Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
15 - Waves, Sand, and Simulations: Modeling with Alex Boland & Ryan Schanta
8 Plays8 months ago
In this Student Research Spotlight, Marissa Torres and Deb Fernandes hear from Alexander Boland (UGA) and Ryan Schanta (UDel) as they share their paths into coastal engineering, their thesis work, and their next steps in PhD research. From modeling storm surge impacts on Georgia roads to improving sediment transport predictions with FunWave and XBeach, this episode explores the power—and challenges—of coastal numerical modeling.
Recommended
Transcript
Introduction to the Going Coastal Podcast
00:00:10
Speaker
Hello and welcome back to the Going Coastal podcast, the podcast of the students and new professionals in the coastal field, hosted by John Miller, Marissa Torres, and Deb Fernandez. I'm one of your co-hosts, Marissa.
00:00:22
Speaker
And I'm the other co-host, Deb Fernandez.
Spotlight on Alexander Boland and Ryan Shanta
00:00:25
Speaker
And for this month's episode, we have another round of our student research spotlight. With us, we have Alexander Boland, a graduate research assistant who recently graduated with his master's and freshly started his PhD at the University of Georgia with Dr. Matt Bilsky.
00:00:41
Speaker
And we have Ryan Shanta, finishing his last semester at the University of Delaware and starting his PhD soon at the University of Washington. So excited to have you both. Welcome, Ryan and Alex.
00:00:53
Speaker
Yeah, happy to be here. yeah Thank you for having me. Yay, we love student research spotlights. Okay, so you both you both recently are
Ryan's Academic Journey and Experiences
00:01:03
Speaker
finishing your master. So Alex just finished.
00:01:05
Speaker
Ryan finishes. ah Well, it defends his thesis on July 2nd, which is the day after this drops. And we'll have an official August summer graduation date. Is that true? Yes.
00:01:21
Speaker
Well, and then you both are going off into PhD land. So that's
Alex's Path to Coastal Engineering
00:01:25
Speaker
exciting. I would like to hear about how you got to this point in the first place. So what was your kind of... personal professional journey to get to and ah into your master's and why you chose how you kind of ended up here in a master's program in the coastal field.
00:01:45
Speaker
So Alex, how about you go first? Yeah, sure. Yeah. So I got my undergraduate from the University of Virginia. in civil and environmental engineering.
00:01:56
Speaker
And after that, I did a brief stint at the University of Miami, i getting my master's there. Although I found, once I found an opportunity with my current PI, Dr. Matthew Bilsky, I was more interested in what he had to offer and in the ah research opportunities there.
00:02:13
Speaker
I've always loved you know coastlines. I love going to the beach. you know And so I was super interested in his research. And I reached out to him ah seeing if he had a position available in his lab, transferred out from the University of Miami to Georgia, ah to the University of Georgia.
00:02:31
Speaker
And then here I am. Well, so where, where are you from? ah How'd you end up choosing University of Virginia? And what made you kind of jump into the coastal field?
Motivations and Influences in Coastal Engineering
00:02:45
Speaker
Yeah, sure. So I'm from Virginia. I grew up just outside of Richmond. And when you're from Virginia, there's kind of a couple of main schools that everyone applies to. And Virginia sort of the the default.
00:02:57
Speaker
And I went there and actually, at while I was at the University of Virginia, they used to have a coastal engineering course. And like on my fourth year, I went on like a whole campaign trying to get that course brought back.
00:03:10
Speaker
ah But the professor had retired and ah no one was really super interested in picking that back up. And yeah, I just think, you know, the coast is just such a ah fascinating, there's so much going on, but it feels like not a whole lot is going on and and and sort of, you know, learning more about that, especially, you know, given certain climate crises that may or may not be going on, I thought it would be worth pursuing, you know, an interest in that field, delving deeper into that when I,
00:03:42
Speaker
found Dr. Matthew Bilsky's options, who he's, you know, sort of a, he's a very intelligent man in this field and sort of one of the one of the leading voices leading the charge forward. I just thought it was ah too good of an opportunity to to pass up.
00:03:59
Speaker
We'll talk more about that once we get into our research section. But Ryan, how'd you end up here? Yeah, sure. So I did my undergraduate in civil engineering at Cornell University.
00:04:12
Speaker
um And i was in I always knew I wanted to do something with water. ah initially thought that meant ah water quality, but it did not because I just really did not like chemistry. ah So I quickly got out of environmental engineering and went more towards civil engineering, hydraulics and hydrology. And I started really liking that a lot.
00:04:30
Speaker
um Some of my favorite classes were more on the modeling side. I didn't really consider myself a coder or somebody that did a lot with computers before that. Uh, but seeing how you can apply it to model water and how it moves in the environment was very interesting to me.
00:04:44
Speaker
um and then the course that really sold it for me was a numerical methods course, uh, taught by the environmental fluid mechanics group, um, there. And that really set me up for success and giving me the skills that I needed to, um,
00:04:58
Speaker
be able to code interesting math. And then from there, it was just a matter of where can I actually use these skills? And that ended up being within the context of civil engineering that ended up being coastal engineering. um So the more I learned about it, the more I liked it.
00:05:11
Speaker
I got a little taste of the professional world of that in some internships. um And I continued to like that a lot. And then as I was moving forward, i was looking at what are the main institutions for coastal engineering.
