 All right, good morning, everyone. And how is everybody doing? Yeah, it's so exciting, for sure. There might be some people online. The camera is over there. So when everyone wants to say hi to those online, there's a few right now, but if you're tuning from other parts of the world, other parts of the US, welcome as well. I'm Paula Pasaraco. I'm from UT Austin. I'm the chair of the steering committee of CSDMS and with that role, I get the honor of welcoming you all to these 2023 meeting on patterns and processes across scales. So we have three days of talks. There's posters that are up, there's clinics and they're gonna get us to think about space and time scales and process scales and how they impact the dynamics at the Earth's surface and does impact the modeling that we do. And in these three days, I also wanna invite you to get to know our community. And I think it's important to remember that CSDMS stands for test. We'll do a quiz on all of you. So it's community surface dynamics modeling system and that's C of community, it's us, right? But I think it's important to remember that that community grows. And so in fact, who is in the room as a first attendee to CSDMS? Let's hear it, oh, no hands, just to make noise. Let's make some noise. All right, well, welcome to all of you. It's quite a number. And so I wanna encourage all of you. I think this is a meeting where lots of us meet friends and see people that haven't seen in a while. But I also wanna encourage all of you to just take advantage of the meeting and make it a goal to meet someone, introduce yourself to someone you never met before. And I know it's sometimes uncomfortable, you know? It's a little creepy sometimes, but since I'm saying it and I'm putting it as a goal, I think you should make it a goal of every break to go to someone you never seen before and just introduce yourself and talk about stuff. And you may become friends or collaborators or things like that. There was a symposium I was at a few years ago where the facilitator did that, I was in the audience. And then soon enough I became the chair of that group. So there's like something to say about just getting to meet people. So with that, so we're talking about CSDMS and community. And so I have two important announcements to make and they're coupled, like our models. The first one, and I just got to warm up for those applause, so I wanna hear it really loud. I'm excited to announce that CSDMS is launching its new five-year mission, just recently funded by the National Science Foundation. And so we're very excited, we're thankful to the team for landing that. We're thankful to NSF for funding it. And the second announcement which is related to the CEO community I was talking about before is the fact that I'm really excited that the CEO community is very much at the center of what the next five years are gonna look like. And we just need a round of applause. We know some of the people in the room, the question that the next five years are gonna be focusing on in addition to other things is who is not in the room yet? Who can we engage with, that we haven't engaged with enough until now? You can think about people of schools that may not be able to afford coming here. You may think about residents, all those communities, we model hazards in our research, right? What about the people that are directly impacted by those hazards, can we engage with them? And so the next five years are really gonna move us closer to these dynamic modeling system for and by the community. And to tell us more about that, I'm excited to introduce Greg Tucker, Professor of Geological Sciences and Executive Director of CSTMS. Thanks, Pella, for emphasizing that C in community. So my task here, before we dive into the keynotes, is just to give you a quick update from the integration facility. And it's great to see so many people who haven't been to a CSTMS meeting before. For your benefit, let me tell you a little bit about what CSTMS is. So first of all, pronunciation. If it's five letters, it's a little awkward, I realized. But if you say it with a soft C, and then you say the D kind of hard, it sounds like systems or systems. So you'll hear people calling it systems. So systems is actually three things. So it's a modeling system, and I'll say more about that in a second. It is also a community. It's all of us and currently 2,300 of our colleagues worldwide, and it is a facility funded by the National Science Foundation. I want to echo Pallas thanks to NSF for supporting this community and the work that you all do. The integration facility is based here in Colorado and is a small team that works in community computing and education to try to advance and support the science that you all are doing. But let me say a few words about this modeling system. What is this modeling system? It's in many ways unique among community models and modeling systems in ways that reflect unique aspects of the sciences of the Earth's surface. So the sciences of the Earth's surface, they're complex, they're dynamic, and they're creative. What do I mean by that? Well, okay, every science is complex in its own way, but if you think about other disciplines like, and very closely allied disciplines, like atmospheric science, solid Earth science, oceanography, these are all sciences that wrap a piece of the Earth's system, a chunk that has substance. But the Earth's surface is different, it's an interface. It's where the solid Earth meets the cryosphere, the oceans, the atmosphere, most of the biosphere. And of course, it's the place where all 8 billion of us live our lives from cradle to grave, right? So the Earth's surface is not a substance, it's an interface. And from the point of view of modeling it and understanding it theoretically, that means it has to be an interdisciplinary modeling system, right? We have to be able to draw on pieces that represent these intersecting spheres. And that in turn means we need a modeling system that is flexible, that is modular, that draws on other communities, and above all is fair in the sense of findable, accessible, interoperable, and reusable, right? So the dynamic piece here, I mean, I don't just mean that the processes are dynamic, though they are that, but that the science is dynamic. It's changing, right? I mean, one of the really exciting things about working in the various sciences that touch on Earth's surface is that they're moving fast. We're getting new data and new discoveries. From the point of view of a modeling system, that means we need a system that is not static, doesn't have one set of equations that we've decided on and are done, but it's gonna evolve. It has to be built to last, and again, it has to be flexible and fair. And finally, the creative piece. I