Transcription
Okay, let's get started again. Before I bore you with more honest, fair stuff, we have an interesting question that I don't know the answer to, and hopefully someone in the audience does. Which is: why did the giant planets rotate so quickly? So why is there this 10-hour, 14-hour day? Why does your Earth rotate slowly? Interaction with the moon is that what I heard? Which kind of starts larger stars have more rapid rotation, so it's just conservation of angular momentum. Okay, so didn't everyone hear that during accretion giant planets secrete more mass, smaller size, faster rotation rate. Asteroids rotating 10 hours may become of collisions. Okay, anyway, that sounds pretty good to me, but I don't know the actual answer. Any other general questions?
There was one raised which I didn't discuss. When, say, you're producing the ion density in your model ionosphere, what is the electron density? The first thing you do is you always say the electron density is just the sum of the ion densities, so you don't actually calculate the electron density explicitly for your thermal population. But that's not always true. Sometimes you could have negative ions; you could have negatively charged dust grains. It's a good first-order approximation, but it's just something to keep in mind. Oh sure, yes. I don't know. Does anyone know for certain? No, right? As we discover different exoplanets.
[Music]
[Music]
Okay, so mm-hmm. So that's, as just going back to the slide, that's the current accepted explanation for why we have this trend in latitude at Saturn with electron density. It's because there's some additional loss that peaks at Saturn's equator, and that loss falls off the latitude, and that loss most likely is an influx of water from, say, Enceladus and/or the Rings. So in two ways that have commonly been suggested to bring the predicted electron densities down so that they can agree with the actual observed values are to convert—they both act in the similar way—that's to convert H+, which is long-lived due to its long radiative recombination loss, into a short-lived molecular ion that it then recombines. So the first way you might do that is H+ charge is changing with H2, giving you H2+ and, of course, H2+ reacts to give you CH3+. This reaction, however, only goes if H2 is in the fourth or higher vibrational state. And so the major uncertainty here is that, well, one, we don't really know this reaction rate; this is a good—yes, we're not certain of it. More importantly, we have no idea what the vibrational levels of H2 are in the outer planets, so we can calculate them explicitly. What if the calculations that have done so when they are then used to estimate electron densities, they don't match observations. So we're missing something in the calculations, so we don't really know; we don't have a good constraint on this population, vibrational levels of H2. But the other example is from water or some other external influx where, again, H+ charge changes with the molecule; you get some molecule that has a much faster loss rate than H+ does, and you reduce the electron density. And so these, these both act to reduce electron density, and in addition, they will slightly raise the peak of the electron density in altitude. Other ways we can shift in altitude are some force vertical plasma drifts, electric field, something like that.
So I'm just going to run through a few brief examples. These, these are old model calculations, but they're sort of state-of-the-art in some ways as well. So, for example, this is number density versus altitude for Jupiter. Curves A in each case are the actual Voyager radio occultation electron density profiles. Then curve B is the model that doesn't use an additional loss process for H+; it's a sort of the nominal model prediction, and you can see that it predicts too dense an ionosphere at too low of an altitude in both cases. And then curve C is the best model fit when they've just arbitrarily enhanced the H2 vibrational populations until they've reduced the electron density to the absolute observed value, and then they apply some force vertical drift as well to sort of match up the altitudes. So if you do that, you can get a pretty good match, but the problem is that doesn't give you any predictive capabilities either, because you can't then take these values and apply them to the next observation and get just as good of a match; it's going to be different. Here's a similar example for Saturn. So curve A, once again, it's the observation; curve B is the nominal model. And then the difference here is that curve C on this panel uses an enhanced vibrational distribution of H2 to remove H+ and reduce electron density, and then on the right it uses a water influx; you sort of reproduce the observed electron density profile. Now the uncertainty with water influx, we have—we have some measurements of water in the atmospheres of all the giant planets, and of course there seems to be more water at Saturn, which makes sense because we have Enceladus and the Rings are mostly water ice, but there is water at the other giant planets as well, but we don't have a good constraint on the actual magnitude of the influx and especially the variation in latitude and in time of the influx. So again, this is sort of a hand-wavy fix to the observations, but you can't use it in a predictive capability. Just a brief example. There are a few models that have attempted matches at Uranus and Neptune as well. I just want to point out here, so here's altitude versus electron density in this model; they reproduce these sharp lower electron density layers. And if you remember from Saturn, if you have a vertical shear in the wind, so positive and negative wind as you go up in altitude, and you also have an inclined magnetic field, then that's going to—it's going to lead to the creation of layers of ionization naturally. And so that's what they've done here is just use a vertical wind shear to produce these layers. So likely a vertical wind shear is present at least in all the outer planet ionospheres because we see these layers almost—see them ubiquitously in the measurements. Here's another example of how you can get a vertical wind shear. So this is coupling from below. Now gravity waves from the lower atmosphere are breaking just into the—past the homopause in the thermosphere. As they break, they deposit energy, momentum, and they induce vertical wind shears. And so in these model simulations, they take—this is altitude versus, in this case, H3+ density, and then here number density of electrons, and they sort of figure out which combination of gravity wave parameters would best reproduce the observed profile. So, for example, in the bottom case here for Saturn, the black curve is the observed electron density profile, which has a lot of structure, and then the red curve is the model reproduction of that. You can see that just by sort of tweaking the model so that you have the right kind of vertical wind shears, you can reproduce a lot of the structure that you want to see. Again, we're limited here because you're using the observation to tell you what the vertical wind shear is rather than the other way around, so you can't really use this in a predictive capability. Um, does it allow you to say something about the tides and planetary waves? Maybe I'm not quite expert enough to say definitively. I would say because there are a family of solutions here that would apply; there's not one unique solution that maybe you—you can't stretch that too far, but we do know there are waves in the outer planet atmospheres. So, for example, the Galileo probe, when it—when it descended through Jupiter's atmosphere, it measured definite wave structure, and some of that structure is similar to what you would then apply to drive this sort of profile.
