Titan's Methane Weather

Titan’s Methane Weather
Henry G. Roe
Lowell Observatory, Flagstaff, Arizona 86001; email: hroe@lowell.edu
Annu. Rev. Earth Planet. Sci. 2012.40:355-382. Downloaded from www.annualreviews.org
by NORTH CAROLINA STATE UNIVERSITY on 10/02/12. For personal use only.
Annu. Rev. Earth Planet. Sci. 2012. 40:355–82 Keywords

The Annual Review of Earth and Planetary Sciences is meteorology, hydrology, climate, atmospheres, planets, hydrocarbons
online at earth.annualreviews.org
This article’s doi: Abstract

10.1146/annurev-earth-040809-152548
Conditions in Titan’s troposphere are near the triple point of methane, the
Copyright c 2012 by Annual Reviews. second most abundant component of its atmosphere. Our understanding
All rights reserved
of Titan’s lower atmosphere has shifted considerably in the past decade.
0084-6597/12/0530-0355$20.00 Ground-based observations, Hubble Space Telescope images, and data re-
turned from the Cassini and Huygens spacecraft show that Titan’s tropo-
sphere hosts a methane-based meteorology in direct analogy to the water-
based meteorology of Earth. What once was thought to be a quiescent place,
lacking in clouds or localized weather and changing only subtly on long sea-
sonal timescales, is now understood to be a dynamic system with significant
weather events regularly occurring against the backdrop of dramatic seasonal
changes. Although the observational record of Titan’s weather covers only
a third of its 30-year seasonal cycle, Titan’s atmospheric processes appear
to be more closely analogous to those of Earth than to those of any other
object in our solar system.
355
EA40CH15-Roe ARI 1 April 2012 8:19
1. INTRODUCTION
Saturn’s largest satellite, Titan, differs conspicuously from all other moons in the solar system by
the presence of a thick atmosphere. Titan’s surface displays great diversity, from fields of dunes
to hydrocarbon seas, with fluid-carved channels common across all regions. Conditions in Titan’s
troposphere are near the triple point of methane, the second most abundant component of its
atmosphere, leading to methane-based meteorology and hydrology. This active meteorology in
the atmosphere and hydrology on the surface interact via precipitation, evaporation, and possibly
cryovolcanism or other mechanisms for releasing methane from the surface. All of this combines
to make Titan the closest analogy in the solar system to Earth, the only other planet or moon
known to host active and interacting meteorological and hydrological systems.
Although Titan is often compared with Earth, aspects of Titan bear resemblances to other
planets; for example, Titan’s zonal wind structure and high atmospheric opacity resemble what is
seen on Venus. Seasonal transport of volatiles from pole to pole on Mars may also be an applicable
analogy to Titan, with the possibility that liquid methane shifts back and forth between Titan’s
poles annually.
All of this makes Titan a superb laboratory for comparative planetology and the testing of
our understanding of fundamental physics and chemistry of atmospheres and surface-atmosphere
interactions. Titan has been the subject of intense study in recent years, including a dedicated
lander (Huygens), >70 close encounters by the Cassini spacecraft, and numerous ground-based
observing programs. Every time technology allows us to examine Titan’s atmosphere in more
detail, it surprises us with its complexity. More than a decade of ground-based observations, seven
years of flybys by the Cassini spacecraft, and the Huygens probe entry have dramatically revealed
Titan’s surface and atmosphere, including the presence of methane-based weather in Titan’s
troposphere.
This review’s primary focus is to give a snapshot of our understanding of Titan’s tropospheric
methane weather, with enough context for the nonspecialist. Observations of Titan’s weather
have been taken for approximately a third of Titan’s 30-year seasonal cycle. Our knowledge of
Titan’s weather has improved substantially in that time. However, as often happens, although
the observations and theoretical advances of the past decade have answered many of the basic
questions, they have revealed many new questions for every old question resolved. Observations
of Titan over the next decades will no doubt bring as many, or more, surprises as the observations
of the past decade.
2. CONTEXT
2.1. In the Atmosphere

As on Earth, the dominant gas in Titan’s atmosphere is nitrogen. In Titan’s case, >90% of the
atmosphere is N2 . The next most abundant molecule is methane (5% at the surface, decreasing
to 1.4% in the stratosphere). Methane plays such an important role in Titan’s atmosphere and
on its surface as to deserve further discussion in its own section below. On Earth, conditions are
near the triple point of water, and the phase transitions of water lead to much of the interesting
weather phenomena. On Titan, conditions are near the triple point of methane, and, as might be
expected, phase transitions of methane play an important role in many of the weather phenomena
observed there.
At the surface of Titan, the pressure is ∼50% greater than the surface pressure of Earth. The
atmospheric density at the surface of Titan is also much greater than that at sea level on Earth
356 Roe
10 –8
a Thermosphere
Titan Earth
10 –6
Pressure (bar)
Mesosphere
10 –4
10 –2 Stratosphere
Troposphere
10 0
600
b
500
400
Altitude (km)
Titan
300
200
Earth
100
0
0 100 200 300 400
Temperature (K)
Figure 1
(a) Titan’s and Earth’s atmospheric structures plotted as logarithmic pressure versus temperature. On these
axes, the two atmospheres appear remarkably similar except that Titan is colder. (b) Titan’s and Earth’s
atmospheric structures plotted as altitude versus temperature, revealing that Titan’s atmosphere is
significantly thicker than that of Earth.
(∼4×). In spite of Titan’s smaller size, with a surface area only 16% of that of Earth, the total
mass of Titan’s atmosphere is nearly twice that of Earth’s. In a plot of temperature versus pressure
(Figure 1a), the physical structure of Titan’s atmosphere appears remarkably similar to that of
Earth except shifted to much colder temperatures.
Titan’s surface temperature (93.5 K at the Huygens landing site; Tokano et al. 2006) is ∼12◦
warmer than the effective temperature that would be expected if Titan had no atmosphere (∼82 K).
This difference is the net of the competing greenhouse and antigreenhouse effects of Titan’s at-
mosphere (McKay et al. 1991). On Titan, far-infrared opacity of N2 -N2 , CH4 -N2 , and H2 -N2
collision-induced absorption in the troposphere creates ∼21◦ of greenhouse effect. An antigreen-
house effect occurs when some part of the atmosphere strongly absorbs visible wavelengths but is
transparent to the thermal infrared. On Titan, the stratospheric haze serves this role and provides
∼−9◦ of antigreenhouse effect.
Although only modest wind speeds of a few centimeters per second might be expected on the
basis of the low solar flux available in the troposphere and the relative massiveness of Titan’s
atmosphere, the lower atmospheric winds observed by Huygens were 0.5–1 m s−1 (Tomasko et al.
2005). The lower atmosphere and surface winds are largely the result of seasonal solar forcing
www.annualreviews.org • Titan’s Methane Weather 357

effects (Tokano & Neubauer 2005). However, an important secondary effect on Titan, apparently
unique in our solar system, is the significant impact of winds generated by gravitational tides
(Tokano & Neubauer 2002). Saturn raises a ∼1% pressure bulge in Titan’s atmosphere that
moves eastward at 22.5◦ per day and creates localized wind speeds as high as 5 m s−1 .
2.2. Below the Surface

With a solid radius of 2,575 km, Titan is larger than the planet Mercury (2,440 km) and only slightly
smaller than the largest moon in the solar system, Jupiter’s Ganymede (2,631 km). Titan’s bulk
density (1.88 g cm−3 ) indicates roughly equal parts water and rock. Titan partially differentiated
into a rocky interior and water-rich mantle and crust. This crust is thought to be an approximately
100-km-thick water-ice shell floating on top of a water-ammonia ocean (Lorenz et al. 2008c).
This outer ice shell may be convectively active (Mitri & Showman 2008). Cracking at the base
of the ice shell would form liquid ammonia-water pockets that would work their way slowly up,
accumulating dissolved methane from methane clathrate deposits along the way (Mitri et al. 2008).
This cryomagma would erupt effusively at the surface, releasing methane into the atmosphere.
The trapping of a pocket of liquid from the ocean into the shell is relatively inefficient but likely
to happen episodically on regional scales, thus simultaneously explaining how Titan’s ocean did
not lose all its ammonia long ago and suggesting that cryovolcanism may inject methane into the
atmosphere.
Because Titan’s thin outer crust floats on a liquid ocean, the length of Titan’s day can vary
noticeably as angular momentum is transferred between the atmosphere and crust (Tokano &
Neubauer 2005). Cassini radar observations of surface features over several years show that the
surface shifted 0.36◦ per year in apparent longitude between 2004 and 2007, which is consistent
with the magnitude and sign of the expected seasonal angular momentum transfer between at-
mosphere and surface (Lorenz et al. 2008c). The change in the length of day, tens to hundreds of
seconds on Titan, is significantly greater than the annual variation in Earth’s length of day, which
is ∼1 ms. This many-orders-of-magnitude difference in effect can be explained by Titan’s more
massive atmosphere, its more dramatic seasonal circulation reversals, and the fact that the angular
momentum is transferred largely to Titan’s thin crust floating on an interior ocean as compared
with the much more massive interior of Earth.
2.3. On the Surface

