[X] My Assistant

[dburt]

We don't really need to debate the semantics on this one...
Doug

Speaking as a geologist, if Cugel is looking for a single word describing the fact that wind is causing dust or sand to move, try "transport". For a sediment like sand, geologists commonly speak of "erosion" of some bedrock at the source (whether by sand blasting, impact, explosive volcanism, matrix dissolution, ice sublimation, or some other cause), "transport" of the sediment some distance from the source (possibly half way around the planet for wind-carried dust), and "deposition" of the sediment at some site where it accumulates for long enough and thickly enough to be described as a distinct unit (mappable layer) or even harden into a rock. On water-rich, geologically-active Earth, rapid preservation via burial beneath younger sediments and hardening (groundwater cementation) into a rock is generally assumed. On dry, frozen, geologically moribund Mars it is difficult to say for how long a given particle might blow around before being deposited somewhere (probably in a crater or other depression), and for how long such a deposit might be preserved. Hardening into a rock could involve ice as the cement, as for frozen dust layers at the poles, or sulfate salts as the cement, as was apparently the case at Meridiani (for whatever reason - acid playa lake or acid groundwater evaporation, fumarolic volcanism, sulfide oxidation, or salt/brine condensation from an impact cloud).

Volcanic (and probably impact) surge deposits, BTW, tend to become cemented as they are deposited ("instant rock"). Home Plate rocks, or at least the basal layers (and I hypothesize the upper layers too) have been described as a type of surge deposit (volcanic is assumed, probably unnecessarily).

-- HDP Don

[dburt]

I've seen scattered reports of clays detected by orbiting instruments within Meridiani, but specifically with regard to your question about clays beneath the sulphate-rich sandstones of the Meridiani plains, I noticed the next to last paragraph in this news release recently. I'd like to hear more about that.

So would I, although as mentioned above, occurrences of older clay-rich layers towards the base of the section would probably be unrelated to the younger cross-bedded layers observed by Oppy at the top. (For example, the many different rock layers exposed in the Grand Canyon, AZ are inferred to have formed in completely different environments that range from beach sands, to shallow seas, to river deltas, to desert dunes on land.) Also, I don't necessarily agree with the closing statement "The big picture appears to be a change from a more open hydrological system, with rainfall, to more arid conditions with groundwater rising to the surface and evaporating, leaving sulfate salts behind" because clays need not indicate rainfall and salts need not indicate liquid water evaporating. Clays and salts might form in a variety of ways, especially on planets other than Earth. In fact, rainfall (other than ephemeral condensation of water vapor released by major meteorite impacts) may never have been typical of Mars. That early Mars must have resembled present-day Earth seems to be an implicit assumption.

-- HDP Don

[dburt]

dBurt

"Can you explain that scour in terms of their model? They appear to have hypothesized a "supersurface" ("Wellington contact") of regional extent on the basis of a single large cross-bed with a large channel-like scour taken out of the top of it - which scour, on the face of it, makes their hypothesis untenable. Do you disagree? If so, why?"
Why not a channel cut or similar for the “scour” in the Wellington contact? If there was surface water in the area as the MER model states all you need is a small gradient and you have flow, which in turn could cause down cutting (I hear the surface rocks are easily-scoured sand). If there was a fluctuating water table at ancestral meridiani perhaps there was also topographic relief (even minor) just enough to generate flow once groundwater breaks through the ground surface (aka springs). Here is perhaps evidence in the shape of “scours” for a local surface hydraulic gradient at meridiani. Seems more plausible IMO (small wart) than a surge generated vortex (big wart). I would hold back on speculating a regional extent for the Wellington contact however. I’m still hoping to see a similar feature in Victoria.

Question – the surge model seems to apply more directly to the layers examined in the Columbia Hills then meridiani, yet there seems a lack of enthusiasm from you to really hit that hard. Are you taking one landing site at a time, or is there a more fundamental reason for that, or am I just completely out-of-the-loop crazy ?

