Pluto Atmospheric Observations: NH Post-Encounter Phase, 1 Aug 2015- TBD Options

[scalbers]

A neat paper by Jonathan Fortney shows this ratio to scale (approximately) with sqrt(Rp/H), with Rp being the planet radius and H the scale height. Both indeed decrease this effect for Pluto.

If we assume the scale height of Pluto's atmosphere is 60km and the aerosols have the same height as the gas, then I was able to get a few numbers in the course of comparing various airmass equations. Earth would be about 39 airmasses in the horizontal and Pluto would be 6.4. These numbers would be doubled when looking at grazing incidence from space as in the NH images. I'd still like to come up with a formula for an isothermal atmosphere (exponential density decrease with height) by integrating the thin shell relationship over height and to compare this with the other formulations in Wikipedia. On the other hand, the isothermal case is within just a few percent of the homogeneous (constant density with height) case.

To check the scale height and see why it is much higher than Earth, we might evaluate this expression for Earth and Pluto:

H = kT/mg

H is scale height
T is temperature (a representative value since this varies with height)
k is Boltzmann's constant
m is molecular mass
g is gravitational acceleration

The Wikipedia link above shows this worked example for Earth:

Taking T = 288.15 K, k = 1.3806488x10-13 J/K, m = 28.9644×1.6605×10−27 kg, and g = 9.80665 m/s2 yields H = 8345m

Roughly speaking, if pluto has .07 Earth's gravity and the same T and similar m we'd get about 120km scale height. If the scale height is 60km, then the temperature would still end up being ~140K. So we can check how much the temperature increases with height over the surface value of 44K.

There are other atmosphere posts in the Near Encounter thread as well (e.g. posts #1238 and #1252).

[nprev]

This topic is for discussion of data concerning Pluto's atmosphere (including exospheric phenomena) received after 1 Aug 2015.

[Bill Harris]

As noted earlier:

That has had me doing back-of-envelope scribbling, too. I dunno.

At the Solstices, one pole is 100% in sunlight (with varying sun elevations) for 1/2-year while the other pole is 100% in stygian darkness for 1/2-year. The Equator, OTOH, varies from 1/2-day sunlight _at_ the Equinoxes to 100% sunset (ie, zero Sun elevation, on the horizon) at the Solstices. For arm-waving simplicity, I'm assuming 90 deg polar inclination instead of the 119.xx deg it is. And add to that the variation of solar intensity due to the orbital eccentricity.

The insolation at any point will be the product of the solar intensity (x) a function of the solar elevation (x) a factor of the Sun's time above the horizon (x) whatever I've not thought of. I suspect that we'd need some sort of Calculus to describe that, with sine/cosine wheels to vary things.

Dang. What a can o'worms. Much easier to crank out purty pictures, but less fun.

--Bill

And, as noted:

"Pluto’s insolation history: Latitudinal variations and effects on atmospheric pressure" by Earle and Binzel in April's edition of Icarus is probably the last word on Pluto's insolation history. They've done all the calculus so we don't have to!
For those without access there are various papers on Arxiv which touch on the subject.

The Earle and Binzel article covers a lot, but there has been a wealth of study done on insolation and the atmosphere of Pluto over the past few years. Google "pluto insolation atmosphere" and start digging. The atmosphere of Pluto, which drives the weathering, erosion, transportation, deposition and induration processes of the surface is clearly solar-powered. I've not give much thought to the variations in insolation on a high axial-inclination planetary-body other than imagining that it is out of the ordinary. And the revelations on the structure of the Pluto atmosphere from the post-encounter occultations will drive more research.

--Bill

[lollipop]

I meant to say that Earle and Binzel had sorted out the insolation, in the sense of watts per square metre at various latitudes, and including the orbital eccentricity, polar orientation etc. Of course the in situ measurements of the atmosphere and the images and composition of the surface will tie up theorists for years.
While the insolation is important, any other heat source will have interesting effects on such cold surfaces and such tenuous atmospheres.

[Bill Harris]

Even though the NH encounter increased our knowledge of Pluto and Charon orders of magnitude and many orders of magnitude, respectively, this knowledge is in the form of snapshots of the system. Along with all of the prior data on Pluto the NH data is much like piles of jigsaw puzzle pieces on the table and we'll spend years seeing how the pieces fit together. Three weeks ago we didn't know what the surface of Charon even looked like but now the geomorph people are supposing terrains and processes. Even though Charon is a small airless world, it does have solids that turn to gases and back to solids, seasonally, which is something that the rocky and ice worlds don't have.

