 is it working? So this is a third part of the workshop now and we're going to talk about lens scales and time scales of coupling surface and and tectonic processes. There are two talks now, it's going to last half an hour and 15 minutes question, then we're going to get a break and another talk after okay so they all going to talk about this I think this this talk especially gonna talk about this problem of scales and lens scales and how much do we need to resolve to looking at different problems, different geological problems. So the first talk is by Jean-Arthieu Olive, he comes all the way from L'École Normale et Supérieure and he talks about feedbacks between return deformation and surface processes inside from extensional settings. Thanks very much Luc and I want to thank the committee for inviting me to pretty much share my journey from a geodynamicist slowly becoming a geomorphologist or at least a partially coupled geodynamicist and geomorphologist however you want to call it. Of course this is not a solitary journey I started this during my PhD with Marc Bain, continued during my postdoc with Water Block and this is where I'm continuing to be now with Vika Manatista who is a real geomorphologist and Paul Betka has been very helpful in taking us outside to look at rocks and I also want to acknowledge support from NSF for this. So I want to talk about brittle deformation and how it interacts with surface processes but let's kind of frame the picture a little bit. When I'm talking about brittle deformation I really know what's going on in the upper on that little crust where the temperature is pretty cold you don't have significant discos creep things happen mainly through either extremely localized deformation when you break faults or in a more distributed fashion when you have broad-scale lecture or broad-scale aesthetic readjustment for example and in particular I want to talk about the feedbacks between surface processes and the dynamics of brittle strain localization but as you'll see and as I hope I'll convince you you can't really treat strain localization in isolation from all these other processes of distributed deformation. So when we think again about surface processes and strain localization something that keeps coming back in a lot of miracle models is this idea that surface processes generally promote strain localization that means that if you run a evolutionary wedge model or rifting model what you generally find is when you rework the surface it helps promoting deformation on fewer longer-lived faults. This is something that comes back in a lot of miracle models. What I really want to get into today is what are the physics that underlie these feedbacks kind of in detail what might make a particular tectonic system more sensitive to the growth of its own topography and order removal of its own topography and then from the surface process side when our surface processes efficient enough to really alter the landscape sufficiently to have a tectonic impact and finally as we're discussing in the breakout groups can we find some evidence or indication for such modulation in real landscapes. For this I want to focus not on convergent boundaries but on rifts and half-grabbins in particular and I view them kind of as really great natural laboratories to study strain localization in relation to broader upper crust deformation. This is a large-scale cross-section through a section of the basin and range in Nevada. The upper crust is about 12 kilometers thick as indicated by depth maximum depth of earthquakes and you can see that although this is a diffuse very wide rift much of the tectonic action happened on major master faults that down half-grabbin structures here so most of the topographic relief and structural relief happen through a high degree of strain localization in extension. If we zoom in on one of these the wassuk range in Nevada well we pretty much have all the ingredients needed to say something hopefully interesting about the feedbacks between the surface and what I mean by that is we have a system where most of the strain is taken up on the master fault slip on this fault creates a well locked at a lift creates a range and subsidence in the hanging wall creates a basin that it's built by sediment here it's covered by a lake and you can see right away that this is a fault on which the total slip has been estimated to be a border almost 10 kilometers and yet the relief you've created was only about a kilometer from lake to drainage device so significant amount of mass has been removed as the system was growing and it's always interesting to contrast that with other settings for example at a mid ocean ridge if you had a normal fault that was growing with such a large offset with limited reworking of the topography it wouldn't look anything like that it would actually expose and preserve most of the football into a dome or oceanic core complex so just these simple kinds of thinking leads you to realize that reworking the topography must have a sizable impact on the tectonic yeah all right so to get a little bit deeper into the feedbacks that might happen between topography evolution and a few years ago we coupled a geodynamic model with a simple parameterization process so we took the code that I developed during my PhD sister simple slope solver with ectotic rheology available online can download and play with it it solves all your favorite conservation equations but the key is that it localizes faults continuously whenever the stress state is right for it so when you meet the model pressure you have a scheme that allows local weakening that