Inasmuch as I was unable to answer these questions by "other Don" last July, I'll touch on them now. Contrary to his claim, that scour is the right scale to be a vortex in a surge cloud - I saw and imaged several similar-scaled scours (1-2 m across) in volcanic surge deposits in Oregon last summer. I cannot see how it could be a water scour, if the planar upper surface marked the water table, as claimed (i.e., the whole idea is that wind or water erosion could occur down to the water table, but no further). If the water table afterwards dropped, before deposition of the overlying sandy layer, allowing water to erode a large channel, then why should the water stop at only one channel, why is the channel filled with sands that obviously were not deposited in water, and why didn't general erosion resume? In this regard, intricate braided stream patterns, not isolated deep channels, generally mark sand-choked streams flowing across very low gradients. I have looked at many such paleowater tables in eolian sandstones over the past few years, and have never seen such a scour in any of them. The boundary surfaces are always near-planar, with thin shale beds apparently marking low spots where standing water was locally present. We're still waiting to see the first shale bed on Mars. This could provide direct evidence for standing water in a dusty environment.

Regarding the Columbia Hills, slightly out-of-the-loop might be my vote. The hills appear to consist mainly of crudely bedded coarse ejecta (i.e., breccias), rather than finely bedded surge deposits. This is fairly obvious from outcrops, and was stated in previous posts, I believe. Members of the MER team seem comfortable with this interpretation, at least in part (although to fit their volcano interpretation, they would like at least some of it to be coarse volcanic ejecta, rather than impact ejecta).

-- HDP Don

[dburt]

...Please do not take this comment as disrespectful but I suspect that your interest in the impact surge concept has become personal rather than scientific and your arguments seem centered on attacking the JPL theory without providing sustainable arguments for your alternative scenario. I do note that JPL considered the impact surge scenario as one of their favoured models and then rejected it based on tests and feedback from the rovers. Lets face it, Innocent Bystander raises significnt questions with respect to your theory.

Given that I can now do so (and I am grateful to Doug for the chance), I'll also reply to this July statement. First, as stated near the beginning of this thread, I have great respect for those who made what I might call the Cornell hypothesis, rather than "the JPL Theory" (my impression is that JPL press releases merely quote the PIs, in general). Based on their initial assumptions about an Earth-like watery past for Mars, and the need for readily studied terrestrial analogs, it was an entirely reasonable hypothesis. Most planetary people's ideas about impact deposits were shaped by studies of the Moon, a planetary body lacking an atmosphere or subsurface volatiles. In such a case, ballistic ejecta must predominate, because there is little or no opportunity for turbulent surge clouds to develop. Finely cross-bedded impact deposits therefore probably are lacking on the Moon (although microimpact surface gardening makes it impossible to know for sure), and all such deposits probably have been eroded on Earth. Given the lack of familiar examples, impact was rejected out of hand for the finely cross-bedded deposits at Meridiani, despite the abundance of impact craters of all sizes and relative ages in the vicinity, and the abundance of tiny spherules. If I hadn't spent several years studying volcanic surge deposits (some with spherical accretionary lapilli) in the 1970s, I'd probably have done the same.

Regarding "Innocent Bystander," that is, as I recall from last July, a piece of ejecta sitting on top of Home Plate that had been run over and partly crushed by the rover. It was interpreted to have formed originally by fumarolic or hot springs alteration (similar to the fragmental rocks of Silica Valley, and the high-silica rock "Fuzzy Smith" initially encountered on top of HP in 2006). Fumarolic and hot springs activity are believed to be common following major impact events on Mars (owing to the formation of impact melts and other heated rocks that can interact with groundwater or ice); they need not indicate volcanism. The isolated, fresh-looking high-silica blocks (as well as the blocks of vesicular melt) emplaced on top of Home Plate (of which "Fuzzy Smith" was the first found, apparently) must have been put there either by relatively recent impacts, or by relatively recent volcanic explosions. In other words, they all are ejecta blocks that originally came from somewhere else. As mentioned in previous posts, the highly vesicular melt blocks themselves could represent either impact melt or basaltic lava. Take your pick (at this point I have no valid basis for choosing).

