 closely affiliated with. And I see a lot of CGD, MQ, and RAL here today. And maybe let's have the second question, which is, which option best describes your career stage? Is that directed to anybody specific? Whoever's here can answer. Yeah, we just want to get an understanding of who our audience is, and that would make it easier for the speakers to know who they're speaking with. And then for us to know who's engaged in the conversations. Excellent, thank you. Sure. Well, I see mostly postdocs, but we do have a range of early career scientists here. So I would just want to make sure that David, John Gagnier, and Julie Dumuth, we can hear you if you could just say hi. Okay, we hear you. That's great. And presumably you can hear me too if you can hear DJ. Yes, oh, you're in the room. That's great. I didn't know that you were there. Awesome. I just got here like 30 seconds ago. That's great. That's great. Well, I think that we can start now. So welcome everyone to the sixth session of the NCAR-UCP Postdoc Grand Writing and Management Series. And this session is going to be focused on convergent science. And we have three speakers that we're really excited about. Julie Dumuth, Project Scientist 3 from MQB at NCAR, Paulette Blanchard, many roles she plays I've heard, but one of them that we're familiar with is the PI at NCAR Rising Voices Changing Coast Grant. And she is here from Haskell Indian Nations University and David John Gagnier, Machine Learning Scientist 2 from CISL and RAL at NCAR. Can I have the next slide, Mariana? Thank you. So this series is organized by the Postdoctoral Fellows Professional Development Committee. And today the moderators are myself, Sudha. And then we have Shema Shams, Mariana Cainz, Diamond Tachera, and Anna Del Moral-Mendes. And all of us are postdocs here at NCAR. And I'll give a chance for our speakers to introduce themselves better, but I just wanted to give you the series schedule. It was a six session series. And as you can see we're on the final and sixth session, which focuses on convergent science. If you've missed any of the previous sessions, we do have the recordings and we will share them with you. And the session structure, which is on the next slide, we are going to talk a little bit about the topics and goals and what we would hope to touch on in this session. And then we'll have short presentations from our speakers talking about the work that they do related to convergent science. And then we will have a roundtable discussion, moderated by Diamond, Mariana, and Shema. And then we'll move on to Slido Questions, moderated by Anna, which will be questions from the audience to the speakers. And we'll finish with closing thoughts from the speakers and also the series organizers. Next slide, Mariana. Thank you. So in this session, we're hoping to at least understand what is convergent science, because it is a relatively new area to most of us at least. And some of us don't really know what it means, me including myself. So I'm really excited to understand what really convergent science means. I've heard a lot of descriptions. And then also to talk about how we can conduct collaborative writing and ideation when we are all coming from different fields and making sure everyone's ideas and science and skills are valued the same. And then we're going to focus on convergent science to include indigenous knowledge and communities and end with some tips on how to get involved in convergent science as an early career scientist, because we might not know how to network, how to get involved in those projects, and it would be nice to hear from the speakers who have been involved in these projects and how to get involved. And without taking more time, I'd like to ask our first speaker, Julie DeMuz, to introduce herself and talk about her science related to convergent science. Julie, you're welcome. Thank you. Just checking again that you can hear me. Okay, great. So I might have quasi-misinterpreted a few things. I will introduce myself, but I've got a couple slides that have definitions that we can quickly go into and maybe cover. I can just give sort of my perspective on this and then if Paulette and DJ have anything they want to say. And then I figured I would talk very briefly about one of my projects, but I really would love there to be more discussion because I feel like talking at you all isn't the best way to learn about this stuff. So I think, well, I guess I will introduce myself first. So like Suda said, I'm a project scientist in MQ, although actually I started in what was called ISI at the time, which I don't even remember what that stands for. I honestly don't even remember what that stands for. And then also had a joint appointment with Rao. My background is in both atmospheric science where I did my undergrad and my master's degree in atmospheric science. So I was one of those people who grew up completely in love with the weather. And then while I was doing my master's work, I got really interested in sort of the human component of things. And so I kind of had a career crisis at that point in time, an existential crisis. I was just reflecting on this the other day. It's funny how these things reflect on this, but it's true. So I thought I would share that. And it's like, what do I want to do? I don't really know. I ended up going to work for the National Academies. They're board on atmospheric sciences and climate for a few years. And then that was actually how I met Eve Gruntfest, who is somebody who, thank you, Scott. I'm like, I knew society was in there somewhere. I got up on the eye. What does this stand for? So I met Eve Gruntfest, which is somebody who had been doing research kind of in this arena looking at kind of human behavior in response to the big Thompson flood. And it was really through Eve that I expressed my interest in kind of social science, even though I didn't really know what it stood for at the time or really what it all encompassed. And ended up coming to NCAR and helping stand up a series of workshops called WESIS, Weather and Society Integrated Studies. I'm happy to talk about that. But while I was doing that work, I had the opportunity to work with Rebecca Morris and Jeff Lazo. And that was when I realized I was really interested enough in this stuff to actually think about getting a PhD. And so I went back and I got a PhD in communication while I was working at NCAR and working kind of alongside them. And so really my research kind of sits at that interface of risk communication, risk perception and decision making, but also really tied to kind of issues of predictability and like predictive capabilities. So, Mariana, if you're, I think Mariana, you're running the slides. If you can go to the next slide. So I just wanted to put up some definitions because I think it's great, Suda, that you said that you're not really sure because this is a term that I think for some of us we use a lot knowing what we mean by it. But it's even interesting to me to encounter people who are like, I don't really know what you mean. I was talking to somebody very high up in weather service this summer and he thought I just invented the term. He's like, Oh, that's a nice idea. I was like, I could tell you what it means. And so I in particular really rely on the National Science Foundation definition, which is that if we have kind of these really challenging research problems, kind of these really wicked or particularly vexing issues that are kind of require deep scientific questions or really are often connected to these pressing societal needs that address these, we really require this deep integration across different disciplines. And kind of I love this part here too, that if we're doing this deep integration, we really are kind of blending our knowledge, right? We're bringing our theories, our methods to the table. But we're also thinking a lot about refining our research questions or, you know, creating new kinds of research questions or approaches or things that we're really doing that are different from if we were doing it kind of solely disciplinarily on our own. And that last part I also wanted to highlight that, you know, these new frameworks or paradigms or even new fields can emerge from when you're really doing convergent science research. But this is one definition. So Marianne, if you could go to the next slide. I wanted to also share this because this is a report that came out, I think last year, yeah, 2021 on the future of Earth System Science for NSF. And this was an Academy's report. And I just wanted to highlight in there this box, which actually has definitions of what disciplinary, interdisciplinary, transdisciplinary and convergent science are, and you can see my highlights there. And for the most part, it's pretty consistent. I know it's tough to read, but mostly I just wanted to make sure people have this reference. But the most part is consistent with NSF's definition. The one thing that you'll see a little bit in here and transdisciplinary and conversions is they also talk about maybe working with communities. So it's not necessarily just researchers who are coming to the table. I think both are valid. And then maybe one more slide, Mariana. I like this, which is Lori Peake. She has led this big effort at the Natural Hazard Center called Converge. And they put out a paper in 2020 where she was taking the ideas and particularly applying them to hazards and disasters. So she's got a definition there. It's kind of small again, for the most part, pretty consistent. But I also really like this representation. I have a colleague who doesn't love this because she thinks it's really simplified. But for me, I just feel like it nicely characterizes disciplinary, multi-disciplinary, where you can have multiple disciplines at the table, but they're really working in parallel, hardly ever interacting. Interdisciplinary, where they're starting to blend a little bit more. But that kind of transdisciplinary or convergent science where you're really getting more of that kind of deep integration idea. I will give credit to Rebecca Morse who says she would actually like to see that transdisciplinary still kind of ordered. The colors are all in the same order, like blended even more, like a lot messier than that. Because I think that's really what transdisciplinary looks like. So this is one more definition or representation. And then, Marianne, if you can go to one more slide, and then I promise I'll talk about one project. I also wanted to share, this is a paper that Rebecca Morse, Heather Lazarus, and I wrote and put in risk analysis a few years ago, where we were