00:05:23
Speaker
And the standard ones that I was seeing were Delaware, Oregon, Florida, and Delaware ended up just making the most sense for my career goals at the time. um So currently, I'm working with the SHU Lab in the Center for Applied Coastal Research, which has just been a really great place to learn coastal engineering because there's such such a legacy here of people to learn from and also just talk with.
00:05:50
Speaker
There's, at any given time, there's about 10 to 20 different coastal people here. so it's really great to have that network and It's been a pretty good foundation, i think. That's pretty cool.
00:06:00
Speaker
Where are
Complexities and Passion for Coastal Modeling
00:06:01
Speaker
you from again? I don't if I picked it up. Yeah, so I'm originally from Allentown, Pennsylvania. So Eastern Pennsylvania, about an hour north of Philly. So I think you're probably the first one to tell us that you fell in love with in coastal engineering because of the interesting maths there behind.
00:06:17
Speaker
correct? am i correct Yeah, I mean, before I started this degree, I wasn't even that particularly interested in the beach, if I'm being completely honest. I've come to appreciate it a lot through what I study, ah but it wasn't like i wasn't a beach person growing up or anything.
00:06:32
Speaker
Yeah, that's okay. We need people that love everything. So if math behind beach stuff are interesting, why not? I love that. I do feel like towards the modeling profession, though, like towards the coastal and ocean modeling, you do just get a lot of geophysicists or mathematicians or computer scientists that just like applying math.
00:06:56
Speaker
to this complex problem because it is complex and there are still a lot of unknowns and we still don't even have the technology to simulate all of it in a reasonable amount of time don't worry that's why we have this young generation here
00:07:13
Speaker
To develop our next set of coastal and oceanic numerical models. Isn't that right? Yes. i was hoping for more support from Ryan and Alex, but... They're not yet convinced.
00:07:25
Speaker
I mean, what can we do to convince you? i mean, this this episode is is a little special because of all of our student research spotlights... We have a variety of, I mean, the coastal field is pretty diverse in ah backgrounds and interest from the sciences to geology, the biology, you know, oceanography, chemical, all of all the components that make up the ocean and how it interacts with our shoreline more or less. But this one's special because both Ryan and Alex are modelers near and dear to my heart and
00:08:00
Speaker
And that's what we're going to have fun talking about today and commiserating about today. yay
Internship Experiences and Career Decisions
00:08:06
Speaker
Yeah. I do want to close the loop on your experiences leading into master's and like during your master's. Ryan, you mentioned having some internships in the private industry. um Wanted to know like which companies did you work for and what was that experience like and how did that help guide you to get you to this point?
00:08:27
Speaker
Yeah, sure. So the first one was in Bergman Engineering, which is now Collier's Engineering. And that was a very interesting ah experience because it was in upstate New York and it was basically in tandem with the Erie Canal Corporation.
00:08:43
Speaker
ah So this firm was more or less using Army Corps standards, steel, concrete kind of design, waterway structures is what the division was called. um and more or less maintaining the Erie Canal and all of the different pieces of it, because there's a lot more pieces than you would think than just the canal itself. There's a lot of infrastructure and overhead and also how it connects with the different waterways of the region.
00:09:07
Speaker
um So that was kind of more in the hydrology and hydraulics side, but that was where I first got a taste for waterways. Coastal engineering as well, because it was my understanding that there's a greater recognition of the Great Lakes as a coastline for engineering purposes.
00:09:22
Speaker
um So they were doing some projects on Lake Ontario where I got to see some interesting modeling work. um So that was probably the most formative one because I actually got to see um a really cool piece of infrastructure with the Erie Canal and how seeing how the history of it and how people are using it now.
00:09:39
Speaker
um So that was one of my main motivations early on. And then ah summer later, I worked at Arup, which is in New York City. And they were recently, they're largely an international firm um based out of the UK is where a lot of their coastal projects are, but they were starting to bring some projects over to the US side.
00:09:59
Speaker
um So it was really cool to kind of see how they were trying to get into the coastal field since it was very new for the American branch. um So I got to talk with some of the kind of pioneers for their side of bringing in and winning coastal work and kind of seeing what the process is of how do you make proposals? How do you bid on it? And how do you actually like win work in coastal engineering, which is a very different side than other stuff that I had seen.
00:10:26
Speaker
Interesting. Alex, what about you? Have you done any internships um in undergrad or or during your grad programs that got
Academic vs. Corporate Paths in Coastal Engineering
00:10:35
Speaker
you here? Yeah. So if I'm honest, I'm i'm sort of a I had no internships in all of my undergrad.
00:10:43
Speaker
ah when when ah When my PI hired me, I was sort of pretty unemployable but for all intents and purposes. i' really i had sort of slacked off on my my um like internship applications in college and and you know I got into a master's program in Miami and and because I had known a lot of the engineering side of it and I had known you know basic modeling and and certainly coding has come up a lot in in in civil engineering courses. But it was at Miami where i sort of, because I wasn't there for engineering, was there for marine conservation and and
00:11:15
Speaker
One aspect that kept coming up is is natural infrastructure, ah especially you know in Miami. A lot of mangroves are there to sort of protect a lot of inland communities.