creates a sheer band that we fault our model setup was fairly simple we did a perfectly elastic plastic brittle layer of a known thickness we initially seeded a normal fault at an optimal tip and we just asked the question what happens if we just pull on the system and let it develop topography the topography is handled through attraction free surface we actually have a whole air layer a very big air layer that we're actually including in the model but in a very simple first order in a way that's just put a fault pull on it and look at kind of what kind of topography we get and how the fault default does it stay stable does it keep growing or breaking a new fault we did this coupling it with a simple set of topographic evolution rules the first step is to erode material at a rate that scales nonlinearly with the local slope so we take away material and the second step is to take all that mass that's been eroded and deposit it that's in the corresponding watershed we have a local mass conservation basin which leads to basically whenever you create a football high you take away a good chunk of the mass and deposit it in there and the one knob that you can try is the overall efficiency of processes for example through the reference so here are three snapshots that synthesize our finding this is for paper we published four years ago where this is the simple exercise of pooling and posing 20 kilometers of extension on our perfected astoplastic layer over an intensity and this is a case with very inefficient erosion very slow erosion rates and what we found is that the initial fault that we had seeded here number one grew accumulated maybe a couple of kilometers of slip and then was abandoned in favor of a new fault here which was actually and this fault accumulates some slip and then it was then itself abandoned in favor presently active of the snapshot we did this again and again increasing the erosion rates by the way this is for a 15 kilometer thick elastic up across and what we found is that if we could have surface processes efficient enough to level the topography our initial fault number one was actually pretty happy this way and it just kept growing and growing creating the basin so this is this basic effect of strain localization promoted by efficient surface processes the other thing we did is repeat this experiment with different rheological parameters and namely in a really simple way varying the effective thickness of that elastoplastic layer as a proxy for its integrated strength so the cases I showed you before here I plotted on this plot of the initial fault lifespan that's how much offset was I able to build up on my initial half grabbin before it was abandoned versus a proxy for the efficiency of the original processes my reference a region rates normalized by false separates the runs I just showed you actually plot along this red trend here where if you have very inefficient erosion this was the snapshot an initial normal fault only grows maybe three kilometers and then dice as you increase the erosion or efficiency you increase the lifespan or the total offset you can be called all the way to infinity in this case we never have but if you repeat this experiment not in a 15-kilometer but in a 25-kilometer layer so much stronger you do get the same kind of promoting strain localization effect but it doesn't ever get you false that live forever it actually only prolongs your lifespan from about three kilometers of offset which is still pretty good it's a sizable impact but it doesn't get you that tells you that the integrated strength of the system has some effect on modulating graphic so that's where we were about about four years ago and we decided to kind of dig a little deeper into the physics and to do this it's always nice to go back to very high school you know basic high school physics just to kind of set the picture we're thinking about how energy is partitioned into the system but a simple way of presenting that is okay let's consider block fine plane you want to take it from initial altitude see I back all the way up to final and you're asking well how much energy does that take you can think about the blocks mechanical energy as a kinetic component will ignore it because it's so slow in our world and there's a potential energy you want to move this block up to a new state you have to push on it right you supply some external energy into the system and a classic convert conservation statement tells you that the change in mechanical energy will be the sum of the work of all the dissipated forces so let's say that's telling you is that the external work you supply to the system needs to counteract the work that's dissipated by friction at the block plane interface if you reshuffle these equations very slightly at the end you can write the total work that you had to supply that media outside observer outside actor had to supply to the block to get it to its new state and that relates to the change in gravitational energy and the friction or generally the work that was dissipated by all kinds of dissipated forces this is in high school so we need to take this to a slightly more intense level but you can pretty much do the same type of argument in the continuum this is just the simple force balance statements divergence of stress plus gravity zero if you're familiar with finite element code you do this kind of thing all the time you formulate it in weak form what that means is you take this force conservation statement multiplied by some displacement right that represents the motion of material through your crust you integrate that over the volume do a little