 Thank you so much. It's a pleasure being here. Thank you James for inviting me to give this talk. So today I'm going to focus more on the process base rather than a lot of modeling because there are other talks during this conference. So what I have here is some research that we are doing in South America. So we are going to go from a cold region to a little bit tropical region. It's in Peru. Here, so since we are going to talk about morphodynamics, let me show you what kind of morphodynamic plan for features that we have in the Amazon region. So just to locate yourself, this is the Maranjon River and this is the Ukayali River and this is the Amazon River where it's born in a city called Nauta. Nauta is somewhere here. So and one of the largest cities in Peru is called Iquitos, which is right there. So basically, my point of, let's say, that my field study is in this region for now. So the Maranjon River is an abranching channel. As you can see here, well there is also the Amazon River that is an abranching channel. So basically what is an abranching channel or the system is when you have a main channel with additional secondary channels, which is kind of different from a truly meandering channel which only has one single channel. So the dynamics are different and we have shown that the dynamics of the meandering channels are more, let's say, the rate of migration are higher than the abranching systems. So but these two systems coexist in the same region. So I have to show this visualization, one of the best that I have seen, just to illustrate how a meandering channel evolves. So you have a flat plane and then you have a single thread channel that migrates, depending on the morphoenamic and hydroenamic condition, it will migrate in hundreds or thousands of years. So what is interesting from this illustration is that the meandering channels migrate and then they have to cut themselves. Basically the cutouts will happen in order to have kind of dynamic equilibrium. So this river itself, if there is no change in the tectonics or geological settings or anything, this river will try to get a dynamic equilibrium condition. So let me show you what I think is a conceptual model for a bend that is going to migrate. So we have, this is a plan for view, one ab and c2, and this is the profile view. So basically, let's assume that this initial channel is under equilibrium condition. So basically the slope that you see here is under equilibrium condition. So if the river starts to migrate, let's say in these two bends, ab and bc, start to migrate, what is going to happen is that the distance between ab and b, of course, is going to be longer. So you will have a reduced reduction in the slope, same for bc here. So what is going to happen is, well, c2 is the same because nothing has migrated. Let's assume that nothing has migrated. So we have now difference in the slope of the meandering setting. So we have an initial equilibrium condition that right now we have a different equilibrium condition, so a different stage. So basically the question is how the river will try to go back to this equilibrium condition. If there is no increase in sediment transport here, let's say there is no more load of sediment, either suspended or bed low, this will take a long time to reach, let's say, to deposit all these flat planes and get a parallel slope to the initial condition. However, so basically what we have here is a depositional system. However, if we go back and illustrate the conceptual model for a cutoff, again, 1, ab and 2, so let's assume that this setting is also under equilibrium condition, this is the slope here. So if a and b are, there is a cutoff here, a bend is, let's say, abandoning this area, it's abandoning this ridge. So what is going to happen is that b will be closer to a, so basically you have an increase on the slope. From b to 2 is the same slope, so it's parallel to the blue line, to the original blue line. So the question is how the river now goes, try to go back to the equilibrium condition. So in this case it's much easier because if let's assume that the dash yellow line is the equilibrium condition that is going to try to reach the river. So in this case, what is going to happen, you will have a wave, an erosional wave going, traveling farther upstream and a depositional wave traveling farther downstream. So by the combination of these very simplified conceptual models, a bend migration and a cutoff, you will have, let's say, the equilibrium condition that any river might do. So if we see large systems like the Ukayali River, this is the tributary, one of the main tributaries of the Amazon River. Remember the Ukayali and the Maranjon form the Amazon River. So this is the Ukayali, farther upstream from the confluence. But you can see here, this is a picture from 1975, 96 and 2010. 1975, you can see that there is a bend here, bend one, two, and there are three and more or less four. In 1996, you don't have these bends. So basically this system, at least in this area, is developing more erosional, let's say, waves traveling upstream and the depositional wave traveling downstream. Again, erosional waves will do is, what will do is produce a straightening of the ridge. And you can see here, the ridge, at least in this area, is more like a straight channel. However, somewhere in 1998, you have here the kind of a complex bend, let's say, three consecutive bends that was cut here. You can see in 2010, this bend does not exist anymore. So you have a huge cutoff here. So again, if you have a cutoff upstream, you will have an erosional process straightening of the channel. And you can see here, you have a meander channel here. And in 2010, due to this cutoff of this bend, that is downstream, you don't have, I mean, you are reducing the sinuosity or the amplitude of this bend. So now, at least in this area, the process is more erosional. But I show you from 1975 to at least 2000 to 1996, this whole area was going under erosional, more cutoffs. And look here now in 2010, the river is trying to go back to the equilibrium condition by migrating or increasing the rate of migration. So basically, the amplitude of these bends here are increasing. So by this combination of cutoffs