Habitability of tidally locked exoplanets

[SW VandeCarr]

There's been a recent upsurge of interest in evidence for rocky planets orbiting red dwarf stars. Red dwarfs are very common in the the Milky Way and can live much longer than larger stars such as our Sun. The possible existence of several rocky planets in the habitable zone of Gliese 581 has sparked a number of investigations and simulations. My interest is not so much whether this particular system pans out, but in possibility that an Earth like planet will eventually be confirmed.

en.wikipedia.org/wiki/Gliese_581_g

From an SF point of view, Glies 581g is interesting because it could be earthlike while at the same time, being very different. First, it will almost certainly be tidally locked meaning one side faces its star at all times. At first this seems like strong negative, but it's argued that a deep liquid water oceanic planet with an atmosphere at least as dense as Earth could distribute enough heat from the bright side to the dark side to keep the the water from freezing.

I've been thinking about the possible atmospheric circulation on such a planet. Assuming it is tidally locked (and there is no axial tilt) any point on the surface of the bright side would receive a more or a less constant level of light according to its position. It would maximal at the subsolar point and minimal at the perimeter of the bright side. There would be no day-night or seasonal cycles. The subsolar point would be an idealized "thermal warm pole" and its antipode on the dark side would be the "cold pole". Constructing a polar coordinate system on this basis might be more useful than the typical system based on the poles of rotation. I would expect cold surface air at the cold pole to circulate in a spiral manner toward the warm pole, and a return circulation at altitude. In the absence of significant Coriolis forces, I would not expect the of kinds wind belts we see on Earth (trade winds and equatorial easterlies, westerlies, polar easterlies) or the many more seen on Jupiter. The spiral pattern would be caused by the same forces that create a spiral pattern in water draining in a sink. Water cannot flow into the drain flowing in a straight line. Whether the flow is clockwise or counterclockwise would depend on initial conditions and not Coriolis forces. Does this seem to be correct?

 

[mfb]

Sunlight can vary if the orbit is eccentric. This would give global seasons for the whole planet, and the border between light and dark would shift a bit during a year.

I would expect a global atmospheric cirulation roughly along constant longitude: cold air from cold to warm at the bottom, warm air from warm to cold at the top. Could give some circular pattern close to the poles, but I don't see a driving force for an angular momentum. An ocean might follow a similar pattern.

 

[Dotini]

From an SF point of view, Glies 581g is interesting because it could be earthlike while at the same time, being very different. First, it will almost certainly be tidally locked meaning one side faces its star at all times. At first this seems like strong negative, but it's argued that a deep liquid water oceanic planet with an atmosphere at least as dense as Earth could distribute enough heat from the bright side to the dark side to keep the the water from freezing.

Would unusual gravity conditions pertain on an Earth-like planet which was tidally locked to a brown dwarf star? Would the ocean rise up on the side locked to the star, and fall down on the dark side? Would creatures perambulating on the star side feel gravity any differently from those on the dark side?

Respectfully submitted,
Steve

 

[cephron]

@Dotini: I am fairly certain that tides consist of a "pull towards" on the near side and a "pull away" on the far side--gravity and centrifugal force (ie. inertia) gently pulling in opposite directions on opposite sides of the body. So the ocean should rise both at the subsolar pole and the opposite pole, and be lowest at the "equator" of twilight.

@mfb and Vandecarr, I asked a question similar to this a while ago, and someone brought up the possibilities of Hadley cells forming.

The thread:
https://www.physicsforums.com/showthread.php?t=570873

My own speculation on this would be that Hadley cells would be much more likely to form at or near the terminator, where the temperature gradients (should be?) strongest. But perhaps they would form even near the poles, due to vertical gradients?

 

[mfb]

With constant tidal forces, the diameter at the poles would be a bit larger than the diameter at the "equator" (sun/dark border). In equilibrium, the whole crust follows this, just as it does on Earth with its rotation*.

*the point with the largest distance to the center is not Mount Everest, it is Chimborazo close to the equator.

 

[SW VandeCarr]

I would expect a global atmospheric cirulation roughly along constant longitude: cold air from cold to warm at the bottom, warm air from warm to cold at the top. Could give some circular pattern close to the poles, but I don't see a driving force for an angular momentum. An ocean might follow a similar pattern.

Thanks everyone for your suggestions. Here's a simulation of "Eyeball Earth" where a habitable zone 90 x 90 degrees centered on the sub solar point is simulated with the particular known constraints of the Gliese 581g model. My conception is not limited to this model. I would like to have a planet very close to Earth size, gravity and atmosphere, but it can have more CO2 to keep the entire day side and possibly the whole planet ice free. However, a frozen night side is OK if the alternative is a day side that is too hot.

geosci.uchicago.edu/~rtp1/papers/Gliese581gPalletesApJL2011.pdf

mfb:

Regarding wind patterns, I think meridian oriented winds near the terminator (equator, zero latitude) are likely, possibly with a Hadley cell extending across the terminator. However at some point, possibly as low as latitude 20 or 30 on the day side, I think winds would begin to gradually shift rightward or leftward eventually becoming generally circular near sub solar point (lat 90). Wind arrows that converge imply increased wind velocity and some physical factor that forces convergence. I don't see what that physical factor would be, other than a whirlpool effect, in this oceanic idealized model. My idea is that of an inverted whirlpool of rising air (and a lot of rain) around the sub solar point . What do you think?

