 I guess I'm up first. So I'd like to thank Lauren for taking good notes today. We had a Pretty substantive discussion as you can see by the bullet points. We spent most of our time on the First question and kind of ran through all of the rest of them I think before I get into the detail what I'd like to do is just spend a minute or two talking about the high-level inputs that I think summarize where everybody was headed and thinking about first being Early stakeholder involvement and what are the questions that groundwater models are trying to answer? I think the consensus among the group was that we develop great models and then present results But it'd be nice to get stakeholders involved early on and the model Questions so that we can develop the right model for the right scale both temporally as well as regionally So that was an important discussion that we had And I think the other the other question that everybody raised was how do we do a better job of integrating the human scale and Socioeconomic issues into groundwater models to help us make decisions So those were two big questions that I think ran through all of our discussions that I don't know the answers to but you guys are smarter than I am So you'll figure that out as we develop the next generation of models So the first question was reflecting on the modeling frontiers. What are the most important groundwater model uncertainties? Obviously people wanted more pumping data so we can get more information About the physical last attributes of the aquifers we're dealing with Using remote sensing data more effectively, I'm not going to read through all of the bullet points. You can go through that just trying to Summarize the highlights Time scales was a very important issue that everybody tried to address In terms of model development and model outputs How do we deal with unconventional aquifers? Most of our models deal with data sets that we collect From aquifers that we know about but there are other untapped aquifers that folks might be trying to model That we don't have a lot of data for so we need to figure out how to deal with those issues and I guess the spatial and temporal scales that we use for modeling and Developing management options and that try to answer management questions for what a resource managers Is always a big problem in terms of how we use our models and develop our data So for those of you that have participated if I missed anything You know, please chime in and I think the last bullet the human elements behaviors. How do they? Impact model development model output, how do they inform policies and governance? How do we develop more effective groundwater regulation in places where it's weak where there are no institutional Programs to manage groundwater effectively can we use models? To help inform those kinds of decisions. So that I think summarizes our Answer to the first question pretty effectively Hi, everyone. So I'm reporter for the group that stays stayed in that room. My name is Loreline I'm from the Columbia Water Center and what's very I think there's great Concurrents between the points that our two groups have been making so I'll be quite short I don't think we can learn more much more on that So the first aspect that we talked about is that we had similar questions that you did Modeling flow and recharge before what where and at what scale really? What are we trying to accomplish and do we really want to do more modeling? It's really important to do that in a way that is context specific and to answer a specific read a specific question In particular, we highlighted this with the mismatch between the global models that are developed and local decision making There's a really mismatch in scales at all levels that needs to be addressed there We then talked about getting into the more specific parts about the modeling questions and uncertainties that needs to be addressed Subsurface properties really trying to understand that Their order of magnitude in some of their variability But also the human factors just as you mentioned so really understanding the water volumes that are pumped where pumping is happening The infrastructure and how these pumping decisions are made. We say that there's a lot of uncertainties on that side We also talked about the coupling surface water and subsurface both physically so understanding the flow process and the physical interactions between Reavers and groundwater but also as surface water can be an alternative Another use another source of water and really understanding really this human Component of that so who's pumping surface water when people decide to use groundwater and going back and forth between these two sources we talked about The aquifer themselves so their storage capacity, but also the quality of the water So saline fresh, but also contaminants whether they are geogenic or anthropogenic One aspect that was also brought forward is that we don't really know how to convey this uncertainty to stakeholders So we cannot solve all of the uncertainties that we have in the subsurface But we need to make sure that we convey this uncertainty to the people that are going to use the results and make decisions We listed also the types of uncertainties We're saying there's uncertainties in the inputs the data that we rely on the modelled physical conceptual the outputs There's also the use of the outputs, but there's also uncertainties of who's going to use The output so all of these uncertainties should be kept in mind all the way through And that was it for our first question Sure Yeah, so the second question is what kinds of data Included proxy data and approaches could be most useful in minimizing modelled uncertainties So the first thing that we talked about is that there's subsurface hydrologic Hydrologic properties that should be included, but also the water properties So really taking into