MRO_first    MRO_second   tagged_old    tagged_new

Notice how the faceting on  the right side of the "cross" image matches the "east wall" detail of the new image in outline. That is a lot harder to explain away than the triangular "G" feature on the enhanced comparisons, since tit involves their own new detail. Even if one was to grudgingly concede that perhaps my processing had pulled more detail out of the dark than  theirs,  that would explain nothing because the angle is different- this hole is supposed to have sides so vertical that a straight-down view can't see them., yet the area of correspondence is about the same size on both. However, if we are looking at a structure which rises above the ground , then it would be the slope of the roof we see, and it would  be visible from either angle of view. The same argument can be made for the detail around the edges. Next, there is that "G" area, a triangular shape within what should be nothingness on either image. There is a slight difference in the left side- it deviates from a straight line in opposite ways, that is to say it has a bit of waver. This is consistent with  either of two things-  if the top surface is curved and transparent, the angle of view would distort that line in such  a way; or  the perspective correction done to the newer image to make it less oval and more of a match to the older one could have introduced exactly that kind of distortion as well. If, on the other hand, it was merely some sort of imaging or processing flaw or if it had been painted in for some unknown reason, that distortion would not be there at all. Only if any processing was acting upon  something that was already there could you see any such effect. 

And on the possibility that we might be looking at  a puddle of oil or a lava  flow, those options both would require discarding the thermal data, as far as I can determine. The heat retention of either would not match the THEMIS  information. Of course, the thermal readings could be bogus...but it probably isn't wise to assume that without some evidence.

A reasonable question to raise would be, "What about the published analysis of these holes?  Surely the scientists who wrote technical papers were not trying to push any lies, were they? What did they find, and do those findings support or refute this crazy idea ?

That is an excellent question, so let's investigate. And don't call me Shirley. I'll cite one little quote from the paper by the team who named the holes. You can read the entire paper (it's only two pages) by clicking  HERE.

Lunar and Planetary Science XXXVIII (2007)

THEMIS  OBSERVES  POSSIBLE CAVE SKYLIGHTS ON MARS

"Figure 2 shows afternoon temperatures for Annie
that are warmer than the shadows of adjacent
collapse pits, and cooler than sunlit portions.
Meanwhile, nighttime temperatures for this candidate
are warmer than all nearby surfaces. Such is the
behavior we would expect of a cavern floor that
receives little or no daily solar insolation."

OK, that sounds like a legitimate observation, but does it lead to any inevitable conclusion, or could it be interpreted more than one way?  First, I should point out a small mistake- "Figure 2" on their paper clearly shows the feature they dubbed Jeanne, not Annie. Check for yourself. Their names, their paper, their illustration- if I made a booboo like that, there are people who would immediately say that it "proved" that I didn't know what I was talking about, or that I was obviously messing with the data. Yet such is the conditioned reverence for Scientists that not only did no one call them on it, many people are now confused as to which hole is which. There is a joke in there I will politely avoid making. In a spirit of charity, I will just consider it a typo, rather than subtle disinformation or sloppy work.

UPDATE: Now that an MRO image of another one of these has been released, which we will look at shortly... I am still not sure of the names. I am simply trying to provide you an accurate reference, here.  I did not intend to make a huge fuss about what seemed to be nothing more than a bit of sloppy editing in that initial paper, but I think its authors should have clarified this already, since their work has circulated so widely. A subsequent paper (not provided here, it says nothing new) by a different author,  repeats the same apparent mistake, making me wonder if he bothered to check. The new MRO image, PSP_005414_1735, is of the previously yclept Jeanne feature, which is the one obviously shown in the "Figure 2" temperature graphic cited above, and the MRO people call it "Annie" (in agreement with  the quote above from the paper). Sigh.

Here's the confusion, if it is at all possible that you aren't following this: the caption for the graphic says, quote:

Moving on, let's consider this temperature business. Suppose you had a building with a somewhat reflective surface on the roof. It would show the same characteristics, catching more sunlight than shadowed areas but remaining cooler than darker solid areas. As an enclosed space, it would be expected to retain more  heat at night than the solid areas, too, as those  would exhibit more convective cooling. So the observed data seems to provide equally valid support for the proposal that these features are indeed structures. Of course, those researchers were not looking for buildings.

Here is the supposed Wendy  Hole. Interesting building, but not unusual by Martian standards. Note the shadows, both there and on the wide view above. Since all the esa versions are from only  three images, it is becoming more possible that none of the holes are holes.