talking about, we call it interdisciplinary, but really we were talking more about convergence and transdisciplinary. And what does it really look like at the working level? Because there's a lot of conversations around structurally kind of at high levels. How do we help people be able to succeed in an organization like NCAR? Or how does an individual do this? But we really wanted to dig into what does this look like on a day-to-day basis when you're working on projects. And so we wrote about what we think are different strategies for doing this. And I wanted to, and there's like, these are basically the highlights of the sections, like how we do come together. We maybe start with this overarching research question, which is what brings people to the table. But when you're all there in the room, that's when you really start refining and maybe articulating your research questions or changing them in different ways. We talked about different mechanisms, and then also like really what this looks like when we're really interconnecting kind of the knowledge and ideas. And we had to synthesize, I think, in a couple of different tables, what signs of successful interdisciplinary work looks like, and then what are some of the practices. So I just wanted to highlight this for people who are interested in that. Okay, so those are a few resources, and I would love to hear from Paula and DJ about if they have other things that they would like to bring to the table in terms of definitions. But just to very briefly introduce one project, well, two projects. So this is one project I'm not going to talk about today, but I am getting an MQ seminar. So interestingly, and we can get into this if we want to, I would probably call this convergence, or at least I think it's on the path to convergence. I don't really want to get into like policing what is and isn't. But I think there are good conversations to be had about how do we decide, again, what are indicators of convergent projects. I would not say I have the answer. I'd love this to be a conversation. But I am going to talk about this project later this afternoon. But the one that I thought I would just kind of give an example of today, if you can go to the next slide, Mariana, excuse me, is a project that began for me in 2015 when there were model developers over in Boulder at the Global Systems Lab who were really interested in developing new convection allowing model ensemble guidance for forecasters. And they wanted to know what is the kind of information that these forecasters actually need? What would be useful? What is actually usable for them? And so they actually, because they knew about Rebecca and me, asked us to participate in this project. It was an R2O-funded grant. And under the US Weather Research Program, that's what USWRP stands for there. And I tried to take, I think with Anna's suggestions about like, what was the objective high level? What's the SOA and what was our role? So that is the objective. I think the SOA really had a couple of components. I mean, first and foremost, that if we could figure out what forecasters need, we could actually design stuff that is more helpful for them to do their jobs, which then functionally should help society. But I also think part of the SOA that came out of this, not something I thought about from the outset, is really changing how we approach these kinds of questions, not to be just disciplinary, right? Not just to be the model developers in the room. And so my role on this was as a co-PI, but I was really the social science lead who did all this research with forecasters to learn about what guidance they look at, how they might make sense of information. What are some of their key like challenges when they're issuing forecasts for warnings? And then how can we derive new guidance or develop new probabilistic guidance for them? But that was a very high level effort. And I wouldn't even have called that convergent, even though so much of what we learned fed back into the developer space. But what it did is it really, and I would actually say for some convergent projects, I'm not sure you can actually achieve convergence in one research grant. This was again, something we can talk about. But it then led to this follow on grant that was funded also by NOAA under what's called the DEDI program. And I put these here in case people are interested. And this is a proposal writing workshop to think about where you could propose to JTTI stands for Joint Tech Transfer Initiative. And it is also kind of an R2O focused program out of NOAA. And this was then a grant that I led. And again, the objective here was kind of similar, but a little bit more focused based off of the kinds of things that we were starting to learn in that first proposal, but much more focused around deriving and verifying and visualizing different probabilistic timing guidance. So really predicting the onset or the cessation of different winter and fire weather parameters, again, for forecasters. The so what is really still the same, but in this case now my role was the PI. And so I was the one who was leading the team of the model developers and the computer scientists. I should also say that the derivation of this information, but then also really programming it into an interface, which is impossible to see there. I'm sorry about that. So that forecasters could actually regularly use this and interrogate this information was a key component. So we had computer scientists who were involved in other social scientists. And then my final slide, and I promise I'll stop talking, is I actually think that this effort then also built a bit of a framework that we brought into the AI2ES framework. Although here instead of thinking about numerical weather prediction guidance or ensemble based guidance, we're taking this and thinking about it in terms of AI and machine learning. And the kind of work that DJ and so many others are doing to derive new AI machine learning output for forecasters. And the final things I'll say about this, because I think it's important for me to articulate like why I think these projects or the suite of projects are convergent. Again, I think it is we are tackling this compelling problem of how do we leverage the state of the art science in ways that are really useful for forecasters. Obviously required this deep integration among the different disciplines. The JETI project also had practitioners involved like in a research level in a way that I didn't talk about, but I'd be happy to. What I loved about this is that we kind of had this iterative driving, like the first project was more driven by the meteorologists with me involved. And by the time we got to the second one, then I was driving and involving the meteorologists. So I think now actually for AI2ES, we're kind of like all driving. You can have an AI car where we're all driving but I'm sure that's feasible. And I think kind of akin to what I mentioned with that paper, Rebecca and Heather and I wrote, we had these really broad research questions initially. We really had to refine them over time. But I would also say I think out of all of this, there has been this new framework that has been developed where NWP guidance and AI machine learning is I think more thought of as a form of risk information. So therefore it requires, you know, really being co-developed and co-produced with social scientists at the table. And this is true from the outset. And I think kind of structurally things are changing to just have this be the way of being versus kind of a novel, you know, having social scientists involved maybe a little bit early on. I think AI2ES is an example of this. But I can also say that the Global Systems Lab here in Boulder has really embraced this. And now they're standing up an entire social science program to help them do this a little bit more from the outset. And since my voice has run out, it's clearly a time to stop talking. Oh yeah, one more additional resource. We can come back to this if folks are interested. But I really wanted to make sure you all have these resources that you could draw on later if you wanted to. All right, I'll be done. That introduction, Julie, and all the resources you shared with us. And Paulette, welcome. And if you could introduce yourself as well. Hi, I'm Paulette Blanchard. I am a citizen of the Absentee Shawnee Tribe of Oklahoma in Kickapoo descent, as well as French and Irish and Scottish and in Dutch. But I'm a citizen of the Absentee Shawnee Tribe of Oklahoma. I got my associates at Seminole State College, my bachelor's at Haskell Indian Nations University at Tribal College. I was recruited to the University of Oklahoma and did my master's work there in geography. So I have a bachelor's in Indigenous and American Indian Studies and master's in geography with specializing in Indigenous geography, Native Americans and climate change. And then my PhD was in Indigenous geography from the University of Kansas. I've kind of always had a little bit of a foot in what is being called convergent science now. In my undergraduate, I did a summer internship where we were taught how to do scientific writing a little bit and help develop a research question and do some research and develop a paper out of it. And mine was looking at how Oklahoma tribes were dealing with climate change. And in the process of trying to find out what tribes thought about climate change or what tribes thought about the weather in general, there was no doubt. I couldn't find anything nobody had ever interviewed the tribes in Oklahoma really. And even during the there was no voice during the during the Dust Bowl era, there was no recordings or interviews of Indian tribes in Oklahoma of how they experienced the Dust Bowl, how they experienced and deal with tornadoes and floods and drought. So my summer internship question that I asked of how tribes in Oklahoma were dealing with climate change basically came my answer was that there's not enough data that the research needed to be done. So I got this crazy idea as an undergraduate at a tribal college to do an intertribal workshop. And so I talked with this Southern Climate Impact Planning Program leadership in Norman, Oklahoma at the National Weather Center and the National Weather Center. And I convinced them to put on a conference or a meeting that if they put the meeting on and hosted it, I would get tribes there. So I was able to get 22 out of 39 tribes to participate in this intertribal meeting on climate variability and change. So that got me that got me not only recruited to the University of Oklahoma to do it again for the South Central Climate