00:11:26
Speaker
that That is actually how I came up. I came across ah Dr. Biliski's opportunities because he, you know, coastal engineering is not or natural infrastructure is certainly not a focus of his, but he's he's not afraid to shy away from it. He understands its value. And like civil engineers will, especially on the environmental side, will sort of go into the mindset of like avoiding like gray infrastructure on a coastline, ah sort of minimizing that because it does, you know, it can hurt like local communities as far as like biodiversity goes.
00:11:59
Speaker
um And he sort of acknowledges like, no, there's values to both because sometimes, you know, a couple of plants is not going to protect the inland community. Sometimes you need a seawall.
00:12:12
Speaker
And I think I really value that sort of perspective on it where it's, you know, it'd be nice to be like, oh, let's just plant things and it'll all be okay. But you to to acknowledge the strengths and weaknesses of both sides of it is super important as well.
00:12:29
Speaker
course. Yeah. So what made you want to pursue the master's in the first place? ah Honestly, a lot of it was feeling rather unemployable without a any ah any internships. But...
00:12:43
Speaker
As I've sort of of gone on and explored academia a bit more, especially comparing it to some of my friends from undergrad who are in a little bit more of the corporate side of it. I really, you know, there's there's good there's benefits and downsides to both the downsides to some of the you know more traditional and civil engineering firms.
00:13:04
Speaker
do seem to be pretty, pretty bad and something I would like to avoid. And I've i've quite liked academia so far. You know, I think long term I would like to stick with academia.
00:13:16
Speaker
I try not to get too attached to that plan because ah that phrase ah God or man plans and God laughs. And, you know, I don't want to you know, have Have that fail, but certainly I've liked academia so far and it seems to benefit me thus far.
00:13:35
Speaker
Sure. Great. So just to clarify, both of y'all went straight from high school, undergrad, master's, and now we're going through PhD, no breaks.
00:13:47
Speaker
Yep. Yep. Yeah. that here and i'm um I'm feeling it. I'm really feeling it. Yeah. I mean, technically, you guys get like a winter and spring break and a summer break. I mean, i have to say the nine to five grind is a trap.
00:14:03
Speaker
And I would love to stay in academia or love to have stayed in academia if I had found a place like you guys seem to have. Kind of found your, don't want to say safe space, but like a place where you feel comfortable working with a good team and working on things that interest you.
00:14:22
Speaker
That's good that you guys found that something to align to. It's never too late, Marissa. I think it is by this point for me. Did you know it's been 10 years since I graduated undergrad this year? That's okay.
00:14:38
Speaker
I'm not far off that. ah Yeah. I mean, it's fine. Alex been talking a lot about Matt Bilsky's research, and I'm grateful that you found him. i actually, i did one of my undergrad internships with was in the Champs Lab at University of Central Florida, where Matt was doing his PhD with Scott Hagen, the late, great Scott Hagen.
Alex's Master's Research with Georgia Department of Transportation
00:15:02
Speaker
And that was my first taste into what is ADCIRC and how do you start modeling? So ADCIRC is the advanced circulation model for anybody who's not familiar. It's pretty popular, widespread across academia for storm surge and general circulation and tide modeling.
00:15:21
Speaker
It has an unstructured mesh versus a rectilinear grid or a just a square grid. We're doing great. I don't know how much detail I wanted to get into with the modeling. I mean, yeah.
00:15:33
Speaker
That was my first introduction to Matt, and I couldn't believe that I eventually ended up in this field running into him at like an estuarine and coastal modeling conference that was held at URI one year.
00:15:46
Speaker
And i was like, oh, hey you're in this field. Look, I ended up here, too. I'm doing storm surge modeling. for my master's at least. And it's just a really small world.
00:15:57
Speaker
And I would like to hear more about what it is that ah you worked on for your master's thesis ah with Matt during your time there.
00:16:08
Speaker
My research was funded by the Georgia Department of Transportation. And sort of the overarching goal of the research project was to identify the vulnerable roads from both sea level rise under ah tides under sea level rise scenarios and ah storm surge scenarios. And the storm surge scenarios, ah that was created using data from the you know ah United States Army Corps of Engineers South Atlantic Coastal Study, um which they have you know point shape files or
00:16:40
Speaker
They have tables that you can convert to point shaped files, recording all this data and they use ADSERC in their models, simulating ah hundreds of different storms to try and determine the annual exceedance probabilities. So like the 1% annual exceedance probability, which refers to there's a 1% probability that the water level will be met or exceeded in any given year.
00:17:08
Speaker
ah ah It's also referred to by the 100-year flood. And then using the mesh that they created for the South Atlantic Coastal Study, we input different sea level rise scenarios to determine what the, and and ran simulations to determine how the tides would flood in October, which is Georgia's most tidally active month, to determine the flood extent for all these different scenarios and then determines the vulnerability of all the roads given elevation data and this water surface elevation data.
00:17:39
Speaker
that's sort of the general broad strokes of it. What makes October the most tidally active month in Georgia? Frankly, um I'm not quite certain. That was something that I found on NOAA's website.
00:17:53
Speaker
on And so we we ran the simulations, we ran them over like a 45 day period. And the reason why we did the 45 days is because that that captures the whole ah lunar extent, just every the full tidal extent. And we took the maximum water surface elevation from that to sort of figure out what's the most risk that these roads can experience.
