extension estimates

[cossatot]

Need maximum and minimum extension estimates across the S Snake Range.

[eqsarah]

Just documenting our conversation via email: Max extension: 35 km --Footwall cutoff from retrodeformable cross section maximum amount would be 24 km from McGrew (1993), but 35 km gives more modeling room. If it seems unreasonable once modeled, it can be decreased.

Minimum Extension: 15 km --Based on interpretation of AFT ages in Miller et al., 1999. There are ~12 km of exposed AFT ages in their transect across the SSR that they interpret to be essentially the same age at ~17 Ma, and then ~3 km of of slip interpreted from "overlying partially annealed samples." McGrew (1993) had a lower bound of 8 km based on retrodeformable cross sections, but this seems an unreasonably low number. Unless of course the geothermal was just crazy hot. Which there might be some evidence of....going to look for the paper from the NSR (which is a debatable proxy for the SSR).

[cossatot]

In the absence of evidence for synextensional magmatism, the geotherms for the north and south should be pretty close.

[eqsarah]

One of the things that E. Miller and folks like to talk about is the NSR is a thermal welt, and that this is the reason why the NSR has higher amounts of extension than the SSR.

Also, there is a synextensional pluton in the SSR. The Young Canyon-Kious Basin Pluton intruded at 36 +/- 1 Ma (U-Pb, Miller et al., 1988). It is in the lower plate of the SSRD, and this pluton has mylonitic areas as well as cataclastic areas of deformation. The age of this intrusion is toward the upper end in the range of my ZHe cooling ages (~41.7 - 21.1 Ma), but none of the ages are exactly 36 Ma. Also, if the simple interpretation (a la Stockli, 2005) is correct, the onset of extension based on the point of inflection in ZHe ages is ~25 Ma.

[cossatot]

Well shit. Magmatism can make a pretty big difference and unfortunately is really hard for a program like Pecube to deal with. We will have to do one of two things: Either 'correct' the data somehow by estimating how much younger the ages are because of the magmatism and then pretending it doesn't exist, or demonstrating that it didn't actually affect the ages all that much.

[eqsarah]

This is a rather difficult problem. I have been thinking about the data and I think it is mostly a tectonic signal really based on the younging of ages towards the SSRD. I think I will need to think some more on both of the options (correcting or discounting).

[cossatot]

RIch Ketcham wrote a paper with a student on the influence of magmatism on thermochron ages from I believe Naxos. he said 'a few million years max' on the influence when I asked him about it a while back. But it of course depends on the size of the pluton, the relative temperatures, etc. It definitely won't make a difference for apatite ages, because they're so much younger.

BUt one thing that we could do is a multivariate linear regression (e.g. regress ages by both along-strike distance and fault-normal distance or elevation) and look at the residuals (maybe take the ages from the pluton out before doing the regression, if possible). If the ages from the pluton are systematically younger, then we can use that to approximate the difference. If they are the same age, it's no problem. If they are non-systematically different, we just bump up the uncertainty on those ages. But I havne' looked at the data in enough detail to see whether this kind of regression can be done given the spatial distribution of data.

[eqsarah]

I am not quite sure that the spatial distribution of the samples would allow for this regression technique because there isn't much along strike variation, but I'm not sure I completely follow the technique exactly.

Just a quick blurb on spatial distribution in case this clears some things up though: the samples are all within one (Jurassic aged) pluton in the footwall with <1000 m elevation difference. All samples were taken (as much as glaciated outcrops would allow) along a transect subparallel to motion along the SSRD as defined by mineral lineations in mylonites of the SSRD. The pluton, Snake Creek-Williams Canyon, they were collected from ranges from ~3-10 km (current map linear distance) from the ~36 Ma pluton (western most sample in range). Only one of the zircon ages is similar to the crystallization age of the pluton (~36.9 Ma ZHe age; ~6.5 km linear map distance from ~36 Ma pluton). The rest are all either older by >=4 m.y. (and not obviously partially reset based on relatively reproducible ages) or are younger than the crystallization age by >=10 m.y. (~26-21 Ma) The contact metamorphic aureole of the ~36 Ma pluton isn't very extensive as far as I can tell from isograd maps. Peak contact metamorphism is in higher green schist facies conditions (grt+andalusite+bt), but the andalusite-in isograd around the ~36 Ma pluton doesn't extend all the way to the Jurassic pluton I sampled from (see McGrew, 1993 p. 4). Just a quick perusal through literature, and found a Tagami and Shimada (1996) paper that sampled ZFT ages around a pluton and found that ages were reset up to ~3.6 km from pluton. This is a higher temperature system than ZHe, but it would suggest that the one ZHe age similar (~6.5 km from pluton) to the pluton emplacement age may just be coincidence?