Science Center's regions and their tribes, but it also got me cited in the National Climate Assessment. And so moving forward into my next research, I interviewed tribes about climate and I asked them what does climate change mean to you? Are you you or your tribe doing anything about it? If so what if not why not basic questions? And I had this incredibly rich dialogue with people from different careers, different farmers, fishermen, teachers, all kinds of different professions came together from different tribes to talk about weather and climate. I had a white male professor tell me the Indians didn't know the difference between weather and climate and so I set out to prove that wrong. But anyways we brought together all these different people from different disciplines, different cultures, different backgrounds, different understandings of science and knowledge and education levels. Had professional scientists from Chickasaw Nation that when asked what does climate mean to you they're like oh precipitation and temperature, the basic generic answer. Whereas some of the farmers and some of the medicine people had this incredibly detailed description of plants and animals invasive species, weather pattern shifts and all of these really rich conversations. So during that same process I met Heather Lazarus and got to meet Julie Maldonado and many other people and they started this program called Rising Voices which was similar to what I was already part of with Haskell which was the indigenous people's climate change working group and Rising Voices became a place where people from different cultures, native cultures, different disciplines, different scientific levels, different education levels all came together to talk about you know the challenges that climate creates and the different ways that people are responding to it, the different sectors that people have interest in. So we had community members, we had teenagers, we had scientists that were modelers and scientists that were social scientists and geographers and all of these people come together and and having these incredibly rich conversations about climate and and so on and so forth. So as I went through my PhD program I applied to be a UCAR fellow and was awarded the Diversity Equity Inclusion Fellowship the same one Diamond was in later on the year after me and it just grew this this pattern of doing science from a multicultural, multidisciplinary, multi education level, multi-gendered place of doing, of talking about climate, talking about weather, extremes and variability, talking about disaster and risk and working with communities to help develop their plans of adaptation and response. So that relationship with Rising Voices over many years culminated in the application of a grant for coasts and people through NSF and Haskell Indian Nations was the lead PI in Dr. Wildcat and I'm one of the co-PIs and there are several others and we put together this grand idea of doing convergent science working with communities in one of four hub sites, Alaska, Hawaii, Puerto Rico or Louisiana and talking with the indigenous communities or the native communities in those spaces about how the coasts are changing about how fishing and farming and subsistence living was being affected by weather extremes and variability and climate shifts and so we're in the middle of this ginormous $20 million five-year project or just at the beginning of it I should say and we have brought together like I said probably 88 different people from different disciplines everything from archaeology, anthropology, all the social you know a few social sciences geographers, indigenous geographers both physical and human geographers and modelers from CI specialists and I mean I just can't even name off all the amazing disciplines that have come together and are trying to work you know together and develop language that is consistent you know because each of us bring our own languages from our own disciplines our own key words our own specialties and we're trying to communicate across across all of these differences and build and build something that is consistent which is a challenge so I'm gonna just admit that it's a little messy it's and it sometimes can feel very overwhelming because there's so many pieces moving um but we're doing it we're working towards I'm not I'm not gonna talk a whole bunch a whole bunch a whole lot longer um because I want to get to the other speakers and all but um it's possible it's happening we're doing it um one to me one of the important components for for my work is that we're working across cultures we're not just working across disciplines we're working across very different cultures very place-based place specific cultures who have unique challenges and so we're going to be able to compare and contrast these different places that all have coastline you know what are the things that are similar they have what are the problems that are different how can solutions in one area possibly be a solution to another area or or a problem be a problem in another area so all of these people have come together to try to develop a new framework to develop a new and consistent way of of doing science including indigenous science as a legitimate science instead of a token um tokenized or um what do they call it a um anecdotal knowledge you know indigenous knowledge is deep it's place-based it's it's generational it's it's communicated in different ways in different communities like one of my favorite stories that I love to talk about is how I learned how Hawaiians talk about the Hawaiian cycle or one of the ways that they've talked about the Hawaiian cycle the the hydrology the water cycle in their culture in their community is um very much in in uh in a chant a hula and the process is is communicated beautifully and effortlessly and passed on through generations through these songs and these dances well many indigenous communities have had if not still have some of those those knowledges and those methods and those um ways of knowing doing and being so it's really exciting to work not only across the disciplines but across cultures so that is something that I think um you need to keep in mind as you move forward with um your interest in convergent sciences is creating space for other ways of knowing doing and being and not trying to deconstruct it with comparing it to western systems of knowledge and methods and methodologies but to recognize that that is western the the mainstream system that we are dependent on is just one way and that if we create space and and respect and be responsible to um the science and accountable to be reciprocal with the communities we're working with you know we're not just taking from them to be um to recognize the relatedness peoples have indigenous peoples have with place they right they identify their bodies being of the soil of the water of the sky you know the air um that there's that physical interconnection with place um of relatedness that the work has to be relevant to them in their communities um that the uh there's a relationship that has to be built and maintained over time so um you know you're not just dropping in taking and then never coming back but you're building relationships over time and maintaining those and that um the redistribution of the information how are we giving back to the communities are just seven basic ours that I've used in my research to work across disciplines across cultures so with that I will say thank you and I look forward to questions thank you so much Paulette for talking about your career path in uh relationship with convergent science and also bringing up the fact that cross culture is also important when talking with convergent science and I think now we have one speaker left and that's David John Garnier if you could introduce yourself and also talk about your background in relation with convergent science that would be great welcome John uh thank you for the introduction uh today uh um let me see if I'm trying to say which way uh I will aim just to be right at you but the difference here uh so uh um currently I'm a kind of a machine learning scientist at NCAR uh but getting to this point was a very long and and somewhat twisted road that I've kind of I think there were a lot of seeds of convergence along the way that that kind of eventually blossomed into like all these big projects like AA2ES uh this goes I can kind of trace it back to about 2007 I was a freshman at University of Oklahoma and got um initially rejected from the there's a research experience for undergraduate site at at the National Weather Center there uh and and I didn't get into I applied for but didn't get into that program but then got instead selected for a another hurry being held kind of in conjunction with it uh by a relatively new CS professor I know you named Dr. Amy McGovern who's like hey do you want to do some machine learning on this these simulations of super cell storms I have and I was like okay I don't know anything about machine learning or AI or at that point but it was it was kind of game to try it at like I growing up being really really interested in weather and interested in computers so it seemed like a combination of things to to try out um so I spent this the I think one of the the cool things about the program because I even with this REI basically was still embedded with the other weather center or people and for those who aren't familiar with the National Weather Center Paula mentioned it in her uh talk the the underlying idea behind it was that wasn't some ways to foster we are now calling convergence uh originally all the weather the government and the academic weather people were in different parts of Norman Oklahoma so like all the NOAA people were on the north side of town and all the all the OU school meteorology people were on campus and so they got this idea to build a a single building where everyone can can be in the same building and closer together so that they'd be more likely to actually bump into each other and come up with new projects and work or things out together and so this was like just opened in like kind of the year before and people were kind of pouring into the building getting settled so so I got the I think some of the key like some of the seeds there were one there were a number of people doing machine learning whether they're kind of the forefront of that uh like Amy as well as uh Valiapalakshmanan and time like Mike Richmond and uh so there's actually there's like a core of of AI weather people uh there there's also kind of a burgeoning social science development because I think I