bit of a vector and circular magic and you end up with this form of the equation so this stress term that ends up split into terms something that has to do with the boundary tractions that has the far-field tectonic forces on the system but it also has to do with this term that is radiance displacement and the gravitational potential appears here so let's go through them real quick this is exactly the same analogy well this is analogous to what I was doing block on an incline plane this is what you would consider the external work from the partial forces applied on the edge of the domain this is the gravitational potential energy term and this is the dissipation of the internal work that's done through this is going to have two components actually so if you think about the most generic modes of deformation in the upper crust like I said there's a very distributed folding flexing and there's very localized slip on the fault and that is something that can actually be handled by that term if you see it doesn't relate to the displacement field view it relates to the gradient in displacement so anywhere where you have strain integrated times stress so part of that gradient might be really sharp really sharp gradient that's localized deformation that's the friction on fault interfaces and another part is just dissipation for example elastic bending over longer length scales now the gravity work is also quite interesting it's basically the integral of the vertical displacement over the area of the system in consideration you can again kind of split it into two terms this first term here that has to do with integrating the vertical motion across this entire area like the bulk the inside of the crust and this is a case where over some length scale all the uplift will eventually compensate a lot of that might actually balance itself out but as you get closer to the surface there's another component to this term that has to do with pushing gravity up pushing topography up and pushing basins down and this actually sums constructively in this case because here you've got positive uplift driving positive topography negative uplift driving negative topography plus plus minus minus is all plus so that's essentially the firm that is linked to topography and that's pretty much the heart of the feedbacks with surface processes because that's the part where you're sensitive to external reworking of the relief and finally the external work like I said has to do with the far field forces exert on the system how much energy as an exterior driver do you have to fly to your upper crust to keep it going it's easier to think in terms of forces at least for me than in terms of work so for the rest of the talk I'll focus on force description instead of saying that the work is the integral force along the path I'll just say the force is the derivative of the work either shortening or extension and the point is that this external force that you have to supply to the system to kind of doing changes as you flex the layer as you create faults as slip accumulates on the fault if you start applying a force on the system and you want to keep it deforming doing lovely tectonic things you have to supply more and more force but part of that force actually has central energy components and part of it as a dissipative internal work component and so the question is how does that force golf when is it easier to go break all the configuration shift the system the system organization or when is it easier to just keep going as you started so another way to again frame the same thing but perhaps the way is to think about well how hard do I have to pull on a half grab and to keep the half grab and active so again the total force I need to exert has a term that's connected to internal energy part of it is additional dissipation of faults the other part is flexing and part of it has to do with that's the process we can write this as the amount of force I have to supply to the system to the half-grabbing as a function of increasing fault offset or fault heat the red line marks the threshold for breaking a new fault that's essentially material property let's say I broke a new fault my fault is weaker than the surrounding material so I instantly have a strength drop but then as a cumulative set on the fault this energy term increases so I have to supply more and more energy at the greater force until I reach this breaking threshold where it's easier to break that's including all three terms that's this curve here if I now remove all the topography completely level it build the basins cut off the mountains then I actually strongly slow down this increase in force within fault offset and I will end up reaching this threshold for breaking the next fault much later this is what's happening in about in a 25 kilometer thick perfectly elastic layer you can also do this for a 15 kilometer thicker layer and that's a case where the relative effect of the topographic term relative to the rest of the dissipated term is greater so removing it as a greater influence on the system overall that's why you can do a fault that would normally just grow a couple kilometers and die to a fault that can live forever essentially so that's all the way and good that's kind of like a normal fault in a vacuum with perfect surface processes that takes topography on or off but the next question is well a what's a realistic rheology right it's not just what is on earth a realistic efficiency for surface processes what's a reasonable middle ground between leveling all topography and conserving all topography and again this goes back to the discussion