 Okay, great. Thanks, Brad. And welcome everybody to sunny Boulder, Colorado. It's where life is full of surprises. My apologies to those of you caught up in the bus debacle. I'm glad everybody seems to have arrived safe and sound, and as they say on airlines, we'll make up the time in route. Okay, so as Brad mentioned, this is the first annual meeting of systems 3.0. So we had systems 1.0 starting in 2007, 2.0 in 2012. Now we're in systems 3.0. So I want to first thank the National Science Foundation who underwrite this this community effort in this community. And this year I also want to give a special thanks to Kuwazi. And you're gonna, you're gonna hear more about Kuwazi and their mission and how Kuwazi and CSTMS have been working together. I think in various ways in this meeting, and you'll hear from Jared Bales in a few minutes to learn a little bit about Kuwazi. So I think Brad has nicely explained a little bit about our theme this year of bridging boundaries. We're bridging boundaries across natural systems, human systems, and we're bridging boundaries between one another, learning from one another, learning new tips and techniques, we're making new friends, and of course that makes the science enterprise better in it, and it makes more fun. So that's bridging boundaries. So I'll say a few words about some updates from the community and a little bit from the integration facility. So our community is growing and on the upper left here is a chart showing our numbers, the CSTMS members through time over the past 10 years. So we haven't been necessarily trying to grow, but growth has happened organically on its own. Earlier this month we hit a milestone, 1800 of you as CSTMS members. And it's an international group. So we're represented currently by 70 different countries. How many of you are here for the very first CSTMS meeting you've attended? Raise your hand. Good. So about half of you. So welcome, special welcome to those of you who are at your first meeting, hopefully the first of many. As you may or may not know, the community is organized into a collection of focus and working groups. You see them listed here. You'll have the opportunity to meet the chairs of the working groups you're interested in later today when we do our breakout group meetings. We have a couple of new co-chairs this year and I want to welcome them. So joining Nicole Gasperini as a new co-chair of the Tresor Group, we have Leslie Schief from the USGS. And joining Courtney Harris as co-chair of the Marine Working Group, we have Mike Stechler from Lamont Doherty. So welcome and thank you to Leslie and Mike. Okay, so on the note of community, there's one other kind of boundary bridging that I want to say a few words about. No, it's it's unfortunately the case that in the world of science, at least in the United States in recent history, that the community, the science community as a whole has not been equally welcoming to everybody. For reasons that have nothing to do with the science and everything to do with social and and and historical factors. So it feels really important to say loudly and clearly and frequently that the CSCMS community is an inclusive and open community. It's for everybody. And so, so if you're a male or female or non-binary, if you are young or old, whatever your height, your size, your color, your background. If you're interested in the things this community does, then you're welcome here and you belong here. So I think that's let's recognize that. So on the other hand, this is an easy thing for someone like me to say, and the reality is that the various science communities have a ways to go. So in geosciences in the US, for example, Although things have gotten better, the geoscience community still doesn't look as a whole like the American population. So to help us all collectively think about what we can do to make that situation better. We're going to have a special event toward the as the capstone event this meeting on Thursday afternoon. We're going to do a panel discussion on the issue of equity diversity and inclusivity And share some ideas and get some information about what what can we all do to make sure that this community is open to everybody and all the talent out there that's interested is at the table. So look forward to that on Thursday. Okay, so a couple of thanks for the integration facility staff. Many of you know these folks. So this is a collection of folks here who are supported by NSF to do the put on meetings like this one and provide the products and services that you've come to know and love. So thank your local integration facility member when you meet them. In the comings and goings. I also want to put a special thanks to those who have put on pre meeting workshops. So we had a pre meeting workshop yesterday on software carpentry led by Mark Piper, Mariela Perignon and Charlie show where you guys raise your hands. You are okay. Thank you. This is a in some ways a kind of a bittersweet. Thank you to Mariela in particular because we say goodbye to her later this month as she heads off to bark on a new career in data analytics. Thank you, Mariela and and good wishes. I also want to give special thanks to Jeff Carras was Jeff. Thank you, Jeff for for leading a fabulous Short course on quantifying uncertainty. Hugely important topic in our science. These folks are volunteers who volunteered their time to make these These lessons available to you. Okay, so I'll say a few words about updates from the integration facility. As you may know, there are sort of three key areas that CSDS works at this community support and development enhancing research and discovery. There's computing resources and there's education. When CSDS