Titan’s surface is unlike any other in the solar system, although aspects have familiar analogs.
Titan’s surface is heterogeneous, even on local scales; for example, the Huygens probe landed on
a damp outwash plane covered in tumbled ice-rocks, but a field of dunes composed of organic
sand grains lies just 10 km north (Lunine et al. 2008). A unifying theme of Titan’s surface is that
it is covered in hydrocarbons, from the equatorial sand dunes to the polar lakes.
Titan’s current thick atmosphere shields the surface from smaller impactors that would create
20-km-diameter craters. The population of larger craters, created by impactors big enough to
penetrate the atmosphere, is small. Only a few craters have been positively identified, although
another several dozen circular features may be heavily eroded old craters ( Jaumann et al. 2010).
This dearth of craters indicates that the surface is relatively young. Roughly half of the putative
craters are smaller than 20 km, suggesting either that the atmosphere has been significantly thinner
at some point in the past ∼1 billion years (∼1 Ga) or that the crater retention age is much longer
on some parts of Titan than is currently thought [100 million years (100 Ma) to 1 Ga; Lorenz
et al. 2007]. Crater destruction on Titan is likely dominated by atmospheric processes, including
358 Roe
methane rainfall and fluid flow on the surface, thus eroding crater features, as well as crater burial
by shifting sand dunes or direct deposition of heavier organics from the atmosphere.
Dunes cover ∼20% of Titan’s surface, confined almost exclusively to within30◦ of the equator.
The dunes were predominantly formed by westerly (eastward) winds (Lorenz & Radebaugh 2009).
The discovery that the dunes implied eastward winds was initially a conundrum as all general
circulation models predicted westward winds near the equatorial surface on average. A more
careful analysis of the wind field predictions (Tokano 2010) found that for most of Titan’s year,
the equatorial surface winds do blow westward. However, the wind speeds are typically less than
the saltation speed of dune particles. Twice per year around each Titanian equinox, the equatorial
surface winds reverse and blow strongly eastward at speeds exceeding the minimum threshold for
saltation. Thus, the dunes are constructed during brief periods each year, and prevailing conditions
for most of the year have little influence on the dunes.
Signs of hydrologic activity are ubiquitous in Cassini radar maps and visible/infrared imaging.
Lunine & Lorenz (2009) give an in-depth review of what is known of Titan’s surface hydrology.
Nearly all the surface liquids discovered to date are contained in the polar lakes, which are largely
confined to north of latitude 55◦ and range in size from 10 to 105 km2 (Hayes et al. 2008). Titan’s
lakes are a mix of methane, ethane, and dissolved nitrogen and heavier hydrocarbons, although
the exact ratio of ingredients is uncertain (Cordier et al. 2012). The rest of the surface is devoid of
lakes, except the south polar region, which is covered in considerably fewer lakes than the north
polar region but notably has Ontario Lacus. Ontario Lacus has been identified as consisting at
least partially of liquid ethane (Brown et al. 2008). The shoreline morphology is indicative of
recent lake-level changes (Barnes et al. 2009), and the lake level has dropped since the arrival of
Cassini, most likely owing to methane evaporation (Hayes et al. 2011).
Fluvial features are seen at nearly all latitudes (Lunine et al. 2008). Their typical morphology,
dendritic with acute branching angles, is suggestive of a rainfall origin as opposed to sapping
springs (Lunine & Lorenz 2009). Analysis of a subset of the fluvial erosion channels observed by
Cassini and Huygens found that their formation required infrequent but strong rainstorms, in line
with the observed and predicted storms, perhaps only once per Titan year but involving several
centimeters of rain in less than a day ( Jaumann et al. 2008). On average, the channels of Titan’s
surface have a poleward and eastward trend, suggesting systematic drainage toward the poles from
the equators (Lorenz et al. 2008a).
Cryovolcanism is yet to be definitively detected on Titan’s surface, but there is tantalizing
evidence for it from Cassini radar and imaging (Wall et al. 2009). Several surface features have been
tentatively identified as cryovolcanic. Most of these are controversial ( Jaumann et al. 2010) and
may possibly be explained by nonvolcanic processes (Moore & Pappalardo 2011). The strongest
evidence for cryovolcanism is at Sotra Facula, a 60-km subcircular feature near 40◦ W, 15◦ S (Kirk
et al. 2010). Cryovolcanism would likely have immediate and significant influences on the weather,
and it could raise the surface temperature the few degrees necessary to initiate convective cloud
activity. Furthermore, any release of methane would raise the local relative humidity and trigger
convective clouds.
Although, overall, Titan’s surface is very flat, with few slopes greater than a degree or two,
mountains up to 1–2 km tall and ridge lines extending a few hundred kilometers are seen
(Radebaugh et al. 2007). On the basis of estimated erosion rates, all of the observed mountains
are thought to be younger than 100 Ma, another indication of the relative youthfulness of Titan’s
present surface. Because raindrops fall slowly in Titan’s atmosphere and are likely evaporating
as they fall, these mountain tops may receive more rain than surrounding low-lying areas do. As
discussed in Section 4, although Titan’s mountaintops are relatively short, they can still play an
instigating role in cloud formation.

Autumnal equinox
November 1995 Aphelion
April 2025 September 1988
Winter solstice
N April 2018
May 1973
July 2047
October 2002
April 2032 days) Northern summ
an er (183
(159 Tit S Titan
n fall d
or ther ays)
N
N N
10.07 AU
S 9.01 AU S
Nor
the rn wi ays)
nter (1 26.7° itan d
56 Ti N g (178 T
tan days) Northern sprin
Perihelion
January 1974 S Summer solstice
December 1987
August 2003 Spring equinox

November 2032 February 1980 May 2017
August 2009 October 2046

January 2039
Figure 2
Saturn’s orbit about the Sun is just under 30 years long, setting the length of Titan’s year. The primary
driver of seasonal differences on Titan is its effective obliquity of 26.7◦ , although the varying distance to the
Sun is a significant secondary factor. The length of each season is indicated in units of Titan days.
Abbreviation: AU, astronomical units.
2.4. Seasons
The average length of a day on Titan is set by the period of its tidally locked orbit about Saturn, 15.9
Earth days. The length of Titan’s year is set by Saturn’s orbital period, 29.4 Earth years. One impact
of the length of Titan’s year is that no detailed and systematic observations of Titan’s weather
span more than a fraction of a Titan year. Systematic ground-based monitoring of Titan’s surface
and tropospheric clouds began in the early 2000s. Even Cassini, with its now-funded extended
Solstice Mission, will at best observe Titan for a total of 13 years (2004–2017; from early northern
winter to just after the northern summer solstice) before being intentionally crashed into Saturn.
Titan’s rotational axis is inclined 0.3◦ to its orbital plane (Stiles et al. 2008), which is inclined
0.33◦ to Saturn’s equator. These values are small compared with Saturn’s obliquity, 26.7◦ , which
is the primary factor forcing Titan’s seasons and just a few degrees greater than Earth’s obliquity
of 23.5◦ . Whereas Earth’s orbit is nearly circular, Saturn’s orbit is eccentric enough to create
significant seasonal asymmetries (see Figures 2 and 3). Currently, Titan is closest to the Sun
near the southern summer solstice and furthest from the Sun in the opposite season. Much as the
Milankovitch cycles on Earth drive a climatic cycle with periodicities of several tens of thousands
of years, Titan experiences long-term climate forcing owing to orbital evolution (Aharonson et al.
2009). In the case of Titan, the precession of Saturn’s perihelion with a period of ∼45 thousand
years (∼45 ka) is dominant. Several other orbital effects modulate the insolation on even longer
timescales. Thus, the asymmetries of Figure 3 will evolve on a multithousand-year timescale
and regularly reverse. The impact on climate and meteorological/hydrological systems has not
been significantly investigated, although Aharonson et al. (2009) suggest that the north/south lake
asymmetry may be a result of this climate evolution.
2.5. Methane’s Important Role

On Earth, water vapor contributes the largest percentage of greenhouse effect of any single molec-
ular species and is also the condensible, leading to potential positive feedback effects. On Titan,
360 Roe
Southern Southern Northern Northern Southern

spring summer spring summer spring
equinox solstice equinox solstice equinox
6 5 4 3 1
2
1
0 2
50N 3 5
um
4 im
ax
Latitude (degrees)
o f mtion
e la olar latitude
ud so Subs
tit in
La
EQ 5
4
3
50S 6 5 2
1 0
2
1 3 4 5 6 7
2000 2005 2010 2015 2020 2025
Year
Cassini Titan flybys
Figure 3
One Titan year of average daily insolation at the top of the atmosphere (erg s−1 cm−2 ) as a function of
latitude and time. Routine Earth-based observations of Titan’s weather commenced in the early 2000s and
have continued into the era of Cassini (2004–2017). By the end of its mission, Cassini will have observed
Titan during more than 125 flybys.
a similar situation exists with methane, which is both the primary condensible in the atmosphere
as well as a significant contributor to the greenhouse effect. Abundance and variability are major
differences. The water vapor in Earth’s atmosphere condensed would form a layer just a few cen-
timeters thick on average, with local variations of an order of magnitude. On Titan, the column
of methane condensed from the atmosphere would be 4–5 m, and the variation from location to
location is <20% (Penteado & Griffith 2010).
On Earth, the tropopause acts as a cold trap preventing water vapor from diffusing into the
stratosphere. Titan’s tropopause is not cold enough to trap methane, and methane diffuses into
the stratosphere. The abundance of methane in the stratosphere is the same as in the upper
troposphere (1.4%). This leads to another difference between water on Earth and methane
on Titan. Photolysis of methane in Titan’s upper atmosphere is a significant and irreversible
destruction mechanism, and the loss rate is governed by the availability of ultraviolet photons
(Strobel 1974, Yung et al. 1984). Calculating this rate of methane loss is straightforward. Assuming
no replenishment, photolysis would deplete atmospheric methane in 30–100 Ma. On Earth,
water is constrained largely to the troposphere, and there is no equivalent removal mechanism.
Titan’s source of atmospheric methane is a significant outstanding problem. The rate of
cometary/asteroid impactors is insufficient to resupply methane at the necessary rate (Zahnle
et al. 1992). Methane evaporation from the lakes may be enough to counter the loss to photolysis
(Mitri et al. 2007), but the volume of the lakes is insufficient to be the long-term source of atmo-
spheric methane (Lorenz et al. 2008b). If the lakes are ethane dominated (Cordier et al. 2009), then
the methane evaporation rate would be too low to balance the loss rate. The source of methane is
almost certainly Titan’s interior, where it could be stored in clathrates, stored in the macroporosity
in the shallow ice crust, or dissolved in the deep ocean. Methane can be formed in the deep ocean
by serpentinization and then be brought to the surface by cryovolcanism (Atreya et al. 2006).

Methane sets the atmospheric structure of Titan, directly through emission and absorption of
radiation but also indirectly through the influence of its photochemical products, including the
stratospheric haze. If depleted of methane, Titan’s atmosphere would cool so significantly that
nitrogen would largely freeze out onto the surface, leaving Titan with a thin atmosphere, more
like that of Neptune’s moon Triton (Lorenz et al. 1997). If methane abundance on Titan’s surface
and in its atmosphere has varied significantly in the past, Titan’s atmosphere will have undergone
significant climatic changes. For instance, if the resupply mechanism of methane from Titan’s
interior to the surface and atmosphere is episodic on 100-Ma timescales, then the atmosphere
may fluctuate between a collapsed frozen state and a warmer wetter state, with the current state
somewhere between the two. In that case, many of Titan’s surface features may be created primarily
during the more tropical state when rainfall and surface liquids are most abundant.
On the surface, liquid ethane and liquid methane mix, and the fluid mechanics is complicated by
dissolved nitrogen and heavier hydrocarbons. For clouds and rain in the atmosphere, the situation
is somewhat simpler owing to the low abundance of ethane in the troposphere. Optically thick
clouds, rain, and hail should be primarily composed of methane with a small amount of dissolved
nitrogen.
Until approximately a decade ago, Titan’s troposphere was thought to be calm and quiescent
without any significant clouds or weather. Confusing Voyager results led to the claim that Titan’s
troposphere was supersaturated in methane (e.g., Courtin et al. 1995 and references therein).
The troposphere was thought to be so lacking in cloud condensation nuclei that as soon as a
raindrop began condensing, it would immediately grow so large as to fall out of the atmosphere;
this idea is known as the rain-without-clouds hypothesis (Toon et al. 1988). We now know that
Titan’s troposphere is not saturated in methane and that cloud particle nucleation proceeds easily
(Curtis et al. 2008).
An upper limit to average rainfall can be estimated on the basis of specific heat of evaporation
and the available insolation-driven surface evaporation. For water on Earth, this yields an upper
limit of 1.2 m of precipitation averaged over the entire surface per Earth year, somewhat more
than the observed estimate of 1 m year−1 . On Titan, the upper limit on methane precipitation is
18 cm per Titan year (Lorenz 2000). The average rainfall in one year on Earth is approximately an
order of magnitude greater than the typical precipitable column in the atmosphere. On Titan, the
situation is reversed: The average rainfall during a Titan year is more than an order of magnitude
less than the precipitable column. Although the average annual rainfall on Titan is modest, it is
delivered in infrequent but fierce downpours. When the effects of precipitation on Titan’s surface
are considered, average conditions are much less important than extremes.
3. OBSERVING TITAN’S ATMOSPHERE

Titan’s size (5,150 km) and great distance from Earth [8–11 astronomical units (AU)] result in an
apparent angular diameter from Earth of 0.7 –0.9 . Turbulence in Earth’s atmosphere distorts
typical telescopic observations; a blurring effect of 0.25 –2.0 thus impedes observers from seeing
Titan with any significant spatial resolution.
The radiative transfer situation in Titan’s atmosphere also limits observations in the optical
(∼0.3–1.0 μm) and near-infrared (∼1–5 μm). At wavelengths shorter than 0.6 μm, the stratospheric
haze layers are optically thick (τ = 7 at 0.6 μm) and dark. At wavelengths longer than 0.6 μm,
the haze particles are strongly scattering. The opacity of the haze layer decreases as a function
of wavelength and becomes optically thin in the near-infrared (McKay et al. 2001, Tomasko
et al. 2008). The haze prevents shorter wavelengths from penetrating deeply into the atmosphere
and pushes observations of tropospheric clouds to the near-infrared (see Figure 4). Methane has
362 Roe
0.5 μm 2.03 μm 2.12 μm 2.22 μm
a b c d
e
Titan albedo
CH4 transmission
g
Solar flux
0.5 1.0 2.0 3.0 4.0 5.0