we've been having and we continue to have over the next day is what do we actually need from our landscape evolution models what's really important to capture that's something I'll get into with rifting models but what we set out to do is try to really understand what are the processes we needed to really capture mass redistribution and the right rates of mass redistribution in the half-grabbing system so we actually went out to beautiful scenic Idaho famous potatoes this is in the lemm high range it's the northern most extent of the basin range has a bunch of really nice half-grabbins where you can see screen localization on the master fault here looking through this profile this is looking north in my range it has about one kilometer again of topographic relief the fault itself is thought to have accumulated as much as four kilometers slip and just from this topographic profile you can see the decaying topography away from the crest which records a flexural readjustment to fault slip and right away you can see something from the topography about the main ways that mass is redistributed at the surface you can see these rivers that incise the football block grain material deposited pretty flat this valley it's fine grain sediment some movable cones at the toe of the mountain and if we actually go inside this canyon back at the hanging wall we see pretty much the main mode mass is being stripped of the mountain and transported outward our fault is here I'm here contributing to hill slope diffusion this is the scree slopes of material falling from the mountain into the river and the river just takes it out this is this coupled system of getting material from the tops to the bottom of the valley and then just blushing it out and incising the football pretty large so with this in mind this is where the landscape evolution formulation comes in these are things that are typically or that people try to capture that we've discussed with these types of formulations we've now coupled our simple 2d static model with the more with a more complex model that incorporates the hill slope diffusion right so this is all this mass falling into the rivers our tectonic uplift is not imposed in its case it's fed by my tectonic simulation I give the vertical component of displacement to the landscape model do a little bit step of landscape evolution average it back and get it to the tectonic solver again and the stream bar incision is really what we will focus on and the key primer we'll play with is k irritability although maybe k sends a little bit for kitchen sink because you can really put anything that you want in there to proxy you know it has to do with climate has to do with irritability mythology that degree of fracturing but we won't get into that I just want to try to do something fairly agnostic let's see if we can find realistic values of k stick it in the model and see if we get some response see if we in a regime where our tectonic evolution is sensitive to surface processes so that's the model setup it's not fully 3d it's still using a 2d cross-sectional numerical models where faults evolve continuously but again what you do is you take the vertical component displacement extend it in and out of the page to infinity so you can do your landscape evolution and once you've done your landscape evolution step collapse back all this topography and see it through the tectonic solver so the tectonic solver only sees the kind of average load from this property average between crests and valleys the other thing that I didn't mention is we don't have a fully consistent rule for sediment transport in the position yet so for this I will just infill all the subsciting area assuming that as soon as I create a piece it's filled you can think of that as perhaps that's not all material coming from the football there could be material coming but the question is what relevant landscape parameters can I put so one way to do this is to actually look at landscapes that you believe to be in some kind of steady state and use the string power framework to actually invert for some of these landscape parameters and then simply stick them into the model this makes no assumption as to what k really means for example it's just saying in that language in the framework of that formulation if I can read these parameters in real landscapes then it's probably reasonable to try to see what the model's doing it doesn't presume anything on what the meaning of this so the approach that I've chosen for analysis is basically looking at river profiles in equilibrium the idea being that if you look at the equilibrium river profile going upstream you'll find this characteristic shape that reflects the competition of uplift and erosion but also to a large extent reflects the fact that as you approach your drainage area is reduced so what killer Perron and wiki rodent did is introduce the framework to correct for that it's based on a clever upstream integral of the steady state profile I won't go into details but basically what you can do is transform your upstream distance coordinate into a tight coordinate it's something that basically corrects for the fact that the drainage area changes at the upstream so instead of plotting a profile of upstream distance and elevation you can plot elevation versus this corrected uplift distance takes away the drainage area and the good thing with this is that the slope has to do with the competition of uplift and irritability that's a direct record of that we did this for again the lemai fault in Idaho using some rivers this is an example for rivers in football we plotted them in a tight