mars’ seismicity. J. Geophys. Res., Planets 111(E11) (2006)
N. Kobayashi, K. Nishida, Atmospheric excitation of planetary free oscillations. J. Phys. Condens. Matter
10(49), 11557 (1998a)
N. Kobayashi, K. Nishida, Continuous excitation of planetary free oscillations by atmospheric disturbances.
Nature 395(6700), 357–360 (1998b)
R.L. Kovach, D.L. Anderson, The interiors of the terrestrial planets. J. Geophys. Res. 70(12), 2873–2882
(1965)
G. Laske, R. Widmer-Schnidrig, Treatise on Geophysics, Theory and Observations: Normal Mode and Surface Wave Observations, vol. 1.04 (Elsevier, Amsterdam, 2015)
P. Lognonné, Normal modes and seismograms in an anelastic rotating Earth. J. Geophys. Res., Solid Earth
96(B12), 20309–20319 (1991)
P. Lognonné, Normal modes of the Earth and planets, in Handbook on Earthquake and Engineering Seismology (2002), pp. 125–147
P. Lognonné, Planetary seismology. Annu. Rev. Earth Planet. Sci. 33, 571–604 (2005)
P. Lognonné, D. Giardini, The NetLander very broad band seismometer. Planet. Space Sci. 48, 1289–1302
(2000)
P. Lognonné, C.L. Johnson, Planetary Seismology, vol. 10.03 (Elsevier, Amsterdam, 2007)
P. Lognonné, C.L. Johnson, Planetary Seismology, vol. 10.03 (Elsevier, Amsterdam, 2015)
P. Lognonné, B. Mosser, Planetary seismology. Surv. Geophys. 14(3), 239–302 (1993)
1150
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
_####_
Prn:2019/01/22; 9:15
Page 24 of 25
p. 24/25»
Y. Nishikawa et al.
P. Lognonné, B. Mosser, F.A. Dahlen, Excitation of jovian seismic waves by the Shoemaker-Levy 9 cometary
impact. Icarus 110(2), 180–195 (1994)
P. Lognonné, J.G. Beyneix, W.B. Banerdt, S. Cacho, J.F. Karczewski, M. Morand, Ultra broad band seismology on InterMarsNet. Planet. Space Sci. 44(11), 1237–1249 (1996)
P. Lognonné, E. Clévédé, H. Kanamori, Computation of seismograms and atmospheric oscillations by
normal-mode summation for a spherical Earth model with realistic atmosphere. Geophys. J. Int. 135(2),
388–406 (1998a)
P. Lognonné, V.N. Zharkov, J.F. Karczewski, B. Romanowicz, M. Menvielle, G. Poupinet, B. Brient, C.
Cavoit, A. Desautez, B. Dole et al., The seismic optimism experiment. Planet. Space Sci. 46(6–7), 739–
747 (1998b)
P. Lognonne, T. Spohn, D. Mimoun, S. Ulamec, J. Biele, GEP-ExoMars: a geophysics and environment
observatory on Mars, in 37th Annual Lunar and Planetary Science Conference, vol. 37 (2006)
P. Lognonné, W.T. Pike, V. Tong, R. Garcia, Planetary seismometry, in Extraterrestrial Seismology (2015),
pp. 36–48
P. Lognonné, F. Karakostas, L. Rolland, Y. Nishikawa, Modeling of atmospheric-coupled Rayleigh waves on
planets with atmosphere: from Earth observation to Mars and Venus perspectives. J. Acoust. Soc. Am.
140(2), 1447–1468 (2016)
J-B. Madeleine, F. Forget, E. Millour, L. Montabone, M.J. Wolff, Revisiting the radiative impact of dust on
Mars using the LMD global climate model. J. Geophys. Res., Planets 116(E11) (2011)
G.M. Martínez, C.N. Newman, A. De Vicente-Retortillo, E. Fischer, N.O. Renno, M.I. Richardson, A.G.
Fairén, M. Genzer, S.D. Guzewich, R.M. Haberle, A.-M. Harri, O. Kemppinen, M.T. Lemmon, M.D.
Smith, M. de la Torre-Juárez, A.R. Vasavada, The modern near-surface Martian climate: a review of insitu meteorological data from viking to curiosity. Space Sci. Rev. 212(1), 295–338 (2017). https://doi.