think Eve Eve was at OU at the time and and there was like I've heard about was is uh like through originally through the aria and there's some talk about social science and and just so I can as part of the aria you want in addition to kind of doing my project on like storm classification I also got to spend some time like in working with like talking with a lot of the like weather forecasters and researchers and uh they were very happy to share their stories and and kind of answer lots of questions kind of like these kinds of round tables and so I got to get lots of advice and and pick their brains and kind of give you some ideas of all these different pathways the thing things to do and where how to intersect stuff uh I continue doing uh kind of some like research with with Amy uh like pretty much since that point uh but but kind of in in different capacities and and a lot of the projects were kind of we had like initially sort of a few like meteorologist team members as well as like and just more on the computer science side now I was taking meteorology classes and computer science classes and kind of building this first intersection of like kind of meteorology and computer science sort of these dual like trying to build depth in both of them to to be able to kind of tackle this joint problem of it's like how do you get make an AI do a weather forecast um and then I had the opportunity to stay at OU for for graduate school uh and I guess also during my undergrad one of the other things I did that I think planted a lot of seeds was was attending the AMS American Meteorological Society annual meeting and meeting the the members of the the AMS AI committee who were kind of coming from a number of different sectors the the conference at the time was really tiny there's only a few people so there's a lot of opportunity to talk with like kind of these all these different experts in different areas a lot of them were working like not just on AI for the sake of AI but they were working with like AI for aircraft turbulence or renewable energy or uh like like kind of starting to embed themselves on other stakeholder groups and try to understand like the ones that did have some make some inroads and success for ones that actually understood the needs of their like the partners they were working with and tailored their AI systems to to to what what like the data and the problems that that that were being faced um and kind of through this I like started started and kind of still have a kind of core into your storms but but through some happenstance I was basically asked to participate in of AI renewable when energy forecasting contest and and then like wanted on kind of like like happened to not put too much effort in but then there's a company there that was like hey we need your help doing machine learning for wind forecasting so then I got kind of this consulting gig with a company in in albany new york called meso incorporated that were like we want to use more more modern machine learning in our in our web tooling so like float there had an interview and then kind of worked remotely for them for for a few years off and on and they basically talked to them about decision trees and random forests and like I think these methods could work well for your your forecasting needs uh and through talking to them I also kind of understood their needs and uh there's also around this time I through the eye committee I had gotten to know so hopped uh who's uh current I guess going to be stepping down as associate director of ral but but had had was at pennstay and then moved to mcar to to work in ral and had developed it like a big wind energy project with excel energy to basically prove their wind forecast that they could reduce the amount of reserves they needed for their their power and and like as part of that like eventually I was able to visit mcar in like 2014 and got to be involved in some of the like meetings that that that suit would organize between like the power companies and the the mcar researchers on like wind and solar forecasting and kind of the needs of the utilities versus the like what we can do on this like on the science side so so this was really helpful and and um but like making sure like the problem we're trying to solve is the problem that they need us to solve rather than like what's convenient uh and sometimes there's a lot of inconvenient aspects so like a lot of power data is is very proprietary and hidden behind NDAs and other stuff so so you have you you really can't just like go out and grab data off the internet you have to you really have to go work with work with these companies and and find the way to build trust and relationships with them and I got to spend a year at mcar through the asp graduate visitor program working on kind of solar energy but also some working on severe weather stuff as part of my dissertation I think around the time I also met Julie and Rebecca I think it was originally at the SLS conference in 2014 that they were giving some I think posters on some of the twitter work y'all were doing and found it really fascinating and also I also know through you and Kim Cloco like while she was still like a PhD student and doing her initial research on like the April 27th 2011 tornado outbreak and so kind of like I understand like kind of through her and and through some of the was is group like like kind of being adjacent to some of those conversations understanding just kind of the need for better communication of forecasts and how how much value it like we like it can't just make a more accurate forecast it needs to be able to get to the public in a way that they can make use of it and make and hopefully make better decisions and work through all of like the barriers along the way and I was also seeing kind of similar problems in the AI side so I built a cool AI algorithm and show like oh it was great verification score but then what do you do with that like forecast like the meteorologists are still really skeptical like how how this actually going to be useful so I was able to like in the process of developing some of the like hail machine learning algorithms I got to be involved with the NOAA hazards for their test bed kind of informally I think I was at and the I think it was like the OU party at AMS and happened to talk to Jimmy Korea who's like hey yeah I heard you're doing cool work you should you should reach out about being in the test bed and then I reached out and kind of was not like formally in there but basically got to hang out and and run my algorithm occasionally show it on the on the monitor and get some feedback from people and and and that sort of led to to kind of understanding some of the needs of operations and talking to like SPC forecasters and weather service forecasters and tailoring like what I was showing not just to like not to show every single thing that could come out of the machinery model but sort of tailoring it towards what kind of products that that they're like looks like what they're already using so they can have a better comparison and gain the context of or what we described now went from AI to that is like the extra contextual information they need to make uh like build their own conceptual model at the forecast then yeah I guess I finished my PhD had a number of opportunities that then decided to come to got an ASP postdoc in the car and my my in-car appointment was interesting because like usually when you get an ASP put you're like you're based in like a single lab and that's your kind of home I was able to finagle a uh let's throw another kind of thing I didn't talk to about as much but the I was involved with the climate informatics workshop which is a kind of small conference held at NCOR on like climate and machine learning and AI and statistics all trying to come together and like build its own convergent community. I had gotten to know Doug Nitschke through that as well as I already knew Sue and so I proposed a joint postdoc where I'd be between Rao and Sizzle and and kind of work on deep learning and solar energy and looking at some uncertainties and how we can kind of plug all these things together. My initial path and that didn't go the way I expected it so I went back to Hale but then was able to look at interpretable deep learning and put together some of the pieces that to make something that works pretty well and get some a couple interesting papers out that were sort of the foundations for some of the proposals for like AI2ES and a number of other like NOAA projects that came out out of that and during my postdoc at NCOR I got to like meet a lot more people in both the Nellie and like like in the Rao community and in the Sizzle community including like some of the statisticians that worked there and got involved with that. There was a there was a math institute called SAMHC in North Carolina that sponsored like a year of climate in AI and statistics so I did I worked with some stochastic parameterization experts including another postdoc at NCOR Hannah Christensen to basically try out a machine learning stochastic parameterization and kind of work through all the challenges of that but also learning kind of the language of the stochastics community and the statistics community and the applied math community and kind of getting a sense like everyone calls things different stuff so that was a like kind of enriched my experience and kind of learning about some other issues in terms of integrating like AI into like a client into a client model but but also including stochastic stuff into AI which is usually run more deterministically and all that then had an opportunity at where whereas my postdoc was ending I was offered a like basically my person would be my boss at it like I guess Sue and Rich Loft ended up putting together a position for me as a this new machine learning scientist so like kind of coming up with a new classification to to emphasize some of the machine learning components because they recognize that this is an important area we need to invest in it and this is kind of the start of where AI like people recognizing that deep learning was kind of a big thing and that we need to pay attention to it so I was able to stay here and kind of start filling out a group that where we were essentially trying to reach across NCAR and build different connections to different labs and got some seed money from the directorate to to kind of start some of these projects on like parameterization in particular and and we've been able to kind of and those projects have sort of evolved into lots of different some that were continuing in different