Wavelength (μm)
Figure 4
An albedo spectrum (e) and images (a–d ) of Titan from the Cassini Visual and Infrared Mapping
Spectrometer (VIMS) compared with a methane transmission spectrum ( f ) and the solar spectrum ( g).
Shortward of ∼0.7-μm wavelength, absorption by haze and Rayleigh scattering dominate Titan’s spectrum.
At these shorter wavelengths, only higher altitudes of the atmosphere are probed, and Titan appears
relatively featureless (a). Longward of ∼0.7-μm wavelength, methane absorption dominates. In each of the
“windows” between methane bands, Titan is bright, and surface features are visible (b). At wavelengths
where methane is very absorbing, the images show only the relatively featureless high-altitude hazes (d ). At
wavelengths of intermediate degrees of methane absorption, surface features are obscured from view, but
tropospheric clouds are revealed at high contrast (c) against the backdrop of the high-altitude haze. Although
this example shows the ∼2-μm window, this technique of separating tropospheric clouds from surface
features and higher-altitude hazes can be used in each of the ∼9 methane windows. Through the fitting of
radiative transfer models to spectra of individual clouds, accurate cloud-top altitudes can be assigned. VIMS
images and Titan albedo spectrum were obtained from the NASA Planetary Data System Imaging Node.
The methane transmission spectrum is a combination of data from Karkoschka (1994) and Irwin et al. (2006).
numerous strong absorption bands across the near-infrared, but in “windows” between these bands,
the surface of Titan is visible. Trading off among the effects of haze opacity, methane absorption,
and solar spectrum leads to optimal observing of Titan’s surface and lower atmosphere in the
clear windows between methane bands in the roughly 1–2.5-μm wavelength range. Observations
at somewhat shorter wavelengths are also possible, although with lower contrast due to haze

scattering, and observations at longer wavelengths are possible, although typically with lower
signal-to-noise due to the weaker sunlight.
Fink & Larson (1979) published the first spectra of Titan covering 1–2.5 μm. At the time, Fink
& Larson were unsure if they had detected Titan’s surface. Had their observations spanned more
than 3 nights of Titan’s 16-day rotation period, they would have almost certainly detected the
light curve of Titan’s surface and been sensitive enough to detect at least the largest tropospheric
storms. However, these observations were not continued on a regular basis.
The first spatially well-resolved observations were made by the Pioneer 11, Voyager 1, and
Voyager 2 spacecraft during their Saturn flybys of 1979–1981. The longest-wavelength filter on
these missions cut off at 0.64 μm, where the haze scattering opacity is still very high and the
returned images appeared to show only high-altitude features such as the dark polar hood (Smith
et al. 1982). Voyager 1 did not detect transient tropospheric clouds, nor could it have detected
any except the most extreme events (Richardson et al. 2004).
Voyager 1 largely resolved the major uncertainties surrounding Titan’s atmospheric structure
and composition. The physical structure of the atmosphere was determined via radio occultations
(Lindal et al. 1983). These data combined with midinfrared spectral observations of molecular
emissions established our basic current understanding of Titan’s atmosphere: It consists of ∼90%
N2 , a few percent CH4 , and numerous less abundant organic photochemical products (see, e.g.,
Coustenis et al. 1989). Voyager 1 was less successful at resolving uncertainties in the troposphere.
The Voyager data prompted the idea that Titan’s troposphere seemed likely to be quiescent with
no significant clouds or weather (Courtin et al. 1995).
In the 1990s, new near-infrared detectors enabled routine high-quality spectroscopy from
ground-based telescopes, especially the United Kingdom Infrared Telescope (UKIRT) and
NASA’s Infrared Telescope Facility (IRTF). Observers initially concentrated on Titan’s surface
(e.g., Lemmon et al. 1993), but these same observations proved useful for detecting clouds in the
troposphere.
The first observation of clouds on Titan were the spectra of Griffith et al. (1998), which showed
a large event on September 4–5, 1995; these clouds covered ∼10% of Titan’s apparent disk,
with the cloud tops reaching the midtroposphere. Titan’s season at the time was just before the
northern autumnal equinox. As these were disk-integrated spectra, the latitude of the feature was
not constrained, although the size and altitude were determined by fitting radiative transfer models.
Contemporaneous Hubble Space Telescope imaging found a cloud at 40◦ N (image published in
Lorenz 2008; see also Figure 5). It is likely that these observations were of the same event. Further
spectral observations in 1999 revealed smaller tropospheric clouds covering ∼0.5% of Titan’s disk
that varied on timescales as short as a few hours (Griffith et al. 2000), although again the latitude
was undetermined. In later analyses of speckle imaging data taken in October 1998 on the 10-m
Keck I telescope, a cloud in the south polar region was found to cover ∼1% of Titan’s disk (see
Gibbard et al. 2004 and Figure 5); thus, the cloud was similar in size to the daily varying clouds
of Griffith et al. (2000). In retrospect, these early observations of Titan’s clouds are now widely
accepted, but at the time they were controversial. Some doubted that clouds could exist in Titan’s
troposphere and attempted to explain the observations as instrumental artifacts.
Major observational advances came with the advent of adaptive optics (AO) on large (5–10-m)
telescopes. With AO, Earth’s atmospheric distortions are corrected in real time using a wavefront
sensor, deformable mirror, and control feedback loop, allowing diffraction-limited observations.
Furthermore, AO has allowed the 10-m Keck telescopes to achieve a resolution of ∼250 km
on Titan at a wavelength of 2 μm. The direct imaging observations of Brown et al. (2002) and
Roe et al. (2002) taken in December 2001 largely settled any debate concerning the existence of
tropospheric clouds on Titan. Brown et al. (2002) presented simultaneously acquired images and
364 Roe
a b c d e
1995 1998 2001 2004 2010
Figure 5
Improvements in technology have led to increasingly better images of Titan’s tropospheric clouds. (a) The first cloud imaged on Titan
was found at 40◦ N latitude in this differenced image of two Hubble Space Telescope 673-nm observations taken in 1995 (Lorenz
2008). (b) In 1998, speckle imaging with the 10-m Keck I telescope revealed surface features and a cloud at ∼40◦ S latitude (Gibbard
et al. 2004). (c) One of the discovery images of the seasonal south polar clouds was taken with a 2.12-μm troposphere-probing filter in
December 2001 with adaptive optics (AO) at the Gemini North 8-m telescope (Roe et al. 2002). (d ) By 2004, when this image was
taken at Gemini, the technology of AO had improved significantly. The improvement is especially noticeable in the higher contrast
resolution of the significant limb-brightening owing to the stratospheric haze (Roe et al. 2005b). (e) In this Cassini Imaging Science
Subsystem image of Titan taken in October 2010, the dark features are the surface, and the bright features are an equatorial cloud field
revealed in exquisite detail.
spectra that revealed cloud locations and altitudes in the midtroposphere, whereas Roe et al. (2002)
used three narrowband filters to separate surface features, tropospheric clouds, and stratospheric
features. Both groups found clouds exclusively in the south polar region and found that south
polar clouds appeared in every one of their observations.
Meanwhile, the Cassini-Huygens mission was on its way to Saturn and Titan. Launched in
1997, the spacecraft arrived in orbit around Saturn in mid-2004. Cassini will have completed ∼125
Titan flybys by September 2017, when it will be crashed into Saturn. The flyby data returned
by Cassini’s Imaging Science Subsystem (ISS) and Visual and Infrared Mapping Spectrometer
(VIMS) are unsurpassed in the level of detail revealed about Titan’s cloud morphology, locations,
and altitudes (see, e.g., Griffith et al. 2005, Turtle et al. 2009).
Observing programs at several Earth-based observatories are active for the seven to nine months
of the year that Titan is available in the night sky. These ground-based programs include nightly
photometry and spectroscopy at smaller telescopes and targeted high-resolution imaging with
large telescopes on up to several dozen nights per year (see Figure 6). Even in the era of Cassini,
there remains a strong need for frequent Earth-based observations to fill in the temporal gaps
between flybys, which are often two to four weeks apart. An example of a large storm that was
mostly missed by Cassini is shown in Figure 7. (More recently, Cassini has begun to take VIMS
observations of Titan between flybys, although with reduced spatial resolution.) After 2017, the
Earth-based observations will constitute the only record of Titan’s weather until some future space
mission arrives.
4. MODELS
Titan provides an excellent laboratory for testing models of atmospheric circulation and precipi-
tation under conditions and with materials vastly different from those on Earth. On Earth, these
complex interrelationships of processes have led to a push for a systems approach to studying
Earth, often referred to as Earth System Science. Titan deserves a similar systematic approach.
However, the complexity of the Titanian system and the general paucity of data have led most
modeling efforts relevant to methane weather to investigate individual phenomena rather than the
surface-atmosphere system as a whole. Titan System Science as an approach to studying weather
phenomena on Titan is in its infancy.

Gemini images
K’ H2 (1-0) Br- γ
1.95 – 2.30 µm 2.098 – 2.15 µm 2.139 – 2.198 µm
Surface Troposphere Stratosphere
2008-05-28 UT
(no clouds)
2008-04-16 UT
(clouds)
IRTF spectra
0.20
0.15
Albedo
0.10
0.05
12 cloud-free spectra
Clouds on Titan
Surface Troposphere Stratosphere
2.05 2.10 2.15 2.20 2.25

Wavelength (µm)
Figure 6
Example of how a “window” between methane bands is used to probe different altitudes in Titan’s surface on
the basis of data from two of the ground-based monitoring campaigns. (Bottom) Whole-disk spectra of Titan
from the program of Schaller et al. (2009) on NASA’s Infrared Telescope Facility (IRTF). Spectra from
twelve relatively cloud-free nights are shown along with one spectrum from a cloudy night. At wavelengths
of low methane opacity (2.10 μm), the rotationally varying surface is apparent. At wavelengths of high
methane opacity (2.15 μm), the stratosphere is unvarying on weeks-to-months timescales. At wavelengths
of intermediate methane opacity (∼2.10–2.15 μm), the tropospheric methane clouds are apparent with high
contrast. (Top) Example images from the Roe et al. (2005b) program at the Gemini 8-m telescopes. This
program uses three filters to separate surface features from tropospheric clouds from stratospheric haze.
IRTF spectra are taken nearly every night, whereas Gemini data are taken on a target-of-opportunity basis
when conditions warrant. Abbreviation: UT, universal time.
366 Roe
2008-03-25 UT 2008-03-27 UT 2008-04-13 UT 2008-04-14 UT 2008-04-15 UT
Cassini IRTF
T42 detects
flyby activity
2008-04-16 UT 2008-04-17 UT 2008-04-18 UT 2008-04-20 UT 2008-04-21 UT

