It's been pretty slow in the Spirit forum recently, but hopefully this will reawaken it.

A new thin dark streak is visible in the latest sol 961 rear hazcam, that wasn't visible in the previous shot, from sol 924. It's visible in both L and R hazcam views. It looks remarkably like a dust devil streak, but the scale is very small. Here's the R hazcam from sol 894 subtracted from the 961 image:
Click to view attachment
The dark streak cuts across more or less from left to right about a third of the frame down from the top. Other differences are visible, mostly around rocks, due to the change in sun angle between the frames.

To demonstrate that the dark streak isn't just an effect due to that change in sun angle, look at these three frames:
sol 856, local time 11:07
sol 961, local time 11:21
sol 894, local time 11:45

You can see that the shadows move progressively from the sol 856 to 961 to 894 frames above, as you'd expect from the local times they were taken. But if you flip between the three frames, you can see the dark streak only in the middle (sol 961) frame. If the streak were due to changing sun angle it would be at least as strongly visible in the 856 or 894 frame. Therefore it's not due to sun angle.

As I said, this is way too small for a dust devil streak. And, I don't think we're anywhere near dust devil season. My guess about the origin of this is a small "crumble event" on one of those dark rocks in the upper left, followed by the normal winds blowing some dark dust downwind.

Anyone know if the streak direction is consistent with what the winds should be doing now? Has anyone else noticed changes near Spirit?

Look at the change in the shadow cast by the aft solar array on the ground - that's quite a lot of solar-angle change - and looking at the three frames - the change you identify would seem to be consistant with a change of illumination more than anything else.

I have - however - been looking hard for changes in lots of images and I've not found anything yet. What WILL be interesting is the changes in tiny ripples upon which the rover is now sat.

Doug

Interesting. An additional new picture would provide a definite answer. I see your point Doug but it seems to contiguous and localised to just be shadows. It would be great if it did turn out to be a streak.

Doug, the local time difference between the sol 856 and 961 frames is 14 minutes, which corresponds to a solar angle change of just 3.5 degrees. For sols 961 and 894, the difference is 24 minutes, or 6 degrees. The change in solar panal shadows looks quite large because of the height of the panels above the ground and the closeness of the panel shadows to the hazcam. 3.5 degrees is not a lot of change!

But the strongest argument that the dark streak is not the result of the change in sun angle is in my third from last paragraph in my post above! Any such putative darkening of a patch of soil due to changing solar angle must be monotonic with that angle over such a small range of angles, whereas my three frames show that it clearly is not.

Notice also that the dark streak crosses areas of ground that are at different inclinations, due to ripples etc. Why would differently oriented patches of surface respond to changing solar angle in the same way? Look for example at the largest rock closest to the rover (pixel location 680x375 or so). There are large variations in surface inclination angle in the soil surrounding it, but these do not show up in my difference image - the entire area (except for the rock's shadows) is very close to uniform on the difference image.

Also we don't see any other such darkened (or lightened) patches of soil elsewhere. Just one curiously linear darkened patch. Finally, the dark streak is darkest at the left, near my putative source, and thins to the right.

Altogether, an argument that the streak is an illumination feature necessarily takes on a conspiratorial character: differently oriented patches of soil, coincidentally lined up in a streak, responding in a bizarre non-monotonic way to small changes in solar angle.

Edit: that's conspiratorial in the sense of "contrived", of course no-one's conspiring here!

Actually there is another simple explanation: the feature is the shadow of something off the frame to the upper right. But that's even more conspiratorial!

Oh crikey - I was looking at something else.....you may have spotted something I was looking in the 800,280 sort of area.... I'll see what I can come up with for anim-difference images

OK - the anim is a work in progress- but this is the difference between the 961 image, and the average of the preceeding three images. Bloody good 'spot' there Fred - nicely found. I'm going to 'phone it in' as it were.

Doug

Actually the change in solar elevation angles are even less than you indicate because of the change in noon solar elevation as she passed through mid winter. Also there is also a change in solar azimuth that might have been relevant had the solar elevation angle been low enough to cast shadows from landscape features off to the left. It isn't and as you point out anything that could cause such a shadow would be really conspiratorial.

Sol 856 LT 11:07 - Elevation 53.5deg Az 3.9deg
Sol 961 LT 11:21 - Elevation 50.7deg Az 351.7deg
Sol 894 LT 11:45 - Elevation 49.5deg Az 347.0deg

I cannot see how you would get an area of shadow across the entire region in question that would only appear on Sol 961. We still need another photo to be sure but it does look like a clear change to me.

I guess I'm missing something -- except for the minor sun angle changes, I can't see what you're talking about in the three image set. I see what appears to be a dark groove in the very-low-sun-angle image, but I can't identify it in any of the three images. Or any albedo change in them.

