Draft version May 29, 2018

Typeset using LATEX twocolumn style in AASTeX62

Cassini Ring Seismology as a Probe of Saturn’s Interior I: Rigid Rotation

Christopher Mankovich,1 Mark S. Marley,2 Jonathan J. Fortney,1 and Naor Movshovitz1
1 Department of Astronomy and Astrophysics, University of California Santa Cruz
2 NASA Ames Research Center
arXiv:1805.10286v1 [astro-ph.EP] 25 May 2018

(Received May 18, 2018)

Submitted to ApJ

ABSTRACT
Seismology of the gas giants holds the potential to resolve long-standing questions about their internal
structure and rotation state. We construct a family of Saturn interior models constrained by the gravity
field and compute their adiabatic mode eigenfrequencies and corresponding Lindblad and vertical
resonances in Saturn’s C ring, where more than twenty waves with pattern speeds faster than the ring
mean motion have been detected and characterized using high-resolution Cassini Visual and Infrared
Mapping Spectrometer (VIMS) stellar occultation data. We present identifications of the fundamental
acoustic modes of Saturn that appear to be the origin of these observed ring waves, and use their
observed pattern speeds and azimuthal wavenumbers to estimate the bulk rotation period of Saturn’s
interior to be 10 hours 35.3+2.3
−1.7 minutes (median and 5%/95% quantiles), significantly faster than
the period measured from Voyager spacecraft radiometry. Structure in the resulting pattern speed
residuals suggests the presence of differential rotation in Saturn’s outer envelope.

Keywords: planets and satellites: individual (Saturn) – planets and satellites: interiors – planets and
satellites: rings

1. INTRODUCTION ground by Mosser et al. (1993) and Gaulme et al. (2011).

The prototypical gas giants Jupiter and Saturn offer
an opportunity to study the processes at work during
planet formation and the chemical inventory of the pro-
tosolar disk, and also constitute astrophysical labora-
tories for warm dense matter. Inferences about these
planets’ composition and structure rely on interior mod-
els that are chiefly constrained by the their observed
masses, radii and shapes, surface abundances, and grav-
ity fields (Stevenson 1982; Fortney et al. 2016). While
the latter have been measured to unprecedented preci-
sion by Juno at Jupiter and the Cassini Grand Finale
at Saturn, there is a need to identify independent obser-
vational means of studying the interiors, and seismol-
ogy using the planets’ free oscillations appears to be the
most promising such avenue.
Preliminary detections of Jovian oscillations were
made through radial velocity measurements from the
Corresponding author: Chris Mankovich
cmankovich@ucsc.edu
The power spectrum of Gaulme et al. exhibited excess
power at millihertz frequencies and enabled the measure-
ment of Jupiter’s large frequency separation describing
the spacing of acoustic overtone (p-)modes of adjacent
radial order, yielding a value consistent with expec-
tations from interior models. However, the frequency
resolution was not sufficient to identify specific normal
modes responsible for the power in the spectrum, a nec-
essary step before the frequencies can be used to probe
the interior in detail.
Saturn provides a unique opportunity for seismic
sounding of a Jovian interior owing to its highly or-
dered ring system, wherein gravity perturbations from
Saturn’s free oscillations can interact resonantly with
ring orbits. Saturn ring seismology is the focus of this
work.
1.1. Background
The concept of ring seismology was first developed in
the 1980s. Stevenson (1982) suggested that Saturnian
inertial oscillation modes, for which the Coriolis force is
the restoring force, could produce regular density per-
2 Mankovich et al.
turbations within the planet that might resonate with
ring particle orbits and open gaps or launch waves, but
he did not calculate specific mode frequencies. Later
in the decade in a series of abstracts, a thesis, and pa-
pers Marley, Hubbard and Porco further developed this
idea. Marley et al. (1987), relying on Saturn oscillation
frequencies computed by Vorontsov (1981), suggested
that acoustic mode oscillations, which differ from iner-
tial modes in their restoring force is ultimately pres-
sure, could resonate with ring particle orbits in the C
ring. They recognized that mode amplitudes of only
a few meters would be sufficient to perturb the rings.
Marley & Hubbard (1988) focused on low angular de-
gree ` f -modes which have no radial nodes in displace-
ment from surface to the center of the planet (unlike
p-modes) as the modes which had the potential to pro-
vide the most information about the deep interior of a
giant planet. Marley et al. (1989) compared the pre-
dicted locations resonance locations of such modes with
newly discovered wave features in the C ring found in
radio occupation data by Rosen (1989). They suggested
that the Maxwell gap and three wave features found by
Rosen which had azimuthal wave numbers and propaga-
tion directions consistent with such resonances were in
fact produced by Saturnian f -modes with ` ≤ 4. As we
will summarize below, we now know that these specific
f -mode–ring feature associations were correct, although
the story for the ` = 2 and ` = 3 waves is complicated
by g-mode mixing.
These ideas were ultimately presented in detail in
Marley (1990, 1991) and Marley & Porco (1993). Marley
computed the sensitivity of Saturn oscillation frequen-
cies to various uncertainties in Saturn interior models,
including core size and regions with composition gradi-
ents, and discussed the sensitivity of ring resonance loca-
tions to these uncertainties. As we will show below, the
overall pattern of resonance locations within the rings
first presented in Marley (1990) agrees well with subse-
quent discoveries. While Marley recognized the impact
of regions with non-zero Brunt-Väisälä frequency N on
f -mode frequencies and the possibility of g-modes (for
which the restoring force is buoyancy), he did not con-
sider mode mixing between f - and g-modes. Marley &
Porco (1993) presented the theory of resonances between
planetary oscillation modes and rings in detail and de-
rived expressions for the torque applied to the rings at
horizontal (Lindblad) and vertical resonances and com-
pared these torques to those of satellites. They also sug-
gested several more specific ring feature-oscillation mode
associations, many of which have subsequently turned
out to be correct. Marley and Porco concluded by not-
ing that because the azimuthal wave numbers of the
Rosen wave features were uncertain, only additional ob-
servations by the planned future Saturn mission Cassini
could ultimately test the hypothesized oscillation mode–
ring feature connection. Consequently there was an es-
sentially two-decade pause in ring seismology research
until those results became available.
Optical depth scans of the C ring from Cassini ra-
dio occultations and Ultraviolet Imaging Spectrograph
stellar occultations presented by Colwell et al. (2009)
and Baillié et al. (2011) confirmed all the unexplained
waves reported by Rosen et al. (1991b) and identifying
many more. Hedman & Nicholson (2013) followed up
with VIMS stellar occultations, combining scans taken
by Cassini at different orbital phases to determine wave
pattern speeds and azimuthal wavenumbers m at outer
Lindblad resonances, making seismology of Saturn us-
ing ring waves possible for the first time. As alluded to
above, the detection of multiple close waves with m = 2
and m = 3 waves deviated from the expectation for the
spectrum of pure f -modes. In light of this result, Fuller
et al. (2014) investigated the possibility of shear modes
in a solid core, finding that rotation could mix these
core shear modes with the f -modes and in principle ex-
plain the observed fine splitting, although they noted
that some fine tuning of the model was required. The
most compelling model for the fine splitting to date was
presented by Fuller (2014), who showed that a strong
stable stratification outside Saturn’s core would admit
g-modes that could rotationally mix with the f -modes
and rather robustly explain the number of strong split
m = 2 and m = 3 waves at Lindblad resonances, and
roughly explain the magnitude of their frequency sepa-
rations.
Subsequent obervational results from the VIMS data
came from Hedman & Nicholson (2014), who detected a
number of additional waves including an m = 10 wave
apparently corresponding to Saturn’s ` = m = 10 f -
mode. French et al. (2016) characterized the wave in
the ringlet within the Maxwell gap and argued it to be
driven by Saturn’s ` = m = 2 f -mode, supporting the
prediction by Marley et al. (1989). The remainder of C
ring wave detections that form the observational basis
for our work are the density waves reported by Hedman
et al. (2018, in preparation) and the bending waves
reported by French et al. (2018, in preparation).
1.2. This work
Here we seek to systematically understand the ring
wave patterns associated with Saturn’s normal modes.
In particular, we aim to identify the modes responsible
for each wave, make predictions for the locations of other
Saturnian resonances in the rings, and ultimately assess
 Saturn’s Seismological Rotation
what information these modes carry about Saturn’s in-
terior. We describe the construction of Saturn interior
models in §2. §3 summarizes our method for solving for
mode eigenfrequencies and eigenfunctions, as well as our
accounting for Saturn’s rapid rotation. In §4 we recapit-
ulate the conditions for Lindblad and vertical resonances
with ring orbits and describe which f -modes can excite
waves at each. §5 presents the main results, namely
f -mode identifications and a systematic comparison of
predicted f -mode frequencies to the pattern speeds of
observed waves and its implications for Saturn’s inte-
rior, principally its rotation. The separate question of
mode amplitudes and detectability of ring waves is ad-
dressed in §6.1, which also lists the strongest predicted
waves yet to be detected. Discussion follows in §7 and
we conclude with §8.
2. INTERIOR MODELS
Our hydrostatic planet interior models are computed
using a code based on that of Thorngren et al. (2016)
with a few important generalizations. To model arbi-
trary mixtures of hydrogen and helium, we implement
the equation of state of Saumon et al. (1995) (the version
interpolated over the plasma phase transition, hence-
forth “SCvH-i”). Heavier elements are included using
the ab initio water EOS of French et al. (2009), extend-
ing the coverage to T < 103 K using the analytical model
of Thompson (1990) for water. The density ρ(Y, Z) is
obtained assuming linear mixing of the three compo-
nents following
Z 1−Z
ρ−1 (Y, Z) = + , (1)
ρZ ρHHe (Y )
where in turn
Y 1−Y
ρ−1
HHe (Y ) = + . (2)
ρHe ρH
Here Y and Z are the mass fractions of helium and heav-
ier elements, respectively, and the densities ρH , ρHe , and
ρZ are tabulated as functions of pressure P and temper-
ature T in the aforementioned equations of state.
The outer boundary condition for our interior mod-
els is simply a fixed temperature at P = 1 bar, namely
T1 = 140 K, close to the value derived by Lindal et al.
(1985) from Voyager radio occultations and mirroring
that used in previous Saturn interior modeling efforts
(e.g., Nettelmann et al. 2013). The envelope is assumed
to be everywhere efficiently convective so that the deeper
temperature profile is obtained by integrating the adia-
batic temperature gradient:
Z mr
T (m ≥ Mc ) = T1 + ∇ad (P, T, Y ) dln P , (3)
M
3
with the core itself assumed isothermal at T (Mc ). Here
mr denotes the mass coordinate,
and the adiabatic tem-
perature gradient ∇ad ≡ ∂∂ ln ln T
P ad is assumed to be that
of the hydrogen-helium mixture alone and is obtained
directly from SCvH-i.
Following common choices for models of Saturn’s in-
terior (e.g., Nettelmann et al. 2013), the distribution
of constituent species with depth follows a three-layer
piecewise homogeneous structure: heavy elements are
partitioned into a core devoid of hydrogen and helium
(Z = 1) and a two-layer envelope with outer (inner)
heavy element mass fraction Z1 (Z2 ). The helium con-
tent is likewise partitioned with outer (inner) helium
mass fraction Y1 (Y2 ) subject to the constraint that the
mean helium mass fraction of the envelope match the
protosolar nebula abundance Y = 0.275. The Z and
Y transitions are located at a common pressure level
P12 , a free parameter conceptually corresponding to the
molecular-metallic transition of hydrogen, although in
SCvH-i itself this is explicitly a smooth transition. We
only consider Z2 > Z1 and Y2 > Y1 to avoid density
inversions and to reflect the natural configuration of a
differentiated planet.
The particular choice of this three-layer interior struc-
ture model is motivated by the desire for a minimally
complicated model that simultaneously (a) satisfies the
adopted physically-motivated EOS, (b) includes enough
freedom to fit Saturn’s low-order gravity field J2 and J4 ,
and (c) does not introduce significant convectively stable
regions in the envelope, such as those that might arise in
cases where composition varies continuously. Require-
ment (c) precludes a viable class of configurations for
Saturn’s interior (e.g., Fuller 2014, Vazan et al. 2016),
but it significantly simplifies the formalism and interpre-
tation because in this case the normal modes in the rele-
vant frequency range are limited to the acoustic modes1 .
As will be discussed in §5 below, a spectrum of purely
acoustic modes is sufficient to explain the majority of
the spiral density and bending waves identified in the C
ring that appear to be Saturnian in origin.
2.1. Gravity field
We generate rigidly rotating, oblate interior models by
iteratively solving for the self-consistent shape and mass
distribution throughout the interior using the theory of
figures formalism (Zharkov & Trubitsyn 1978), retaining
1 While the isothermal cores of our models do admit g-modes,
the maximum Brunt-Väisälä frequency attained there is only
N ≈ σ0 , where σ0 = (GM/R)1/2 is Saturn’s dynamical/natural
frequency. Since g-modes
√ have frequencies at most N , and f -mode
frequencies follow σ ≈ `σ0 (Gough 1980), g-modes in such a core
will not undergo avoided crossings with the ` ≥ 2 f -modes.
4 Mankovich et al.