00:18:18
Speaker
What were your primary findings or conclusions ah from that study in evaluating the total water level and and identifying these vulnerable roads? How is that important to your average Georgia coastal driver or beachgoer?
00:18:36
Speaker
Right. So the the whole idea was to mainly just as a as a tool to identify vulnerable areas, it's sort of like an asset management for the Georgia Department of Transportation.
00:18:48
Speaker
ah So they can make plans around that, you know whether that be managed retreat, increasing the sub-base elevation of the road, or just straight, like just abandoning the roads altogether.
00:18:59
Speaker
That sort of gives them a few options. I like one of the outputs of my project, I made like a... yeah Well, we've provided them with the shapefiles that they could pour through the data themselves, but also I made a um an Esri dashboard so they're able to click around and look at different scenarios, identify like where there are multiple roads that are that are sort of flooded. I also, as a part of my research, I used graph theory to try and determine how the network accessibility is going to change under these different scenarios. so
00:19:29
Speaker
and Are they going to have to drive a lot farther when the road is flooded? Is it even connected at all? or Are people isolated when the roads are flooded ah to sort of help inform future decision making there?
00:19:41
Speaker
Yeah. How did you like building the Esri dashboard and and applying graph theory? The Esri dashboard was time consuming. ah you know i sort of yeah I've had this data and and that was sort of you know frustrating, but I think, I mean, look, I made it really pretty. So I think you know that'll be good for them.
00:20:00
Speaker
ah The graph theory stuff was super interesting ah just because you know that's not something I look at every day or frankly, I'd never even heard of until I was looking into how people assess transportation, like flood impacts on transportation networks.
00:20:15
Speaker
because before that, it was just like, oh, I found the wet roads. And like you know that's not that's helpful on some level. But I thought using the graph theory to sort of identify like, oh, no, like these areas are going to be trapped if we don't.
00:20:28
Speaker
like do something about that. I thought that brought the project a little bit to the next level. Plus I got to, you know, flex my Python muscles. And that was, that was a lot of
Hydrodynamic Models and Their Applications
00:20:37
Speaker
fun for me. not good at a lot, but I'm, I'm real good at Python.
00:20:40
Speaker
Good to know. I'm always looking for Python coders, right, Ryan? Yeah, love Python. Big, big fan. Dude, that's awesome. So it sounds like FEMA's regional flood maps.
00:20:53
Speaker
And, you know, you did something on a local scale for Georgia DOT. but applying and taking that data and applying it and and using that data in a certain way that makes it applicable for a specific application, right? It's like, it's the asset management. It's the, are people trapped or are they not? it's It's taking it to that usable, like functional level, right? making For decision-making, decision-making tools. Is this something that ah you guys could see this method or platform being developed further for other vulnerable coastlines?
00:21:27
Speaker
Yeah, yeah. I mean, you know, it sort of feels egotistical for me to say like, yeah, people should do what I did. um But absolutely, I think, you know, I think it has its merit, especially on the East Coast and and the Gulf Coast in particular. um yeahre We're seeing a lot of similar studies, especially in like, you know, Norfolk, Virginia, or, ah ah you know, New Orleans.
00:21:51
Speaker
So other similar studies are going on across, especially using the ADSERC, because a lot of times they're using ah the bathtub method, which can sort of basic model that oftentimes, especially under higher sea level rise scenarios, will over predict flood extent.
00:22:06
Speaker
I think having the hydrodynamic model gives it a little bit more accuracy and a little bit more, I don't want to say like meaning, but it I think i think it is it does provide more meaningful data. I actually also part of my master's, I carried out the bathtub method as Noah uses it just to see how it compared to my data, just ah how the how the flood extent was similar and how how the roads impacted were similar. And it underpredicted for the present day, it underpredicted the flood extent compared to the AdCirc model. And then, like I said, in the higher sea level rise scenarios, it
00:22:38
Speaker
it really quite overpredicted ah flood extent. And so I think the more meaningful hydrodynamic model sort of takes a little bit more into account, a little bit more natural processes into account. I think think that was something that might my project had. Thank you to Dr. Pilsky that sort of gave it that extra edge.
00:22:55
Speaker
Sure. For our listeners who may not know, what is the bathtub method? Yeah, so the bathtub method is, you know, you have your initial water surface elevation, you increase it by x amount, whatever the the change in sea level is, you do a sort of check to compare our adjacent tiles that were previously dry, are they now below that water level? And
Ryan's Thesis on Sediment Transport Modeling
00:23:19
Speaker
if they are, it sort of fills in that water level uniformly across the surface to to determine a flood extent.
00:23:27
Speaker
And what's the difference with the add circ method that you actually used? it's going to take account a little bit more. It's going to take account roughness. It's going to take account like actual behaviors of water. that I mean, that's sort of a weird way to say that, but it takes account more more natural processes of of water surfaces.
00:23:49
Speaker
And also, I think when you assume that the water level just extends uniformly across without any sort of drawdown, you're sort of over predicting the flood volume that is sort of being filled in there.
00:24:00
Speaker
Cool. I learned something. I'm the only non-modeler here, so I'm trying keep up. Yeah, I hope that made sense. Welcome to the club, Deb.
