Session 2 Discusisons

SESSION 2

DISCUSSION AFTER Norm Sleep's TALK

Alain Leger: What would you expect if the Earths diameter were twice as large as it is?

Norm Sleep: The convective heat flow scales very weakly with radius, maybe even more weakly if plate tectonics is included. However, the radioactivity that needs to be gotten out per unit area scales linearly with the radius, so a larger planet will stay hotter in the interior much longer than the Earth did. It will have a thinner crust, the crust will penetrate downward, so we would still expect tectonic activity. We should also expect to get the analog of continents to some extent because there would still be hydrous rocks to melt. The danger is that the larger planet will accrete too much air, and wont be able to get rid of its hydrogen. So it could end up with an ocean of H, even a few bars, so it will retain its reducing atmosphere very long and that doesnt help the Pale Blue Dot. If its too big, wed end up getting Neptune. The tectonic part of the story is subtle. If the planet cant get rid of the reducing gases early on, there would be problems. The mantle of the Earth is oxidized. To get the ferric iron in the mantle requires an ocean of water to be oxidized, to get the ferrous iron oxidized maybe 10 oceans.

Alain Leger: Where would you put the limit on radius?

Norm Sleep: I cant put a limit this way; it would be set astrophysicallyfrom accreting gas.

Alain Leger: By gas you mean hydrogen?

Norm Sleep: Solar nebula gas. Therell be a limit to the Blue Dot, where hydrogen does not escape. Say a temperate planet, but one where there is too much hydrogen.

Carol Stoker: I was surprised to hear you say that if the Earth had about 1/2 its present amount of water, it would mostly be buried?

Norm Sleep: Yes, it would mostly be buried. Wed still have lakes dynamically maintained. The tectonics would still be producing continents. Thered still be a hydrological cycle, but only small bodies of water would exist, photosynthesis would be difficult, and, though this might still be a Blue Dot, the conditions for life would be far less attractive.

Anonymous: Planetary density is an important parameter, especially for convection. So, even for a terrestrial planet, Earth, if the core were twice as large the convection would be much larger.

Norm Sleep: Right, if had a larger core, convection would be much stronger, greater dominance of plumes. But also there would be less radioactivity in the total planet. So youd have a situation more like Mercury, where the internal heat would get dissipated more quickly.

Same person : So, if you change a parameter like density, plumes would form all over the planet and it would be very difficult to maintain a crust for very long. Correct?

Norm Sleep: No, because the limit is the total amount of heat available. The vigor of convection only scales as gravity**1/3. So no real difference between the Moon and the Earth in this regard.

Same person: Another aspect of planetary convection would be that, as the core cools, its radius increases, by crystallization. When a critical radius is reached, the dynamo is shut down. Right?

Norm Sleep: At some stage, but when you do the scaling, you see to first order that that is independent of planetary radius.

Same person: But when you do that, you no longer have a B field to shield the atmosphere from being sputtered away. So, thats the end of life as we know it, right?

Norm Sleep: Well, Venus has survived without a magnetic field. So, my guess is: if the Earths magnetic field were done away with, the people who sell compasses would be out of business, as would the folks who work on the Aurora Borealis. But I dont think the Earth would rapidly lose its atmosphere.

Alan Boss: You mentioned that it didnt look as if Mars had plate tectonics, at least not for a long time. Could you comment on recent results from Mars Global Surveyor that show magnetic striping on the surface, which has been interpreted as evidence for plate tectonics? If you believe it, what does this tell us about the age that plate tectonics ceased on Mars?

Norm Sleep: This would on be the older part of the planets surface. So it would tell us that plate tectonics with a relatively thick crust, existed maybe 4 billion years ago. And this is the one process we know of that generates large linear magnetic stripes on planetary surfaces. The north-south boundary looks like an analog of the margin on the East Coast of the US that formed when the continent split up and a mid-oceanic ridge formed and spread out. So, in the status of what we understand about plate tectonics, this would be the kind of process. There may be other processes that we don't know about, but at least in the case of plate tectonics, we can do the analysis, make predictions, and see if it fits.

Pat Cassen: Carol asked earlier about the case where there wasnt as much water as the Earth currently has. What about the other extreme? What if there are no exposed land masses and the Earth is entirely covered with water?

Norm Sleep: The typical burial then would be burial on sea floors that would then get subducted. And the oxygen would react with basalt, the carbon would end up on the seafloor and get subducted, so youd get rid of both of them at the same time. Then it would be difficult to build up a large surface of oxygen that could, in turn, build up an atmosphere.