HP Summary: Finely cross-bedded surge beds, fumarolic/hot springs alteration, ballistic ejecta, accretionary lapilli, and vesicular basaltic melt could indicate either volcanism OR impact cratering (or both). They are not unique to either. There are dozens of nice impact craters close to HP. Is there a nearby volcano? (Keep in mind that most exploding basaltic volcanoes are energy pip-squeaks compared to a decent impact, and so the volcano would probably need to be quite close indeed.) Lacking a volcano (which might be eroded or buried under sand), are there igneous dikes that could have fed a volcano? Rather than silica-altered pieces of ejecta, are there quartz veins or cemented alteration mounds that appear to have formed by hydrothermal activity in place? And so on. As I see it, the question of volcanism vs. impact is still completely open for HP, strictly speaking. However, the rocks are so similar to those at Meridiani, where explosive basaltic volcanism seems highly unlikely, and so like bedded rocks found to be extremely widespread in orbital images, in places where volcanism likewise seems unlikely, that impact deposition still seems to be the simplest general explanation (Occam's razor). Exceptions are possible, of course. Just show me the volcano (or dikes, or quartz veins, or whatever) at HP.

BTW, if you still believe that I have failed to provide "sustainable arguments for [my] alternative scenario" please cite some specific examples. As mentioned above, valid counter-arguments were what I was most hoping to obtain when I started this long and rather repetitious thread. Rejecting disproven hypotheses is the only way that science can advance.

And thanks for your valuable input.

-- HDP Don

[dvandorn]

So -- if there are nothing but volcanic and/or impact-emplaced materials anywhere in Gusev, including in the Columbia Hills, how does one explain the rather obvious large-scale landforms that argue very, very strongly that a river once flowed into the crater?

I'll also somewhat diffidently point out that Gusev began as a large crater, with a floor presumably paved with impact melt and breccia. Isn't it possible that the Columbia Hills are uplifted remnants of materials that once lay *below* a lakebed? The brecciated nature of at least some of the materials that make up the Hills would support this theory, I think.

Also, it's obvious that a layer of basalt was extruded onto the top of whatever materials made up the floor of Gusev prior to the lava emplacement. There are no obvious large flows that come from outside of Gusev (unless you want to argue that the river valley which debouches into Gusev is actually a lava channel) -- the lava must have escaped from vents within the crater itself.

I see a lot of large-scale morphological evidence for an early crater-lake which dried up and was then modified by relatively benign and non-explosive lava extrusion. Small-scale violent outbursts in the lava emplacement episode (possibly associated with hydrothermal effects) would seem to account for what we see at Home Plate, IMHO.

Then again, I'm just an amateur... *smile*... I think that what I'm seeing is consistent with the "Cornell Theory," here, but of course I could be wrong.

-the other Doug

[dvandorn]

As I see it, the question of volcanism vs. impact is still completely open for HP, strictly speaking. However, the rocks are so similar to those at Meridiani, where explosive basaltic volcanism seems highly unlikely, and so like bedded rocks found to be extremely widespread in orbital images, in places where volcanism likewise seems unlikely, that impact deposition still seems to be the simplest general explanation (Occam's razor). Exceptions are possible, of course. Just show me the volcano (or dikes, or quartz veins, or whatever) at HP.

Several interesting statements, here.

First off, the rocks at HP are quite dissimilar to those at Meridiani in everything except their fine layering. They are not shot through with concretions *or* lapilli, they are composed of hard-set basalt and not loose grains assembled into very friable rock by sulfur salts, and they have withstood erosive winds for far longer than Meridiani sulfate rocks have.

Secondly, and I know it's an imperfect analogy, our Moon is covered with tens of thousands of square kilometers of mare lavas for which there are no obvious volcanoes or even obvious vents to point to as origins.

Lavas of very low viscosity can seep up through a brecciated subfloor, "igniting" pockets of volatiles (such as ices) and creating localized violent explosions, but not creating cinder cones, volcanoes or other obvious signs of volcanic landforms. Our Moon is proof of that.

Show me all the dikes and quartz veins that "prove" that, for example, the Mare Imbrium lavas were emplaced by volcanic extrusion, and I'll take my hat off to you, sir...