00:24:12
Speaker
Ryan, how did your thesis go and what's that about? Yeah, sure. So my thesis is kind of at the um intersection of how models talk to each other at different scales and how we model things at different scales.
00:24:26
Speaker
And I think the main motivation can be summed up as like sediment transport is hard. um just because it's a lot of physics going on. ah So when you think about all of the things that go into sediment transport, you have the water, you have the individual grains of sand, ah you have potentially the atmosphere as well. So there's three phases and you have to make different modeling choices when you do this, right?
00:24:48
Speaker
So in coastal engineering, we're all really familiar with wave models for the most part, um but sediment transport modeling is almost at the grain of the sand. um So there's this kind of scale imbalance issue there.
00:25:01
Speaker
So when you look at some of the main sediment transport and beach erosion models, such as like X beach or not actually sure what some of the other ones are, but um a lot of them don't necessarily do a great job at the waves because they have to spend so much effort on the sand part.
00:25:17
Speaker
um Or you could just choose to resolve everything. That would be CFD. So computational fluid dynamics. And that's when you need supercomputers to run for like six months, which is just not feasible. So my thesis is kind of splitting the difference between these different scales and saying, OK, well, we have these sediment transport ah coastal models that don't necessarily have a great sense of the waves.
00:25:40
Speaker
And we have these hydrodynamic models that do a good sense of the waves, ah but don't model the sand at all. Is there a way we can kind of have them talk to each other and pass information on to each other?
00:25:52
Speaker
um So my thesis is using FunWave, which Marissa, I know you're very familiar with. But our listeners aren't Yeah, so fun wave is in the coastal sphere somewhere in the middle. So it's not super quick, but it's also not super slow either.
00:26:09
Speaker
So it does waves pretty well. um It doesn't do sand, as I mentioned. um And it can actually resolve the shape of waves as they shoal and go on through all of these nonlinear transformations.
00:26:21
Speaker
um So those shapes end up being really important for sediment transport because the way that the wave shape changes as it reaches a sandbar, breaks, maybe refracts, or does other processes in the near shore, will move sand around under it as well.
00:26:36
Speaker
And that ah process of the wave shape changing is estimated by other models. So the main idea of my thesis is what if we just run in fun waves like thousands and thousands of times, try to get those wave shape parameters out and plug those into other models.
00:26:52
Speaker
So that that's the larger vision. um And right now, the main focus is just leveraging kind of toy cases for 1D transects and then also some well-studied beaches from the Army Corps.
00:27:04
Speaker
And the main way that those models are fit are through machine learning methods, um mainly convolutional methods, so um which machine learning sounds a lot scarier than it is sometimes. but ah convolutional methods just means that it's spatially aware. So a beach obviously has spatial features, right? So um it varies in space and this kind of machine learning architecture can handle that really well.
00:27:31
Speaker
um So the hope is that these parameterizations based on these wave resolving models like fun wave could hopefully improve sediment transport models. um And I mean, it's hard to always talk about sediment transport because the uncertainty bounds are always so high.
00:27:47
Speaker
Hopefully that having this information will improve something in the model. I love that you talked about sand. I love sand. um but Okay, so FunWave does the wave, not the sand.
00:28:01
Speaker
Which other model does the sand? So Xbeach, I think, is probably the most common for um those kind of purposes. That's the one you've used, trying trying to couple it with FunWave? Yes.
00:28:13
Speaker
Yeah, Xbeach only has a weak sense of what the wave shape is ah because it doesn't actually directly model the wave in the same way that a model like FunWave would. What about Xbeach non-hydrostatic?
00:28:25
Speaker
So that's a good question. So that actually does do what you're saying, but it's more or less FunWave at that point. Or it might be more like Splash. I haven't actually fully looked into that yet, but um I don't know how many people are using non-hydrostatic for the purposes of like ah engineering projects because that still is a pretty intensive code.
00:28:48
Speaker
You'd be surprised. no What's non-hydrostatic, XBeach, Ryan? The framework in general could be applied to non-hydrostatic as well. I think the main novelty of us using FunWave is that it's a Delaware product.
00:29:02
Speaker
Um, so we know how it works. Um, it's developed in house here. ah but in theory, the idea is that this general framework could be used for any wave resolving model, um, which would include non-hydrostatic as well.
00:29:14
Speaker
Um, so, and also things like SWASH, any of the other wave models that are out there. So towards Deb's question, what does non-hydrostatic mean? That's a good question. So non-hydrostatic without getting too far into the math, uh,
00:29:29
Speaker
my understanding is that it just does pressure in a much more intense way than some of the other models do. um So you have a much better sense of the hydrodynamics in terms of how that pressure field is changing and interfering with the other velocities.
00:29:44
Speaker
um So it sounds minor, but it takes a lot more computational power to actually get that pressure ah just because of the kind of equations you're solving. um But the kind of payoff to that is that you have a much more accurate sense of what is actually going on hydrodynamically.
00:30:00
Speaker
um For the hydrostatic kind of things, you're just often assuming the weight of the water is one of the main ah contributions to pressure versus the non-hydrostatic, you're considering those other dynamic effects.
00:30:13
Speaker
Thank you. And I feel like you're ready to give us big um mathematics equation with your blackboard for listeners here who don't see our cameras. Ryan has a huge board behind him with some equations there already. i feel like he's ready to explain us in detail how it works.
