 Well, let's go ahead and get started. I get the good fortune today of introducing Yvall Schmalovitz, who's going to be giving our webinar today. He has a master's degree from the Hebrew University of Jerusalem. And he's now finishing up his PhD there. And I've had the good fortune of working with Yvall over the last year. I've learned so much from him around Air Atlanta and the Scarborough Retreat. And so I'm really looking forward to his talk today where he's going to be talking about some of that work that he's been doing. So I'll hand it off to you, Yvall. Thanks, Matt. I will show the screen. OK, so hello, Yvall. It's very nice to be here. And so my name is Yvall, and I am a PhD student from the Hebrew University of Jerusalem in Israel. And today, I will discuss a work that I've been doing with colleagues from CU Boulder, from Utah State University, and from the Hebrew University of Jerusalem. So the title of the talk is Modeling Long-Term Arid Cliffs and Subcliffs of Evolution under the Shodoration Extreme Restroom Events. So in this project, we focus on arid clips and the slope for main processes. And we're also working on integrating rainfall forcing at the scale of individual rainstorm into landscape evolution modeling. So I will begin. And as in the picture, the satellite picture that you have on your screen, you can see Sinai Peninsula. OK, so you can see Sinai Peninsula with the red or black colors of crystalline rocks from the Precambrian crystalline rocks from the Arabian Nubian Shield. On top of these rocks, you can see a yellowish, many carbonite rock that these hard carbonite rocks form roughly 100 kilometers of escarpment with cliffs. Also in this picture, you can see an individual cloud in the blue sky. This way to intermediate develop a cloud and similar clouds like this one, in maybe in more advanced stage, are responsible for significant and large rainstorm events in deserts. Such events also induce significant hydro-geomorphic responses. And this project is aimed to couple this discrete rainstorm forcing with the long-term evolution of cliffs in deserts. So cliffs are a very common land from dry areas. These features are often developed in areas where there are sequences of sedimentary or volcanic rocks. Usually, the more resistant layers create these steep caprocks that laid over much softer lower and erodible layers. Through time, the cliff is retreated from its local base level, and sub-cliff slope or talus slope or active slopes are developed beneath these cliffs. These slopes are also covered by sediment that delivered from cliffs. And this sediment on top of the sub-cliff slopes are experienced weathering, and through time, they are getting into smaller and smaller fine pieces. The longitude profile of the slope is usually is trending from being concave, so it slightly convex normally. So such features, cliff features, are actually common in many of the arid lands all over the world. This map shows some example for larger cliff systems. We talk already about the Sinai escarpment, and we talk more in the presentation on mega sites. Both of them are experiencing today a hot climate, but we have cliffs also, for example, in the Colorado Plateau that experience in a generally cold desert climate, arid climate. Obviously, this climatic context of cliffs all over the world affect the processes and the forms of these cliffs and sub-cliff slopes. Some specific examples from the map that I just showed, here is the Mesa Verde in Colorado, and you can see here a clear contact between the Cape Rock and the lower layer slope. In other cases, other examples such as the Ethiopian Highlands or the southern Morocco example, the contact is much less clear, and there is some height of Talus accumulation or sub-cliff slope accumulation that still receive material from the cliff. Also noticeable in the picture is the weathering pattern of the cliff that is varied between different sites. For example, in this site in Spain, you can see these large borders that weather from the cliff. While in other places, the material supply from the cliff are much finer in the scale of the sentiment. So research on desert cliff is, of course, motivated by the dominant areas. In addition to that, surface processes along cliffs such as rockfall, debris flow, flash flood, impact infrastructure, and also civilization nowadays. And in addition to these two reasons, researchers were historically attracted to cliffs for studying climatic or base-level changes because they are a relative systematic geomorphic system and they are also sensitive to the local conditions. So over the years, there are many studies regarding cliff and sub-cliff slopes. Some of them have been achieved to correlate between sub-cliff slopes evolution or detached slopes evolution with climatic focusing at different scales or climatic oscillation at different scales such as glacial or interglacial cycles. One of the general conceptual model claimed that during wet phase, the sub-cliff slope form a relatively continuous slope, undisected slope, while alternation into dry conditions and detached this continuous slope from the cliff to what we sometimes call the Talus flat iron. And then the cliff is turning to more sediment-starved However, still, this topic is really on debate. The correlation to general climatic condition is not well-established