-the other Doug

[Aussie]

I thought that Irvine and Backstay were identified as probable dike rocks and that since differentiated volcanic rocks were not identified on the plains before the ascent of Husband Hill this almost necessitated a local magma source? Backstay was an essentially unaltered float high on Husband Hill and I thought that the concensus was that it was a local dike intrusion after uplift of the Hills. I can't find the reference but remember an article identifying Irvine as a surface exposure of a dike.

[Doc]

I thought that Irvine and Backstay were identified as probable dike rocks and that since differentiated volcanic rocks were not identified on the plains before the ascent of Husband Hill this almost necessitated a local magma source? Backstay was an essentially unaltered float high on Husband Hill and I thought that the concensus was that it was a local dike intrusion after uplift of the Hills. I can't find the reference but remember an article identifying Irvine as a surface exposure of a dike.

Indeed Irvine and Backstay were identified as dike rocks due to thier unaltered mafic nature

[marsbug]

I'm getting the impression that the asking 'was it volcanoes or impacts' is a flawed approach. The only consensus being reached seems to be that there are any number of features which could have been formed by either, or both. The 'impacts only' argument seems to rest on the lack of an obvious nearby volcano. Dvandorn makes the good point :

Lavas of very low viscosity can seep up through a brecciated subfloor, "igniting" pockets of volatiles (such as ices) and creating localized violent explosions, but not creating cinder cones, volcanoes or other obvious signs of volcanic landforms. Our Moon is proof of that

.

The volcanoes only argument seems to over look the role impacts have almost certainly played in shaping every square foot of mars. If a large impact can cause volcano-like effects, and an obvious volcano is not needed for volcanism to occur then it seems that, in the absence of some other clue, both could have had equal responsibility. Given that a major impact may be capable of triggering volcanic activity the divide between the two ideas becomes still more blurred.

Dvandorn:

Also, it's obvious that a layer of basalt was extruded onto the top of whatever materials made up the floor of Gusev prior to the lava emplacement. There are no obvious large flows that come from outside of Gusev (unless you want to argue that the river valley which debouches into Gusev is actually a lava channel) -- the lava must have escaped from vents within the crater itself

I believe that flood basalt vulcanism in india has been linked to the chixilub impact, on the other side of the earth.

To sum up, I very humbly suggest it would be worth considering a halfway house theory, taking into account both possibilities, as well as the possibility that the two could be intertwined.
Just for fun, and for volcano pele if he's following this thread Ive attached something on the possibility of impact induced vulcanism on io!

Attached File(s)

 iovolcanism.pdf ( 174.38K ) Number of downloads: 277

 

[dburt]

So -- if there are nothing but volcanic and/or impact-emplaced materials anywhere in Gusev, including in the Columbia Hills, how does one explain the rather obvious large-scale landforms that argue very, very strongly that a river once flowed into the crater?

I'll also somewhat diffidently point out that Gusev began as a large crater, with a floor presumably paved with impact melt and breccia. Isn't it possible that the Columbia Hills are uplifted remnants of materials that once lay *below* a lakebed? The brecciated nature of at least some of the materials that make up the Hills would support this theory, I think.

Also, it's obvious that a layer of basalt was extruded onto the top of whatever materials made up the floor of Gusev prior to the lava emplacement. There are no obvious large flows that come from outside of Gusev (unless you want to argue that the river valley which debouches into Gusev is actually a lava channel) -- the lava must have escaped from vents within the crater itself.

I see a lot of large-scale morphological evidence for an early crater-lake which dried up and was then modified by relatively benign and non-explosive lava extrusion. Small-scale violent outbursts in the lava emplacement episode (possibly associated with hydrothermal effects) would seem to account for what we see at Home Plate, IMHO.

Then again, I'm just an amateur... *smile*... I think that what I'm seeing is consistent with the "Cornell Theory," here, but of course I could be wrong.