2008-04-22 UT 2008-04-23 UT 2008-04-25 UT 2008-04-28 UT 2008-05-01 UT
2008-05-03 UT 2008-05-04 UT 2008-05-05 UT 2008-05-12 UT
Cassini
T43
flyby
Figure 7
In April 2008, a major storm system erupted on Titan (see Schaller et al. 2009). The first detection of the
event was on April 13, 2008, universal time (UT), in a spectral monitoring program at NASA’s Infrared
Telescope Facility (IRTF). Subsequent observations with the Gemini 8-m telescope (shown here) revealed
the evolution of Titan’s atmosphere in reaction to the event over several weeks. The initial event originated
at 15◦ S, 250◦ W ( green box), and a small cloud persisted at this location until at least May 1. The initial event
triggered atmospheric waves that induced cloud formation at latitudes from the equator to the south pole.
This event occurred entirely between Cassini’s T42 and T43 flybys of Titan and after a long period of
relatively clear skies across Titan. All images from the Gemini observing program of Roe et al. (2005b).
Even with the complexity of the Titanian system, much can be learned with straightforward
scaling arguments. For instance, a basic energy-balance calculation predicts that only ∼1% of the
annually averaged insolation is available to drive tropospheric convection and that this translates
to average cloud coverage of ∼1% (Lorenz et al. 2005). As discussed above, a similar analysis of
energy flux predicts a maximum average precipitation rate of 18 cm per Titan year. At the other
end of complexity from these simple scalings are general circulation models (GCMs).
For simplicity, the initial Titan GCMs were dry, meaning that they excluded methane con-
densation processes such as clouds and rain, although regions of convergence and updrafts can
be interpreted as likely locations for cloud formation (e.g., Hourdin et al. 1995, Richardson et al.
2007, Tokano et al. 1999). Titan’s atmosphere is not dry, however; methane meteorological pro-
cesses are at work in the troposphere. Therefore, more relevant are the further-developed GCMs
that include methane condensation and evaporation processes (e.g., Mitchell et al. 2006, 2009;

Earth Titan
Southern Northern
Southern Northern summer winter
summer winter
ITCZ SCZ
Southern Northern
Southern Northern fall spring
fall spring
ITCZ SCZ
Figure 8
A simplified view of the global atmospheric circulation and precipitation patterns on Earth and Titan. On Earth, the intertropical
convergence zone (ITCZ) remains consistently near the equator, shifting into the southern tropics in southern summer and northern
tropics in northern summer. At the ITCZ, air ascends owing to surface winds converging from both hemispheres. On Earth, the
majority of the annual precipitation occurs in the ITCZ. Titan’s equivalent to the ITCZ, the seasonal convergence zone (SCZ), is at the
south pole in southern summer and the north pole in northern summer, passing over the equator twice each year. As on Earth, the
majority of precipitation on Titan is thought to occur in the SCZ. On Titan, the descending winter branch condenses a thin high ethane
cloud that covers the entire polar region, is stable over many months, and does not lead to precipitation. Modified from Tokano (2011).
Rannou et al. 2006; Tokano et al. 2001), as well as modeling of localized methane cloud formation
and evolution (e.g., Barth & Rafkin 2007, 2010).
One of the major differences among the various GCMs is how they handle the stratospheric haze
and its seasonal evolution. The optical depth of stratospheric haze strongly influences solar forcing
of the troposphere. The latitudinal distribution of stratospheric haze shifts dramatically with
season, with generally more haze in the winter hemisphere. One model (Rannou et al. 2006) that
attempts to include the impact of the seasonally varying stratospheric haze predicts that circulation
is roughly symmetrical between hemispheres. However, this model appears to be ruled out by
significant mismatches between predicted and observed cloud locations, especially at northern
midlatitudes in late southern spring where the model predicts significant clouds and none have
been observed. Perhaps surprisingly, models that essentially ignore the haze for computational
efficiency (e.g., Mitchell et al. 2006) succeed much better at predicting the seasonal locations
of clouds, although these models are still far from perfect. They predict that the tropospheric
circulation varies dramatically with season: In the summer hemisphere, there are convergence
and uplifting that are matched with downwelling in the winter hemisphere, thus giving a single
pole-to-pole circulation cell (see Figure 8). This pattern reverses twice a year, and the models do
not always agree on the timing and nature of the reversal—e.g., a discrete jump in the convergence
zone from pole to pole versus a more gradual migration of the convergence zone during a period
of split circulation cells between north and south.
One might conclude that seasonal variation of stratospheric haze can be ignored in tropospheric
weather models. However, a closer examination of the stratospheric haze model of Rannou et al.
(2006) finds significant discrepancies from observation—e.g., the model predicts a permanent
thick polar haze cap, whereas observations show that the polar haze in summer is extremely thin
368 Roe
(Lorenz et al. 2010). The correct conclusion is that accurate representation of stratospheric haze
is critical. Further analyses of existing haze observations, from both Cassini and ground-based
data, are needed to then feed the modeling with more accurate representations of the haze and its
seasonal evolution.
Models of cloud formation by Barth & Rafkin (2010) show that the highest altitude attained
by a convective cloud on Titan is primarily a function of the local environment’s level of methane
humidity. A convective cloud formed in a 50%-humidity environment, typical of Titan’s lower
troposphere, will climb to reach the midtroposphere at 28–30 km. A modest increase in humidity
to 65% will cause the resulting cloud to drive convectively upward to the upper troposphere at
40 km or beyond. Convective clouds that reach the tropopause or beyond are predicted to be rare
or nonexistent.
One of the other intriguing results from detailed cloud formation models is that although the
typical lifting condensation level for Titan’s atmosphere is 6 km, or three times higher than the
highest mountain, such mountains can nonetheless play a role in forming tropospheric clouds
(Barth 2010). The modeling shows that under typical near-surface atmospheric conditions and
modest winds, these shallow mountains can form thin clouds within ∼100 km downwind, although
these clouds may be too optically thin to detect with most remote sensing techniques, are non-
convective, will not generate precipitation, and do not have a significant influence beyond their
local environment. However, under somewhat increased humidities, the orographic lifting of air
parcels can confer enough buoyancy to reach the lifting condensation layer and kick off convective
cloud activity. Once a convective system is triggered by this mechanism, it quickly grows large
enough and high enough in the atmosphere to become observable. Upon reaching the higher-
speed winds of the middle to upper troposphere, a cloud can detach from its original instigating
mountain. Thus, clouds on Titan can form by orographic instigation, but a convective cloud
triggered orographically need not appear coincident with, or even near, its location of formation.
The coupling of more advanced 3D GCMs with these detailed cloud formation models and a
better understanding of the seasonal evolution of the stratospheric haze are the next logical steps
toward a more Titan System Science approach to the problem of Titan’s methane meteorology.
The most promising observable constraints on these models are the timing and location of clouds
in ongoing and future observations. Additional constraints on the GCMs are the magnitude and
timing of the seasonal angular momentum transfer between Titan’s atmosphere and surface, which
are detectable in Cassini radar observations.
5. DISCUSSION
Whereas the observational record of Titan’s clouds dates back half a Titan year to the first observa-
tions of Titan’s clouds in 1995, that record is significantly incomplete for the first five to eight years.
Even in the present day of Cassini flybys every few weeks and often nightly Earth-based observa-
tions, significant weather events can be missed, although that is less likely now than a few years ago.
Each observing method has inherent strengths and limitations, which sometimes interact to
make combining and comparing results challenging. Cassini’s instruments return spectacular high-
spatial-resolution images and spectra of clouds, including many features that are far too small or low
contrast to be detected in Earth-based observations. However, Cassini’s observations are largely
concentrated during the few days around each Titan flyby that occur only every few weeks. Earth-
based observations are less sensitive, detecting only moderate and large clouds and storm systems
and revealing only coarse morphological details. However, Titan is observed with Earth-based
telescopes much more frequently, even nightly. To oversimplify the situation somewhat: Cassini
observations capture the “daily” clouds, many of which are far too small or faint to be seen from

Earth, but often miss the less frequent and more extreme “big” events. Meanwhile, the frequency
of Earth-based observations is important for capturing and documenting the full life cycle of the
extreme and rarer weather events that occur between Cassini’s less frequent observations (see, e.g.,
Figure 7). Ultimately, it is the combination of the two data sets, one focused more on the “daily”
weather and one focused more on the extreme rare events, that leads to our current understanding
of Titan’s tropospheric weather.
5.1. Cloud Heights

Opacity variation as a function of wavelength (see Figures 4 and 6) is used to measure the
altitudes of clouds in Titan’s atmosphere. Precision of a few kilometers can be obtained with
moderate-resolution infrared spectroscopy, as is available with Cassini’s VIMS. Several strato-
spheric condensation phenomena have been observed, such as an optically thin cirrus-like cloud
at the tropopause covering much of the Southern Hemisphere (Brown et al. 2002) and a similar
northern cloud identified as consisting primarily of ethane (Griffith et al. 2006). However, these
stratospheric phenomena appear quite uniform and vary only on slower seasonal timescales. In
this review, we are concerned with Titan’s methane clouds.
To date, all of the observed short-timescale variable clouds on Titan have been constrained
to the troposphere. Modeling of tropospheric convective cloud formation shows that methane
humidity is the most significant factor controlling the maximum height achieved by a convective
cloud (Barth & Rafkin 2010). When cloud-top altitudes are measured, they generally fall in the
range of 14–25 km, well below the tropopause at 40–50 km, and match the expectation for convec-
tive clouds formed in Titan’s typical surface environment of ∼50% humidity. With the highest
spatial resolution available from Cassini, which allows separating convective cloud cores from
broader cloud envelopes, a few convective clouds have been observed to reach 42 km, nearly to
the tropopause (Griffith et al. 2005). Cloud formation modeling suggests that these higher-altitude
clouds formed in moister environments of ≥65% relative humidity.
5.2. Timescales
How quickly a cloud forms, how long it lasts, and how quickly it dissipates are all useful clues to
the type of cloud and phenomena at work. All, or nearly all, of the tropospheric clouds observed
to date are formed convectively. Tropospheric clouds on Titan may evolve by other mechanisms,
such as wind shears that can pull clouds out into long streaks, but the initial triggering mechanism
was convection. Because convective clouds form rapidly in just a few hours and then often last for
days or even weeks before dissipating, only a handful of cloud events have been observed from
start to finish. The rapid formation of clouds is used as a key piece of evidence toward identifying
Titan’s tropospheric clouds as convective. In at least one case, Cassini data showed definitively
that the clouds were convective, with the cloud-top altitudes rapidly evolving upward in tightly
confined cores (Griffith et al. 2005).
Whereas convective clouds form rapidly, often in just a few hours, they can remain visible for
days or even weeks before dissipating. In the cases of smaller clouds, the timescale to dissipate
may be indicative of the timescale for diffusively mixing the cloud’s air parcels and evaporating
the cloud particles. In the cases of the largest observed systems (e.g., Figure 7), these relatively
long lifetimes are likely indicative of ongoing convective activity continuing to form new clouds
within the region. Although the frequency of observations and the number of large storm systems
observed do not allow firm conclusions to be drawn yet, it appears that the largest storm systems
maintain themselves for several weeks but then dissipate almost entirely in the course of a few
days or less. This suggests that, once in existence, these large storm systems are self-sustaining
370 Roe
until they run out of surplus methane, e.g., until humidity in the localized lower troposphere
decreases from its elevated levels back to a more typical ≤50%.
5.3. Seasonality
The first extensively observed clouds were the south polar clouds of early southern summer. After
its discovery in late 2001 (Brown et al. 2002, Roe et al. 2002), this south polar cloud field was seen
by a variety of observers in 2002–2004 (e.g., Bouchez & Brown 2005, de Pater et al. 2006, Gendron
et al. 2004) and was still apparent in the first Cassini images (Porco et al. 2005). These south polar
clouds are a seasonal effect, thought to result from global circulation creating a convergence zone.
Being the point of maximum insolation (see Figure 3), the region also probably warmed by a
few degrees during this period, which would be enough to enhance convective activity and cloud
formation (Brown et al. 2002).