mean, I'm guessing that at least in my experience, friends and family who are not in research don't always understand how creative an enterprise it is, right? People sometimes think of it as being kind of mechanical, but you all know, right? This is a creative business. And when you're coming up with new theories, new ways to interpret data, when it comes to the computational side of that, you ideally need paints and pallets and canvases, things that you can quickly sketch out and explore, rather than getting bogged down for five years just debugging code, right? So again, flexible and fair. So what does that actually end up looking like? The F and A piece is the findable and accessible. One way that the CSTMS community-wide infrastructure tries to address that is through the model repository. So many of you know this is an online portal that currently catalogs over 400 community written codes and tools, and that's a lot. So to help you sift through that, there are advanced search tools. And often the question comes up, is Model X gonna be the right tool for the job for any particular application? And to help you answer that question, Albert Kettner has engineered the system so that it can pull in a vast bibliography of references about these 400 odd models and their applications. When I say vast, I mean over 20,000 references and growing. So that's the model repository. But a model repository by itself, of course, doesn't constitute an integrated modeling system. And if you want to, for example, couple models together or compare them side by side or chain them together in an integrated workflow, you need to solve the I piece. That's the interoperability piece. And that's what, among other things, the basic model interface is for. So anybody here heard of the basic model interface? Or hand fight? Okay, now you've all heard of it, so I just told you. So this is basically, it's an interface standard for numerical model codes. And it's effectively a list of roughly 30 or so language agnostic functions that when added in a particular language to a particular code, make that code operable as an experimental object where you can exchange data, you can run it, you can pause it, you can query it and so on. So the basic model interface has proven useful for all kinds of modeling, model coupling projects as used, for example, now by the National Weather Service for their water modeling. We're gonna hear, I think, later this morning about an application in the US Geological Survey it's used in Deltaris and E-WaterCycle to mention a couple. So it's useful, but where it really comes into its own is when you also have some kind of a framework for executing and exploring models. BMI itself is framework agnostic, but to provide a framework for our community, we have the Python modeling tool. So this is, effectively this is using the Python language itself as a framework, putting a Python front end on two numerical codes written in different languages so that you can bring them together in a common environment. IMT adds some tools like re-gritters and it gives you access to several model codes that have already been wrapped as they say. And the point is that a relatively short Python code can then get you up and running quickly with coupling models. So I don't expect you to read all these lines of code but the point is that there's only a few of them and they are already coupling two models together. So PyMT is useful, but it does not address one of the things that you all know comes up all the time when you're working with numerical models and as wrangling data sets, right? I mean, when you think about wrangling a big data set, often the first thing I think of is headache because I'm dealing with big files, I'm dealing with different websites and different formats, it can be a barrier. And so to try to overcome that barrier to some extent anyway, we've been experimenting with the concept of data components. So a data component is basically a piece of Python code that programmatically goes out and fetches a subset from a particular data set, whatever subset you're interested in. Currently we have a handful of data sets, data components that are listed here that include things like meteorological and weather data, surface water, soils, topography, ocean model output, waves. Anybody can write a data component by the way, and if anybody's interested in doing, if you have an idea and you want some guidance, let us know and we're happy to help. So between BMI, PMT, data components, we have the makings of an integrated modeling system, but there is one need that these tools don't really address and it gets back to that creative piece I mentioned earlier. Now, you all will probably know from experience that when you're thinking about new ideas about how to interpret that data set, let's say, and you have a new idea about some concept that you want to explore and you want to explore it computationally, it's awfully helpful if you can pull on individual elements from existing numerical models and then add your own twist to it, your own insight, your own innovation. And to that end, it's helpful to have some kind of aspect that is more granular than complete numerical models that allows you to take chunks of them. And so to try to meet that need, there is a toolkit called LandLab. This is a Python language toolkit and it's designed to promote fair modeling by giving you a toolkit where you can build new numerical models relatively efficiently compared to starting from scratch. And a common design of a LandLab built model is kind of sketched here in this slide. The idea is that you have a little Python code that makes a grid object. So here's one particular type of grid, there's several and then it might instantiate one or more process components written by you or written by others in the past and shared. And then combine that with some new elements you've come up with to create a new novel numerical model without spending five years debugging. So that's LandLab. And this is a good point to reinforce what Paula said earlier. The idea here is not that there's a handful of wizards sitting in Boulder who are producing things for you to use but never touch. The idea is this is a community built modeling system. And I'm really happy to see that we're up to now 35 different contributors to the code base of LandLab and that's growing. So many of you have contributed it, contributed to it. So thank you. It's been used now in over 60 different publications in applications ranging from like frog habitat to tectonic geomorphology and like marine and coastal stuff. I should mention, despite the name, it's not just for land. So that's good to see. Clearly though, there is a learning curve associated with these tools like any tools and to