Okay, let's briefly get back to SEDs, Saturn electrostatic discharges. Remember I said they're a broadband radio emission? Here I'm showing a frequency-time spectrogram, so frequency versus time as Cassini passes—well, basically Cassini is stationary, more or less, in this reference frame, and Saturn is rotating. So Saturn rotates, and the storm rotates into and out of view; you see the intensity measured in the contours, and then here's the actual frequency. And instead of seeing a broadband emission, you see these short bursts, and that's simply because the instrument on Cassini is scanning very quickly through frequency space. So instead of measuring all the frequencies at once, you're getting a little bit—a little snippet at a time, but you can draw sort of like a lower boundary to where you don't see any more SEDs; that's this white curve here. And then this shows you what the lower frequency cutoff is for this SED episode. And in this case, the Cassini also imaged the storm clouds, so you know you're measuring the low frequency cutoff, and you also know where the storm is, and so you can figure out which part of the atmosphere is transited by those radio waves, and then you can use that—that's the angle of incidence alpha that I mentioned here—you can use that to very simply derive the peak electron density in that intervening ionosphere. And then eventually you do this enough times, and you'll have a diurnal trend in electron density in peak electron density. Can you define the cloud? So yeah, so the storms at Saturn are sort of unusual; there's something—there's a great storm that happens sort of on a 30-year time scale, every 30 years going back to 18—late 1800s. There's some giant storm somewhere at mid-latitudes, and then these storms in the Cassini era are—SEDs, for some reason, tend to only occur at once; there's only one major storm at a time at Saturn for unknown reasons, and it's usually at a similar latitude, and this latitude in this case is 35 degrees south, planetocentric latitude. It's not—it's not known why, but it's very obvious when you take an image of—in the visible. Here's a cloud top; there's no cloud tops over here. We're also measuring SEDs, so that—that's the radio signature of the lightning from the storm. When those clouds go away, we don't see the SEDs anymore. Does that explain it? Okay, it's a—it's a broadband radio emission from the—from the lightning. The measurement itself, the physical emission mechanism of lightning is—I don't know, right? Yeah. Um, it could be Whistler waves. Yeah, it's generating these—um, yeah, so this—does anyone have another answer? I mean, Whistler waves is a good one, I think, but yeah, exactly. You see these at Earth, and you see also lightning broadband emission at Jupiter and Venus as well, other planets. Good enough. So here, then, using SEDs for two different eras, the Voyager era and the Cassini era, there have been a few different derivations of what is the diurnal variation of peak electron density throughout the Saturn day. So here from Cassini, you see something like a two-to-three order of magnitude diurnal variation in electron density, and then in the Voyager era it's maybe more like a one order of magnitude diurnal variation. So I don't want to spend too much time on this because it's sounding specific, and the actual answer is—or the question is: what is causing this diurnal variation? And the answer is: we don't know. So the models can't reproduce such a strong variation. In fact, if you want to go from an electron density of 10 to the 4 at dawn and then 10 to the 5 at noon, that's only two and a half hours of real time. And so the known ionization sources at these latitudes, which are not in the world—so in other words, the Sun just doesn't—it doesn't give you enough ionization by about an order of magnitude for what you need to go from 10 to the 4 to 10 to the 5 in only 2 hours. So the explanation may be some sort of drastic new ionization source or some significant transport processes, such as maybe a diurnal exchange of protons that flow into the magnetosphere, the pauses here, and during the day, and then come back during the night, during the dawn, or maybe some of these sharp lower atmosphere layers are what's happening. So this isn't the classic photochemical peak of the atmosphere; these sort of layers, maybe SEDs are actually sampling those layers and not the main peak. It's an open question.
