 in nesting, particularly, that he's going to talk about some work he's been doing that will cover both fluid dynamics and humidities. Thank you, Jai. So in preparing for this talk, one of the things that I wanted to do is to communicate what's been fundamental to my research program. If you look at the bottom of this introductory slide, you'll see the northern Gulf of Mexico spanning Louisiana, Mississippi, Alabama, and the Florida Panhandle. And we have worked for the past couple of decades to try to develop techniques that will allow us to describe the coastal region and the flood plain so that we're characterizing with a high level of accuracy, levees, roadways, railroads, any natural impediments to flow and any natural and engineered conduits to flow. And I also thought about my introduction in terms of Jai's retirement. And I note that 40 years ago, he would have found himself at the University of British Columbia, right? And at that time, in 1977, the powers of 10 came up. I don't know if you remember this, but what the powers of 10 does for us is it allows us to appreciate scales. And I hope that it can allow you to appreciate the complexity of adequately describing coastal regions. Now, I'm sure back in 1977, Jai had a few picnics maybe or two, but this is not him. But we go from 10 meter square to 100 meter square. And mind you, this is in 1977. And we can begin to see some of a coastline that we may want to characterize. You may or may not know where this is. We have to zoom out a little bit further. We're at a kilometer. We're now at 10 kilometers. And you may have guessed it by now when we get out to 100 kilometers that we're looking at Lake Michigan in the Chicago region. We're at 1,000 kilometers. And at 10 million meters, we encompass the entire globe. So one of our goals with my research group anyway is we want to be able to describe some very fine features relative to those kind of scales. And we want to have flexibility in our description. So I have some Center for Coastal Resiliency collaborators on the research that I'm going to be presenting today. And most of these folks have been working with me over the last seven years on an ecological effects of sea level rise study in the northern Gulf of Mexico. And we are recently transitioning that into a new study. And I'll provide some description of that later on. But again, the point that I want to make here is that we want to transition from a large scale approach. We want to think about what our resolution is going to allow us to be able to describe and what it is that we're going to have to parameterize or what's at the sub grid scale that we're not going to be able to describe. So if we really want to get down to that level where we're describing this lawn area along Lake Michigan, we have to have a pretty fine resolution. When we began our ecological effects of sea level rise project in the ecological effects of sea level rise in the northern Gulf of Mexico project back in 2010. And in fact, when we were writing a proposal or sketching it out in 2008, we were thinking about what are some of the processes that we want to describe and what is it that we want to accomplish with this project. And at that time, there was very little work that was ongoing that was not in the bathtub modeling. That is, the vast majority of the work was taken in an existing digital elevation model, raising the sea, determining how much of that existing digital elevation model was inundated, and then inferring what the coastal response would be. And this was true with hydrodynamics, this was true with astronomical tide flow, this was true with hurricane storm surge flow. And with this and the following cartoon, what I hope you'll agree with me on is that the coast is rather dynamic. And if we raise the sea level, we're going to have changes at the coast. Now we may be able to do beach nourishment and we may be able to protect the coast, but that's going to get more and more expensive. And if we allow the sea level to rise or if we experience sea level rise and we don't do anything at the shore, then the shoreline is going to be eroded away and the infrastructure is going to become more vulnerable. So these features are these characteristics at the coast, whether they're with a marsh or whether it's with infrastructure along the coast. We wanted to build into our modeling system. Further drive that point home, take a look at the green line, representing the 1848 shoreline of Grand Bay. At that time, there was the Grand Boutour Island. And you'll notice in my talk that I have a little bit of a focus to Grand Bay. I can't present everything that we did in the last seven years, but I can tell a little bit of a story. And in 1848, the Grand Boutour Island still existed. But by present day, it's been completely eroded away. And in fact, by 1960, it had been mostly eroded away. And it's one of the reasons that the Grand Bay Marsh in Mississippi, it has one of the fastest rates of marsh edge erosion of any marsh along the continental United States, because its protection is gone. Because the coast is dynamic and if we want to model historically or into the future, we need to consider that dynamic system. So our process has been one that encompasses not just the modeling, not just field experiments, but also from the very beginning, within the first three months of our project, we engaged stakeholders, coastal resource managers along the northern Gulf of Mexico. And the stakeholders have added a lot to our project, and they've helped to constrain our project to some extent. They've helped us to define global climate change scenarios that we ought to be considering, and those that they considered to be unreasonable for their purposes. Why include a 40 meter rise in sea level, that's going to occur 