This is similar to an amateur astronomer "seeing" the Encke Gap in Saturn's rings with a telescope who's theoretical resolution limit is greater than the ~0.2 arc second width of the gap. It's the contrast!

Perhaps a more relevant comparison is resolution of the individual wheel tracks of the rovers by HiRISE, though in that case the wheels are almost a pixel wide. Or the detection of the rover tracks by MOC, who's pixels spanned the wheel spacing plus change.

best,
-Tim.

A first preliminary attempt at 'fingerprint-scanning' for hidden craters in the dunefields, craters (both visible and 'filled in') are marked red. Ranger and Surveyor crater are correctly found, as are some others. Smaller craters are not marked as they were outside the limits set for this search (I can define a minimum and maximum diameter to search for). Still some craters are missed which should have been found and there are a lot of seeminly 'false positives', but this remains 'work in progress'...

Click to view attachment

I'm mostly trying to find signs of 'dusttraps', craters which are filled in by sanddunes and mostly hidden from view, but it remains a nasty puzzle to get the software to find the exact signs for that ;-). Still, the dune-fields are a nice testbed, in flat area's the software now correctly finds almost all craters but inside these dunefields it's a lot harder...

Regards,

Geert

Combined view of terrain model and crater-search around the present position

Click to view attachment

Green is normal driving area. Rough sand dunes as well as craters and potential sand-traps (craters hidden by dunes) turn red. Bedrock remains unaffected unless blocked by potential crater. Starts to look 'interesting' but it remains preliminary, no guarantees given ;-).

Regards,

Geert

It has been remarked in this forum already that the area we are now entering looks like a old crater with 350-400 mtr diameter, completely filled in with sand.

Actually looking at the terrain ahead there seem to be quite a lot more of these structures, or at least there are 'round holes' which do not seem to have any exposed bedrock inside of them while they are surrounded by patches of bedrock.

Click to view attachment

Above image are just a few examples of area's which seem to comply to this description, don't know whether I'm seeing ghosts or whether some of them are really old craters?

Click to view attachment

In above overlay of IR relative brightness only a few of these 'craters' seem to show up but basically the sand inside does not seem to differ from the sand outside of these patches.

Click to view attachment

Looking at the relative brightness of the HiRISE image itself (visible red light) the result is even less pronounced then in IR and once again seemingly there is no difference between sand inside and outside of these area's.

Click to view attachment

On the variance model these areas seem to mostly show less dune ripples then their surroundings, but this might be influenced by the fact that there is no exposed bedrock inside these areas (which influences the variance model).

Don't know whether we are seeing here shades of a much older landscape (later covered by the sand)? Anyway, I don't seem to find any indication that the sand itself (=drivability) inside these area's is any different from the sand everywhere else, apart from maybe slightly less dunes.

Regards,

Geert.

Yes, but some of us don't really read this discussion because we find it drier than,... well, Mars.

Ah, but we're searching for the water!

Regards,

Geert.

Drive Map

I was looking at the latest map. It looks like a rough circular area without any exposed bedrock. And ancient and very eroded crater?

I am thinking Opportunity should avoid it and head southeast.

It looks like you guys beat me to that conclusion.

Click to view attachment

Terrain model for Sol 1795, looks like mostly smooth driving ahead with only one suspect area abt 120 mtr almost due south.

Regards,

Geert

Terrain model for area south of Sol 1797 position at 1 mtr/pixel

Click to view attachment

This is a combination of two mathematical filters, one filter is mostly based on relative variance and marks "dangerous" dunes and craters, the other filter attempts to mark area's with potentially higher then usual dust concentrations ("dusttraps").

Excessive Quoting culled - Mod.
excuse me, but how you know the age of craters
Then how old is the Victoria?

I think the assumptions are based on how the craters look.

A crater with big chunks of ejecta, and smooth walls (like Heimdall crater) are assumed to be geologically "recent".

A crater like Victoria, where the ejecta blocks have been wind ablated down to sand/blueberry piles and the rim has been crenellated as the wind-driven sands eroded it's way into the joints, is going to be older.

And craters where the rims have been paved flat and almost completely filled in are assumed to be much much older.