org/10.1007/s11214-017-0360-x
R. Millot-Langet, E. Clévédé, P. Lognonné, Realistic long period synthetic seismograms using normal modes
of a 3d anelastic elliptic rotating Earth, in EGS General Assembly Conference Abstracts, vol. 27 (2002)
D. Mimoun, N. Murdoch, P. Lognonné, T. Pike, K. Hurst et al. (SEIS Team), The seismic noise model of the
InSight mission to Mars. Space Sci. Rev. (2016)
L. Montabone, F. Forget, E. Millour, R.J. Wilson, S.R. Lewis, B. Cantor, D. Kass, A. Kleinböhl, M.T. Lemmon, M.D. Smith et al., Eight-year climatology of dust optical depth on Mars. Icarus 251, 65–95 (2015)
J.P. Montagner, Treatise on Geophysics, Deep Earth Structure: Upper Mantle Structure: Global Isotropic
and Anisotropic Elastic Tomography, vol. 1.19 (Elsevier, Amsterdam, 2015)
N. Murdoch, B. Kenda, T. Kawamura, A. Spiga, P. Lognonné, D. Mimoun, W.B. Banerdt, Estimations of the
seismic pressure noise on Mars determined from Large Eddy Simulations and demonstration of pressure
decorrelation techniques for the InSight mission. Space Sci. Rev. 211(1–4), 457–483 (2017)
J.R. Murphy, C.B. Leovy, J.E. Tillman, Observations of Martian surface winds at the Viking Lander 1 site.
J. Geophys. Res., Solid Earth (1978–2012) 95(B9), 14555–14576 (1990)
T. Navarro, J-B. Madeleine, F. Forget, A. Spiga, E. Millour, F. Montmessin, A. Määttänen, Global climate
modeling of the Martian water cycle with improved microphysics and radiatively active water ice clouds.
J. Geophys. Res., Planets 119(7), 1479–1495 (2014)
K. Nishida, Earth’s background free oscillations. Annu. Rev. Earth Planet. Sci. 41, 719–740 (2013a)
K. Nishida, Global propagation of body waves revealed by cross-correlation analysis of seismic hum. Geophys. Res. Lett. 40(9), 1691–1696 (2013b)
K. Nishida, Source spectra of seismic hum. Geophys. J. Int. 199(1), 416–429 (2014)
K. Nishida, Ambient seismic wave field. Proc. Jpn. Acad. Ser. B 93(7), 423–448 (2017)
K. Nishida, N. Kobayashi, Y. Fukao, Resonant oscillations between the solid earth and the atmosphere. Science 287(5461), 2244–2246 (2000)
K. Nishida, J-P. Montagner, H. Kawakatsu, Global surface wave tomography using seismic hum. Science
326(5949), 112 (2009)
E.A. Okal, D.L. Anderson, Theoretical models for Mars and their seismic properties. Icarus 33(3), 514–528
(1978)
M.P. Panning, P. Lognonné, W.B. Banerdt, R. García, M. Golombek, S. Kedar, B. Knapmeyer-Endrun, A.
Mocquet, N.A. Teanby, J. Tromp et al., Planned products of the Mars structure service for the InSight
mission to Mars. Space Sci. Rev. 211(1–4), 611–650 (2017)
A. Petrosyan, B. Galperin, S.E. Larsen, S.R. Lewis, A. Määttänen, P.L. Read, N. Renno, L.P.H.T. Rogberg,
H. Savijärvi, T. Siili et al., The Martian atmospheric boundary layer. Rev. Geophys. 49(3) (2011)
R.J. Phillips, R.E. Grimm, Martian seismicity, in Lunar and Planetary Science Conference, vol. 22 (1991),
pp. 1061–1062
A. Pottier, F. Forget, F. Montmessin, T. Navarro, A. Spiga, E. Millour, A. Szantai, J.-B. Madeleine, Unraveling
the Martian water cycle with high-resolution global climate simulations. Icarus (2017). https://doi.org/
10.1016/j.icarus.2017.02.016
1200
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF
AUTHOR’S PROOF
« SPAC 11214 layout: Small Condensed v.2.1 file: spac579.tex (ramune) class: spr-small-v1.4 v.2019/01/11
« SPAC 11214 doctopic: ReviewPaper numbering style: ContentOnly reference style: aps»
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
Prn:2019/01/22; 9:15
Mars’ Background Free Oscillations
p. 25/25»
Page 25 of 25
_####_
P.L. Read, S.R. Lewis, The Martian Climate Revisited: Atmosphere and Environment of a Desert Planet
(Springer, Berlin, 2004)
J. Rhie, B. Romanowicz, Excitation of Earth’s continuous free oscillations by atmosphere–ocean–seafloor
coupling. Nature 431(7008), 552 (2004)
J. Rhie, B. Romanowicz, A study of the relation between ocean storms and the Earth’s hum. Geochem.
Geophys. Geosyst. 7(10) (2006)
A. Rivoldini, T. Van Hoolst, O. Verhoeven, A. Mocquet, V. Dehant, Geodesy constraints on the interior
structure and composition of Mars. Icarus 213(2), 451–472 (2011)
B.A. Romanowicz, B.J. Mitchell, Treatise on Geophysics, Deep Earth Structure: Q of the Earth from Crust
to Core, vol. 1.19 (Elsevier, Amsterdam, 2015)
M. Schimmel, E. Stutzmann, J. Gallart, Using instantaneous phase coherence for signal extraction from ambient noise data at a local to a global scale. Geophys. J. Int. 184(1), 494–506 (2011). https://doi.org/10.