everywhere. And the understanding of the coupling between sediment supply, weathering, sediment transport on the slope, and climate is still poorly understandable. Nowadays, many of the cliffs around the water are actually located inside the climatic gradient. Here is an example from the booklets in Utah. You can see this roughly 100 kilometer escarpment. You can see the really wet and vegetative, relatively wet and vegetative part in the western part relative to the southwestern part. And also, by looking on the picture from the field, you can see there are differences in morphology. You can see by eye regarding the cliff thickness, the slope, and so on. On a finer scale, this is the central nagel. We are looking on a 40 kilometer escarpment that is also lying today within a climatic gradient. The northeastern part of this crater or this escarpment is receiving in a year about 40 to 50 millimeters, while the southwestern edge of the escarpment receive around 100 millimeters. I guess for some of you that work in much more human places, this difference of 40 or 50 millimeter in a year don't sound a lot. But in desert, it can alternate the vegetation pattern. It can induce differences in the hydrological. And geomorphic responses. This is our two pictures from two edges of the Ramon escarpment. And I should say that along this escarpment, we don't see a significant variation of the pathology. But as you can see, there are differences in the morphology. So this side is the drier side. You can see the thickness of the cliff is about 50 meters. And the debris cover is really relatively limited, while in the more wetter side, the cliff is more covered by debris. And the thickness of the debris seems to be much higher. So the point of these two pictures is that although there are relatively minor changes in the annual rainfall, as I said, about 40, 50 millimeters in a year, we do notice significant changes that are related probably to climate. And even though everything is inside the arid and even the hyper arid rain. Going deeper, even to the microclimate scale, this is a site from the Eastern Hegel. It's called Tsuktamur. Here, the caprock is a church. And the softer white rock or chalk. And in this site, we see that there are differences in aspects of the cliff retreat and the cliff morphology that we claim the effect of the microclimate conditions. So generally, we see an increase in the cliff retreat toward the south facing aspect. And we also observed changes in the morphology of the cliff, of the south cliff slope. And even in the long-term transport rate of sediments that we obtained by cosmogenic nucleus. We, as I said, we consider this change to be affected mainly by the microclimate conditions along the south relative to the north facing aspect. So arid landscape and arid catchment in general are influenced by rainfall characteristics. Rainfall in arid area is quite unique. And it is firstly characterized by high spatial and also temporal variability. An example for that, you can see the map on the left side. This is a map from the Wollongad experimental watershed. And this is isoed from rain gauges. And you can see that along the 100 meters, there is a decreasing rainfall about 40%. And another example for that is from the Renat 2021 paper. This is a radar map for three days rainfall, a single event. And you can notice on the very spotty rainfall calls in the large gradient space, there are places with 150 millimeters during the storm, while in few kilometers away from there, the rainfall is almost zero. This spatial variability in rainfall is in part a result of the convective nature of the rainfall in this area. So this is an example for convective sale from the Easternegev as well. You can see first how local it is. This is a scale of three kilometers. And first you can see how fast rainfall are decreasing, rainfall intensity are decreasing from the center outside. This convective nature of rainfall also responsible for very high rainfall intensities, and especially for the shorter section. So the diagram on the left is connected to the rainfall intensity, the retained period for the rainfall intensity, and different durations. You can see that the red curve, which is for the arid area, are diverged from the two other curves for the two other climate zones, which means that arid area are characterized for the rare return levels, is characterized by very high rainfall intensity, especially for the short durations. So these short durations intensities are important, and especially for several processes. Many of the pioneer works in the deserts, experimental studies, but also more recent studies have pointed out that rainfall intensity and the profile of rainfall intensities within stones is influenced the infiltration rate, it's influenced the rainfall concentration and generation, and also the exceeding of erosion pressures. I want to give an example for that by looking on this video and rather on the random map. So this combined video will show on the left rather maps for a very short event in the eastern agave. And the video on the right will show a field picture from a timer's camera that has been installed in the field during this storm. And the picture are correspond to each other, the picture show exactly the same minutes as the rather data. So at first, you can notice this is a look downwards