first started one of the first experiments that the integration facility tried was, hey, if we make A web portal and invite people to openly share their codes. Will they do it. This was a time when codes were trade secrets, circa 2007 and the answer was a resounding yes to that experiment. Yes, people are willing to share and happy to share and eager to share. And so as a result, we now have a model repository that has 229 models 89 tools and the list keeps growing every year. So here's a list of some of the codes that you all have donated over the past year. So thank you all for making your hard work accessible to the rest of the community so that your, your impacts are, are bigger and deeper. Another thing that the integration facility does is to provide support for you all in doing your science, both through developing proposals and through the sort of project base. So some of the things that we can, we can offer you are in the proposal support category. We write a lot of letters of support. We're happy to do that. We can provide help with broader impacts thinking about how to turn your models into educational resources or to make them more broadly accessible. We can help you with data management plans. In terms of project support. So we've, we've supported PIs for a long time in an informal way we want to make that a little bit more formal so it's more visible to the entire community rather than people who sort of feel they know us well enough to do that. We can do this in a variety of ways. We can provide us a limited amount of consulting with our software engineers are there to sort of help you deal with things like how do you, how do you engineer a model so that it'll work in a framework and that kind of. If you have a more extensive need, we're happy to collaborate with you. We do a lot of sort of sub awards on on proposals and that kind of thing. We often have site visits so somebody will come out for a day or two days or a week or something like that and we'll work with you here in Boulder. So know that those resources are available and you can find out more on the website if you go to services. One new service that we just rolled out last month is the help desk. So here's the URL. And so this is essentially a web portal through GitHub where if you have questions about one of our frameworks about basic model interface which we'll talk about in a minute about particular models. Post your question here and we'll do our best to help you and if we don't know the answer we'll try to steer you to people who do. So take advantage of that. As you know we also coordinate meetings like this one. Brad mentioned last year's meeting on the theme of natural hazards. So we have a special issue in the journal NHS natural hazards and earth system science on computational modeling of due processes and due hazards. So you're in luck because the deadline for submission isn't for another month. So you have something relevant. See Albert Albert is there. The guy with his hand up. I think it'll be an exciting issue. So here's a sampling of some of the papers that that have come in so far the topics range from droughts to storm surge and coastal storms, riverine floods, post wildfire flooding, seismic hazards, and you name it. So I think that'll be a great issue. Hardware and software. So so for many years CSDMS has provided us sort of a modest supercomputing facility for those of you who want to get into supercomputing and try it out and develop develop a familiarity with it. So we still do that. Our old facility beach has wound down and we're now using and offering you all if you need it a cluster of nodes on a shared system here at CU Boulder called Blanca. And what it lacks a number of cores it makes up for in speed and memory. Once you've developed your sort of bona fides and demonstrated that you know what you're doing on an HPC facility we can help you translate that to a bigger higher performance system. There's one system that several of you are using here at CU. It's an NSF funded facility called Summit has nearly 10,000 cores. So there's lots of lots of computing power to be had. Let me tell you a little bit about model coupling and model frameworks. So another experiment that CSDMS tried early on was having heard from the community from you all that we don't want one single large model. That would have in some ways been simple like know how to make a code and maintain it. But it was clear that from you all the community that what we needed as a solution was a modular plug and play environment that would be able to span the many different scales processes environments that you all work in. And that's a harder challenge. So one of the first questions was is this possible. Is it feasible a and B what would be required. And so one of the lessons learned that one of the you need a couple of things you need multi language capability, you need a vocabulary and that's led to Scott's work on geoscience standard names, and you need a model interface standard. So let me tell you about an interface standard. So, you know, probably most of you here have a driver's license and have driven some kind of a motor vehicle. Maybe some of you even rented a motor vehicle when you flew in and you probably took it for granted that regardless of what you rented whether it was something small or something slightly larger, you know, sometimes they upgrade you that whatever you're driving with is a dump truck or, you know, Humvee or whatever. It's going to have a standard interface. So the standard interface for a model code is the same idea. Right. And so we've developed at CSMS this basic model interface. It's really just a list of functions. And if the model provides those