terms in the expansion of the background potential to P (bar)

fourth order in the small parameter mS = Ω2S a3 /GM , 107 106 105
where ΩS is the uniform rotation rate, a = 60268 km
101
is Saturn’s equatorial radius (Lindal et al. 1985), and
GM = 37931207.7 cm3 s−2 is Saturn’s total gravita-

ρ (g cm−3)
tional mass (Jacobson et al. 2006). To this end we use
the shape coefficients given through O(m4 ) by Nettel-
mann (2017) and implement a similar algorithm. 100
With Z1 , Z2 , Y1 , P12 , and m as free parameters, an
initially spherical model is relaxed to its rotating hydro-
static equilibrium configuration. The mean radii of level
surfaces are adjusted during iterations such that the 10−1
equatorial radius of the outermost level surface for a con- 0.00 0.25 0.50 0.75 1.00
verged model matches that of Saturn. As the mean radii r/RS
are adjusted and the densities are recalculated from the
EOS, the total mass of the model necessarily changes; P (bar)
107 106 105
therefore the core mass Mc is simultenously adjusted
30
over the course of iterations such that the converged
model matches Saturn’s total mass. These models in-
clude 4096 zones, the algorithm adding zones late in it-

erations if necessary to speed convergence to the correct
total mass.
The values for the gravity used for generating inte-
rior models are those of Jacobson et al. (2006), ap-
propriately normalized to our slightly different adopted
reference equatorial radius of 60268 km according to
0
J2n = (a/a0 )J2n . Although dramatically more precise
harmonics obtained from the Cassini Grand Finale or-
bits will soon be published, the values of J2 and J4
from Jacobson et al. (2006) are already precise to a
level beyond that which can be used to put meaning-
ful constraints on the deep interior using our fourth-
order theory of figures, where in practice solutions are
only obtained with numerical precision at the level of
|δJ2 /J2 | ≈ |δJ4 /J4 | . 10−4 .
For the purpose of fitting the gravity field, we cre-
ate models using mS = 0.13963 corresponding to the
10h 39m 24s (10.657h) rotation period measured from
Voyager kilometric radiation and magnetic field data
by Desch & Kaiser (1981). We sample interior mod-
els from a bivariate normal likelihood distribution in J2
and J4 using emcee (Foreman-Mackey et al. 2013) as-
suming a diagonal covariance matrix for these gravity
harmonics. Because the numerical precision to which
our theory of figures can calculate J2 exceeds its ob-
servational uncertainty, the former is used in our like-
lihood function. We take uniform priors on Z1 and Z2
subject to the constraint that 0 < Z1 < Z2 < 1, a uni-
form prior on 0 < Y1 < 0.275, and a uniform prior over
0.5 Mbar < Ptrans < 2 Mbar. The mass distributions
and sound speeds for models in this sample are illus-
trated in Figure 1.
cs (km s−1)
20
10
0
0.00 0.25 0.50 0.75 1.00
r/RS
Figure 1. Saturn interior models with two-layer envelopes
of varying Y and Z distributions, surrounding pure-Z cores.
Models are sampled based on J2 and J4 from Jacobson et al.
(2006). Mass density (top panel) and sound speed (bottom
panel) are shown as functions of the mean radii of level sur-
faces (bottom horizontal axes) and pressure coordinate (top
horizontal axes).
3. MODE EIGENFREQUENCIES AND
EIGENFUNCTIONS
Our approach is to perform the pulsation calculation
for spherical models corresponding to the converged the-
ory of figures models, with the various material param-
eters defined on the mean radii r of level surfaces. The
influence of Saturn’s rapid rotation is accounted for after
the fact using a perturbation theory that expresses the
full solutions in the presence of Coriolis and centrifugal
forces and oblateness in terms of simple superpositions
of the solutions obtained in the non-rotating case.
For spherical models, we solve the fourth-order sys-
tem of equations governing linear, adiabatic oscillations
(Unno et al. 1989) using the open source GYRE stellar
 Saturn’s Seismological Rotation
oscillation code suite (Townsend & Teitler 2013). The
four assumed boundary conditions correspond to the en-
forcement of regularity of the eigenfunctions at r = 0
and the vanishing of the Lagrangian pressure perturba-
tion at the planet’s surface r = R (Unno et al. 1989,
Section 18.1). As will be discussed in §4, comparison
with the full set of observed waves in the C ring requires
f -modes with angular degree in the range ` = 2 − 14,
and we tabulate results through for the f -modes through
` = 15.
In what follows, we adopt the convention that m > 0
corresponds to prograde modes—those that propagate
in the same sense as Saturn’s rotation—so that the time-
dependent Eulerian perturbation to, e.g., the mass den-
sity corresponding to the (`mn) normal mode in the
planet is written as
ρ0`mn (r, θ, ϕ, t) = ρ0`mn (r)Y`m (θ, ϕ)e−iσ`mn t ,
where σ`mn is the mode frequency in the frame rotat-
ing with the planet, and r, θ, and ϕ denote radius, co-
latitude, and azimuth respectively. Analogous relations
hold with the pressure P or gravitational potential Φ
in place of density. The Y`m (θ, ϕ) are the spherical har-
monics, here defined in terms of the associated Legendre
polynomials P`m as
1/2
(` − |m|)!
 
m+|m| 2` + 1
Y`m (θ, ϕ) =(−1) 2
4π (` + |m|)!
× P`m (cos θ)eimϕ .
The solution for the displacement itself has both radial
and horizontal components, with the total displacement
vector given by
  
∂ 1 ∂
ξ(r, θ, ϕ, t) = ξr (r) r̂ + ξh (r) θ̂ + ϕ̂
∂θ sin θ ∂ϕ
× Y`m (θ, ϕ).
3.1. Rotation
In reality, Saturn’s eigenfrequencies are significantly
modified by the action of Saturn’s rapid rotation be-
cause of Coriolis and centrifugal forces and the ellipticity
of level surfaces. We account for these following the per-
turbation theory given by Vorontsov & Zharkov (1981)
(see also Saio 1981) and later generalized by Vorontsov
(1981) to treat differential rotation, using the eigenfunc-
tions obtained in the non-rotating case as basis functions
for expressing the full solutions. In this work we calcu-
late corrected eigenfrequencies for a range of rotation
rates, treating Saturn as a rigidly rotating body.
5
Denoting by σ̃`mn the eigenfrequency obtained for the
`mn mode in the non-rotating case, we write the cor-
rected eigenfrequency as an expansion to second order
in the small parameter
ΩS
λ≡ (7)
σ̃`mn
so that
σ`mn = σ̃`mn 1 + σ`mn,1 λ + σ`mn,2 λ2 .


