Titan's topography, strange.... Options

[Juramike]

Recent article in Science by Zebker et al.:

Zebker et al. Science in press, "Size and Shape of Saturn's Moon Titan". doi: 10.1126/science.1168905
(published online April 2, 2009)

Link to abstract (pay-for article): http://www.sciencemag.org/cgi/content/abstract/1168905

Article on spaceref discusses this paper: http://www.spaceref.com/news/viewpr.html?pid=27912

Figure 3 from the Science article is a global elevation map relative to barycenter.

Key points of article:

• Poles are squished - might explain why lakes are up there
• "Mountains" aren't necessarily elevated - they might've sunk down into the crust to form "a basin of their own creation".
• (quote from spaceref. article)
• Large scale features do not appear correlated with elevation
• Xanadu and Tseghi are BASINS! (ca. -600 m to barycenter)
• Adiri is higher than Xanadu (by almost 1 km)
• Dilmun is also pretty elevated (+ 400 m relative to barycenter)
• highest elevated terrain on Titan seems to be region around "Adiri junior" in the S Senkyo "basin" at ca. + 600 m above barycenter
• Shangri-La "basin" is elevated +400-800 m ABOVE Xanadu (using barycenter elevations in Fig. 3)

"Xanadu seems to be systematically lower than other parts of the equatorial belt, and not uplifted like most mountainous areas on Earth." (quote from Fig. 3 caption in article)

-Mike

[Juramike]

Here's a cartooon showing how Xanadu could have formed:


1) A diapiric rise of warm low density (mostly due to warmth) water ice
2) reaches surface; erosion begins
3) cools and contracts; ice now denser (heavier) than surrounding crust; crustal deformation begins
3) regional depression begins as old diaipir sinks due to now-cooled higher density material

Quote from the spaceref article:

"Zebker said that if you look at images of the surface of Titan, you see surface features that look every bit like mountains on Earth but don't have the high elevations compared to the plains stretching out around them.

'One of the really surprising finds that we have from this, is that the largest apparent continent is lower than the average elevation on Titan, as opposed to higher than the average elevation, as we have on the Earth,' Zebker said.

'My favorite explanation is that the material that forms the mountains is simply more dense than the material surrounding them,' he said. That would result in the mountains pushing down the surrounding crust, effectively putting the mountains in a basin of their own creation.

On Earth, the situation is the reverse: The crust that lies under the oceans is denser than the material that makes up the continental crust, where mountain ranges are built up.

'The things that we would expect to exist on the surface of Titan would either be solid hydrocarbon materials, essentially frozen ethane and methane, and that is fairly light, and then frozen water ice, which is denser,' Zebker said. 'If the mountains are composed of water ice and the plain features in between are composed of these solid hydrocarbons, that could lead to this kind of a situation.' "

[Juramike]

The Science article is in reference to the Barycentric center of Titan.
This could be different from the equipotential surface.

If I understand this correctly....

In a perfect world, with a uniform density gradient, the elevation and equipotential surfaces pretty much match:


A large mass concentration (high density crustal materials) can cause a localized gravity well, and cause the equipotential surface to bulge out. (More mass, more gravity, more "pull"). So you can get an elevated ocean that covers a elevated (based on barycenter distance) rise.


A mass deficit (lower density crustal materials) can cause a localized gravity deficit and cause the equipotential surface to dip in. (less mass, less gravity, less "pull"). This could cause a basin that one would think should be flooded, to be dry. The water table responds to the equipotential surface, not the elevation.


http://en.wikipedia.org/wiki/Geoid

[Juramike]

So now a Big Question:
Why are the poles wet, the equatorial Sand Seas dry, and Xanadu dry and sand-free?

The elevation difference can explain why the poles are wet:

(from Zebker et al. Science in press, "Size and Shape of Saturn's Moon Titan". doi: 10.1126/science.1168905
(published online April 2, 2009):

If we posit that the lakes are surface expressions of a more or less continuous liquid organic "water table," then the lower elevations of the poles could lead to the observed preponderance of lakes at high latitudes. However, whether the polar surface intersects a methane table depends on its distance from a constant gravity potential surface, and not the on its elevation from the barycenter, because the equipotential may be depressed as well at the poles.

