For anyone following @curiositylog and excited about the reported 125 meter drive on sol 356, it looks like there's a glitch in the SPICE data files. Here are the last few records of the track sampled every 100 seconds:



So the longitude jumps from 137.436561 to 137.437050 and back again at the end of the track.

I changed the code to filter out impossible rover speeds, so now the track log is cleaner and the sol 356 distance is 50.82 meters.

Well done, Joe... I just updated my odometry file/plots. I see you changed also drive numbers, suppressing previous #44 which was a 0-meters shift!

I was merging some of the Sol 352 ML frames and Aeolis Mons looks really beautiful

Good work getting that detail out.

BTW, what is the deal with the MARDI shots on this trek? They've been taking a MARDI shot on every drive sol, always around the same time, about a half hour after sunset (~7pm).

The MARDI shots are a way to survey the diversity of the pebbles on the ground. The straight-down view makes it easy to get precise measurements on their sizes and shapes. Jim Garvin is doing this work; here's his LPSC abstract.

My take on the Sol-352 M34 Pan sofar.

Related update from Ken http://astrogeology.usgs.gov/news?p=2&...f-twilight-wins

It's fascinating how a patch of the mountainside appears to disappear (arrowed) at certain times of day, when the lighting is just right:

I guess with just the right angle of illumination, the surface brightness of that patch matches almost perfectly the surface brightness of the adjacent sky (at least integrated over the hazcam spectral response), so it looks as if we're seeing through a "gap" in the mountainside. It reminds me of some kind of atmospheric "mirage" effect, though I'd guess that's unlikely with the low air density on Mars.

Here are the original frames:
http://mars.jpl.nasa.gov/msl-raw-images/pr...FHAZ00304M_.JPG
http://mars.jpl.nasa.gov/msl-raw-images/pr...FHAZ00323M_.JPG
And another one which shows the "mirage" on a different sol, presumably near the same local time:
http://mars.jpl.nasa.gov/msl-raw-images/pr...FHAZ00323M_.JPG
I haven't been able to find navcam or mastcam imagery showing the "mirage", but I'd love to see some...

That is very weird, and the same part of the mountain, too, in those two images in which it appears. On the "flicker-GIF," were the two images taken nearly around the same time?

Clearly not - just look at the shadows in the imagery - Fred has give you links to it.

Thanks, Emily & wildespace, that really explains what they are doing with the MARDI shots. Who knew that Mars pebble size follows a log-hyperbolic distribution?

Per the abstract, Dr. Garvin found the preliminary data to be most consistent with volcanic ejecta.

My take on 356:

http://www.flickr.com/photos/84750994@N05/...269592/sizes/o/

Didn't expect the hazcam compression to average-away the ridge quite like that, but i do remember seeing many instances of disappearing horizons in the MER data sort of like this but much worse (some even had jagged serrations at the clipping point). perhaps they have improved those algorithms but this is still an an edge case exploiting a sweet-spot, helpful specs in section 2.7 though i haven't quite digested..

We've definitely seen examples with MER where the sky is overexposed and it "bleeds" across the horizon because of how the CCD works. But that's not what we're seeing here - the sky is nowhere near overexposed. I think the mountainside in that direction really does happen to match the brightness of the sky fairly closely, at the right time of day. But navcams or mastcams would settle it.

PS - haven't noticed this sort of white layer since back around Yellowknife Bay:
http://mars.jpl.nasa.gov/msl-raw-images/ms...4000E1_DXXX.jpg

I recall some Phoenix albedo pans where the terrain blended into the sky without a horizon line, due to a radiance match. That's not possible with the color cameras, or in color pans in general, due to the sky being bluer. But that's not as funky as this.