(8)
Here the dimensionless factors σ`mn,1 and σ`mn,2 cap-
ture the effects of Coriolis and centrifugal forces and
ellipticity of the planet’s figure in the presence of ro-
tation. In the limit of slow rotation, it is appropriate
to truncate the expansion at first order in λ, in which
case Equation 8 reduces to the well-known correction of
(4) Ledoux (1951) in which rotation breaks the frequency’s
degeneracy with respect to the azimuthal order m. Ex-
pressions for σ`mn,1 and σ`mn,2 are obtained through the
perturbation theory; in practical terms they are inner
products involving the zeroth-order eigenfunctions and
operators describing the Coriolis and centrifugal forces
and ellipticity. This formalism is constructed to retain
the separability of eigenmodes in terms of the spheri-
cal harmonics Y`m , so that each corrected planet mode
may still be uniquely specified by the integers `, m and
n and the expressions 4 and 6 hold for the corrected
eigenfunctions. Finally, we note that corrections related
to the distortion of level surfaces require knowledge of
the planetary figure as a function of depth, and these
(5)
are provided directly by the theory of figures. Further
details on the calculation of these rotation corrections
are given by Marley (1990), which the present imple-
mentation follows closely.
Besides the intrinsic effect on mode eigenfrequencies
resulting from Saturn’s rotation through Equation 8, a
final aspect of Saturn ring seismology that is sensitive
to the planet’s rotation rate is the Doppler correction
(6) necessary to transform a frequency σ`mn in the frame
inertial
rotating with the planet into the frequency σ`mn seen
by an observer in inertial space, namely
inertial
σ`mn = σ`mn + mΩS . (9)
This frequency in inertial space can in turn be related
to the pattern speed by
1 inertial
Ωpat = σ , (10)
m `mn
which is suitable for direct comparison with the pattern
speeds observed for waves in the rings.
As an illustration of the relative importance of these
various contributions to the modeled pattern speed, we
6 Mankovich et al.
10 3 an ILR an orbiting particle instead overtakes the slower
δΩpat (deg day−1)
Sectoral modes
Zeroth-order frequency
First-order intrinsic
Second-order intrinsic
102 Frame shift
5 10
`
Figure 2. Magnitude of the contributions made to the
modeled pattern speed by each of the four terms in Equa-
tion 11. Quantities take positive values with the exception
of the second-order intrinsic correction (green, dashed curve)
which takes negative values. Second-order (Coriolis, centrifu-
gal, ellipticity) rotation corrections are important at the level
of > 50 deg day−1 for derived pattern speeds.
may substitute the frequency expansion 8 into 9 and 10
to write
σ̃`mn σ`mn,1 ΩS σ`mn,2 Ω2S
Ωpat = + + + ΩS .
m m mσ̃`mn
These four contributions are shown in Figure 2, which
demonstrates that the second-order rotation corrections
effect the pattern speeds at the level of & 50 deg day−1
for modes with ` below 15. These corrections are thus
essential for comparison with the observed wave pattern
speeds, whose uncertainties are no larger than approxi-
mately 0.1 deg day−1 (P.D. Nicholson, private commu-
nication).
4. SATURNIAN F -MODES IN THE RINGS
4.1. Resonance conditions
The condition for a Lindblad resonance is (Murray &
Dermott 1999)
m(Ω − Ωpat ) = ±qκ
with the upper sign corresponding an inner Lindblad
resonance (ILR) and the lower sign corresponding to an
outer Lindblad resonance (OLR), and with q a posi-
tive integer. Taking the lower sign in Equation 12 to
consider an OLR, it physically represents the condition
that the perturbing pattern speed overtakes an orbiting
ring particle once every q/m epicycles. This prograde
forcing in phase with the ring particles’ epicycles leads
to a deposition of angular momentum that may launch
a spiral density wave propagating toward the planet. At
perturbing pattern once every q/m epicycles, leading to
a removal of angular momentum that may launch a spi-
ral density wave that propagates away from the planet.
Such waves are common in Saturn’s rings at mean mo-
tion resonances with Saturnian satellites.
Vertical resonances follow an analogous resonance con-
dition, namely that the perturbing pattern speed rela-
tive to the ring orbital frequency is related to the char-
acteristic vertical frequency µ in the rings by
15 m(Ω − Ωpat ) = ±bµ, (13)
where b is a positive integer and the vertical frequency
µ(r) in the ring plane can be obtained from (Shu et al.
1983)
µ2 + κ2 = 2Ω2 . (14)
As with Lindblad resonances, there exist both inner
and outer vertical resonances, depending on the sign of
Ω − Ωpat . IVRs are common in the rings as a result of
Saturnian satellites, e.g., those whose inclinations pro-
vide resonant vertical forcing. In what follows we limit
our attention to OLRs and OVRs because in practice,
the prograde f -modes have pattern speeds that exceed
Ω(r) throughout the C ring.
The orbital and epicyclic frequencies Ω and κ for orbits
(11)
at low inclination and low eccentricity can generally be
written as a multipole expansion in terms of the zonal
gravitational harmonics J2n , namely
∞
" #
GM X  a 2n
Ω2 (r) = 3 1 + A2n J2n (15)
r n=1
r
and
∞
" #
GM X  a 2n
2
κ (r) = 3 1 + B2n J2n , (16)
r n=1
r
with the J2n values scaled to the appropriate reference
equatorial radius a. The A2n and B2n are rational coef-
ficients and are tabulated by Nicholson & Porco (1988).
We use the even harmonics of Iess et al. (2018, submit-
ted) through J12 for the purposes of locating resonances
(12) in the ring plane, although the gravity field only affects
radial locations of resonances and has no bearing on f -
mode pattern speeds. We thus use the latter for quanti-
tative comparison between model f -modes and observed
waves.
The above relations constitute a closed system allow-
ing the comparison of planet mode frequencies to ob-
served waves at resonances in the rings. In cases where
we do compare resonance locations, the resonant radius
for a Lindblad or vertical resonance is obtained by nu-
merically solving Equation 12 or 13.
 Saturn’s Seismological Rotation
4.2. Which modes for which resonances?
Each planet mode can generate one of either density
waves or bending waves. The type of wave that the
`mn mode is capable of driving depends on its angu-
lar symmetry, and in particular the integer ` − m =
(0, 1, 2, 3, . . .). Modes with even ` − m are permanently
symmetric with respect to the equator, and so are not
capable of any vertical forcing. However, they are an-
tisymmetric with respect to their azimuthal nodes, and
so do contribute periodic azimuthal forcing on the rings.
The reverse is true of modes with odd ` − m, whose per-
turbations are antisymmetric with respect to the equa-
tor and so do contribute periodic vertical forcing on ring
particles. Meanwhile their latitude-average azimuthal
symmetry as experienced at the equator prevents them
from forcing ring particles prograde or retrograde.
In what follows, we restrict our attention to prograde
f -modes, namely the normal modes with m > 0 and
n = 0. Acoustic modes with overtones (n > 0; p-modes)
are not considered because that they contribute only
weakly to the external potential perturbation due to self-
cancellation in the volume integral of the Eulerian den-
sity perturbation; see Equation A11. We further limit
our consideration to prograde modes because while f -
modes that are retrograde in the frame rotating with
the planet can in principle be boosted prograde by Sat-
urn’s rotation, we find that the resulting low pattern
speeds (. 500 deg day−1 ) would place any Lindblad or
vertical resonances beyond the extent of even the A or
B rings. Finally, azimuthally symmetric modes (m = 0)
do not lead to Lindblad or vertical resonances.
5. RESULTS FOR RIGID ROTATION
Figure 3 summarizes the OLR and OVR locations
of prograde model Saturn f -modes with ` − m be-
tween zero and five, together with locations of 17
inward-propagating density waves and four outward-
propagating bending waves observed in Cassini VIMS
data. A visual comparison in this diagram provides a
strong indication that the f -modes are responsible for
the majority of the wave features shown. In particu-
lar, we can make unambiguous identifications for the
f -modes at the origin of 10 of the 17 density waves, and
all four of the bending waves; these visual identifications
are summarized in Table 1.
The remaining seven density waves at m = 2 and
m = 3 exhibit frequency splitting that is likely at-
tributable to mixing with deep g-modes as proposed
by Fuller (2014), and which our model, lacking a sta-
ble stratification outside the core, does not attempt to
address. We note that in each of these multiplets, the
outermost observed wave is consistent with the OLR we
7
predict for the sectoral Saturnian f -mode. In the m = 2
case, this corresponds to the wave in the ringlet embed-
ded within the Maxwell gap French et al. (2016). Of the
three m = 2 OLRs found closer in to the planet, wave
B9 may be particularly challenging to explain due to
its large splitting from the other three, although mixing
with deep g-modes seems to be a promising interpreta-
tion.
The model f -modes whose resonance locations coin-
cide with the remainder of the observed waves include
the rest of the sectoral sequence up to ` = m = 10, as
well as seven non-sectoral modes with ` − m taking all
values from 1 through 5. These waves are evidently the
result of time-dependent tesseral harmonics associated
with Saturn’s nonradial oscillations.
Although general agreement for these is evident at
the broad scale of Figure 3, the observed wave pat-
tern speeds are known to a precision better than
0.1 deg day−1 for the weakest waves yet measured (P.
D. Nicholson, private communication). This high pre-
cision warrants a closer inspection of the pattern speed
residuals with respect to our predictions. What follows
in the remainder of this section is an analysis of these
residuals and their dependence on the assumed interior
model and rotation rate.
5.1. The seismological rotation rate
Saturn’s bulk rotation rate has to date been deduced
from a combination of gravity field and radiometry data
from the Pioneer, Voyager and Cassini spacecrafts (e.g.,
Desch & Kaiser 1981, Gurnett et al. 2005, Giampieri
et al. 2006, Anderson & Schubert 2007). Along differ-
ent lines, Helled et al. (2015) optimized interior models
to the observed gravity field and oblateness to extract
the rotation rate. Since we have demonstrated that the
frequencies of Saturnian f -modes depend strongly on
ΩS through the influence of the Coriolis and centrifu-
gal forces and the ellipticity of level surfaces, a natural
question is, what interior rotation rate is favored by the
waves detected so far that appear to be associated with
modes in Saturn’s interior?
Given an observed C-ring wave with a pattern speed
Ωobs
pat that appears to be associated with a predicted Sat-
urn model f -mode resonance with pattern speed Ωpat
and azimuthal order matching the observed number of
spiral arms m, we calculate the pattern speed resid-
ual ∆Ωpat ≡ Ωpat − Ωobs pat . For each Saturn interior
model and rotation rate considered, we calculate the
RMS value of ∆Ωpat over the set of mode-wave asso-
ciations that we have identified in Table 1. The result-
ing curves are shown in Figure 4 for rotation periods
between 10 h 30 m and 10 h 42 m. The relation between
8 Mankovich et al.
m=` 4 ` 2 ` m=` 5 ` 3 ` 1