I'm guessing that crater counts and comparison of morphologies were used to estimate absolute ages.
(Anybody know Victoria's estimated age? Was it 3 Gyr to 1 Gyr?)

-Mike

In the next few days Oppy will get into terrain that will challenge the predictions of the current UMSF Terrain models:
Click to view attachment

In about a week with the current heading, Oppy will hit a patch that most of the models predict as "scary", while the Canvin FT model predicts as only "not-so-scary".

Depending on where/if the rover drivers take a turn to the E, there is another patch that the Canvin model predicts as "scary", but the other models predict as "not-so-scary" that Oppy might pass through.

Stay tuned...

-Mike

Perhaps Rui/Ustrax is right and we will be heading towards "Porcupine".

The area you mention is mostly the 'Porcupine' area itself.
With my 'updated' terrain model I agree with Canvin, presently I don't see much problems up till Porcupine, the misery starts further south of Porcupine where there is very rough terrain which in my humble opinion should be avoided by turning more to the east.

Click to view attachment
Click to view attachment

'Dune-filter' for area until Porcupine, there are some 'red' patches but they can easily be avoided. Both images overlap slightly for a complete overview.

Click to view attachment
Click to view attachment

'Dust-filter' for same area, marking area's with possible dust-traps in red. Once again patches can easily be avoided by staying on the yellow brick road of the bedrock. Interior of Porcupine should be avoided.

Click to view attachment

This CRISM IR-overlay of the complete area shows nicely the 'bad' area starting south of Porcupine, if you look in IASViewer or Google Mars it's really rough over there, better keep away.

Personally I expect Oppy to follow the 'yellow brick road' proposed by Ustrax as soon as she is back on bedrock at least until Porcupine, after Porcupine she will need to turn east and pass between the 'red' area's shown on the IR overlay.

From what I have seen it looks like general strategy is to stay on bedrock as much as possible, and pass sandy/dune area's in a southbound heading when they can not be avoided.

Ahead the terrain might be clearing a bit, but we are not yet on 'parking lot' terrain and it will worsen again further on, it's still a long way to go before we are really out of the dune-area's.

Regards,

Geert.

Click to view attachment

Terrain-map for sol 1803.

There are a few craters and a some blue (potentially problematic) area's ahead but mostly good driving.

Scorecard for Sol 1786:
Click to view attachment

For the UMSF Terrain Models, here are 10 m x 10 m normalized grayscale pixel averages for Oppy's 1786 position:
HiRise (raw): 113.77
HiRise (normalized): 128.55
Malaska 20081003 normalized grayscale Terrain Model: - (stupid modelization artifact )
Canvin 20081001 normalized grayscale Terrain Model: 142.80
Ontanaya 20081001 normalized grayscale Terrain Model: 140.25
Butler 20081006 normalized grayscale Terrain Model: 103.30
Sassen 20081031 normalized grayscale Terrain Model: 142.61

Scorecard for Sol 1791:
Click to view attachment
For the UMSF Terrain Models, here are 10 m x 10 m normalized grayscale pixel averages for Oppy's 1791 position:
HiRise (raw): 118.59
HiRise (normalized): 157.03
Malaska 20081003 normalized grayscale Terrain Model: 198.53
Canvin 20081001 normalized grayscale Terrain Model: 145.13
Ontanaya 20081001 normalized grayscale Terrain Model: 140.33
Butler 20081006 normalized grayscale Terrain Model: 105.13
Sassen 20081031 normalized grayscale Terrain Model: 144.02

Scorecard for Sol 1793:
Click to view attachment
For the UMSF Terrain Models, here are 10 m x 10 m normalized grayscale pixel averages for Oppy's 1793 position:
HiRise (raw): 112.47
HiRise (normalized): 121.36
Malaska 20081003 normalized grayscale Terrain Model: 146.13
Canvin 20081001 normalized grayscale Terrain Model: 102.30
Ontanaya 20081001 normalized grayscale Terrain Model: 142.38
Butler 20081006 normalized grayscale Terrain Model: 96.50
Sassen 20081031 normalized grayscale Terrain Model: 129.09

