 As long as there's space up there, you might want to go up there. Oh, they're going to be. Yeah, sure. Yeah. I'll leave you 10 talks, all right? Yeah. Okay. Good morning, everybody, and welcome to this joint meeting of the Geographical Sciences Committee and the Mapping Sciences Committee, with our topic today being the Federal Landscape of Geographical and Mapping Sciences. We're happy to see all of you who are here in the room with us, and we're glad that a number of people were able to join us online. So as a note to the people in the room, I will say, remember that we are not the entire group, and please do use a microphone when you speak. For those in the room, the exit for this meeting, if there should be need for exit, is the door you came in and the restrooms are down the hall. We're really delighted to have all of you here with us today, and I think what we're going to do first is to just do a pretty quick round of introductions. We have two committees here from the National Academy, so we have a number of people in the room. Well, first go around the table, and it would be nice if you would just say and your institutional affiliation, and then we'll pass a microphone around the rest of the room. So why don't we begin up in the front? Let's begin with Bill Selecki, and then we'll go across to Marianne around this way. Bill Selecki, Geographer, City University of New York, Hunter College. Hi, I'm Mary Malewski, National Weather Service. I'm the Chief of the Water Resources Services Branch. Burl-Monts, Geographer at East Carolina University. Sandra Knight, Consulting, WaterWox is my company. Good morning. I'm Deirdre Bishop. I'm the Chief of the Geography Division at the United States Census Bureau. Hi, I'm Grady Tool. I have a company, 3D Ideas. I specialize in design and manufacture of LiDAR systems. Martha Wells, Spatial Focus. I specialize in addressing systems and addressing development. Mark Raker with Open Geospatial Consortium. We're focused on interoperability and innovation in geospatial. We do how to leave pretending to be a geographer. Oak Ridge National Laboratory. Kathleen Stewart, Professor of Geographical Sciences, University of Maryland College Park and Director of the Center for Geospatial Information Science Center. Nancy Jackson, University of the Institute of Technology. I'm also a geographer. Ann Lynn, National Academy of Staff. I'm Harvey Miller, Professor of Geography and Director of the Center for Urban and Regional Analysis at the Ohio State University, and I chair the Mapping Science Committee. I'm Carol Hardin. I chair the Geographical Sciences Committee. I'm Professor Emerita of Geography at the University of Tennessee. Good morning. I'm Cara Laney. I'm a staff officer here with the National Academies. Glenn McDonald, Department of Geography, University of California, Los Angeles. And I'm on the Geographical Sciences Committee. Dan Brown, Professor and Director of School of Environmental and Forest Sciences, University of Washington. I'm on the Mapping Sciences Committee. Mike Tischler, I'm the Director of the National Geospatial Program at the U.S. Geological Survey and one of the sponsors of the Mapping Sciences Committee. I give a Junic Geography and Spatial Sciences Program Officer at National Science Foundation. Right, and let's go around the outside of the room here. Maria Zamankova, National Science Foundation, the Computer, Information, Science and Engineering Directorate. Good morning. Laura Klein, State Department Office of the Geographer. I'm Hanan Samet from University of Maryland, College Park. I do spatial databases and reading news with maps. Hi, I'm Macarena Ortiz. I'm a physicist at the National Geospatial Intelligence Agency in research. Good morning. I'm Lauren Steen. I'm also a scientist with National Geospatial Intelligence Agency's research office. Good morning. I'm Mike Brady, National Geospatial Intelligence Agency. I'm in the Maritime Safety Office. I'm Mindy Bishop. I'm an engineer managing the Smart Columbus program for the city of Columbus. Gary Burg Cross, formerly of the Spatial Ontology Community of Practice, interested in semantic harmonization. Rebecca Summers, GIS Management Consultant, representing Eurisa today. Good morning. Elizabeth Aida with the National Academy of Staff. Dave Wandsfuck, Retired Member of Public. Thank you. I'm C. K. Shum. I'm a professor of geodesy in geodetic science at Ohio State University. I'm also with the School of Earth Sciences. And I met, well, Rainer and Lane about seven years. Okay. Thank you. Okay. Thank you. Thank you. I'm so glad that people continue to come into the room and will be coming and going throughout the day. So thank you all for joining us. Oh, introductions on the phone. Did you want, can we have people on the phone introducing themselves? Looks like we've been unmuted. This is Josh Murphy with the office. Hold on for just a minute. We're figuring this out. Amarita, could you introduce yourself? Hi, I'm Amarita Gupta. I'm a PhD student at Georgia Tech. We're having a problem with broadcasting just a minute. I think we're just having a problem getting it to come through the speakers. We can hear her here, but I'm Amarita. Go again. Hi, I'm Amarita Gupta. I'm a PhD student in computer science at Georgia Tech. And I work with spatial planning and flooding. Thank you. Frederick Chorman. Joshua Murphy. Good morning, Joshua Murphy. Geographer with Noah's office for customer management. Thanks, Mark Brun. Hello, I'm Mark Brun. I'm a research GIS analyst. I formerly worked for RTI international currently unaffiliated was one of the developers of the national hydrography dataset. Thank you. Pinky Mondale. Good morning, everyone. This is Pinky Mondale assistant professor of geospatial data science at the University of Delaware. I'm Sean Olu. Good morning, everyone. I'm Sean Olu from the Boston University. I'm now visiting professor here and I'm majoring in geoinformation systems and the remote sensing. Thank you. Hi, I'm David Smith with DC government and I'm a GIS analyst there. I think that's good. Thank you all again for being with us this morning. Before we have our first session, I'd like to talk to you with my colleague here, Harvey Miller, a little bit about our committees to sort of set the stage for this meeting and tell you what's going to be happening today. So I wonder if we could have the first slide, please. I chair the geographical sciences committee and we're a standing committee as is the mapping sciences committee. And we're both committees of the board on earth sciences and resources. And the slides are coming. Great. Next slide, please. Just to give you a sense of what a standing committee does, we serve, we do meetings and workshops such as the one today. We serve as a forum for engagement among federal agencies and states and non-governmental organizations and other organizations. So we convene gatherings around a theme or topic or a set of topics. We also do public outreach. We have webinars, videos, public lectures. We will in fact have a public lecture this evening. More about that in a minute. We sometimes provide information and testimony to congress and we have a presence on the internet with our website as well as Facebook and Twitter. The main product of meetings like this is to find places where there's a knowledge gap or just a need for more scientific input in policy and decision making. And when we identify those needs, what the standing committees do is to convene an ad hoc committee, not of ourselves, but of a different group and that would then produce a consensus report on really what is the state of the science on that topic. Next. You've heard introductions from members of our committee. A few are not here or are not here yet. And here you can see our names. Next slide, please. The Geographical Sciences Committee was originally created as the Committee on Geography in the 1990s with the recognition that spatial, place-based science and human environment science were of increasing importance and increasing importance in the national academies. In 2004, the name was changed slightly to be the Geographical Sciences Committee just to better reflect what we do and to reflect the fact that the breadth of expertise in the areas of human environmental science and spatial, place-based science may extend beyond the formal discipline of geography that there are many geographical scientists out there who aren't card-carrying geographers. Next. The mission of our committee is to provide high quality scientific, technical and policy advice and recommendations to society and to government at all levels using the methods of spatial analysis and representation. We have another mission in addition to our work of providing science to policy and that is in the latter two points that you can see on the screen to foster international cooperation by serving as liaison to other national geographical organizations. And then finally to serve, our committee is the official U.S. liaison to the International Geographical Union and we facilitate the participation of U.S. geographers in that group. Next. Just to give you a sense of the sorts of issues we work on and meet about, here is a list of some of the recent topics. Land change modeling, wildland fire management, equity and access and health effects of exposure to nature, opportunities and consequences of using sensors to capture human geographical behaviors and vulnerability of U.S. energy infrastructure to coastal flooding. As you can see, we have a wide range of topics as geography is a broad field. Next, please. And over the years, we've had quite a number of collaborators and sponsors. We are currently sponsored by the National Science Foundation. Okay, next. Okay, thank you Carol. I'm Harvey Miller and I chair of the Mapping Science Committee. The Mapping Science Committee was established in 1987 with a very simple but I think powerful and profound goal is to promote the informed development use of spatial data for the benefit of society. And our sponsors have been traditionally the USGS, the NGA and the U.S. Census Bureau. Next slide please. And again, this is our committee. Many of them are in the room. Some of them are not or not yet. But you could see that the Mapping Science Committee does draw from academia but also geospatial organizations and from the private sector. We really look at the intersection between science and technology and between academia and the private sector. Next slide please. So our meetings are usually two per year here in Washington, D.C., here in the Keck Center where we see updates from sponsors. We have usually a public of workshops on a geospatial topic and develop ideas for future workshops, studies and outreach similar to the Geographic Science Committee. And we have a wide range of participants as well. Committee members, federal, state, local government and organizations at these meetings, private companies and also professional societies and NGOs such as ASPRS, AEG and COGO. Next slide please. And this is some of our recent topics just in the last few years. Some of it centers on mapping but also the impact of these mapping technologies on broader science and broader society such as 2D, 3D and 4D mapping of cities. They'll come up with 3D urban models and 3D urban models with temporal expressions. LiDAR, the next generation of LiDAR technology and how that's going to change mapping and change a geographic science. Smart cities, grand challenges facing smart cities and the agency perspectives and cloud and based mapping, cloud enabled, excuse me, mapping science. What happens when we move mapping science, data, technologies into the cloud? What are the implications? What are the new capabilities that we can perform in this new cloud based environment? Next slide. So we'll talk now about today's workshop. I guess I'll jump in with that and Carol can go ahead when she wants. What we're trying to do today is we want to basically as the title implies the federal landscape for geographical and mapping sciences. We want to generate potential topics for additional meetings and ultimately studies showcase the scope and importance of geography and mapping sciences in the federal landscape, hear from agency and other public and private sector speakers about the issues, perspective, interest and needs with respect to geography and mapping and also provide an opportunity for interaction and collaboration among these agencies. Next slide please. So I'll let Carol take over and talk about the first session. Okay, our first session, which will be this morning, is on analyzing and communicating flood risk. This is an enormous topic and it really is central to the work of both of our committees. Billions of