I’ll be attending a µSymposium before the Lunar and Planetary Science Conference this coming weekend, and I just got a reminder e-mail today. Included in that e-mail was a link to an animation that shows Shackleton crater, a crater that is ON the moon’s north pole. As such, its interior is in permanent shadow.
BUT!! The Lunar Orbiter Laser Altimeter instrument (LOLA) on the Lunar Reconnaissance Orbiter (LRO) has plenty of data that allow it to be viewed: Click Me!.
I find this very neat — until the last decade, we could never see stuff in permanent shadow because we didn’t have the instrumentation. LOLA and LROC have allowed us to do that. And there are thousands of craters in permanent shadow on the moon that may hide water (which is what I’ll be presenting at the µSymposium).
For reference, the north pole of the moon is just about smack dab at the 10:30 position on the large crater’s rim. Just inside the rim, along a line from that small crater just outside the rim to the center of the crater.
But for pseudoscience, you may also notice that there are some artifacts in the data. There are radial streaks from the center of the frame (usually). There’s a prominent one diagonally from upper right to lower left on the upper wall of Shackleton itself. Others are more prominent towards the edges of the animation.
These are not lunar roads nor subways nor trollies nor anything else made by an ancient civilization. They are artifacts in the data itself. LOLA is very well calibrated, and the “average” (root-mean-square) uncertainty is under 5 meters in elevation data. But some tracks (orbits) are a bit off. And since LOLA is fundamentally measuring the time it takes light to bounce off the surface from a laser beam from the craft, it needs to know exactly where the craft was to get an accurate surface elevation.
And some are off by a bit. These manifest in this kind of product as ridges or troughs that are perfectly in a straight line, along the line of the orbital track. It’s something that scientists who use these data see and ignore because we know exactly what they are. But pseudoscientists will look at line artifacts like this, or at image seems in a mosaic, and claim things like they are artificial tram lines.
Exposing PseudoAstronomy 
The movie combines the results of LOLA data from a lot of orbital passes. How do I know this? Because in comparison to a LRO Narrow Angle Camera image which has a horizontal scale of 5000 pixels or 2.5 km at the LRO’s nominal 50 km orbital altitude, whereas the same LOLA data has a horizontal scale of only 5 pixels per 2.5 km at nominal orbital altitude. Shackleton crater’s diameter is 21 km. Clearly dozens of LOLA data sets have been combined in order to achieve the resolution and subtle details which we see in the movie.
Obviously “strips” of LOLA data sets which cover a given swath of lunar terrain have been combined in order to achieve much higher resolution, and then adjacent “strips” for another swath of lunar terrain were similarly combined. The final step would be to achieve a best fit of the bundles which comprise the data for adjacent bands in ISIS. That gets to be tricky to do when you are working in polar coordinates and since the LRO’s orbit rarely passes exactly over the lunar pole.
These radial streaks are not artifacts in the data itself, but instead the radial streaks are the result of not precisely adjusting the strips of LOLA data in ISIS in order to achieve the best fit to adjacent strips. These radial streaks inherently prove that the LRO is in fact in orbit around the moon and that the LRO’s LOLA instrument did in fact acquire this data. How do these radial streaks in the raw data prove this? Because the radial streaks are the result of the LRO being ever so slightly higher or lower in its orbit, and/or being ever so slightly to the left or right in its orbit, as a result of flying over denser or less dense lunar features which slightly perturb the LRO’s orbital altitude and direction. In the Apollo era, areas of the moon which perturbed the orbits of the Lunar Orbiters and the Apollo CSMs and LMs were called MASCONS which is an acronym for mass concentrations.
Many research papers have been published about methods for adjusting light bundles in spacecraft images, as well as for methods to identify and remove erroneous data. It is not an easy thing to do — especially when working with bundles which have very small angles since very small changes to the angles when fitting will result in large changes in elevation.
Comment by GoneToPlaid — March 15, 2016 @ 1:10 pm |
Yes, I could’ve gone into the data processing/gathering a lot more, but … brevity. 🙂 Any LOLA track is a 1D track, just giving you the elevation of the points in that track, so you have to build up a map with thousands of passes going over slightly different areas.
I would still count the misregistration as an “artifact” of the data, though. I didn’t realize it was likely because of mascons, though.
Comment by Stuart Robbins — March 15, 2016 @ 1:25 pm |
Hi Stuart,
Funny how the movie is of Shackleton Crater, one that I spent alot of time with, especially using LOLA raw data, and this crater is on the South Pole! Even the best of us make mistakes from time to time!
I had a lot of fun working with the raw LOLA data. The mappers do alot of averaging over fixed area sizes to fill in the blank spots. LOLA, great though it is and may it keep on working forever, cannot “paint” the whole polar surface with laser light. It they just showed the spots, it would look like they hardly covered any of the surface. Even operating continuously, LOLA has likely only “painted” <2% of the polar surface. It can't be helped, its alot of surface and alot of data. And of course what it hits and reflects back is an average of the surface height over a beam diameter of 5 meters. Some higher, some lower within the spot.
Still it is a neat movie.
James
Comment by J Fincannon — March 22, 2016 @ 12:05 pm |
Also, when I used the raw LOLA data, I found some anomalies which I duly reported. Feedback from the instrument folk state that ” any return with a slope greater than ~37 deg is filtered out” but their autoediting algorithms are not perfect. I found some errors which you can only detect using your eye looking at the data. In those cases, for some reason, the LOLA orbit dataset for a pass is too high or too low compared to the surrounding data points. Feedback shows these are likely due to being near “orbit adjust” so had either bad orbit knowledge or off-nadir pointing. If you created an average map at a large enough resolution, these bad points would disappear through averaging.
Comment by J Fincannon — March 23, 2016 @ 8:02 am |