00:30:33
Speaker
Well, thank you for the breakdown for hopefully other listeners like me don't know much about it. So I'm learning a lot. i love this. Yeah. Check out more about models. So what what I like about modeling, um there are a lot of models out there.
00:30:50
Speaker
And there, don't want to say that there are fights, but there are models that have overlapping applications or overlapping physics.
00:31:01
Speaker
Some might overlap a lot versus overlap just a little bit. For example, like you have XBeach. so xbeach is delterus version so delterus is the company out of the netherlands so it's um it's not open source right it's it's you have to get the software from you to license it essentially get that get it from delterus um that's same same people make delft 3d um do they they're swash as well yeah i don't know if that's deltaris or i know that comes out of delft don't know if it's deltaris specifically
00:31:32
Speaker
Okay. Comes out of Delft. they're They're like, are they different? One's a university and one's like that organization. and then there's also IHE. It's confusing.
00:31:43
Speaker
don't know. The Netherlands smart modeling people group that have all of these these products that come out. um Like Sphinx, like that new one's like super fast, implicit,
00:31:57
Speaker
Non-linear code? some i don't know. Sphinx. That's S-F-I-N-C-S. That's one of their new ones that's coming out, for example. And so I was saying like X-Beach, right? That is a Delterra's product.
00:32:09
Speaker
um The Army Corps, more or less equivalent... Loose equivalent is like seashore, which I think is just stands for cross shore modeling. So that's like usually your 1D transect. And that does it's a phased average model, ah but does the sediment transport in transect like ah fashion ah with your.
00:32:30
Speaker
depth of closure type things. I don't want to get too too fancy with it. But um I do want to see, and I'm not going to put you guys, I guess, put Ryan on the spot, but what is the what is the difference between a phased averaged model ah a phase average wave model and a phased resolving wave model?
00:32:50
Speaker
Yeah, sure. This is a big part of my thesis for sure. So the phase that average model is you're actually This is going kind of be using a definition for a definition, but you're averaging in time and you're not directly resolving the wave surface.
00:33:06
Speaker
You're more kind of looking at energy propagation in space or more often wave action, which is similar to energy. um So often you're not actually directly modeling what the wave surface looks like at any given time. You have an average sense of what it looks like.
00:33:22
Speaker
And then you can back out by using some theory what that wave shape is. um but it's still just an average sense in time versus the wave resolving models. You actually are directly modeling the wave surface that every time step in your simulation. So if you go into the code, there's actually directly a line that says, this is what the surface elevation is at this point.
00:33:42
Speaker
um So they take a lot longer generally than wave average models, just because the, uh, computations The physics have a lot more terms generally and often a lot more dimensions.
00:33:54
Speaker
um You have different ways around that by Taylor series or doing like layered models, as they call it sometimes. um But yeah the main idea is it's that you have that surface at every point in time directly and there's not really an approximation to get it.
00:34:09
Speaker
Yes. Perfect. No notes. You're going to pass. So there's there's a lot of different models out there, and it depends on the processes that you're interested in, the time scales that those processes need to go over. For example, when when Ryan was saying the the the um bridging the gap between the different scales of the processes from the wave time scales and spatial scales down to a granular piece of sediment um that's being acted upon by the waves.
00:34:37
Speaker
Requires different equations, different processing power. um And there's only no there is no one model to rule them all. there There is no one model that can do everything from atmosphere to ocean, to propagating waves near shore, to the surf zone, also compound flooding, like rain events, ah the hydrology component bringing that riverine input down and interacting with, you know, at that barrier ah with the sediment transport or even aeolian driven sediment transport. So windblown sediment.
00:35:16
Speaker
There is no one model to capture all of those processes. And if there were, it would do it very poorly because where there's no way that you could resolve all of those processes at the resolution, spatial or temporal, that you would need to accurately represent the full picture.
00:35:38
Speaker
So that's why we have all these collections of individual models that are targeted at these individual processes and potentially at these timescales. So um what is great about Ryan's research is that it's taking some more or less like more complex wave model that or a more high fidelity wave model.
00:36:04
Speaker
teasing out some of the important parts that can help inform some of the lower fidelity sediment transport models to make things more accurate, but also keep computational costs down ah and just save time, money, and resources for for everybody at ah at the end of the day.
Successes and Challenges in Coastal Modeling
00:36:26
Speaker
So I think even though Ryan's applying this towards XBeach,
00:36:30
Speaker
ah or towards XP applications. I think this could be useful for seashore, for example, are and that's for the Army Corps ah seashore. ah And ah potentially any other kind of wave, but also sediment models like CMS, the coastal modeling system, that's with Army Corps.
00:36:49
Speaker
That does, it has a waves, but it's mostly circulation. And it also does sediment transport and particle tracking pretty well. i think CMS, like tends to compete with AdCirc. Is that true?
00:37:00
Speaker
Alex, do you know, do you guys use CMS at all down there? No, we're pretty devout to AdSerk. Sure. I guess they're not one-to-one, no, because AdSerk has the unstructured mesh and CMS has the quad tree grid.
00:37:17
Speaker
And you can Google what that is, but it's like a non-uniform Cartesian grid. So I just thought it was interesting. Thanks for coming to my TED Talk. Any fun moments you guys had or things that you liked using your models?
