APOD: Dark Craters and Bright Spots on... (2015 Feb 18)

[Chris Peterson]

I was addressing Nit's definition of pseudoscience. And generally, the scientific method - as it is taught - is something that can be done by individual scientists. Testable results with a hypothesis that can be shown to be true or false is the key.

Absolutely, good science can be done by individuals. But if they do not subject that work to peer review, it ceases to be good science. For the most part, the easy stuff has been done. These days, new results are seldom simple enough or obvious enough that anybody can draw reasonable conclusions from them. Review by specialists is essential. Otherwise, they will not (and should not) be taken seriously. And it is worth noting that one of the principle hallmarks of the pseudoscientist is lack of peer review and self-publishing.

I hope you realize, however, that there is a lot of subjectivity built in to the peer-review process, as opposed to the objectivity of experimental results.

Both are required.

[Chris Peterson]

The mystery link: http://space.io9.com/just-what-are-thos ... 1684160250 claims the spots are 9% brighter than the surrounding areas.

I like this idea from this article posted by Alter-Ego:
Ceres could have a rock core, an ice mantle, and a thin crust of debris and dust. Meteor impacts could burst through the crust, exposing ice below

Certainly possible, but given how dark these light areas apparently are, they would be very old impacts (ice tends to be white). It takes millions of years for space weathering to reduce ice to the color of coal.

[Nitpicker]

The mystery link: http://space.io9.com/just-what-are-thos ... 1684160250 claims the spots are 9% brighter than the surrounding areas.

I like this idea from this article posted by Alter-Ego:
Ceres could have a rock core, an ice mantle, and a thin crust of debris and dust. Meteor impacts could burst through the crust, exposing ice below. (For more on this theory, tell NASA to fund Andy Rivkin's research.)

I'd vote to fund Andy Rivkin's research.

Yeah, but to paraphrase Chris, who has already pointed it out, such a small variation in brightness (9%) would barely be noticeable in an image that had not been contrast-stretched.

To put it in another context, the average albedo of the Moon (~0.12) is slightly higher than Ceres (~0.09). But the variation in localised albedo is such that the brightest illuminated part of the Moon has an albedo 150% of the average, and the darkest illuminated part has an albedo 50% of the average. Source: http://the-moon.wikispaces.com/Brightne ... d+Features

[geckzilla]

I'd like to see some raw data but in the meantime, some clues about Ceres's albedo can be found in these HST datasets. It is notable that the bright spots seem to be quite a bit brighter in blue to UV (<330nm) light than they do in green (555nm) light. http://sbntools.psi.edu/ferret/SimpleSe ... ESHST-V1.0

[BMAONE23]

I'd like to see some raw data but in the meantime, some clues about Ceres's albedo can be found in these HST datasets. It is notable that the bright spots seem to be quite a bit brighter in blue to UV (<330nm) light than they do in green (555nm) light. http://sbntools.psi.edu/ferret/SimpleSe ... ESHST-V1.0

Interesting Ceres images. I note that in the distant series of Ceres,
from 2004-1-23 23:52
Until 2004-1-24 08:52
Approximately 9 hour span, something Long passes by the Asteroid showing up in 7 frames
Was this looked into at the time and was any determination made?

Entering the frame

Passing Ceres

Leaving field of view

[alter-ego]

I'd like to see some raw data but in the meantime, some clues about Ceres's albedo can be found in these HST datasets. It is notable that the bright spots seem to be quite a bit brighter in blue to UV (<330nm) light than they do in green (555nm) light. http://sbntools.psi.edu/ferret/SimpleSe ... ESHST-V1.0

Interesting Ceres images. I note that in the distant series of Ceres,
from 2004-1-23 23:52
Until 2004-1-24 08:52
Approximately 9 hour span, something Long passes by the Asteroid showing up in 7 frames
Was this looked into at the time and was any determination made?

The streak you're seeing are from an ~11th magnitude star (1200-05453718, USNO designation) during the 108-sec exposure time. That is the only star aligns well with Ceres.
It's convenient that the star grazes Ceres. It made identification relatively easy.

[mollwollfumble]

The light spots on Ceres have to be either "Unusual light-colored rocks sitting on the surface" or "Underground light soil uncovered by recent meteor impacts".