of the slope, a Talu slope in the Tsukhtamur table mountain. And you can see that before the rain comes, the Talu slope is totally dry. Whenever the storm starts to arrive the area, you can see this very high intensity is all over. And if you can see there are small rurals and gullies that fill with water. After the storms ends, the water infiltrates and about five to seven minutes after the rainfall arrived, the slope is dry again. So up to now I was trying to summarize all the challenges for coupling surface processes or arid landscapes to specific climatic conditions in the desert. And I mentioned the high special temporal variability of the rainfall, complex hydrometeorological responses, the evaluation of extreme events, and also I should mention that the fact the records are limited in space and also in time. This is all together bring us to the reasoning that the processes on dry land cliffs and slopes meet storm scale examination and approximation are often not valid in such areas. However, this brings a great challenge because a storm scale forcing is obviously in scale of minutes or even smaller than minutes, but landscape evolution is of course in much larger timescale. This brings a lot of challenges including computational power and so on. So to tackle this challenge we, I will present our agenda or the main approach so first we collect rainfall and topographic data in high resolution and then we are design storms for different and brilliant and I will elaborate about this in the next slide. And our final or last stage is to develop a landscape evolution model that can be fed by the design storm. So what is the design storm? We call it density duration frequency design storm I will explain it by an example. So we can have a diagram as we saw before when we have the rainfall densities, the return level and different attempt periods. Using these curves we can build that conceptual storm design one or artificial one that is correspond to specific attempt period. It can be the five year attempt period, 50 year or 100 year attempt period. By that we actually design storm that is correspond to specific attempt level and then if we simulate this storm we can bridge the gap and evaluate the impact of extreme event on the landscape if we are using the appropriate model. So our next mission was to develop a model and we do it using the LandLab toolkit. I will try to give a general background or introduction into the model components now. So the model is in general has two units, the Kepprock units and another erodible, much more erodible layer. We also include a soil layer which is much thin layer that is mainly developed from a cliff derived sediment and it's up to a few meters in depth. The hydrological component in the model are the infiltration by grain and up concept and runoff generation is we are using diffusion wave approximation for overland flow. The fusion wave approximation is actually the simplification of the shallow water equation that submits the momentum term and the flow of velocity is calculated based on the linearization of the Manning equation. This both components are already available LandLab toolkit. The soil layer is represented here by 30 size classes of soil fragments or soil crusts. We include here an important feature of Aritz-Locke which is classed for the mutation. We done it by the concept that we follow or adopt the concept of wells and the coin of the soil for the mutation. It's mainly built on some transition matrix that transfer mass of classed from specific sites fraction to another one. And I really hope that soon our version to this model will be free and available in LandLab repository soon. The sediment transport of the class is done by is calculated or determined by the shear stress exerted by the overland flow. We choose to include here a selective transport scheme in which the transport of a given size fraction is not depends only on the size itself but also on the size of the surrounding grains. The total flux from a specific node will be the sum of fluxes for all the size fractions and for all the size classes of fragments that are included in the specific nodes on the slope. There are two additional components in the model. One is the incision. Sorry, I will say that this component I hope that also will be available in the repository. And our two additional component one is incision of the soft rock. We done it by the already existing component of the depth-slope water erosion and we also included a heat-slope diffusion using the linear diffusion component in the repository as well. Okay, so the next process that we invest much to represent is cliff retreat. And here we are including two types of retreat. One is cliff weathering, what we call kind of grain by grain weathering. This process will act on all the cliff sections that are not covered by sediment. So all the cliff sections that are not covered by cement are experienced in weathering, they are weathered in a parallel way. And then we determine the critical node following some concept of the highland component of Benjamin Campfortz. So we define a critical node and then all the sediment that has been eroded above this critical node are run out using the highland component run out algorithm. So they are run out in the kind of landslide like way. So the sediment are run out after weathering the