Scorecard for Sol 1795:
Click to view attachment
For the UMSF Terrain Models, here are 10 m x 10 m normalized grayscale pixel averages for Oppy's 1795 position:
HiRise (raw): 111.05
HiRise (normalized): 112.94
Malaska 20081003 normalized grayscale Terrain Model: 147.11
Canvin 20081001 normalized grayscale Terrain Model: 128.17
Ontanaya 20081001 normalized grayscale Terrain Model: 147.33
Butler 20081006 normalized grayscale Terrain Model: 97.97
Sassen 20081031 normalized grayscale Terrain Model: 110.45

Scorecard for Sol 1796:
Click to view attachment
For the UMSF Terrain Models, here are 10 m x 10 m normalized grayscale pixel averages for Oppy's 1796 position:
HiRise (raw): 109.45
HiRise (normalized): 103.58
Malaska 20081003 normalized grayscale Terrain Model: 151.89
Canvin 20081001 normalized grayscale Terrain Model: 125.03
Ontanaya 20081001 normalized grayscale Terrain Model: 122.30
Butler 20081006 normalized grayscale Terrain Model: 91.66
Sassen 20081031 normalized grayscale Terrain Model: 107.25

Scorecard for Sol 1797:
Click to view attachment

For the UMSF Terrain Models, here are 10 m x 10 m normalized grayscale pixel averages for Oppy's 1797 position:
HiRise (raw): 114.74
HiRise (normalized): 135.28
Malaska 20081003 normalized grayscale Terrain Model: 173.10
Canvin 20081001 normalized grayscale Terrain Model: 146.99
Ontanaya 20081001 normalized grayscale Terrain Model: 120.73
Butler 20081006 normalized grayscale Terrain Model: 127.13
Sassen 20081031 normalized grayscale Terrain Model: 134.00

Graph of Sol1784 to Sol1797 observables (as we cruise into easier, but softer, terrain):
Click to view attachment

Here is an EXCEL file covering the trip from Victoria Crater to Sol 1797: Click to view attachment

Thanks for the Excel sheet, Mike, I'm still busy trying to correlate wheel-resistance to measured values from orbital images, in order to try to predict wheel resistance on the remaining route.

Click to view attachment

This is the graph my software gives for terrain till sol 1813. THEMIS Night IR complies quite nicely with the band of dunes which has been crossed, Variance and Variation as measured from HiRISE more or less follow (note these are only values at the 'end of sol' positions, so there might be higher values in between two positions which are not plotted in the graph). Variance and variation is for a 12 X 12 mtr grid.

Click to view attachment

Chemical composition values as measured by CRISM, note that these are also influenced by IR brightness so should be treated with caution.

Regards,

Geert

Same values at same scale as above, now calculated for the remaining route till Mini Endurance.

Click to view attachment

The peak in CRISM IR near waypoint 40 can probably be explained as the route is running outside the CRISM coverage. However, the rise in THEMIS Night IR starting at waypoint 46 is somewhat weird as there is no notable rise in dunes or terrain roughness. This might indicate a change in soil-composition/thermal properties. Not sure what to think of it.

Click to view attachment

Chemical components as measured by CRISM, note after waypoint 38 the route is leaving the CRISM coverage.

Below route to Mini Endurance which I used for above graph's. This route more or less follows the route proposed by Rui/Ustrax. Presently oppy still seems to be heading for Porcupine conform this route, although it is unsure whether they will pass this old crater on the east or on the west, both routes are possible although the west route seems slightly better (alternative would be to pass straight through Porcupine, but no idear what the soil would do inside the crater).

Click to view attachment

Click to view attachment

Click to view attachment

Regards,

Geert.

This is (at least) a bit premature, but when we get to the rim of Endeavour, are there expected to be problems with crossing the crater-rim hills? Oppy will be pretty old by that point...

I guess a lot would depend on how many wheels are still operating if/when she reaches Endeavour. Below is a 'quick' analysis of some of the hills on the crater rim, basically the terrain around the hills looks nice and flat and there are plenty 'holes' in the hills to pass inside the crater, but the hills themselves look quite rough.

Click to view attachment

Personally I think the area just on the foot of the hills looks a bit like an old shoreline, if this is true that might be the most interesting place to study and it might be within reach.

Regards,

Geert

Nice!