From the discovery of the south polar cloud field in December 2001 until October 2004,
south polar tropospheric clouds appeared in essentially every observation taken (see Figure 9).
Occasionally, a much larger storm would cover a larger region of the south pole, as much as
Southern Geographically controlled clouds

30 (clustered in longitude) Cloud observed
summer No cloud observed
solstice
40
50
South latitude (degrees)
60 Circulation-driven clouds
(random longitudes)
70
80
90
November July March November July March November June February October
2001 2002 2003 2003 2004 2005 2005 2006 2007 2007
Month
Figure 9
The seasonal dissipation of the south polar summer cloud field is apparent in the locations of clouds observed
with several ground-based observing programs over the period 2001–2007. The relative size of a cloud is
indicated by the size of its blue diamond symbol, and observations in which no clouds were seen are noted at
the bottom with red diamonds. In the years around the southern summer solstice, clouds appeared exclusively
and frequently within ∼30◦ of the south pole. In 2003–2005, midlatitude clouds appeared at ∼40◦ S and
tended to cluster near 350◦ W longitude (Roe et al. 2005b). Following the large storm of 2004, the south
polar clouds largely disappeared, and the atmosphere was relatively quiescent until 2007, when additional
midlatitude clouds at 40–50◦ S and random longitudes appeared. Updated from Schaller et al. (2006b).

several percent of Titan’s apparent disk (Schaller et al. 2006a). Following the largest of these
observed episodes in October 2004, the south polar clouds essentially disappeared from Titan
(Schaller et al. 2006b). The timing of this change coincides roughly with a jump in the point of
maximum insolation from the south pole to midsouthern latitudes. Although the change in the
clouds preceded the insolation shift by approximately a year, the dissipation of the south polar
clouds appears to have been a seasonal effect.
Around the time of the last of the giant south polar clouds and shutdown of the south polar
cloud field, a new class of clouds appeared at midsouthern latitudes, typically within 1–3◦ of 40◦ S
(Griffith et al. 2005, Roe et al. 2005a). In ground-based observations, these clouds were seen
episodically a few times per year between late 2003 and early 2005 and were found to cluster
in a range of longitudes near 350◦ W (Roe et al. 2005b). Cassini also saw these clouds during
its first Titan flyby in mid-2004 (Brown et al. 2010). In Cassini and ground-based observations,
these clouds largely disappeared between early 2005 and mid-2006, with only a few small clouds
detected in Cassini VIMS observations. (See Brown et al. 2010 for an analysis of the first four
years of Cassini VIMS cloud observations.)
The localization to a narrow latitude range and the timing of the appearance of the clouds,
roughly coincident with a shift in the point of maximum insolation from the pole to ∼30◦ S latitude,
suggest that they originate from seasonal changes in global circulation. However, the first appear-
ance of these midlatitude clouds occurred two Earth years before that insolation shift, whereas
most seasonal phenomena would be expected to lag changes in insolation rather than precede them.
Furthermore, it seems suspiciously coincident that one of the largest midlatitude cloud episodes
preceded the massive south polar storm of late 2004 by just weeks. Finally, the longitudinal clus-
tering seen in the ground-based observations of the largest clouds is not explainable by circulation
alone. Possible explanations for the 2003–2005 midlatitude clouds are discussed in Section 5.6.
In Cassini VIMS observations, midlatitude clouds reappeared as a regular feature in mid-2006,
although many of these clouds were too small or faint to be detected in ground-based data (Brown
et al. 2010, Turtle et al. 2011a). In ground-based observations, the midlatitude clouds were not
apparent again until early 2007 (see Figure 9). In both data sets, the midlatitude clouds appeared
different from the earlier pre-2006 clouds. The post-2006 clouds were generally smaller, were
not clustered in longitude, appeared less episodic, and occurred over a wider spread of longitudes
ranging further south to ∼50◦ S. These post-2006 midlatitude clouds were more consistent with
expectations of how circulation-driven seasonal midlatitude clouds would appear.
On Earth, at the intertropical convergence zone (ITCZ), southern and northern hemispheric
surface winds meet, generating a band of clouds, typically convective thunderstorms, that encircle
the globe near the equator. This zone shifts modestly with season, tracking the subsolar point into
the Northern Hemisphere during northern summer and into the Southern Hemisphere during
southern summer, but typically remaining in the tropics. This seasonal shift causes many tropical
regions, e.g., the Indian subcontinent, to experience dramatic shifts between dry and wet seasons.
The seasonal shifts of Titan’s weather patterns can be explained by analogy to Earth’s ITCZ, with
the primary difference being that on Titan, the convergence zone is not constrained to the tropics
but instead migrates annually from the north pole to the south pole and back again. Because of
this pole-to-pole movement, well outside the tropics, we suggest that on Titan this be termed the
seasonal convergence zone (SCZ) (see Figure 8). Until at least 2004–2005, Titan’s SCZ resided
in the south polar region, generating the continuous sequence of clouds seen by Earth-based
observers and Cassini. Sometime thereafter, the SCZ began its seasonal migration northward,
ending the predominance of south polar clouds and leading to cloud formation at midlatitudes.
With the discovery of equatorial rainfall (see Section 5.4), there is evidence of the SCZ having
reached southern tropical latitudes by late 2010, as would be expected in this model of seasonal
372 Roe
shift. Further observations of the timing and nature of these seasonal shifts in the SCZ will provide
important constraints on the circulation models.
Although this review is focused on tropospheric meteorology, the discovery of a thin ethane
cloud at the polar tropopause in midnorthern winter is worth noting (Griffith et al. 2006). This
cloud, unvarying on timescales of weeks and months, is unlikely to directly influence the more
active tropospheric weather. However, its appearance has been interpreted as an indicator of
the tropospheric-stratospheric circulation. Stratospheric air, rich in ethane and other heavier
hydrocarbons, cools as it subsides at the winter pole, leading to condensation and the formation
of this stable cloud feature at the tropopause. Presuming that this formation hypothesis is correct,
the timing of the appearance and disappearance of this feature, and of a similar feature at the
southern winter pole, will provide a strong diagnostic of tropospheric-stratospheric circulation.
Whereas Earth’s orbit is close enough to circular that its varying distance from the Sun has at
most a modest asymmetrical influence on northern versus southern seasons, Saturn’s orbit around
the Sun has a significant eccentricity (0.056), leading its distance from the Sun to vary by more than
10% (Figure 2). The observing record of Titan’s tropospheric weather is not yet long enough to
have determined how this asymmetry manifests itself in the tropospheric weather.
Finally, we should not assume that each year on Titan is the same as the last. Like Earth’s
weather, Titan’s weather almost certainly has year-to-year differences, e.g., the differences
between La Niña and El Niño years. Not until the mid-to-late 2020s will the observing record
be long enough to be able to test for year-to-year differences in Titan’s weather. However, one
group using broadband visible wavelength filters has consistently observed over more than a
Titan year (Lockwood & Thompson 2009). Their filters probe the stratospheric haze layers
and intriguingly show that Titan’s stratospheric haze changes from one Titan year to the next.
Year-to-year changes in the troposphere are likely as the stratospheric haze modulates the
insolation received in the troposphere.
5.4. Large Events and Rainfall

The observational record is neither long enough nor complete enough to allow firmer statistics,
but on intervals of roughly 3–18 months, a large cloud system forms on Titan. Unlike the largest
“daily” clouds, which cover the typical ∼1% of Titan’s disk, the clouds in this system tend to cover
several percent of Titan’s disk, up to 10%, translating into several million square kilometers. The
first observation of one of these large events was coincidentally the first observation of clouds on
Titan (Griffith et al. 1998), which appeared at 40◦ N latitude at the end of northern summer. Since
then, several more have been observed from onset to dissipation (e.g., Schaller et al. 2006a, 2009),
which typically takes several weeks.
To date, the best-observed event, from onset to dissipation, was the system that begin in April
2008 (Schaller et al. 2009). In that case, nightly spectral monitoring of Titan revealed a dramatic
change on April 13, leading to sustained near-nightly observations for more than a month until
the event dissipated (see Figure 7). Unfortunately, Cassini largely missed this event owing to
the timing of its flybys. These ground-based observations demonstrated that these large weather
events can give a large enough “punch” to the atmosphere to generate Rossby waves that then
initiate cloud formation elsewhere. In this case, the initial event was at ∼15◦ S and triggered cloud
formation from equatorial latitudes to the south pole, where significant clouds had not been
observed in several years.
One of the most important questions about Titan is: When and where does it rain? Part of
the difficulty in answering is that not every cloud produces precipitation that reaches the ground.
Although evidence of prior rainfall is abundant on Titan’s surface, proving that any particular cloud

led to precipitation that reached the surface has been elusive. If one makes the reasonable, but by
no means certain, assumption that the amount of rainfall from a given storm is proportional to
cloud area, then the bulk of rainfall comes in infrequent, but torrential, downpours such as during
the events of April 2008. Cassini radar observations would be capable of detecting raindrops in the
atmosphere but have not done so to date (Lorenz et al. 2008d). The likelihood of coincidentally
targeting the radar observations during a flyby to the time and place of one of these large storm
systems is low.
Cassini ISS imaging of the surface has found two examples where the surface darkened in the
wake of large storms, thus providing direct evidence in favor of the hypothesis that rainfall reaches
the surface. Comparison of Cassini ISS images of the south pole taken approximately months
before and approximately seven months after the large south polar storm of 2004 (Schaller et al.
2006a) revealed a new region of dark patchiness, which was attributed to methane rain from that
storm pooling on the surface. After one of these large storms in 2010, Turtle et al. (2011b) observed
Titan’s surface to darken over a region covering ∼2,000 km east–west and ∼130 km north–south,
centered roughly at 20◦ S, 250◦ W. The surface then recovered to its pre-event brightness over the
course of several months. The best explanation for this darkening of Titan’s surface followed by
a gradual return to normal brightness is that the preceding large storm had wet the surface with
methane rain, which then evaporated and/or seeped or flowed away over a period of time.
Many questions remain about these infrequent large weather systems on Titan: What triggers
them and governs where they appear? What controls their lifetimes (typically a few weeks)? The
complexities of atmospheric dynamics may be enough to provide the trigger for such events.
However, as discussed in Section 5.6, there are hints that active geology on Titan’s surface might
also play a role. The latitudes of every observed large event to date have coincided roughly with
the expected location of Titan’s SCZ. This is consistent with seasonal circulation determining
where these events occur, even if circulation alone may not be the only trigger.
What governs the dissipation of these events is unknown. It appears that these large storm
systems are self-sustaining as long as there continues to be a source of methane-rich air. In that
hypothesis, the storm system would dissipate only after it had dried out the nearby atmosphere
via raining methane out of the atmosphere to the surface. In this case, the process of drying out
is relative and probably involves decreasing the relative humidity by only a few to a few tens of
percent. With the combination of the migration of Titan’s SCZ to the north polar zone in the
next few years and the abundance of lakes in that region, the frequency and size of these events
may increase dramatically.
5.5. Geographic Connections: Lake-Effect Clouds

