My idea:
First the wheels of Opportunity had loosen the little rock out of the ground, increasing its wind attack surface.
Second step is the wind that came and blew it down the slope.
Do we have any knowledge about the weight of this piece?
And what about wind speed during the last sols?

If there was a wind so powerful to move and turn over this rock, the solar deck would as clean as new and we would definitely know about it.

Paolo

PS: Happy 10 to all of you.

Sorry, Paolo, but I can't imagine even a storm cleaning the solar cells.
I never had a clean car after I ran the German Autobahn.

Here's an interesting factoid I learned while doing book research: wind more readily lifts sand-sized particles than clay-sized particles, both on Earth and on Mars. It was a mystery until pretty recently how dust was getting lifted off the Martian ground into the atmosphere, when sand didn't seem to move much. Dust devils seen by MGS provided one answer, with their vacuuming effect -- that could lift dust, while straight-line winds couldn't. Then Spirit's MI provided another answer, showing that dust on rocks in Gusev had clumped into aggregates large enough for the wind to effectively pick up before the aggregates poofed apart into their component dust particles. So if you see a surface that's really coated by really fine dust, it's not obvious that even a strong wind will clear that dust off, yet another reason that the fact that cleaning events happen is such a pleasant surprise.

Click to view attachment

Parallel-eye, not cross-eye, for better depth perception.

Yesterday's rear hazcam gives a better view of the divot area. The moved rock seems to have broken into two pieces. The upper piece has a cavity.

Very interesting. It's obvious there's something amiss with the theory. On Mars the dust is clearly more mobile than the sand or else everything would be grey not red.

Once lifted in the atmosphere, clay-sized particles stay there for a very long time, mixed up by the wind again and again and spread all over the planet. The sand falls down immediately. Remember there is no rain to wash out the atmosphere.

Looking at post #86 from mhoward, I noticed a coloured patch on the ground above PI the same as it's 'interior'. Does anyone think there is a link? Maybe this is an impact site where the rock landed (launched by your favourite method) and then rebounded, performing a half sommersault, and then landing in it's present position.
Or am I as mad as a bag of badgers?

That's the brush mark.

You sure?

Strange shaped brush.

Was the inside of the rock brushed as well?

The brush spins, so normally makes a disc-shaped mark. In this case it looks like the plane of the brush was tilted a bit compared to the ground, so only made contact on a portion of the disc (we can see the lower part of the disc).

I think it's pretty unlikely that they'd try to brush a small, loose rock like PI - I doubt the rock would sit still!

I mention brushing the rock because of the similarity in colour. The reason for my post in the first place.

Also, being freshly dug up it has no dust on it so it doesn't need brushing.

We have been told that the interior of the rock is chemically unusual but so far nothing about the bright teeth around the perimeter, or have I missed something?

Because the colour of the inside of the rock is so similar to the 'brush' mark would that indicate that it hasn't been carried that far, or would it put a limit on how far it could have been carried, or an indication of where it may or may not have come from?
And would the place that it came from now be the same colour as the brush mark?

Yes, pretty sure in this case: the mark wasn't there in earlier Pancam images before the RAT was placed on that spot. And it looks like a typical glancing brush mark.

I don't think it has been carried far, just kicked out by a wheel scrunching around in a turn or whatever. Should it match the place where it came from? Yes.

As for whether the corresponding hole in the ground (assuming it exists) exposes the same colors, I believe Steve Squyres said at the Planetary Radio Live 10th Anniversary Celebration that's something they'd like to find out, and they will be moving the rover soon to try to take a look.

Thanks for the reference. I finally had a look at it, and, judging from Fig 4b, the observed cumulative crater frequency is a bit over 2*10^-6 craters/km^2/year. That's for all craters larger than the smallest observed (roughly 1.5 m diameter). So in 10 years, the odds are about 2*10^-5 that a crater would form within one square km of Oppy.

Even if you follow the trend of the models, which predict higher rates than the observations, the odds are perhaps only 10 times larger (hard to be sure, since the model predictions aren't given for the smallest craters). I guess you should also add meteors small enough that they don't produce craters, but the trend of the cumulative curve (4b) is not promising.