This slice shows a little bit of everything... but there is all sorts of detail in the  apparently chaotic area on the right side. Click  HERE for a higher res look at a tiny spot in the lighter toned part. I thought the most interesting single feature at a larger scale was the tall Inca with the qena (flute) in the dark area on the left.  Click  HERE for a better look at the flute player- and notice the way the edge has broken away on the lower left. It is quite visible on the full size original, too, which should be of interest to some. Ahem. The only real mystery about this "hole" is why the nearby features (visible in the long view of the esa image) look like hills there but depressions  ( "collapse pits", as they call them) on the MRO image. I suspect it is at least partly a perspective effect. There is a long examination of just that phenomenon on the Inca City page.  

PSP_003317_1975 detail

 Thumbnail to the left leads to a higher res close up. On the right,   the original as-posted appearance. Click HERE to view a  fully  enhanced version done as if the "pits" were actually dark like that. Looks interesting, but there are reasons why it cannot be the correct orientation, which topic we'll get to in a minute. I wonder what the technical distinctions are between pits, holes, and craters?

The Dance of the Pleiades by Elihu Vedder (engraving by F. E. Fillebrown), superimposed on  image UKS 18  ©  Anglo-Australian Obs./Royal Obs. Edinburgh..

If you would be so kind as to click on the thumbnail below and study the full-sized positive-negative comparison for a few moments, we can proceed. The hole was not quite all captured on that one RAW image strip, in case you were wondering. Compare the ground as well, not just the putative hole. Which way looks right? I can't help noticing that the interior part  really looks more like it is an elevated slope on both. That might be an optical illusion, of course. But look at the perimeter- on the left image, it is bright, like the sun was reflecting off a ridge, but there is an area immediately to the right which is dark, with no easy explanation why it should be in shadow, if the light was coming from that side. If the light was instead coming from the left, consistent with the idea of the interior section being down below the surface of this "rimless pit", the lighting still doesn't really work out logically. On the right hand image, which looks like a big chunk of chalk, the shadow around the base seems much more natural, and the light-toned area looks like a reflection off the side. Very confusing, isn't it?

 Look closely at the left side of each. See how there is a faint indication of some angular detail? See how it all looks "squashed flat", as if  ( but it isn't) pressed up against a plate of glass? That's the next thing we need to talk about...

 http://joshuacolwell.com/blog/index.php/2007/very-strange-mars-hole/

It was in answer to a question about the utter blackness of the center of the Annie hole image. We can assume it applies to the others as well, in principle. I highlighted one part of his remarks.

'It looks black because of the DN cut off in the lookup table. HiRISE is a 14 bit camera, but we usually LUT the data to 8 bits based on what we predict the histogram to look like. An unpredicted dark spot would get cut off."

Well, I'm certainly glad we cleared that up.  I will now attempt a translation. Apparently, this means they  set thresholds at the top and bottom of the histogram ( the total range of values from darkest to lightest)  for what data is translated into an image. Put another way, someone is deciding in advance that any values darker than "x" are not going to provide useful information, and can therefore be discarded. It does occur to me, however, that if one is concerned with deriving maximum detail from a hole, as opposed to shadow, this might not be the best approach. It explains the squashed effect noted above, however- and when you clip an image that way, there will be false-plane imaging artifacts generated at the threshold as the cut-off slices across and artificially joins areas of equal value. If, that is, you do any enhancement of that image which involves stretching the histogram. The modified parameter has effectively created a new, separate "channel" (in a Photoshop sense) which would need to be treated individually.

That is one of the reasons I never use any selective-area enhancement protocols. Those threshold settings can cause problems, so it is safer to not try too hard to recover that lost part of the histogram, lest the rest of the values get skewed by the attempt. In many cases. it has been clipped anyway, so there is nothing to recover. By way of example, on the "dark spot" enhancement ( the link in the text above the Pleiades) you'll see the interiors are even darker than the original shows because I did not  separately treat those areas.

Oboy. Maybe we need to take another look at the Cross image? We shall do that in a moment.  Isn't it odd that Dr. Mellon seems to specifically admit  this was done to the very image which was presented as so carefully stretched, that is to say, expanded? This is going to be good, I promise you. But first, one more look at 3317_1745, this time  in B&W for easier comparison with the others. Notice how the tops are so flat, unlike what can be seen on the ones from the full-resolution RAW file. Obviously, that threshold was set pretty high for the browse image (remember, this is reversed). Eerily similar to the Annie image, in the sharp edges and more believable shadows at the bases. Pretty intriguing for "rimless pits".

Armed with our new understanding of how they fiddle with the images, or at least one frank admission of one of the things they can do, let's see if we can detect anything amiss in that Cross image. I have been stating from the onset that I was suspicious. The posting of an image replete with apparent Christian symbolism right before the feast of Pentecost seemed awfully convenient, if not contrived. Yet only by applying some fairly (forgive me) sophisticated enhancement technique could that image be brought out, so exactly who was the intended target audience? And how could such a risk be justified? There would have to be, at minimum, a considerable degree of smug assumption that no one could discover or prove anything...about something THAT WAS DESIGNED TO BE DISCOVERED. How bizarre is that?  I have made few changes in the previous pages as this situation has developed so as to preserve some semblance of continuity and context. As so often happens, the missing piece of the puzzle came from an unexpected direction. The key is that discussion of thresholds and clipping above.