and migration, the river will try to reach again the equilibrium condition. And if we perform, let's say, a power spectrum analysis based on the channel centerline, you can see that for the Ucayali River in this area, it's around 25 kilometers, the wavelength, the dominant wavelength, given the sinuosity of the channel 1.7. So what we did this year is we went to this area that is near Pucalpa. It's a big city there in the Peruvian Amazon area. And we have done single beam measurements. So basically, we get the bathymetry in this area right here. It's around 25 to 30 kilometers long. The channel width is somewhere 800 meters. And you can see here the solutions, basically the measurements. You have the positional region at the inner bank, erosional region in the outer bank here, the blue color. So basically here, the coloring are showing pink are, let's say, the positional areas and blue are erosional areas. So this is a typical configuration for a meander. You can see here also downstream the position and erosion in the outer bank. So the key question here is, these large systems, as you can see, they have the old traces. I mean, these were abandoned channels. This whole system has small channels that are connecting these large channels. So now the city that is located there, it has more population. And the linkage between the city, the population, people and the river, it's stronger now. Why? Because they all depend on the commerce based on the river, if they can navigate. And also they depend on all the water intake and outflows into the river. So they are, let's say, concerned about if the river will migrate and get away from the city of Pukalpa. So that's what we were trying to do here, understand the dynamics of the Yucayeli River near the city. First we did some matematric measurements. And then we have done this ADCP measurements. We have done every, we have several cross sections in this area. And this is just one of them near the city here. You can see outside of this bend. And you can see that the velocities are higher in the outer bank. These are depth average velocities. And this is the cross sectional section here, the cross section here showing again also in the water column where you can expect higher velocities. These are the streamwise velocity contours. So again you have a scour region near the outer bank, so typical behavior of meandering channel here. So what we are trying to do is, we have done, at least I have been working for a few years in meandering channels, mostly for the laboratory experiments and modeling for lab scale. And this is a challenge now that we are trying to model a large river, this is Yucayeli again, 25, 35 kilometers long. And right now we are concentrating on validating our modeling. So basically we are using a 2D modeling for now. We also plan to use later on a 3D model, but for now 2D modeling in order to validate our measurements and our modeling with our measurements. So the idea is then what we can do is develop the morphodynamic model and try to predict the evolution of this meandering configuration. Again, let's go back to the conference of the Amazon here, the Maranjon and the Yucayeli river. So we have seen a little bit about the Yucayeli river. Now let's look a little bit about the Amazon river. Again, the Amazon river is born right here. Again, the Amazon river is not a meandering channel. It's an abranching structure, main channel and secondary channels. If we see the, let's say this is the starting, this is Nauta where the Amazon is born, this is Iquitos and this is going through Brazil. So basically this is the boundary between Peru and Brazil. So if we identify the abranching structures again, what I'm calling an abranching structure is a main channel with secondary channels. If we identify along this flat plane, we can see here in these yellow circles where you are observing an abranching structures. The interesting thing is that for example here, let's pick this one. You have a high sinuosity channel, let's say local sinuosity, but still you have secondary channels. Let's pick this one here. You have very low sinuosity of the main channel and still you do have secondary channels. And here for example, this one is a, let's say, medium sinuosity and you still have secondary channels. The question is, how many secondary channels can we expect either on low, let's say, medium or high sinuosity of the main channel? How are they produced? This is a erosional system. Basically there is, this is not like the Parana river where the Parana is still in the positional system. So here the dynamics of these secondary channels might be important for the dynamics of the whole an abranching system. So that's what we are trying to do, trying to understand, let's say, what kind of, let's say, what is the importance of these secondary channels on the dynamics of each an abranching structure and why, how is that important in order to understand the whole Amazon system? So let's focus on only one here. This is the Muiui area. This is near Iquitos here. You can see there Iquitos city. So this is clear. You have a main channel right there and secondary channel and a small, let's say, secondary channel. But let's consider only two secondary channels here for simplifying things. From 1987 to 2009, if you compare the images, maybe things might not change too much, right? You can see that the whole an abranching structure is more stable. However, if you see here, and I don't have the images here, for example, when you have this high-synosity main channel, the dynamics are quite similar to a meandering channel. So basically it's the rate of migration for this an abranching structure is higher than the rate of migration for this other one. So let's go back here to Muiui. However, if we see in detail here, okay, the main channel does not migrate too much because it might be stabilized by these secondary channels. However, the secondary channels you can see here, this is an IT7. This secondary channel is starting to develop like its amplitude. It's starting to migrate somewhere here. And in 2005, it already merged with this other secondary channel. So basically, the message here is that we can treat this an abranching structure as the