dollars of damage occur every year in the U.S. and human lives are lost as a result of flooding and it still happens. We should be able to do better than this and so we're wondering what societal factors affect the ability of federal and local agencies to reduce and prevent flood damages and we chose to talk about flooding today because we had to have some area of focus here but we understand that whatever we learn from flooding may also be quite applicable to some of the other hazards that people face such as wildfire. Session two also this next slide please, sorry. Session two also this morning we'll be mapping the new Arctic. As we know there's a lot of environmental and climate dynamics occurring at the high latitudes in the Arctic. The new North as described in this article Nature from 2011. We're seeing interest, commercial interest in the Arctic, new shipping corridors being opened up by the lack of ice up there. Natural resources and there's scrambling and competition to acquire those. We're also seeing strategic issues geopolitical issues surrounding US, Russia, Canada, Northern Europe and even China are scrambling for this new landscape in the new North and there's also scientific interest disappearing sea ice to find a permafrost changing ecosystems and the impact on traditional cultures and the global climate and our question here in this session is how can geographic and mapping sciences inform our understanding of this rapidly changing region and the implications for the rest of the globe. Next slide please. In our session this afternoon, session three will be on smart technologies and communities and the smart technologies that are being introduced into our cities and communities are profound technologies. They're going to be going to have deep changes on our daily lives, our businesses, our social practices, the delivery of community services, the range and deep impact in our communities. And one thing we'll be focused on this afternoon in particular are smart mobility technologies, ride sharing, mobility as a service, mobility beyond ownership but actually may have to configure mobility as another type of service and eventually connected in Thomas vehicles, more commonly known as self-driving vehicles. So these are going to really change our lives. They're going to change our lives on the order of how the railroads changed our lives in the 19th century and how cars changed our lives in the 20th century. So we need to get a handle on these smart technologies and these smart mobility technologies. They also are very much geospatial technologies. Smart technologies are fueled by geographic information about sensors in the environment and also are really trying to shape processes and activities across space and time. So our question is how can we shape smart technologies so that communities are more environmentally social and economically sustainable? It is unclear like any profound technology that necessarily will lead to better communities and that's something that we have to get a grip on as scientists, as practitioners and as different levels of government is how we can make sure that our communities are better in places after the better places after this smart technology revolution. And that'll be our session in the afternoon. So next slide please. This has just given you a sense of our schedule. Again, we'll jump right into the Human Contributions to Flooding session. We'll have a keynote talk about half an hour then some panel reactions about 10 minutes each with open discussion. We'll have lunch. I guess I got the Arctic rung one that's after lunch. We'll have a mapping a new Arctic and then we'll have a keynote talk again and some panel reactions discussion and break. Next slide and then the smart technologies and communities to wrap up our third session this afternoon again same format and then around 4.15 we'll have around the room general discussion across all these topics and talk about what have we learned, what are the next steps forward, how do we continue to engage in this dialogue and this conversation. But don't think of leaving at five o'clock because there's more. Next slide please. Think about, stick around for the Gilbert White Lecture this afternoon at 5.30 where Boudou Baduri will be talking about synergy between geography and mapping within nation's energy mission which I think will be really a perfect capstone to today's workshop. We're all looking forward to that Boudou. Unfortunately for our online visitors that will not be live streamed but the lecture will be videotaped and will be on the website later on. Just as a kind of procedural note, today when we have our speakers and panels we'll have the opportunity to take just a quick moment for any burning question or question of clarification right after each speaker. Otherwise we'll move right on and we'll save those questions for the discussion period. So we have one discussion period with each session and then we have a broader discussion period where we can call everything back at the end of the day. We are right on schedule. We are right on schedule and I think we're ready to proceed with our first session. Great. And I was actually just checking my notes because I know it's sort of embedded in this sort of topic by the way I built Selekyl on Moderate. I'll introduce myself. So there's been a lot of discussion even of the title, like in this kind of connection between humans and the flood risk question and I know it's gone through a couple of iterations so actually I think what was up there was slightly an old title and here's our new title, yes. Analyzing and communicating flood risk. So we'll have a keynote speaker sort of present some of the broader questions and then we'll have two follow on panelists. We'll do about 30 minutes for the keynote and then some clarifying questions as needed with respect to that presentation and then we'll go on and have two roughly 15 minute presentations as well from each of the panelists with clarifying questions and then we'll open it up around 1130 for general discussion. So I'm really excited to sort of hear this conversation even within the geographical sciences community or committee there's a lot of discussion about how to sort of frame it and sort of present this kind of question I think in many ways as Carol has already said it's sort of flood risk in many ways as a touchstone to sort of looking at much larger questions so I think that's sort of part of our objective. So I'll just sort of present each of the panelists. Well actually I'll just do the keynote first and then I'll introduce each of the panelists as we go through. So we'll have a keynote presentation Sondra Knight, Dr. Knight has a background in engineering, a PhD in engineering civil engineering from University of Memphis she's worked extensively I think you've probably taken a look at her bio but has extensive sort of experience sort of both on the academic but more critically I think professionally as well from the agency perspective so she brings that that kind of expertise to bear from NOAA FEMA and the Army Corps currently she's the she splits her time as president of Water Wonks which is located here in the city and then also has a position as a senior research engineer at the Center for Disaster Resilience at the University of Maryland so it's very exciting to hear her expertise and perspectives. All right thank you Bill very excited to be here this morning I hope I meet the expectations of both titles and the discussion we had about what I would present the nice thing about being retired fed and being on your own is you can be a little bit of a provocateur and so I'm going to start out with being that I hope and get us to think but I'm also going to share with you some research I've been doing over the last year and maybe a big idea I have I don't know how big it is but I'm trying it out on you today so thank you for giving me this opportunity okay so we all love and hate maps right I I'm not a geospatial person I will say when I took GIS in 1991 I thought oh my gosh it's going to save me from the mylar and USGS maps I used to have to use to delineate my drainage facins for my hydrology studies right and so on the left these are maps that people but these maps are things that people use on a daily basis right and I love them and I hate them so the one on the left is you know your map app right so I want to go from my place on this map to the Kennedy Center and I call it up and it gives me several routes and I think okay I get it and I push go and then all I see is this triangle that keeps jiggling around and saying proceed to the route proceed to the route and so as if you're geospatially used to looking at large scale maps when you're then zoomed in to just a few feet in front of you you really get disoriented I was in a parking garage in Rhode Island a few weeks ago on the fifth floor and my my map kept going procedure rerouting proceed to rerouting and a little triangle was just spinning like crazy you know so so I think when we think of these things we have to think about the user and how it works for some reason I had an Uber driver my husband and I almost always use public transportation or walk but we were treating ourselves to ride to the Kennedy Center in an Uber and our driver by the way had an app and he didn't take any of these routes I don't know where he was going he went way up in North DC and now or so to get there I thought we were going to miss the performance the map on the right is of course it's just a simple trail map my husband and I go to Sugarloaf Mountain have any of you ever been there hiking yeah okay Leather it's a really crummy map right but I keep it in my pocket because then at least when I get to the fork in the road I can decide if I want to go a little further or a little less I really don't can't exactly figure out if it's going uphill or downhill even though there's some sort of and then finally I would say the map I've used the most I moved up to DC in 2007 and my husband I both got one of these folding maps with the metro center the metro map I carried it with me probably for five years I'm just now figuring out now what end of the platform to stand on and what exit to get off right you know so this is very useful so for all our high tech stuff that we do we really sometimes it's simple and as users and I the title of my presentation was presumptions and assumptions which means basically there's uncertainty regardless but presuming means that you assume that you presume that somebody has some real data and this is pretty accurate and maybe on a map like this where it's been surveyed and it's to scale it is but when you start adding other geospatial data it gets so I'm going to give you some examples to introduce this idea of flooding but I'm going to one's going to be at a global level I'm going to talk I'm going to get a little bit into Mary space on Noah's hurricane predictions and some of that and then I'm going to talk in detail about FEMA maps because when I was at FEMA I was over the flood mapping program one of many things so this is not the maps that came out of this study but you've all probably seen this new study that was in nature communications about the impacts of sea level rise and that we've grossly underestimated the impacts not the climate science not because of the climate science but because of the geospatial data we used so and this means more to y'all than me because this is my space but the satellite data that had been used was used primarily used for these large scale looks at global impacts of sea level rise because it was a standard standard that could be used everywhere. LiDAR grade is here LiDAR is not available everywhere so detailed information is then available everywhere so this study didn't change the climate science but it used some interesting techniques too right it used LiDAR data from the US and Australia where there's a lot of spatial resolution and somehow normalized it to the biases that are in this large scale satellite database so the biases were wrong in the really worst way because it was all too tall so everything you know canopies and dense urban areas appeared to be higher above sea level than they actually were so redoing this study says that and I wrote down the cities. Cities like Vietnam, Ho Chi Minh City Bangkok Shanghai Mumbai Alexandria, Egypt Basra, those all those cities are going to experience much worse impacts from just regular tidal flooding because they're going to be so far below sea level so what this tells us is when we're looking at global scales and big scales we often use less accurate information we try to get a common data set because we want it to be at least that we've used the same stuff