Interesting times, these.

--Bill

[Herobrine]

I've previously "unwrapped" the atmosphere/haze from two different sets of LORRI images from SOC.
The second set were nice because they covered the full disk, but this came with substantially lower resolution.
Since the Sun is lighting the haze unevenly around the limb, trying to merge data from around the disk to get a cleaner, higher-resolution profile of scattered light is not a simple task.
Here, I've taken the two LORRI JPEGs and, for each pixel, mapped the value of the sample (0-255) with (the center of) that pixel's distance from my best estimate of the center of the planet. What that gives me is a plot of the combined scattering gradients of all of the different Sun-haze-sensor geometries present in the LORRI frames.

Here are simple plots of that data for both LORRI images.

The vertical axis is sample value (0-255). The horizontal axis is distance (in pixels) from estimated Pluto center. The darkness is an indication of how many pixels are mapped to that spot. Each data point is rendered as a single pixel of black, but with subpixel precision, so the black pixel is distributed across as many as 4 pixels.
Shown in those plots are the mappings of about 32,000 pixels from each image. Specifically, all pixels that have a distance-from-center of between 105 and ~145.65 pixels.

Here is a combination of both images' data, with lor_0299323899's values scaled up slightly to correct for an apparent, small difference in brightness between the two JPEGs.


Pluto center used for lor_0299323899 (pixels): 412.125, 483.75
Pluto center used for lor_0299323929 (pixels): 337.125, 460
^{(center of image's top-left pixel is 0.5, 0.5)}


Edit (2015-08-10): Remade and replaced the "combined" images more carefully, by merging the original data, rather than combining images of both, and using a more precise subpixel render.

I don't have a solid explanation for some of the repeating patterns seen in some parts of the plots. They are not processing artifacts in the sense that they are present in a simple plot without any processing of the data, but given that they have a horizontal spacing of very close to one pixel of distance and occur very close to the mid-point between integer pixels of distance, it seems likely that they are in some way a result of the distribution of pixel distances near certain angles. That isn't to say that they don't represent the actual scattering gradient, but rather probably that near the vertical and horizontal, pixel center distance clusters, horizontally compressing the data in that plot due to a lack of pixels that fall at distances just below and above it near those angles.

[Gennady Ionov]

To be seen the glow of the atmosphere due to Rayleigh scattering need to raise the sensitivity of the image in 8000 times. Then it will look something like this
(old lovely version) (fixed version, really Rayleigh glow is dimmer then Charonshine)
but we does not seen Charonshine on the LORRI frame

Thus the main optical depth falls on the aerosol.
If optical depth of aerosol in 1000 times bigger then for Rayleigh scattering and exponential scale equal to 20 km we obtain a proper image:

[fredk]

Are you saying that, for the geometry of that image, Charonshine on Pluto is very roughly of the same brightness as atmospheric Rayleigh scattering around the limb of Pluto?

What atmospheric parameters did you use for Pluto? And the intensity of Charonshine depends on knowing the absolute albedos of Charon and Pluto - what did you use?

[Gennady Ionov]

Are you saying that, for the geometry of that image, Charonshine on Pluto is very roughly of the same brightness as atmospheric Rayleigh scattering around the limb of Pluto?

What atmospheric parameters did you use for Pluto? And the intensity of Charonshine depends on knowing the absolute albedos of Charon and Pluto - what did you use?

Yes, Charonshine on Pluto is the same brightness as atmospheric Rayleigh scattering.
I used N2 atmosphere with scale 60 km and surface pressure 1.0 Pa.
Albedo of Charon = 0.375, and albedo of Pluto depends from a surface position. Pluto's albedo just is Pluto map.

[fredk]

Sure, Pluto's albedo depends on position, but how did you choose the absolute albedo corresponding to the Pluto map pixel values 1-255? I don't think anyone has made any claim for their map such as that 255 corresponds to albedo = 1 and 128 to 0.5, eg.

Of course, however you chose that shouldn't affect your conclusion that Charonshine is comparable to Rayleigh.

[Gennady Ionov]

Sure, Pluto's albedo depends on position, but how did you choose the absolute albedo corresponding to the Pluto map pixel values 1-255? I don't think anyone has made any claim for their map such as that 255 corresponds to albedo = 1 and 128 to 0.5, eg.