composition of a main channel and secondary channels that are quasi-freely meandering channels. Basically, non-developed meandering channels. And what do I mean by that? You have a boundary condition here upstream, at least for this secondary channel. You have boundary conditions given by the an abranching structure. And also, you do have a downstream boundary condition. It's not like in the case of, let's say, a freely meandering channel when let's go back here. In a case of a freely meandering channel that the upstream boundary condition is farther upstream. And let's say, in this case, there is no, let's say, a force in downstream boundary condition that is modifying the structure. So you allow the river to migrate along the flat plain, develop this dynamic equilibrium, get the dominant wavelengths, get the dominant amplitudes, get the dominant migration rates, all these parameters, statistical parameters. However, for the case of the, for the case of the an abranching structure here, you don't have the opportunity, you don't let these meandering channels to develop. So they are non-developed and they are not freely meandering because they interact with other channels and they also have the placement of the all, let's say, the, you can see here all traces. So it's not the same as those freely meandering channels. So what we have seen here is, again, meandering channel, purely meandering channels migrate much faster than an abranching channels. And you can see here, this is a meandering channel that the migration rate is much higher than, let's say, it's much higher than the whole structure. So, but still we need to understand under what conditions here this is more stable. And this is important at least for management because this is the Amazon River. Brazil is somewhere here, is trying to get to Asia through the Amazon River, going to, to, trying to, let's say, go to Peru and get to the Pacific Ocean and then go to Asia there, especially, specifically to China. So this is going to become very important. And a lot of, let's say, hydraulic structures or other kind of civil structures will be built somewhere in this region. So people might think that this is a very stable an abranching structure. But the idea is to understand the dynamics of each of these an abranching structures because the question is why do we have this 30 kilometer wavelength of presence of these an abranching structures. That's something that we don't know. And at least my, my group is trying to understand that. So again, we have done two field campaigns in this area, 2010 and 2011. But in 2010, we have used multi-being and single-being here, but this is the single-being data for this, for different an abranching structures. Here, I'm only showing you for the MUI area. You can see this is water depth. So this is not the bed elevation. So higher water depth, more scour region, lower water depth is the positional region. So you can see here for the main channel, typical erosional region here, in, let's consider this is a vent, outside of the vent, let's consider this is a higher erosional region, outside of the vent, higher erosional region. It behaves like a big, big meander. So if we change the contours here at least the scale and just look at the individual secondary channels like this one here, it's interesting because you start to see a typical meander. You see here, in this case, water depth again, erosional region in the outer bank, erosional region in this outer bank changes. So you start to see, let's say, development of this vent. What is interesting here is look at the colors. Here, the erosional region is more or less pink, but here it's more white. So basically, what it means is that this meander is trying to develop higher and higher erosional rates or erosional, let's say, scour power, to scour the vent farther downstream, but it cannot do it as a freely meander because you have the influence of the downstream boundary condition. So let's look at another one here. Again, typical meander configuration. Upstream in this area, in the outer bank, red colors, erosional region, here pink colors, erosional region. And this is showing what I have said, downstream, the farther downstream that you are in these non-developed channels, the higher the erosional rate. So what is telling us is that, let's say, we need to understand the dynamics of this in order to understand the dynamics of the whole anabranches structure. And you can see here clearly the inflection point, basically from one bank, from the outer bank of this vent to the outer bank of the other vent. Again, these are single-bin data. We were playing with a multi-bin data in order to get the bed morphology, basically, bed forms in this anabranches structure because that's another question. How about the bed forms in these non-developed secondary channels? How are they compared to the main channel? For example, if we have dunes, how are they compared? So that's something we are doing. Also, we have done a lot of ADCP measurements in different places, but I'm showing you right here just to give the message. Typical meandering configuration. Higher velocities are, again, depth average velocities. Higher velocity to the outer bank changes to the other outer bank. So, these are not the typical, let's say, I always play with more freely meandering channels. These are not the typical ones, and we need to understand these non-developed scenarios. Let's talk a little bit about what kind of morphoenamic, bed morphoenamic features that we encountered. So this goes back to my research when I was at the University of Illinois, where we have developed this Kinochita channel, and I have measured some dunes migrating along this high-synosity channel. You can see here at different hours, you can see this is an animation at different hours, so where dunes were migrating along this high amplitude meander. So you can expect that dunes are migrating, maybe modifying the flow structure. The question is, how much do they modify? How important are those bed forms in order to consider for long-term meandering migration? So this is a time equal one hour. You can see one bed form here. This is a dune, right? What I have done is we have measurements at each