-the other Doug

Sorry, I got behind on this old thread, what with spring semester classes beginning today. The basic problem at Home Plate may be that we are speaking of somewhat different things. The rovers and orbiters can see only the surface at Home Plate and elsewhere on Mars, and are therefore seeing only a record of the most recent events. (Note: the rovers and orbiters also happen to see things at vastly different scales.) The rock record of older events is buried, in general. I won't argue with you that Gusev was originally a large impact crater paved with impact melt and surrounded by planetwide ejecta. It then may have been filled with a frozen-over lake that deposited sediments, as indicated by the large channel that seems to drain into it. After the ice sublimed away, these lake or river sediments, if any, may have been covered with lava flows and injected with dikes, and these dikes may even have interacted explosively with wet sediments or groundwater (or ground ice) to form small pyroclastic surge deposits surrounding a maar (volcanic explosion crater). The problem is that the rovers and orbiters haven't directly imaged any of this at the surface. The rovers haven't imaged anything resembling lake beds at either landing site. Rather than intact lava flows or dikes, the Spirit rover has imaged only broken rocks. What broke them up and scattered them about, if not later impacts? Rather than typical pyroclastic surge (which, anywhere near a vent, should contain a major component of coarse material), it has imaged only layered fines (at Home Plate, anyway, just as at Meridiani). IMHO, these are most simply explained as the deposits of a distant impact or impacts, just as coarser broken material (Columbia Hills) may have resulted from closer impacts. The single bomb sag in HP, if that's what it is, is not diagnostic of either volcanism or impact.

Unlike Earth, with its extremely active tectonics (via plate tectonics), Mars (like the Moon) is tectonically nearly dead. Hills and mountains, such as the Tharsis volcanoes, can be built up or erupted, layer upon layer ("the hard way" with youngest on top), but not "lifted up" ("the easy way"), as you propose for the Columbia Hills (although an impact can deform and uplift rocks in and near the crater). One might argue for volcanism as the dominant recent event at Home Plate, but the evidence for late impact is far stronger, IMHO, in that craters are everywhere, everything is broken up (no intact lava flows or dikes) and an extremely heterogenous assortment of rocks is scattered across the surface, including the fines that make up Home Plate itself (and seem extremely common elsewhere). I realize that the surfaces of lava flows can themselves be broken up as a result of flow movement after cooling (e.g., aa in Hawaii), but such a process could not scatter basalt chunks across the surface of Home Plate (and nearly everywhere else, including the far Northern Plains of Mars).

I've personally looked at dozens of examples of pyroclastic surge deposits related to various types of small volcanoes, and, again IMHO, they don't closely resemble anything yet imaged by the rovers (either one). In fact, this very lack of close resemblance is usually used against us, especially for Meridiani. Of course, we are arguing for impact surge, not volcanic surge, but people seem to ignore this point whenever it suits their convenience. Again, a single bedding sag produced by a chunk of ejecta is not necessarily diagnostic of volcanism. I ask you, are any of the other features of Home Plate (and vicinity) necessarily diagnostic of volcanism?

Impact craters of all ages, up to and including the present, are the dominant geological features across most of Mars. Impact cratering appears to be the dominant "modern" process - the one that should be most represented in the modern (topmost or most visible) rock record. So far no mission has identified a single layer or type of sediment related to impact, other than perhaps small erosional remnants surrounding a single class of crater, rampart craters. Are isolated ejecta blocks otherwise the only rocks produced by impacts on Mars, or are important features still being missed by many in the planetary community?

Thanks for your comments. I could, of course, be wrong instead. Many here seem to think so.

-- HPD Don

[dburt]

Several interesting statements, here.

First off, the rocks at HP are quite dissimilar to those at Meridiani in everything except their fine layering. They are not shot through with concretions *or* lapilli, they are composed of hard-set basalt and not loose grains assembled into very friable rock by sulfur salts, and they have withstood erosive winds for far longer than Meridiani sulfate rocks have.

Secondly, and I know it's an imperfect analogy, our Moon is covered with tens of thousands of square kilometers of mare lavas for which there are no obvious volcanoes or even obvious vents to point to as origins.

Lavas of very low viscosity can seep up through a brecciated subfloor, "igniting" pockets of volatiles (such as ices) and creating localized violent explosions, but not creating cinder cones, volcanoes or other obvious signs of volcanic landforms. Our Moon is proof of that.