functions, then you can say it's standardized and it can be easily brought into a coupling framework, whether one of the ones we've developed here or others out there elsewhere. And, you know, it's not it's not rocket science in a way it's pretty simple that kind of functions you need or something like I want to start my model and give it some inputs, I want to run it for some period of time or if it's a one time thing ask it to do its thing. I might want to interrogate its values. What's the C level. What's the sediment thickness. What's the discharge. What's the biomass whatever it may be. When I may even want to modify values and this is the key to coupling that I can modify that values in the model while it's running. We do that with cars too. Right. And then when all is said and done we want to clean things up and shut down smooth right so that those are five of the BMI functions. And so we have if you want to learn about the BMI and how you can add it to your code if you're a model developer. There's several places you can look there's a paper by Scott and Eric and boy Anna. There's a clinic happening later today. Led by Mark there's a website where you can learn all about the BMI we even have now a webinar about BMI so you have no excuses to learn about the BMI. So once you have wrapped a model with a standardized interface then it can be brought into a model framework. And that was another of the experiments early on OK can we build a model framework that would accept different codes written in different languages and allow people to couple them together where that makes sense or to run single ones. That led ultimately to what's called the web modeling tools some of you have used before. It's basically a browser application where you configure one or more models as components string them together. You set their inputs and you say go when you say go the run is launched on the supercomputer behind the scenes. So this was cloud computing before cloud computing was a thing. We didn't exactly invent cloud computing. We didn't know we were doing it but we were doing it. And we're still doing it. So another thing we learned though from by by looking at the community and studying how you all work we learned that many of us need including me need for many of the things we do we need a way to get our get our hands dirty so to speak with code. We need to be able to write code that gives us the level of control that we can do the kinds of problems we want to do whether to bring in the right sorts of data managing our models and so on. And so to try and meet that need we've developed a programmatic version of our coupling framework called the Python modeling tool or pi mt. So this is essentially a little Python package that allows you to run and if you need to couple models as components through a Python interface notebook a web or a command line environment or what have you a script. There's an example coupling together a Irregular grid based landscape model with a grid based delta model right because they have standardized interface each of the models in this system looks and feels the same it has the same steering wheel and so on. So here's a list of the model components that currently exist in the system the list is going to keep growing we hope it's about a dozen of them now roughly color coded according to Themes so there's some cryosphere related one some fluvial one some coastal and marine ones some hill slopey sorts of things and so on. I'll highlight one thing to make a There's one point I want to make about this this easy simple snow model is most recent model added the snow model from environment Canada. Dr. Kong Wong did the engineering to create a BMI around this model. And he was able to successfully take a model written in a fairly old flavor of Fortran and get it to be a component operating in a Python environment. So that's that's an achievement there's a little bit of output from the simple snow model. So it turns out we now have some of you have heard of or have worked with land lab land lab components can now be easily translated into pi mt components if you want to do that here are some of the land lab components around. 20 or so of them total I think at the moment. So that's pi mt and there's going to be a clinic on that I think tomorrow on the educational front arena and gang have been working hard to try to translate some of our lab exercises into pi mt so you can look for those in the pi mt. Sort of learning collection if you look at for examples on the pi mt website you'll find notebooks like this that will walk a group of students through some exercises using one of these models. Speaking of land lab on development continues on land lab so this is a team with that started out within the terrestrial group. Creating a basically a get a Python language environment for being able to efficiently assemble models from component parts so you can make a grid regular or irregular or hexagonal or whatever you like you can populate with data. You can create or pick up and use interchangeable process components. Here's a picture of a few of the applications that have been written and published with land lab in the last couple of years. On the teaching front, Nicole Gasperini has produced a wonderful set of teaching tools for land lab including exercises, both in geomorphology and surface water hydrology so you can find those through the land lab website. They're set up and they've been they're now the veterans of several different class exercise. Finally on the educational front we've introduced to some of you know a set of webinars so we have now in the library about six webinars they're online so if you missed it when it was live you can go