 

[cephron]

Nice find! Wish I'd seen this earlier.

 

[Czcibor]

Thanks everyone for your suggestions. Here's a simulation of "Eyeball Earth" where a habitable zone 90 x 90 degrees centered on the sub solar point is simulated with the particular known constraints of the Gliese 581g model. My conception is not limited to this model. I would like to have a planet very close to Earth size, gravity and atmosphere, but it can have more CO2 to keep the entire day side and possibly the whole planet ice free. However, a frozen night side is OK if the alternative is a day side that is too hot.

geosci.uchicago.edu/~rtp1/papers/Gliese581gPalletesApJL2011.pdf

In cold version there is one problem. Already in version shown in the paper in figure 2 they assume one atmosphere and temperature on the dark side below 200 K.

In 194.7 K under 1 atm. carbon dioxide sublimate. Given a chance you can get quite quickly with all your carbon dioxide collected from atmosphere and frozen on the dark side, which would further cool the whole planet.

For a hard SF story I'd think about much denser atmosphere (absolutely reasonable for superearth) which should distribute heat and forbade to reach temperature in which carbon dioxide is solid.

 

[mfb]

However at some point, possibly as low as latitude 20 or 30 on the day side, I think winds would begin to gradually shift rightward or leftward eventually becoming generally circular near sub solar point (lat 90). Wind arrows that converge imply increased wind velocity and some physical factor that forces convergence. I don't see what that physical factor would be, other than a whirlpool effect, in this oceanic idealized model. My idea is that of an inverted whirlpool of rising air (and a lot of rain) around the sub solar point . What do you think?

You propose a model with a net angular momentum at each pole. At both poles, there would be some friction on the surface, reducing the total angular momentum of air in that hemisphere. So how do you produce "new" angular momentum?

@Czcibor: CO2 would fall down, and probably heat up during this process, on the cold side. If this is sufficient to keep the atmosphere above the sublimation point, it might work. Ice and snow are bad heat conductors, so temperature loss could be restricted to radiation into space.

 

[Borek]

Hothouse, Brian Aldiss.

 

[Vanadium 50]

There's tidal locking and then there's tidal locking. The one tidally locked planet in the solar system is in a 3:2 resonance, not 1:1.

 

[SW VandeCarr]

You propose a model with a net angular momentum at each pole. At both poles, there would be some friction on the surface, reducing the total angular momentum of air in that hemisphere. So how do you produce "new" angular momentum?

Some Coriolis effects are present with synchronous rotation. The link below discusses orbital periods of one day and 365 days. In the Gliese 581g model the orbital period is 37 Earth days. According to this link (p3), Coriolis forces are still a factor in the slow rotation model. I would expect for period of about 37 days, zonal flows (which normally means latitudinal) would still be a factor close to the sub solar point. This simulation, as well as the previous one indicate flat temperature gradients on the dark side, so the thermal "cold pole" would probably not be significant in terms of wind patterns.

www.princeton.edu/~tmerlis/Merlis_tidally_locked.pdf

 

[mfb]

Ah right, we still have rotation.
An interesting reference. I miss the other side in most plots of the slow rotation scenario somehow.

 

[Czcibor]

@Czcibor: CO2 would fall down, and probably heat up during this process, on the cold side. If this is sufficient to keep the atmosphere above the sublimation point, it might work. Ice and snow are bad heat conductors, so temperature loss could be restricted to radiation into space.

How should it heat up?

 

[mfb]

It gets in an area with higher pressure, and adiabatic compression would heat it.

 

[SW VandeCarr]

Here's a set of simulations specific to Gliese 581g and its 36.5 earthday rotation period.

http://www.google.com/url?sa=t&rct=...rHql_u1a8shLcYashmPnw&bvm=bv.1357700187,d.cGE

(Strange looking url but it seems to work)

Note the figures on page 2. It shows wind patterns (right) the when the estimated synchronous rotation rate of Gliese 581g is taken into account. The warmest regions are not centrally located and the wind patterns are not predominantly longitudinal compared to the simulation (left) which is not adjusted for the 36.5 day rotation rate. (Use the sizing capability to get a better view of the figures).

The temperature ranges indicated are surprisingly narrow and cool, but these may be equilibrium temperatures as opposed to actual surface temperature estimates. For example the planetary equilibrium (black body) temperature for the Earth is about 255 K. In any case, small adjustments in orbital radius, stellar mass/luminosity or atmospheric CO2 levels should be able (if necessary) to allow for a good Earth-like model. I haven't yet read the entire paper so an explanation may be provided.