Show me all the dikes and quartz veins that "prove" that, for example, the Mare Imbrium lavas were emplaced by volcanic extrusion, and I'll take my hat off to you, sir...

-the other Doug

Doug - I beg to differ about rocks at HP vs. Meridiani. I am not arguing that they are exactly the same (they are on opposite sides of the planet, after all), only that they appear to have formed by the same general process. That is, they resemble each other far more than they differ, mainly in bedding features. Also, HP rocks are likewise friable and salty (although not so sulfate rich as the high-albedo uppermost layers at Meridiani); they likewise contain no coarse material, except slightly coarser sand at the very base; they definitely are not "composed of hard-set basalt" although loose chunks of vesicular and non-vesicular basalt appear to be scattered around on their surface (chunks interpreted as impact ejecta). Spherules (called accretionary lapilli) have been reported from within HP proper and appear to be locally abundant in hematitic rocks in the immediate area (e.g., "King George Island" outcrop). The HP spherules are not reported as being rich in specular (hydrothermal or metamorphic shiny blue-gray) hematite, however. Most terrestrial surge beds contain no accretionary lapilli. The spherules at Meridiani do not appear to resemble actual concretions, other than being rounded, as enumerated many times above - wrong type of hematite, wrong trace element enrichment, wrong shapes, wrong size distribution, wrong spatial distribution in the rock.

Why do you then move the remainder of the discussion to features that might or might not be present on the Moon? How is this relevant to Mars? I will note that Mare lavas, other than obvious local impact-related disruptions, are all still "in place" (intact flows), unlike the wholly fragmental rocks so far found by Spirit. "Igniting" is perhaps an inappropriate word for explosive interactions of magma with water (phreatomagmatic being the technical term). These are mainly pipsqueaks geologically (the sorts of feature easily overlooked from orbit), but they do invariably leave a visible explosion crater that is usually called a "maar". AFAIK, these are utterly lacking on the Moon, owing to lack of water. They also appear to be lacking near HP, although they may be common in younger volcanic areas elsewhere on Mars.

Finally, almost no one argues that the lunar suface has been totally disrupted by impacts. Quartz veins are virtually impossible there, given the lack of water, but dike exposures would likely have been destroyed. It's a little unclear to me why many people are so unwilling to accept that impacts have similarly reworked the surface of Mars, and why many others discount the effects of the atmosphere and subsurface volatiles on the martian impact process (impossiblility of lunar-style microimpacts on Mars, for example).

Thanks again for the discussion.

-- HDP Don

[dburt]

I'm getting the impression that the asking 'was it volcanoes or impacts' is a flawed approach...

Marsbug - Impacts vs. volcanism may be the wrong discussion for the Home Plate area considered in a vacuum (and for heavily cratered, obviously volcano-rich Mars in general), but is it the wrong discussion for the enigmatic finely layered rocks at the two rover sites considered together, or for similar finely layered, sulfatic rocks that orbital evidence suggests may be extremely widespread elsewhere in the equatorial regions of Mars? Impacts are fine for craters and their immediate vicinity, and volcanism for volcanoes and their immediate vicinity, or for plateau-forming lava sequences (such as Mare basalts on the Moon), but what are we to do with all the rest? I am arguing that the effects of major impacts are probably more planet-spanning than those of volcanism, and that volcanism on Mars has, over time, become so slow and so localized to specific areas, that it can safely be neglected in other areas, such as Meridiani. (In this I actually agree with the MER team.) As for HP, if impacts can explain everything you see, and volcanism can't, why do you need to invoke volcanism? I'm not saying that volcanism is impossible or didn't once occur, only that invoking it doesn't seem necessary (Occam's razor, a.k.a., the parsimony principle), especially for the finely cross-bedded layers of HP proper. Finally, if the HP layered rocks formed by the same process that formed very similar-appearing rocks at Meridiani, then volcanism can probably be excluded for both.

BTW, the linkage of major impacts to roughly coeval volcanism on the opposite side of the planet is easier to hypothesize than to prove. It has been hypothesized also for Mars (inamuch as Tharsis is roughly opposite Hellas), but is not yet proven, AFAIK.

Many thanks for your common-sense input.