everywhere so it's not biased in that way and maybe that's good for strategic planning or communicating something that a broad scale but at the end of the day maybe this growth underestimation of what the real impacts of sea level rise are has really impacted how we as nations and as leaders have thought about how impactful it is and how critical it is so just a consideration I am a weather junkie after working at NOAA and meeting all the best meteorologists oceanographers and you know atmospheric people and climatologists I'm not one but I like to pretend I know a few I do know if you would like to pretend to be once I watch the weather channel just all the time plus disaster resilience is my gig so I really want to know what's going on and just watching Hurricane Dorian this last time I just kept getting frustrated I kept thinking we're really really really not communicating what people need to know we have great information so I love that we have you know the hurricane prediction center has gotten so much better on it hurricane tracking forecast I like to see the spaghetti models myself that really gives me a wow look it's kind of all over the place but you know the European ones land in here and the NOAA ones landing so I love that but the cone of uncertainty if I if to the casual observer makes gives you a sense of safety that's probably not there the better the weather service gets the narrower the cone of uncertainty but look at the size of the actual hurricane on the right relative to the width of the cone of uncertainty if you are you know the casual public observer and you see this you're thinking oh I'm outside of the cone of uncertainty so I'm not going to have any impacts and yet we know these big big hurricanes with all their wind that they've been pushing it for days and days and days they're gathering up more moisture with all these big hurricanes we know we're going to have substantial rainfall and probably surge and so the weather channel that people showed the cone of uncertainty then they spend another five minutes explaining well that's really not the biggest problem the biggest problem is probably rain and surge but it's hard to get this information it's hard to understand it so let me explain so we saw in Harvey what 60 inches of rain in Houston I don't know I'm 61 inches so is that what that means 60 inches of rain no that's not what it means does 60 inches mean the creek in my backyard is going to overflow does it mean that it's going to be you know what does it mean so what is 5 inches 5 inches can be rain I'm glad we have someone from the District of Columbia the Federal Triangle in 2006 had an intense short rainfall and I forget what it was but it was much less than 20 inches and yet it flooded flooded there was interior flooding right so it can rain in your backyard this is very you know random right and so that also gets into some other uncertainties that we don't really understand we I'm just learning some new words this year aleatory very random uncertainty and that's what weather kind of is until we get to know more right and so you have these random uncertainties but then you have also these you know things that we introduce whether it's through the model or through the data or the assumptions or the parameters in the model so we we add uncertainty to all of these products in addition to it just being kind of weather random and so I'm I'm watching and I'm thinking well okay but what does that mean to me if I live in New Bern North Carolina is it going to flood and our name is is this going to be a problem because we've had some bad events for the past few years and my home has been inundated before with you know name one Florence Irma somebody Matthew Floyd you know North Carolina always gets whacked so what does it mean to me personally and so when we take talk about precipitation which is highly variable putting it in terms that means something to the end user and even to you know working at FEMA and with emergency managers that can mean a different thing I'm not specifically looking for an address but I do want to know what critical infrastructure is impacted and what roads may be closed due to due to a lot of rain or a lot of flooding or a lot of surge and so that gets me to the math on the right try to find a surge map I know Noah is starting to put them out like 2448 hours in advance pretty crude this is a probabilistic one and they quickly stamp guidance on it and so if you really want to know it's the because we have high high Dorian had high tide they didn't know when the hurricane was going to hit relative to high tide that kept changing I'm in Charleston I want to know how many feet above the seawall the surge is going to be oh and by the way surge doesn't even talk about the waves on top of the surge right so we really don't know how to prepare if we don't know it's over topping our dunes or infrastructure in some way so I do go I did add this slide this is yesterday we have big nor'easter coming through this is this is at the University of North Carolina coastal emergency risk assessment this is not official anything you have to accept that you understand that to get into their website you have to know where to get into their website but it's great it's really good information so they use a high resolution ad-circ model it doesn't run quite as fast as the weather service slosh model but it has more details and it's almost at the pace where it can keep up with the next hurricane forecast so a little different with nor'easter you know have the updates like the same kind of updates but you can scale in you can pan in on this pretty pretty close now I would argue that once you get to a certain place that it's going to be inaccurate you know so let's talk about my next favorite subject so I'm kind of panning down right we kind of looked at global regional now what is the flood in does everybody know what the flood insurance rate map is okay so this blue outline is not a surveyed line because if it was we would feel pretty comfortable about it it's based on a bunch of probabilistic stuff right and then it's narrowed down to a point a line a base flood elevation that delineates the line and so people look at this and they're either in the flood zone or they're out of the flood zone and oh if I'm out of the flood zone I don't need insurance and I presume this to be accurate up and less I find out that I'm in the flood insurance zone then I'm going to pill your map because it's crummy data right so anyway so I mean and spoke to me before the mapping the zone kind of hits a couple of points really one is it was a report done by the national academies but this whole idea of more detailed elevation data is important but no matter how good elevation data is this line implies that this hundred year flood is a single value and it's a probability so we update these things but not as frequently as we'd like because there's not enough money in the program there's billions of dollars investment in flood maps around the United States for things they shouldn't be used for because communities don't have anything better to use now you go to New York City or someplace else they got their own stuff because they want finer details they're doing large scale development Washington DC has more information but you for small communities this becomes their planning tool this is what drives floodplain management in 22,000 communities so we need to be good at it or we need to communicate what's not good about it so that's kind of where I get it and I do have there are this is just examples depending on the floodplain manager and their resources they do provide more detailed information I love DC's flood risk tool I can type in my own address and I can zoom in and out and see what the flood zones are but gosh that alone is confusing what's an A zone versus an A zone versus a B zone versus an X zone and what does that mean and I can't tell you how many friends and family call me and ask me to explain it and it's pretty hard and I kind of know what's under the hood and it's hard to explain so I'm going to shift gears to my big idea after but first I'm going to introduce this because I actually I'm actually your second string speaker the first one was going to be David Alexander who leads the flood apex program for the Department of Homeland Security Science and Technology Directorate and I've been working with that program for the last five years and doing projects but managing external review board but it's been a really interesting program and I bring it up because if you're looking for good geospatial tools and new state-of-the-art things it would still be worthy to have David come back and talk to you but this program also helps study and fund something I've been working on the past year so as a former hydraulic modeler and administrator of things that depend on models I've been thinking about this for a long time what is the quality how is a community of practice do we evaluate our own stuff so I'm going to kind of shift from the map but to the data that goes on the map that's based on uncertainty and presumptions and assumptions and modeling and so I had a two-part study one the Coastal and Hydraulics Lab of the Army Corps of Engineers helped support and I started out and then the one that S&T is supporting I've interviewed I don't know 70 or more people hundreds of references but it's and so I've kind of been collating some ideas on this and so you're like any pigs see which thing so what do I mean by evaluation well I'm calling it this when I talk to people I didn't want to I didn't want to set I didn't want to say what is your validation process or how do you benchmark I wanted to understand what they thought quality in the model meant so I think it's kind of any process or check box or that somehow give you some information on the quality of this model the software, the data the performance some of these methods that you use and Mary is going to talk about Hydraulic Ensemble Forecasting. Ensembles help produce uncertainty so there was new methods I learned about methods can be qualitative and quantitative and so qualitative would be does it have and this is kind of commonly used stuff particularly in the core does it have a is it documented is there is a user's manual was there was it peer reviewed those are not necessarily quantitative things but but it does it is their training provided is there you know software version updates those are kind of qualitative quantitative or far more can be far more technical and I could deep dive into that but I'll spare you the point is that that anything we do to show what we've done to evaluate these models gives us more confidence in it and it also helps us define what the uncertainties are in these models so if if I haven't convinced you that you need evaluation strategies that this community of practice and by the way there's no standard I've talked to all sorts of people that do similar types of modeling and it's kind of left to the individual modeler maybe the agency maybe not maybe the acquisition is requirements drive it maybe not so there is where in other communities of practice there are third party certifications and verifications there is not one in our community of practice for what I would call hydro analytic sorts of stuff so if that doesn't convince you then what I've heard from people was what might make them think about it is well well our organization is world class we've always been the best modelers did you decide that yourself because best best of class isn't how you perceive yourself it's how others perceive you and so what are you doing to convince others that you're still best of class you know you might have gotten a mutual and star for your restaurant a few years ago but you know it's not any good anymore so let's you got to keep up we have a new class of problems so what before might have been easy enough to do a biology model or maybe we're just looking at a coastal wave model now we know that really what we need to do is talk about compound events the ocean meets the river and when these big hurricanes come in we have a lot of phenomena going on and these models are typically done independently we have world class problems about how we connect this very technical physics based kind of stuff to socio-cultural data to ecological data we have world class problems on how we look regionally globally and so we have new challenges that the same old way of doing things isn't going to work you can always if you have some idea of what the uncertainty is in your data in your model that also can drive where you put resources where you're going to spend money to reduce uncertainties or it may decide like on global models if we wanted to spend