10 OLR 10 OVR
2⇡/⌦S
8 8 10h 30m
10h 35m
m

m
6 6 10h 40m
10h 45m

4 4 HN13,14
F+16
2 2 H+18
D ring C ring D ring C ring F+18

70 75 80 85 90 70 75 80 85 90
3 3
Saturn-centric radius (10 km) Saturn-centric radius (10 km)
Figure 3. Locations of resonances with our model Saturn’s f -modes (vertical ticks) and wave features observed in Saturn’s
C ring using stellar occultations in Cassini VIMS data (open symbols; see references in Table 1). The number of spiral arms
m (or equivalently, the azimuthal order of the perturbing planet mode) is shown versus distance from Saturn’s center in the
ring plane. Left panel: Outer Lindblad resonances, which can excite inward-propagating spiral density waves in the rings. The
three roughly vertical model sequences correspond to modes with m = `, m = ` − 2, and m = ` − 4 from right to left. The
three observed m = 3 density waves are offset vertically for clarity. Right panel: Outer vertical resonances, which can excite
outward-propagating bending waves in the rings. The three vertical model sequences correspond to m = ` − 1, m = ` − 3, and
m = ` − 5 from right to left. Model resonances are colored by the assumed Saturn rotation rate as described in the legend.

RMS ∆Ωpat and ΩS always exhibits a distinct minimum, 2π/PS (deg day−1)
owing to the strongly correlated response of the f -mode 820 815 810 805 800
frequencies to varying ΩS . In particular, the predicted 10 1.0
RMS ∆Ωp (deg day−1)

pattern speeds uniformly increase with faster Saturn ro-

tation. 8 0.8

Period CDF
The precise location of the minimum RMS pattern
speed residual with respect to assumed Saturn rotation 6 0.6
period depends on the interior model chosen. Given that 0.4
4 DK 81
we have no reason to prefer one model in our set over
G+05
any of the others on the basis of their modeled gravity 2 G+06 0.2
fields, and because the minimum magnitudes of RMS AS 07
H+15
pattern speed residuals attained by all interior mod- 0 0.0
els are virtually the same (the full set spanning only 30 35 40 45
2.9 − 3.3 deg day−1 RMS residuals), we assign equal PS − 10 h (min)
weights to each interior model in our estimate of Sat-
urn’s rotation period. Figure 4. Saturn’s rotation rate from fits to the set of ob-
For the posterior distribution of interior models con- served waves identified with Saturnian f -modes. RMS pat-
sidered, the distribution of optimized rotation rates may tern speed residuals across the full set of waves are shown
as a function of Saturn’s assumed rotation rate. Each grey
be summarized as PS = 10.588+0.039−0.029 h where the lead- curve corresponds to a single interior model from the sam-
ing value corresponds to the median and the upper ple shown in Figure 1. The thick blue curve shows the cu-
(lower) error corresponds to the 95% (5%) quantile. mulative distribution of rotation periods minimizing RMS
In terms of a pattern speed, this can be expressed as residuals for each model. The resulting distribution can be
+2.99
2π/PS = 816.00−2.25 deg day−1 . summarized in terms of its median and 95% and 5% quan-
Although these seismological calculations vary the as- tiles as PS = 10.588+0.039
−0.029 h corresponding to a pattern speed
−1
sumed rotation rate, the underlying interiors randomly 2π/PS = 816.00+2.99
−2.25 deg day . Vertical dashed lines indi-
cate Saturn rotation rates from the literature, references to
sampled against J2 and J4 using the theory of figures
which are given in §5.1.
as described in §2.1 assumed the Desch & Kaiser (1981)
Voyager rate, in principle an inconsistency of the model.
As a diagnostic we generate a new sample from the grav-
 Saturn’s Seismological Rotation 9

Table 1. C-ring wave patterns and model f -mode associations

Observed Model
Reference Wavea Symbolb m Ωpat c Type ` Ωpat c |∆Ωpat |c
French et al. (2018, in preparation) B9  2 2169.3 OLR — — —
Hedman & Nicholson (2013) W84.64† ♦ 2 1860.8 OLR — — —
Hedman & Nicholson (2013) W87.19† ♦ 2 1779.5 OLR — — —
French et al. (2016) Maxwell† 4 2 1769.2 OLR 2 1762.3 − 1771.8 2.6 − 6.9
Hedman & Nicholson (2013) W82.01† ♦ 3 1736.6 OLR — — —
Hedman & Nicholson (2013) W82.06† ♦ 3 1735.0 OLR — — —
Hedman & Nicholson (2013) W82.21† ♦ 3 1730.3 OLR 3 1729.2 − 1733.1 1.1 − 2.8
Hedman & Nicholson (2013) W80.98 ♦ 4 1660.4 OLR 4 1657.7 − 1661.1 0.8 − 2.6
Hedman & Nicholson (2013), Hedman et al. (2018, in preparation) W81.02 5 5 1593.6 OLR 5 1593.7 − 1596.2 0.1 − 2.6
Hedman et al. (2018, in preparation) W81.43 5 6 1538.2 OLR 6 1540.1 − 1541.6 1.9 − 3.4
Hedman et al. (2018, in preparation) W81.96 5 7 1492.5†† OLR 7 1495.0 − 1495.8 2.5 − 3.4
French et al. (2018, in preparation) W76.46  7 1657.7 OLR 9 1656.2 − 1657.2 0.6 − 1.5
Hedman et al. (2018, in preparation) W82.53 5 8 1454.2†† OLR 8 1456.9 − 1457.5 2.7 − 3.3
Hedman et al. (2018, in preparation) W83.09 5 9 1421.8 OLR 9 1424.0 − 1425.0 2.1 − 3.2
French et al. (2018, in preparation) B5  9 1626.5 OLR 13 1624.9 − 1626.7 0.2 − 1.6
Hedman & Nicholson (2014) W83.63 ♦ 10 1394.1 OLR 10 1395.2 − 1396.8 1.1 − 2.8
Hedman et al. (2018, in preparation) B14b 5 11 1450.5 OLR 13 1450.9 − 1453.3 0.4 − 2.8
French et al. (2018, in preparation) B3  4 1879.6 OVR 5 1868.3 − 1871.8 7.8 − 11.‡
French et al. (2018, in preparation) B1  7 1725.8 OVR 10 1723.3 − 1724.4 1.4 − 2.5
French et al. (2018, in preparation) B6  8 1645.4 OVR 11 1644.2 − 1645.3 0.2 − 1.3
French et al. (2018, in preparation) B4  9 1667.7 OVR 14 1665.8 − 1667.7 0.1 − 1.9
a Labeled as in Hedman & Nicholson (2013, 2014), or Hedman et al. 2018, in preparation; otherwise as in Baillié et al. (2011).

b cf. Figures 3 and 6

c deg day−1

† Member of a multiplet of waves of the same type having the same m but different frequencies, possibly the result of resonant coupling between the f -mode
of the same m identified here and a deep g-mode as demonstrated by Fuller (2014). Pattern speeds and residuals presented here are with respect to the
outermost wave in each multiplet, which most closely match the pattern speeds of the predicted ` = m = 2 and ` = m = 3 f -modes.
‡ This wave exhibits the largest residual with respect to the f -mode prediction; see §5.

†† Weakest among the sectoral sequence; pattern speeds should be taken as provisional.