First of all, I want you to take notice that all the images of Ceres from Dawn so far are monochrome, not true colour. Because of the way that Dawn takes photographs it only takes monochrome images in whatever filter is placed in front of the lens. There are seven colour filters designed not to replicate true colour but to study the geology. The only true colour images of Ceres so far are those from the Hubble space telescope. I eagerly look forward to high definition colour photos of Ceres, but it will probably be at least two months before the first one is made public.

Secondly, when APOD says "The enigmatic spots were first noticed on Texas-sized Ceres a few weeks ago during Dawn's approach", that's not true. They were seen in they much earlier Hubble images. However, they were much more startlingly prominent in the Images from Dawn.

Thirdly, white spots like this can be seen on Mercury, where they are clearly "Underground light soil uncovered by recent meteor impacts". The Aristarchus crater on the Moon is similar but not as bright. The white spots on Ceres have similar size to the craters on Ceres, which suggests that the same mechanism is responsible. However, I want to see the image that I hope will be taken on 19 Feb before calling it for sure. The reason is that whereas the white spots on Mercury and the Moon have white rays radiating away from them, the white rays for the brightest spot on Ceres have yet to be seen.

[Ann]

alter-ego wrote:
The streak you're seeing are from an ~11th magnitude star (1200-05453718, USNO designation) during the 108-sec exposure time. That is the only star aligns well with Ceres.
It's convenient that the star grazes Ceres. It made identification relatively easy.

Thanks, alter-ego! You are a sleuthhound indeed, in the best possible sense.

Ann

[alter-ego]

Thanks, alter-ego! You are a sleuthhound indeed, in the best possible sense.

Aw, gee. Thanks Anne.

[BMAONE23]

Nice sleuthing indeed it makes sense that it would be a star streak given the motion of Ceres in the time lapse images

[geckzilla]

I've been waiting for a good place to stick the link to this funny story. Here seems as good as any.
The "False" Supernova 2003lr

[Ann]

I've been waiting for a good place to stick the link to this funny story. Here seems as good as any.
The "False" Supernova 2003lr

That's incredible!!!

Ann

[sallyseaver]

The mystery link: http://space.io9.com/just-what-are-thos ... 1684160250 claims the spots are 9% brighter than the surrounding areas.

I like this idea from this article posted by Alter-Ego:
Ceres could have a rock core, an ice mantle, and a thin crust of debris and dust. Meteor impacts could burst through the crust, exposing ice below. (For more on this theory, tell NASA to fund Andy Rivkin's research.)

I'd vote to fund Andy Rivkin's research.

Yeah, but to paraphrase Chris, who has already pointed it out, such a small variation in brightness (9%) would barely be noticeable in an image that had not been contrast-stretched.

To put it in another context, the average albedo of the Moon (~0.12) is slightly higher than Ceres (~0.09). But the variation in localised albedo is such that the brightest illuminated part of the Moon has an albedo 150% of the average, and the darkest illuminated part has an albedo 50% of the average. Source: http://the-moon.wikispaces.com/Brightne ... d+Features

I am not sure that we can trust 9% as a definitive measurement of variation in brightness; (the images shown on http://space.io9.com/just-what-are-thos ... 1684160250 show little contrast). I looked at the link you provided, Nitpicker, and it says that there is the scale for Albedo goes from absolute black [0] to total reflection [1]. The albedo of the highest peak of the Moon is 0.183. If you look at this image from the Hubble telescope - no contrast tweaking (original image link provided by Geckzilla), you will see that the gray is in the .5 neighborhood and there are 2 whole pixels of total white. This is a much higher difference than 9%, it is closer to 50%.



This picture has a central filter wavelength of 2255.4224.

I know that there are concerns about whether the high-res image posted to APOD on Feb 18, 2015 has had its contrast modified. But it might be interesting to know that a 5x5 pixel eyedropper color sample taken from the middle of the boxes shown in the image below showed 99% brightness for the brighter boxed area and 55% from the darker boxed area. And this lines up with the low-res image from the Hubble Space Telescope.



According to Andy Rivkin's conjecture, there is a small layer of debris on the outside, so of course we would not see the whole extent of the ice underneath unless we get past the debris, which fortunately has happened to a certain extent in a few spots.