I'm interested in the Themis bump near waypoint 15-16, to the S of Porcupine Crater. I wonder if that is due to the brigthness (or signature) of the bedrock exposure. It looks similar to the spike near Conjunction Junction (Sol1710).

That also looks like the last major piece of bedrock for a while as the terrain smooths out to flatter sand on the proposed route.

-Mike

Mike, I have been quite busy these last weeks in trying to find correlation functions between your wheel-resistance figures (as analyzed from the images of wheelrim, starpatern, etc) and the various data from orbital satellites (HiRISE, CRISM, THEMIS). Previously I worked with 'filters' which more or less did the same but I now made the software 'smart' enough to start looking for patterns itself and match terrain to your data-points.

Although I can create images for terrain roughness also using this method, it works best for wheel resistance and soil-type.

Click to view attachment

Above image is the result of the 'soil-analyses' for the area immediately surrounding Victoria, carried out by comparing your data on wheel resistance with the orbital data. The whole thing still needs more fine-tuning, but it is starting to produce results. Blue is 'moderate' wheel resistance and green is light wheel resistance. It is quite amazing to see patterns appearing where HiRISE seemingly does not see any difference. According this analysis, the area to the east of Victoria would have shown less wheel-resistance and better driving then the westerly part, might be long before we ever find out whether that's true but it is interesting ;-).

Click to view attachment

The present area, it seems to pick up the larger bedrock patches (owing to the lower CRISM resolution this method doesn't work well on very small area's). The sand area inside Porcupine 'looks' different to the eye, however this type of terrain has as yet not been 'sampled' so the software won't find a match with historical data.

Note that these colors only show expected wheel resistance and bearing-strength of the soil, it doesn't say anything about terrain roughness etc!

Disadvantage of the thing is that it needs CRISM data to work, I can make it work with only THEMIS and HiRISE but then results are far less accurate, so hopefully there will be some more CRISM images of the route further south.

Distribution of Ferric and Iron minerals in Victoria and Endeavour region.

Note the colors in themselves are meaningless, however area's with the same measured quantities of minerals (within a certain bandwidth) are given the same color.

Click to view attachment

Click to view attachment

Images are created from the average of 5 different CRISM observations, basically the software checks the quantities for ferric and iron minerals for every position, then creates a mapping giving all positions with the same measured quantities the same color.

Distribution of (high-Ca and Low-Ca) pyroxenes in Victoria and Endeavour region, using same method as above.

Once again colors in themselves are meaningless but area's with the same distribution are given the same color.

Click to view attachment

Click to view attachment

Terrain model created by comparing Mike's measurements of wheel resistance and dune heights with observed data from HiRISE and CRISM.

Note in this image the colors in themselves are meaningless, however area's with same properties are given same color.

Click to view attachment
Click to view attachment

Once again, note that colors in themselves are meaningless, so green does NOT mean 'perfect driving', however terrain with the same color can be expected to have the same properties (both wheel resistance and dune size/hight).

This is VERY cool. Would it be possible to add in the topographic data?

I think this is creating a pretty geologic map of Meridiani planum with the exposed strata having slightly different mineral components. With the topographic data added, it should be possible to create a side view of the different layers.

Here is a previous graphic based on an abstract by Sullivan et al., 2008. (post 731, this thread)

The topographic data might also explain the curviness of mineral exposures seen near Endeavour crater seen in your lower graphics.

IIRC, as we head south we are going into younger terrain. So each band southward sat above the older northern band in the layer stack.

-Mike

Here's a graphic based on Geert's CRISM (Fe and Ferric minerals) map and the Sullivan et al. abstract of the layering sequence in Oppy's location in Meridiani:

Click to view attachment

-Mike

Technically it should be possible, as I mentioned before I developed a dataset with all data (CRISM, HiRISE, THEMIS, etc) in digital format linked to a lat/lon grid, so I can relate anything to anything and project it in any which way and just let the computer crunch out the results (I'm more or less writing the software as I go along).

Problem with the topographic data is that as yet I'm not very impressed with MOLA data, I know it's a wonderful tool but at the distance scales we are talking about it's not giving me very much fabric to cling to. Basically the easiest format for me would be a greyscale map-projected 'image' with brightness relating to height but it needs to be very high resolution in order to spot the correlation you're trying to find. An other option might be to simply take a number of lat/lon positions and check their heights in Google-Mars, and from that I can probably let the thing produce a graph relating height to composition...

