Pretty picture: An unexplained chain of elliptical craters on the Moon

Here's the first cool pic I've managed to produce from the recently-released Lunar Reconnaissance Orbiter Camera data set. Figuring out how to work with these data turned out to be an unexpectedly arduous process, which I detail in the previous post. This post is intended to show you a pretty pic -- and save the rest of you some pain!

Check out this weird chain of elliptical-shaped craters:

This is a small segment of an LROC observation made on July 17, 2009, less than a month after its launch. At the time, the spacecraft was in its preliminary orbit, higher above the lunar surface than its planned science orbit, so at its full resolution LROC got 1.55 meters per pixel on the surface here, about three times poorer than its eventual science images. That hardly matters; it's an astonishing view even at this scale. The version of the image at left clicks to an enlarged view that is 10 meters per pixel; the whole image covers an area about 15 by 80 kilometers in size.

What process could be responsible for these weird elliptical craters? They are definitely not impact craters. They are in a chain on the volcanic plains and have a somewhat sinuous outline, so it is tempting to assume that they are some kind of volcanic feature, maybe collapse pits over a subsurface lava tube. But I'm not very happy with that explanation. Why? Look at the spaces in between the elliptical pits. In between the pits there are funny little hills. At the southern end of the elliptical chain, the elliptical pits disappear and the feature turns into a raised hill with a somewhat jagged outline.

That raised hill with a jagged outline is, I believe, a wrinkle ridge, a feature that is very, very common on the lava plains of the Moon and Venus and is even found on the rare flood basalt lava plains on Earth (such as in the Columbia River plateau flood basalts). But I've never seen a wrinkle ridge turn into a chain of collapse craters like that before. Of course, I haven't studied the Moon very much; maybe there are more of these kinds of features elsewhere! I'll have to keep looking.

What do those collapse pits look like at full resolution? Scroll down.

This version of the image is shown at its full resolution of 1.55 meters per pixel (once you click to enlarge); it covers an area about 2.8 by 7.4 kilometers in extent. The dramatic lighting here is throwing deep shadows into the floors of the elliptical craters. Since the Moon lacks an atmosphere, there is basically no light being scattered into those shadows (although there should be a teeny tiny amount of light in there that's been reflected off of the sunlit crater walls).

In the sunlit areas, you can see delicate little fractures concentric to the elliptical craters; those kinds of cracks form when a slope is slumping, when there is not enough strength to hold up the materials in the wall and they start sliding down. There's some kind of fissure below the surface that is opening up, removing the support from underneath these features, and the lunar soil is sliding downward. The pairing of what are clearly collapse pits with other features that look like wrinkle ridges has me very confused; wrinkle ridges form when you compress (squeeze) the ground, while collapse pits and slumps form under tension, when you stretch the ground apart. Now, there are places on wrinkle ridges where tension happens, but still it doesn't seem like there should be enough tension over the folds in a wrinkle ridge to make collapse pits that are as wide as the ridge itself. It's not adding up. But it's cool to look at!
Some of you may be saying: cool, now how do I get my hands on pictures like that? Here's my recommended method. (For more background on this, read my previous post.)

First, make sure you have the latest version of IMG2PNG. Browse the LROC data set to find an image of interest. I just discovered yesterday that they actually have a map interface to their released images, which seems very handy. Once you have settled on an image that you think is worth downloading, follow the link to "Download CDR." Make sure you don't get the EDR; IMG2PNG is tested only with the CDR versions of the data. Go get some lunch, maybe work on something else; it's been taking me about an hour to download the 500 MB files.

Run the file through IMG2PNG, and you should wind up with an enormous PNG file that you can open in Photoshop or GIMP; many other image processing software packages might choke on these files because there are so many pixels. If you are using GIMP you may want to experiment with the -s switch to stretch the image during conversion, since GIMP will squash the 16 bits down to 8 bits when you open it. Once you have the image open, you will need to determine whether it needs to be mirror-flipped or not. To do that, note the center latitude and longitude of your image, then go to the PDS Map-a-Planet page and generate a context map for yourself.