Are Double Craters Caused by
"Image Smear?"

By Mike Bara
and
Steve Troy

CONTACT: MICHAEL BARA

Recently, some web pages have appeared on the net alleging various "explanations" for data posted both here and on the Enterprise Mission web site and publications. Most of these "explanations" have (as usual) already been dealt with in previous publications, but at least one is new and has not yet been addressed in public.

A highly critical page out of Europe has claimed, among other easily refuted things, that the double craters noted by Hoagland in Lunar Orbiter frames LO-III-85 H1, H2, and H3 were "obviously" caused by image smear. This persons claim was that the movement of the spacecraft had caused stretching or distortion of the high-res frames. This person was unaware of the story previously posted here "Double Craters in Hortensius." Despite my attempts to make him aware of it, he continued to insist that this was the explanation.

Naturally, Hoagland had considered this explanation when he did his original research in the early 90's, as Steve and I did when we wrote our report last year. We rejected it because quite frankly it was refuted by studying the images. We never included it in our reports simply because it was not a remotely plausible explanation.

Still, now that the issue has been raised, Steve and I have decided it may be necessary for the ill informed to address this "explanation" directly.

The Lunar Orbiter cameras were equipped to deal with the issue of spacecraft motion during exposure. The film was moved during exposure to compensate for the spacecraft velocity, which was estimated by an electric-optical sensor. Despite this, some longer exposure times (.04 sec or longer) did create some occasional minor smearing. (The longer exposure times were dictated by illumination conditions and camera obliquity).

It is noted in CR-Report 984, Lunar Orbiter III Photography that excellent surface detail is shown in the telephoto frames (H-frames):

Can the doubles be explained away by the presence of image smear? There was a real problem with smear on the first two Lunar Orbiter missions. Much of the photography from them was not released due to the poor quality of the images resulting from smear. On Lunar Orbiter III, the images improved substantially. Indeed, some of the frames where the doubles have been found were taken at a shutter speed of 0.04. However, the tilt angles of the photo sites where they were found on this mission range from 57 to 69 degrees -- well outside the "normal or nearly normal" (90 degrees) parameters where image smear was described in the Contractor Report above.

Sectional Enlargement from LO-III-123-H2

As you can see from this sectional enlargement in the Hortensius region, there is some blurring or lack of sharpness in the image. This minor bluring is the image smear referred to in the report. There are several factors which we considered when judging the possibility that the "Doubles" may have just been a product of this minor distortion.

1. The direction of travel of the spacecraft. Any possible smearing must coincide with the spacecraft path. If the orbital path is somewhat vertical, and the smearing is horizontal, this absolutely cannot be the reason for the "doubles" effect. In addition, the observed "doubles" must uniformly be parallel to each other without exception. The spacecraft can only travel in one direction at a time, so if smearing is the explanation for the "doubles" you cannot have divergent "smear paths."

2. The shadow patterns in both sides of the double craters must match exactly. If the shadows are different, the only explanation is that the two halves have a different interior composition (in other words one side is a bit more filled in than the other) or are actually laying at differing angles to the light source (the sun). In either case, the only explanation for this observation would be that the two different halves of the "double" are actual features on the lunar surface.

3. The central divider of each "double" must match at least one "side wall" of the crater. If the explanation of the central divider is that it is just one side of the crater "smeared" over, then this must also be the case in each and every double visible in the frame. In addition, the same side of each and every double must be the one that matches the central divider. (In other words, one double cannot be matched to the left side wall, and one to the right side wall. This would be a direct contradiction of the "smear" explanation.)

4. Finally, if smearing is the explanation, than all doubles visible must be smeared to the same degree and all other features in the frame must also demonstrate this same degree of smear.

It was with these four tests in mind that we began our work.

Direction of Travel

The Lunar Orbiters generally traversed the equatorial regions, and in the case of the Hortensius frames LO-III-123-M, H1, and H2, the direction of travel was left to right. The orientation of the camera to the ground track can be determined by assessing the framelet lines in the images. Since LO cameras took images in strips that were reassembled after transmission to Earth, the framelets act as a guide to the spacecraft's course. In order for the doubles to be caused by an optical illusion -- image smear -- the "smearing" must take place parallel to the framelets in each and every case of doubles. Clearly this is not the case.