plot so elevation above base level and guy and assuming that we know the uplift and really this is just an order of magnitude type of exercise extremely precise because there's of course lots of the order of magnitude k that we get from this exercise assuming an overall uplift rate in the football point here with your order 10-5 that's very much in line with the values that were in this morning in the talk so that's encouraged but and by the way this was done with the football toolbox speed of matlab tools is really useful and helpful suite of uh uh software to actually do this really easy so we did this in Idaho we also did this across a number of half-grabbins from different rifts your brown rift and east african rift this is just to highlight the kind of natural variability you see in this slope of elevation versus sky which remember reports the competition of uplift and irritability that's not directly telling you about irritability it's just telling you that natural systems different half-grabbins span different degree of competition but again through a simple order of magnitude exercise what you typically find are irritability coefficient in the range of 10-60 10-4 10-4 per year regardless of what that actually is and of course there's lots of limitations with this approach one being that we're only in the framework of string power that we ignore the spatial variability and uplift rates and we have to assume that uh landscapes are extinct but it's a good place to start so that's how we kind of try to calibrate the landscape part of the coupled model try to fit to something that is reasonable and documented in nature the other part of that is we also need some realistic rheological profiles for the upper crust and the lower crust so rather than doing a perfectly elastic plastic brittle layer now we incorporated a little bit more complex flow loss to describe creep in the lower crust we use the wet part side flow loss which produces strength or peak stresses that are on par with what was inferred from paleo-physometry in the Whipple-Malton complex so that's a way of kind of constraining what you might think is a plausible strength and flow and that's because strength modulates the degree of the feedback between surface processing techniques like I was showing in the simpler model so we really wanted to get this to be a plausible strength for us so here comes the fun part these are the fully coupled models I have three slides they're pretty much the same just showing you the same outline this is the cross-sectional view of the tectonic model you're looking at plastic strains that tells you where irrecoverable deformation is happening tells you where faults form we always start with a fault seeded just to start with a half carbon that's enclosed and then look how it evolves we're pulling at a rate of one millimeter per gear on the sides the black areas the brittle domain the brownish areas the ductile domain that creep in a distributed way and here is a 3D view of the landscape evolution model showing you the development of the biography both from the area shown right up there so they're on the loop but they basically show you how the fault grows creates the biography here's the foot wall here's the angle basin and evolves through time with increasing so this is our reference case with the rotability 10-7 which is lowered and anything that I've documented in the real systems and no sedimentary input if I have a basin in this case I just let it grow as you can see here it makes a pretty big impression and when I do that my initial half-graben quickly becomes extremely unhappy and it wants to localize a new fault in this case to form a full-graben after 3 kilometers so that's the reference case let's keep the irritability the same and just fill in the basin and see what happens in this case what I'm adding is the the code actually tracks the material that's being deposited as these yellow areas that fill the angle block you're still developing sizable topography almost a kilometer well see it's not very strongly incised by rivers but the topography now in the angle is really flat because I force it in that way I just say I've sedimented up to that level and in this case my half-graben is still not stable for very long it's stable just a little bit longer instead of dying after 3 kilometers it turns to a grab and now let's go to a rotability coefficients they're actually documented in the field 10-6 per year in this case with the full sedimentary infill and you can see here the river networks that that incise progressively through the through the football you can see the retreat of the crest of the football and you can see the thickening of the sedimentary basin as it goes and in this case the half-graben is quite happy it's pretty similar result in what we had before except this time we're doing it in much more realistic or at least plausible strength assumptions and landscape boundary conditions that we do some degree calibrating in name and what this new formulation also allows us to do is to really think about what control the behavior of the of the model in relation to topographic growth here we're looking at the relief that developed as a function of increasing extension relief in the range in the football blocks so positive and a basin relief that put the basin as it in all the runs that I showed you before going from super inefficient surface processes that's no erosion no infill it's the blue line so you start increasing the offset on the fault create a football that reaches about 600 meter of relief while you create a basin that's 600 each you do that over 3 