our money on Ho Chi Minh city, Bangkok, Shanghai you know the we can prioritize what's important resource line and then I would say disruptors in science and technology LiDAR was a huge disruptor GPS was a huge disruptor GIS was a huge disruptor and now we've got machine learning AI a lot of new things we haven't gotten our arms around the quantifying the uncertainty in the way we do our physics based stuff and now we're going to throw in just some data stuff that we don't know how to quantify and the most important thing that needs to happen is kind of this idea of transparency and collaboration if I say my map is made on best available data well you made the map in 1984 is that still the best available and what was best available data then is it still good today so we've got to be able to communicate that and I'm not going to go into this but I'm a lumper and so when we talk about evaluation strategies we can talk about the techie things and that's if you talk to modelers and mappers I know that we want to get right into the technical stuff you know kind of what the what the numbers were and that requirement but what's the most important is actually the user requirements and then if you want to get it done you better have some institutional framework to do it so just flood analytics what's it used for well we tend to think oh well we're just real time forecasting okay well that's one thing there's probably another this many more things we can use this for flood plane management insurance rate investing prioritization in the UK they use their national risk assessment of floods to prioritize where they're going to invest in infrastructure repairs and improvements emergency planning strategic planning project design I want some really good stuff if I'm designing a seawall or maybe putting a nuclear or power plant somewhere near the water I better have some really detailed information so use is important real-time forecasting I need it now I don't need it next year so you've got to think about the timing the level of quality that's needed for each of these uses and that can be different so what's next on the technical pieces I mean that's easy place to start I would say part of the study for the coastline hydraulics lab was they're going through a process called numerical technology modernization and they're looking to employ things like validation verification and certainty quantification into some of their hydrodynamic codes that's a big that's a whole new area of study these days BBUQ so a lot smarter people than me can figure that out but there's test beds we know in the weather service and others that we use those to help validate what we do and I loved I love the idea that using meteorology of skill scores and that kind of works in real time forecasting because you have every day you make a forecast you have another data set to prove if you did it right or wrong right and so then you can decide how skillful you were at predicting but if you're doing if you're doing probabilistic sorts of things like flood insurance rate maps you don't get a chance to you know it's hard to find out because the 100 year event doesn't happen as 100 year event maybe it happens as something else and it's not a smudge so you don't get to prove through skill scores but there should that still should be a consideration there's no data there may be for mapping or LIDAR or specific individual things but there's no real data modeling standardization the certification process that I would say that's in this this area what kind of performance do you want and then I'm going to this is my last next one is my last slide so when I went to the UK and spent I went to Del Tars last year spent a week with them talking about how they do their modeling I spent a lot of time with weather service people a lot of time with core people and other random people that would would talk to me but the UK they're moving out and there's actually two different I talked to environmental agency people met people modelers you know all around there's kind of two levels of things going on there but what what the big idea is kind of embedded in this what I've got next and so you know that remember the weather service guidance this is under construction I'm just bouncing off some ideas just on what the construct might look like so if you had we know that even if you get the technical stuff right so in the past FEMA primarily looks at where the highest risk areas are and when the last map was to determine where its next investment in mapping is but it really doesn't consider things like current community resilience you know remember I said places like New York don't need your updated flood insurance map they've got better data but communities that don't have it or communities that are rural or that have never been mapped so you need to build it can't you can't just have ideas based on one technical thing so I have this idea that you could use and they're doing this in the UK they have a confidence index that they're building and it's just for the data input model performance but they give under each of those categories they have sub categories you know the modeling input how accurate was the precip weather how accurate was the elevation data what were the initial parameters how were they validated and then performance didn't meet certain you know marks and high performance computing you know is it efficient you know all those sorts of things and then for each of those elements they rate them whether from good you know from one star to bad or poor to very good and then they have a composite score called a confidence index and that way they can then tell the public that your map your model area and your region is a two or five if it's a two this is probably good for maybe some general planning activities and to consider you know where you might want to make further investments in data if it's a five man maybe you can go ahead and use that you don't need to invest more and reduce an uncertainty you could probably use this for most of the things that you want to use it for so it's just an idea so but I didn't want the rating just to be solely on technical pieces I think there's aspects of community resilience like if you look at societal the social vulnerability index that Bill and I were talking about earlier what is the capacity of the community and its users and then we we don't think the maps for instance really aren't funded to consider climate change even though they should we have an option to actually okay sorry I'm almost done anyway so and then economic development could we rate how these these might be used in quality of life so it's just a concept I'm starting to pull together some of the details up under there just another thing to think about today not suggesting that we adopt my concept just how do we frame this so that so we actually have a moment if anyone has a what we're trying to do here is any clarifying question like a question that was unclear or piece of information that you need to sort of you know think about the presentation maybe just one or two quick ones and then we're going to move on to the panelists so we have one here just a question about data and models and how we're defining sorry introduce yourself sorry Dan Brown University of Washington a question about data and models and how you're defining them I like where you've gone with the thinking about assessing and how they're useful data and models as we move into questions on the right hand side of this last slide economic development human response human vulnerability that implies different kinds of data and models than we started with which was terrain models and hydrologic models which are useful for engineering solutions but not necessarily for social solutions so just a question how do we define the necessary data and models in this case I think that is the question right and that we know I said that we're getting we have complex problems now and they can't be solved by just looking through a technical lens so I think in some cases we do have some of this data we do have the U.S. global climate change assessments and if they were incorporated into our precipitation forecast or hurricane forecast we do have some information on where things are developing you know where cities are growing you know Detroit or some other small city you always see that in the news there are there is some information out there and the social vulnerability index is really quite widely used now and so it may not fall under a B.C. ratio benefit cost ratio and so we got to get beyond that and you could add weightings you could argue what the index could look like which parameters you would use but that's a good point okay great one quick comment if you don't mind Harvey Miller from Ohio State and mapping science committee I just want to point the audience to a MSC report that we produced in 2009 called mapping the zone which actually looked at the inadequacy of land elevation data to determine if buildings need flood insurance so it's a 10-year-old study report but I think it's still very relevant to this conversation I mentioned that and it has stayed on my desk the whole time it was still in my office yeah great I think we'd like to move on if that's okay but certainly sir keep your question and we'll bring it back into the larger conversation but thank you so we're going to actually move in through now the two panelists so the idea of the panelists would be to both sort of provide their own insight into some of these questions further some of the conversation broaden in some cases narrow as well so we have two experts we'll have first Mary Malowski who currently serves as the chief of water resources service branch in the analyzed forecast and support office makes long title of the national weather service so she oversees a lot of the weather service plans and policies and water resources warning and forecast operations so bring stuff right from the the very sort of front line so Mary about 15 minutes and then also clarifying questions as needed do you want me to stand up there or sit here the chairs either okay thank you okay thank you and it's obvious alright hi I'm Mary Malowski I'm chief of the water resources services branch I realize that I started my slides more in the where we're going and not so much the where we are today and so I just wanted to take a minute with how we communicate flood risk how we analyze and communicate flood risk today and primarily it's use of polygons and points we have polygons flood warnings or flash flood warnings and we have point specific hydrologic forecast data at rivers so we have about reducing daily forecast that that we provide a river level and the river flow for the next either three to seven days depending on how confident we are in the precipitation forecast the polygons are primarily we're comparing the amount of rainfall to some of the hydrology that we've modeled for that large basin hydrology if we think that it's going to exceed that we issue the warning for either a flash flood or an aerial flood so what what happened in we can go to the next slide right right button thank you wasn't so obvious okay so back in 2012 there were beginning to be a lot of issues too much water too little water poor quality water and congress recognize and there are 24 federal agencies that have some fingerprint in water and congress recognize that we could do more we could better serve the public if we work together work better together so they ask the iris agencies the integrated water resources science and services agencies which is the army corps engineers noa usgs and fema to go out to our stakeholders and ask what else what other information do you need and so we started this in 2012 we started first with being directed to go to the mid-atlantic river basin commissions and we're very focused on river basin commissions at that time so we went to sepcahana there petomic and new york the Hudson river after that they asked us to expand to other parts of the country so we went to the Ohio river basin and the Russian river basin and recognizing that what we were hearing was primarily river focus we also held a flash flood summit as the iris agencies to understand what the needs were for flash floods and what did whoops I hit that again alright I'll get it next time there you go there we go ok so what did we learn really we heard that flooding is still the number one priority that our stakeholders absolutely need flooding information at all time scales and spatial scales but they also need water quality information water availability information, drought information and they need that information all in the context of a changing climate and they need that integrated