Here's part 3 of my stroll through this year's LPSC abstracts that I found particularly interesting (keeping in mind that this is strictly my own judgment):

(1) #1957 and 1269: Magellan's count of large Venusian craters indicated an average surface age of roighly half a billion years -- but it also showed two major surprises. First, the geographical distribution of craters seemed very close to randomly even --, indicating that most of Venus' surface was the same age -- a situation completely unlike that on the Moon or Mars. Second, there was a major shortage of craters that appeared to have been partially "embayed" (covered over) by lava flows. The conclusion drawn by many was that Venus had undergone "catastrophic resurfacing" about half a billion years ago -- a spectacular spasm of planet-wide volcanic eruptions that virtually completely covered over its existing surface and then suddenly and almost completely died away, allowing new craters to start building up on its surface since then unmolested by further volcanic activity, except in a few small regions that do show more recent volcanic activity and fewer craters. Indeed, some models of Venus call for it to have a thick lithosphere that tends to trap mantle heat which builds up under it until it's all suddenly released in a dramatic spasm of planetwide eruptions and lava flows, after which the lithosphere solidifies again and the heat starts building up again underneath in preparation for the next such event. (Strictly on the basis of the crater patterns, though, it would be equally possible for Venus' surface to have stayed continually and dramatically active until half a billion years ago, at which point its surface volcanism and tectonics suddenly and almost totally shut down.)

But there are always dissenters, and in abstract #1957, T.M. Bond reports a series of statistical analyses he did which revealed, "much to our surprise", that there are alternative models that can also explain Venus' even crater distribution and shortage of partially lava-covered craters with a GRADUAL drop in its surface volcanic activity, running over as much as the last 2 billion years. The reason why his simulations canproduce such results for the modern surface, while earlier models of a gradual drop in surface activity couldn't, is that his version assumes not just a gradual drop in the FREQUENCY of Venusian lava flows, but also one in their SIZES -- whereas previous models assumed a slow drop in their frequency, but assumed that recent ones would be as big as the earlier ones. "The reduction in the size of magmatic events with time is critical in maintaining both spatial randomness in the crater population and low numbers of partially destroyed craters. If the flow size is too small early in the model, too many partially destroyed craters are retained; if the flow size is too large late in the model, spatial randomness is not maintained."

Meanwhile, in abstract #1269, Richard Ghail proposes a model of Venusian internal structure that could explain such a dropoff. His analysis of the planet's internal structure calls for Venusian crustal tectonics to be going on right now almost as actively as on Earth -- but for its results to be concealed from surface view by the fact that Venus' high atmospheric temperature actually creates a second asthenosphere in the planet's crust! The crust below 3 km depth is all soft and viscous, so that the hard lithosphere just consists of the topmost layer of mantle, instead of also including the crust as on Earth. This lithosphere's hard plates slide around on top of the planet's main asthenosphere, collide, fold upwards or subduct just as on Earth -- but the second asthenosphere above them separates all this from the planet's thin, hard 3-km "topmost skin" effectively enough that only modest, mild signs of all that crustal tectonics actually show on Venus' surface.

Moreover, Ghail's model shows that before about 2 billion years ago, Venus' greater early internal heat flow kept the mantle lithosphere from existing at all, so that "the crust [which was also thicker then] couples directly to the convecting mantle" and the planet DID have very dramatic surface tectonic and volcanic activity that erased all its craters soon after their formation. Then the intervening mantle lithosphere began to appear, sliding around in pieces underground while allowing the new separate plastic crustal asthenosphere above it -- which was also getting wider and wider -- to shield the planet's topmost solid crust from more and more of the effects of that sliding, producing just the sort of gradual decline in rate of surface change that Bond's calculations show to be possible.

(2) In #1588, R.R. Herrick continues his declaration that a careful inspection of Magellan's images actually shows a lot more craters by be partially embayed by lava flows than had been thought earlier.

(3) In #1763, N.P. Lang and Vicki Hansen propose a new model of the strange Venusian canali (which should please Emily). To wit: earlier melting of a surface basalt crust separates it into a surface layer of granitic rock with a much lower melting point, and a lower layer of basalt residuum with a higher melting point than regular basalt. Then a current of regular basalt lava later trickles along through this sandwich, melting the granitic layer easily to carve a tunnel while not melting into the underlying layer of very mafic basalt residuum. This "piping" erosion explains why the canali often chop straight through local surface hummocks. And when the tunnel is melted to a certain width, the current of hot basalt in its center no longer transfers enough heat to the tunnel's walls to widen it much further -- which explains why the canali stay the same narrow width for astonishingly long distances. Ultimately, if the basalt current finally melts its tunnel wide enough for the roof to fall in, it loses its heat to the air fast enough to solidify and shut down, although that may take a while.