course and they are deposited beneath the slope and they form a tall slope. The second process of retreating is cliff under mining. So we assume that whenever a softer layer is exposed and the contact between the cap rock and the lower layer is exposed, we oil the cliff section above is collapsed according to some determined structure plan and the sediment are run out again using the highland component downslope. This type of retreating is maybe some of you are familiar with this, maybe it's similar to the conceptual model of Coons that determine the cliff retreat would be mainly as a function of the evacuation of the Talus debris and the exposure of the softer layer. So I would like to summarize some of the advantage of the approach or the model we developed here. So first we have explicit representation of ideology. Second we calculate the sediment flux for shear stress and we also include glass selected transport scheme. We are using the landslide-like sediment run out of the Camp Force 2020 and we also include two type of retreat mechanism. Using this model we now share with us some of the results regarding two main research questions. One is what are the main factors that drive diversity of added cliff and sub cliff slope morphology and second what are the conditions under which cliffs becomes buried or under. So the main simulation scheme is more like in the hydrological modeling way. We do it event-based. That's mean that we are simulate the 100 year storm a 1000 time. This should be equivalent to 100,000 years in total. So the 100 year storm, we define the base storm. This base storm is based on statistical analysis of rainfall in the central mega. You can see here in the graph of the rainfall intensity just notice that the scaling is flipped and based on the statistics, we use this storm as a base case. The cliffs to determine the factor that govern what makes cliff to be buried or not. We played with many parameters regarding the model. So it could be the cliff inclination and the sediment size from the cliff, the lower layer gradient and so on. I will focus here today mainly on the rainfall intensities and on the size of sediment supplied from the cliffs. To emphasize why it is important to include the intensities pattern and not just the constant intensity storm, I will use this example. So here we have the diagram as before with the hydrograph of the 100 year storm and we have the water depth in blue regarding this storm. In dash to black line, it's a constant intensity storm and the corresponding water depth. We can translate the water depth into a critical grain size that's meant the critical grain size for motion. So we can do it for all the hydrograph and then we have in y-axis we translated the hydrograph the water depth to critical grain size. Now we can ask ourselves what's really happened to a 30 millimeter class. So you can notice here that if we are talking about the constant intensity storm, the class will not mobilize because the critical grain size of the storm is lower than 30 millimeter plus while time varying intensities, the peak is getting higher and therefore this sediment is predicted to move. Integration of many of such moments may result in totally different morphology and this is why density fluctuation are important. Okay, so the first test for the model was to check it against the measurement and we used an analytical solution for cliff degradation, one of the Fischer lemma and one of the Baker and Leroux and both of these analytical solution were tested at the field by Hodgson 1998. Their model assumption was that weathering produced an equal retreat of all parts of the exposed free phase as we assume here. The resulting debris accumulated at the cliff base and cliff bedrock that is covered by sediment is protected from the future weather. So here is the result, the numerical simulation relative to the analytical solutions. So you see the both analytical solutions of the bedrock, this black curve of Fischer lemma and the Baker and Leroux, both of them are here and the pluses are the numerical solution of the model. Assuming we force the model that will be no transport as the analytical model require. So the acceptance between these two make us to trust the model and maybe prove that it present a relatively realistic pattern of cliff retreat and then we were motivated to move on. The next test was to examine the model against the field data in different conditions. So you saw this picture already, this is from the Ramon escarpment and what I didn't say to you before but we also have measurement except of picture, we have measurements and we know that from the north-eastern side there are smaller grain sizes and there are higher extreme intensities that we calculate from reinforcement statistical analysis. This is relatively to the southwestern side where we recognize that are much larger grain sizes from the cliff and lower extreme intensities. So our test was to simulate the model with disposed cases and to see if we can get similar morphology of cliff and sub cliff stop as we observe today at the field. So just before I go into the simulation videos I want to give you I was to invest one moment to just show description how to read this video so we'll have quite funny diagrams like that when the red lines is the initial conditions and through time we'll have