Anonymous : What do you think about the importance of continental growth? Because, according to our calculations, it would have a main influence on the habitable zone. If the area of the continents is small, say in the past, then its hard to get the CO2 out of the atmosphere. But if continents will grow larger in the future, say at their present rate, then CO2 will go quickly from the atmosphere to the solid Earth and the planet will cease to be habitable.

Norm Sleep: There is another major sink for CO2, thats the oceanic crust. Now this oceanic basalt will react with sea water that contains CO2 dissolved in it and will form carbonates. At the present time, this process doesnt occur strongly because the continents are getting the first crack at the CO2. But if the continents go away, then the ocean crust becomes hard to overwhelm. So if the continental cycle could be turned off completely, leaving just the oceanic crust to do the job, the CO2 in the atmosphere might increase by up to a factor of 10, but there would still be a buffer. The Urey type buffer would still operate. If there is ejecta on the early Earth, thats a very efficient sink of CO2 and we could end up with an ice-covered planet, early on. If we just have small impact bodies opening lanes in Antarctica, where the CO2 in the air is strongly out of equilibrium with the water, then the atmosphere and the ocean will stay in more or less dynamic equilibrium. So, for an ice covered planet, where this occurred globally, you would have to take this into account.

DISCUSSION AFTER Kevin Zahnle's TALK.

Anonymous : Considering large impacts late in a planets history, what would be the effect on Norm Sleeps larger-than-Earth planet? Can a giant impact turn such a planet into an Earth-like body? Conversely, what would have happened if Orfeus struck Venus, instead of the Earth? Would Venus be more Earthlike today than it is?

Kevin Zahnle: I dont have a good quick answer. What is intriguing about a late, large impact on Earth, is that it places this massive injection of metallic iron into the system well after the main accretion. So the beneficial aftermath of this event is not as likely to be undone by later impacts. Life that originates after the giant impact is more likely to survive, than life that originates before. Note that Norm did not say that larger planets would not support life, just that the transition to an oxygen-rich atmosphere would be delayed.

Anonymous: Both Kevin and Norm seemed to take the position that processes that happen to the Earth create a surface environment for the orgin of life. But its also true that life requires a great deal of energy from the Sun, and the resulting chemistry changes the atmosphere and affects the surface. How important is the fact that life exists, in controlling the environment we have now and how would it change if that werent the case?

Kevin Zahnle: The discussion about methane as a greenhouse gas implies a fully developed biosphere, because only a biosphere can supply that methane at the rate at which its being destroyed photochemically. So there is an example of what you are talking about, namely, where the biosphere is essential for establishing and maintaining the environment.

Anonymous : Is the relative magnitude of the methane peak which you show on your historygram of the Early earth an estimate?

Kevin Zahnle: Not even that, its just freehand! Its drawn bigger than the nitrogen peak, and that was intended. Im thinking it would be tens of bars.

Anonymous: Thats interesting because to get that might require embracing the Stevenson model: any impactors being incorporated into the mantle would incorporate their metal as an emulsion.

Kevin Zahnle: Yes, its tough to avoid this emulsion. The core of the impactor is much denser than the rock through which its plowing, so Rayleigh-Taylor instabilities cause the iron to be stirred into the silicate to make an emulsion.which I think is Stevensons idea.

Anonymous : OK, but we have a time scale given by the hafnium-tungsten isotope systematics, which suggest very rapid core closure.

Kevin Zahnle: We have 100 M years on Earth, based on this. My impression is that hafnium-tungsten only gives for Earth a lower limit on the length of time after formation at which the impact could occur. And there are other numbers..the xenon closure age of Earths atmosphere, and mantle..the lead-lead closure age of the core. They all give roughly the same number, namely, about 100 M years after the formation of the solar system.

Dave DesMarais: An important question for this meeting comes from this idea that the big methane pulse will produce an abiotic Earth with that sort of methane background, against which we would have to look for evidence of life. So how could we get a handle on the abundance of methane in the absence of a biosphere?

Kevin Zahnle: My guess is that the methane would be short-lived. At the rate at which Ly alpha photons hit Earth, the methane would be destroyed.

Robert Chatfield: I think I follow the logic that leads to a catastrophic model for evolution of early Earth. But are you saying that we understand the high mass end of the impactor distribution well enough to get to your conclusions? Or are you just arriving at a purely mathematical solution, based on an extrapolation of a not very well known distribution?