As you can see from the version of LO-III-123-H2 above, the double craters visible in the frame are not in alignment, either to each other or the framelet (The green line represents the framelet reassemble line). By simply drawing a line over the straightest edge visible on each of the double craters, we can show that while essentially parallel, the edges of most of these objects are not precisely parallel, which they must be for image smear to be the cause of the "doubles" effect. In some cases, the angles at the point of intersection are up to 20°. . Since the spacecraft cannot be travelling at so many divergent paths at the same time (four one-hundredths of a second exposure), this cannot be the cause of the double effect. This represents the first proof that image smear is not a valid explanation for the anomalous double craters.

Interior Shadow Patterns

The second test requires examining the doubles one by one. By taking individual crops of many of these doubles, we can enhance them and examine them closely to determine if they meet the conditions of the second test.

All of the images above were taken from LO-III-123-H2 (with one exception). They were enhanced using the Extensis Intellihance Gray Scale enhancement filter. The preferences were set as:

• Contrast - Normal
  Brightness - Balanced Tone
  Sharpness - Medium
  Despeckle - Off

Following this, the Gamma was adjusted to a value of 1.3, and a Gaussian Blur filter applied with the radius set at one (the lowest setting) to counter the effects of the sharpness filter. These same enhancement techniques were applied uniformly to the entire cropped image. The results are as you see them in the table above.

Probably the most convincing and obvious fact to be mentioned regarding the doubles is the difference one can easily see between the hemispheres or lobes of each double. One hemisphere of one double is DIFFERENT than the other. If one looks carefully even at the doubles in the upper sections of the H2 frames of either area, one can see how the sunlight plays on each lobe differently. This affects the shadows that are cast. The lobe of the double that is farthest from the direction of the sun’s light is different in appearance both on the inside and outside than the lobe that is closest to the sun.

In each and every case above from LO-III-123-H2 (and one from LO-III-123-H1) the left and right side of the double craters do not match. In the "smear model," they must match. Since they do not (and they are not even close) the only viable conclusion is that the two separate sides of the doubles are receiving different levels of illumination from the light source (the sun). Note that it is not necessary to use the enlargements or enhancement filters to make this assessment. The differences in illumination are plainly visible in the unenhanced wide angle images.

The enhancements are for illustration purposes only. We did not selectively choose just the "doubles" that fit the pattern we were trying to support. We  provided a sample cross-section of the doubles on these images, and would be glad to provide the full resolution original scans via e-mail upon request. The simple truth is that none of the doubles on these images have identical interior shadow patterns. In fact, it is quite clear that the two halves are completely different in most cases.

Since the "smear" argument cannot be supported by a close examination of the images above, what could be the reason for the different shadow patterns in these doubles? One obvious consideration is that the doubles follow the local topography (this in itself argues that the doubles are actually on the Lunar surface). It also appears that in some cases one side of the double is more filled in than the other with debris. This has the effect of raising the bottom of that side of the "double" into the sunlight.

It is important to remember that if even one of these doubles follows the local topography and has two distinct sides, the "smear" argument is refuted. Since we have multiple doubles showing this to be the case, this represents a second proof that image smearing cannot explain the presence of the doubles.

Other Tests

The final two issues to be resolved are the questions of symmetry and evidence of smearing on other "non-double" features in the frames.

This wide angle view of some of the enhanced doubles in the LO-III-123-H2 shows that the central "bar" dividing the two lobes in most cases does not match either sidewall of the double. This qualitative assessment is a further indication that the "bar" itself is a unique feature of the terrain and not simply the right edge of the crater "smeared" to the right in the image.

The final nail in the coffin of the "smearing" explanation is the enhanced feature in the upper left. It appears to show recessed openings or bunkers with supports in between. Note that this feature is about the same size of the largest doubles, and it clearly has three distinct sections. There is no evidence of smear, certainly not to the length of the doubles visible in the field right next to it.

Conclusions

Our assessment was (at the time of our initial report on the Hortensius doubles) that smearing cannot be the explanation for the double effect for the following reasons:

1. The observed doubles are measurably not parallel to the flight path of the Orbiter or to each other.

2. The shadow patterns on the floor of each lobe of the doubles are not identical or even similar.

3. The central dividing "bar" does not match either side of the double and therefore cannot be simply a "smeared" version of the crater rim.