account the quality of the water, but also its age really making sure that we understand that and we use this information to understand recharge and and lateral flow We say that we wanted to maximize we didn't talk too much about new types of data But really maximizing the use of existing data sets so that we can do data assimilation But also we're thinking that there's a lot of data products that are there that are not fully exploited And about that there's this wealth of data that is available But not not all data is useful not all data is easily accessible and reusable for Assimilation so that is something that needs to be happening. So maybe we should ask ourselves also as we publish new data sets What is the demand for that data was going to use it and how? We talked also specifically about model parameters, so there's a tendency to publish these it's greatly happening From the academic sector, but we're thinking that there's a lot of other models that are developed and maybe not as much Exploited so maybe consultants that develop models government agencies That develop models that could be useful and we could rely on their parameters, too We also talk specifically of the wealth of data that is available in the US And that could be exploited more to understand how to develop methodology that allows to deal with the sparsity sparsity of Datasets in other region of the world We talked more about model outputs So really trying to distribute them for reuse so really having rear We're thinking that actually repository Can be a challenge so how to host host these model outputs correctly But we thought that one of the biggest challenge that needs to be addressed for proper reuse is actually having proper metadata and Format and we're thinking that policies and agencies have a role to play in this to coordinate effort on maximizing the reuse of this data and Again, we stress once more that model outputs are relevant But there can be too much data at times also So we should keep that in mind especially to address Stakeholders needs making sure that the model outputs are useful We also mentioned that raw data is not efficient that we want synthesis analysis But in particular confidence intervals to make sure that the data can be adequately Integrated to have a better assessments and better assimilation of the wealth of data that is out there so again, I think there was a lot of Concurrence on some of the results of our discussion on the kinds of data I think there was a lot of discussion associated with using Large-scale AEM surveys more geophysical data better spatial representation of Actors Again we got a lot of discussion associated with citizen science and the impact of human activities on our groundwater resources How do we integrate a lot of these other socio-economic? Data sets into our groundwater models and how do we make our models more efficient and effective in predicting outcomes? I think looking at Data to help us understand the differences between the developed world and the developing world How do we model aquifers where data is sparse? Where aquifers are not well managed like they might be so is so to speak here in the United States That's a debatable question, but we won't get into that discussion and what lessons could we learn from Well-managed aquifers and translate that into models and the developing world where water resources management Suffers from a lack of good information Looking at remote sensing information grace data. How can we use it as input into models? How can we use the data to evaluate the models in a diagnostic way to identify uncertainties and improve parameterization and Utilization of new data science techniques if we think about the proliferation of censored data Looking at citizen science and cell phones is a lot of information That people are collecting and we don't know how to manage that data or integrate it yet and can we look at new computational tools and Utilize AI to refine modeling capabilities So that was the output on what kinds of data and what kind of Approaches we could use for improving model outputs So for the third question So we were discussing basically writing a wish list for NGA what we think we would love to see happening So the first thing we mentioned is actually has special resolution satellite imagery To really identify water infrastructure. So pumps Also dams and reservoirs to really understand groundwater use and really get an insight on that we talked in particular for irrigation and We mentioned geostationary Satellite imagery because that would give us, you know, high resolution special resolution But also high temporal resolution to really have proper time series and better understanding and to address the scale gap that you were talking about Before regarding Groundwater use so we think there's a great potential there and we believe that NGA might have access to these higher resolution data That could be very useful for us. We also mentioned other Social or other soft data Typically that could inform a behavior and human Relationship with groundwater so typically understanding better when failing crops electricity usage when deeper wells are curing to really model better groundwater and the human component of groundwater We also thought we were not sure about this that maybe there's potential for NGA to host certain large model outputs Really by using repository capacities for very large data set from models or or not and really facilitate Usage of this data for local water planners. So really like tools for formatting sampling Transform writing transforming the data set so that we can maximize their reuse and their integration at a local at the more local Scale and facilitate decision-making We were also thinking that maybe for the ideas that we were proposing in previous questions Is that maybe there's a role for NGA to develop through BA's Specific RFPs for