forms under different funding streams but like enabling to kind of embed in a lot of different communities including that with the kind of social science community through AI2ES and I also have a couple slides kind of a long journey but I did have a couple slides. Sorry to interrupt you David John would it be possible to go back to the slides during the roundtable if the questions come up? Yeah I think that'd be fine. I'm sorry to stop you. It's an exciting conversation and we're realizing we didn't give it enough time we should have had a two-hour session because it's so enjoyable you all come from such different backgrounds and have such interesting experiences. Yeah I'd be happy to kind of jump into the question. Whenever you get to a point to answer a question that you know you it's on the slide let us know and we will share that slide. Okay thank you I appreciate that so I'm going to hand it over to Mariana Diamond and Shima to moderate the roundtable Mariana. Yeah thank you so first I want to say that we have some prepared questions but if anyone from the audience has questions they want to ask as well they're more than welcome to and you can also add them to the slide link so we've heard from a broad range of background both in what you do where you do it and how you do it so I was wondering across the three speakers if you could talk about how it is that you go about writing or we're ideating in a collaborative space when you might have different ideas or different perspectives than the other people that you're working with and I'll open it up to whoever wants to talk first. Mariana one clarification question do you mean this writing and ideating when you're writing a proposal is that what you're that's what I mean yeah so when it comes to the proposal aspect yeah. I guess I can feel for it. I can talk a little bit so one of the bit I think AI2ES was probably the biggest experience in terms of trying to ideate a collaborative proposal with bringing together a lot of different bodies and I think some of the key parts of it was like going into it we are there was already a group of us that had known each other for quite a while um so so we already had kind of a working relation like informal working relationship built up so that that sort of helped that we understood our language but then there's still a lot of people that were kind of newer to the to the process but to like help integrate them I think some of the key parts were like actually getting everyone in the room together like we we literally like rented a room at the at the Denver airport like the hotel this is right before COVID and and and like kind of hand murdered out over day like presented each of our different like backgrounds and what they can contribute and then kind of I think initially there's sort of like a big exploratory period where you have to kind of get everyone's ideas out there and then be respectful and kind of like an improper sort of yes and as much as you can initially uh to to like build on all these pieces uh so that's like kind of the ideation phase so so get all the ideas out there kind of explain them find find that language and figure out where like there's a lot of points where there's a lot of common terms that mean different things to different people so so making sure you're clear in your definitions of those terms is really important uh then there's the ruthless cutting of ideas to get into that proposal limit uh and and that's that's kind of where we're like some of the weaker ideas make you have to like be willing to not be too in love with a single idea and kind of be willing to like kind of give and take and and figure out like we cool to do this but we don't have the time or scope but maybe we can do it in this this form and then also making sure you like if you want to be a have a serious commitment to like social science it makes bring in extra people like recently we've brought in Julie and Julie was like oh we like I'm not enough expertise on this we need to bring in this other Ann who we didn't really know but like it was a game for it so so we uh uh brought her in as well and and like made some amazing contributions to the proposal that if we hadn't had them at the table yeah I don't think it would necessarily would have gotten funded because I think like the social science aspect has been one of the kind kind of key like like I think key is to bring the whole thing together uh so so that like kind of being yourself from being open to to kind of changing course and giving and taking thank you Paula or Julie Paula you want to chime in and then go last sure sure um that's a that's a really good component is being fluid being able to adapt to be able to let go of something you might be clinging to to create space for other ideas um especially ones that maybe are outside of your comfort zone um for us also in our project it was um we were fortunate to be um working with community members from different indigenous communities for a period of time who had their own ideas of what they wanted or what their interests were for research in their region and um that was for us a key component is listening and implementing what the community wants um which you don't often get because often you know projects are created without the community engaged in the beginning in the planning process and that's something that is is critical when working with indigenous communities is recognizing not only the sovereignty of these sovereign these independent communities um both politically and socially but the um the engagement on the planning level on the in the very beginning and coming from where you know they are and what their needs were and listening to them and you know kind of setting all of our preconceived ideas and and plans aside to hear what their needs were to try to plan around that um so uh for working with you know indigenous communities specifically um it's really important that when you're planning something um expect it to take more time and which equates to extra money um and beings that there's a certain amount of reciprocity expected um and you're dealing with a nation to nation relationship um it's really it's important to understand how best to work with any community but being extra sensitive to the political status and the sovereign status of um indigenous communities uh and I think that if that kind of respect can be managed across to any community I think the project can only benefit um and I'll just I'll just leave that at that point so that next speaker can go ahead yeah I think one thing I want to emphasize that I think Paula and DJ have both said is time like you need the time to plan to do this um because it does take a while for the ideas to come together in some coherent way and it's also you know frankly a microcosm of what working on the research is going to be like because it takes time and a lot of interaction and if you don't invest in that so that it will actually come together in a coherent way the reviewers will see it or I have been involved in proposals where my work was maybe more tacked on and I thought well maybe eventually during the research we could figure out a way to connect but that was also an indicator of what the research was going to be like so we didn't you know it didn't really achieve some of what I thought had a lot of potential to be achieved um so I think time is a key component I think I've got a lot of different I one thing I forgot to mention is that I'm a project scientist so I write a lot of proposals and I've got a lot of experiences in a lot of different ways where I've either been a co-PI or a PI and I think I think AI2ES was a different experience in the sense that we were all kind of I think we did come together and have some ideas but we were all like actively writing I don't know how Amy managed that but like there were like 20 people in the document writing which and it came together well that if I were leading it would not have worked for me so I think it's helpful to know what does work for you um I think it was really interesting to be part of that and see how it is though like I think again it's a it's an indicator of the fact that the whole team works together so well and it's very like non-linear but dynamic and very like generative way for proposals that I have led I think I've been more of the person who's led most of the writing I will like ask other people to contribute parts that I just don't feel like I can fully represent that I kind of after doing a lot of the talking with everyone have the ideas in my mind and feel like it's on me to pull things together in a in a way that is really coherent so I don't think it just looks one way for every single project um but you know those I haven't done a 20 million dollar grant so I think if I were like cope and ai2s I think are really different kind of scales and I think if I were to lead something like that I don't know that I could be the one to like write all of the like be the one who manages most of that I think you do need a lot more people who are contributing their expertise um than just the you know the ways that I've done this in the past so I don't know if that kind of starts to answer questions or if there's follow-up questions kind of about that process but I do think time and getting together in the room talking through these ideas and knowing that I think this is the other thing that when you're writing things at the proposal phase you this is true for any proposal it's not just conversions like you have enough of what you want to do enough of the ideas in terms of what are the interesting research questions and how you're going to do it you don't really know what it's going to look like until you start doing it right like I just think this is how we do science so yeah chris I have a follow up one of my questions is like when you're writing the proposals how do you walk the line about being like concrete enough so that people like this is a good investment versus like we will converge like we want to have space to work on stuff together you know so like showing enough detail without structuring yourself into like different silos in some ways like how can you come together and like you have maybe a couple months and like this is how we're going to all meld together and you're like well how will you meld together how do you know that will work you mean like so how do you convince people of like the right yeah kind of balance there of giving yourself the freedom but also like being convincing I mean sometimes it doesn't happen like like I've been I mean I probably could point to some successes I've also had