All of that being said, there was probably plenty of "extra" thermal heat during the late Eocene to early Oligocene based on the numerous volcanics (ignimbrites and flows) that were erupted all over eastern Nevada and western Utah at this time (which leads to a chicken and the egg question: did the extra heat from magmatism cause extension in the first place in the area, or did extension allow for magmatism?). So, "unmixing" the magmatic heat signal from cooling associated with movement along the SSRD is going to be next to impossible (although it would be way awesome if you could--might get at that chicken and egg question at least for this one MCC). You mentioned it is difficult to deal with changes of thermal parameters for the crust over time in PeCube, so I imagine that is just a limitation that we have to deal with--many other workers have used thermochronology in the basin and range and do not "unmix" the signals either (to my knowledge).

I will look for the Naxos paper to see if it helps make an argument that we can discount the affect of the Young Canyon pluton as well--thanks for the lead.

[cossatot]

The paper is here: http://www.sciencedirect.com/science/article/pii/S0012821X05006515

Also, w/r/t the chicken and egg thing: From a causal perspective, it is pretty clear that in orogens that extend during regional compression (say from gravitational collapse), heating and weakening of the crust won't cause extension. The Houseman and England papers on thin viscous sheet modeling from the 80s demonstrated this. Nonetheless, it's a widespread belief because it's intuitive but it's probably not right--if the crust is viscous (at some approximation) and gets weaker, it actually thickens more under compression. It would have to get really really weak, and therefore have a really really low viscosity, before it would start to extend from weakening--basically it would have to be a magma/crystal mush. This extension would also be perpendicular to the direction of maximum horizontal compression. But the early basin and range extends in the direction of NAM-Farallon plate convergence (roughly). Don't ask me what's causing that (yet... I'd have to read up to get some ideas).

One easy way to think about this is from a different perspective (that of brittle deformation) to simply think about the strength of rocks. The innate strength of rock is basically the strength under tension, or under no confining stress. This is the y-intercept of the failure envelope on a Mohr circle diagram. For large a volume of rock that has fractures/weaknesses, it's like 1 MPa, basically zero compared to the stress magnitudes at 10 km depth (pressure here is 270 MPa, neglecting pore fluids). The other parameter that can change here is the slope of the failure envelope, which is determined by the 'internal friction' and elasticity of rock, which doesn't change appreciably with temperature. The deformation state of rock is really controlled by the stress state, not by the rock strength. Extension begins because vertical stress overcomes horizontal stress.

[eqsarah]

Thank you for the paper link! This got my wheels turning, since their modeling shows cooling (even with a heat pulse in the middle of their cooling ages) is still controlled by tectonic processes. I want to be able to show that the heat pulse doesn't affect the cooling ages in the SSR, so I am going to first try to run some very simple 1D models for thermal equilibration of the crust around the Oligocene pluton. Ehlers (2005) published some MATLAB code that makes it pretty straightforward (give assumptions for intrusion temperature, background crustal temperature, thermal diffusivity of the rock) that even a Luddite like me can handle. I intend to input the background country temperature as a function of depth of intrusion (assumed), geothermal gradient (also assumed) and some available higher temp thermochron (K-Ar Mica), the temperature of intrusion as a function of the well(ish) known chemistry of the pluton, and then the thermal diffusivity as a function of the chemistry as well. If you have any insights or suggestions to this approach let me know.

Hopefully this code will show that the thermal pulse does not affect the locations where I sampled, or if it does it will give us a decent idea about the thermal state of the crust so that we might be able to correct things somehow. I'll let you know how it goes (in a short course for the next couple days though and I know you're still busy too).

In regards to your information above, I absolutely agree that heating and weakening of the crust alone will not cause extension. In terms of some literature on the driving forces of extension in the NBR specifically, there is (of course) debate (two end member/possible models mentioned above is extension as a result of magmatism and extension from gravitational collapse). As you so neatly pointed about above, the chicken and egg/end member model arguments aren't correct but it still gets addressed sometimes when discussing extension in the B&R.