00:37:32
Speaker
I guess it's it's always fun to see things come together in the end. Like, even though I've done it probably hundreds of times at this point, it's always nice to actually animate the wave field and kind of see what it looks like and to see that it's like actually reasonable and that your inputs make sense and that you're actually modeling something that's real. So if you have data, that's also exciting if it actually matches up somewhat.
00:37:55
Speaker
um So I think always the animation part and actually visualizing it is really fun because it's kind of, um to get there, you had to have had a successful model in the first place. So that's, that's half the battle.
00:38:06
Speaker
um So that's always a nice product, making fun little um videos, figures and all of that. um And then also just kind of the physics that you learn throughout it and the little tidbits you pick up here and there as you troubleshoot a model.
00:38:22
Speaker
Because I, like even, Today, like I've been using the same model like for two years. I'm always back on the wiki and the manual. um And you just really get to know it really well. and I think that's a really cool part about modeling. Once you actually like know a model really well, and you can start to say, oh, well, this part's doing this, this part's doing this. And I think it's just cool to know a system, like a complex system like that inside and out.
00:38:47
Speaker
I was just going to say, sort of piggybacking off of that, one of my favorite things is really when when people come to me with questions that I can answer. that That feels good. That feels like this is what it's all for is you come you needing me is is you know sort of the nice thing.
Alex's PhD Research and Future Goals
00:39:06
Speaker
He's employable, guys. i was about to say, see, not so much unemployable. Yeah.
00:39:15
Speaker
Well, let's jump right into what do you say. So, Alex, you're going to do AdSir plus one and some ex speech for your PhD. What is your research going to turn into? Yeah, so my research, and and I'm going to speak sort of generally, ah one, because you know I'm about three weeks in, and two, i don't I don't know how much I'm allowed to say, but you know ah for the project, it's funded by the Department of Defense.
00:39:42
Speaker
the idea The idea behind it is we're going to be using act speech Not exactly coupling or maybe maybe coupling. I don't know. We're we're still early on. i don't want to. But ah not not exactly. I'll say for now, not exactly coupling AdSerk and XBeach, but um using outputs from AdSerk in putting them into XBeach to try and ah you know determine an optimum design.
00:40:08
Speaker
with different structures ah to to figure out what what is going to work best for a given goal. And as we sort of go on, we're not quite sure what that goal is going to be. you know Maybe it's the main goal is sediment accretion.
00:40:24
Speaker
Perhaps the main goal is ah you know like reduce wave forcing. ah Perhaps it's finding some you know even balance between the two.
00:40:34
Speaker
ah but it Coming up with a bunch of different designs for for different structures to determine what is going to give us our best ah possible output.
00:40:46
Speaker
Like I said, we're we're early on. I'm speaking super generally, but I'm excited to really sink my teeth into it as as I learn more and more. It seems like this is more of a meeting the need for a process to answer those types of questions for what kind of structure is going to be your best option for this purpose.
00:41:12
Speaker
Right. Yeah. Yeah. Whether hard or natural, gray or natural infrastructure. Yeah. Yeah, absolutely. Structure has a lot of different meanings. True. In this context. And so, yeah, like I said, I don't know how much I can say, but. um Well, I mean, you mentioned DOD funding. So, yeah, let's let's not. ah Count our chickens before they get in the coop there. Hopefully that funding still exists. We didn't say sea level rise.
00:41:40
Speaker
Right. no it's resilient communities and future conditions. Irreversible flooding now. Irreversible flooding. Insert existential crisis here So Ryan, you are starting a PhD at University of Washington.
00:41:58
Speaker
Who are you working with again? Morteza? Yeah, Morteza Dirac and the environmental fluid mechanics a group of people there. So it spans a few departments, but I'll be in civil engineering.
00:42:09
Speaker
All those civil engineers. They're just flirting with coastal. Do you know, um so what is Morteza's research and what do you anticipate working on in your PhD or is that way too soon to even know? I just know you've been in contact with him for a little bit, so...
00:42:28
Speaker
Yeah, so um I know the general, it's using a different type of numerical method and a different kind of modeling that we actually haven't discussed yet. More in the realm of CFD, so computational fluid dynamics. So it's this model called, or it's this method called smooth particle hydrodynamics, which is really interesting.
00:42:47
Speaker
um because it can resolve a lot of things. So when you think of traditional computational fluid dynamics on like supercomputers, those can take a really long time to run.
00:42:59
Speaker
ah But because of all of the advances in GPUs and just computing power in general, some of these um smooth particle hydrodynamics methods are really fast, um almost as fast as FunWave for some purposes.
00:43:11
Speaker
um And you can resolve just a lot more. um So I think it's kind of a cool field to be in right now because um My understanding is that we're still seeing what processes we can resolve, what processes do we want to resolve?
00:43:22
Speaker
Can we do things like ah wave structure interaction, particle tracking? um So to my understanding, that's what I'll be working on um in some capacity. um The specific problems we're looking at with SPH, a little bit too TBD right now, um but kind of seeing how it compares to other models and frameworks, I think is going to be a big part of it, which I think is exciting.
00:43:46
Speaker
Sweet. I mean, and you get to go to Seattle. So yeah, leaving us for the West Coast.