understanding it's not we were talking about having separate models we have separate models for flooding we have separate models for water supply we need to integrate that because our users are making integrated decisions they're deciding today if we build a community for 20 years from now how that's going to impact the flooding the water availability so they need an integrated understanding of both the near term and the long term risks and they want that in a very high resolution in space and time and they want that information to be linked to all sorts of intelligence information like infrastructure, economic, demographic and environmental information because what they ultimately want is actionable information information that they can make decisions on that can help their communities we're going to get it right this time ok alright so that led to two major investments on the weather services side one is the national water model and the second is the hydrologic ensemble forecasting service so what is the national water model it's a continental scale, high resolution in space and time hydrologic and hydraulic water physically based model of water parameters and it gets us that integrated understanding it's good from droughts to floods and it's everywhere it's based on that NHDPlus infrastructure and so it's being able to commute those water resources parameters at every single grid scale down to 50 meter grid scale so what you're looking at on the right there is the red points are those existing points where we provide river forecast and the blue are the streamlines for where we can now produce information based on the national water model so this is based on what our stakeholders were telling us they were telling us they wanted this integrated understanding they wanted it from floods to droughts and they wanted it from summit to sea so we began implementing this in August 2016 and as I said it goes from floods to droughts and it really provides that information at those underserved locations, those locations where we don't have forecast points today where the red dots are there's a lot of space across the country that is not served by those red dots it produces a spatially continuous national estimate of all of our water resources parameters so everything in the hydrologic cycle needs to be represented in this model and it implements the modeling architecture that allows for rapid infusion of new science we're still working on that we're trying to get into more of a community based model where we can have more of a sandbox with a plug and play capabilities but the idea is to get there so that academia and other components other federal components could then help to improve the national water model and it really is the foundation for improved growth we know that if you look real close we don't get very close to the coast we don't have that coastal connection but what this provides the national water model provides is a capability in a consistent way to connect to ad-circ models to be able to represent the coast and the river green components together the national water model to meet a variety of different users is run on a bunch of different scales it's four different scales and those are the rows that you're looking at now so the first scale is an analysis scale it's the current conditions and it basically looks back either the last three hours or the last 28 hours or the last 12 hours and builds to the current time and that provides the best understanding of what the current conditions are the next cycle is a short range forecast so we produce this hourly and we go out for the next 18 hours and so that's going to inform more of your flash flood decisions much of that fast responding the next is the medium range which we run four times a day and it goes out for 10 days and so that's more of the flooding in the next three to seven days what you're concerned about and then the final one is a daily daily model a daily run that goes out for 30 days and that's used for more of the water supply type of questions that we're trying to answer thank you the other investment is the hydrologic ensemble forecasting service and what you see here is for those 4,000 locations that we today we provide river forecast using a lumped hydrologic model what we're trying to do is account for the uncertainty we know when we put out a river forecast for the next three to seven days there's so much uncertainty in the rainfall in the hydrologic models that we've used even in the observations so what we're trying to do and you see it on the left the left there is show the range of possibilities for a forecast and so for the locations in green we have currently implemented the hydrologic ensemble forecasting service the locations in purple they have been implemented but they're displayed slightly differently because it's more of a water resources customer out in California and not the 10 day forecast and then the locations in blue are where we're working to validate the techniques today and we're continuing to expand the hydrologic ensemble technique over the next the next two to three years where we hope to have a base coverage of the entire country so what we've learned though is now we have a we have a fire hose of information between the national water model and HEFS there is a lot of information and how do we make that useful for our stakeholders so we went back out to our stakeholders to ask that very question and we went to in 2017 we did a number of forums where we had integrated stakeholders so we had a spectrum of water resources managers, emergency managers fisheries different people and then in 2018 we went to much more partnered specific topics so we just went to emergency managers and asked them what they thought of these products and data and we talked to our own forecasters of what information are you communicating what are some of the gaps in that information and what we came out with is a logic model and it's a very busy picture but what we're trying to show here is that we need to provide information if you're looking down the columns at three main categories if you look all the way to the right we have to provide information for the high flow flood risk that's the number one piece of information that people come to the weather service looking for that middle column need to provide low flow information and then also we need to have just general map services forecast that we provide every day for maybe a navigation or a water supply they need to know what the forecast is regardless if it's flooding or drought because they're making decisions on how much to load the barge or how much to operate that reservoir we need to provide that information now I'm going down the scales we need to provide it at national levels we have national customers like FEMA position resources all the way down to neighborhood details where we have an emergency manager deciding what road to close and how to have actionable information at that level as well and also if you go across the bottom we need to provide this information at many different timescales we have users who need it in the next 18 hours because they're making decisions on do I close this route out to the next the daily time step to the weekly time step out to the seasonal time step where we're making decisions and the other pieces of information in those blue boxes are the pieces of information we need to communicate with our forecast it's not enough for us to put out a stream flow forecast our users want to know what drove that forecast what was the precipitation that went into that forecast what was the soil moisture how did does it they want to do that intuitive does this forecast make sense for my application thank you so now we're trying to flesh out that that logic model and try to put in some of that information that we provide today and what information is coming in the future so now I'm looking at if you remember the logic model all the way to the right at the national scale so I'm talking about flood risks at the national scale and so we have a number of things that we can do today at our point forecast we can provide the observed conditions at our point forecast we can provide a we can provide the forecast over the next three to seven days we can provide forecast at those points for the next 30 to 90 days and then we also have the seasonal flood outlook which is their first guess our first look at what's going to happen this for the spring flooding and be able to communicate that at very national levels and then we see some of that national water model filling into that information so we have a model soil moisture of the current conditions we have an 18 hour stream flow forecast or a 10 day stream flow forecast or a 30 day stream flow forecast and how that can fill in and provide other information and then if we're looking at that logic model down at the local neighborhood scales there's a lot of information and as Sandra was saying that flood and nation mapping is the number one information so we have we have techniques today at 150 locations where we use very similar mapping technologies to FEMA with hydrologic and hydraulic models to provide flood and nation maps we also can go to the new national water model and using a more approximation approach be able to provide that for across the country and that's what we're beginning to see as people are now using the height above nearest drainage it's not nearly as precise as what FEMA has for the regulatory maps or it's not nearly as precise as what we can provide with the hydrology and hydrology hydraulics but it's a first guess and sometimes when there's nothing else that first guess is very meaningful information we have our hydrograph that we provide at 4,000 locations for the next 3 to 7 days we have our ensemble forecast providing that range of possibilities so what's next for us and where we're going is trying to validate these different services and see what is what provides our users with that actionable information what makes sense what helps them make the decisions that they need to make great so I do I'm sensitive to the time so we wanted to try to start discussion at 1130 but if there's a like a burning clarifying question we can pause for a moment otherwise I'd like to bring Burl Amantz on let me do that if that's okay so let me introduce Burl Amantz I've known for a long time a fellow geographer she's at the department of geography planning and environment at East Carolina University and she's the co-director of the university's natural resources and environment research center written widely on a number of topics Dr. Montz's work particularly in the context of natural hazards looks at the question of the effects of flooding on property values perceptions of risk and responses to warning and vulnerabilities of communities so certainly kind of wraps on many of the issues that we've already been touching on thank you and I'm delighted to be here I'm hopeful that the social science perspective adds to the more scientific natural science hydrologic science and so I'm going to be talking about as you can see vulnerability perception and response which are really broad topics I'm going to try to do it in 15 minutes bill will shoot me if I don't so I will I'll get it done but they are very broad and complex topics so bear with me as I just sort of scratch the surface at the outset to recognize that both vulnerable yep vulnerability and perception lead to response for better and for worse and that's what we're going to be getting at so when we look at vulnerability we think about the physical environment so we talked about exposure already exposure living in an unsafe place so you can see the houses that are pictured here in Norfolk Virginia smartly they have elevated some of these houses in order to reduce their exposure susceptibility is a different issue with respect to the physical environment because people who live there not equally successful susceptible to harm these folks who've elevated are less susceptible to harm than others and we'll talk about I'll just talk briefly about each of those again and then vulnerability is also defined by the socioeconomic situation where individuals and communities however you define communities are not equally do not have equal coping capacities nor adaptive capacities and I'll talk about some of the factors that influence that so we know flood exposure in the United States this map of cumulative claims of insurance shows where we have high flood risk where there's a lot of flood insurance claims there's a lot of flood risk we know this right we know where the exposure is it may vary locally it may vary from tight obviously varies from time to time but we have a good sense of