500 years in the future, if all you're going to do is turn off your stakeholders because of doom and gloom. Let's try to focus a little bit more on the more near term, maybe going out to about the year 2100, and providing them with information that they can consume and that they can build into their adaptive management strategies. We benefited from an enormous wealth of data, as we all do, collected Earth data that's been collected, and it is provided by NOAA and USGS to name just a couple of the federal agencies. We have an integrated suite of models that produce dynamic results that allow us to make these coastal dynamic assessments. And our ultimate hope is that with the involvement of our coastal resource managers, ultimately we're going to have some benefit to the coastal ecosystem and society itself. Our project really is a transdisciplinary research project. It has presented transdisciplinary research outcomes, and a colleague of mine, Denise Goodorm, has written a wonderful paper that is included in a recent special issue we did in Earth's future, and it's all about how we incorporated this into the ecological effects of sea level rise in the northern Gulf of Mexico project. It's not an easy thing to do. This project started with 10 co-PIs. One of them had to be let go along the way. Managing 10 faculty members is always a challenge. Not all of them are really interested in doing interdisciplinary research, but if you stick with it and you work with your strengths, particularly the graduate students, you find that such research can be accomplished and can be very fruitful. So before I get into any of the modeling results, I want to make a point that I hope all of you are able to take with you and to consider and to ponder. Now, we all know how valuable LiDAR has been, especially those of us that do coastal modeling, that do tide and hurricane storm surge modeling. The improvement over the national elevation data sets that we used to have is so dramatic. Going from uncertainties of three to five feet along the coast in the upland areas where upper reaches of the flood plains to an uncertainty that's on the order of a foot is a tremendous improvement. But what happens, I think, is that it's such a tremendous improvement that we get blinded by that improvement and we think that the data is perfect. And like any data, it's not. And so we're trying to develop on the one end a very localized marsh response modeling capability and understand that if we're modeling a marsh and we're doing this over a decade old period, we still only have accretion rates that are on the order of millimeters per year. So if you don't start the marsh surface as closely as possible as to where it's at, you're not going to integrate it forward in time appropriately. And what we demonstrated at the East Point Marsh and at Grand Bay Marsh and at transects all along the coast and what we're beginning to document in Louisiana is that there are unacceptable levels of error in the LIDAR within the marsh systems and we have to devise procedures that we can use to correct that data. Again, that's important if you focus on the lower left graphic, the marsh platform out of which in this case, Spartina Alterna Flora grows is going that that Spartina Alterna Flora can only grow, can only function over a period of time if it's between mean low water and mean high water. If it's below mean low water, then that marsh is going to convert to open water. If it's above mean high water, then that marsh is going to convert to up. It makes perfect sense, right? So what I would suggest to you as just a simple check is if you can run a transect through a marsh that you're interested in studying or if you simply run a transect through the LIDAR that you're going to apply and look at it. Is that LIDAR transect between mean low water and mean high water? If it's not, that's your first indication that the LIDAR DEM has some level of inaccuracy. That's unacceptable for the work that you're doing. In the project we developed hydromem or hydrodynamic marsh equilibrium model. This is a coupling of, in our case, it's an ADCIRC model. It doesn't have to be an ADCIRC model, but that's just what I work with. Coupling of the ADCIRC model with Jim Morris's marsh equilibrium model. With this coupling of the ADCIRC model and the marsh equilibrium model, we're able to estimate biomass productivity and accretion rates for a marsh system. Here's an example in Grand Bay, Mississippi. And so if you look in the upper right of the animated graphic, you'll see the year and the sea level rise for that particular year. We're looking at a two meter sea level rise by the year 2100. Blue of course is water, fluorescent yellow is medium, red is low, and green is high biomass productivity. So as we iterate through time, we're seeing the marsh migrate up one and we're seeing more open water areas created. Now a fundamental part of our project is to put on top of all of this across the three state region, across Mississippi, Alabama, and the Florida Panhandle, Hurricane Storm surge simulation. So we want to simulate what is the surge going to be like in the future? What's it going to be like in 2050 or like in 2100? And if we don't take into account the impact to the marsh system and these massive areas of open water that are created, we're not going to represent the flood plain because we're not going to drive the surge through there like it will now be able to. We're applying hydro mem at numerous places along the Gulf Coast and along the East Coast. One of the more interesting projects that we have right now, we've developed a model that, the hydro mem model that's spanning Chesapeake Bay inlet up to Ocean City, Maryland. So again, the backbone of the work that we do, as Jai said, is the unstructured mesh