ity field, but set mS = 0.14145 consistent with the 10.588
h rotation period derived here. Repeating the remain-
der of this analysis we find a very similar distribution of
optimal rotation periods, the median shifting to longer
periods by less than one minute as a result of the slightly
different interior mass distributions obtained. The fre-
quencies of the f -modes themselves are more sensitive to
Saturn’s assumed rotation rate than are the low-order
gravity harmonics J2 and J4 , a consequence of the f -
modes extending to relatively high `: at higher angu-
lar degree, Saturn’s rotation imparts a larger fractional
change to the frequency (see Figure 2).
5.2. Inadequacy of rigid rotation?
The lack of any perfect fit among the range of inte-
rior structures and rotation rates we have considered
is evident in Figure 4, where the RMS pattern speed
residuals reach approximately 3 deg day−1 at best, an
order of magnitude larger than the typical observational
uncertainty of approximately 0.1 deg day−1 associated
with even the weakest waves we compare to here (P.D.
Nicholson, private communication).
The absolute residuals are shown mode by mode in
Figure 5, including the full span of residuals obtained
over the sample of interior models, each one evaluated
at its optimal rotation rate. Points lie on both sides
of zero by construction, but again no model provides
a satisfactory fit. The (`, m) = (5, 4) wave stands out
as having the largest residual by far, the observed pat-
tern speed larger than that predicted by approximately
10 deg day−1 . Perhaps the most striking result is the
strong systematic trend evident in the residuals for the
sectoral modes, the modes of ` = m > 4 all having fre-
10 Mankovich et al.
quencies overpredicted relative to the detected waves.
Furthermore, the residuals exhibit a clear maximum at
` = m = 8 before decreasing again with higher `. That
the residuals are so strongly correlated with ` likely in-
dicates a missing piece of complexity in our model. A
natural generalization that would impart systematic `-
dependent changes to the f -mode frequencies would be
a depth-dependent rotation rate.
Finally, it is notable that the model pattern speed co-
variance (the diagonal elements of which set the vertical
spans in the residuals of Figure 5) varies so strongly and
non-monotonically with m. This can be understood as
a consequence of the tradeoff between the decreasing
zeroth-order frequency and the increasing contribution
from the first-order rotation correction with increasing
`, as can be seen from Figure 2 for the sectoral modes.
At high `, the zeroth-order frequency loses out to the
first-order correction. Since the latter is proportional to
ΩS , the overall pattern speeds vary more strongly with
rotation than at intermediate `. At low `, where the fre-
quency is dominated by the zeroth order contribution
and so rotation plays a smaller role, the large model
covariance is due mostly to sensitivity to the locations
of the core boundary and envelope transition, sensitiv-
ity that decays rapidly with increasing ` as modes are
increasingly confined close to the planet’s surface.
6. STRENGTH OF FORCING
The adiabatic eigenfrequency calculation that forms
the basis for this work provides no information about ex-
citation or damping of normal modes, which have yet to
be fully understood in the context of gas giants. Recent
work from Dederick & Jackiewicz (2017) demonstrated
that a radiative opacity mechanism is not able to drive
the Jovian oscillations, although they noted that driving
by intense stellar irradiation is possible for hot Jupiters.
Markham & Stevenson (2018) studied water storms as a
mode excitation mechanism and found it insufficient to
generate the mHz power excess that Gaulme et al. (2011)
attributed to Jovian p-modes, although they found that
deeper, more energetic storms associated with the con-
densation of silicates were viable. Stochastic excitation
of modes by turbulent convection such as in solar-type
oscillations is one obvious candidate for Jupiter and Sat-
urn because convective flux dominates the intrinsic flux
in each planet. Indeed an order of magnitude estimate
by Markham & Stevenson (2018) following the theory
of Kumar (1997) for resonant coupling with a turbulent
cascade of convective eddies, when adjusted for Saturn’s
mean density and intrinsic flux, yields a scaling relation
 3/2
σ0
∆r(R) ≈ 200 cm (17)
σ`mn
5
⌦p (deg day 1)
0 0
PS (min)
5
-5
OLR OVR
` m=0 ` m=1
10 ` m=2 ` m=3
` m=4 ` m=5
-10
2 4 6 8 10
m
Figure 5. Pattern speeds residuals (predicted minus ob-
served) for models each calculated at their optimal Saturn
rotation period. Circular markers are for one interior model
randomly chosen from our sample, while vertical lines show
the span of residuals obtained across the full sample. These
vertical spans thus indicate the amount of freedom avail-
able from the low-order gravity field as applied to three-layer
Saturn models, when the rotation rates are tuned using the
seismology. Note that these spans do not represent random
uncertainties because the residuals for the various modes are
correlated. The vertical axis at right expresses the residu-
als in terms of minutes of Saturn rotation, i.e., the degree
to which Saturn would need to be spun up or down to fit a
given wave’s observed pattern speed. The pairs of modes at
m = 7 and m = 9 are slightly offset horizontally for clarity,
and the four pairs of modes that are members of multiplets
(` = 5, 9, 10, and 13) are linked by grey dotted curves; see
discussion in §7.
for the surface amplitude of the radial displacement in
the `mn mode. Here
r
GM
σ0 ≡ (18)
R3
is Saturn’s dynamical frequency. For completeness, the
velocity amplitudes in this model are given by
 1/3
−1 σ0
v ≈ 0.01 cm s . (19)
σ`mn
As Markham & Stevenson (2018) noted, the rapid decay
of mode amplitudes with increasing frequency in such a
model would fail to explain the excess power at mHz fre-
quencies in Jupiter that Gaulme et al. (2011) attributed
to Jovian p-modes. However, substantial power (dis-
placement amplitudes of order 10-100 cm) is predicted
at the low frequencies σ`mn ∼ σ0 attained by the Sat-
urnian f -modes of low angular degree. In what follows,
we normalize our f -mode eigenfunctions in accordance
with the amplitude estimate of Equation 17 and derive
the resulting torques applied at OLRs and OVRs, find-
ing them to be roughly consistent with the apparent
 Saturn’s Seismological Rotation 11

observability of waves at these resonances, with several applied by the f -modes at OLRs and OVRs in the C and
noteworthy exceptions. D rings assuming the spectrum of amplitudes follows
the scaling relation of Equation 17. Because the torques
6.1. Torques and detectability (Equations 20 and 22) are proportional to ring surface
In deriving the magnitudes of torques applied at ring mass density Σ, itself strongly variable across the rings
resonances we follow the approach of Marley & Porco at a variety of spatial scales, we instead plot the normal-
(1993). In a ring of surface mass density Σ, the linear ized torques TL /Σ and TV /Σ. These are straightforward
torque applied at a Lindblad resonance is (Goldreich & quantities to calculate even with imperfect knowledge
Tremaine 1979) of the mass density itself. When comparing to detected
wave patterns should be kept in mind that Σ can play
mπ 2 Σ 2 an important role in whether a given wave is likely to
L
T`mn =− (2m + ` + 1) (Φ0`mn ) , (20)
DL be driven to detectable amplitudes.
where Saturnian waves can also be obscured by more promi-
 2 ! nent eccentric features, such as those associated with
9 a satellite resonances. Of particular importance is the
D L = − 3 − J2 Ω2 (1 ∓ m)
2 rL strong Titan 1:0 apsidal resonance, which Nicholson
(21)
 2 et al. (2014) studied in Cassini radio and stellar oc-
21 a
+ J2 Ω2 + O(J22 , J4 ) cultations and found responsible for driving the m = 1
2 rL
wave in the Titan/Colombo ringlet (77,879 km) and also
and Φ0`mn is the magnitude of the perturbation to the dozens of other m = 1 features from 74,000-80,000 km.
gravitational potential caused by the `mn mode, evalu- Their test-particle model (cf. their Figure 19) predicts
ated at the Lindblad resonance r = rL in the ring plane maximum radial deviations in excess of 100 m as much
cos θ = 0. Similarly the linear torque applied at a verti- as 3,500 km away from that resonance, posing a seri-
cal resonance r = rV is (Shu et al. 1983; Marley & Porco ous challenge for the reconstruction of weaker wave fea-
1993) tures from stellar occultation profiles obtained at differ-
2 ent phases. This substantial region of the C ring thus
mπ 2 Σ dΦ0`mn

V
T`mn = , (22) may be concealing waves driven at Saturn resonances,
DV dθ
and Figure 6 accordingly indicates the region where the
where maximum radial deviations are larger than 300 m ac-
 2 ! cording to the model of Nicholson et al. (2014).
9 a
D V = − 3 + J2 Ω2 (1 ∓ m) For context, the torques associated with four satellite
2 rV
 2 (23) resonances that open gaps or launch waves in the C ring
21 a are also shown in Figure 6. Prometheus 2:1 ILR opens
− J2 Ω2 + O(J22 , J4 )
2 rV a gap in the C ring while the Mimas 4:1 ILR launches a
wave. The Mimas 3:1 IVR opens a gap, while the Titan
and (dΦ0`mn /dθ) is to be evaluated at the vertical res-
-1:0 nodal resonance launches a wave. Estimates for the
onance r = rV and cos θ = 0. In the expressions for
strengths of these satellite torques are taken from Rosen
DL and DV the upper (lower) signs correspond to in-
et al. (1991a,b) and Marley & Porco (1993).
ner (outer) Lindblad or vertical resonances, as in Equa-
tions 12 and 13. Expressions for Φ0`mn and dΦ0`mn /dθ
6.1.1. Conspicuously missing waves?
are derived as in Marley & Porco (1993); these are re-
produced in Appendix A for completeness. These ex- Inspection of Figure 6 reveals a few f -mode resonances
pressions rely on integrals of the Eulerian density per- that this simple excitation model predicts to experience
turbation ρ0`mn over the volume of the planet. While strong forcing, but where no waves have yet been de-
accuracy to second order in Saturn’s smallness param- tected. Four of the OLRs with m = ` − 2 have nor-
eter mS would demand that this density eigenfunction malized torques predicted to be greater than that of the
include second-order corrections from the perturbation detected (`, m) = (13, 11) OLR. The most obvious of
theory described in §3.1, the fact that only an order of these is the (8, 6) OLR, which this model predicts to
magnitude calculation of the torques is required for the lie at 74,940 km, happening to be almost exactly co-
present purpose leads us to simply calculate these using incident with the detected B3 and B4 OVR features
the zeroth-order density eigenfunctions. (Baillié et al. 2011). The fact that these OVR waves
To illustrate which modes are likely to excite the apparently dominate the signal at this position betrays
strongest ring features, Figure 6 summarizes the torques some tension with the spectrum of amplitudes we have
12 Mankovich et al.