Alright, so just to summarize from part one: the main ionization sources are UV and x-ray, solar protons, and of course magnetospheric energetic particles. I only talked about energetic electrons, but you can also have ions and protons and neutrals, any other—other energetic particles. The dominant ion—ionization species produced as H2+, but that's quickly converted to H3+, and the major ions then are some combination of H+ and H3+, and you need to introduce some additional unconstrained chemistry in order to reduce H3+ densities and therefore reduce the electron densities if you want to match the observed observations. Those, those unconstrained chemistries are, of course, what are the populations, vibrational excited H2, and what is the water influx? How does it vary with time and with latitude? There are some remaining unknowns, such as those low-altitude electron density layers; it's probably some sort of vertical wind shear, maybe from gravity waves; there could be other causes. The latitudinal structure—it's consistent with what's expected in terms of a water influx from Saturn's rings and from Enceladus, but it's not definitive yet. And of course, the SEDs—what—how do you get such a strong diurnal variation in electron density related to that? We see strong, powerful lightning at Jupiter, but we've never measured an equivalent SED, so that's an open question: why not? All right, part two then is going to focus more on the coupling between the atmosphere into the ionosphere, and I want to go through that in a couple different ways. First, let's look at what all emissions are present, and we're going to separate those with different categories of emission, and in particular the most common that you consider at the giant planets is sort of the UV aurora versus the IR, or the IR is primarily from H3+. Then we'll go into a few examples of ionosphere-thermosphere and magnetosphere-ionosphere wind coupling. One is this interestingly titled "Saturn ring rain," which is a relatively new phenomenon, and then there's something called the "giant planet energy crisis," because every discipline needs a crisis so you can generate funding. And in particular, the upper atmospheric temperatures that you would predict in the model and if—in theoretical temperatures just from solar heating are far, far below what is observed, so we need some additional sources of heating, and that coupling is—is driven by the ionosphere electrical—electrical conductance—conductivity. So we'll go into that a little bit more detail, even though we saw some of that on Saturn, and I'll very briefly mention some future prospects, such as Juno and JUICE.
All right, so overall emissions: the general definition of aurora is some photo manifestation of interactions between energetic extra-atmospheric electrons, ions, and neutrals with the atmosphere. The extra-atmospheric is the important distinction here because remember if you have a photon that excites—it excites some emission; that's referred to as dayglow or airglow. Nightglow encompasses both dayglow and airglow, but basically what you need is some source of plasma that is then accelerated. So these extra energetic particles that have been excited, it now precipitates into the atmosphere; they collide with the ambient neutrals; they excite, ionize, and eventually giving off a roll—and of course they contribute to the energetic electron, ion, neutral populations. So the suprathermal electron populations that then drive heating of the thermal electron population and also will be responsible for closing the magnetospheric currents in the ionosphere. Here's a few pretty pictures—just—I'm sure you've seen similar pictures before, but just to remind ourselves. Aurora are prevalent throughout the solar system and basically anywhere you have an atmosphere and some extra-atmospheric energetic particles interacting with that atmosphere. So at Earth, for instance, this green line is associated with oxygen 5577 angstrom emission, and then at the giant planets that UV emission. So Jupiter, Saturn, Uranus UV emission is primarily a prompt emission in the H2 Lyman and Werner emission bands. Here, Fran described this sort of world context for Jupiter earlier. We have the main auroral oval, some footprints of different moons—Io, Ganymede, Europa, and Callisto is somewhere right over the main auroral oval, so you don't typically see the footprint, but it's there, and some—some complex polar emission, and typically also a dark region of emission in Jupiter. Then there's also a corresponding—at the other end of the footprint—Acadia; there's some corresponding emission there, civil Rory Acadia. Aurora can be used—you study the emission, and you can learn a lot about the properties of the precipitating particles, the properties of the atmosphere, the participating magnetospheric configuration and topology, such as where—to the words of the magnetic footprints from the moons map to, and in the magnetospheric dynamics as well. That's for ya.