But keeping lower lying downwind Xanadu free from Shangri-La's dune sands is harder to explain.

If the equipotential is depressed at Xanadu (due to a lower overall density of crustal materials), this could explain why the mobile dune sands remain in Shangri-La (which is higher barycentric elevation) and don't fill in Xanadu.


Otherwise some funky physical barriers would need to be invoked.

[The sand seas could be isolated from the north polar lakes due to many reasons: an isolating temperate airfall deposit, lack of sand reservoirs, lack of winds carrying material into the poles, and also sands getting trapped by the lakes]


This Science paper is awesome. It raises a lot of really tough questions.....

[ngunn]

Thanks for starting off a discussion on this. One fairly safe conclusion would seem to be that for Titan a triaxial ellipsoid is a very poor fit to an equipotential. The latter must be very bumpy compared with what we're used to. But as you say there are problems everywhere with this (which likely means that it will be a very productive observation in the long run). Are Xanadu's mountains dense, or light and porous? We have observational evidence for the latter, yet here it is suggested that they are denser than surrounding terrain and thus capable of making their own basin to sit in.

A general observation (not necessarily helpful): topsy-turvy topography is less difficult to explain away if ALL the surface materials (except the liquids) are of very low density, i.e. fluffy or aerated.

[Doc]

Forgive my ignorance ngunn but exactly what was that observational evidence that indicates light and porous material in Xanadu?

[ngunn]

It came from RADAR reflectance observations of the dielectric properties if I remember correctly. It was a while ago now, but it led to extensive discussion here and elsewhere about possible underground cave systems (likened to Xanadu in the poem). Emily's Planetary Society blog had at least one entry on it at the time, which may or may not have been called something like 'Caverns measureless to man'. Sorry not to be more specific - I wish I was better at keeping references.

[Juramike]

Nigel's right, the most recent evidence comes from a Janssen article on dielectric constant (discussed here: Titan's Mid Latitudes thread, post 6). The dielectric constant in some parts of Xanadu was waaay too low to be of any normal materials, even very low dielectric organics.

Dielectric constant is modified by fluffiness (pores). Volume scattering can make the material appear to have a much lower dielectric constant. It was postulated that volume scattering was responsible for the apparent low dielectric constant.

[Juramike]

The channel drainage pattern in W Xanadu in the T13 SAR RADAR Swath is very consistent with the Zebker elevation data.

(For E Xanadu, check out Fig 2b the freely available LPSC abstract: http://www.lpi.usra.edu/meetings/lpsc2009/pdf/1037.pdf. For W Xanadu, check out Fig. 7 a and b in Lorenz et al. Planetary and Space Science 56 (2008) 1132-1144. "Fluvial Channels on Titan: Initial Cassini RADAR observations." doi: 10.1016/j.psss.2008.02.009

The dendritic drainage pattern across the T13 RADAR Swath pattern is to the S, towards the lowest regional barycentric elevation determined to be in SW Xanadu.

Thus, the elevation data seems to agree with the local geoid in W Xanadu.

Back to the tough question, what prevents the Shangri-La dune sands from dropping down into the SW Xanadu low elevation basin???

****

The Zebker paper also makes the airfall origin of the sand sea dunes difficult to explain. If atmospheric deposition and sintering is supposedly occuring all over, it should occur over Xanadu. The regional gradient should make it collect in the SW Xanadu basin. According to this idea, there should be a dune sea filling in the SW Xanadu depression. But there isn't. (At least no dunes are seen by RADAR, and ISS shows a bright coating rather than a dark dune sea.) There is, however, a smooth (RADAR) dark area seen with channels crossing into it, again going downgradient.

[rlorenz]

Forgive my ignorance ngunn but exactly what was that observational evidence that indicates light and porous material in Xanadu?