Either way, Oppy would be very very very lucky to experience a meteor hit that close.

On the other hand, the odds are good that Oppy has been within very roughly a 130 km radius of a meteor strike in these 10 years...

The Navcam L0 view on Sol 3555.

Jan van Driel

Click to view attachment

This is a quick polar version of Jan's new panorama.

Phil

Click to view attachment

And this, by tilting the projection plane and shrinking radially, puts the area in more of a map-like geometry.

Phil

Click to view attachment

The first of Phil's polar images gives a real sense of the slope Oppy is on. Staring at it in my half awake Monday morning state, I get a real feeling that I may fall off the top of the rover and roll down Solander Point.

xeye of the 'Disturbed Clast' from SOL 3350
Click to view attachment

I don't see much disturbed in 'Green Island' SOL 3352, maybe a scuff at left..
Click to view attachment

After a short bump downhill we finally have a view of what was under the rover:
http://qt.exploratorium.edu/mars/opportuni...0M1.JPG?sol3566
Pinnacle Island is the small rock just above the left edge of the sundial. We can now see a messed-up area farther uphill that looks like a potential source for PI.

Here is the panoramic view stitched together from L0 Navcam images
taken on Sol 3559-3560 and Sol 3564.

Jan van Driel

Click to view attachment


and Sol 3566

Click to view attachment

It's clear from examples like Pinnacle Island, the deliberate crushing of gypsum veins at Cape York, Spirit's serendipitous discovery of silica and other examples that rover wheels are valuable geological tools in their own right, so I have a question for the engineers (including armchair engineers). How might you modify wheel design to maximise this function? Add a thin outstanding flange somewhere in the otherwise wide flattish tread profile perhaps? Have occasional small cup-shaped scoops or spikes at intervals around the wheel perimeter? Extra cameras? Any ideas?

I've always wondered if a rake has ever been considered, one that can be routinely dragged when roving over promising terrain, just enough to turn over some rocks and stir things up for a better idea of what the dust layers are hiding. Extra science could be had by taking advantage of the routine long drives, by dropping a rake at intervals just behind the rover, followed by a some pictures, and dragging would be well within the capabilities of the wheels. In case the device got stuck in the down position, some explosive bolts could get rid of it!

There are probably 50 engineering reasons why that is a bad idea, lol.

The rake idea is pretty much what Spirit did when it was dragging a frozen wheel - digging a trench in the soil (making major discoveries as it did so, salty soils and silica) or moving small rocks. So basically that could be recreated at will by driving five wheels and not driving the sixth, without adding the complications and mass of a rake and the mechanisms needed to raise, lower and eject it.

Phil

There are a number of different configurations you could use for a rake-like implement off of a rover, though. A large flap-like rake would serve to disinter buried rocks and pebbles, at the cost of possibly getting hung up and also of piling up large mounds of dirt. However, a dixie-cup-sized device, open at one end and with a mesh at the other, could be used to collect cobble-sized samples which are shallow-buried for a sample return mission.

For in-situ analysis, you would need to identify a need to disinter buried cobbles as part of your sampling methodology for this kind of device to become important. For example, if you come up with a theory that only buried rocks will display a given desired-to-observe chemistry, you might want to develop a way of disinterring such rocks for easy transport to your sensors.

I get the gut feeling that such a sampling methodology would be more useful for sample caching in preparation for a sample return mission than for in-situ analysis. But in either case, I'd think you would need a compelling reason for wanting access to shallow-buried rocks before getting into the complexities of designing such a raking mechanism.

-the other Doug

A wheel with a partially sandpaper-like surface may be usable for an on-the-fly rock abrasion.

Edit: E.g. round patches on a wheel; rock could be abraded by steering in-place to the left and right.

We have the RAT ( on MER ) and DRT ( on MSL ) for that. Using a wheel for it would be a messy affair.

And with all the notions of a 'rake'....you have to consider what would you take off MSL or MER to make room for it in terms of mass, volume, power..and budget. Such notions are not free...and you have to consider the cost to the vehicle.

The wheels do a great job when required for simple trenching, crude abrasion....and at the same time are useful for getting around. Extra hardware really isn't justified.