Sorry. Yes. Okay, interesting. I haven't heard that before. So these are solar particles that are then accelerated and come back down. Okay, good. I've heard that before. Excellent. Yeah, this is a good point. Yeah, some auroral experts are very picky about the exact definition because sometimes you'll see—we saw aurora at Mars, and they say, well, no, it's not exactly because it wasn't, you know, driven by a magnetic field or something like that. This definition seems fine with me; some might disagree and say you need to have neutrals, so I'm not actually absolutely certain one way or the other. What is the ionization fraction in the solar photosphere? Okay, 10 to the minus 4 is a similar ionization fraction to many ionospheres, so let's go with that. So also the Sun—this is not to scale. Let's focus a little bit on the aurora at Saturn. So here are a few UV aurora images and then infrared images of aurora, and in general the rule of thumb you—you want to take home with you is that UV spectroscopic analysis gives you the characteristics of the particle precipitation, whereas infrared H2 repulsive gives you ion densities, temperatures, velocities, and also neutral temperatures in the atmosphere. How do you actually go about doing that? Here's one example for UV spectroscopic analysis. So basically you take a color ratio. So, of course, I'm going to use—these terms will come up—up with—where E_m is the mean energy of the precipitating particles, so you can have a Maxwellian distribution, for instance, and then q_precipitating is the energy flux of those particles. Okay, so at Earth, for example, you might use the N2LBH bands; at Jupiter and Saturn and outer planets you would use the H2 Lyman and Werner bands. And basically what you do is you look at different spectral windows across these spectral bands; some of those windows will be absorbed by O2 at Earth or methane at Saturn or Jupiter; some of those windows won't be absorbed. So you can take a ratio between the windows that are absorbed and those that aren't absorbed, and then you can drive the column density of—of methane and there—and then since on Saturn you may recall, if we assume hydrostatic equilibrium, which is a good assumption in the upper atmosphere, then you know something about the distribution and altitude of methane, and so the column density also tells you something about the deposition altitude of these precipitating particles. The deposition altitude, of course, is related to the energy of the particle because the higher—higher the energy, the lower the altitude it's deposited. And then you can also back out the—for example, the H2 column depth. So just by looking at different spectral windows in H2 or N2, say, at Jupiter and Saturn versus Earth, you can drive quite a lot about the properties of the precipitating particles. So this doesn't work all the time, however, because sometimes if the energies are—of the particles are too low, they're not going to go down all the way to the thermosphere or the homopause, and they're not going to be absorbed by methane because methane—remember, as soon as the molecular diffusion takes over, the large mass of methane means that it's essentially removed from the upper atmosphere, so it falls off very quickly with altitude. So the low-energy particles won't be absorbed by methane; you can't use this—this. The very, very high-energy particles are absorbed at much lower altitudes, and they start to have spectral contamination across all the windows, so you don't have an uncontaminated window in a—in a contaminated window, so you can't use this—of this ratio again. So basically you need the energies to be just right. At Earth, that might be something like 0.2 to 20 keV for the energy ranges; at Jupiter and Saturn something more like 10 to 100 keV. And similar techniques can be applied at all kinds of different—for all kinds of different auroral emissions with just slightly different limitations based on what's absorbing and what the emission is and so on. And in order to back out some of these quantities, you usually need comprehensive modeling to go along with the observations. Okay, now let's—let's go through four basic categories of giant planet auroral emissions. The first one is
That the precipitating particles themselves will emit. And, for example, you can generate radio emission as the electrons are accelerated into the atmosphere along magnetic field lines. So this is a cyclotron emission, similar to the discussion of synchrotron emission on Saturday, except these are non-relativistic particles. Typically, here's an example of where in space that mission would originate: so Saturn radius here and X and the in the vertical direction, it's a schematic of Saturn. And then there's a little dark spot here, so it's significantly above the atmosphere but somewhere maybe in the acceleration region or just after the acceleration region. These particles are accelerated and they give you radio emission. Saturn is called SKR, Saturn kilometric radiation; Earth, AKR; and other planets, other KRs. Okay, so it's sort of anywhere you have these energized particles, you'll have the emission. And then this is where they map to in the polar atmosphere of Saturn. So this maps to approximately here, which is it coincides with the UV main emission in the infrared main emission; in other words, the main auroral oval.
So where you have the precipitating particles, they're energized, they emit, and then they're going to the atmosphere and they generate a traditional auroral oval that we think of in the UV. And you can also have x-ray emissions produced from two different processes: bremsstrahlung emission from high-energy precipitating particles that are sky-bird atmosphere; and then also that's a minor effect at Jupiter. It turns out the main effect are the main generation of x-ray emission and Jupiter is highly charged heavy energetic ions. So these ions come in with a heavy ionize, I mean a high ionization, and eventually as they recombine they can give off an x-ray in the atmosphere. So bremsstrahlung is not too dissimilar from a cyclotron emission, except what's happening here is the electron, say, is decelerated by collisions, maybe Coulomb collisions, and that loss of kinetic energy gives you the corresponding emission.
I was hurrying.
Yes, yeah, you infer that this is the physics.
Okay, for the bremsstrahlung. Thank you.