data storing tools development and processing of these large data sets and maximizing their integration or future users I Think that's all we say that's all our wishes for you So we had the benefit of Tony being in our group So the answer to your last question is there are a lot of ways that NGA can collaborate with Modelers and I'll let him explain that maybe after the discussion is over for the benefit of everybody in the room But he asked us a couple of questions on what can we help NGA with in terms of Identifying decision-aid tools for groundwater models to inform policy makers Collaborative efforts with modelers and others We need to identify the kind of water quality information That we would like that may be available through NGA I asked the question about classified versus unclassified information and Tony indicated that they have a lot of Unclassified data that is publicly available and we may not be accessing that effectively So I would suggest everybody check out NGA's website and figure out how to access that data to inform our marbles Sorry, I Wouldn't say a lot of it's publicly available Some of it. Yeah, but but but yeah, but there are I what I was trying to say Was that there's a there's ways there's mechanisms where you can get access to NGA's unclassified data So it may not be available to the public, but there are mechanisms if you join in some of these If you participate in we have what's called a crada, these are there's no money exchange uses cooperative research agreements between NGA and Typically academic scientists, but we also do work with commercial companies as well There that's where an opportunity where we need to provide data To an individual or group and then they would then be able to apply their their scientific knowledge on advancing on a problem But then there's also opportunities even with the within the Like the NARP and NURI programs, which is our academic the NGA Academic research program where we provide funds where you could then have a legitimate reason to access some of our data source So I think one of the things that came out in our breakout session yesterday was NGA needs to communicate Some of the these data holdings better to the academic community So that way when they do write these proposals They know what? Opportunities there are to you leverage our current current resources because we do have access to things like planet lab Right and digital globe imagery that we can't provide under some of these agreements It's just that we can't just give it away because there are certain license restrictions that we have And so that's where the lawyers come in into play and they tell us what we can't do With a lot of these these data sets But there are opportunities that we could leverage that to the academic community To better better serve, you know our nation's needs here Just to kind of wrap up. There were a lot of specific asks for information That could inform models rather than go through all of the individual ones I think everybody has an idea of the kinds of information we'd like access to That would make for a better modeling Outputs and inputs and decision-making tools. So we appreciate it Tony's input on that discussion So the last questions what about was about some examples of successful collaboration Opportunities and we really didn't answer that question at all Instead what we did is to I think that's the accumulation of all of our previous comments is actually with a suggestion And we think that it would be fantastic to see a collaboration that would be between USGS and GA NASA and Universities to really try to do the most a lot of assimilations and maximize use of Data that is currently existing Process all of this information rely on model simulations But also augment that with machine learning tools to really maximize of our understanding and really maximize analysis of the existing data product We were thinking of doing this first US centric so that we can really do this As integrated as possible this assessment But that could serve then as an analog for missing data in other regions of the world So we will have only part of this information and we could rely on US centric information to augment this data set and really provide an analog to Maximize our understanding of groundwater and groundwater dynamics including human component of it so wrapping up We Discuss some examples of interdisciplinary efforts. I don't know what all these acronyms mean, but you probably do I think the Earth Scope effort one of the Discussions was about Standardizing the way data is collected Standardizing the approaches to the data once it's collected and Stored so that we can make more effective use of it the Earth Scope project. I think is a good example of that We talked about tools that could be more widely accessible to a variety variety of stakeholders Bringing stakeholders in at the beginning. I think I discussed that a little bit earlier and some examples of Better water management was in Tampa Bay one of our Researchers talked about how they integrate Water supply plans and more effectively manage The variety of water resources that they have both from groundwater and diesel Sources in order to effectively manage resources during Whether variability or use variability and things like that So we ran out of steam as well because we had a lot of discussion in the first two or three Didn't really get to explore a whole lot of successful collaborations. I think yesterday. We did mention if you're talking about Collaborations on the federal level. There is the advisory committee on water information Which is federally funded all federal agencies participate in that and there is a subcommittee on groundwater That deals with what are the federal government agencies need in terms of? Groundwater information and I would encourage all of you to check out that website and participate if you can