The last time someone (to my knowledge) tried to synthesize a bunch of data from all over the NBR to hash out the extensional driving forces was Sonder and Jones (1999), others have certainly done it using a less broad scope since this paper (they also address the Central and Southern B&R too). I don't know if you've ever read it but just in case you haven't here are the cliff notes (disregard if you have):

They came to the conclusion that in the NBR (1) over thickened crust with high potential energy after the compression from the late Cretaceous to early Tertiary (internal body forces), and (2) possible lithospheric mantle delamination and certain foundering of the Farallon slab below the NBR followed by asthenospheric upwelling (basal normal forces). But also like you said, this still doesn't really address syn-convergent extension in the NBR, since it definitely is recorded within the rocks prior to the foundering of the Farallon....it's an interesting puzzle for sure.

[cossatot]

I think the sonder and jones papers are really interesting and great works. However, it's definitely interesting to compare them to the GPS velocity field (which started coming around 5 years after they published all that), because their predictions on where extension should be happening is really off-- But the theoretical work they did is great.

There are probably big confounding effects due to different crustal rheology due to heat, composition, etc.--this can localize deformation for sure, but not control the kinematic type of strain.

I think that work coming out by the UNR geodesy crew (Kreemer, Hammond and pals) is pretty great, because it's very data-forward and they're quite clever and capable. THey don't always do a lot of speculating over long-term orogenic dynamics, though.

[cossatot]

Re magmatic heat modeling: Here is a paper discussing a model with a GUI (I think) that is probably a bit better than Ehlers' TERRA model. I think I downloaded it a couple of years ago while I was at Dalhousie living in a dorm room and doing cosmo work day in day out (it was awesome). I don't remember if I emailed the authors or found it on the internet or supplementary info or whatever, and both that computer and that email address are no more.

[eqsarah]

Thanks for the paper link. I will try using this model instead of the other to estimate any affect of the pluton heating. I had originally intended to just use not even the TERRA model, but rather a very simple calculation that he describes in "Crustal Thermal Processes and Thermochronometer Data Interpretation" (p. 335 of the low temp thermochron reviews in mineralogy and geochemistry) as a first pass to see if the heat pulse is something to be mildly or wildly concerned about. I will of course keep you updated on my progress.

[cossatot]

I'm going to open a new issue about the magmatism thing so we can keep this thread on topic.

[eqsarah]

Ok, great. Otherwise everything for the extension estimates is detailed up top (post 2), and I don't think these numbers should change with synextensional magmatism since they were defined using retrodeformable cross sections.

[cossatot]

What other extension constraints can we add? I can do things like:

• 15 km <= extension <= 35 km
• SSRD extension >= 2 * WPF extension
• WPF extension <= 2 km

(the last two are made up but examples of additional constraints)

[eqsarah]

As to other constraints I do not think anyone has ever suggested amounts of total extension on the WPF (it's kind of an ignored fault, sadly).

I am hesitant to use the amount of basin fill in Spring Valley (the valley west of the SSR) to get even a rough estimate amount of extension (assuming purely dip-slip motion, and a dip of the fault from the McGrew cross section). I think this might be a an unreasonable estimate since: (1) this would assume the fault is completely planar and dip-slip, (2) that we actually know the thickness of the sediments closest to the SSR side of the valley, and (3) the range bounding fault on the Schell Creek side of the valley is a "greater" structure since it has exhumed something like ~7-9 km of stratigraphy in it's footwall and therefore is likely to be the major contributor to the amount of accommodation in the valley. Just musing here, if you think of something that might be able to track down or find to help us with at least a first order estimate let me know.

I do know that the slip rate on the WPF is quoted as <0.2 mm/yr in the Quaternary from the Quaternary Fault and Fold Database report on the WPF. This is their comment on the rate estimate: "No detailed data exists to determine slip rates for this fault. dePolo (1998 #2845) assigned a reconnaissance vertical slip rate of 0.01 mm/yr for the fault based on the presence of scarps on alluvium and the absence of basal facets. The late Quaternary characteristics of this fault (overall geomorphic expression, continuity of scarps, age of faulted deposits, etc.) support a low slip rate. Accordingly, the less than 0.2 mm/yr slip-rate category has been assigned to this fault."