a lot of failures along the way yeah we're getting failures yeah we're you like it like it's really many times it's tacked on or basically the way the research is playing out is basically these two parts working in parallel and never actually converging I've been on a few projects like that I've still gotten good work out of it but but basically I kind of you know feel like when you feel tacked on it's hard to it's hard to engage more deeply and at a certain point you just kind of you're like along for the ride and you do you just kind of make it through and what happens happens but I think when it has like tried to convert some some of it I think does require someone to have a stronger vision to like push people toward a certain direction Amy is certainly good at that like like she even there was like one institution that like wasn't really contributing a lot and she just like straight up cut them and then like the site visit reviews there the MSF program managers had had some questions that like are you going to be a strong and a force to do this and she was like she explained I had this one institution that wasn't contributing and I just cut them and and but like you know kind of that you can make the tough decisions on that and that someone has to have a final call to to I think move it forward there's also just a lot of conversation about yeah how do you like how do you come up with a conceptual structure that that actually ensures things are talking to each other and like in our if I go to the slides I think a good example of this is our AI2S diagram that has that sort of has this all right I guess I could share and when we were originally trying to to workshop this diagram with the kind of the three circles originally we there's like kind of these various line kind of one goes like after the other and you're real thing but like like there's not that doesn't so how how it should work is that then applies like some kind of pipeline then we don't actually like iterate it all and so we kind of converge on the circle the head cram that like each part kind of feeds into each other and kind of this virtuous cycle I think was the the the terminology that that came out of it and I think it's mostly come to fruition as not every connection has worked out but but I think like building it into the underlying structure your proposal that that there should be some iteration cycle about there should feed things should feed into each other at at certain points I think it's pretty key I'll let you have anything you want to add otherwise but then I can jump in no go ahead please I think um well there's like five things I want to say I'm trying to prioritize them I think this is an important um clarification that you can do convergent science I feel really strongly about this this is why I often say that you know many of us have been doing convergent research before convergent research was really called this particular thing and so you can propose to grants that aren't just specifically calling for convergent science research right but you believe that this is how you're going to address some of those questions but then there are calls to do more convergent science stuff is in particular from NSF so I think kind of more directly to answer your question because sometimes it depends like how strongly you emphasize that depending on what you're proposing to so like if I were proposing to something like the JTTI like I did it wasn't really a thing I felt like I needed to come through that much because they didn't care about it right so I just wanted to say that but it's also a frame that I had for how we were going to achieve answer address these research questions but I think kind of to your question about how to make sure you're not too stovepiped so I think doing things like this is part of it but I think it's also reflected in the kinds of research questions you frame in the proposal right I think a reviewer and I've been reviewers for a lot of these proposals you can tell if they're not really likely to get to true convergence based on how they're even approaching what they're trying to do if that makes sense I think the one other thing I'll quickly add to this kind of jumping around in terms of kind of what it looks like is I think it really matters if you have a PI who truly cares about and believes in this stuff they set the tone like that's what I think like Amy has done so well on other projects that hopefully the ones that I've led or the other people have led like they're really going to fight hard to actually bring everybody to the table to value their perspectives we talk about this in our paper it's not that everybody has to necessarily have equal contributions but their value needs to be equal right it depends on kind of you know how you structure the proposal and I think that tone is we can't overemphasize this enough that if the PI believes in it then they will help everybody else even people who maybe are entirely sure about some of this like what is this whole why why do we have social scientists involved right some of those people can come along if you actually have a PI who cares about it yeah and one other thing I want to make sure I have a chance to say at some point but Paula I would love to make sure you have an opportunity to answer the question I'm good no you guys have done done great I mean like I said it's it has to do with diversifying your community diversifying your ideas and creating space for everybody like you said that that the equity and justice will come out if the the project is nurtured that way from the through the whole entire thing you know you can't just drop it in in certain spots and say okay we've done it it's it's it's gonna be great but it has to be somehow woven through the entire project so yeah that's it yeah thank you and I will pass it over to diamond yeah I think um this is a great conversation and I think I as an indigenous scientist I think I want to um bring a little bit more of that part of convergence into this conversation and I know like Paulette and I have a lot to say about this um but you know as as I've been working in this like climate change realm now um I've heard a lot about um this term nature-based solutions um and as an indigenous scientist I don't really know how I feel about that term because to me nature-based solutions is just indigenous ways of knowing and being um and so I kind of wanted to hear um all of your thoughts on how maybe we might be able to reframe the way we discuss nature-based solutions you know as we engage with communities as we engage amongst ourselves in our scientific community and as we think about these these projects these proposals how can we reframe the way we talk about this that honors our ancestors and other relatives who came before us who who did this work and laid this foundation for what we call sustainability um so yeah I don't know if anyone wants to start Paulette I don't know sure I'll jump in on that um well first and foremost western science likes to tokenize and and um and silo uh its ideas and thoughts into different specific disciplines and and everybody hoards their information because the uh academic cannibalism is is a thing and and people lift each other's ideas and and try to claim credit for other people's work it happens no matter how much we try to convince ourselves that you know this is the system is set a certain way um the reality is is that western science co-ops other knowledges and claims it as its own and um indigenous knowledges are um how uh Lewis and Clark survived it's how America learned how to survive in the new world was from indigenous knowledge um because when they first came over none of their science mattered because it didn't it didn't um all their methods of farming fell apart in in in the new world and so um when we talk about indigenous knowledge as this other this other thing this um almost this this taboo thing that you know somehow or another is blasphemous if you if you somehow or another believe in it but it's just in the the redistribution of how that knowledge is transferred across generations and across cultures is different it's it's um got tens and tens of thousands of years of of trial and error and survive survival and thriving of different indigenous communities um to it doesn't take it's not a far look to realize indigenous knowledge and indigenous science is is real um so first and foremost western science systems have to come to terms with its own hubris i think first and foremost um and kind of recognize that um indigenous knowledge has very important things to offer um it will supplement some of the holes in sciences in science because western science likes to take thing apart and look at it in its most minute form where indigenous knowledge is a system science it likes to look at how that how all of these things interact with one another and how each of those cycles or systems interact with other systems and so it's a very holistic system of looking at things um so i think as the science culture is shifting um being on the front edge of that shift and being willing to let go of some of your preconceived theories and ideas that Eurocentric science has brought into the system that we depend on and put so much emphasis on but recognize that there are um other ways of knowing and doing things that can lend to the tools and technologies that western science systems have and together build this like we talk about this convergent night this convergent system that develop new frameworks and new systems and new methods and new ideas and new um breaking down old barriers of assumptions and preconceived ideas of what is acceptable by science and what is rejected um is really important so um as we go forward just remember that we can't be so stuck and so so form fitted in our idea of what is knowledge and what is accepted and what is um what is yeah what is accepted what is what is considered a truth and by who and you know um because science truly western science is not truly objective because everything about who informs us of who we are and how we see the world and all the experiences we've we've grown up with and all the influences you know they will flavor our opinion of how things will be how the question will be formed how the analysis will be done what discussion will come from it and what the you know the the so what is um and that can very very differently between a western science and an indigenous scientist um so being willing to cross that that bridge and let go of some of our own hubris of science um and being more respectful and open and patient with under with recognizing our