Yeah, it's it's kind of similar. So it's a little bit confusing to distinguish between the mechanisms at work, but these are both—I mean someone might disagree—but I think in general these are relatively minor for the aurora, etc., at least in terms of what people study typically. Though the main, the classic aurora, that is the actual generation of emission from excitation and then emission of a photon generated by precipitating particles, is it's just that it's direct atmospheric excitation. So it's usually a prompt emission, and this is this is what we more normally think of when we say aurora. And it's similar on different planets; again, only the different compositional differences, you'll get different wavelengths. The brightest giant planet emissions are in the UV Lyman-alpha in the and H2O Lyman runner bands, and then they're also visible, like Balmer series. So here is an example at Jupiter and Saturn, different UV aurora configurations, and then similar ones at Saturn. One thing to point out here is that the scale of the emission at Jupiter is about a factor of 20 larger than the scale of emission in Saturn. So Jupiter were very, very bright, something on the order of hundreds of kilo Rayleighs versus more like ten kilo Rayleighs in Saturn. And the benefit of this prompt emission process is that it provides you an instantaneous view of what is the actual precipitation into the atmosphere because the brightness tells you something about the flux and the deposition altitude tells you something about the energy as well as the brightness. And so you can back out a lot of the properties that are happening then in the sphere. And you also know exactly where the coupling is happening because you have narrow regions of acceleration.
There are two other categories that are sort of related, and if if you're more like me and you're not really a world expert, you might bulk them together at some point. Well, let's try to distinguish them for now. So the first is thermal auroral emission. So this is emission, thermal emission produced generally from heating of the upper atmosphere and that when that heating is is generated by ionosphere currents driven by the precipitation driven by the coupling to the magnetosphere. Okay, so that can happen in hydrocarbons that can have an H3+ and and these are also important heat sinks for the upper atmosphere. So, for example, hydrocarbons provide a majority of cooling in the thermosphere. H3+ is highly temperature dependent and will emit very strongly when it's above a certain temperature. It's also a heat sink, and it's the most easily observed. I'll come back to why that is in a minute. Okay, so that's sort of a vague definition of thermal auroral emission. Let's contrast that to ionization aurora. Ionization aurora are distinguished from the atmospheric excitation aurora because now we're actually looking at the emission from the ions that are generated by the precipitation. So, for example, H3+ emission and energetic electron ionization plus. And now we're seeing that emission; that's the ionization aurora. However, H2+ itself has a relatively short chemical lifetime, and so while the ionization aurora maps fairly well to the atmospheric excitation aurora—in other words, the UV and the infrared share similar morphologies—in general, the H3+ aurora is more dominated by the temperatures. So you have ionization aurora where the H3+ densities are important; those those short H3+ lifetimes mean that the overall global structure in the auroral region, say, might be dominant more by the temperature structure than the density structure. So these are very closely related.
So let's go back to protonated molecular hydrogen, H3+. It's a very useful probe of the upper atmospheres of all the giant planets. And in fact, it's expected anywhere you have H2 and some ionization source, you will have an H3+. So giant molecular clouds, exoplanets that aren't too close to their stars, their parent stars, because if they're too close you're going to dissociate H2 and then you can't produce an H3+. As Fran mentioned, the first astronomical spectroscopic detection in the universe was at Jupiter, and that was in 1989. So it's a relatively recent, but it's been tremendously useful diagnostic of the upper atmospheres of the giant planets ever since then. And in particular, that's because there are bright emission lines in the K-band and L-band atmospheric windows, as you can observe from the ground. And on top of that, there is strong methane emission in the L band. And remember, H3+ is generated above the homopause, so it's generated in above the majority of methane, which means that any emission from the interior of the planet is absorbed in this L band, is is predominantly absorbed by the methane. And the net result is you have a bright emitting region of H3+ sitting on top of a dark region of methane. So it's very easy to see. Here's an example of that. This is just a narrowband filter centered near some H3+ emission lines near three or four microns, and you can clearly see the aurora and even some a little bit of structure across the planet.
So as I mentioned, it's highly temperature dependent, and therefore you can use ratios of different lines, emission lines, to back out the temperature of the emitting H3+. Then you know temperature, then you know emission, then you know densities. You can also look at the velocities of the ions. H3+ is also expected to be thermalized to the neutrals, so the temperatures you measure are good proxies for the neutral temperature, which is otherwise fairly difficult to get. And I mentioned it's an important point as well. So coming back to the exoplanets, if you don't have an H3+, so you're too close to your parent star and you lose another important coolant of the upper atmosphere, and you might have some runaway heating processes.