As another post noted, Xanadu has very peculiar dielectric properties - its radar echo is quite
depolarized, suggestive of porosity (see the Janssen paper)

There is also the result published by Buratti et al of the VIMS team suggesting that the near-IR
phase function of Xanadu suggests rougher material (at the microscale, IIRC)

[rlorenz]

Thanks for starting off a discussion on this. One fairly safe conclusion would seem to be that for Titan a triaxial ellipsoid is a very poor fit to an equipotential.
...
A general observation (not necessarily helpful): topsy-turvy topography is less difficult to explain away if ALL the surface materials (except the liquids) are of very low density, i.e. fluffy or aerated.

For your first remark, we'll need to wait to see what the published gravity field is. If the 3rd order terms are, in
fact low (at one time the retrievals suggested they might not be), then in fact a triaxial ellipsoid IS a good fit
to the EQUIPOTENTIAL.

Maybe (since you are referring to the shape measured from radar) you mean that a triaxial ellipsoid isnt a good
fit to the figure ?

As far as liquids draining downhill goes, there does seem to be a consistency between lakes at high latitudes, and
the hint of a tendency (with incomplete sampling) in my fluvial paper for rivers to flow poleward, and the
difference between the topographic shape and reasonable geopotentials.

The low-latitude longitudinal contrasts (viz, low Xanadu vs the dunes) is harder to explain, as is the orientation
of the dunes themselves. I sometimes wonder about albedo-driven wind (i.e. sea-breeze type circulations -
with effectively katabatic flow away from high albedo regions).

The Zebker et al (you can call me al !) paper about the shape - like the Lorenz et al spin paper last year -
is I think just going to be the start of a long and complicated story.

[ngunn]

On the ellipsoid vs. equipotential I'm happy to be corrected. (I can't pretend I meant something different.) Like you say, we're just at the beginning.

I'm very interested in your idea of albedo-driven winds. For some time I've been trying to shake off habitual assumptions about the importance of gravity in distributing surface materials on Titan. At the Huygens landing site for example there seems to be evidence of fairly recent fluid flow from at least two directions. To me that suggests something like a storm-driven flood, followed up by some gravitational trickling. Of course lighter materials would pay even less heed to gravitational constraint.

Even on high-g Earth with heavy silicate sand there is striking evidence that sand seas don't follow equipotentials. The Taklamakhan in western China is a good example. It is much higher at it's western end. I was lucky enough to be there last year and was amazed to see sand 'mountains' piled high on the Pamir plateau, deposited near the head of the steep valley that leads up there from the desert.

[Juramike]

Looking at channels in the RADAR Swaths, I haven't seen any obvious places where the apparent flow direction is different from the proposed barycentric elevation differential (Fig 3 in Zebker et al.).
[BTW, the actual swath traces and their altimetry data can be seen in the figures in the text. This was really helpful.]

As far as I can tell, the barycentric elevation data seem to fit the equipotential surface (geoid).

Why Xanadu is not filled in baffles me...

-Mike

[ngunn]

Why Xanadu is not filled in baffles me...

Perhaps dune particles have a finite lifetime on Titan, and some areas like Xanadu are simply made of the wrong materials for producing them. I agree with your point in an earlier post that they can't be falling from the sky. It has always been awkward to explain how some areas are swept clean if that were the case.

[Juramike]

The low-latitude longitudinal contrasts (viz, low Xanadu vs the dunes) is harder to explain, as is the orientation
of the dunes themselves. I sometimes wonder about albedo-driven wind (i.e. sea-breeze type circulations -
with effectively katabatic flow away from high albedo regions).

Here is a detail of the dune pattern near the E Shangri-La/W Xanadu boundary using Amazing Dune-O-Vision (gamma-modified, contrast enhanced, negative image) for a portion of the T13 Swath. I highlighted the dune pattern orientation in some areas using yellow lines.


There seems to be a sudden turn southward near Xanadu as the dunes smack into Xanadu. For a sea-breeze circulation, I would have thought that the off-Xanadu circulation would make the dune pattern "confused" or dissappear in a marginal zone around Xanadu. Instead, while the dune orientation changes, it remains very clear right up until hitting Xanadu. (Also indicating that the Xanadu margin is not much of a topographical barrier)

And the dune sand migration seems to bend to the S when it starts to go across Xanadu.

Is this vector change sufficient to divert sand flow around the depression instead of into it? Where does the sand eventually end up?

-Mike