and watch the recording. You can even hit the speed up button so you can hear you know you can hear Mark talking at double speed. His voice doesn't change it's a miracle of modern technology. So if anybody has ideas about webinars that you would like to give or ideas about webinars you'd like to see let us know and and we'll put it on the schedule so we're we're going to continue this series again in the fall. Okay, so that was a quick tour of some of the resources I want to just tell you one quick story before wrap up here and it's a story about why modular standardized modeling software is a good thing. Well I think you everyone gets the idea that if you want a couple models together to look at crossing boundaries and interface problems. Yeah you need to standardize a couple of what if you're doing something simpler. Here's a little story with thanks to Simon Kubler for contributing this he didn't know he was contributing it contributed it. So Simon is a visiting scientist from the University of Munich and he's gotten very interested in lakes and how lakes in tectonically extensional regimes like East Africa and Oregon which is where this picture is from Simon. How the rise and fall the lake levels would have influenced animal migration patterns in these rugged landscapes and how those animal migration patterns might in turn have influence where prehistoric humans went to hunt them. So it has archaeological implications sort of an interesting problem. But here's the thing if you're going to worry about paleo lake elevations and if differences of a meter or two matter because there's very low topography between lakes. Then you have to think about isostasis right DK Gilbert taught us this in the Bonneville Basin back in the 19th century. So gosh we really ought to do an isostatic calculation but what a pain in the neck that would be right if you've ever written in isostatic 2d flexure code Andy. This is a lot of work right I mean it's you have to think about an elastic sheet and you have to think about Kelvin vessel functions need to set up a grid and you need to debug it and then you need to debug the bugs that you introduced by debugging it right. You know where I'm talking about but here we had a component two components in fact we had our pick that already did 2d elastic flexure with a standardized interface. So instead of having to spend months writing this and debugging it and so on in a few hours we had a little application that would take that component and give it a dm and a lake level and it would find the flexural response to the load of that lake and so end up with pictures like this that show this area with the red color being the flexural rebound once the water is evaporated. And all the same so so plug and play in addition to letting you couple models it turns what starts out as would be months of work into hours of work and that's a good thing that lets us do I mean think about if your papers took hours to write instead of months. That would be we'd all get tenure in six months. Okay. All right so last couple things I want to say that our awards program we're going to have the awards banquet on Wednesday night we're going to be honoring a few people one of them is our student. So Vitsky student modeler award to Letty Roach from the National Institute of Water and Atmospheric Research in New Zealand. She's being honored for developing modeling software for ice flows and doing a good job of that. She won't actually be able to join us but we will honor her and she'll come to the meeting next year. We're going to be honoring the winner of the poster competition. So that winner is a mystery person because we haven't had our poster session yet so but you all have to vote so this afternoon and tomorrow afternoon you're going to see some amazing posters and you can't vote for all of them you can only vote for one. You can vote early but you can't vote often and so make sure to vote and we'll announce the winner tomorrow night at the bank. Our other award of course is the is the Lifetime Achievement Award and I'm thrilled that we're going to be honoring Jaya Svitsky as our Lifetime Award winner. I'm going to embarrass Jaya and ask her to raise her hand. Hello. So you'll learn a little bit more about Jaya's achievements and hear from some folks who, rumor has it, may be doing a little light roasting. Hopefully not. Hopefully fairly light. We'll see. Okay so here's the plan of the next three days. We're a little bit behind schedule but that's okay. We're going to make this up en route as they say. We're going to start with a set of keynote talks next. We'll have then breakout groups, clinics in the afternoon, a few more keynote talks and then posters and we'll be sort of alternating between those activities through the next three days. Let me say quickly about breakout groups. So today you're going to meet with CSDMS Focus and Research Groups and the breakouts will be led by the chairs of those groups. Tomorrow and Thursday they're going to be topic-based breakout groups and so you can sign up for the topics that you find most intriguing out of the front tables. What we're really looking for in these Wednesday and Thursday breakouts is your ideas about potential proof of concept application. How can we use the concepts and the technologies and the ideas that CSDMS One and Two created to create demonstration projects that essentially the technology helps turn high-hanging fruit into low-hanging fruit. So we're looking for your ideas in that. I think these should be fun conversations. Okay, I think that's all I have to say. Lynn, am I forgetting to say anything? Anybody integration facility people? Are there any other logistics we need to cover?