All these theories makes Bruce Campbell's proposed "VISTA" Discovery Venus orbiter -- equipped with a copy of the MARSIS radar capable of piercing 1-2 km down through Venus' rock -- look like a more and more attractive mission to find out just what is going on on this planet.

Finally, I should mention http://www.cosis.net/abstracts/EGU06/06066...6-J-06066-1.pdf ,in which a French team's new computer model of Venus' atmosphere provides "a satisfying super-rotation... with winds of the order of 150 meters/second near the cloud tops". We may hope that this major Venusian puzzle is near solution -- at any rate, the ability of Venus Express' VIRTIS to sensitively track the rate at which cloud features at different altitudes blow along has a good chance of finally nailing down the answer.

Thanks, Bruce, this kinda sets the stage for the current round of Venus Express Science. The crater distribution papers kinda quench the argument that recent or current volcanic activity is providing maintaining gas vents to the atmophere.


Thanks, Bruce, this kinda sets the stage for the current round of Venus Express Science. The crater distribution papers kinda quench the argument that recent or current volcanic activity is providing maintaining gas vents to the atmophere.

One of the oddities I have found no explanation for (other than 'new physik'), is that the canali and chasma of Venus all demonstrate positive Bouguer anomalies, while similar features on Mars always produce negative anomalies. I assume this is tied into the interpretation of the basaltic flows, and any light you have seen on this subject would be appreciated.

Point 1:
The abstract says, in non-native English: "The clouds layers... are not yet included in our computations (clouds layers are fixed for radiative calculations). " I interpret this to say that in the model, the clouds are represented either as a plane or as an isotropic layer of fixed thickness.

Point 2:
The thing that makes this model new is an account of infrared radiative transfer. Was this really lacking from previous models?

I'm not bowled over with confidence if the state of the art is still playing around with whether to include major factors or not. Maybe these authors are the ones to finally nail it, but I am reminded once more of the Pashler and Roberts paper I posted on the planetary sciences Yahoo group: They are cognitive psychologists, but their paper is more about the methodology of science in their field and is relevant, I think, to some extent, other fields like planetary atmospheric dynamics.

The first question in my mind is: could 25 credible models of Venus's atmosphere all explain super-rotation with one leaving out this factor, another leaving out that factor, etc. If so, it's not really science. Pashler and Roberts suggest that a minimum standard of acceptance ought to be that a model do more than just explain the observed phenomenon; in addition, a model should make predictions that would surprise (some of) the field, and be tested against those predictions. I suggest, with no slur of these authors, that it wouldn't be that hard to simply construct a model that demonstrates super-rotation if you get to decide which factors to include. Now if the same model can be used to refine predictions of the weather of Earth, Mars, or Titan, then it is a probable winner. However, the spatial resolution of the model is pretty coarse, and definitely wouldn't refine earthly weather models for want of resolution alone.

So I'm not so sure this matter is close to being resolved even if this team did the best they could given their computational resources.

Visible/Near-IR and Thermal IR radiative transfer in Venus' atmosphere is *HARD* to model. Beside remote sensing from Earth and spacecraft of the atmosphere down to unit optical depth in the haze above the clouds (about 65 km), we have radio-occultation measurements from above the clouds down to the middle atmosphere, and limited measurements of light, infrared, cloud structure, cloud/gas composition in the atmosphere, and atmosphere dynamics. The only really high quality data from most probes is vertical temperature profiles.

A lot of this is variable and where it's most variable is in the clouds, and in the fine details of atmospheric structure and stability which turn out to be very imperfectly measurable from the atmosphere structure data. A tiny change in the rate temperature changes with altitude is all the difference between convecting and stable atmosphere, at times. Pioneer Venus' Small Probes tried to measure the difference in up vs down heat flux in the atmosphere, but the net flux radiometers ended up with a design problem that made interpretation of the data rather severely instrument model dependent and thus uncertain.

In addition, while Soviet probes measured the Visible/Near-IR spectrum of skylight as they descended, and transmission spectra of the deeper atmosphere as they descended, any good thermal infrared spectra and really high spectral resolution Vis/N-IR spectra do not exist.

Finally, molecular physics based models of the spectral behavior of the complex, hot and HIGH PRESSURE <very important... non-linear physics> gasses in the atmosphere are sill only partly able to match what we see in Earth and spacecraft observations of the emissions from the hot sub-cloud atmosphere.

So models of radiative transfer in the atmosphere are included <necessarily> in atmosphereic dymanics models, but the radiative transfer models are more than a bit "iffy" and that has real problems in the energy that drives the models.

another model:

poster from last DPS

http://www.atm.ox.ac.uk/main/research/posters2005/2005cl.pdf (1.5 Mb)