kilometers offsets and the star marks where the initial half-graben dies if you slightly increase efficiency of surface processes by filling the basin you no longer get any topography in the basin it's all flat but the football still creates pretty sizable photography that's the black line and it terminates after 40 kilometers of stay and as we get to more realistic values is the ones that are documented in the field this is where we see the half-graben growing indefinitely and reaching a topographic state after a couple of kilometers so the fault is still growing and growing and growing accumulating slips as large as 810 kilometers but the topography now no longer changes this is the case of the movie that I showed you we have another run with even more intense irritability so this process reworking that just reaches topographic states quicker level well that's telling you perhaps it's really topographic steady state and the timing to getting to the correct state that controls the dynamic and in the framework of the force balance that's because as soon as you've reached steady states well you no longer need to spend any more energy on the system you're done your whole topographic contribution is pretty much done and what really matters is whether you can do this quickly relative to the time scales of the system or if you do this slowly relative to the final time scales I have two very quick slides on whether we can see fingerprints of this in uh in real systems I'll go through that pretty quickly this is a compilation of topographic relief versus fault offset across uh different half-graben from the basin range um and New Zealand what's interesting is that a lot of them follow this uh overall trajectory if you had no erosion right the topographic relief would follow structurally to some extent but in this case there seems to be a pretty uniform trajectory which is in line with what the models are predicting telling us that in a lot of these systems you do reach topographic steady state perhaps after a couple of uh extension that's of course a little more complicated than that because these represent varying slip rates varying topologies very much but that's an avenue to look at is to um take different half-graben and I'm going to conclude recapping what we've seen and try to generalize it in most systems the first question that we need to ask is is your tectonic system likely to be susceptible to reworking topography right regardless of what the surface processes are doing all right you have a predisposition to being sensitive to topography removal and that's mainly controlled by um how important in a relative sense is that gravitational energy term relative to the overall dissipation so we could define some kind of dimensionless numbers that describes the rate of increase in energy with increasing deformation related to topography only to the overall increase in energy with increasing retention related to dissipation processes and um that's a dimensionless number that would essentially tell you if you're able to remove all topography this is how much you'd extend the lifespan of a simple and that's promoted in weaker crust because in weaker crust the overall contribution dissipative flexural frictional energy terms become lower and the relative share of the gravitational grams higher so we can have and finally once you've established that your tectonic system is sensitive to topography the question becomes are realistic or at least earth documented rates of topographic reworking enough to do something on the relief so that it can be active and then the time scales of interest become the time scale that it takes to reach topographic state it's function of irritability and uplift sorry irritability and the width of the mountain relative to the internal time scale of what would be the lifespan so again we could define a dimensionless number that has to do with localized efficiency with weaker crust slower tectonic rates greater irritability or faster deposition favoring the high value of this number which means enhanced localization on faults like we're seeing in all these miracle models whether they're stationary wedge or rifts this is my conclusion slides what do we need from coupled models well my answer based on simply this would be we need to really capture the time to topographic steady states and the rates of topographic build-up because once you get to topographic steady states topography build-up no longer has an influence on the evolution system but it's unclear to what extent the height and width of the actual steady state will have an influence on the system it's actually not clear how important it is to really reproduce the fine-scale morphological features like triangle facets like that might be important get additional constraints on your landscape might not be critical for tectonics one thing that we do need to discuss but i've conveniently put the very last slide in my thought is the effect of grids and the brits and solution the fact that if you refine your grid you'll get smaller rivers that are less efficient at eroding the landscape you get overall longer times to steady state greater relief but overall what's key is to use parameterizations in our coupled models that can directly benchmark and calibrate them real landscapes just so we're sure that we're not doing something too crazy either taking too much mass or just not altering that's something that we really need to take to other processes that could be incision hill slopes lateral erosion or threshold effects and i come with it and especially sedimentary so lots to be done so we're gonna take some