what our physical exposure is I'm thankful to Sandra for mentioning Newburn, North Carolina because this is Newburn and while we know that exposure this is a very busy map I recognize but I think it gets that susceptibility because it looks at water infrastructure in Newburn as it is susceptible to storm surge sea level rise and riverine flooding and suggest that even if for instance those houses that I showed you that are elevated might not get flooded directly they may not have clean water they may not have access to their let's put it this way they may not be able to flush their toilets well so again that susceptibility is separate from exposure but and it so we have to think about it in a broader perspective and then we have the socio-economic situation and I listed here a dozen factors that influence vulnerability of individuals and communities the SOVI method was mentioned that's a useful method in mapping however there's two issues associated with that and that is the scale at which it is the spatial scale at which we map the temporal scale at which we map because communities change, neighborhoods change and so if we're using census data which I love don't get me wrong but if we're mapping census data in 2018 well does that really tell us what the neighborhood or what the region is like in 2018 and these do not these are not mutually exclusive criteria right so for instance age much of the research shows that the elderly I prefer to call it particularly mature adults and the very young so the older and younger are more vulnerable because of their relative inability to perhaps protect themselves that's a gross generalization on the other hand particularly mature adults may have more experience and therefore might would make better different decisions based on that experience so these are interactive and they will obviously influence one's both perceptions to an event or to the risk of an event and their ability to make certain responses so again I'm not going to go over those you'll see them again and I want to go to perception perception is a really complex criterion it's a really complex concept it consists of our judgments beliefs and attitudes and how do we measure those how do we find those out we know we can ask people but we know that sometimes what people say and what they are really thinking or what they're going to do are very different so we have to think about those but obviously how we perceive our risk is going to have a big influence on our response now recognize that our perceptions derive from cognitive factors how we view the world how we view our risk and I'll talk about a few of those in a minute but also our situations right we may view ourselves to be at risk but we don't have transportation to evacuate for instance or we don't have anywhere to go so while we may perceive ourselves to be at risk we don't take action because we can't so vulnerability and perception are integrally related and both of them are influenced by knowledge and they influence our knowledge so we do know things so that's a whole other layer on this diagram but both of these influence response I'm just going to dig a little deeper into the cognitive factors so just to give you an example of psychological element within the cognitive factor the locus of control is a concept or a theory that was developed in the 1960s by Gottman and it still sort of holds today the locus of control is one where some people tend to be more externally oriented where they sort of believe that they don't have a lot of control over their fate and others are more internally oriented where they tend to say I can control this I've got this going now it isn't quite as black and white as I suggested but I think that it gives you an overview an attitudinal we have different outlooks on life some are more optimistic than others some are more pessimistic than others some of us are more risk-averse while others are more risk-taking those are kind of our outlooks on life and we also make decisions within certain biases so some people are really wedded what is the wrong word some people have the anchoring bias where they rely on the first piece of information that's the one that really grabs them that's the one they're going to make their decision on or a confirmation bias where you focus on the information that confirms what you already believed and there's a number of others I won't go through all of them but I think that this table that came from some work done following Hurricane Sandy by folks at Yale and other institutions looks at that if you just look at the the difference between the 21% who were the first out and the 22% who were the diehards think about their attitudes toward their outlook their biases their locus of control so again I think that this really gives us a good sense of how perception plays out in our responses and of course these are people who had the opportunity to evacuate these are ones who had the choice their vulnerabilities did not hold them back from evacuation and so we're looking at response so I'm going to deal with warnings I don't know about messaging because people don't take action unless they're told to mostly so we know that people have trusted sources and Mary I promise you the Weather Service is truly a trusted source one of the best everybody says that having said that the role of official advisories is limited people always look for additional information so they might go online to look at the Weather Channel to see if what the Weather Channel is saying is the same as what the Weather Service is saying friends and family always have something to say about this or people will rely on what they say you might call your family member who's the weather geek and ask them or the one who works in the Weather Service and then there's the environmental cues for a long time no matter how much technology we have people use what my colleague Eve Grunfest and I call the stick people go out and look at either the stick or the rock in the river doesn't matter doesn't matter what's going on what the Weather Service is saying they're going to go see where is the water on the stick or the rock that they're used to looking at and how fast is it rising that's one of the things on which they make their decisions so one of the issues is we need the right messages through the right channels to the target audience yeah that's way easier said than done what is the right message what are the right channels those are the things that we're still trying to work out and the message alone again is unlikely as per action those environmental cues are really important and that doesn't matter whether it's floods with tornadoes people go out and look at the sky because they get a tornado warning like whoa sunny hell why would I worry about this so again those environmental cues and the influence of family and friends is also very important and there's a lot of talk about the cry wolf syndrome if you will it sort of exists but it's overstated much of the research shows that you know people do look for other information people do take environmental cues so while maybe cry wolf some people say well I didn't do it last time it didn't happen last time the way I said and in fact that might have been a problem with sandy and Irene some people did say that well they got it so wrong with Irene they must be wrong with sandy but a lot of people didn't say that so again it exists but it's probably overstated and complicating factors this diagram is like my first map of Newburn I'm sorry but this shows the diversity of decision makers this is a diagram that comes from some work that we did funded by NOAA a number of years ago that looks at the center of it is emergency managers and once they get a briefing from the weather service this is how it goes out these are all you know it's like that telephone thing if we said something and went around the room well this is what happens then and people take that briefing or the information that they get and some take things out some add things to it so think about how that might change as we have this diversity of decision makers with a diversity of needs obviously there's uncertainty inherent in the warnings so how do we present that uncertainty and the hydrologic ensemble forecast system I think is a really good way to get at that the role of lead time we've often talked heard that the longer the lead time the better well sometimes that isn't the case the longer the lead time well now I can go by beer now I can go you know we talk about winter storms right and we have what we call French toast syndrome you go get milk eggs and bread and so you know a long lead time might lead people to do the wrong thing on the other hand think about the diversity of decision makers emergency managers need a long lead time we talked about storm surge for instance emergency managers in the outer banks of North Carolina want to know what the onset time is of tropical storm force winds because they have to have everything evacuated and all their equipment away by then so they need a really long lead time others maybe not so much and obviously the context in which the warning is received so where are people what are they doing what are they thinking what are their responsibilities so some flat factors influence response so we know from the research that those that influence response positively is that trusted source of information higher income suggest a more positive response to take action because they can they're less vulnerable they have the they don't have those vulnerabilities that say they can't evacuate and so forth having children in the home for which you're responsible and having strong social connections and networks just some of the things that suggest positive responses to warnings for instance negatively the sense that the home was safe every event is different but that experience may work and we'll see that experience works both ways traffic concerns for evacuation for instance people are concerned about delayed reentry or they're going to caught in traffic anyway work responsibilities my boss won't let me go I have to be at work and pets and while we've had some initiatives that have tried to help have shelters where people can take pets some people aren't either not aware of them or still aren't going to aren't going to go so we know that those negatively affect response and then here's the great news it might go either way experience well I've experienced the big one I can handle anything or I experienced the big one I'm never going through that again the estimation of the threat what are people's perceptions of their risk the cost versus the benefits and that is the perceived cost versus the benefits like the traffic concerns and delayed reentry versus having your kids at risk so what what you're going to do your location information how you understand your location compared to what you're the information you're getting and then friends and family can help you or hurt you I'm sorry so what are the takeaway messages the message matters but some won't hear it because they're not listening or they don't have access to the information some won't understand it maybe because of the language that's the technical language it's used or it's not in a language that they speak or that they understand and again we're making strides toward that but it happens and some won't hear it and there are many reasons for not responding on some can't they have vulnerabilities where they simply can't take actions and some won't because their perceptions their understanding than their experiences may lead them to say this isn't a big deal I don't need to do anything and they have might have a lack of awareness of the risk that they actually face and then there is that whole issue of hubris where I won't go there and in a true academic here's my references and thank you great great thanks so I think in some ways that the three presentations I mean again sort of the objective here is really to sort of think about some of these thread points between the mapping sciences committee and the geographical sciences committee and I think these three presentations together sort of illustrated a lot of those interesting potential areas of overlap or existing overlap and opportunities for others so we have until noon for some Q&A what I'd like to do is to throw I think if the chairs co-chairs are okay maybe take two or three in a bundle and then sort of rather than one by one just sometimes feel that you get more questions and then the panelists and keynote can reflect on them so maybe while we take two or three immediate comments and then the panel can respond so gentlemen in the back sir and if you could introduce yourself thank you my name is