Okay, here's a complicated example of how you might apply H3+ observations to derive magnetospheric properties. So the spectroscopic analysis of H3+ observations. So first we have cartoons of the polar regions of Jupiter and Saturn showing the main auroral ovals and then the bright regions, different regions within the aurora. Okay, and then on the top is a velocity versus planetary radii diagram for Jupiter and for Saturn. What essentially what you might measure along this sort of slit placed across the auroral oval. So the velocity, the dashed curve in the velocity represents rigid corotation with the planet. And then this line here is the actual H3+ emission intensity, Jupiter, then over here at Saturn. And then there's also a line that mostly follows a dashed curve, but you can see that there are some deviations from the dashed curve. That line represents the actual ion velocity in this room. So wherever there's a deviation, and that means there's a sub- either a sub-corotation or a super-corotation, mostly of sub-corotation. So just as an example, remember that equatorward of the main oval we expect there to be rigid corotation in the magnetosphere, so that's this A region, the equator region, and indeed this line follows right on top of the dashed curve. So there's a rigid corotation. Then the the breakdown in corotation is associated with the main auroral oval at Jupiter, so right here you see a strong breakdown or deviation from corotation, and you also see this is region B. Now you'll see very bright emission there, and so you can kind of step your way through the different regions and derive not only, you know, what is happening in the ionosphere, but then you can map that back to what's happening then in the sphere. Though it's it's directly measured from spectroscopic H3+ observations. Yeah, so it's I don't know why it the figures like this, but there are two solid lines that look similar to me; one of them is H3+ emission, yeah, and the other one that more closely follows a dashed curve, that's the actual H3+ velocity, ion velocity. So I mean this this figure, you can spend a lot of time on, and people have. I just want to give a sense that you can map out what's happening in the ionosphere, and you can map that back to the magnetosphere. And just what is one final point. So this curve is the emission, H3+ emission in Saturn. For this case, the main Saturn aurora is not associated with the breakdown in corotation, but you can see a little blip here. This blip in emission corresponds to breakdown in corotation in Saturn. Remember where this line deviates from the dashed curve, and so there is a secondary oval or secondary aurora. Saturn is associated with the breakdown in corotation. But that's not all. H3+ can also be used to make measurements across the disk of the planet, in particular Jupiter. Remember, it's highly temperature dependent, and Jupiter is warmer in the upper atmosphere than the other outer planets. It's also closer; it's also the H3+ densities are higher, so it's a lot easier to measure at Jupiter than in Saturn, Uranus, and Neptune. And so it's just tremendously useful because you can not only map H3+ temperatures and densities across the entire planet, and then you can compare those to your model, and you get a very valuable constraint that combines with the limited radio occultation constraints, which are only electron density profiles at dawn and dusk, and it really helps constraining ionospheric models. But this is, remember also, H3+ temperatures are a fairly good proxy of neutral temperatures in the lower atmosphere.
Yes. Oh, dear. Okay, they developed. Let's go through Saturn ring rain very briefly. What is Saturn ring rain? Saturn ring rain is charged water group particles originated from Saturn's rings falling along magnetic field lines into the atmosphere of Saturn, altering the local photochemistry there and therefore altering the observed H3+ densities. So what we're seeing here, since I'm short on time, I'm going to skip through what the observations are, but basically two H3+ emission lines, the the slit is north-south at Saturn. If you map those emission lines and latitude, you see H3+ densities all sorts of interesting structure with latitude that you wouldn't expect for a solar-produced ionosphere. You expect a very smooth structure in latitude, and in fact these extrema, so these local minima and maxima, are mapped into the conjugate hemisphere along magnetic field lines. So it's a very clear magnetic connection from the atmosphere to Saturn's rings. So this is an example of low and mid-latitude atmosphere and magnetosphere coupling. And in fact, you can use those emissions and H3+ to get an estimate of what the water influxes might be from the rings. So here's an example. These contours are modeled H3+ column densities, and remember there are two sort of unconstrained losses in the atmosphere: what is the population of vibrationally excited H2 and what is the water influx? So these model simulations explore a wide range of each parameter, and then the dashed curve you can barely see here is what the H3+ emissions measured at this particular latitude. So basically, the dashed curve describes a family of solutions, that is combinations of this parameter and this parameter, they would reproduce the observed H3+ density there. So if you do this sort of similar sort of thing for each ring rain latitude, then you can drive a trend in latitude of the expected water influx from the rings. And then you can also take that water influx and map it back to the ring plane. That's what's shown here. So water influx versus equatorial radius. The green curve is the water influx estimated from H3+ observations at the atmosphere, and then just to guide your eye, this is the ring, this is optical depth of ring structure, and the peak in the green is at 1.5 Saturn radii, which is actually predicted to be a radius where you would expect the siphoning of material from Saturn's rings to its its atmosphere base. It's more complicated than this, but basically because this is where the some typical force balances the gravitational force, and so an ion rotating here in the magnetosphere would be free to move up along magnetic field lines with any perturbation.