[geckzilla]

The processing on the images makes a huge difference in our perception of what is there, though. Here the detail within the white spots have been destroyed as the histogram has been clipped on that end. The unfortunate result of this is that then our brains see this as a glint since it is a typical pattern recognized in photography here on Earth. Anyway, we will know more soon enough. I'm not holding my breath for any exotic explanations. It looks...mundane, unfortunately. Nowhere near as exciting as 67p. Speculation always grows free and wild in the gaps between actual knowledge.

[Nitpicker]

Sally, I think you should look up the details of the HST images. They are not at all what you think they are, especially the uncertainty map you've re-linked to.

And as for the Moon data, I think you've missed my point completely.
Average Moon albedo = 0.12.
Aristarchus central peaks = 0.183/0.12 ~ 150% of average albedo.
Grimaldi floor = 0.061/0.12 ~ 50% of average albedo.

On Ceres, an albedo 1.09 times the average of ~0.09 would be 1.09*0.09, or a maximum albedo of about 0.10. Basically, the average shade and the brightest shade are really two slightly different shades of dark grey, barely distinguishable to the eye. (Unless you stretch the contrast.)

[geckzilla]

A few more details are now available from some new images of Ceres. The spot appears smaller (apparently, not actually) and is now two dots instead of one.
http://www.nasa.gov/jpl/dawn/bright-spo ... companion/

[ta152h0]

headlights

[Nitpicker]

Just for fun, I've taken the latest images from the Dawn spacecraft, taken on Feb 19, and scaled them approximately, so that Ceres appears in more or less its true size against a couple of my own Moon images. (All images have been reduced significantly for upload to The Starship.)

I see white spots all over the Moon, too:

[THX1138]

My bet is that a hard dirty snowball hit that spot hard and hot, the surface stayed hot long enough for the dirt mixture in the ice that was melted to settle to the bottom and that left a slightly cleaner ice to re-freeze on the surface around the impact area.

[sallyseaver]

Sally, I think you should look up the details of the HST images. They are not at all what you think they are, especially the uncertainty map you've re-linked to.

And as for the Moon data, I think you've missed my point completely.
Average Moon albedo = 0.12.
Aristarchus central peaks = 0.183/0.12 ~ 150% of average albedo.
Grimaldi floor = 0.061/0.12 ~ 50% of average albedo.

On Ceres, an albedo 1.09 times the average of ~0.09 would be 1.09*0.09, or a maximum albedo of about 0.10. Basically, the average shade and the brightest shade are really two slightly different shades of dark grey, barely distinguishable to the eye. (Unless you stretch the contrast.)

Dear Nitpicker – Thank you for continuing the dialog.

My understanding is that the histogram image has a precision that forces each pixel to be one of a limited set of values; and the error range for uncertainty is 1 pixel . I visited the page again and this is still my impression, so I am interested to know what the proper interpretation is.

Thank you for more words and explanations about the Moon albedo data. I understand what you are saying and I agree that .09 and .10 are both different shades of dark gray. I still have not received data from a trustworthy source telling us “that the areas of increased reflectiveness have only a 9% higher albedo than the surrounding surface.” The http://space.io9.com article does not say where the 9% figure is coming from; it could be from old data before the recent approach of DAWN. Please let me know why you trust this reported number.

ALSO I do not think that the Hubble histogram would have 2 pixels of 0.9-1.0 (with 2 pixels of about .6 below and 2 pixels of 0.4 to the right) versus the surrounding gray pixels in the Hubble histogram if the difference was only 9% of .09 (Ceres avg. albedo).

However, I see that my attempt to calculate a percent difference was wrong. If you grant that the lightest spot (i.e. average of 2 brightest pixels) is .9 albedo, then (.9-.09)/.09 --> 900% difference.

I have written to the Media (public) Relations person for the Dawn Mission (at JPL) to see if we can get JPL to release some better albedo data in an upcoming press release. It will probably take months; I hope it's not years. Please let me know if you see any data on Ceres albedo from the DAWN mission.

On another note, please forgive me for dropping you and other readers into the middle of my Mass Vortex theory for formation of the solar system in my first post, with no preamble or preparation or allowances for the need to shift gears from the current paradigm to a new one. I went back a few days ago and could see how jarring and outrageous it must seem to everyone. I truly appreciate the graciousness I have received from you and others.

[sallyseaver]

My bet is that a hard dirty snowball hit that spot hard and hot, the surface stayed hot long enough for the dirt mixture in the ice that was melted to settle to the bottom and that left a slightly cleaner ice to re-freeze on the surface around the impact area.