Kevin Zahnle: For any impact smaller than those that gave the South Pole Aitken basin on the Moon or the Hellas basin on Mars, the distribution is well known. Those impactors follow the power law. That pins down the distribution up to an order of magnitude smaller than that required to evaporate the Earths oceans. You could then arbitrarily terminate this power law distribution, after following it for some 20-30 orders of magnitude, and at some point it does shut off. But the origin of this power law is thought to be a collisonal cascade, where bodies of a given size are formed by the collision of larger bodies. So the only real limit, where it breaks down, would be when you get to the very largest object in the distribution. OK, what would that object be? I would argue that it would be something like a planet in size. But you are right, there is an extrapolation of an order of magnitude required at the high end for my logic to be correct.

Norm Sleep: Another way to view the statistics.typically 15 objects larger than the one that made the Aitken basin on the Moon, about a 200 km sized object, would hit the Earth just because its larger than the Moon and has a higher gravitational cross-section. And, as Kevin said, since we only have the distribution confirmed for smaller objects, we cant say much more than this, but clearly we are dealing with the statistics of small numbers.

Jack Farmer: How does the scenario involving drawdown of CO2 by the reactive ejecta play against the mechanismsvulcanism, higher heat flowthat tend to replenish the atmospheric CO2?

Kevin Zahnle: To a first approximation the rate of outgassing from the mid-oceanic ridges will be equally enhanced compared to the rate of subduction of the oceanic basalts and whatever ejecta is sitting on them. So to first approximation, running both cycles faster doesnt change the system. To go into more detail, wed need to consider second order effects to see how the CO2 would move away from todays value. Weve made an assumptionweve looked at the present timescale for cycling the crustal carbonate through the mantle, about 3 billion years with todays conditions. In the past that timescale was of order 300 M years, very short, so it must have had an effect. But its not short enough that you would expect to really have steady state. We really should do a time dependent model, then wed probably see CO2 walking all over the place.

Anonymous Is it still thought that Uranuss axis tilt is the result of an impact? What would be the mass?

Kevin Zahnle: I think so. People usually quote a few Earth masses for that collision.

DISCUSSION AFTER Jim Kasting's TALK

Alain Leger: Do you still need a methane greenhouse if you have the CO2 clouds for early Earth?

Jim Kasting: Not for early Earth. Even CO2 will work. CO2 clouds dont tend to form on early Earth. Might form them if get into global glaciation on early Earth, but the Forget and Pierre Humbert (??) paper showed that that was not a problem. However, where you may need methane is on early Mars. We are working with a GCM (Haberles) to see if putting CO2 clouds into a realistic Mars model will be sufficient, or if you need additional greenhouse gases there as well.

Alain Leger: OK, so we dont know if there was a lot of methane on the early Earthits an open question?

Jim Kasting: Its an open question, but methanogenic bacteria are evolutionarily ancient by just about anyones analysis. Theres other evidence that suggest methanogens were a very important part of the ecosystem back then. Dave DesMarais talk will probably get into this.

Kevin Zahnle: Karhou and Holland (???) had a paper in which they find a carbon isotope excursion in carbonates between 2.2 and 2.06 B years ago. I had always thought that the glaciation was coincident with the rise in oxygen, but the timing doesnt seem to work. Is there a way around this?

Jim Kasting: Yes, be suspicious of the dating on all of these things! There are some old papers by Rosco where he points out there are areas where you see unoxidized sediments, followed by glacial deposits, followed by oxidized sediments. Thats the type of dating you should believe, you dont have to rely on dates derived by some other method, youre looking at a record of what actually happened.

Rob Rye: Getting back to methane..our group has provided data on isotopically light carbon from this time period. This is not only an indication of methanogens, producing methane, but also of methane-using organisms that fixed the methane, called methanotropes. Finding those organisms in very shallow water, which seems to be the case, would indicate that there was a substantial amount of methane very near the surface, if not actually in the atmosphere. An amount of methane within an order of magnitude or so of what Jims models suggest.

Jim Kasting: Back in the Archean there is data for lots of light kerogens, so thats consistent with this whole methanogenesis story.

Nick Woolf: Your first slide seemed to show a mechanism for an inorganic Gaia, and also seems to support the notion (Lovelocks statement) that Gaia prefers cool, moist conditions. Could you comment?