Maybe my eyes are just getting too old for this...

-the other Doug

The attachment that Fred did - and the one I did - are basically the difference between the Sol 961 image, and the other RHAZ images. So that dark streak is something that has changed between the last RHAZ image, and that Sol 961 one. It reminds me of the sort of trail left behind after a DD in Mark's enhanced DD movies.

Doug

Not joking (realy). How far is this? Look prety close though. Any notice of a side effect on power supply (clean up)?

I'm not sure to be honest - probably best to match these features Nav or Pancam and work the range from that.

Perhaps - if we're lucky - they will do another Navcam pan from this position and then we can compare with earlier Navcam pans and vertically project it to see the track.

Doug

Dvandorn and anyone having trouble seeing this in the original images, here's a 3-frame gif that should help. Frame one is sol 894, unprocessed. Frame two is sol 961, unprocessed. Frame three is frame 2 minus frame 1, and then brightened:
Click to view attachment
Look for the (subtle!) change between the first two frames at the position of the streak in the difference image (frame 3). It's subtle but I was able to spot it before I made the difference image.

As I said above, I doubt this is a devil. I think small "crumble event" followed by ordinary winds is our best bet. I think we expect some level of steadyish winds all winter?

Not yet, power has just begun to climb again from the 280whr/sol low point it has been at for the past month or so to 287whr/sol on or about sol 957/958. We've had very regular power reports from Spirit over the past six weeks and there has been no evidence at all of any cleaning (yet).
We'll see if the next power figures show any change but I wouldn't expect it - I think it is very unlikely that this just happened so whatever effect it would have had would already have been seen.

Fred, your observation is interesting but what disappoint me is that this streak coincide exactly with regions with higher slope (even if with different sign comparing left portion with right). Based on this, I tend to think is a consequence of small illumination differences and/or deposition of some dust layer with slightly different light scattering properties... (not necessarily localized only in this point as for a dust devil track).

Dilo, the ground here has regions with greatly differing inclinations. Recall what I said in my second post about the area around the big rock. Why don't we see these kind of changes anywhere else?

Obviously a new image (after sol 961) would be great to have. But we've waited over a month since the previous rear hazcam image! Hopefully we won't have to wait as long for the next one...

That's what I thought dilo - but the change is real imho.

Doug

ok, let's wait for next images...

I identified the putative source region of the dark streak in pancam imagery. Since the streak is darkest on the left end, I'm guessing that's near the source. That end is between the red, green, blue, and yellow marked rocks in this crop of the sol 961 hazcam:
Click to view attachment
Here's a crop from James Canvin's (stunning!) McMurdo pan, where I've identified and marked those same four rocks:
Click to view attachment
I've circled the area corresponding to the left end of the streak.

It's pretty clear what may have happened here: a piece of the very thinly layered outcrop fell down onto the sand below, kicking up loose dust. There are a number of potential loose pieces, such as the one I've arrowed.

Boy, would I love to see some new pancam imagery of this area!

Yes -- with the flicker gif I can make out the change in albedo. I kept looking in the area of the streak that is apparent in the subtraction image, but by opening each image separately and just moving from one browser window to another, I honestly couldn't see the albedo change. With the flicker gif, I can see the actual albedo change quite clearly, even without the third subtraction frame.

-the other Doug

Nice hypothesis, Fred.
I was looking for other Pancam images of this region, when I noted a huge DD track not previously visible in the eastern Eldorado!
Original imagese were taken with PanCam L2 filter on Sol 924 and 961 (37 Sols apart), with almost exactly the same pointing and illumination:
Click to view attachment Click to view attachment
(start images are 2P208392405ESFAS00P2411L2M1.JPG and 2P211675276ESFAS00P2411L2M1.JPG).
The track is already evident, anyway I made subtraction and used it also for a elaboration image showing the change over the first image through the hue/sat change:
Click to view attachment Click to view attachment

This story is encouraging, because we aren't yet in the DD season and, with some luck, one of these precursor could hit Spirit soon, making possible anticipated return to full rover activities...

That's something !
I note that, pending the hypothesis of a DD is right, the direction it took would have bring it little bit South of Spirit but not that far. Since the DD desapears in the depression on the other side of the ridge we loose its track. But if it has continued, it should be visible on some images taken in the South direction. Do you agree on this? Do we have such images ?

Dunno... My impression is that dust devil had effect only on Eldorado dunes, terrain outside wasn't disturbed. If I'm correct, based on prevailing wind direction in old DD movies, devil's direction was toward South (toward Spirit) and the intriguing fact is that terrain in the foreground (the Homeplate, essentially) changed slightly overall hue in the last picture, suggesting a similar (weaker) darkening occurred!
However, if DD invested Spirit, I imagine we should have more whrs now (and a lot less dust on solar panels)...