D ring C ring
101 D ring C ring
2

-TV /⌃ (1015 cm4 s 2)

1 Prometheus 2:1
TL/⌃ (1015 cm4 s 2)
10 Mimas 3:1
1
` m=2 `=m 10 4
` m=3
<latexit sha1_base64="Q17XMYmFW8OWMqZEav7bKUBfmvc=">AAACGHicbZDLSgMxFIYz9VbrrerSTbAILmTMdKztRii6cVnBXqAdSiZN29DMhSQjlGEew42v4saFIm67821Mp7PQ1gOBj/8/Jzn53ZAzqRD6NnJr6xubW/ntws7u3v5B8fCoJYNIENokAQ9Ex8WScubTpmKK004oKPZcTtvu5G7ut5+okCzwH9U0pI6HRz4bMoKVlvrFy7iXXtIVI9eJkWmVbatmXSDTrtUqNtJwbVcRqiQ9yvmNl/SLJWSitOAqWBmUQFaNfnHWGwQk8qivCMdSdi0UKifGQjHCaVLoRZKGmEzwiHY1+tij0onTnRJ4ppUBHAZCH1/BVP09EWNPyqnn6k4Pq7Fc9ubif143UsOaEzM/jBT1yeKhYcShCuA8JThgghLFpxowEUzvCskYC0yUzrKgQ7CWv7wKrbJp6WQerkr12yyOPDgBp+AcWKAK6uAeNEATEPAMXsE7+DBejDfj0/hatOaMbOYY/Clj9gNkV5wz</latexit>
sha1_base64="joy757jNFkANCVfmGIDF9XqaxoI=">AAACGHicbZDLTgIxFIbP4A3xhrp0MxFNXBjsgAgbE6Ibl5jIJZmZkE4p0NC5pO2YkAmP4cZXceNCY9yy820sAwtFT9Lky/+f057+XsSZVAh9GZmV1bX1jexmbmt7Z3cvv3/QkmEsCG2SkIei42FJOQtoUzHFaScSFPsep21vdDvz249USBYGD2ocUdfHg4D1GcFKS938ReKkl9hi4LkJKlqlslWzzlGxXKtVykjDVbmKUGXiUM6v/Uk3X0BFlJb5F6wFFOonkFajm586vZDEPg0U4VhK20KRchMsFCOcTnJOLGmEyQgPqK0xwD6VbpLuNDFPtdIz+6HQJ1Bmqv6cSLAv5dj3dKeP1VAuezPxP8+OVb/mJiyIYkUDMn+oH3NTheYsJbPHBCWKjzVgIpje1SRDLDBROsucDsFa/vJfaJWKlk7m/rJQv5mnAVk4gmM4AwuqUIc7aEATCDzBC7zBu/FsvBofxue8NWMsZg7hVxnTb+wjnJc=</latexit>

<latexit sha1_base64="OagMNZ0YZtxlcc2eYI64YA1bquU=">AAACGnicbZDLSsNAFIYnXmu9RV26CRbBhZYkjbRdCEU3LivYC6ShTKaTdujkwsxEKCHP4cZXceNCEXfixrdxkmahrQcGPv7/nJkzvxtRwoWufysrq2vrG5ulrfL2zu7evnpw2OVhzBDuoJCGrO9CjikJcEcQQXE/Yhj6LsU9d3qT+b0HzDgJg3sxi7Djw3FAPIKgkNJQNZJBfonNxq6T6NVmvWlZ+rletQyzfpmBYdZqDSsdYEov/CszHaoVvarnpS2DUUAFFNUeqp+DUYhiHwcCUci5beiRcBLIBEEUp+VBzHEE0RSOsS0xgD7mTpJvlWqnUhlpXsjkCYSWq78nEuhzPvNd2elDMeGLXib+59mx8BpOQoIoFjhA84e8mGoi1LKctBFhGAk6kwARI3JXDU0gg0jINMsyBGPxy8vQNauGTObOqrSuizhK4BicgDNggDpogVvQBh2AwCN4Bq/gTXlSXpR35WPeuqIUM0fgTylfP2j0nK8=</latexit>
sha1_base64="EkMkf2zxUh1BwHVrHsCAJx+L7E0=">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</latexit>

6
<latexit sha1_base64="3kQNX+04N7vOmWwsDlnD/1fPPTk=">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</latexit>
sha1_base64="0irKpClmbJEd8TjX36pJKvucs2E=">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</latexit>

1
10 Titan -1:0
3
6 10 7
7

Mimas 4:1
` m=4 ` m=5
10 3 ` m=1
<latexit sha1_base64="cmTzBDzL5SyFdm2JEhxKzsBine4=">AAACGnicbZDLSgMxFIYz9VbrbdSlm8EiuNCSyNTpRii6cVnBXmA6lEyatqGZC0lGKEOfw42v4saFIu7EjW9jOp2Fth4IfPz/OcnJ78ecSQXht1FYWV1b3yhulra2d3b3zP2DlowSQWiTRDwSHR9LyllIm4opTjuxoDjwOW3745uZ336gQrIovFeTmHoBHoZswAhWWuqZKO1ml7hi6HsprCAHwdrlGaxUHQhrtgZUtWtVZ9qlnJ8HV/a0Z5ZhBWZlLQPKoQzyavTMz24/IklAQ0U4ltJFMFZeioVihNNpqZtIGmMyxkPqagxxQKWXZltNrROt9K1BJPQJlZWpvydSHEg5CXzdGWA1koveTPzPcxM1qHkpC+NE0ZDMHxok3FKRNcvJ6jNBieITDZgIpne1yAgLTJROs6RDQItfXobWRQXpZO7scv06j6MIjsAxOAUIOKAObkEDNAEBj+AZvII348l4Md6Nj3lrwchnDsGfMr5+AHGNnLU=</latexit>
sha1_base64="2zZ5FilKYIfrOp1imh8KchCctOA=">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</latexit> <latexit sha1_base64="t+/95rFtxILFsLDzorwTYC0ojBY=">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</latexit>
sha1_base64="lumsfglqmO8OZbSm3jUAoc3wH7A=">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</latexit>

8 9 5 9
10 10
<latexit sha1_base64="BawJ3cDXBTpyfCzHmU7cL4KdVws=">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</latexit>
sha1_base64="KcRPEGG2+FA41E+GodiS5oIKhQg=">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</latexit>

12

OLR OVR 12
5 7
10 10
70 75 80 85 90 70 75 80 85 90
3 3
Saturn-centric radius (10 km) Saturn-centric radius (10 km)
Figure 6. Strengths of torques per surface mass density exerted on the C and D rings by model Saturnian f -modes, with
amplitudes assigned according to the scaling relation of Equation 17 for excitation by turbulent convection. Model points (filled
circles) are colored by their value of ` − m. Open symbols enclose model f -modes that we have identified with observed waves
as in Table 1; these are rendered in grey to emphasize that they are not measured torques. The grey shading in both panels
represents the region where maximum radial variations in ring orbits caused by the Titan 1:0 apsidal ILR exceed 300 m, making
the detection of wave features more difficult; see §6.1. A subset of resonances have been labeled to their right by their azimuthal
wavenumber m for ease of identification.

assumed, which predicts B3 and B4 to have torques one parable to or larger than than the smallest predicted
to three orders of magnitude lower than that predicted torque associated with a wave that has already been ob-
for the (8, 6) OLR. Given the close proximity of these served. Likewise, Table 3 reports resonances predicted
resonances, an appeal to the spatial dependence of Σ to lie in the D ring, although it is not clear whether any
seems unlikely to resolve this tension. wave patterns there will ultimately be detectable given
Of the remaining m = ` − 2 OLRs stronger than the ring’s faintness.
(13, 11), none among (10, 8), (11, 9), or (12, 10) have had
7. DISCUSSION
associated wave detections. This may be attributable to
strong perturbations from the Titan apsidal resonance This work offers interpretations for the set of inward-
as discussed above. Among the predicted m = ` − 4 res- propagating density waves and outward-propagating
onances, the (12, 8) OLR at 74,556 km is quite close to vertical waves observed in Saturn’s C ring in terms of
the inner boundary of the C ring where there are a series resonances with Saturnian f -modes. It also demon-
of gaps that have yet to be fully understood. Falling in strates that Saturn’s rotation state is of critical impor-
such a gap could render such a resonance unobservable, tance for Saturn ring seismology, a fact made evident
although within the model uncertainty, this resonance by the systematic mismatch with the observed pattern
could lie between the gaps or on gap edges. speeds of these waves obtained assuming that Saturn
As for the OVRs, the only resonances that yield waves rotates rigidly at the Voyager System III rate of Desch
that have been detected so far in the C ring are the & Kaiser (1981) or slower (see Figure 4). The interior
four that fall closest to Saturn, and indeed the strongest configurations considered to arrive at this conclusion
predicted waves in each ` − m have been observed. It accounted somewhat generously for the freedom in the
warrants closer attention that three of the four strongest low-order gravity field, because the likelihood function
OVRs predicted in the C ring have not been associated used to obtain our posterior distribution of interior mod-
with any wave feature, while waves have been observed els assumed an inflated variance on J2 to accord with
at what should be weaker OVRs with ` − m = 3 and the numerical precision of our theory of figures imple-
` − m = 5. These three “missing waves” correspond to mentation (see §2.1). Because the resulting distribution
the (`, m) = (6, 5), (7, 6), and (8, 7) Saturn f -modes. included a diversity of heavy element and helium distri-
Because of their location, it is possible that these waves butions, envelope transition locations, and core masses,
are present but obscured by the Titan apsidal resonance. the seismology suggests a tension with the Voyager ro-
To aid in the search for Saturnian resonances in the C tation rate commonly assumed for Saturn’s interior that
ring, Table 2 lists the pattern speeds of all model OLRs different three-layer models seem unlikely to resolve.
and OVRs in the C ring with predicted torques com- Our model is deficient in two major ways. First, the
model is not suited to address the close multiplets of
 Saturn’s Seismological Rotation 13