questions very nice presentation thank you i'm patricia passat from louisiana state university um you show the the effect of sedimentation on you consider the effect that fluids would have on this that's what's our fluids fluids as it's with with higher sedimentation or bringing sediments against the fault zone what do you mean by fluid so fluids water on the so you would have you would you would make it easier to slip on so you're talking about fluid inside the crust yes okay so you could think that in those systems when you're creating a lot of sediment you're actually building the crust as you go you're creating a large part of the crust that has sediment property perhaps high porosity and you're right that that might do two things that might help fluids percolate into the system change the effective stresses on the fault but on the other hand you're also increasing the burden of increase the strength of the fault so anything possible yeah i think i'm listening to you up and go i thanks so much for you know carrying this down to the essential physics to driving the processes and and but given the nature of the meeting and the way that brings communities together i wondered if you could share with the group here where your tectonic models have strong fits with real observations say the time sequence of strain distribution and versus and where there are our large misfits maybe other aspects of the process that we're not understanding and and specifically you're taking a half robin model the flexural model is the crustal strain the crustal thickness varies it's consistent and it's a time evolution the fault system is consistent with that tectonic part what do we know what helps me know very well when you compare your models with true observations obviously just going to answer a fraction of that because there should be a lot to talk about but there are the cool thing with working with half-grabbins is like you said there's many observable different strains so one of them being the wavelength of flexure that is something that we do reproduce fairly well the fact that topography over a roughly 20 kilometer wavelength is something that's seen uh current rift seen in the basin and range i'd seen it's seen in the number of places that's a measure of increased strength from the crust from a flexural perspective you could also look at here the degree of strain of pollution you can see this kind of secondary faults that do exist but never take over and if you think of half-grabbins in the east african rift for example there's these classic studies of how much strain being taken up on the master fault maybe 50 percent the rest being partitioned into the hanging wall that's something we do see to some extent here we do see a lot of little secondary faults that never quite went what you've shown doesn't match we just said the east african rift that won't have with the patterns and i mean that's okay because you know first scale but when you actually look at dimensions and other aspects it doesn't and i i think that's not that the point that i was trying to make it's it's about the details how are we going to work forward to evaluate what parts we know well and what we don't and what data sets you know like the questions we had this morning and i'm i'm just wondering if they have basins and i the reason i want to caution the tiny bit is that i saw in some of your diagrams that you've taken topography say from the ruins worries where there are two faults on both sides and glacial on the top yes and there are multiple processes to be taking care in the way that you benchmark observations is going to be really important to teasing out that's the coupling between these two processes so um and we've gone so far but can we can we do some of these benchmarking benchmarking exercises like that actually calibration exercises or that's when the choice has to be made right how much do you want to fit so might take it's the approach i've taken is to really fit the first order integrated proxies like like for wavelength as a proxy for flexibility the overall average relief that you get the degree of strain localization and the amount of offset and perhaps the pattern of where the next whole break might be another whether i could or should push it to a finer detail maybe not maybe not at definitely not at this stage but yeah that's we must be having this discussion for techniques in the same way we're having it yes a little polite to know from the universe of Pittsburgh and i was uh quite impressed by your tectonic model and one thing i wanted to ask is how much sort of the horizontal displacement on because you're in an extensional setting and in your surface mode i saw your input is up and down and you reach very easily steady statement up and down yeah what if you have that's a great point i didn't talk about it but it's there so the lateral infection is there you can see the spot actually migrate migrating and rotating a little bit part of the retreat of the crest is actually that effect the whole fault is moving and retreating so i'm not just feeding the vertical component i'm also effecting the landscape and stretching the landscape which is something that a lot of landscape evolution models don't do but it's hard to track the horizontal set that is there and so you're right the topographic steady state is mainly in a vertical sense but at some point the river profiles on either end do tend to balance each other out and you do get to a steady state also with respect to lateral lateral variation but yeah it's it's in there okay thank you george let's thank george once again we're gonna go on