Climber & Dilo,

Be careful!
Spirit + DD ==> cleanup-event = Doug jumping off its seat.

This is not necessarily a dust devil in the usual sense but simply a turbulent swirl in the atmosphere as the wind rounds the Husband Hill obstruction.

--Bill

I know we are walking on eggs here (dunno if the expression works in english, actually, it's a french one )
1-it's not because it didn't happen before that it'll not happen
2-I took some precautions before on a post here above
3-we all know it's a remote possibility
4-the first one to spot such an event will win THE gran price

PS : I'd better like dust jumping off solar panels that Doug out off his seat (don't like " unecessary post deleted")

Edit : you're right Bill, a kind of dd instead of DD.

I don't want to seem like a spoil sport but I feel the need to reiterate that there has been absolutely no evidence of any dust removal activity at all recently from Spirit's solar panels. Zero, nada, zip. Power output has followed the expected mid winter insolation curve with no additional dust deposition (which is good) but no dust removal either over the past six weeks or so. That was up to Sol 957/958. The next power update might change that but for now any speculation about cleaning is a bit premature.

Opportunity has scored a noticable (~20%) power benefit from something recently, presumably a cleaning event or events but that is not yet clear.

Nice find, Dilo!

I have to wonder like Bill Harris if this isn't a dust devil proper. The thing is, we know that there's something unique about the topography associated with Eldorado, since we have the rippled sands there. Could we have captured here a step in the normal buildup of Eldorado?

It's also interesting that the track ends (or starts!) at the edge of Eldorado.

But still, remember we're seeing Eldorado quite foreshortened here, so the track is actually quite long and thin (and probably longer still - we can't see the other end of it). That to this non-expert sounds like a devil or something very close.

Returning to this places makes me shiver...
dilo, do you remember this images?

It looks to be (in the image above) the same area you are making reference to, seems like this is a previliged DD corridor...

The darkening in El Dorado isn't very long -- it likely was caused not be a dust devil, but by a seasonal change in the direction and speed of the local winds. At least, that's what I would suggest, barring any evidence that dust devils actually occur during mid-winter here in Gusev.

-the other Doug

Eh, Ustrax, not only I remember these images... in fact, I was inspired by them in searching changes inside Eldorado! I remember also our wild theories about these features (cracks in a lake made of lava or something else...).
About other Doug's explaination, the absence of any trace outside the dune field is suspect, but also a seasonal wind change impacting a so delimited region is a little hard to accept, IMHO...

Is there a second image to confirm the darkening? This seems *very* early in the season for there to be dust devils (the shape of the ground track seems wrong too). Strong gusts could be responsible, but I wanted to throw out another possibility.

Given that Opportunity has recently seen some clouds, could this darkening be a cloud shadow projected on the ground?

edit: So, the changes are real and lasting, as evidenced by this L1 image taken on sol 969.

Tosol we have a pancam of this area (360 L1 albedo pan) - the little bit you point to is just off the bottom but the rest of the outcrop looks undamaged (by eye).

Personally, I don't buy the 'bit falling off' theory I think it's much more likely that as a small gust of wind blew over the ledge and the vortex created in its wake produced a 'mini dust devil'.

http://nasa.exploratorium.edu/mars/spirit/...00P2269L1M1.JPG

Another streak has appeared at El Dorado, also a couple of narrow streaks to the left on the hillside (arrowed) look like they were caused by rocks rolling down the side of the hill.
Click to view attachment

The newest streak is darker than the previous, and much wider and longer.

Here's a 3 frame animation, comparing the L2 images from sol 924, L2 from sol 961 and L1 from 969 showing the 2 new dark streaks:

Very good, Alan/Slinted!
I noticed too a couple hours ago, but this time you beat me in posting it!
Not only this streake is darker and longer... it is clearly visible outside the field of dunes, ant this reinforce idea that dust devil season started, at least in this point! probably, the South side of Husband hill is a good place for local atmospheric warming which cause this phenomenon...
Correct me if I'm wrong, but Spirit didn't yet activated the new navigation software implementing the smarter DD catcher. If so, at this point they should do asap!

Fine work catching and displaying those streaks, dilo, alan, and slinted! I think I've said it before, but the collaboration that takes place in the UMSF community is becoming legendary, and this is yet another wonderful example. That animated gif is sweet.

I dunno -- I'm not necessarily convinced that you have to invoke a dust devil to get these changes. Surely an unusual set of straight-line winds could have caused the same type of track? I mean, wouldn't it all depend on where, when and how straight-line winds blow as to what effect it might have on the ground? And this *is* the leeward, "protected" side of Husband Hill, which has allowed the dust to settle into El Dorado in the first place... you'd think that any straight-line wind events would be fairly rare, and thus a lot more noticeable here than elsewhere around the hills.