1111/j.1365-246X.2010.04861.x
M. Schimmel, E. Stutzmann, S. Ventosa, Low-frequency ambient noise autocorrelations: waveforms and
normal modes. Seismol. Res. Lett. 89(4), 1488–1496 (2018). https://doi.org/10.1785/0220180027
J.T. Schofield, J.R. Barnes, D. Crisp, R.M. Haberle, S. Larsen, J.A. Magalhaes, J.R. Murphy, A. Seiff, G.
Wilson, The Mars Pathfinder atmospheric structure investigation/meteorology (ASI/MET) experiment.
Science 278(5344), 1752–1758 (1997)
F. Sohl, T. Spohn, The interior structure of Mars: implications from SNC meteorites. J. Geophys. Res., Planets
102, 1613–1635 (1997)
S.C. Solomon, D.L. Anderson, W.B. Banerdt, R.G. Butler, P.M. Davis, F.K. Duennebier, Y. Nakamura, E.A.
Okal, R.J. Phillips, Scientific rationale and requirements for a global seismic network on Mars (1991)
A. Spiga, Elements of comparison between Martian and terrestrial mesoscale meteorological phenomena:
katabatic winds and boundary layer convection. Planet. Space Sci. 59(10), 915–922 (2011)
A. Spiga, F. Forget, A new model to simulate the Martian mesoscale and microscale atmospheric circulation:
validation and first results. J. Geophys. Res., Planets 114(E2) (2009)
A. Spiga, S.R. Lewis, Martian mesoscale and microscale wind variability of relevance for dust lifting. Mars
5, 146–158 (2010)
A. Spiga, D. Banfield, N.A. Teanby, F. Forget, A. Lucas, B. Kenda, J.A. Rodriguez Manfredi, R. WidmerSchnidrig, N. Murdoch, M.T. Lemmon, R.F. Garcia, L. Martire, Ö. Karatekin, S. Le Maistre, B. Van
Hove, V. Dehant, P. Lognonné, N. Mueller, R. Lorenz, D. Mimoun, S. Rodriguez, É. Beucler, I. Daubar,
M.P. Golombek, T. Bertrand, Y. Nishikawa, E. Millour, L. Rolland, Q. Brissaud, T. Kawamura, A.
Mocquet, R. Martin, J. Clinton, É. Stutzmann, T. Spohn, S. Smrekar, W.B. Banerdt, Atmospheric science
with InSight. Space Sci. Rev. 214, 109 (2018). https://doi.org/10.1007/s11214-018-0543-0
T. Spohn, M.H. Acuña, D. Breuer, Geophysical constraints on the evolution of Mars. Space Sci. Rev. 96,
231–262 (2001)
N. Suda, K. Nawa, Y. Fukao, Earth’s background free oscillations. Science 279(5359), 2089–2091 (1998)
T. Tanimoto, Continuous free oscillations: atmosphere-solid earth coupling. Annu. Rev. Earth Planet. Sci.
29(1), 563–584 (2001)
T. Tanimoto, The oceanic excitation hypothesis for the continuous oscillations of the Earth. Geophys. J. Int.
160(1), 276–288 (2005)
T. Tanimoto, J. Um, Cause of continuous oscillations of the Earth. J. Geophys. Res., Solid Earth 104(B12),
28723–28739 (1999)
V.C.H. Tong, R.A. García, Extraterrestrial Seismology (Cambridge University Press, Cambridge, 2015)
T. Van Hoolst, V. Dehant, F. Roosbeek, P. Lognonné, Tidally induced surface displacements, external potential
variations, and gravity variations on Mars. Icarus 161(2), 281–296 (2003)
S. Ventosa, M. Schimmel, E. Stutzmann, Extracting surface waves, hum and normal modes: time-scale
phase-weighted stack and beyond. Geophys. J. Int. 211(1), 30–44 (2017). https://doi.org/10.1002/
2017JB014354
S. Watada, H. Kanamori, Acoustic resonant oscillations between the atmosphere and the solid earth during
the 1991 Mt. Pinatubo eruption. J. Geophys. Res. Solid Earth 115(B12) (2010)
S. Watada, G. Masters, Oceanic excitation of the continuous oscillations of the Earth, in AGU Fall Meeting
Abstracts (2001), art. S32A-0620
S.C. Webb, The Earth’s hum is driven by ocean waves over the continental shelves. Nature 445(7129), 754
(2007)
J.H. Woodhouse, A. Deuss, Theory and Observations – Earth’s Free Oscillations, vol. 1.03 (Elsevier, Amsterdam, 2015)
V.N. Zharkov, T.V. Gudkova, On the dissipative factor of the Martian interiors. Planet. Space Sci. 45(4),
401–407 (1997)
1249
1250
Journal ID: 11214, Article ID: 579, Date: 2019-01-22, Proof No: 1, UNCORRECTED PROOF