C.K. Shum from Ohio State University I'm a geologist we're on sea level and cause those zones this is very open eye-opening very interesting for me to see how geospatial geodesy actually could extend to so many applications so I have actually I'm a technical person so I want to ask like two technical questions the first one is for the first talk very interesting so if I like to ask for example commenting so this measuring of the digital elevation model right so I think our technology will require us probably to better forecast strong surge sea level rise you will have to measure the changes the changes with adequate spatial resolution along the coastal regions for example and in the ocean I think the first priority against technical comment is probably to do hydrography and measuring the perfirmity and we're not going to be able to actually in my opinion and knowledge at this point to measure changes of the perfirmity for example sediment flows down the changes so your model will be off so the the second let's see the second question is just a simple question for the national water model it's very interesting I wonder whether you would consider groundwater groundwater actually maybe even transporting nutrients and some other phenomenon like for example near the lake with which causes harmful algal and so on so these are just technical questions thank you any other immediate questions to bring forward just to bring a cluster of things please introduce yes hi I'm Grady tool with 3d ideas member of mapping sciences committee question also on your presentation Sandra I think very very generally is excellent ideas to quantify uncertainty through the entire decision-making process if you would but I think what you were getting at was essentially to ultimately put uncertainties on these on the actual maps right in the in the end and I'm just wondering if that doesn't sort of transfer the problem from where we drew the line as to what the what the height was to what the probability of of flooding actually is and then how does that get resolved so a flood flood insurance company is going to make a decision based on a probability is that is that sort of the direction you're proposing there's a question for you thank you and I would sort of translate also into Burl's presentation like the sort of ranking how would people perceive that you know as questions so there are three two from the gentleman on the back and this one here questions that sort of bring up a lot of the issues maybe we'll pause here let folks respond and then take another set OK I don't know that I have answers so you folks help me I just told you I was going to be a provocateur I didn't say I had answers so the DEM piece going back to mapping the zone what we learned from that national Academy study was elevation elevation elevation right and so we know the more accurate the elevation information the better quality we have and it's also true and I agree with you on the first question things change so particularly coastal zones but also rivers and so even if you have good light our data thank you Grady it changes with every storm and so being able to get information before and after storms is a goal but it's not doable in all parts I mean it is it's not affordable in all parts of the world so even in places in the U.S. you can't always get the technology may not work but them a tree is absolutely that what you can't see is as important as what you can see particularly when you're doing hydrodynamic modeling right and so if we know that that I mean coastal surges really driven by the shape of the coastline as well as the but them a tree so we tend to funnel this water up through our estuaries and up into our rivers and and riverine modeling is really focused I mean I know national water model is high above near drainage and there's very little resolution inside the channels but if you want real information about where to dredge or where to place your port or where to where flood where to put flood protections you really do have to have detailed information I I think from that perspective there is a lot more that we can do I think we have technology that can do it it's a matter of how much we're willing to invest and that's why I think you have to think about these investment strategies of where it's more important to get the best data you can I'll and so I'll speak to the national water model and where we're going with the ground water and other advancements as I mentioned before we're trying to look into constructing the national water model into its components and creating a sandbox so that people can play in there and put you know plug and play a different model or a different capability and definitely ground water is one of the areas that we know needs improvement for the low flow estimates so that's certainly an area that we want to explore USGS is working on more of some of their ground water models and how applicable those would be for a national implementation we expect others to be more appropriate in areas so we do see some growth in that area we want to move towards modeling stream temperature as a precursor to some of the water quality parameters if we can get into that area then maybe expanding into some of the other water qualities like sediment loading and those sorts of questions the question of uncertainty do you want to go take that first that's certainly something that we hear as we're issuing flood validation maps that they want to know how confident we are in that map and for us it's a forecasted map so the confidence comes from how confident are we in the precipitation how confident are we in our models which are our approximations one of the things that we're talking about is stitching together a model that we would use our hand technique where there are no other models where there's nothing better like the Army Corps of Engineers models where they have projects they would have the best models at that time and so we would use inundation mapping from there and then we would use the more at specific river gauges at the USGS gauges where they have hydraulics and hydrology study we would want to be incorporating that as the best model there so kind of making the best best of maps map to communicate that flood risk and then doing other things like taking the maximum extent the worst possible case to certain people because we recognize the social science challenge of that the emergency managers want to see worst case but they don't want their people seeing the people in their community seeing that worst case scenario you want to react to that but actually this is about communicating risk I think you probably have the better insights but back back to the flood map when do I want to see a probability distribution on it I don't think so I don't think people will understand that so what I'm saying is though when we put a line on a map then it's like if you surveyed my lot it's like a property line to me you know I as a consumer of this product now believe that that's the truth and only the truth and so furthermore it doesn't really it really doesn't speak to your risk because the 100 year flood is only about the hazard itself not the impacts of the consequences so if we talked about having a consequence map I don't know what that looks like maybe that would be different or just forget the probabilities at this point and just say if you had if you know the forecast is going to be a foot above mean high water level whatever that is then and then it's going to be in my back yard so you know the depth maybe you just put depth of flooding that could occur for certain events but I think to the simple public not that the star rating but we are used to that right travel advisory if we had some simple way to start thinking about it now I'll also flip that and say FEMA and I don't know what's under the hood yet they're just rolling it out but risk rating too that FEMA is putting out is about the perspective for each home so it will be what the actual risk is for that home so even if I'm inside the special flood hazard area my footprint may be a certain size and so if I have a big home big expensive home I should have you know to pay more insurance right and if I'm above the base flood elevation I should get a break and if I'm so they're going to rate it based on the structures within the thing but I still don't think it speaks to risk it speaks to the hazard but communicating risk I think well I agree about probabilities on a map I think that that won't work although maps work and a haps is really helpful to have people understand their risk and not everybody a lot of people really who live in flood zones do understand their risk you know some of the work we've done in eastern Pennsylvania for instance in the Delaware river valley in Pennsylvania New Jersey they absolutely understand their risk and they have the stick and the rock and they look at the hydrograph and they look at a haps and they look at everything so they do understand it so you have the other problem is that it's not just when we're looking at decision makers it's not just the public versus the officials there's a whole range of public out there and they have very different perceptions and very different understandings and probabilities the other issue is that Mary brought up is the and I think we see this with the HEFS the hydrologic ensemble forecast system is the statistical uncertainty versus the forecaster confidence and they're two different things and so people want confidence they want to know how confident the forecaster is in his or her forecast they sort of want to know the statistical probability and we're trying to figure out how best to put that out there one of the issues is that the weather service faces as soon as you put something on the internet everybody has access to it so you don't want to confuse those who don't understand it as well as others so if you have something for emergency managers if it's confusing to the public that can create more problems than you really want so there are those issues but you're absolutely right about the flood maps too that I think probabilities won't work but there are some things where I remember I used to be at Binghamton University in New York and it flooded a lot in the region and I had one student who said well I didn't think I was at risk because I wasn't in a flood plain the map told me I wasn't in a flood plain this was a geographer so she was an undergrad geography major that worked great okay maybe we'll take some more questions so here and then come up here yeah I assume you need to yes I know Harvey Miller I'll issue an MSc I'm wondering whether or not there's other ways that we can use geographic information to help people imagine these risks for example I was reading a book recently called the Optimist Telescope by a former Obama administration official who talks about how do we get people to think more long term and there was an example where they put some people in a virtual environment and they got to see themselves older and then they asked them later on about preferences for saving for retirement and they had increased preferences attitudes towards saving for retirement after imagining themselves older I also heard something in the news recently about climate scientists are using VR as well to try to imagine well what will your city look like in 2050 so I'm wondering whether or not we can talk about just these probability maps and these flood maps to try to create imagined futures that really resonate with people as to this is what the world will look like in the future under this particular trajectory we're following okay Dan Dan Brown University of Washington I wonder I've heard recent reports about the disparities and inequities in flood response such that those who have more value on land get more payout from response programs and I wonder the degree to which certainly that's an element built into flood insurance programs and who pays who can afford to pay who knows but I wonder to the degree to which geographic information and the availability and perception and value of that information feeds inequities and disparities in outcomes okay great so I mean stretching the envelope a little bit both questions in some ways we also have a question online I think we want to take as well I'm Simon from Boston University and I'm very interested in the presentation of from Mary Mosulis Tai and I think for the national water model there is a big assistance with a lot of field observation data and remote data so my question is these kind of data from different apartment what kind of cooperation strategies between NOAA and like USJS and other institutes or universities and that means how certain data and field observation