help meet that learning curve and make it a little more gentle, we've really pushed the development of tutorials and lab exercises. So there's a growing collection of tutorials online for technical things like particular LandLab components for concepts like deltaic sedimentation or glacial advance, you name it. Many of you have contributed to that collection. Know that this is for all of us to create together. So I encourage you to contribute and take advantage of that resource. Now, for educators, one of the things you know is that, okay, there's some nice lab exercises here in Jupyter notebooks that you can have your students work through to learn about a particular concept, often using a model as part of the learning process. But when you think about trying to get a class of 50 people to install on their laptops, you think, oh God, I'm not going there, right? So to try to overcome that limitation, we have been running a set of what are called Jupyter servers. Many of you know what these are. These are essentially cloud-based programs that you can access through a login. I think everybody here now has a login whether you want it or not. You can log in, you can run notebooks, you can have a Python session, you can have a text editor and so on. We have three of them at the moment. There's the Jupyter instance, which is for running tutorials and testing stuff out. There is the lab version, which is for labs and classes and things like that. And then there's a relatively new one called Frontier, and that's for actually doing production runs. Like if you have a model and you want to do 200 different sensitivity analyses with different parameters and your laptop is gonna die if you try and do that locally, we can set you up to try it there. So this is a little bit experimental, but so far it's worked really well. Now, even this sort of level of convenience still doesn't take away the need to have a certain basic skill set in cyber techniques, if you like, to access the community modeling system. And so to try to meet that need, we've also been running a variety of educational programs. So we've just concluded the fourth iteration of the Earth Surface Processes Institute, and thanks to Irina and Mark and their team for running that. So this is a six-day training program for early career scientists that provides, gets you spun up with numerical computing in Python, working with numerical models, various topics. We realized we're not able to meet the full demand so far this program has been oversubscribed by three or four, but we're getting there. We've also started a new program called Roadshows in which Mark or Eric or Tien will go out and do a short course on something like introduction to scientific computing for let's say three days at a different institution around the US. And we're especially eager to target minority-serving institutions. So we've done one so far, we're hoping to do about two of them per year over the next couple of years. We don't have the schedule fully mapped yet. So if you're at a minority-serving institution or you are connected to one in some fashion and you'd like to see us go there, let us know. And we can look at putting it on the schedule. I can't promise we'll be able to serve everybody, but that's what we're trying to do. So these are in-person events that doesn't always scale up to the 2,300 people currently in our community. So we're also working with online resources. I wanna point out just a few of these. Many of you know about the CST-Miss Help Desk, so you can go online and post a question and one of our engineers, Mark, Eric or Tien, will get back to you hopefully with an answer that solves your problem. We have an Office Hours program that's new so you can basically sign up for a video conference to have a one-on-one consultation with a research software engineer to try to get sorted out. We also have a new forum so where the Help Desk is meant for you to ask questions of a technical expert. The forum is for posing questions and discussing among one another as a community, from the scientific to the technical and everything in between. There is the webinar series many of you know about that currently we currently have, I think a few dozen now recorded webinars. So take a look at that. And finally, we're working toward building up a set of self-paced learning resources where someone can go online and get ahold of a series of notebooks on topics that you're interested in learning about and just work through them at your own pace. And one iteration of that is called CSTMS, IV that Mark has put together. So I encourage you to check that out as well. Let's see, so finally on the stuff from the integration facility front, if you have a project or a writing proposal for a project and you would like support from a research software engineer to think about for example, how would I turn my code into a fair and lasting resource that lives on after the project has a broader impact? We're happy to work with you on that, just let us know. Okay, final thing to do is to say some thank yous. First of all, as again, as Paula said, this is a community organization and it's led by the community. The leadership body is the executive committee and that consists of the chairs of the various working and focus groups. So I wanna thank all of the chairs that chairs who are here, can you raise your hands just so we'll see who you are. So these are your contacts. Also wanna thank Paula and the other members of the steering committee for their advice and guidance. This meeting would not have happened without the work of a lot of people. And I wanna thank the staff of the integration facility. I wanna thank the meeting program organizing committee who put together the program that we're about to enjoy. And I wanna thank the various volunteers who are helping out in the day-to-day running of the meeting. Okay, so last thing here is a map of the meeting over the next three days. We're gonna have keynotes in the morning. Later morning, we'll leave there a breakout sessions or tomorrow we're gonna hear project presentations from the summer institute along with a brief update from NSF, Unrelevant Programs there. First thing in the afternoon, we'll do clinics. So if you have forgotten which clinics you signed up for, it's in the back of your badge. And then in the afternoon, we've got poster sessions today and tomorrow, we have an award ceremony tomorrow afternoon followed by after the poster session of banquet here in this building. Whether it should be nice or it will pour rain one or the other. And then on Friday, we'll wrap up with a panel discussion on applications of AI and ML and Geoscience. So that should be fun. Okay, I think that's all I got. I'm gonna pass it over to Albert Kettner now.