additional lines the blue lines will be the topography the grey line will be the bedrock and green line will be the the logical contact. So this is our videos from Ramon Ramon like or central negative like simulations at the left we have the case when we include lower intensities and larger cliff derive class size this is the you see the legend B50 for each node within the model and in the right figure you see the eastern north eastern site like case when higher intensity and cliff supply smaller class and you can see that the cliff from Ramon site it start to get buried whenever relatively to the quasi steady state of the north eastern site like so this diagram our end of simulation figures the red line is the end of simulation for both cases and this is observations as I showed you before we can compare the cliff height and you can see that the model the pattern the model predictor quite similar we have thinner cliff in the west and larger cliff in the south in the northeast and also the grain size pattern are similar the blue lines here are grain size according to the second one y axis and this grain size of fish node can be compared and the trend and also the numbers are quite similar yes ok so we want to have more systematic examination of the of the rule of grain size and here we conduct to experiment all the parameters are identical but the differences are in the sediment of grain size supplied from the cliff so the right figure we are looking on cases where the cliff provide block in a scale of one meter to the slope and in the left case the cliff provide much smaller material and you can see them how the cliff is getting buried in the right case relative to the to the left case we can also look on how the grain size change through the simulations so this is base time diagram and this is the time and this is distance from the lower node upwards so if we are standing on this node let's say this is the 50 meter node from the outlet and we are going upwards and you can see that we had kind of a warp change in the sediment side this is because of kind of landslide or rockfall and then we have slowly reducing in grain size due to soil or class fragmentation we also see the cliff retreat here as different nodes that didn't have any material on them because they were cliff now retreating and now delivered large material is delivered to these nodes we can also look on the distribution in each node at the first like between stone 1 to 300 there is a warp change because of this let's say rockfall and the grain size is much closer and then it's go fine again and we end up with a relatively nice distribution of grain sizes next we want to examine the rule of rainfall intensity here we we act in kind of a naive way we use our base scenario case of central negative case and we change it by a factor and so here are two cases where the factor we reduce the storm intensity by half and this is where we increase by half and you can see that this kind reaches steady state where the cliff sediment are not really getting buried cliff while in this case obviously maybe sediment are just deposited and the cliff is getting thinner no incision this case relative to this case where we start to get incision of the panel taking these two factor together so we have in one side the cliff sediment side this is the sediment size of cliff delivered from the cliff to the slope and the other case is the rainfall intensity factor we can now look on different combination of these two parameters and we the contour map is the end of simulation cliff height and this is an I think it's a nice way to see that under high rainfall intensities rainfall the end of simulation cliff height is relatively thick and while if the grain sizes supplied from the cliff are larger or the rainfall intensity is lower the cliff is the end of simulation free face cliff is relatively thin and we have many of combinations in some of example we have many of them of different combination of the clear grain size to the specific rainfall intensity factor interesting to see that in this example there is no cases where the lithological contact is exposed because the lithological contact is here but in the reality we do have such case like in the Soviet right we have a case where there is a gap and we obviously see that the cliff is retreating along maybe along this lithological contact and a possible explanation for that could be the initial cliff height so here are two here is a comparison of two simulations the initial cliff height of 15 meter and everything else except of the initial cliff height is similar but as you can see in the right example the cliff is retreating along the lithological contact because the overland flow is able to evacuate all the material supply from the cliff in the other case we have much the flux from the cliff is much larger because the cliff itself is much larger and then the cliff is actually started maybe to buried a bit if you are looking if we are looking on the same diagram as before but now I just normalize the end of simulation height by the initial height that's mean that 100% that's mean that the end of simulation would be exactly as the height of the cliff is exactly as was in the start of the simulation and we have these two maps we have two cases one with 15 meter initial height and one with 15 meter initial height so and