data integrity in these big systems and another question is what is the accuracy of the system because we have another for forecasting functions if this results can a concept for the people's requirements that means are there any successful cases can prove for this forecasting model the result is correct or not that is all my question thank you great thank you so the question is kind of clustered together in this idea we stretch out it's a different context different technologies to sort of look at some of the questions issues of equity and also of course interagency collaboration so nice thicket to go through and I'll throw it open to anyone so we'll start with the second one on the flood payouts and the inequities there and I'm sure these they can contribute too so it's true right and a lot of ideas because of these social vulnerabilities let's look at home buyouts that might be accessible to people first there's some really stupid rules with them with some of these programs good intentions but bad rules so substantial damage what does that mean you know it's a stupid 50% cutoff you know it makes no sense you if you got 25% substantial damage then you're not eligible but your neighbor might have had 50% substantial damage and they are and we're all in this together right we're a neighborhood we're a community we're doing buyouts one place at a time and it's not we're not considering the social cohesion of that neighborhood in that group so we missed that opportunity plus we wait so long so one of the reasons you know more affluent communities have more political savvy they can get their act together and come up with proposals to submit for grants bigger infrastructure that protects hot more important costly neighborhoods ends up with a BC ratio that's better than in poor communities and we've seen it play out everywhere but we certainly saw all of this play out in Puerto Rico and unfortunately people are not getting the money in time what money they're getting isn't justified now there's arguments about whether they even deserve it because it was informal settlements and not they can't show a title some buyouts have been halted because they don't have access to someone to do title insurance you have to move the deed over to somehow so not having access to the resources you need to actually carry out this information much less understand what the process is and how you get in the queue so there's a lot of problems and they're not related to maps they're related to the policies that we have been implementing for a long time having said that I think they could be related to maps in the sense that we know enough about where people live and what their risk is that we could map we do map who lives where once it's flooded we can say here's what they are and then devote the resources to those areas where for instance the marginalized or disenfranchised populations exist and send resources there to help them navigate the systems I think that's really part of it is navigating the bureaucracy in time to get the kinds of resources that are needed to recover or to move or to adapt somehow so I think that we have we can map it but why not use that mapping to say here's where we need to devote our resources rather than those who as you say have the political power and can pound their fists and say I need it now can I well I have this what the heck I got the floor a virtual reality I think is a great idea because we've got to get people to understand what their risk is and you know if the AAPS maps work virtual reality is going to work better I mean AAPS they can see what's going to flood under certain circumstances just in a 1D kind of interesting but being able to see it in a 3D would be very useful the trick is getting people to it's time consuming to get to do that but I think it's really useful and one of the issues when we get back to those cognitive biases one of the issues is a myopia bias where we tend to think short term and very narrowly and that would I think help us to avoid that that bias I think it's a great idea I just think it's really hard to implement and again maybe prioritize places where it's absolutely critical and of course for now until the technology advances I'll just say I like that idea too but I think when we added to our 4000 locations where we provide river forecast when we added impact statements to what would be impacted at what level it changed response it changed how people responded to that when we added pictures to that it increased it again so getting to that level making it as real as possible gets to that actionable information same question I love it took me a long time to go away from paper to online newspapers but it's fabulous because you get all this cool graphics and so we've seen emerge you know in Washington policy or times or these graphics editors that take our brilliant data and put it in ways that communicate to the public and I love it but I'm not sure and so I think they've come up with some smart clever ways and you will talk to some scientist and say well they didn't get it exactly right well that's okay they did communicate something but I would also argue that virtual reality the access to these kinds of resources not everyone has that and so how do we make that more accessible to people that don't have access to those sorts of things yeah okay so I wanted to address that third question which was the accuracy of the national water model and the data interoperability we've been working with the Army Corps of Engineers in the USGS for decades to do river forecast we cannot do river forecast the weather service cannot do river forecasting without the help of the USGS with the gauge locations and without understanding how the Army Corps is operating their reservoirs data interoperability is key for us it's been so for decades when it was when we were on the phone communicating information paper trials it's just getting better and better with big data with interoperability systems with data services we have access to USGS data in a much faster way the Army Corps of Engineers went to their core water management system which is more of that data services capabilities so I see that as a growth area the more we can communicate in data services the more they can communicate in data services the more we can make those available to everybody in data services will help the growth of that in terms of the evaluation we're always looking for ways to evaluate our models there's lots of opportunity with just observations but as I mentioned before the USGS it only has point observations at about 8000 locations across the country and now we're trying to simulate stream flow at 2.7 million stream reaches so millions of stream miles so it's hard to do an evaluation verification but that information is different than something that's never been there before and providing a service that's never been there before so we're looking for ways can we use satellites, can we use drones to evaluate some of our especially the legendation mapping but definitely having those real-time observations and to simulate it into our models is something we're striving to do great so maybe looking at the time we'll take these last two questions and then responses and we'll probably press up towards lunch but sir and then back yep if you could press your little red then introduce I'm Mike Tishler at the National Geospatial Program at USGS I'll be talking about the Arctic but a big part of my job is producing a lot of typographic information for the country so the 3D elevation program collects coast-to-coast that was stood up I think in response to issues like mapping the zone flood mapping and FEMA is a big partner of ours and we'll also produce a National Hydrography Dataset which fits right into the National Water Model so it's really fascinating to hear the conversation and I just want to make a quick comment about where we're going with both of those you mentioned an HD plus in addition to groundwater which is a terrifically difficult problem we're also working on how to capture the urban stormwater systems as part of the surface water particularly in urban areas where people live very, very complex and not all the cities have the same information and we've had a good successful pilot in Washington DC where they did have good data where you've been able to integrate the urban system into the surface water network with the 3D elevation program in LiDAR you're right, refresh and measuring change is hard and it's expensive but once over for the country it's supposed to be about a billion dollars we're working with NOAA on a study to figure out what the next generation of that will be incorporating more coastal information bathymetric data and inland bathymetric data across the rivers it all starts with a good justification and a return on investment so we can talk to the legislators and talk to other stakeholders to recoup that investment I think we've got a good model that's what we did with the 3D elevation program and we put over 400 million dollars over the past five years into LiDAR collection across the United States we want to be able to do that on the coast so that you can have the symmetry data with the systems and models that's just the one server though how do you do it again when you need it the inland bathymetry and the coastal bathymetry has a much shorter shelf life than topography in Utah last question before lunch this is really interesting and the comment about virtual reality and everything I'm glad McDonald from UCLA the comment about virtual reality and everything a little bit but when we present a map to someone, a user we're at least of a risk or hazard we're at least two steps of abstraction away from their usual reality which is when we're looking at the world in a plane view and then we're trying to then on top of that put some color coordinate it gets very confusing for people now I'm thinking about private partnerships Google spend a lot of money driving cars around that take pictures of the street right that's how people see their environment they see their home and for real estate you use that a lot in the MLS for the street view of your house and all of that it seems to me that we would be able to provide a model of where the flood water is going to be relative to those streetscapes and people could just simply type in their address without having to put on a virtual reality hat and see what their house might look like based on the best projections we have it doesn't seem to me that that is any stretch of current day technology and it's the way people are accessing information I could grab this out and I could see what my house is going to potentially look like now I understand there would be legal ramifications and things like that but there is a huge data set that's out there Google's put a lot of time and energy as has various real estate firms into these kind of georeferenced databases that take you out of the bird's eye view and put you on the street so just a thought Google was a partner to do something like that and I think people saw them going around they were giving out first floor elevations and that got had some legal ramifications so North Carolina does an excellent job they have detailed data and their structures and I don't know how they look at depth or what their websites look like but they have the information where people could look so I think we're close we've got to get past some of the legal issues if we had structure by structure information which don't need for risk rating too and for some other things that would be great being able to visualize it is another thing but also adding this real time forecasting back to the issue of what is five inches of rain mean in my house or in my neighborhood we're a long way I think from getting to that place we might can take a statistical storm or a statistical event and say well if you're if a hundred year event were to occur your water would be this many inches under or something but the more real time piece of it is a long way way Any other thoughts or reflections maybe to sort of close down the panel any sort of comments or questions otherwise maybe I'll put it back to the co-chairs to close us down for the day or the morning well thank all three of you very much for your presentations this morning and we're now going to adjourn for an hour for lunch we reconvene at one o'clock and does the staff does anyone want to comment about the lunch there's a cafeteria two floors up the elevators are just around the corner from where you came in and there's an open cafeteria as well as an atrium to enjoy lunch and then we'll reconvene in here at one o'clock it's on the third floor somebody will be in the room the whole time