4. There is no indication of other nearby features showing a similar "smear" effect.

It is our conclusion that the "very small amount of image smear" noted in the Contractor Report refers to the slight blurring observed in the "H" frames, since many other "H" frames from Lunar Orbiter 3 display a generally sharper focus and better image quality (see here). However, it is possible that this entry may been in reference to the doubles themselves, since they are plainly visible on the full size prints and clearly do not fit the traditional model of Lunar geology. It may simply be a way of the project scientists "explaining way" features they cannot account for through conventional models.

In addition, Steve made the following assessments based on his examination of the analog images:

If one carefully examines the III123H1 and H2 frames one is able to see single craters and features that are AS SMALL as the double-formations that are resolvable and in focus and that are ADJACENT to them. One is able to see the difference between the features and the doubles that isn’t due to smear. Occasional level terrain shows textural surfaces unchanged by smear. Smear would have defocused all features and would have stretched all features the size of the doubles including surface delineation.

Another area where they were extensively found was in the area of the Eratosthenian crater Cavalerius and the adjacent, Cavalerius Hills on the western border of Oceanus Procellarum. LOIII214H1 and H2 photos were examined and sent to members of our research team back in 1996 along with several others from other missions. It is secondary site IIIS-30. The Luna 9 site lies to the northeast of Cavalerius near Planitia Descensus. Cavalerius crater is said to have been formed after the emplacement of the last local mare units.[2] It has a relatively preserved ejecta blanket and a diffuse ray system. Interestingly, NASA has studied both Hortensius and Cavalerius as future potential lunar resource sites. (to be discussed in future reports)

A double crater from LO-III-214-H2 (Cavalerius hills)

Double craters are evident on BOTH H1 and H2 frames of III214. This is significant because the higher the number of H frame from 1 through 3, the farther back in distance one sees (remember that the Contractor Report above emphasized the presence of smearing in the FOREGROUND). H1 is the fore-area of high-resolution coverage while H3 is the background toward horizon-area. The alignment of the doubles here is almost south to north. As reported in my second Hortensius report, the doubles are again similar in appearance and are located on relatively level as well as extrusive terrain. Again, even in the foreground areas (H1) there is clear resolution near to the doubles. They are seen in middle ground areas, as well.

More importantly is the fact that their configurations are also very discernable toward the TOP of the H2 frames of both III123 and III214. That means they’re visible well into the MIDDLEGROUND areas of the frames. On III214H2, Planitia Descensus is seen. On the wide-angle frame III214M, it is located more than halfway to the top of the frame. The moon’s horizon is about two thirds of the way up from the bottom of the M-frame. Doubles can STILL BE DETECTED on Planitia Descensus on the H2 frame which is farther back than simply "middle ground". Furthermore, with a good lupe and a good negative, they can be seen north of Descensus out into the mare area!

Double crater archologies can be seen on a variety of lunar terrains well back into the distances away from the foreground areas. It must be mentioned that the III214H2 frame is remarkably focused and all features are very clear and sharp! It must also be noted that the doubles like these have been found on VERTICAL Lunar Orbiter photography, as well. They have also been discovered on Apollo 8 and Apollo 11 photography. They aren’t relegated to one mission’s photography data.

This final observation is significant because if the doubles were restricted to only one kind of photographic process and just the Lunar Orbiter missions, the possibility that they might be a product of a defective imaging system increases. Since this is not the case, our assessments regarding the "smear" argument are strengthened.

Since the arguments that might support the smear conclusion all fail, we are left with no alternative but to argue that the doubles likely have an artificial origin. Such regular repeating features are indicative of construction rather than random meteoric impact. We will be providing additional reports on doubles in other regions, and on the doubles found on the Apollo 70mm and belly camera photography.

FOOTNOTES

1 NASA CR- REPORT 984, Lunar Orbiter III Photography, Boeing Co., Feb. 1968, pp.56-69.
² Shultz, Peter, Moon Morphology, Interpretations Based on Lunar Orbiter Photographs, Univ. of Texas Press, 1976, p.420.

COPYRIGHT © 1999 MICHAEL BARA AND STEVE TROY