Table 2. Predicted OLRs and OVRs in the C ring without waves observed to have the same azimuthal wavenum-
associated wave detections ber m, namely the mutiplets of waves in the C ring with
` m Type Ωpat (deg day−1 ) Remark (see §6.1)
m = 2 and m = 3. The bulk of these seem naturally ex-
plained by the model of Fuller (2014), wherein avoided
11 11 OLR 1367.5 − 1374.6
crossings between the f -modes and deep g-modes of
12 12 OLR 1345.5 − 1352.4
higher angular degree give rise to a number of strong
13 13 OLR 1325.9 − 1332.6
8 6 OLR 1740.0 − 1749.5 Coincident with B3, B4
perturbations with the same m. However, in the wealth
10 8 OLR 1584.7 − 1592.9 Near Titan apsidal of new OLR and OVR wave patterns that have been
11 9 OLR 1530.5 − 1538.3 Near Titan apsidal measured from increasingly low signal to noise VIMS
12 10 OLR 1486.0 − 1493.4 data since Hedman & Nicholson (2014), it seems that
14 12 OLR 1416.5 − 1423.4 only two waves add to the mixed-mode picture, both
12 8 OLR 1692.0 − 1700.1 Among gaps with m = 2: the close-in B9 wave, and the Maxwell
14 10 OLR 1565.1 − 1572.4 Near Titan apsidal ringlet wave that matches our model Saturn’s pure (2, 2)
15 11 OLR 1517.9 − 1524.9 Near Titan apsidal f -mode. Acoustic modes of higher angular degree have
6 5 OVR 1736.9 − 1747.3 Near Titan apsidal less amplitude in the deep interior and so are less likely
7 6 OVR 1646.1 − 1655.6 Near Titan apsidal to undergo degenerate mixing with any deep g-modes
8 7 OVR 1577.3 − 1586.1 strongly. Indeed, there is not yet any direct evidence
9 8 OVR 1522.9 − 1531.1 for f -modes with ` > 3 undergoing avoided crossings
10 9 OVR 1478.5 − 1486.3
with deep g-modes, although the outlying (5, 4) OVR
11 10 OVR 1441.4 − 1448.8
warrants closer scrutiny in the mixed-mode context.
12 9 OVR 1578.0 − 1585.8 Near Titan apsidal
13 10 OVR 1527.0 − 1534.4 Near Titan apsidal
The second deficiency of the model is the strong as-
15 10 OVR 1600.8 − 1608.1 sumption that Saturn rotates rigidly. While Lorentz
forces would likely suppress shear in the metallic part of
Note—Pattern speeds can be mapped to physical locations given
Saturn’s equatorial radius and J2n using the relations in §4.1.
Jupiter or Saturn’s envelope (Liu et al. 2008), evidence
gathered from spacecraft indicate that zonal wind pat-
terns penetrate to significant depths (Smith et al. 1982;
Kaspi et al. 2018). It has been proposed that the insu-
lating molecular regions of these planets may be rotating
differentially on concentric cylinders (Ingersoll & Pollard
1982; Ingersoll & Miller 1986), the zonal winds being the
Table 3. Predicted OLRs and surface manifestation of these cylinders of constant an-
OVRs in the D ring gular velocity. A significant observational development
has been made by Hedman et al. (2018, in prepara-
` m Type Ωpat (deg day−1 )
tion) in that Lindblad resonances with Saturnian nor-
5 3 OLR 2309.9 − 2323.6 mal modes have now been detected for the full set of
6 4 OLR 2031.4 − 2043.1
sectoral f -modes from ` = m = 2 up to ` = m = 10,
7 5 OLR 1858.8 − 1869.2
constituting direct frequency measurements for modes
9 5 OLR 2058.3 − 2068.8
10 6 OLR 1897.3 − 1906.8
that sample an uninterrupted sequence of depths within
11 7 OLR 1780.7 − 1789.4 Saturn. One of the significant findings of the present
work is the evident structure in the pattern speed resid-
2 1 OVR 3378.6 − 3393.6
uals (see Figure 5) within this sequence, structure which
3 2 OVR 2420.6 − 2433.5 is robust with respect to the different interior mass dis-
4 3 OVR 2061.1 − 2073.6 tributions considered. One natural interpretation is that
7 4 OVR 2172.6 − 2184.5 in reality the modes are acted on by strong differential
8 5 OVR 1964.3 − 1974.8 rotation in the interior, as offered in §5.2. If this is the
9 6 OVR 1822.7 − 1832.2 case, the non-monotone behavior of the residuals with
11 6 OVR 1965.7 − 1975.1 ` would seem to indicate that angular velocity is not a
12 7 OVR 1837.1 − 1845.8 monotone function of depth, the modes of ` ≈ 8 most
13 8 OVR 1739.6 − 1747.7 strongly sampling a global minimum.
The identifications we have made also reveal that the
set of waves at Saturnian resonances so far contain four
instances of a pair of modes belonging to the same mul-
14 Mankovich et al.

tiplet, i.e., they are described by the same angular de-
gree ` but different azimuthal order m (see Table 1 and
Figure 5). This carries significance for the prospect of
deducing Saturn’s rotation profile from the frequency
splitting within each multiplet, although the important
centrifugal forces and ellipticity due to Saturn’s rapid
rotation complicates the picture compared to the first-
order rotation kernels commonly applied to helioseismol-
ogy (Thompson et al. 2003) and asteroseismology (e.g.,
Beck et al. 2012).
8. CONCLUSIONS
We have presented new Saturn interior models and
used them to predict the frequency spectrum of Saturn’s
nonradial acoustic oscillations. Comparison with waves
observed in Saturn’s C ring through Cassini VIMS stel-
lar occultations reveals that the majority of these waves
that are driven at frequencies higher than the ring mean
motion are driven by Saturn’s fundamental acoustic
modes of low to intermediate angular degree `.
The frequencies of Saturn’s f -modes probe not only
its interior mass distribution, but also its rotation state,
especially those modes of higher `. We used the fre-
quencies of the observed wave patterns to make a seis-
mological estimate of Saturn’s rotation period assuming
that it rotates rigidly. Using these optimized models, we
argued that small but significant residual signal in the
frequencies of the observed waves as a function of ` sug-
gests that Saturn’s outer envelope rotates differentially.
Forthcoming work will extend the present methods to
address interiors rotating differentially on cylinders.
Saturn ring seismology is an interesting complement
to global helioseismology, ground-based Jovian seismol-
ogy, and asteroseismology of solar-type oscillators. Be-
cause the rings are coupled to the oscillations purely by
gravity, they are fundamentally sensitive to the modes
without nodes as a function of radius, and the observa-
tion of modes from ` = 2 to ` ∼ 15 stands in contrast
with helioseismology where the vast majority of detected
modes are acoustic overtones (p-modes) and f -modes
only emerge for ` & 100 (e.g., Larson & Schou 2008).
Likewise ground-based Jovian seismology accesses the
mHz-range p-modes and Saturn ring seismology fills in
the picture for frequencies down to ∼ 100 µHz. Because
of their point-source nature, main sequence and red
giant stars with CoRoT and Kepler asteroseismology
means that typically only dipole (` = 1) or quadrupole
(` = 2) modes are observable because of geometric can-
cellation for modes of higher ` (Chaplin & Miglio 2013).
In contrast, the proximity of the C and D rings to Saturn
renders them generally sensitive also to higher ` so long
as the modes exhibit the correct asymmetries. We finally
reiterate that Saturn is a rapid rotator (ΩS /σ0 ∼ 0.4),
more in line with pulsating stars on the upper main se-
quence (Soufi et al. 1998) than with stars with CoRoT
and Kepler asteroseismology, and to our knowledge this
is the most complete set of modes characterized to date
for such a rapidly rotating hydrostatic fluid object.
This work buttresses the decades-old hypothesis
(Stevenson 1982) that Saturn’s ordered ring system does
act as a sensitive seismograph for the planet’s normal
mode oscillations. The set of Saturnian waves detected
in the C ring so far thus provide important contraints
on Saturn’s interior that are generally independent from
those offered by the traditional static gravity field. Fu-
ture interior modeling of the solar system giants will
benefit from joint retrieval on the gravity harmonics
and normal mode eigenfrequencies.
We thank P.D. Nicholson and M.H. Hedman for ex-
tensive discussions about their detection and modeling
of waves in the occultation data. Nadine Nettelmann
provided invaluable guidance regarding the theory of
figures. C.M. further thanks Steve Markham, Ethan
Dederick, Daniel Thorngren, and Jim Fuller for help-
ful conversations. This work was supported by NASA
through Earth and Space Science Fellowship program
grant NNX15AQ62H to C.M. and Cassini Participat-
ing Scientist program grant NNX16AI43G to J.J.F. The
University of California supported this work through
multi-campus research award 00013725 for the Center
for Frontiers in High Energy Density Science. Some of
these calculations made use of the Hyades supercom-
puter at UCSC, supported by NSF grant AST-1229745
and graciously mantained by Brant Robertson.

APPENDIX

A. PERTURBATIONS TO THE EXTERNAL POTENTIAL

The density perturbations associated with nonradial planet oscillations generally lead to gravitational perturbations
felt outside the planet. These perturbations can be understood as time-dependent components to the usual zonal and
tesseral gravity harmonics, and these are derived here closely following Marley & Porco (1993).
As in the standard harmonic expansion for the static gravitational potential outside an oblate planet (Zharkov &
Trubitsyn 1978), the time-dependent part of the potential arising from nonradial planet oscillations can be expanded
 Saturn’s Seismological Rotation 15

as
∞
( ∞ ∞ X
`
)
GM X X  a ` X  a `
0 0 m 0 0
Φ (t) = − J`n P` (cos θ) + P` (cos θ) [C`mn cos mϕ + S`mn sin mϕ] . (A1)
r n=0 r r
`=2 `=2 m=−`

0 0 0
The coefficients J`n , C`n and S`n are analogous to the usual gravity harmonics, but with the background density
replaced by the Eulerian density perturbation ρ0 (r, t) due to the oscillation in the `mn mode:
Z
0
M a` J`n = − ρ0`mn (r, t) r` P` (cos θ) dτ,

0
! ! (A2)
−
Z
` C `mn 2(` m)! 0 ` m cos mϕ
Ma 0
= ρ`mn (r, t) r P` (cos θ) dτ,
S`mn (` + m)! sin mϕ

where dτ = r2 sin θ dθ dϕ dr is the volume element and the integrals are carried out over the volume of the planet.
Given that our solutions for the density perturbation take the form

ρ0`mn (r, t) = Y`m (θ, ϕ)ρ0`n (r)e−iσ`mn t

(A3)
= c0 P`m (cos θ)ρ0`n (r)ei(mϕ−σ`mn t) ,

where
1/2
2` + 1 (` − |m|)!

m+|m|
c0 ≡ (−1) 2 , (A4)
4π (` + |m|)!
the integrals in Equation A2 are separable:
Z 2π Z π Z a
2
M a` J`0 = −c0 e−iσt eimϕ dϕ [P` (cos θ)] sin θ dθ ρ0n` (r) r`+2 dr (A5)
0 0 0
! !
2(` − m)! −iσt 2π imϕ cos mϕ
0
Z Z π Z a
C`m 2
Ma `
0
= c0 e e dϕ m
[P` (cos θ)] sin θ dθ ρ0n` (r) r`+2 dr. (A6)
S`m (` + m)! 0 sin mϕ 0 0