See, if you try to alter a digital image, the result is not the same seamless invisibility you can try for when  assembling (or drawing)  one from scratch- which has a separate set of texture problems, lest the result look too much like a cartoon.   There are markers that can be detected. Even in Hollywood, the special effects rely on some suspension of disbelief. How many times have you seen a movie again a few years later, and been surprised  how cheesy the effects seemed, that were so convincing the first time? As the standards for viewing improve, the level of deception needs to be improved as well.   The safest place to alter an image is at the source. That is why the idea that something might have been added  to an image, as opposed to simply obscuring or even removing some element, seemed preposterous and improbable.

If  working at the source is not an option, the next best way to effect change is by rephotographing the altered image, thus making that the "source".  Difficult to do when the originating camera is unavailable for the reshoot. This where the thresholds come in to the story.

Remember the histogram? There is all the data, spread across a spectrum of values. If you arbitrarily limit the extent ( the range) of data across that table  by clipping anything below a certain threshold, you have created an artificial space, into which the remaining values could be expanded, effectively increasing the details  of that data, or where new data could be inserted, especially if that data represented something dimensionally and graphically simple, like say,  a cross and a few curlicues.  Only when the new amalgam was contrast- enhanced to an extreme degree would that added data buried at the bottom of the histogram interact with the "real" contents and become noticeable.  Instead of separating from the original data, it would merge with it, picking up missing parameters from the main data matrix.  Resampling the image would have much the same effect as the rephotographing option mentioned above. Seems like a lot of trickery to perpetrate just to booby-trap an image , doesn't it?  OK, does any of this have any relevance to the Arsia Cross? And how would one go about proving it?

There is something called steganography which is at least conceptually related to this sort of image manipulation. The term refers to various methods for embedding text or another image in a picture, invisible until  revealed by some appropriate decryption protocol. If you are a spy, a politically unpopular group or a pervert of some sort and feel the need for such stealth, the software to do it is readily available. Aha. Maybe there is a bit of relevance here- what if the Cross was intended as a Secret Message, perhaps some sort of arcane Masonic ritual devotion? NASA has often been accused of indulging in displays of ritual symbolism...

The Jet Propulsion Laboratory (JPL) was often jokingly called "Jack Parson's Laboratory" in its early days, and that founding member was a cohort of the most notorious mystic of the 20th century, Aleister Crowley. Those who seek can find seemingly endless quantities of Sacred Geometry,  ritual dates, astrological alignments, and magick-related stuff in the doings of the space programs. This may indicate a continuing connection with the ancient traditions and Secret Societies, or some of it may simply be technical staff with extra time on their hands trying to drive researchers like Richard C. Hoagland crazy. As it happens, he has a new book exploring those associations, Dark Mission,  so we won't spend any more time on  them here.

You can see, in this exaggerated view, a hard line around the edges of the cross, the signature of old-fashioned fletchett knife trimming. A transparent mask of irregular shape ( the lighter area on top, one of the "layers" visible on the enhancement a few pages back) was applied to the center of the dark area. The cross itself was applied to that (probably first), and if you look closely you can see the horizontal piece of tape overlapping the vertical piece. There is also a tiny shred of "scruff" on the right side of the crossbar where it had to extend beyond the edge of the irregular-shaped main mask and the trimming was not precise on the corner. When I said "masking" in explanation of how I derived this image I was speaking of a digital product, but here I mean the old-fashioned darkroom physical kind. So somebody took the original image created from the downloaded data and mixed it with the cross overlay - not  the same way that I made that Pleiades Boogie picture, but rather mechanically , physically assembled on a light box ( a backlit glass table). Old-fashioned, like I said, just like that notorious Lee Harvey Oswald photo from years ago. Then the result was digitally re-imaged, and only the lowest range of values saved, to be mixed in below the cut-off threshold of the processed-for-the-public version of the original. That is the most likely scenario. The one thing it can't be is accidental, but I still cannot fathom the purpose .  It is almost enough to make me embrace the deepest theories of arcane Masonic ritual symbolism. Almost. It would be quite easy to   theorize about some sort of symbolic display or even "secret message" involved here, though I 'd prefer to have a bit more context information first. It is probably  a pretty safe bet  that any such message was not embedded in there by any supernatural entity.

This may not be the end of this. I may find some flaw in the exposé, or even uncover some motivation for the stealthy  image alteration.  If that happens, it will be presented here. For now, let's move on to other things.