I had always wondered what kinds of pressures could be exerted using the arm, for instance to move a rock, perhaps the RAT would be the most appropriate contact point, my thought is that it would seem possible of being useful, though clearly such purpose was not designed in and re-purposing the existing assembly to do so is evidently fraught with unacceptable risk. I assume some thought went into this general concept for MSL, if not MER, though apparently not chosen as a route worth pursuing..

You enter a world of many many unknowns when you start trying to shove things around. Will it move at the toque the arm can handle - will something roll back and put torque on the system etc etc.

http://www-robotics.jpl.nasa.gov/publicati..._2005_final.pdf has some good IDD spec - including typical preload forces on the RAT of around 30N. The IDD is a very fragile thing. The Phoenix arm was a brute of a thing that could really bring some torque to the game, MSL's is about inbetween I believe.

Thanks for finding that. very illuminating of the ingenuity involved in solving the myriad dilemmas and giving us 5mm positioning accuracy.
a favorite highlight (p4):

"The mass allocation for the IDD of 4.4 kg, including launch restraints and cabling, was also an extreme design driver. The cabling mass alone, mostly to support instruments, was almost 1 kg. The arm had to support a turret of instruments with a mass of 2 kg, provide preload for the RAT, and be capable of achieving the absolute and repeatable positioning requirements described in Section 2.
The mass allocated and load magnifications possible with the 5 DOF robot arm did not allow the structure and joints to be designed in such a way that the arm could not damage itself in all possible operational cases. Instead, an approach which integrated flight software fault protection, operational constraints, and mechanical capability was developed that reduced the risk of damaging the IDD during normal operations. Joint flexibility became particularly important since arm configurations that cause dangerous load magnifications also allow for significant travel past the point of instrument contact before the loads reach their critical stage. A redundant method of terminating IDD motion based on contact sensor feedback and tight limits on over travel past the expected contact point keep the arm in the safe load zone. While the torque used to get the joints started is capable of creating unsafe loads, software quickly ramps the current limit down after movement starts before more movement occurs than arm flexibility can accommodate. Software also limits the torque of the motors when the arm is in configurations that can magnify loads."

and finally a stereo pancam view from sol 3567
Click to view attachment

Click to view attachment

Thanks! That explains how the rover might kick a rock over some distance. But could the rover motion also produce a fresh crater like the one (above image) at Stewart Island?

That doesn't seem to fit any definition of crater. Seen in context in atomoid's much appreciated x-eye it is pretty clear that the wheels gave that rock a fair working over, pushing the regolith into a mini berm.

Are you saying the wheel rotated the rock, causing it to carve out a bowl-shaped depression? Yes, that would explain it.

I don't think anyone was expecting to see a 'crater' ( this location was, after all, directly under the rover when Pinnacle Island showed up so you would have needed a neat hole straight thru the rover to cause it ) - the 'crater' hypothesis was such that somewhere nearby ( 10's of meters, hundreds of meters, km's away perhaps ) a crater was formed and PI was a piece of ejecta from that.

However - I think we have a smoking gun in the form of wheel/rock/digup/tiddlywink interaction.

sol 3568 same disturbed area, slightly different angle I guess
Click to view attachment

The Navcam L0 panoramic view on Sol 3568.

Jan van Driel

Click to view attachment

Mi Cam view on Sol 3569.

Jan van Driel

Click to view attachment

Thanks, Jan. I've made your panorama into a circular projection to help show where we are relative to the outcrop.

Phil

Click to view attachment

'Stuart Island', Opportunity Sol 3571 (L257)

wonder if Oppy cracked that rock driving over it.
little piles of soil dumped at intervals suggest PI could have been picked up along with soil to exit during one such dump and roll further downslope?
Click to view attachment

Looks to me as though Pinnacle Island would fit into the missing piece at the top left of Stuart Island.

The complete Navcam L0 view on Sol 3568.

Jan van Driel

Click to view attachment

Nice pan, Jan!

This is the same pan in a circular format.

Phil

Click to view attachment

There we go. PI mystery solved. http://www.jpl.nasa.gov/news/news.php?release=2014-051