Just as geography on Earth strongly influences weather, there are indications that geography
on Titan is important. The most firmly established link between geography and weather is the
lake-effect clouds in the north polar region.
In midnorthern winter, as the north polar region emerged into sunlight, Brown et al. (2009)
noticed a new class of streaky and knotty clouds north of 55◦ . These clouds reached midtropo-
spheric altitudes (20–30 km), and one of them was observed to form in only a few hours. The
altitude, morphology, and timescale of formation indicate convective origin. Convection is sur-
prising for winter in the polar region as the air in this region is expected to be stably stratified and,
if anything, subsiding; neither of these characteristics is conducive to convective cloud formation.
Aside from being located north of ∼55◦ , there was no pattern in their latitudinal distribution.
However, in longitude, nearly all of the clouds appeared downwind of the largest seas, suggesting
lake-effect clouds (see Figure 10). One objection to the lake-effect hypothesis is that the clouds
374 Roe
180
210 150
240 120
270 90
5˚
300 60
30
330
Figure 10
A polar projection map of Titan’s north pole (5◦ lines of latitude) shows the locations of lakes from Cassini
radar (blue) and Imaging Science Subsystem (ISS) (black) maps. Superimposed are the locations of clouds
detected in Cassini Visual and Infrared Mapping Spectrometer (VIMS) (dark yellow) and ISS (light blue) data.
The gray dots show the longitudes of polar clouds detected in ground-based images, where poor-viewing
geometry and lower resolution hindered accurate measurements of latitude. The clouds appear almost
exclusively downwind of the largest lakes. Reproduced from Brown et al. (2009).
are downwind of the lakes, not solely clustered over the lakes or at the edges of lakes. However,
these clouds appear to be convectively active enough to rise up out of the boundary layer and be
carried downwind, thus detaching themselves from their point of origin.
On Earth, cold, stably stratified air parcels move across warm lakes, increasing the humidity and
temperature in the parcel until the parcel becomes unstable and condensation and convection are
initiated. This can be seen particularly prominently over the great lakes of North America as their
size and thermal mass keep them relatively warm compared with the wintertime air temperature.
On Earth, these lake-effect clouds are driven by both the transfer of thermal energy from the lakes
to the air and the humidification of the air from lake water. Because of the differences between
liquid water (Earth) and liquid ethane/methane (Titan), it is unlikely that both effects are at work
simultaneously for a lake on Titan.

On Titan, a lake of methane can saturate the parcel of air via evaporation, but that evapo-
ration has a strong cooling effect on the lake (Mitri et al. 2007). This has two consequences.
First, evaporation from methane lakes can be self-regulating; the evaporation essentially shuts
off when the surface layer of the lake cools until insolation or some other mechanism of heat
transfer (e.g., mixing or diffusion within the lake itself ) can rewarm the surface of the lake.
Second, the temperature of the surface layer of the lake will likely never exceed the local air
temperature by any significant amount or for any significant length of time. Thus, a methane
lake may generate episodic, or even cyclical, lake-effect clouds via humidification of air parcels
but will never have enough of a positive temperature differential from the air to warm the
air.
Methane is not the only liquid on Titan’s surface. Whereas lakes dominated by methane
can humidify the atmosphere with methane, a primarily ethane lake will have a very low vapor
pressure and will not be a significant source of methane humidity. As a consequence, such a lake
will not experience evaporative cooling and can retain a significant temperature differential with
the atmosphere. Thus, an ethane or primarily ethane lake in the polar region could warm with
summer insolation and slowly release that thermal energy back into colder air parcels during the
winter, triggering winter convection.
Both an ethane-dominated lake that drives convection via thermal energy transfer and a
methane-dominated lake that drives convection via humidification of air parcels would be ex-
pected to create lake-effect clouds during the season of observations to date (northern winter
emerging to northern spring). However, as the pole emerges further into sunlight during the
spring, the two scenarios diverge and will be distinguishable in future observations. Additional
insolation on a methane lake should drive evaporation, and hence cloud formation, more rapidly.
Meanwhile, in the case of an ethane lake, the increased springtime insolation will warm the lower
atmosphere, decreasing the temperature differential between lake and atmosphere and suppressing
convective cloud formation.
5.6. Geographic Connections: Hints of Geologic Activity

The existence of mountains, which are estimated to erode in ≤100 Ma, and features that appear
to be cryovolcanic suggest that Titan’s surface has been geologically active relatively recently
and is possibly active in the present day. Geologic activity is also thought to be the most likely
culprit for releasing methane stored deep in Titan into the atmosphere to replenish the losses to
photochemistry. However, no “smoking vent” has been discovered, and whether the surface is
geologically active in the present day remains uncertain.
Observations of Titan’s clouds provide additional intriguing hints that geologic processes on
the surface may play a role in Titan’s weather, although the evidence is far from definitive. The
first episodes of the 40◦ midlatitude clouds were found to be largely clustered in a region around
350◦ W (Roe et al. 2005b), and this was suggested to possibly be a sign of episodic geologic
venting of methane. These clouds appeared several years earlier than would be expected if they
were a purely seasonal phenomenon following the point of maximum insolation. Although later
analyses of Cassini observations found no longitudinal clustering of clouds in this latitude band, this
disagreement can be explained by the differences in spatial and temporal sampling of the Cassini
and Earth-based observations. As discussed above, the clustered midlatitude clouds were observed
from late 2003 to early 2005, during which Cassini had only a few observing opportunities. The
conclusion about the lack of longitudinal clustering made from Cassini data is based primarily on
observations in 2006 and later. Ground-based observations agree that the post-2006 midlatitude
clouds were not clustered in longitude.
376 Roe
Furthermore, the Earth-based observations sample Titan on a much more frequent basis, often
nightly, than Cassini’s much less frequent Titan flybys. However, the Earth-based observations
are sensitive only to larger clouds, whereas most of the clouds reported in Cassini data are smaller
and fainter than can be detected in the observations from Earth. Therefore, it seems likely that
the observations are effectively sampling two different populations, with Earth-based observations
showing the rare and large events and Cassini revealing smaller but more frequent clouds. If the
initial midlatitude clouds were formed via geologic pulses of methane into the atmosphere, then
that could possibly explain the temporal coincidence that the large south polar storm event of
October 2004 (Schaller et al. 2006a) appeared just a month after some of the most intense midlat-
itude cloud activity (Roe et al. 2005a). Schaller et al. (2006a) suggested that the large south polar
storm could have been triggered by poleward transport of methane injected into the atmosphere
at latitudes further north.
The April 2008 storm, one of the largest observed to date, originated near 15◦ S, 250◦ W. A
faint small cloud persisted in the region for at least 19 days, even after the ongoing storm system
had moved 90◦ in longitude to the east (Schaller et al. 2009). Why the region where the storm
initiated remained covered in thin clouds for so long is mysterious, although it could be a sign
that the large storm was triggered by cryovolcanic or other geologic activity injecting a pulse of
methane into the atmosphere, after which the activity tapered off over several weeks. Intriguingly,
this region of Titan’s surface is the same region where, 2.5 years later in the wake of another large
storm, Turtle et al. (2011b) observed evidence of rainfall wetting the surface. However, none of
the candidate cryovolcanoes are found in this region, and the colocation of these events years apart
may simply be a coincidence. Another possibility is that topography played a role in influencing
cloud formation, as models now demonstrate that even the modest terrain on Titan (1–2-km-high
mountains) can have a significant role in cloud formation (Barth 2010). It is possible that all of
the hints of geographically clustered clouds that are not fully explained by circulation are actually
orographic clouds linked to the underlying terrain, although this has not been thoroughly tested.
The triggering mechanism for the longitudinally clustered midlatitude clouds of late 2003
to early 2005 remains unknown, but some type of geologic event, such as cryovolcanism, is the
leading hypothesis. The coincidence of large storms at 15◦ S, 250◦ W is also unexplained. Further
Cassini observations of Titan’s surface and frequent ground-based monitoring of Titan’s clouds
should help determine whether Titan’s surface is geologically active in the present day.
SUMMARY POINTS
1. Titan has active meteorological and hydrological processes in its atmosphere and on its
surface that are directly analogous to those on Earth.
2. Titan’s methane meteorology is most manifest in the observations of tropospheric clouds
accumulated over more than a decade of Earth-based observations and during dozens of
flybys of Titan by the Cassini spacecraft. Frequent observations of Titan’s clouds have
been ongoing for less than a decade, from the early-to-mid-2000s to the present day,
which covers only early northern winter to early northern spring.
3. Observed cloud locations are largely explained by seasonal circulation, although geog-
raphy appears to play a secondary role in the formation and location of certain clouds.
Lake-effect clouds have been observed, and there is suspicion that active geology, such
as cryovolcanism, geysering, or other means of injecting methane into the atmosphere,
may play a role in forcing certain episodic clouds.

4. Extrapolating from the existing cloud and precipitation record reveals that no portion of
Titan’s surface is excluded year-round from cloudy skies and the possibility of precipita-
tion. Even the arid equatorial dunes can experience cloudy skies and precipitation.
5. The maxim that “extremes are more important than averages” is prominent throughout
our understanding of Titan. For instance, the equatorial dunes are formed by strong
westerly winds that only briefly interrupt the dominant easterlies around each equinox.
Many regions of the surface are dry for the majority of each Titan year but have numerous
fluvial features that are formed during infrequent but violent downpours. This concept
may even extend to multimillennial timescales if Titan’s atmosphere oscillates between
warmer, wetter and colder, dryer states.
FUTURE ISSUES
1. One of the biggest outstanding questions is: When and where does it rain on Titan?
Continued observations along with improved “wet” GCM models following a Titan
System Science approach are required.
2. Over Titan’s lifetime, photochemical models predict the production of several hundred
meters of ethane, which is far more than the volume of all of Titan’s lakes. Even if the
ethane is consumed into higher hydrocarbons, the volume of hydrocarbons produced in
the photochemical models is far more than the lakes and dunes combined can account
for. Either Titan’s atmosphere has produced a much smaller volume of hydrocarbons
over history than the models predict or significant volumes of hydrocarbons have been
sequestered from the surface into Titan’s crust or deeper interior by as-yet-unexplained
processes.
3. The source and mechanism for replenishing methane to the atmosphere remain uncer-
tain, including whether the replenishment is episodic or continuous. In either case, it
seems unlikely that the replenishment rate always perfectly balances the destruction rate,
which is controlled by the solar ultraviolet flux. The extent to which the surface lakes play
a buffering role when the rates of methane injection and destruction are not balanced is
unknown.
4. Much remains uncertain about material behavior at Titanian conditions, including how
methane cloud particles form; the physics of how hydrocarbon liquids sculpt the surface;
and evaporation rates of mixtures of methane, ethane, nitrogen, and heavier hydrocar-
bons. A rich vein of laboratory research is available in this area.
5. Further observations are needed to constrain GCM predictions, including wind mea-
surements, cloud observations, and Cassini radar measurements of changes to the length
of Titan’s day.
6. When and how will seasonal asymmetries due to Saturn’s eccentric orbit about the Sun
manifest themselves?
378 Roe
7. Cassini will be crashed into Saturn on September 18, 2017, after more than 100 Titan
flybys over 13 years. NASA has shortlisted the Titan Mare Explorer as a possible future
mission. This mission, which would be much simpler than Cassini, would deliver an in-
strumented buoy to land and float on one of the northern lakes and return observations
for several months. Other mission concepts for exploring Titan include blimps, heli-
copters, and airplanes. However, for understanding Titan’s methane weather, the most
scientific value would be gained from a long-lived Titan-orbiting weather satellite.
8. How variable is Titan’s weather from one year to the next? The stratospheric haze is
known to change from one Titan year to the next. Long-term multidecadal studies of
Titan’s tropospheric weather are needed.
DISCLOSURE STATEMENT
The author is not aware of any affiliations, memberships, funding, or financial holdings that might
be perceived as affecting the objectivity of this review.
ACKNOWLEDGMENTS
Thank you to Emily Schaller and Michael Brown for numerous fruitful discussions. This work
was funded in part by NASA Planetary Astronomy grants NNX07AK74G and NNX11AH46G.
LITERATURE CITED
Aharonson O, Hayes AG, Lunine JI, Lorenz RD, Allison MD, Elachi C. 2009. An asymmetric distribution of
lakes on Titan as a possible consequence of orbital forcing. Nat. Geosci. 2:851–54
Atreya SK, Adams EY, Niemann HB, Demick-Montelara JE, Owen TC, et al. 2006. Titan’s methane cycle.
Planet. Space Sci. 54:1177–87
Barnes JW, Brown RH, Soderblom JM, Soderblom LA, Jaumann R, et al. 2009. Shoreline features of Titan’s
Ontario Lacus from Cassini/VIMS observations. Icarus 201:217–25
Barth EL. 2010. Cloud formation along mountain ridges on Titan. Planet. Space Sci. 58:1740–47
Barth EL, Rafkin SCR. 2007. TRAMS: a new dynamic cloud model for Titan’s methane clouds. Geophys. Res.
Lett. 34:L03203
Barth EL, Rafkin SCR. 2010. Convective cloud heights as a diagnostic for methane environment on Titan.
Icarus 206:467–84
Bouchez AH, Brown ME. 2005. Statistics of Titan’s South Polar Tropospheric Clouds. Astrophys. J. 618:L53–
56
Brown ME, Bouchez AH, Griffith CA. 2002. Direct detection of variable tropospheric clouds near Titan’s
south pole. Nature 420:795–97
Brown ME, Roberts JE, Schaller EL. 2010. Clouds on Titan during the Cassini prime mission: a complete
analysis of the VIMS data. Icarus 205:571–80
Brown ME, Schaller EL, Roe HG, Chen C, Roberts J, et al. 2009. Discovery of lake-effect clouds on Titan.
Geophys. Res. Lett. 36:L01103
Brown RH, Soderblom LA, Soderblom JM, Clark RN, Jaumann R, et al. 2008. The identification of liquid
ethane in Titan’s Ontario Lacus. Nature 454:607–10
Cordier D, Mousis O, Lunine JI, Lavvas P, Vuitton V. 2009. An estimate of the chemical composition of
Titan’s lakes. Astrophys. J. 707:L128–31
Cordier D, Mousis O, Lunine JI, Lebonnois S, Rannou P, et al. 2012. Titan’s lakes chemical composition:
sources of uncertainties and variability. Planet. Space Sci. 61:99–107