own misunderstanding of what we've been taught um because science is a social construct right so we have to be conscientious of that so I'll just stop there because I can get off on a tangent on on indigenous science really easy. DJ and Julie do you guys have anything you want to add? One thing I'll add is I'm learning um I would I'll be really honest like your question diamond I don't even know this term it's new to me or kind of how it's coming into these conversations so I've got nothing but I'm really valuing hearing about it. Uh I'm I can't say I'm as I really appreciate Paulette's answer and kind of explaining a lot of the the details of it I my probably one thing to add is I certainly seen some of this like kind of it there's like a fundamental tension of this in like the broader machine learning community because there is a lot of people who I guess as you call kind of the Eurocentric science approach like want to just come into a community and like here's our big AI model that will solve your problem boom uh and and it's like in a way actually like because of things like you know Facebook and social other social media systems and how they essentially plug the same kinds of models that are trained in the United States on across the world it uh not understanding various like other countries cultures and languages and like it's caused a lot of social unrest by by like not monitoring the disinformation in different communities uh or thinking that they solved the problem because they solved the benchmark data set that someone created for them when without actually engaging in that community uh and in a lot of ways causing even more problems a lot a lot of like the fail AI experiments that have happened uh have been because of this lack of respect for whether it be indigenous or other populations uh and not and not working with them uh and I want to try to be better with that and try to engage more clearly there's a lot I can do the further improve that uh and like my network diverse next piece is not as diverse in different cultures as I would like uh and and so that's definitely an area to I think we need to kind of be keeping aware of and work to work to improve on so can I ask just maybe everybody else is better informed than I am is there a definition of what you mean by nature based solutions or how this term gets invoked yeah I so one one way that I've heard it used um in Hawaii specifically is um there's like a lot of pollution um in the groundwater and ocean water um through like um just you know runoff and things like that and um shellfish I think there's like a specific like oyster or clam or something that does a really good job of filtering um this pollution and uh like for a lot of us indigenous people like our ancestors grew up harvesting and knowing that these species were important for the environment um but now I've heard it use more as like it's a nature-based way to resolve this like this issue that we've created as humans and in my head I'm like yeah like my people knew that like this isn't a new thing um but uh frequently uh as I like when I was applying for my postdoc I was looking specifically for positions that would allow me to work with indigenous communities and a lot of the time I noticed that this term nature-based solutions was attached or within those job descriptions um and so I felt like it I could see where they were going but it didn't quite honor the ancestors who um are from those places which are interesting but yeah I think this was like a perfect segue to what Shima's question was going to be so I'll hand it over to her absolutely yeah thank you everyone it's so much fun to listen to the organic conversation that we have so another interesting topic in this realm is community science which is kind of like what we talked so far uh and to put everyone on the same page so it means like designing and conducting the research based on priority and needs of the community or a group of target community which is an important step to have a truly actionable science um so what we wanted to discuss is that do you think that we need a culture change in traditional and common way of science to be connected to the priorities of the communities and if we need those changes what are those in your mind especially considering the lack of prior funding before proposals and what are your suggestions for early career scientists how they can connect with communities and design their proposal to be more community oriented great question well let's start with relevance one of the seven hours I talk about you know relevance you know and and that's the so what you know the so what is the work we're doing mean or matter to the community that it's supposed to help um and I think that um some of this work comes from building relationships over time and relationships are built at the speed of trust I've heard that so often from from Bill Thomas and Dr. Wildcat and so many of my mentors that I look up to is is you know relationships are built at the speed of trust and trust takes time and trust can be broken and trust is so delicate sometimes and um work you know if you're gonna work with a community you know know your community know what community you're gonna work with no um you know and take the time to sit at the kitchen table and have coffee and tea um be willing to spend time just being part of a community contributing reciprocity you know what are you giving to that community that makes them want to support giving something to you um so and then what is the responsibility you carry being a member or a contributor to that community are you being accountable with your research um so for me you know I work with indigenous communities because I come from one so it's easy for me to um to identify and connect with other indigenous communities because I too you know come from that space um so if you're you know I think it really comes down to where you want to spend your career because once you start those relationships you really should nurture them over time I have I've got this incredible network that I've built over many many years starting with my my undergraduate you know my um undergraduate associates uh programs where uh I made I became friends and extended relatives of different communities or different people of different families and so when I am tasked with doing a project I've kind of got one foot in already um because I'm already either part of that community or identify with that community so you know coming from you know go go with what you know that that's helpful um be fearless be kind be generous don't come empty-handed you know don't show up expecting a community to give you something without somehow or another having something to reciprocate I mean I've I've gone out and helped farm helped pick weeds helped uh clean up the beach you know I've I've done given physical time and labor to reciprocate with a community to show them I'm invested and I think that you know find ways to invest in those communities um because those networks will grow because word of mouth you know people will tell their their their peers and their community members and their relatives if you're if you're a good one or not you know good scientist or not because um you know especially with indigenous communities we have a long torrid history with science and scientists but I think that is is a safe assumption if you're working with an urban community inner city urban um or farmers and rural you know just having an understanding of their community understanding some of their history some of their politics some of their you know uh challenges uh just coming prepared educate yourself on on them so that you can be respectful and be you know address the etiquettes and protocol of that place and in a proper in a proper way in a good way DJ you have any thoughts yeah I agree with everything Paula said in in terms of like building your network and be I think say yeah getting involved in the care work of the community and yeah I've seen way too many machine learning people come in to to like different even like even coming to the weather and climate community like give me your data said I will hand you my wonderful algorithm that will solve all your problems and inevitably it doesn't happen or or they like they don't engage and leave and and that's like yeah it's unsuccessful that helps diminish the reputation of like that we're trying to build that but this is a useful tool that we want to engage and integrate properly and yeah like there's a whole lot of different communities and even like in the science is a little subset every different subfield of science is kind of its own community and has its own uh like people who are the experts in history and stuff and making sure to I think respect that as well as the yeah like indigenous communities and and different you know racial gender groups and like there's a lot of different communities and I think if you want to engage with one yeah just yeah spending time with it it's kind of the only way to and building that those relationships of trust is the only way to really do it successfully I think one other aspect I'll add is that as you're going through your career and kind of going from being like a grad student or postdoc or very early career where you're kind of maybe a member of a team or working on your own to like building maybe maybe end up building your own group I think paying it forward and setting the culture of like mentoring your students and colleagues and other people working in this area to kind of follow these practices I think is also really key and recognizing that if you yourself can't maybe spend as much time and with engaging with this community making sure there is a point person who has been trained in this area and has like kind of your like you've vetted well that to be kind of your focal point there or if you do want to build a bridge to a new community grow that person to be that bridge so so you make sure they still get get the time and attention they need I certainly had some failures in that regard where I've kind of engaged within like moved on to other projects or just gotten signed off on too many things and then not been able to keep up that engagement with that community and thus lost that connection or in that trust and or there was the potential for something and it didn't and it didn't play out successfully so you can't be everywhere at once so it's kind of get the prioritized but then if it's important or if you like want to make that trade off there can be other ways to still engage but but be still maintaining that respect I think those are great