 Lock Benjamin Comfort, who is gonna bring everybody up in the drainage basins to higher grounds, and we'll talk about landslides to slide or not to slide. Hello everyone, can you hear me fine? We can hear you. So thanks Irina, and I want to acknowledge my co-authors on this story, and especially Charlie, with whom I collaborate closely to get these things together. So I will start with the question for all of you. And the question is, which process mobilizes most sediment? Is it either rivers or are it landslides? And the answer is not so simple, and it probably depends on the time scales you're studying. But one thing is for sure, landslides do mobilize a lot of sediment. And this is illustrated on this picture where you see the effect of co-osasmic landslides triggered by magnitude 6.7 earthquake in the Hokkaido Island of Japan. And not only is it obvious that these landslides mobilize a lot of sediment, they only completely changed the landscape and the outlook of the landscape. And yet, in most of our current landscape evolution models, we do not explicitly simulate the impact of these landslides. And that kind of makes sense, because if you look at landscape evolution models simulated with only rivers, they look realistic. And that's because of the time scale on which we are running them. However, during the last decades, there has been many papers saying that landslides do affect channel morphology and landscapes in general. And also that landslides are a main driver of landscape evolution and a dominant source of sediment. Which is why we developed Highlands. And Highlands is a hybrid landscape evolution model, which does simulate the interaction between landslides and sediment dynamics. And it's hybrid in the sense that it combines both fluvial processes and landslide processes at every point in the landscape. Now, I'm not going into the details here, but what's important to understand is that it exists out of two components. There's a fluvial component based on work published by Charlie in 2017 in GNB. And basically the river model simulates rivers so that they can dynamically switch between detachment-limited incision and transport-limited states. Second component are landslides, and we simulate the area and the volume of landslides using the Morcunal stability analysis and then we route the sediment in the landscapes using multiple flow direction algorithm and we deposit them based on nonlinear function as a function of hill slope gradients. And if you want to know more about the model, I would invite you to go and check our GMD paper, which is currently under review. So I will show how Highlands works through a series of three movies. And in the first movie, what we're going to do is we're going to evolve a landscape towards a steady state. So what you see here in the left upper plot is the landscape going towards steady state. The blue line indicates the location of the river. Right-hand side you see the same landscape but now the sediment accumulating in time. The more red, the more sediment. And in the bottom, you see the profile of the river and you see that it's evolving towards a concave up steady state shape. And the orange line indicates that thickness of the sediment, which is also going to evolve towards a constant thickness. Now what we'll do now is we will impose a period of 100 years of landslide activity on this steady state model. And you can think of it as a period of increased seismic activity. So what you see here in the left upper plot are the black diamonds showing the location of those landslides. In the right plot, you see the sediment being produced by the landslides. And in the bottom, you see the impact of landslides on the river profile. And you see both the orange bumps and the gray bumps and the orange bumps are landslides produced dams consisting out of sediment, whereas the gray bumps are irregularities in the bedroom profile, which result actually from drainage reorganization as a consequence of landslide activity going on. And the blue areas represent water bodies being accumulated behind the landslide dams. So that was our intense period of landslide activity. And what we're going to do now is we're going to re-establish a steady state. So we take this model output and we run it again to steady state. So again, in the left upper plot, you see elevation. You see the sediment thickness in the right plot. And then in the bottom, you see the river profile. And slowly, you see that the sediment is going to be evacuated from the river profile. And what's also interesting to see is that also this backdrop irregularities are going to be smoothed out. So I increased the speed of the movie here. So you will see more frames in a number of seconds. So what you will see is that effectively, these gray bumps in the backdrop will propagate upstream in your system as nick points. So this is auto-genetic formation of nick points because of landslide activity. So there you go. It goes to a steady state. And something else I want to highlight is that you can use actual DEMs to feed into highlands. And we did that for the Namshabarwa region in Tibet, China. And what you will see here is again, landslides being formed in the left-hand side. So that's the black diamonds. The red areas represent erosion and the blue ones represent deposition. And in the right-hand figure, what you see is sediment being accumulated through time as a consequence of this landslide activity. And if you would look at this in time snapshots, what you see is something like this. Again, the red colors represent the area of the landslides and the blue colors represent the deposition. Now surprisingly, even without tweaking model parameters to a large extent, we were able to reconstruct these relationships, magnitude, frequency distributions for observed landslides in that area. And even the area-volume relationships were quite realistically simulated. So left-hand plots, the red dots are what we simulated. The black and the gray lines are what we have measured there from other data sets. And in the right-hand plot, the area-volume relationship, the gray band is what we measured. The red dots are again what we simulated. So basically we develop violence as a tool to answer all kinds of questions where you want to look into the interaction between sediment dynamics and fluvial processes. And maybe at one time, to answer this challenging question, if it's rivers or landslides, which in the end mobilizes the most sediment. So that's it. Thanks a lot.