Jim Kasting: The carbonate-silicate cycle is modulated by biology, but not dramatically so. Thinking back to Lovelocks Gaia theory, motivated by the faint young Sun problem, where Sun was fainter, CO2 higher, organisms evolved that buried organic carbon and drew down CO2 in such a way as to compenste for the increase in solar luminosity. Well, its probably not the organic carbon cycle thats most important in that scenario, because the organic carbon cycle is buffered by redox control. The big strong feedback in that cycle involves oxygen. Whereas, the reason the carbonate-silicate cycle is most important is that there is no redox control, so that cycle is free to act as a buffer on the climate system.

Conrad. ??: Concerning the slide which shows the habitable zone. In our calculations, we find that even the width of the habitable zone is strongly time dependent, much wider in the past. Venus was never in it, but an Earth-sized planet at the orbit of Mars would have been within the zone, leaving it only about 1/2 billion years ago. Eventually the habitable zone width will diminish until it vanishes in about 1.4 billion years.

Jim Kasting: In our calculations the width of the zone is constant, but the zone migrates outward. Its the continuous habitable zone that eventually goes away.

Conrad.??: No, I mean the habitable zone. And its the geodynamics of the planets themselves that causes this effect.

D. Williams: The solar flux falls off as 1/r**2, so the H.Z. grows.

Conrad??: Yes, for a geostatic model. So if the Earth and the carbonate cycle would have always been the same as today.

Jim Kasting: Youre saying that at some point in time the carbonate-silicate cycle slows down so that you no longer cycle CO2?

Conrad ???: Yes, and the continental era grows and this drives the habitable zone to zero.

Jim Kasting: Yes, this model assumes that there is an active method of recycling carbonates. Now Norm Sleep just told us that plate tectonics may shut down in 1/2 to 1 billion years. But this doesnt necessarily mean that carbon metamorphism will shut down..there are other mechanisms, vulcanism for example. Remember that the current cycle recycles all the carbon in maybe 1/2 million years, so very short geologically. So even if slow down by factor of 10 or 20 or so, that negative feedback still works. The danger is that you might get stuck for long periods of time in the global glaciation periods, which evidentally can get triggered by certain kinds of events.

Dave Blake: Can you say a few words about the origin of the ocean, in view of the fact that the D/H ratio from comets looks a little too high and there was a late moon-forming impact, so where might the water have come from?

Jim Kasting: There are essentially three sources: from planetesimals, from asteroids, and finally from comets. These later in the heavy bombardment period. For comets the D/H ratio, as you point out, is too high, at least for three comets in which it was measured. Theres an equally serious problem for asteroids, as pointed out by Toby Owen. If the noble gases come in at the same time as the water, then get 20 times too much Xenon. So, I dont know the answer to the puzzle.

Jeff Cuzzi: Can you get around the noble gas problem if the water is accreted further out, as a solid, maybe adsorbed onto the asteroids?

Jim Kasting: Well, my understanding of water in meteorites is that it was originally condensed out as ice, then melt it, then get liquid water by heating from radioactives, then hydrate the silicates. So a lot of the water is hydrated silicates. So cant form directly from gas phase reactions. So needed to come from a region that was originally cold. And problem is, in carbonaceous chondrites, there is a lot of Xenon relative to Argon and Krypton. So if those are indicators of the water, then bringing it in this way leads to an overabundance of xenon by a factor of 20, compared to what we see. And geologists have looked very hard for the missing xenon, but havent found it.

Kevin Zahnle: Can estimate how much Earth accreted from Iodine isotope ratioand those estimates say that Xenon is then overabundant by a factor of 100. So the Xenon had to escape somehow, and the factor of 20 problem you mention is subsumed by this factor of 100!

Jim Kasting: Kevin, youve thought a lot about this, what do you suppose the answer is?

Kevin Zahnle: I think the water comes from asteroids. The Xenon issue is not a show stopper. Have to lose 99% of Earths radiogenic Xenon, so why not 99% of its non-radiogenic Xenon as well?

DISCUSSION AFTER Wes Traub's TALK

Alain Leger: Without life, what is the origin of the ozone feature?

Wes Traub: It exists mostly high up in the atmosphere, where the reaction rate is lower, and it comes from the little bit of oxygen that is there.

Alain Leger: OK, but where does the little bit of oxygen come from? Is that abiotic?

Wes Traub: That was in the Margolis and Lovelock model, and I cant answer where it came from in their model. It does look consistent with the amount of oxygen we might have expected 2 or so Billion years ago. Water vapor in the atmosphere can produce some ozone.

Jim Kasting: But you dont get enough oxygen abiotically to produce a strong enough ozone signal to detect.

Wes Traub: Yes, as I said, there is an issue with self-consistency of these models, though the spectra are correct for the conditions given. On Mars there is a small amount of oxygen, hence ozone, but not nearly at the level in these models.