This is an interesting idea and it seems plausible. A comet could have struck Ceres (your dirty snowball) and the white areas appear to be within an impact crater. Non-ice components of the comet would be heavier and sink. Heat from the impact would melt the ice which would refreeze over time - in cold space. However, this explanation does not cover the other larger regions of lower-contrast lightness. On the other hand, via the dialog in this thread comparing the light and dark regions of Ceres to those on the moon, maybe the other regions are not as significant as they appear in the Feb 18 APOD image. I vote that your explanation go on the short list.

[alter-ego]

Thank you for more words and explanations about the Moon albedo data. I understand what you are saying and I agree that .09 and .10 are both different shades of dark gray. I still have not received data from a trustworthy source telling us “that the areas of increased reflectiveness have only a 9% higher albedo than the surrounding surface.” The http://space.io9.com article does not say where the 9% figure is coming from; it could be from old data before the recent approach of DAWN. Please let me know why you trust this reported number.

You'll recall I cited the original reference as having "claimed" the 9% figure. Even though it is consistent with the spot's relative brightness appearing to have been accentuated by contrast stretching, I've not been fully convinced. To add to the fray, recently Chris Russell (principle investigator for the Dawn mission) has been quoted as saying the bright spot's albedo could be as high as 0.4. Based on the fact the spots are not fully resolved he considers ice to be a tenable candidate for the higher albedo.

"Right now, all we can say is that the material reflects 40 percent or more of the light falling on it," UCLA astronomer Chris Russell, the principal investigator for the Dawn mission, told NBC News in an email. "This limit is because of the resolution of the camera at this distance from Ceres. If the final answer ... is that it reflects all the light that falls on it, then the most probable reflector would be ice.

This report may be closer to the truth, but it all is still hearsay as far as I'm concerned.

[Nitpicker]

My understanding is that the histogram image has a precision that forces each pixel to be one of a limited set of values; and the error range for uncertainty is 1 pixel . I visited the page again and this is still my impression, so I am interested to know what the proper interpretation is.

Thank you for more words and explanations about the Moon albedo data. I understand what you are saying and I agree that .09 and .10 are both different shades of dark gray. I still have not received data from a trustworthy source telling us “that the areas of increased reflectiveness have only a 9% higher albedo than the surrounding surface.” The http://space.io9.com article does not say where the 9% figure is coming from; it could be from old data before the recent approach of DAWN. Please let me know why you trust this reported number.

ALSO I do not think that the Hubble histogram would have 2 pixels of 0.9-1.0 (with 2 pixels of about .6 below and 2 pixels of 0.4 to the right) versus the surrounding gray pixels in the Hubble histogram if the difference was only 9% of .09 (Ceres avg. albedo).

However, I see that my attempt to calculate a percent difference was wrong. If you grant that the lightest spot (i.e. average of 2 brightest pixels) is .9 albedo, then (.9-.09)/.09 --> 900% difference.

The description of the Hubble dataset (which explains the content of the images, including the uncertainty maps, without any mention of histograms) is here:
http://sbntools.psi.edu/ferret/datasetD ... ESHST-V1.0

The "9% higher albedo" came originally from this (rather low resolution) Hubble dataset, and the small number of papers written around it, which until a few days ago, provided perhaps the best evidence and analysis for the variable albedo on Ceres. As alter-ego is now suggesting, the evidence from Dawn is quickly surpassing the evidence from Hubble (as expected). But regardless of whether the brightest portions on Ceres can ultimately be shown to be ice, space-weathered ice, or not ice at all, the "9% higher albedo" was the correct conclusion based on the analysis of the Hubble dataset.

[geckzilla]

It seems that we've taken too much away from that conclusion, though. At Hubble's resolution, the spot could have been a diffuse area of slightly different albedo or, as we've come to discover, a small area of apparently significantly different albedo.

[Nitpicker]

It seems that we've taken too much away from that conclusion, though. At Hubble's resolution, the spot could have been a diffuse area of slightly different albedo or, as we've come to discover, a small area of apparently significantly different albedo.

Yes, but we were right to stick with that conclusion, rather than speculate over a handful of new, higher resolution images from Dawn, with hardly any published technical details. Once technical details of the newest images become available publicly, the situation will be different.