Okay, let's let's go through this energy crisis. So there's no crisis from the giant planets' point of view; they're just a temperature they expect to be. The crisis is because we can't explain it from the theoretical point of view. So this plot shows exospheric temperature versus distance from the Sun. So Jupiter, Saturn, Uranus, and Neptune; these are what values are actually observed, and then these are the model values if you only consider solar heating as a source. So, in other words, there's a significant shortfall between theory and observations. So hence the crisis. How might we make up that shortfall? What are the heating sources available? Of course, there's solar heating; we've gone through there's particle heating, and particle heating can be broken up sort of into two categories. First of all, there's sort of the the billiard ball of a particle smashing through the atmosphere and imparting some collisional heating; that's actually very minimal, and most of the the heating comes from what's called ionosphere Joule heating, which is the the currents that are driven through the ionosphere and the ion drag heating at the ion at high latitudes that results from these neutrals and ions interacting. Finally, there's dissipation of upward propagating waves from the lower hemisphere that may provide heating. We know the waves are there; the problem here is that for the various analyses of size, for example, gravity waves and acoustic waves, it seems that the amount of heat that you would get from these these gravity wave breaking is not sufficient to explain the temperatures; it's far it's a very insufficient. Instead, let's take a look at the Joule heating available. So here's the ratios of the solar heating at Earth, Jupiter, and Saturn, and then the auroral particle or Joule heating. So at Earth, solar heating is dominant, but at Jupiter and Saturn, you would have quite a bit of heating that we would expect from the Joule heating process. So let's go into that in a little bit more detail, but before that let me just mention that these temperature estimates of the upper atmosphere were based only on a few measurements until now for Saturn. So only recently at Saturn, notice this is the observed temperatures versus latitude for Saturn in the infrared and the red points and the UV and the blue points. Most of these points will notice come from this reference here, Cost me et al. 2013. And so prior to that, we had no really good idea of what the global structure, the global trend in temperature was. Even now we can't say for certain, you know, what the trend is; maybe it's like this, maybe it's more or less constant, but it doesn't matter much. The point here is that you can say with some certainty that at low latitudes the temperature should be somewhere around 450 Kelvin. Okay, so we need to at least get up to 450 Kelvin before we try to explain the 450 Kelvin low-latitude temperature. Let's ask the question of is there any crisis in temperature at high latitudes? So here are temperature versus pressure model simulations of temperature for a series of different auroral particle forcing. Okay, and then these black bars represent measurements, constraints; it's constraints from the UV stellar occultation and from infrared ground-based observations. So if any, there's certainly no problem in reproducing the high-latitude temperatures, and if anything we might start to worry about maybe producing too much, too hot of a high-latitude thermosphere. So instead, what happens when we want to transport all this energy that's in the polar region equatorward? Well, remember first of all the giant planets are fast rotators, so the very strong correlated Coriolis force. So even though we have a strong temperature gradient that sets up an equatorward wind, that equatorward wind is very quickly turned zone by the fast, the Coriolis forces. And in addition, the there's what's called an ion drag fringe effect. So ion drag fringe effect represents the ions moving relative to the neutrals, slowing the ions, I mean the neutrals, and as the neutrals slow they can't maintain the sort of oblate shape that's required at whatever latitude they're at, and that drives an inward collapse of the neutrals. So there's an inward collapse in the pole. This inward collapse pulls in low-latitude particles to fill the void, and that drives a circulation pattern like we see here in this circular circulation pattern. So we're pulling in the low-altitude neutrals from low latitudes, and we're depositing at high altitude neutrals from from polar latitudes. That in fact cools the lower, the latitude, the atmosphere, rather, the thermosphere, rather than heating it, cools it. And in fact, if you add more auroral energy here and drive stronger currents, you have a stronger ion drag, and this accentuates the effect, and so you get even more cooling at low latitudes, and it's a runaway effect. But so we can't explain the temperatures; that's why it's still a crisis. These simulations were for steady-state. What if, you know, the reality is that the auroral forcing is very strong with with time and with location across the auroral oval, and so what if you have a more dynamic simulation, can you drive some energy equatorward? In such a case, so here we can ignore these panels for now. Basically, we have temperature profiles versus latitude and altitude at Jupiter. Now you have a high-temperature polar thermosphere and a low temperature, far below the observed values at the low latitudes. But then this is sort of the nominal case. Then we have two cases where there are many auroral compression events that that drive some waves to the thermosphere, and then these show temperature differentials, so the before case minus the after case, and you can't really read the scales here, but they're small. So you you see that you give some temperature increases on the order of maybe 10 or 20 Kelvin at low latitudes. This is only 5 or 10% of what you need to explain the observations, but you can imagine that it's this sort of, if you can continually heat the low latitudes a little bit at a time, eventually you can heat them to the observed values, maybe.