generation. So for this three state region, Mississippi, Alabama, and the Florida Panhandle, we needed to be able to describe the flood plain up to the 15 meter elevation contour because we wanted to be able to describe Hurricane Storm surge under a sea level rise of two meters. And so we needed to increase the flood plain beyond what it would be today. And we still wanted to have this capacity to describe, as in Grand Bay, the tidal creeks. And so we have this ability to employ the unstructured mesh and have higher resolution out in the Gulf of Mexico and a very finer fine resolution in Grand Bay. The model that I work with, ADSIR, is a finite element model of the shallow water equations. And as such, it is a current condition based model. So the smallest element size over the deepest of the symmetry is going to drive the time step. And when we have a high resolution like this, our time step is on the order of one second. And we'd like to be able to simulate four to six days of Hurricane Storm surge. We'd like to be able to simulate at least 45 days of astronomical tides. We do this with high performance computing capabilities. We provide, or excuse me, we do domain decomposition on a model like this or on a mesh like this and we'll decompose it onto 400 to 600 cores just depending on what we're running. The ADSIRC model frame, ADSIRC plus one model framework is relatively simple. I've talked a little bit about the lower left, the ADSIRC mesh. I've talked about how the coastline is going to change with respect to the marsh. We also need to take into account that if I have a two meter rise, I'm going to have a different barrier island configuration or at least dunes on the barrier island than if I have a 30 centimeter sea level rise by the year 2100. Society will be able to do more beach nourishment if we have less sea level rise. But you know, as an aside, what I like to tell people is think about how expensive that's going to be. Think about the heyday of the United States, post-World War II, all of the infrastructure that we build along the coast, all of the beach nourishment that we've done up until the present day and how much we've invested along the coast when we underwent just a five inch rise in sea level or just a 10 centimeter, excuse me, a 12 centimeter rise in sea level. We've literally best invested hundreds of billions of dollars along the coast just in the northern Gulf of Mexico. What is that going to cost if we have sea level rises that are multiples of that, that are order of magnitudes greater than that? But I digress. And the bottom line here is we need to be able to describe future surface roughness. And with surface roughness, I'm meaning not just Manning's end to describe the roughness of the surface to impede the flow of water, but also the roughness of the surface to impede the transfer of momentum from the wind onto the water surface as it's blowing across the flood plain. We need to have astronomical tides filled into our model. And of course, we need to have a wind field. And the wind field is, in my opinion, just as key as the actual mesh itself. So we employ this in our modeling system. And we have a different configuration for the marshes, for the land use land cover, for the shoreline and the dune morphology, depending on whether I'm dealing with a high sea level rise, present day, low, intermediate low, intermediate high, and again high. And when you see the changes in the colors, when you see a change to a light blue, you're seeing more open water opened up along the coast and more vulnerability to hurricane storm surge. So what I like to think that we have accomplished in contributing to a paradigm shift away from bathtub modeling to coastal dynamics of sea level rise is not just coastal dynamics of sea level rise, but recognizing what is it that drives sea level rise. That's the temperature variation that we have, depending on whether we have a high carbon emission scenario or low carbon emission scenario. So if we have a high carbon emission scenario, we're going to have a higher average temperature, we're going to generate more sea level rise. Now, what our system doesn't take into account, but can, what we haven't modeled as of yet, is what's going to happen if the Antarctic ice cap starts to break up, like we've seen some recent articles in the New York Times about. What's going to happen if we have huge amounts of ice breaking off of its shelf and into the ocean, and we have less than a smooth rise in the sea level, and we are likely to have more of a stair stepping. And I submit that we need to be able to take those kinds of projections into account as well. For the hurricane storm surge for this particular project, we wanted to generate future flood plains, future 100 and 500 year flood plains. And since we worked with FEMA and developed the coastal inundation model for Alabama and the Florida Panhandle, and we had access to the same kind of data from Mississippi, we decided to take a similar approach to our future flood plain assessments. So we down selected from the 295 synthetic storms that were used for the Alabama Florida Panhandle study by looking at the return periods and the surge heights by going back to the joint probability method with optimal sampling, and down selecting only those storms that generated 100 and 500 year flood plains. The result is that we're able to estimate return period still water extents that are a function of the coastal dynamics of sea level rise. Our new project, which we just got underway at the end of last year, we've coined the acronym NGOM plus N2E2, N2E2 because we're incorporating natural and nature-based features along with an economic impact analysis and an ecosystem services valuation. So just as an example, looking in the lower left at the Appalachicola