hour, and we have computed a flow structure using a three-dimensional full-in-averse equation model. So what I'm showing you here is cross-section 15, cross-section 16, right here. 15 is the apex of the bend here. You, more or less, these are the velocities what you're observing here is just the cross-sectional velocities. So more or less, you recover what is the secondary flow, right? We have learned from fluvial mechanics like 101 that near the surface flow goes to the outer bend, and near the bed flow goes to the inner bend. That's why you have a depositional region in the inner bank. And you, more or less, recover that here, this secondary flow or this helical motion. However, if you are in cross-section 16, right there, you have a presence of a dune. You can see there. And these are not two-dimensional dunes. These are three-dimensional dunes. And if you plot the vectors, at least from the simulations, we, we couldn't measure completely when we have a progressing way because dunes. Because, you know, you have to have a statistic at least three minutes or five minutes of measurement. So under bed forms, we're migrating. So, but using this three-dimensional model, we describe here that there is no secondary flow. So it's almost the opposite. Look at this. Near the surface going inward. So what does it mean, this, is that the presence of this three-dimensional dune is modifying the secondary flow, the natural secondary flow. And why is this important? If we go to the field, let's say, to the Cayalli River, if we have dunes that are dependent on the dimension, of course, whatever you measure, you might be able, if you measure, let's say, in this condition, you will say, oh, we didn't observe the natural secondary flow. But that may be wrong because it is, the secondary flow is there, but it's being modified by the boundary layer above this dune. So that's what we need to consider, that, that, you know, what kind of bed forms do we have in that, in the channel is what we need to consider. And if we compare, how many minutes do I have? Two? Okay. So if we compare, let's say, this is an instantaneous value with, like, say, a time average bed morphology. This is just the bank, shear stresses on the bank, the outer bank. You can see here, if you have a dune, you have high shear stresses compared to a, let's say, a steady condition. So basically, most of the morphodynamic models that, at least what I have worked, we consider only the steady condition and not the instantaneous values that, for shear stresses, and depending, of course, on the bed morphology, you may have high shear stresses, higher fluid erosion. And that's what it's something we are doing. One of my students is trying to characterize these bed forms, separating dunes and ripples for large river systems. And the, my other student is trying to do is trying to understand the flow structure. At least this is for a laboratory scale. You can see this is in millimeters. But at least what we want to do is understand this bed form interaction with the flow structure and see if that's important for long-term migration or not. So with that, I want to finish just saying this. We have a GIS-based long-term migration model that we initially worked, you know, back in the day, 2006. It was a part of my master thesis. Now, we have improved a lot with the help of Eddie Langendon from the USDA. And we have some capabilities for bank erosion, physically-based bank erosion model. So this is the website. And we will be contacting the, in order to put this into the CSDM website. So again, just want to finish. I have to acknowledge without this support from these agencies and from specifically the Peruvian Navy that they are, we are, we have an agreement with them. Basically these field campaigns will be impossible. And well, my students and the support from my university. So thank you so much. Thank you. Questions? No question? Yeah. Yes, just one small question on the 3D flow field you showed on the, on the dune in the riverband. I was very surprised by, by the stronger velocity upwards near the bed. Do you have any explanation on that? Yes. So basically that cross-section, that cross-section is located at almost the tip of the bed phone. When you have, you know, flow more like a developing sheer layer going to the free surface. So that's the location. That's, that's, that's, I don't know if I can go here. Here. You can see here. So this dune, we have the developer boundary layer right there. And because the flow is accelerating, that's why you have these vectors going up. So if we go, let's say for just a simple, simple configuration, let's say you have, if you don't consider two bed phones, only one, I mean, always you will have a sheer layer right here, correct? So that's, that's what we are showing in the other simulations. I don't know if that answers question. We can look into the boundary conditions. You know, you assume the vertical velocity to be zero. No, in, in, in, these are large, yeah, these are large simulation, but these one were more runs. So we were putting like a logarithmic, let's say, wall functions into this, in the bed, at least in, in this simulation. These were done because of the domain. We were modeling the whole three consecutive bands in this case. Okay. I'm curious to know that when you are doing modeling, you must be measuring the flow, river flow and the sediment discharge, right? Yes. How many, how many places, how many locations? Oh, we have at least for the Amazon region, we have, I mean, we are covering the entire Amazon, the Peruvian Amazon, up to our boundary with Brazil. And the Navy is trying to get an agreement with the Brazilian Navy in order to get more, let's say, go, go into the Brazilian side also. And so here is what I want to acknowledge is that this is not only Peruvian Navy and University of Peace, but we have all institutions. That's why we created this center that is CREAR, it's a Spanish word that means create. So basically we will be doing, you know, long-term monitoring of this region. And if we are talking about also sharing data, that's also that we are trying to do. Maybe in the future we can put all our measurements for other people that want to do more analysis on that.