this area is the much interesting area we can see that the more red colors here or the darker red colors here relative to this side that's mean that the end of simulation cliff height is it's very similar to the initial cliff height and this result of the undermining pattern of cliff retreat relative to this case when the cliff reaches kind of quasi steady state in time going even deeper on the pattern of cliff retreat we can look at the diagram of the x axis is time or another of storms and on y axis there is the cliff height and we can see how in this specific simulation it's a specific simulation of rainfall intensities and sediment size supply from the cliff and the cliff is reached a quasi steady state relatively fast there is the opposite case where the cliff is very very fast get buried and then we end up the simulations and we have many other cases and more complex cases and we can recognize three main patterns of retreat the pink pattern is the pattern of reaching relatively fast the quasi steady state the cliff is the transport power is much more efficient relative to the flux from the cliff and therefore the cliff reaches quasi steady state quite fast the blue pattern is the one that the cliff is getting buried really fast while the grey pattern is the one that the cliff is started to get buried but then at some point because the slope is increasing the runoff is also increasing and therefore it's emerge or it's turn into this quasi steady state stage we can put numbers on this and to recognize in under which cases cliff is getting buried or under which cases is staying let's say alive so mainly we can see that for cliff getting buried under this analysis that is similar to the central negative case rainfall intensities that are equal or smaller than what we have today but if the cliff will supply a 50 centimeter grain sizes it will ultimately get buried we can look on a similar diagram but now for a case of thin initial cliff before it was for 50 meters and now it's for 15 meters here we see only two patterns one pattern that the cliff ends up with a similar height as it started with and this is the under model pattern and another case is when the cliff is getting buried relatively fast also here we can put numbers and we can see that if the cliff initial height was relatively seen in this kind of few meters rainfall intensity smaller than today will ultimately get him to be buried the last thing that I want to show you this is more a very preliminary result is to see what will be the impact of changes in this extreme intensity in time so this is also quite a naive but we take the 1000 stones that we simulate and we split into half half of them 50,000 years period the lower intensities and then we have other section with higher intensities so we can look on them on the diagram the video that I guess you are familiar with already and you can see that around 500 stones sediment are getting transport much more fastly and ultimately the sediment reaches a stage of sediment star so a possible that changes in the brainstorming intensities could alter the cliff from two different geomorphic mode from sediment accumulation to sediment star okay I'll reach you to the conclusions so first we developed here a cliff model, a cliff slope model that can account for observed characteristics and diversity of other cliffs that are cliff slope deserts the model captured geomorphic analytical solution for bedrock below a retreat cliff we found that the cliff slope forms are highly depends on cliff the sediment size and the storm intensities and we saw a high dependency of the cliff retreat pattern to the cliff rock thickness we also have some things that increase in reference intensities may alternate the geomorphic modes of the cliff from accumulation to sediment star phase under future work we have much to do first we want to get into deeper into the rainfall forcing we want to change the rainfall patterns to their frequency and duration and so on and we also want to incorporate rainfall from different return levels not only the 100 and we want to explore more field size and other cases and the next major step will be to make simulation in 2D now we do it effectively in kind of 1D and then explore healing and gutting pattern below the cliffs and that's it I will say thanks again to Mark and also people that helped with field measurements and of course all the colleagues that helped me with the work and I will invite you to join our session the upcoming EGU you can scan this back or just submit according to the type that's it well we can open it up for questions now you can either raise your hand and ask your questions directly to you all or I will be monitoring the chat for any questions that appear there to pick us off I can ask a question which is what are your thoughts and you touched a little bit on your next steps in the future but what are your thoughts in terms of taking the design storm approach towards simulating sediment transport so using one recurrence time and how do you think you're going to be able to test thinking about a whole population distribution of events by taking such an approach yeah that's true so now we are actually simulating we are actually simulating all the time we repeat on the same storm this is the 100 year storm that is based on well this is based on rare fault statistics from