Notice from the symmetric integrand over azimuth that the J`0 only have contributions from axisymmetric (m = 0)
0 0
modes, while the C`m and S`m only have contributions from nonaxisymmetric (m 6= 0) modes. Using the orthogonality
of the associated Legendre polynomials
Z π Z 1
0 0 2δ``0 δmm0 (` + m)!
P`m (cos θ)P`m
0 (cos θ) sin θ dθ = P`m (µ)P`m
0 (µ) dµ = , (A7)
0 −1 (2` + 1) (` − m)!
Equations A5 and A6 reduce to
 1/2 Z a
4π
Ma `
J`0 =− e −iσ`mn t
ρ0`n (r) r`+2 dr, (A8)
2` + 1 0
1/2 a
4π (` − |m|)!
 Z
m+|m|
0
M a` C`m = (−1) 2 e−iσ`mn t ρ0`n (r) r`+2 dr, (A9)
(2` + 1) (` + |m|)! 0
0 0
S`m = iC`m . (A10)

The coefficients S`m are identical to the C`m up to a phase offset and can thus be ignored. These expressions for the
coefficients J`0 and C`m
0
can be substituted into the expansion A1 to write the `mn component of the external potential
perturbation as
  1/2 Z a
G 4π −iσ`mn t
ρ0`mn (r)r`+2 dr,

 `+1 P` (cos θ) e m = 0,


0 r 2` + 1 0
Φ`mn (r, t) = 1/2 Z a
4π (` − |m|)!

G m+|m|
 `+1 P`m (cos θ)(−1) 2 e−iσ`mn t cos mϕ ρ0`mn (r) r`+2 dr, m 6= 0.



r (2` + 1) (` + |m|)! 0
(A11)
16 Mankovich et al.

As above, we restrict our attention to prograde f-modes, namely those normal modes having m > 0 and n = 0. Thus
for the modes of interest the amplitude of the potential perturbation felt at a radius r outside Saturn is simply
1/2 Z a
4π (` − |m|)!

G m+|m|
Φ0`m0 (r, θ) = P m (cos θ)(−1)
`+1 `
2 ρ0`m0 (r) r`+2 dr (A12)
r (2` + 1) (` + |m|)! 0

where the time dependence and azimuthal dependence are omitted for the purposes of estimating the magnitudes of
torques on the rings.
Software: GYRE (Townsend & Teitler 2013), emcee (Foreman-Mackey et al. 2013), Matplotlib (Hunter 2007), SciPy
(Jones et al. 2001–), NumPy (Oliphant 2006)
Facilities: ADS

REFERENCES
Anderson, J. D., & Schubert, G. 2007, Science, 317, 1384
Baillié, K., Colwell, J. E., Lissauer, J. J., Esposito, L. W.,
& Sremčević, M. 2011, Icarus, 216, 292
Beck, P. G., Montalban, J., Kallinger, T., et al. 2012,
Nature, 481, 55
Chaplin, W. J., & Miglio, A. 2013, ARA&A, 51, 353
Colwell, J. E., Nicholson, P. D., Tiscareno, M. S., et al.
2009, The Structure of Saturn’s Rings, ed. M. K.
Dougherty, L. W. Esposito, & S. M. Krimigis, 375
Dederick, E., & Jackiewicz, J. 2017, ApJ, 837, 148
Desch, M. D., & Kaiser, M. L. 1981, Geophys. Res. Lett., 8,
253
Foreman-Mackey, D., Hogg, D. W., Lang, D., & Goodman,
J. 2013, PASP, 125, 306
Fortney, J. J., Helled, R., Nettelmann, N., et al. 2016,
ArXiv e-prints, arXiv:1609.06324
French, M., Mattsson, T. R., Nettelmann, N., & Redmer,
R. 2009, PhRvB, 79, 054107
French, R. G., Nicholson, P. D., Hedman, M. M., et al.
2016, Icarus, 279, 62
Fuller, J. 2014, Icarus, 242, 283
Fuller, J., Lai, D., & Storch, N. I. 2014, Icarus, 231, 34
Gaulme, P., Schmider, F.-X., Gay, J., Guillot, T., & Jacob,
C. 2011, A&A, 531, A104
Giampieri, G., Dougherty, M. K., Smith, E. J., & Russell,
C. T. 2006, Nature, 441, 62
Goldreich, P., & Tremaine, S. 1979, ApJ, 233, 857
Gough, D. 1980, in Lecture Notes in Physics, Berlin
Springer Verlag, Vol. 125, Nonradial and Nonlinear
Stellar Pulsation, ed. H. A. Hill & W. A. Dziembowski, 1
Gurnett, D. A., Kurth, W. S., Hospodarsky, G. B., et al.
2005, Science, 307, 1255
Hedman, M. M., & Nicholson, P. D. 2013, AJ, 146, 12
—. 2014, MNRAS, 444, 1369
Helled, R., Galanti, E., & Kaspi, Y. 2015, Nature, 520, 202
Hunter, J. D. 2007, Computing In Science & Engineering,
9, 90
Ingersoll, A. P., & Miller, R. L. 1986, Icarus, 65, 370
Ingersoll, A. P., & Pollard, D. 1982, Icarus, 52, 62
Jacobson, R. A., Antreasian, P. G., Bordi, J. J., et al. 2006,
AJ, 132, 2520
Jones, E., Oliphant, T., Peterson, P., et al. 2001–, SciPy:
Open source scientific tools for Python, , , [Online;
accessed ¡today¿]
Kaspi, Y., Galanti, E., Hubbard, W. B., et al. 2018, Nature,
555, 223
Kumar, P. 1997, in IAU Symposium, Vol. 181, Sounding
Solar and Stellar Interiors, ed. J. Provost & F.-X.
Schmider, ISBN0792348389
Larson, T. P., & Schou, J. 2008, in Journal of Physics
Conference Series, Vol. 118, Journal of Physics
Conference Series, 012083
Ledoux, P. 1951, ApJ, 114, 373
Lindal, G. F., Sweetnam, D. N., & Eshleman, V. R. 1985,
AJ, 90, 1136
Liu, J., Goldreich, P. M., & Stevenson, D. J. 2008, Icarus,
196, 653
Markham, S., & Stevenson, D. 2018, Icarus, 306, 200
Marley, M. S. 1990, PhD thesis, Arizona Univ., Tucson.,
doi:10.5281/zenodo.229754
—. 1991, Icarus, 94, 420
Marley, M. S., & Hubbard, W. B. 1988, in BAAS, Vol. 20,
Bulletin of the American Astronomical Society, 870
Marley, M. S., Hubbard, W. B., & Porco, C. C. 1987, in
BAAS, Vol. 19, Bulletin of the American Astronomical
Society, 889
Marley, M. S., Hubbard, W. B., & Porco, C. C. 1989, in
BAAS, Vol. 21, Bulletin of the American Astronomical
Society, 928
Marley, M. S., & Porco, C. C. 1993, Icarus, 106, 508
 Saturn’s Seismological Rotation
Mosser, B., Mekarnia, D., Maillard, J. P., et al. 1993, A&A,
267, 604
Murray, C. D., & Dermott, S. F. 1999, Solar system
dynamics
Nettelmann, N. 2017, A&A, 606, A139
Nettelmann, N., Püstow, R., & Redmer, R. 2013, Icarus,
225, 548
Nicholson, P. D., French, R. G., McGhee-French, C. A.,
et al. 2014, Icarus, 241, 373
Nicholson, P. D., & Porco, C. C. 1988, J. Geophys. Res.,
93, 10209
Oliphant, T. E. 2006, Guide to NumPy, Trelgol Publishing,
[Online; accessed ¡today¿]
Rosen, P. A. 1989, PhD thesis, Stanford Univ., CA.
Rosen, P. A., Tyler, G. L., & Marouf, E. A. 1991a, Icarus,
93, 3
Rosen, P. A., Tyler, G. L., Marouf, E. A., & Lissauer, J. J.
1991b, Icarus, 93, 25
Saio, H. 1981, ApJ, 244, 299
Saumon, D., Chabrier, G., & van Horn, H. M. 1995, ApJS,
99, 713
Shu, F. H., Cuzzi, J. N., & Lissauer, J. J. 1983, Icarus, 53,
185
Smith, B. A., Soderblom, L., Batson, R. M., et al. 1982,
Science, 215, 504
17
Soufi, F., Goupil, M. J., & Dziembowski, W. A. 1998,
A&A, 334, 911
Stevenson, D. J. 1982, EOS Transactions, 63, 1020
Stevenson, D. J. 1982, Annual Review of Earth and
Planetary Sciences, 10, 257
Thompson, M. J., Christensen-Dalsgaard, J., Miesch, M. S.,
& Toomre, J. 2003, ARA&A, 41, 599
Thompson, S. L. 1990, ANEOS–Analytic Equations of
State for Shock Physics Codes, Sandia Natl. Lab. Doc.
SAND89-2951, http://prod.sandia.gov/techlib/
access-control.cgi/1989/892951.pdf, ,
Thorngren, D. P., Fortney, J. J., Murray-Clay, R. A., &
Lopez, E. D. 2016, ApJ, 831, 64
Townsend, R. H. D., & Teitler, S. A. 2013, MNRAS, 435,
3406
Unno, W., Osaki, Y., Ando, H., Saio, H., & Shibahashi, H.
1989, Nonradial oscillations of stars
Vazan, A., Helled, R., Podolak, M., & Kovetz, A. 2016,
ApJ, 829, 118
Vorontsov, S. V. 1981, Soviet Ast., 25, 724
Vorontsov, S. V., & Zharkov, V. N. 1981, Soviet Ast., 25,
627
Zharkov, V. N., & Trubitsyn, V. P. 1978, Physics of
planetary interiors