Courtin R, Gautier D, McKay CP. 1995. Titan’s thermal emission spectrum: reanalysis of the Voyager infrared
measurements. Icarus 114:144–62
Coustenis A, Bezard B, Gautier D. 1989. Titan’s atmosphere from Voyager infrared observations: I. The gas
composition of Titan’s equatorial region. Icarus 80:54–76
Curtis DB, Hatch CD, Hasenkopf CA, Toon OB, Tolbert MA, et al. 2008. Laboratory studies of methane and
ethane adsorption and nucleation onto organic particles: application to Titan’s clouds. Icarus 195:792–801
de Pater I, Ádámkovics M, Bouchez AH, Brown ME, Gibbard SG, et al. 2006. Titan imagery with Keck
adaptive optics during and after probe entry. J. Geophys. Res. 111:E07S05
Fink U, Larson HP. 1979. The infrared spectra of Uranus, Neptune, and Titan from 0.8 to 2.5 microns.
Astrophys. J. 233:1021–40
Gendron E, Coustenis A, Drossart P, Combes M, Hirtzig M, et al. 2004. VLT/NACO adaptive optics imaging
of Titan. Astron. Astrophys. 417:L21–24
Gibbard SG, Macintosh B, Gavel D, Max CE, de Pater I, et al. 2004. Speckle imaging of Titan at 2 microns:
surface albedo, haze optical depth, and tropospheric clouds 1996–1998. Icarus 169:429–39
Griffith CA, Hall JL, Geballe TR. 2000. Detection of daily clouds on Titan. Science 290:509–13
Griffith CA, Owen T, Miller GA, Geballe T. 1998. Transient clouds in Titan’s lower atmosphere. Nature
395:575–78
Griffith CA, Penteado P, Baines K, Drossart P, Barnes J, et al. 2005. The evolution of Titan’s mid-latitude
clouds. Science 310:474–77
Griffith CA, Penteado P, Rannou P, Brown R, Boudon V, et al. 2006. Evidence for a polar ethane cloud on
Titan. Science 313:1620–22
Hayes A, Aharonson O, Callahan P, Elachi C, Gim Y, et al. 2008. Hydrocarbon lakes on Titan: distribution
and interaction with a porous regolith. Geophys. Res. Lett. 35:L09204
Hayes AG, Aharonson O, Lunine JI, Kirk RL, Zebker HA, et al. 2011. Transient surface liquid in Titan’s
polar regions from Cassini. Icarus 211:655–71
Hourdin F, Talagrand O, Sadourny R, Courtin R, Gautier D, McKay CP. 1995. Numerical simulation of the
general circulation of the atmosphere of Titan. Icarus 117:358–74
Irwin PGJ, Sromovsky LA, Strong EK, Sihra K, Teanby NA, et al. 2006. Improved near-infrared methane
band models and k-distribution parameters from 2000 to 9500 cm−1 and implications for interpretation
of outer planet spectra. Icarus 181:309–19
Jaumann R, Brown RH, Stephan K, Barnes JW, Soderblom LA, et al. 2008. Fluvial erosion and post-erosional
processes on Titan. Icarus 197:526–38
Jaumann R, Kirk RL, Lorenz RD, Lopes RMC, Stofan E, et al. 2010. Geology and surface processes on Titan.
In Titan from Cassini-Huygens, ed. RH Brown, J-P Lebreton, JH Waite, pp. 75–140. Dordrecht: Springer
Karkoschka E. 1994. Spectrophotometry of the jovian planets and Titan at 300- to 1000-nm wavelength: the
methane spectrum. Icarus 111:174–92
Kirk RL, Howington-Kraus E, Barnes JW, Hayes AG, Lopes RM, et al. 2010. La Sotra y las otras: topographic
evidence for (and against) cryovolcanism on Titan. Presented at 2010 Fall Meet., AGU, Dec. 13–17, San
Francisco (Abstr. P22A-03)
Lemmon MT, Karkoschka E, Tomasko M. 1993. Titan’s rotation: surface feature observed. Icarus 103:329–32
Lindal GF, Wood GE, Hotz HB, Sweetnam DN, Eshleman VR, Tyler GL. 1983. The atmosphere of Titan:
an analysis of the Voyager 1 radio occultation measurements. Icarus 53:348–63
Lockwood GW, Thompson DT. 2009. Seasonal photometric variability of Titan, 1972–2006. Icarus 200:616–
26
Lorenz RD. 2000. The weather on Titan. Science 290:467–68
Lorenz RD. 2008. The changing face of Titan. Phys. Today 61:34–39
Lorenz RD, Brown ME, Flasar FM. 2010. Seasonal change on Titan. In Titan from Cassini-Huygens, ed. RH
Brown, J-P Lebreton, JH Waite, pp. 353–72. Dordrecht: Springer
Lorenz RD, Griffith CA, Lunine JI, McKay CP, Rennò NO. 2005. Convective plumes and the scarcity of
Titan’s clouds. Geophys. Res. Lett. 32:L01201
Lorenz RD, Lopes RM, Paganelli F, Lunine JI, Kirk RL, et al. 2008a. Fluvial channels on Titan: initial Cassini
RADAR observations. Planet. Space Sci. 56:1132–44
380 Roe
Lorenz RD, McKay CP, Lunine JI. 1997. Photochemically driven collapse of Titan’s atmosphere. Science
275:642–44
Lorenz RD, Mitchell KL, Kirk RL, Hayes AG, Aharonson O, et al. 2008b. Titan’s inventory of organic surface
materials. Geophys. Res. Lett. 35:L02206
Lorenz RD, Radebaugh J. 2009. Global pattern of Titan’s dunes: Radar survey from the Cassini prime mission.
Geophys. Res. Lett. 36:L03202
Lorenz RD, Stiles BW, Kirk RL, Allison MD, Persi del Marmo P, et al. 2008c. Titan’s rotation reveals an
internal ocean and changing zonal winds. Science 319:1649–51
Lorenz RD, West RD, Johnson WTK. 2008d. Cassini RADAR constraint on Titan’s winter polar precipita-
tion. Icarus 195:812–16
Lorenz RD, Wood CA, Lunine JI, Wall SD, Lopes RM, et al. 2007. Titan’s young surface: initial impact
crater survey by Cassini RADAR and model comparison. Geophys. Res. Lett. 34:L07204
Lunine JI, Elachi C, Wall SD, Janssen MA, Allison MD, et al. 2008. Titan’s diverse landscapes as evidenced
by Cassini RADAR’s third and fourth looks at Titan. Icarus 195:415–33

Lunine JI, Lorenz RD. 2009. Rivers, lakes, dunes, and rain: crustal processes in Titan’s methane cycle. Annu.
Rev. Earth Planet. Sci. 37:299–320
McKay CP, Coustenis A, Samuelson RE, Lemmon MT, Lorenz RD, et al. 2001. Physical properties of the
organic aerosols and clouds on Titan. Planet. Space Sci. 49:79–99
McKay CP, Pollack JB, Courtin R. 1991. The greenhouse and antigreenhouse effects on Titan. Science
253:1118–21
Mitchell JL, Pierrehumbert RT, Frierson DMW, Caballero R. 2006. The dynamics behind Titan’s methane
clouds. Proc. Natl. Acad. Sci. USA 103:18421–26
Mitchell JL, Pierrehumbert RT, Frierson DMW, Caballero R. 2009. The impact of methane thermodynamics
on seasonal convection and circulation in a model Titan atmosphere. Icarus 203:250–64
Mitri G, Showman AP. 2008. Thermal convection in ice-I shells of Titan and Enceladus. Icarus 193:387–96
Mitri G, Showman AP, Lunine JI, Lopes RMC. 2008. Resurfacing of Titan by ammonia-water cryomagma.
Icarus 196:216–24
Mitri G, Showman AP, Lunine JI, Lorenz RD. 2007. Hydrocarbon lakes on Titan. Icarus 186:385–94
Moore JM, Pappalardo RT. 2011. Titan: an exogenic world? Icarus 212:790–806
Penteado PF, Griffith CA. 2010. Ground-based measurements of the methane distribution on Titan. Icarus
206:345–51
Porco CC, Baker E, Barbara J, Beurle K, Brahic A, et al. 2005. Imaging of Titan from the Cassini spacecraft.
Nature 434:159–68
Radebaugh J, Lorenz RD, Kirk RL, Lunine JI, Stofan ER, et al. 2007. Mountains on Titan observed by Cassini
Radar. Icarus 192:77–91
Rannou P, Montmessin F, Hourdin F, Lebonnois S. 2006. The latitudinal distribution of clouds on Titan.
Science 311:201–5
Richardson J, Lorenz RD, McEwen A. 2004. Titan’s surface and rotation: new results from Voyager 1 images.
Icarus 170:113–24
Richardson MI, Toigo AD, Newman CE. 2007. PlanetWRF: a general purpose, local to global numerical
model for planetary atmospheric and climate dynamics. J. Geophys. Res. 112:E09001
Roe HG, Bouchez AH, Trujillo CA, Schaller EL, Brown ME. 2005a. Discovery of temperate latitude clouds
on Titan. Astrophys. J. 618:L49–52
Roe HG, Brown ME, Schaller EL, Bouchez AH, Trujillo CA. 2005b. Geographic control of Titan’s mid-
latitude clouds. Science 310:477–79
Roe HG, de Pater I, Macintosh BA, McKay CP. 2002. Titan’s clouds from Gemini and Keck adaptive optics
imaging. Astrophys. J. 581:1399–406
Schaller EL, Brown ME, Roe HG, Bouchez AH. 2006a. A large cloud outburst at Titan’s south pole. Icarus
182:224–29
Schaller EL, Brown ME, Roe HG, Bouchez AH, Trujillo CA. 2006b. Dissipation of Titan’s south polar clouds.
Icarus 184:517–23
Schaller EL, Roe HG, Schneider T, Brown ME. 2009. Storms in the tropics of Titan. Nature 460:873–75