the mega and yeah it's quite an naive approach but we just repeat this storm assuming no chance in the intensities we accept I think that first it's reasonable in some way because we know that sediment transport in deserts happen they're not happen often at least so we need to examine what exactly is the threshold like what is right what is the storm return level threshold for this transport and we don't know if it's exactly if it's the 100 years or not we do have some simulation results from a paper in 2020 it's a different site that I mentioned there we know that the frequency for mobilization is around 100 years so let's say if it was a tamur case I guess the effect of including lower return level will not have much because we know that like we calculate or we have some conclusion that the sediment transport occurs during this rare storm at least 100 years storm but it's probably different all over the world right so yeah it could be it could alternate the geomorphic modes for sure and we this is one of the next tasks to include storms from different return levels and to examine actually if they have impact that is important to have interlocked simulation or not thanks Benjamin I'm going to ask you a question no we can't hear you Benjamin now you are muted you're still on mute Benjamin I guess it's typing I cannot answer you can you hear me now yeah okay I couldn't unmute I think it was a setting or something so thanks for changing that so hi you've all been good to see you man very nice talk and very rich in experiments and data super nice I was just wondering and you talked a bit about that but you have like higher precipitation amounts you see more evacuation of your sediment and your cliff gets taller but I was wondering how like can you weigh the difference between the total precipitation versus an increase in storm intensities so I'm kind of wondering how you set your threshold to consider the impact of like big storms versus a lot of medium size smaller events yeah so actually we don't have it here now like this let's say this kind of scenarios are assuming that the most effective or the most geomorphic effective events are the 100 extremes the 100 years storms this is our assumption and this is something that came out from other walks in the negative and also from other people over the world that extreme storm are let's say more let's say important in terms of geomorphic differences and I think if I connect even better to what you are saying that we have we need to improve the fact that maybe the total rainfall may not only affect a lot of generations but it could alternate also the weathering pattern let's say how specifically the cliff is retreating and for now we don't have like a mechanistically way to do it we like the weathering pattern and the weathering class size and so on there are parameters so my solution to your question is maybe first to include maybe to as I said to Matt maybe to include rainstorm with lower intensities and dissimulation as well and also maybe to find some connection between the total rainfall and maybe the weathering patterns and then maybe we will have some findings regarding this change but here we are assuming that the most effective changes will occur under extreme storms very clear thank you Albert Thank you for your awesome talk I have two questions since you are simulating over long long time skills it could happen for example that whole climate regime changes right so there could be a desert environment that changes into a more temporal and therefore you know plant species might take a more important role have you thought about somehow hooking up maybe I'm not sure if this possible with other landlap components so that they start to you know interact basically with your component when such happens so that's the first question and the second question is maybe a little bit related about chemical erosion I can imagine that if you change your rain intensity there could also be a change in chemical weathering could that impact any of your outcomes you are generating right now yeah so I think the most of the question I relate and also I think it's related to Benjamin I understand right so yeah we don't have now the connection between the let's say the mean conditions the mean climate conditions and the process that happened on the slope actually I should say that we do have it but we don't have it in a physical way like these are parameters let's say the weathering patterns the retreat rate there are all parameters in the model actually the only parameter that are not physically based overland flow the full ideology is a physical based but there are kind of external forcing like what is the patterns of retreating and what are the patterns of grain sizes which are all depends as you said on the general climate the mean conditions so we can examine actually we can I guess yeah it will be easy to simulate to make an experiment of different climate let's say a more humid climate when the weathering is maybe in greater and the cluster I don't know maybe smaller or larger we can think about it but this is the way that now we can include such a chance like we don't have a really physical way now to do it okay thank you any other questions well let's thank you all for just an awesome talk and sharing this with us today yeah thank you everyone thanks for being here