Let's say I'll work on it ;-)

With regards to CRISM note that this is all more or less 'raw' data, as has been mentioned before CRISM data is influenced by IR-brightness but it is difficult (for me) to know how much I need to correct for this, it's really nice to see pieces fit together and I love to play with this data and see what I can get out of it, but it's not a proper scientific study ;-).

Further attempt on terrain specification for whole area. As CRISM data is not available for the complete area, this map is created by using HiRISE average brightness and variance, with THEMIS Night IR brightness data. I'm definitely not a geologist, so this is only an amateur-overview of what it looks like to me.

Click to view attachment

Note colors in this map do not necessarily define driving-characteristics, they just define terrain-type.

Click to view attachment

The map for the area north of Victoria seems to conform with my earlier CRISM analyses and with Mike's post here .

It looks to me like the green area notes the hermatite-rich strata (basically flat surface with few ripples due to stronger top 'coating'). Further south the sulfate rich strata results in terrain with more dunes and ripples, while around Victoria crater the hermatite-rich ejecta has resulted in re-appearance of the "green" flat terrain. Darker 'windstreaks' downwind of several features might be area's were darker, basaltic, dust was deposited by the wind.

I don't know why the sulfate rich strata seems to come in to 'varieties' in this map (light purple to the northeast, then a small band of 'green' and then the dark brown/purple which define the larger ripples north of Victoria and around Erebus).

Click to view attachment

Click to view attachment

Further south of Victoria the hermatite-rich strata surfaces again, resulting in re-appearance of the 'green' terrain. Unfortunately the MOLA data I have in my dataset at the moment is not yet accurate enough to correlate this map with topographic data.

Click to view attachment

Finally, when entering Endeavour crater it seems to me that there is more darker (basaltic?) tinted sand, which is also present on the planes south of endeavour. There do seem to be several layers of terrain types when descending into Endeavour. There is still a lot to explore ;-)

Can't resist to post a few more images made with above mentioned computer technique.

Don't know how much (or how little) of this really relates to anything and how much is just computer-enhanced garbage but otherwise we can always put it down as abstract art ;-).

Click to view attachment

Endurance crater and surrounding area

Click to view attachment

Vostok area

Click to view attachment

Victoria crater interior

Click to view attachment

Porcupine area. (Note: this is NOT a 'driving' map, colors refer only to terrain-type and have no further meaning as such).

Click to view attachment

NW Entrance into Endeavour crater

Click to view attachment

South slopes of Endeavour (Medusa Rock can be seen on the right in middle)

Click to view attachment

Ridge on the plane south of Endeavour

Your "artwork" is much appreciated Geert!

They do look cool, and they are the right size for CD artwork as well



I know this probably exists in the thread earlier on, but is there a key to those colors?

Nope, in this case there's no key, the colors in themselves have no meaning, they just mark area's with similar characteristics (within a certain bandwidth).

What the software does is searching the dataset for as much data as possible about each pixel position (as CRISM is unfortunately not available for the complete area and THEMIS has only low res, most important are average grid-brightness and variation within a grid centered on the position). Each pixel then gets a string associated with it, which basically states all measurements relating to this pixel divided by a bandwidth factor.

Once this is done for all pixels, the computer assigns random colors to all measurement-strings and marks all pixels with this string in this color.

So each color marks a specific set of measurements and all positions with this color will have the same terrain characteristics, but the color in itself is meaningless.

Although these images are mainly created from HiRISE data, it is remarkable how much resemblance there is with the CRISM measurements I posted earlier, which gives a bit of confidence that at least some of these layers the software seems to detect indeed do relate to 'something', however you would need a lot more data (or ground-checks) to know what chemical or geological feature is causing one specific layer to appear in the image. My software only shows there is a layer but it doesn't say much about what it is made of... (and as I already mentioned, there is probably a lot of 'garbage' in it, sun-shadows, etc).

Some terrain South.

Click to view attachment

Hi res image here (13MB).