Advice for Aspiring Coastal Engineers
00:43:53
Speaker
Well, I don't want to take up too much more of your time. But there is one parting question that we ask all of our guests who appear on the show.
00:44:03
Speaker
So Alex, first, what advice would you give to a student or ah young professional who's interested, I guess, in this case, a student ah who's interested in pursuing a career or position similar to yours?
00:44:16
Speaker
Yeah, I mean, if you're if you're looking to you know get into academia, it really is almost as simple as like just sending out an email. like Looking into like who has your research interests, what opportunities are available. look at you know go Look at different universities' websites, look at different labs, seeing what's out there and and just sending out an email and and really anything can happen.
00:44:43
Speaker
Don't be afraid to cold email. Yeah, I think the ASBPA's ah Coastal Universities Guide could maybe help you get started in identifying universities that have coastal related programs from undergrad to master's and PhD level.
00:45:01
Speaker
So feel free to check that out, listeners. And yeah, again. Brush up on your email skills. Those are important through the rest of this life, I guess.
00:45:16
Speaker
Ryan, what would you but what advice would you give? so first of all, I'm going to plug the Coastal List from the University of Delaware. It's an email server. If you just Google Coastal List, there's opportunities sent out almost every day.
00:45:32
Speaker
um and it really spans every type of research you can imagine. with coastal so you can find interesting stuff there. um And kind of piggybacking off um Alex's too, um just talking to people in general.
00:45:45
Speaker
In general, most coastal people will be very happy to talk to you about their research or what they do. I feel like this is a very talkative profession. So don't be too scared about reaching out and asking, hey, what is it that you do for a job? Or like, how did you go there? Or um you know, um what company would you recommend or something like that. So ah most people are very happy and just very legitimately excited that other people are excited in this stuff as well.
00:46:10
Speaker
um So I've had a lot of good discussions just from reaching out to people And just talking to people just based on, hey, you do coastal, I do coastal. And, you know, you can meet some interesting people that way. um And it really is a super small world, too. So once you meet enough, once you meet a critical mass of people, you can kind of, your two set you're like two degrees of separation away from everybody. so um yeah, so I think just talking to people and getting comfortable with that is really important.
00:46:36
Speaker
Yeah, we're always ah mentioning on this show that coastal is a small field. I mean, at the end of the day, there seems like there's a lot of folks that are doing coastal, but in the end, there's only ah a niche of civil engineers or geologists, mathematicians, like from all walks of life who are focusing on this narrow process from the foredune to the depth of closure or like where the waves stop feeling the bottom.
00:47:07
Speaker
Yeah, at the end of the day, networking, it seems like. Yeah. And I guess the other thing too, is just that um part of the reason I think coastal field so small is just because we're split between probably like 10 different departments in a lot of places.
00:47:20
Speaker
This is true. You also need to look at departments that you might not have taken directly. Like in civil engineering in undergrad, I would have never thought to go to like oceanography or something like that. But, um you know, there might be and college of the environment, engineering, geology, geography, and you know, that's where I think the coastal list is also helpful. Yeah.
00:47:38
Speaker
Cool beans, check that out. Also, just as a last thing, um sort of building on that, like, you know, being aware of things that are outside of your department, like take the classes, you're going to continue, take the classes that really, really aren't fun at all, like that are really you know gonna hurt you emotionally. Last semester I was in Bayesian statistics course and like I would say most of the time I didn't understand what was going on. I really struggled, but I gained a lot of knowledge ah that I'm gonna hopefully you know bring with me forward.
00:48:11
Speaker
It's true. there's There's certainly no harm in taking fun like fundamental classes. And Bayesian statistics is not fundamental, but it's in like math is a fundamental to coastal modeling. Physics is a fundamental to coastal modeling. Statistics is a fundamental Even if there are classes in or seminars in um like data visualization and data processing, that is most of our job is data processing at the end of the day. Also writing.
Closing Remarks and Sponsorship Acknowledgments
00:48:48
Speaker
All right. As we wrap up this episode, we would like to thank the New Jersey Sea Grant for sponsoring the podcast. And announcement, in case you didn't already know and weren't already prepared, the ASBPA National Coastal Conference is taking place from the 7th to the 10th of October this year and will be back in Long Beach, California.
00:49:10
Speaker
While the presentation abstracts period has closed, the poster abstracts period is still open until August 15th. Get in those poster abstracts.
00:49:21
Speaker
And I believe ICCE is the International Conference of Coastal Engineers is coming to the United States for the first time in a little while. And that is taking place in May of 2026.
00:49:36
Speaker
Abstracts are open for ICCE through October of this year. Check that out ICCE and get your abstracts in. Now, if you found this episode insightful, including all of my blabbering, please share it with your colleagues and let's keep the conversation going. We'd very much appreciate it if you like, rate,
00:49:56
Speaker
subscribe, or follow. Lastly, do you enjoy listening to this podcast? You too can support Going Coastal with New Jersey Sea Grant and sponsor episodes that promote your brand.
00:50:08
Speaker
Contact us at our new email at podcastgoingcoastal at gmail.com. Thank you both, Alex and Ryan, for being here. We really appreciate it. Yeah, thanks for having me Yeah, also thanks for having us on. It's always fun to talk about modeling. Hell yeah, same.
00:50:25
Speaker
Great, thank you guys. Bye-bye.