points what DJ just said I think I'm going to sound like a broken record but this is something that's taken me a while to learn is continuing to build in the time when you're proposing something to do that kind of engagement and thinking about it longer term because it is so key and what DJ just said was one of the challenges that I have also faced where even though I care so much about certain communities especially as a project scientist and you're just trying to juggle this one and then move on to the next one and keep yourself funded I've felt like that reciprocity that I care so much about maybe I haven't always been able to fully see that through and again for me this comes back to continuing to learn how to truly budget how much time it's going to take me to do this kind of work all the pieces all the hidden labor pieces that maybe you write into the proposal or maybe you don't but you know that this is part of what how you're going to spend your time you know I think it's an interesting question to think about I think funding agencies are increasingly starting to value if you explicitly say you're going to spend time to do that stuff but even if they don't again building it in and then I would say for me I mean I complete with everything completely agree with everything that has been said I think the community that I work with most closely and have been able to start to bring more formally into the proposals is the National Weather Service like really valuing them their expertise I think you know I've been so appreciative that like Marianne and Chris have really I think they have fallen in love with the forecasters right like and this is really important for me because they have tremendous expertise and knowledge and they have really difficult structures that they work within and they're often not valued compared to you know the the research scientists in the atmospheric science community and even they themselves will often be differential even though you know they have more expertise than some of the people are doing some of the development and so I think kind of working knowing what their structures are like one of the things that Chris has done that I really value is like when he wants to interview them like making it available on a Sunday or in the evening because that might be the best time that they can actually you know talk with you or I've worked in overnight shift with forecasters I've spent a lot of time in their WFOs with them just to kind of really understand this that landscape and have done this for 15 years right so it's not the kind of thing that you do one time I do think it's hard when you want to I think pull out what you said about like knowing who you want to work with early on is so key but then how do you start to like bridge into new areas and can you like work with other people who have kind of those trusted relationships I think is a good way of doing that so I think it's a great question really important awesome so we can go to last question with Anna yeah I'll just shoot the last question and it's just to rub up the entire session um so we are all here early career scientists most of us are or either studying in this conversion science projects or either planning on starting that so we would like to know in one word for each of you what piece of advice would you give to us or kind of like start building these connections that are really important like where to look and how to put together a group and start brainstorming if you have an idea like an early career where did you start like going to conferences or just like chatting with colleagues and how is that evolving um I built my network um through my peer cohort through my community cohort and I I built a lot of connections through going to conferences if it wasn't the indigenous people's climate change working group it was AGU, AAG, AMS, B don't silo yourself in your disciplines conference definitely get outside of your disciplines conferences present at different conferences at different you know um different sessions learn go to and check out other sessions at these different conferences and meet the people that have work that inspires you or excites you or confuses you that you have questions reach out and talk to these people get cards email questions after the conference if there's no time to talk to somebody be be brave um and be patient but definitely don't limit yourself to just your disciplines conference um yeah that's my my my two cents thank you for that um DJ I know I'm like handing off to DJ I was like what do you think but yeah I want to second the getting and go to conferences that are outside your your kind of core domain uh I've I've attended all kinds of random stuff uh from like I went like the last conference I went to before COVID was the catastrophe risk conference like run by the reinsurance companies uh and I've also been to like the cyclone workshop and the kind of like statistics conferences and the GPU technology conference is a particularly fun one because they have all kinds of like everyone's using GPUs but it's like Pixar and then Walmart and BMW and Japanese companies building robots that pick up clothes like toys and it's a very eclectic kind of mix of like especially if you can target some of the like like kind of some of those kinds of conferences where where it is maybe a little bit more eclectic and you might have some very unexpected combinations uh in or places where you're like the one person of your expertise in the room in addition to like building your community kind of conferences where you are with your people I think those are really important too um and like one of my other strategies is especially a place like NCAR is go have lunch just go sit down at different tables that especially now that we're back in the buildings again like maybe you know one person the table but you don't know anyone else that's a good way to to like kind of have an in uh on on kind of making new connections I mean it's for the postdoc happy hours and stuff for useful and and these kinds of Q&A sessions uh I think Thompson like her series was also really good for like inviting like people that are way outside your community you know uh you never know kind of where these lead and also keep in mind that some of these connections aren't like next day you're going to be writing a giant research proposal with this convergent team it's stuff that takes trust like the speed of trust I like the line from Paulette uh it can take years for for some of these things to play out but then when like you know everything all the conditions come together boom you know some so so yeah try to cultivate and see where things take you there's a lot of kind of serendipity in this process um so maybe a couple other things to add I think I think these kinds of fora are really helpful I think if you are interested I think those of us who do this work should always be willing to talk with you more I think a couple of things like we could um I would even be happy to share past proposals because I think there's a lot of knowledge here that's hard to externalize right um in terms of like even how you frame things or you know we try to externalize some of that but just some of it it's just so hard to capture and so being able to talk with people who've done this I think if like knowing you're interested is is helpful and then I think I do think it's valuable if you can also start being involved in other proposals that other people are leading like I think it would be really hard for me to have just started off even though maybe I was interested in this I love DJ seeds of convergence I wouldn't have known what I was doing right like so it helped work alongside people who were doing this for a while and then to learn the process but I think the final thing I'll say because this is such a big important question all these questions are big important and I'll say this to everyone is that NCAR is trying to stand up more coordinated convergent science initiative at the directorate level and I'm going to be helping with that and so part of this is and will involve an implementation plan to like help facilitate this work at NCAR better and including leveraging the existing amazing stuff that's going on and so I would love to maybe have another conversation about these kinds of things to think about I mean this will slowly be ramping up over the next couple of months and be announced but I think for all of you to think about what are additional ways that we can support you and you're interested in doing this kind of stuff but be really helpful to hear about well that's great yeah so I think that we're on time I think that Mariana is gonna yeah it's gonna wrap up this session well this session's the entire seminar series is it just yeah so I think thanks Anna first thank you to all of our three speakers that we had today Julie Pula and DJ we really appreciate you spending the time and sharing the trial and tribulations if you will and we also appreciate you sharing your own unique perspectives on this but on a greater level I want to thank everybody who's in attendance and past attendance that aren't here right now but to just show you what it means when postdocs have an idea and they want to see it through so we were able to put on a six part series as you can see here these are the great faces of those who from internal to and car new car and also external to and car new car who are able to help us better understand the grant writing proposal process I don't think any of us are necessarily experts at this point but we're able to have a little bit of our feet wet to go forward with and the last thing I want to say is keep an eye on your email because we will be doing a post event poll to see how we can better serve the postdoc needs within the community and also just ideas that you might have for other types of the events that you would like to see going forward so with that I'll close it with a physical round of applause and then also Google's new feature you can give virtual rounds of applause and hearts so thank you to everyone this is great all right and now I understand mine doesn't do that they just added like this week yeah yeah but it will just like it will do what it looks like anesthesia I don't get the whole animated and I was like I don't know what you're talking about being this all over and no Maurice did you have a comment think someone raised their hand up no all right well with that we'll close it out and again thank you everyone so much this was great to have everyone's participation in the discussion thank you so much have a good one thank you