In the latest Mars Exploration Rover Update at The J.P.L. Site dated June 9.
Ashley Stroupe says in the next 2 weeks they need to make a decision on 2 paths for
Opportunity to get to Endeavour Crater.

Path 1 is the shorter route with taller sand ripples &
Path 2 is the longer route with smaller ripples.

If they do the shorter path. When will we start heading that way and
can anyone predict when this might start and where we might make that turn???

From the maps I have seen. I think we know what the longer path is.

On SOL 1899 we hit the 10 mile mark

Here is the link to the Update:

http://jpl.nasa.gov/video/index.cfm?id=843

The impression I got was the 'short' path would be something like my prediction the 'long' path would scoot around the west side of the large ripple patch we are approaching.

That would take us off the current HiRISE image - does anyone know if there is a HiRISE image further west that I could ripple map

I found this one (ESP_011765_1780) via Google Earth. On the selection box at the left you should check "Spacecraft Imagery" > "HiRISE Image Browser" and the outline of the images will show up. This image has a lot of overlap with the current one (PSP_009141_1780) but covers some terrain to the left.

This site should be useful too.

You know, I've always suspected that applications would be more useful to me if I would just take the time to RTFM.

That new view shows that the really nice ripple free terrain extends due south of Oppy. That certainly makes the westward route even more appealing. Apart from one tricky patch, it looks like fairly easy trough driving along the western edge to get to it. Not having to go cross-ripple for the eastern leg could make up time-wise the added distance.

I have to admit I can't see any advantage to going west around the large ripples. It looks like we would have to go considerably out of our way to avoid them safely. BrianL indicates that there is one "tricky patch", but it doesn't look to be any easier or shorter than the tricky patch we've got awaiting us to the southeast right now. The westward route may add a lot of good terrain to the journey, but it doesn't seem to remove much bad terrain. And making the trip any longer than it needs to be has to be considered a meaningful risk.

Has anyone tried mapping a tentative westward route yet? It would really help visualize what the trade-offs are if we could see some educated guesses mapped onto the HiRISE images.

Mike

Click to view attachment

OK, here's my lame attempt at the route I envision. The jog around the "tricky part" doesn't need to be as dramatic as they way I've drawn it I'm sure. To my eye, that bottom east leg screams 200 m drives (well, but for that pesky wheel current thing).

Yes, ESP_011765_1780 and ESP_012820_1780 extend to the west of the PSP_009141_1780 image that I've been using for my own route maps. 9141 is great because, though it's not the sharpest image of the landing site, it has the lowest emission angle of all the HiRISE images so far, so there's minimal distortion due to high-frequency topography at the scales relevant to the rover route planning. The other two images have high emission angles, because they were acquired for high resolution stereo of the path ahead (something greater than 30°). Of these, 12820 is less dusty than 11785, so I'm adding these two to my route map with 12820 overlying 11785. 11785 extends slightly farther west than 12820, which is why I'm including it in my mosaic (in the off chance we need to go that far west around the Purgatory fields or triangulate to features in that direction from the rover).

planetarily,
-Tim.

Thanks for posting that, BrianL. Crude though it may be, it gives me an idea of what the scope and range of the westward route would be. I was thinking they could probably cut to the east a couple km farther north, just past the rough patch that you indicate. Either way though, we're probably looking at an additional 2 to 5 km added to the journey. It may also preclude the possibility of stopping at "mini Endurance" crater, depending on where the turn is made.

It still seems to me that the best chance to turn south and east is now. The RP's have negotiated challenging terrain before, and I'm confident that they could pick a safe path through the upcoming rough patch. I would also expect the shorter route to maximize the opportunities for resting that RF wheel, which is clearly a priority at this point.

Of course my own experience in driving robots on other planets is somewhat meager, so I'll just wait and see what the experts decide...

My suggested path is certainly very roughly plotted and likely overly cautious. I'm approaching it from the standpoint of minimizing stress on the wheel. To me, this would mean keeping the ripple size small (further west, smaller ripples) and in following the troughs till they are clear of them. I'm sure a healthy 6-wheeled rover could make it either way without problem. Perhaps the deciding factor in this decision will be which course offers them the best chance should the RF wheel lock up.