-- HDP Don

[dburt]

I thought that Irvine and Backstay were identified as probable dike rocks and that since differentiated volcanic rocks were not identified on the plains before the ascent of Husband Hill this almost necessitated a local magma source? Backstay was an essentially unaltered float high on Husband Hill and I thought that the concensus was that it was a local dike intrusion after uplift of the Hills. I can't find the reference but remember an article identifying Irvine as a surface exposure of a dike.

Aussie - I feel an urge to point out that finding isolated chunks of dike rock, if that's what they are, is not the same as finding an igneous dike in place (i.e., they probably formed elsewhere from where they were found), and that even "dike rocks" in place could result from impact melts being forced up a crack (i.e., they might not indicate local volcanism). Thanks for remembering those two rocks, though. Solid dikes, BTW, commonly are more erosionally resistant than the rocks that contain them (i.e., they stand out clearly as linear ridges or slabs). Dikes have been identified from orbital images elsewhere on Mars, as I recall.

A pattern of resistant dikes radiating away from and/or concentric around a former volcanic center might be more convincing. The dike pattern is one way that geologists know that Shiprock, New Mexico, for example, was a former basaltic volcano, although the volcano itself (except for its neck) has been long eroded.

-- HDP Don

[Doc]

Aussie - I feel an urge to point out that finding isolated chunks of dike rock, if that's what they are, is not the same as finding an igneous dike in place (i.e., they probably formed elsewhere from where they were found), and that even "dike rocks" in place could result from impact melts being forced up a crack (i.e., they might not indicate local volcanism).
-- HDP Don

If you look at the arrangement of Irvine and the adjacent rocks akin to it,and also if you take into account Irvine's chemical makeup (mafic but less oloivine then plain basalts) and morphology, I have to say at least Irvine should be classified as a dike rock.

Furthermore, if Irvine were a product of impact melts forced into cracks, shoudn't it have a more different chemical composition than what has been detected?

[dburt]

If you look at the arrangement of Irvine and the adjacent rocks akin to it,and also if you take into account Irvine's chemical makeup (mafic but less oloivine then plain basalts) and morphology, I have to say at least Irvine should be classified as a dike rock.

Furthermore, if Irvine were a product of impact melts forced into cracks, shoudn't it have a more different chemical composition than what has been detected?

McSween et al. (2006, published in JGR) are a little more cautious, stating in their abstract that Irvine and Backstay "are found as float on the flanks of the Columbia Hills" and "Irvine and Backstay are lavas or possibly dike rocks". In their text they say "More likely, the Backstay- and Irvine-class rocks were lofted onto the Hills as ejecta blocks or were intruded from below to form small dikes or sills." Their photos shows broken blocks crudely alligned horizontally, but no more than might be expected from multiple pieces of similar ejecta being emplaced together (i.e., definitely NOT "in place" to a geologist's eye - and many rocks of the Columbia Hills are similarly crudely layered or alligned). The lack of obvious vesicles (vapor bubbles) might argue for a dike origin, but many lavas lack vesicles too. Also, many dikes are visibly more coarsely crystalline than lavas, especially near the center of the dike, whereas no such wall to center zoning is visible for Irvine or Backstay. So I would would have to agree with the authors, no particular evidence favoring a dike origin, although it cannot be excluded. (Of course, dikes could be impact excavated too, if you want to have your cake and eat it too.) I found the complete JGR article here:
http://www.mars.asu.edu/christensen/docs/M...ev_JGR_2006.pdf

Many loose chunks of non-vesicular lava (or impact melt) are scattered across the top of Home Plate, along with chunks of highly vesicular lava (or impact melt). These all clearly are young ejecta.

As for the composition of martian impact melt, none has yet been identified as such, so I have no basis for including or excluding any particular chemical composition. Furthermore, millions or billions of years of soaking in an alkaline brine or of contact with crystalline salts might have altered these compositions before analysis, especially near the rock surface. FWIW, the above-discussed paper does note that Backstay and Irvine are unusually enriched in nickel, and then notes in passing "other rocks from the Columbia Hills have unexplained high nickel concentrations."

-- HDP Don