Alan Boss: Realistic planets will have a composite blackbody due to the latitudinal dependence of the surface temperature. So how hard will it be to deconvolve the range of blackbodies vs. vibrational-rotational spectrum of water, for example?

Wes Traub: These models were constructed by building up a model where we took account, using a crude grid, of the different temperatures on the Earth. Then we compared this with a single blackbody temperature model, at something like 45 degrees latitude, and using a characteristic airmass at a representative latidtude. The two models give results that are very close, so the single T model is fine for these purposes. Its surprising, but the single T calculation works fine.

Lou Allamandola: Do you include scattering and ice grains in your models, or just use a pure gas phase calculation, for CO2, for example?

Wes Traub: We use pure absorption and emission, no scattering, so no solids. These effects need to be checked. We can introduce effects of solids by putting in opacities that would be characteristic of solids. There is an interesting question about whether or not molecules really behave with the van Vleck-Weisskopf profile, that arises due to collisions, out to many times the central frequency.

Lou Allamandola: My second question is this: I noticed in many, maybe all of your spectra, that there was a shart drop or break in the spectra around 1300 cm**-1. Is this due to the methane or the end of the water band? And if caused by methane, could this be a good way to detect methane in the spectrum of a distant planet, even at low resolution?

Wes Traub: This is a composite and we need to work harder to separate the components and see how it could be used. Our resolution of R ~ 1000 is a little too high to get a good feel for how a low resolution spectrum might appear, so we need to do more work on that. And, yes, we could pull this methane feature out of the spectrum, so we should look more at that.

Bill Borucki: You would expect that most planets with water in the atmosphere would have clouds, as well. Do your calculations include clouds? If not, how do you think it would change your results to include the effects of clouds?

Wes Traub: The way we have included clouds so far is to put in a source of opacity in the middle of the atmosphere, essentially a solid surface. What we need to do is to include the broad band features of water; and then the ice spectrum, as we saw for example in Mars. Now we dont include multiple internal scattering in the clouds or from cloud to cloud, because it doesnt really make a difference for the spectra that were calculating.

Bill Borucki: If you have clouds, will you see the features as clearly as they appear in your calculated spectra, or will clouds make it more difficult to see them?

Wes Traub: Terrestrial clouds are pretty opaque, you cant really see through them. So, for example, if 50% of the surface is obscured by clouds, then you will see essentially a composite of two spectra, one with features and the other the black body emission from the clouds.

Anonymous: I recall seeing an ozone feature in the Mariner infrared spectra of Mars, at least in some of the spectra. I dont see that in your work. But if an ozone feature is present in Mars spectra, then we certainly need to rethink the notion that ozone would be a strong indication of a biota.

Wes Traub: I did not see ozone in these spectra. Now you can see ozone at the poles on Mars, in emission, out of thermal equilibrium. We have seen this. Its high in the atmosphere, an airglow effect, and very faint compared to the planets emission. You need to work hard to see it.

Anonymous: Can you clarify what T and p profiles you were using, especially in the prebiotic case where you had 1 bar of CO2?

Wes Traub: A standard Earth atmosphere, but modified in the stratosphere to account for the removal of ozone from the stratosphere. We estimated that. The tropospheric profiles are the same as for todays Earth. Its not self-consistent, as Ive said.

DISCUSSION AFTER Imke de Pater's TALK

Alain Leger: Do you have a recommendation to make?

Imke de Pater: I think we should seriously consider what we can learn from observations at radio wavelengths, particularly with the new, large arrays that will be coming on line in the future. Also, detection of non-thermal emission from extra-solar planets should be considered for what it can tell us about these planets.

Jim Kasting: Did I understand correctly that, though we can do heterodyne spectroscopy on the planets in our solar system, thats not feasible for extra-solar planets, even with the new arrays that are planned?

Imke de Pater: Yes, but I dont really know what IR sensitivities are. But detector technology is always improving; plus we could use very long integrations. Maybe combining arrays can help, too. But I think that improvements in detectors will be the key.

Ted Kostiuk: Now at longer wavelengths, beam dilution will really cut the signal down, so we really need to concentrate on high resolution, almost certainly from interferometry, in order to get the sensitivity we would need to make detections at the radio wavelengths. And this is being done, with VLBI and Space VLBI. But sensitivity is an issue, of course.

Anonymous: Dont you still have the problem of separating the planetary radiation from the stellar radiation?

Imke de Pa