Smith BA, Soderblom L, Batson RM, Bridges PM, Inge JL, et al. 1982. A new look at the Saturn system: the
Voyager 2 images. Science 215:504–37
Stiles BW, Kirk RL, Lorenz RD, Hensley S, Lee E, et al. 2008. Determining Titan’s spin state from Cassini
RADAR images. Astron. J 135:1669–80
Strobel DF. 1974. The photochemistry of hydrocarbons in the atmosphere of Titan. Icarus 21:466–70
Tokano T. 2010. Relevance of fast westerlies at equinox for the eastward elongation of Titan’s dunes. Aeolian
Res. 2:113–27
Tokano T. 2011. Precipitation climatology on Titan. Science 331:1393–94
Tokano T, Ferri F, Colombatti G, Mäkinen T, Fulchignoni M. 2006. Titan’s planetary boundary layer
structure at the Huygens landing site. J. Geophys. Res. 111:E08007
Tokano T, Neubauer FM. 2002. Tidal winds on Titan caused by Saturn. Icarus 158:499–515
Tokano T, Neubauer FM. 2005. Wind-induced seasonal angular momentum exchange at Titan’s surface and
its influence on Titan’s length-of-day. Geophys. Res. Lett. 32:L24203
Tokano T, Neubauer FM, Laube M, McKay CP. 1999. Seasonal variation of Titan’s atmospheric structure
simulated by a general circulation model. Planet. Space Sci. 47:493–520

Tokano T, Neubauer FM, Laube M, McKay CP. 2001. Three-dimensional modeling of the tropospheric
methane cycle on Titan. Icarus 153:130–47
Tomasko MG, Archinal B, Becker T, Bézard B, Bushroe M, et al. 2005. Rain, winds and haze during the
Huygens probe’s descent to Titan’s surface. Nature 438:765–78
Tomasko MG, Doose L, Engel S, Dafoe LE, West R, et al. 2008. A model of Titan’s aerosols based on
measurements made inside the atmosphere. Planet. Space Sci. 56:669–707
Toon OB, McKay CP, Courtin R, Ackerman TP. 1988. Methane rain on Titan. Icarus 75:255–84
Turtle EP, Del Genio AD, Barbara JM, Perry JE, Schaller EL, et al. 2011a. Seasonal changes in Titan’s
meteorology. Geophys. Res. Lett. 38:L03203
Turtle EP, Perry JE, Hayes AG, Lorenz RD, Barnes JW, et al. 2011b. Rapid and extensive surface changes
near Titan’s equator: evidence of April showers. Science 331:1414–17
Turtle EP, Perry JE, McEwen AS, Del Genio AD, Barbara J, et al. 2009. Cassini imaging of Titan’s high-
latitude lakes, clouds, and south-polar surface changes. Geophys. Res. Lett. 36:L02204
Wall SD, Lopes RM, Stofan ER, Wood CA, Radebaugh JL, et al. 2009. Cassini RADAR images at Hotei
Arcus and western Xanadu, Titan: evidence for geologically recent cryovolcanic activity. Geophys. Res.
Lett. 36:L04203
Yung YL, Allen M, Pinto JP. 1984. Photochemistry of the atmosphere of Titan: comparison between model
and observations. Astrophys. J. Suppl. Ser. 55:465–506
Zahnle K, Pollack JB, Grinspoon D, Dones L. 1992. Impact-generated atmospheres over Titan, Ganymede,
and Callisto. Icarus 95:1–23
382 Roe
EA40-FrontMatter ARI 1 April 2012 7:38
Annual Review
of Earth and
Planetary Sciences
Volume 40, 2012 Contents
Reminiscences From a Career in Geomicrobiology

Henry L. Ehrlich p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 1
Mixing and Transport of Isotopic Heterogeneity

in the Early Solar System
Alan P. Boss p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p23
Tracing Crustal Fluids: Applications of Natural 129 I and 36 Cl
Udo Fehn p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p45
SETI@home, BOINC, and Volunteer Distributed Computing
Eric J. Korpela p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p69
End-Permian Mass Extinction in the Oceans: An Ancient Analog
for the Twenty-First Century?
Jonathan L. Payne and Matthew E. Clapham p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p89
Magma Oceans in the Inner Solar System
Linda T. Elkins-Tanton p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 113
History of Seawater Carbonate Chemistry, Atmospheric CO2 ,
and Ocean Acidification
Richard E. Zeebe p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 141
Biomimetic Properties of Minerals and the Search for Life in the
Martian Meteorite ALH84001
Jan Martel, David Young, Hsin-Hsin Peng, Cheng-Yeu Wu, and John D. Young p p p p 167
Archean Subduction: Fact or Fiction?
Jeroen van Hunen and Jean-François Moyen p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 195
Molecular Paleohydrology: Interpreting the Hydrogen-Isotopic
Composition of Lipid Biomarkers from
Photosynthesizing Organisms
Dirk Sachse, Isabelle Billault, Gabriel J. Bowen, Yoshito Chikaraishi, Todd E. Dawson,
Sarah J. Feakins, Katherine H. Freeman, Clayton R. Magill, Francesca A. McInerney,
Marcel T.J. van der Meer, Pratigya Polissar, Richard J. Robins, Julian P. Sachs,
Hanns-Ludwig Schmidt, Alex L. Sessions, James W.C. White, Jason B. West,
and Ansgar Kahmen p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 221
viii
Building Terrestrial Planets

A. Morbidelli, J.I. Lunine, D.P. O’Brien, S.N. Raymond, and K.J. Walsh p p p p p p p p p p p p 251
Paleontology of Earth’s Mantle
Norman H. Sleep, Dennis K. Bird, and Emily Pope p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 277
Molecular and Fossil Evidence on the Origin of Angiosperms
James A. Doyle p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 301
Infrasound: Connecting the Solid Earth, Oceans, and Atmosphere
M.A.H. Hedlin, K. Walker, D.P. Drob, and C.D. de Groot-Hedlin p p p p p p p p p p p p p p p p p p p p 327
Titan’s Methane Weather
Henry G. Roe p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 355
Extratropical Cooling, Interhemispheric Thermal Gradients,

and Tropical Climate Change
John C.H. Chiang and Andrew R. Friedman p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 383
The Role of H2 O in Subduction Zone Magmatism
Timothy L. Grove, Christy B. Till, and Michael J. Krawczynski p p p p p p p p p p p p p p p p p p p p p p p p 413
Satellite Geomagnetism
Nils Olsen and Claudia Stolle p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 441
The Compositions of Kuiper Belt Objects
Michael E. Brown p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 467
Tectonics of the New Guinea Region
Suzanne L. Baldwin, Paul G. Fitzgerald, and Laura E. Webb p p p p p p p p p p p p p p p p p p p p p p p p p 495
Processes on the Young Earth and the Habitats of Early Life
Nicholas T. Arndt and Euan G. Nisbet p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 521
The Deep, Dark Energy Biosphere: Intraterrestrial Life on Earth
Katrina J. Edwards, Keir Becker, and Frederick Colwell p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 551
Geophysics of Chemical Heterogeneity in the Mantle
Lars Stixrude and Carolina Lithgow-Bertelloni p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 569
The Habitability of Our Earth and Other Earths: Astrophysical,
Geochemical, Geophysical, and Biological Limits on Planet
Habitability
Charles H. Lineweaver and Aditya Chopra p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 597
The Future of Arctic Sea Ice
Wieslaw Maslowski, Jaclyn Clement Kinney, Matthew Higgins,
and Andrew Roberts p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 625
The Mississippi Delta Region: Past, Present, and Future
Michael D. Blum and Harry H. Roberts p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 655
Contents ix
Climate Change Impacts on the Organic Carbon Cycle at the

Land-Ocean Interface
Elizabeth A. Canuel, Sarah S. Cammer, Hadley A. McIntosh,
and Christina R. Pondell p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 685
Indexes
Cumulative Index of Contributing Authors, Volumes 31–40 p p p p p p p p p p p p p p p p p p p p p p p p p p p 713

Cumulative Index of Chapter Titles, Volumes 31–40 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 717
Errata
An online log of corrections to Annual Review of Earth and Planetary Sciences articles
may be found at http://earth.annualreviews.org
x Contents

Titan&#39;s Methane Weather

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Titan&#39;s Methane Weather

Uploaded by

Copyright:

Available Formats

Titan’s Methane Weather

Annu. Rev. Earth Planet. Sci. 2012. 40:355–82 Keywords

This article’s doi: Abstract

2.1. In the Atmosphere

www.annualreviews.org • Titan’s Methane Weather 357

2.2. Below the Surface

2.3. On the Surface

www.annualreviews.org • Titan’s Methane Weather 359

August 2003 Spring equinox

August 2009 October 2046

2.5. Methane’s Important Role

Southern Southern Northern Northern Southern

www.annualreviews.org • Titan’s Methane Weather 361

3. OBSERVING TITAN’S ATMOSPHERE

0.5 μm 2.03 μm 2.12 μm 2.22 μm

0.5 1.0 2.0 3.0 4.0 5.0

www.annualreviews.org • Titan’s Methane Weather 363

1995 1998 2001 2004 2010

www.annualreviews.org • Titan’s Methane Weather 365

Surface Troposphere Stratosphere

2.05 2.10 2.15 2.20 2.25

2008-03-25 UT 2008-03-27 UT 2008-04-13 UT 2008-04-14 UT 2008-04-15 UT

2008-04-16 UT 2008-04-17 UT 2008-04-18 UT 2008-04-20 UT 2008-04-21 UT

2008-04-22 UT 2008-04-23 UT 2008-04-25 UT 2008-04-28 UT 2008-05-01 UT

2008-05-03 UT 2008-05-04 UT 2008-05-05 UT 2008-05-12 UT

www.annualreviews.org • Titan’s Methane Weather 367

www.annualreviews.org • Titan’s Methane Weather 369

5.1. Cloud Heights

formation (Brown et al. 2002).

Southern Geographically controlled clouds

www.annualreviews.org • Titan’s Methane Weather 371

5.4. Large Events and Rainfall

www.annualreviews.org • Titan’s Methane Weather 373

5.5. Geographic Connections: Lake-Effect Clouds

www.annualreviews.org • Titan’s Methane Weather 375

5.6. Geographic Connections: Hints of Geologic Activity

www.annualreviews.org • Titan’s Methane Weather 377

www.annualreviews.org • Titan’s Methane Weather 379

by Cassini RADAR’s third and fourth looks at Titan. Icarus 195:415–33

www.annualreviews.org • Titan’s Methane Weather 381

simulated by a general circulation model. Planet. Space Sci. 47:493–520

Reminiscences From a Career in Geomicrobiology

Mixing and Transport of Isotopic Heterogeneity

Building Terrestrial Planets

Extratropical Cooling, Interhemispheric Thermal Gradients,

Climate Change Impacts on the Organic Carbon Cycle at the

Cumulative Index of Contributing Authors, Volumes 31–40 p p p p p p p p p p p p p p p p p p p p p p p p p p p 713

You might also like

Titan's Methane Weather

Titan's Methane Weather