region, we can assess the marsh biomass productivity. We can assess the hurricane storm surge impact and various quantities associated with that. And we can now do this for proposed nature-based features. So maybe it is proposed to do thin layer disposal when dredging is done in this region, do thin layer disposal over the marsh. Maybe it's proposed that we do more nourishment of the dunes and this approach will allow us to be able to assess trade-offs between ecosystem services valuations provided by the marsh and the economic impact that occurs whether the marsh is there or whether the marsh is not there. You'll look or you'll notice again we're focused on these hucks, these hydrologic unit codes. We're doing this over all of these hydrologic unit codes so that someone that's well removed from the coast can begin to appreciate how their tax dollar that's going to support thin layer disposal over the marsh or do nourishment is actually beneficial to them in that they're likely to have less hurricane storm surge with an extreme event. And then we'll also be doing this with nuisance flooding as well. My concluding remarks are relatively simple. One that we all ought to know and that is with respect to the climate we no longer have the luxury of stationarity. I hope in this brief talk I've shown that we can model this dynamic system and we have the basis of a system of systems if I haven't shown that already and we can do this at amazing scales. If you would have asked me back in 1980 if we were going to be able to build a model like this it was inconceivable. Climate change is a generational problem. We can address it but we can't will it away. It's here and we have to work with it. But I think the most important point that I'd like to make is our numerical modeling technology is awesome. It really is. But with respect to climate change the models only serve as advanced diagnostic tools. We're not making predictions about the future. We're providing information for the policymakers to make informed decisions. Few directly related publications that came out of that project and some acknowledgments. Thank you for your time and Jay, thank you for this opportunity to present. So Robert Nichols, I'm just curious about stakeholders because what kind of time scales were they really interested in? You mentioned 2100. In my experience often it's maybe a bit shorter than that and also you mentioned two meters of change. I sometimes find resistance to these large scenarios. I'm interested in sort of take on what they were sort of most interested in. So it's a very insightful question and it really gets to the point and I have to say that my answer disappoints me because the majority of the coastal resource managers that we worked with really what they're interested in is affecting change within their career. And their career span is 10 or 15 years. They've already been at it for 15 or 20 years. They've got 10 or 15 more years left and they want to do something that they can see the results from. And the problems that we face, as I said, are generational. We have to go back to what made us a great nation. I'm sorry I'm not running for office, but it's the truth. My great-great-grandfather didn't come to the United States and cut trees down and pulled the stumps out so that he could be wealthy and have a great life. He did that so that I would have an opportunity to get higher education. I really believe that. He had a generational vision and people back then did and so Robert it's it is the case that they are looking at 10 or 15, but we were able to encourage them to consider 50 and 100 years into the future. And as far as two meters, well forgive me I show the most dramatic, but we looked at low, intermediate, low, intermediate, high and high and the low was a crazy low of 20 centimeters. Just the linear interpolation of the last 150 years onto 2100 the intermediate low is just a modest increase in sea level expansion or an ocean expansion. The intermediate high actually starts to take into account some glacial melt, but very little at that and it is a rise of 1.2 meters and the two meter is just a little bit more of glacial melt. Well a lot more than we have at present because at present the sea level rise that we've seen over the last 150 years is 90, 95 percent of that is just thermal expansion of the seas and if we have two meter rise by the year 2100 then about 80 percent of the contribution is going to come from glacial melt. So I personally think it's still conservative, but the manager of the Apalachicola National Estuarine Research Reserve really doesn't want to show the results to her locals. Sorry if I took a long answer. I was just wondering about the the results you showed for the northern Gulf of Mexico modeling. Did I understand that you're modeling the evolution of the shoreline and the barriers in the dunes or was that? You did, but not me and with a project like this it is really wonderful that you not only have copi eyes but you'll have scientists who if you gain their respect they'll begin to contribute to the project and and two that are of great importance are Rob Dealer and Nathaniel Plant. Rob Dealer is at Woods Hole with USGS and Nathaniel Plant is at St. Pete with the USGS and they developed a Bayesian network approach and that provided reasonable based on historical data. They're Nathaniel and his scientists who used to be my PhD students are working with us on the new project and they're building more numerical modeling into the Bayesian network. Thank you. Okay so thank you Scott very much it was a great talk. So it's poster time and remember what my dad says, vote early, vote often, but remember the vote we're going to only see and have talked about 50% of the posters the other 50% comes tomorrow. So you're voting not