Of course, however you chose that shouldn't affect your conclusion that Charonshine is comparable to Rayleigh.

The resulting LORRI frames in addition to the absolute surface albedo of Pluto is influenced by many factors: the quantum yield of CCD, telescope aperture, mirrors albedo, etc. Consider all this I was not needed, because everything gives approximately linear contribution. Thus, for each of the simulated image, I introduced two parameters - the color of the background and gain. Varying its manually I achieved similarities with LORRI images. By the way, LORRI frames was pretreated and a value of zero and gain different for different images. Therefore, the correlation of the absolute value of the albedo to a pixel value on the map is not required.

By the way, the faint traces of the atmosphere can be seen even daily images of Pluto, if you know what to look for:

Haze brightness in simulated pre-encounter frame is about 2 steps in 256 colours.
On difference (right picture) we can see, that haze glow almost completely compensate glowing on the source frame.

[fredk]

Varying its manually I achieved similarities with LORRI images.

Normalizing to the LORRI sunlit images and taking into account the different exposure times means you can display the Charonshine as it would have appeared in the image (after scaling 8000x, of course). Without knowing the CCD QE etc, you don't know the absolute intensity of Charonshine. But what about the Rayleigh scattering? Is that calculated absolutely given the atmospheric parameters? There are no other images of Rayleigh scattering for reference, unlike the surface of Pluto. How do you decide how bright the scattering will appear on your image? I don't understand how you can compare the intensity of Charonshine and Rayleigh if you calculate the scattering absolutely but not the Charonshine.

By the way, the faint traces of the atmosphere can be seen even daily images of Pluto, if you know what to look for:

That glow around Pluto is probably just scattered light inside the camera. You can see a similar glow in images taken much farther out before closest approach, but the glow's thickness in those images is much thicker relative to the angular (or pixel) size of Pluto. That strongly suggests camera scattered light. The real atmosphere should always appear the same thickness relative to the size of Pluto.

[Gennady Ionov]

But what about the Rayleigh scattering? Is that calculated absolutely given the atmospheric parameters? There are no other images of Rayleigh scattering for reference, unlike the surface of Pluto. How do you decide how bright the scattering will appear on your image? I don't understand how you can compare the intensity of Charonshine and Rayleigh if you calculate the scattering absolutely but not the Charonshine.

I checked my code and found that really I believe that value 255 corresponds to the albedo of 1.0 (in sense all falling light isotropically scatter in half-space). Charonshine light flow is determined relative the Solar light flow from angular size, phase and geometric albedo. Also, the intensity of the Rayleigh emission is determined with respect to the intensity of emission of the white body which isotropically scatter sunlight. Therefore it is possible to compare the Charonshine and Rayleigh glow.

[Bjorn Jonsson]

That glow around Pluto is probably just scattered light inside the camera. You can see a similar glow in images taken much farther out before closest approach, but the glow's thickness in those images is much thicker relative to the angular (or pixel) size of Pluto. That strongly suggests camera scattered light. The real atmosphere should always appear the same thickness relative to the size of Pluto.

There is definitely scattered light around a bright target like Pluto but I wouldn't completely rule out a very faint layer of aerosols/haze in addition to the scattering. Also one problem with the glow around Pluto in the farther-out images days before closest approach is that it may be caused partially by JPG compression artifacts in addition to scattering. Things really do not become completely clear until we see much higher resolution images of Pluto's limb taken before closest approach (i.e. not at high phase angles). Also let's not forget that a haze layer is visible in images of Triton where the phase angle isn't very high.

[Sherbert]

The general conclusion seems to be that Pluto's "sky" would be "space" during the day, but near sunrise and sunset the "haze" of the atmosphere should be discernible. That would be something to see, two "night skies" in one day.

This being the case would "shooting stars" be visible? That is, is the atmosphere "thick" enough to "burn up" incoming debris? Its not very dense, but it is "tall".

Whether visible or not, not all the atmosphere comes from the surface ices. Icy objects entering the atmosphere and "evaporating" could add significantly to Pluto's atmosphere over time. Its not going to replace 500 tonnes per hour thats for sure, but might help maintain the atmosphere's depth and seed ice crystals, via dust and organics, at higher levels of the atmosphere than those materials being transported from the surface reach. Pluto's equivalent of Noctilucent clouds. One might expect atmospheric dynamics would mix the two, but it is very cold and the evidence is that there are distinct layers within the atmosphere.