Okay, I'm a little short on time, so I'm going to skip a little bit here and just mention that at high latitudes, not only is there an exchange of particles from the magnetosphere coming into the atmosphere, the ionosphere can provide particles to the magnetosphere as well. That doesn't appear to be as dominant at Jupiter and Saturn because we have Io and Enceladus which provide massive amounts of ion of plasma in their magnetospheres, but it might be more important at Uranus, for example. In addition to the exchange of particles, we have an exchange of momentum and energy along magnetic field lines. So if you remember Fran discussing on Saturday, Io generates or emits all these neutrals into a neutral torus; they are quickly ionized as the original neutral there at the corotation velocity. When they're ionized, that's picked up by the magnetic field, and they need to be accelerated to corotation. And in addition, they're moving radially outward, which would mean you would expect those velocities to go down further, but instead that sets up a current loop; it's closed through the magnet at the ionosphere, and and essentially the the momentum that's required to maintain corotation in the magnetosphere, it's extracted from the ionosphere through the closure of this current loop. That holds up to a point until the load on the atmosphere is just too strong, and then you have breakdown in corotation.
At some further out radial distance, okay. So what these these current loop, it's a closed nanosphere, and it's driving primarily Pettersen currents. So just a brief review: the the conductivity tensor and its longitudinal and latitudinal terms; in other words, a Pettersen term and a Hall term. And we'll go into what those are briefly. But then the Joule heating is proportional to the conductivity, dr. field, and then u being the relative ion and neutral velocity.
Now, if we set up these currents and we want to derive the amount of Joule heating present from the currents, if we need to know what the conductivity is. This is just here for reference; I'm not going to go through all the terms in the equation, but they're all they're all pretty straightforward, and we also saw them on Saturday. Just a reminder that Pettersen currents are perpendicular to the magnetic field in the direction of the electric field in the atmosphere, and then Hall currents are perpendicular to both electric and magnetic fields. The conductivity is given here, and then the conductance is the height-integrated conductivity. A typical unit you will see for conductance is the mho, because con- the conductance is sort of the inverse of the resistance; it's ohm spelled backwards. But you also might come across Siemens, and that's the SI unit. So just so you're not confused, one mho is one Siemens.
How much time? Mm-hmm. Okay. Um, let's just let's just briefly say then that you can estimate conductances at Jupiter and Saturn and whatever planet, and you can make a simulation with some mean energies and energy fluxes. And what you find is that conductances of Jupiter are much smaller than conductances at Saturn. So does anyone have an idea why that might be the case? So we have more ionization in Jupiter, a stronger mean energy precipitating flux, more energetic particles, and yet a significantly lower conductance. So the conductivity is associated with the mobility of the charges perpendicular to the magnetic field. So in essence, the magnetic field, the strength of the magnetic field is inhibiting the mobility. So Jupiter's magnetic field is much much stronger, and so you're reducing the conductance sort of by the 1 over B there, and that's essentially what this slide says. So I can skip that. Just remember that conductivity depends on the charge, on the magnetic field, and so on. And in fact, you can see where it would peak in altitude. So there's a conducting layer and a correspondent Pettersen layer and Hall layer, and that peak is where the gyro frequency corresponds to the ion-neutral collision frequency. And so even if you had a perfectly constant density with altitude, you would expect to see conductivity peak right around that layer. You can use electron density profiles along with some estimate of what the atmosphere and the ion fractions are, using modeling, say, to actually make estimates of conductivity at the giant planet.
So here's Jupiter's electron density profile vs. altitude, and then the corresponding estimated conductivity and conductance is that you might expect. So again, the important thing to note is that really what matters is what is the density in the conducting layer. If you don't have the density in the conducting layer, then you don't have conductivity there. So you need to make it up or have some other way to estimate it. Just so, for example, the red curve is one of the largest electron densities here; there's a small electron density here, and the corresponding conductivity is sort of middle-of-the-pack because in the conducting layer, as a small density. Future prospects: we already heard about Juno. I just want to emphasize that it's going to be exciting that Juno will fly in the auroral acceleration region, or at least near in the polar acceleration region. So you can measure for the first time, not only the energy of the precipitating particles, but you can see what sort of auroral emissions they're generating at the same time. Okay, and I'll just leave the summary up then. Any other questions? [Applause] Typically optical thing? Yes, I see resolution and very likely, and and that's because, and on occasion, you can actually get a limb profile of the emission, and then you see, you know, clearly the altitude, no structure, and you can sort of compare that to: does it agree with what you predicted when you couldn't see the limb profile?