I know this isn't a total amount of extension, and is only really helpful for the Quaternary. Ugh, I just wish there was more data on that fault! (A phrase I am sure that has been uttered by almost every scientist in history. Repeatedly. At least there will always be more to learn)

I like where you're going with the second idea, and I think that the extension should be greater than the WPF extension. However, I am unable to make an informed guess on how much greater. Although, at least 2 times greater seems reasonable to me, but not defendable with hard data that I can think of currently.

I'm starting to run out of literature to peruse for the SSR specifically, everything else is either for the NSR or is much more regional that I'm reading now.

[eqsarah]

Ok, so I talked to Wanda about my first point about basin fill for a first order estimate that I was hesitant to use. She thought that it was better than no parameter; however, if you disagree based on what I'm about to outline then no need to use it.

So, I found some inverted gravity data to provide the depth to basement (red area in box for Spring Valley adjacent to the xsxn of McGrew). This of course means, that we assume the fault makes the accommodation space for the basin, and that the Schell Creek Fault's influence is minimal on this side of the basin. It's a relatively uniform thickness basin on the scale of the gravity data. This also means we are assuming pure dip-slip motion, planar fault, and I assume a 60 degree fault based on brittle failure in Mohr-Coulomb space. See my sketch below (also note, this assumes overlying stratigraphy in intermediate stage of extension is taken out by the SSRD and/or erosion and not by the WPF). If you don't like it we don't have to use it, but it might be a useful and a defendable (ish) first order estimate--it turns out not really be either a min or max estimate though. Either way, I come up with an extension estimate of ~1.3 km. We could put the parameter boundary at 1.5 to 2 km to take care of the slop in the numbers I use since they are pretty "fuzzy." Me and GW, doing some fuzzy math. :)

[eqsarah]

Sorry the bottom of the sketch got cut off, but you might be able to guess that it's just the tan 60=2371/x and x=1.34 km

[cossatot]

OK, this mostly sounds good, but I have some questions about the approach. The WPF dips at 51° in your Pecube sections, so we'll go with that. 60° isn't supported by McGrew's x-sec (it looks shallower than 45°) and actual active normal faults more commonly have dips of 45-50 in the seismogenic zone--it's hard to keep a normal fault w/ any real displacement from rotating as it extends. This gives 1.919 km extension.

Also, if you use the intersection of the WPF and the top of the Pz section (the erosional surface with the Tc overlying) in Plate 2c and 2D, it looks like there is about 3 km of footwall uplift, in addition to ~0.6 km of HW subsidence (basin depth), for 3.6 km throw. Now, I realize that most of that uplift probably was due to slip on the SSRD, but this wouldn't have made the basin in the west, so (to my mind) the WPF basically downdropped the basin (or at least its eastern side), so all of that 3.6 km throw is its own.

So that gives 2.9 km extension. I will set bounds between 1.5 and 4 km. If this is too liberal or I am missing something, we can dispense of those excessive models later. Sound good?

[eqsarah]

I think in the intermediate step, it looks like it is dipping at 60 (just using a protractor) but going the 51 is totally fine with me, so the 1.92 km of extension estimate is great and the 1.5 as a lower bound is probably fine too. The upper bound may be too generous, but my claim is hard to evaluate.

For me in that intermediate step (Plate 2C) the thickness of the section above the WPF is pretty unlcear without a scale (Dg-Guillmette Formation to Clp-Cambrian Lincoln Peak Formation). I talked to Wanda about that, and she said you can assume a thickness (by using the same scale as in Plate 2D, but McGrew didn't conserve strat thickness or line lengths in his cross sections, only area so it might not be a good assumption), and have the WPF as the structure removing that section--but that section seems to be what is exposed in the Schell Creek Range to the west due. And that is thought to be from motion on the Schell Creek Fault--not from motion on the WPF. That's why she and I thought to go with the more conservative estimate using only the relief exposed today (which absolutely turns out to be neither a max or min estimate). This estimate of course does not take into account the thickness of the Pole Canyon Formation (Cpc) to the Lincoln Peak (Clp) that has been removed by the Plate 2D (The complexly faulted Dg and older section has all been removed by the SSRD in his cross sections)....and the thickness of the stratigraphy beneath the SSRD is pretty unclear as well. To me that duplication of section in Plate 2C is not absolutely necessary from the SSR data (it may be that there is something the in the Schell Creeks that argue for that, but I am not sure and it is not explained in the text).

It's also possible that I completely missed the boat on what you mean, so if so, try to explain to me again. Or it might be worth an actual conversation eventually. For now those wide ranges are fine and like you said we can tweak them in the future.