It just seems awfully soon past the winter solstice to be seeing actual dust devils. For all of me, I could be wrong, though.

-the other Doug

Hopefully this means that we'll start seeing some stronger winds, which might kindly come by to clean up Spirit.

You certainly may be right about this being just due to wind - the recent streaks on Eldorado support that.

What intrigued me about this from the start was the narrowness of the dark streak I found. That's what led me to the localized "crumble event" idea. It's still possible there's something just below the sol 969 pancam frame.

However, I did a difference image between the sol 969 L1 and sol 859 L2 images (not the same bandpass, of course):
Click to view attachment
You can make out the left end of my dark streak, in the lower right part of the frame, just to the right of the jaggedy outcrop. With this high resolution, there's clearly no single point source for the darkening. Also, there appears to be more darkening farther away, in areas not visible to the hazcam. This all starts to sound like wind gusts affected by small scale local topography.

Has anyone 'in the knows' commented on these changes? And what may have caused them?

I haven't noticed any official statement indicating any cleaning but the last reported power level , 296 whr sometime between Sol 969 and 976 is 10-15whr higher than I'd expected and is the second notable up tick in power in a fortnight following two months of flat power output. That is well within the error in my estimates so you can't read too much into and the most likely explanation is simply clearer skies but it might indicate some cleaning is actually happening.

Another streak on El Dorado this time at the south (left) edge.
http://qt.exploratorium.edu/mars/spirit/pa...00P2411L2M1.JPG
I think they show up better at El Dorado because the sand is bluer and provides more contrast when the dust is removed. How many more are we not seeing in the areas where the contrast is subtler?

Yes, Alan... I noticed too! Look to this animation from Sol 969 and 980 (filter is not the same, L1 vs L2, but this do not introduce big difference here):
Click to view attachment (...hey, Spirit changed orientation! )
Even tough the streak direction doesn't seems toward the rover position, could be related to the power level increase highlighted by Helvick...?

Damn, when did that happen? I haven't seen any recent hazcams.

It didn't move, at least by commanding; current site/drive id is still AS00.
The reason for that effect might be related to a different pancam heading on the two shots. Remember that the image from sol 980 is not a full frame (EFF) but a sub-frame (ESF).

Edited: I cheched the camera pointings an the offset between both of them is about 3º. On the other hand, the rotation which can be seen on Dilo's animated gif is about 0.5º and it's too big to be related to the 3 degrees offset. Spirit should be on a quite steep side-slope in order to account this image rotation to a change in camera heading. Therefore, the "mistery" is still alive.

Edited again: I searched for previous shots like the one taken on sol 980 and found that on sol 924 they commanded a sequence named "pancam_ElDoradoChange_L27". This image was downlinked, among others, and its pointing info is: azimuth=-23.5035deg, elevation=0.0875386. On sol 980 they commanded the same sequence again. This other image was downlinked, among others, and its pointing info is: azimuth=-23.5013deg, elevation=0.048464. Those two pictures were taken pointing to virtually the same position and if you open the following animated gif the new streaks can be seen and a small displacement between the two shots, but no rotation at all.
Man! I changed my mind three times in the last few hours.

Click to view attachment

Edited once again and hopefully for the last time! I calculated the pancam mast tilt (deviation from the absolute vertical) hence the rover's tilt needed to create a 0.5 degree image rotation effect as a consequence of a 3 degree pointing offset. If I did my math correctly the rover should be on a 10 degree east-west slope to account for that effect. Spirit is on a north-facing slope, but does this slope has an east-west component too?

Here's an enhanced-by-difference-image version of the changes from sol 924 and the most recent images on sol 980



The movie showing all 4 panoramas taken over the last 60 sols is available here

Great movie, slinted! looking to it, I have impression that events intensity is going to increase...
At this point, I think they should absolutely install the new smarter DD-catach SW because spectacle should be fantastic!

An attempt to understand if a DD really hit Spirit is this movie of the sundial in the last 40 days:
Click to view attachment
I do not see any significative change in the dust level, however... . And you?

I've been checking the sundial images myself since the changes started appearing at El Dorado. I haven't seen anything yet either. Nice movie, I like watching the shadows of the rock behind the sundial dance.

Movement at Low Ridge Haven but now on the solar deck

of Spirit. Not only the shadow but look at the wire in front

of the mast.

jvandriel

That thing's been ratting around that little pocket for aaagess.. (well, more than 1000 sols anyway) I think RM mentioned it in another thread at one point.

Doug