 Thank you, Jeff. And thank you, Steve, for inviting me. So yes, I'm going to give you an overview today, really emphasizing three things about how the natural capital project is using environmental data and socioeconomic data. So we use big data in the small case letters, as someone said this morning, to really help inform decisions that institutions are making. So that's number one. It's just to show you what's possible in using big data sets around the world, socioeconomic and biodiversity data. The second point I want to make is that we're seeing some impediments. We're seeing limitations to scaling some of these really interesting innovations that we're seeing around the world. So we're trying to figure out how to get past one-off projects. So I'll talk a little bit about those impediments and then finally I'll share some ideas we have for a software development platform and a data server that might help overcome some of these impediments. So NatCap, as we call ourselves, we, as Jeff mentioned, were a four-way partnership and the really the ultimate goal of all of our science is to help institutions and individuals make better decisions that lead to better outcomes for people in nature. So our job as kind of the science engine is to get that information into their hands so that they can take those decisions. And we do that in three main ways. First, we, by listening to the end users, their questions inevitably drive advances in science and software and engineering. And we then create user-friendly approaches and tools once we've learned what kind of information and at what scales will really help them answer their questions. And then we build capacity through training and try to disseminate these stories so people realize it's not actually quite as daunting a task as it might seem at first blush. So at the heart of all of our work is this concept of natural capital and in the academic literature we call them ecosystem services and most simply put they're just the benefits that people get from nature. So they can be material benefits like food and fiber and fuel, clean water, protection from coastal hazards. They can also be spiritual and aesthetic and recreation kinds of benefits and then other services such as pollination, which obviously leads to a lot of the foods that we eat. So benefits from nature is really the common element in all of the spatial modeling that we do to prepare people to take different kinds of decisions. And our group really is based on all of our work where there are 40 of us at NACAP and the work I'm going to show you is based on a lot of people's work is really based on these really simple premises that people's lives and livelihoods as Jeff mentioned depend on nature. We know this and that understanding when and where nature matters most to people will really help transform decisions. So this is kind of the crux of the why behind what we do. So I'll give you now some examples. So this is the portfolio of projects that NACAP has done in our first six years. So we've worked all around the world and because we're kind of science wonks, we have purposefully chosen very distinct geographies and very distinct decision context to find out what is the different kinds of natural capital information people really are wanting. When they say I need this information, what does that mean? What scale? What metrics matter to people? And really testing this in different decision contexts from formal payment or compensation schemes, climate adaptation, government work in either spatial planning or looking at development impacts, and then corporate risk management is a growing part of our work. So really looking across these really different kind of test cases to see what is the spatial kind of information that people need and what are the data and software engineering requirements to get that into the decision-makers hands. So we have an open-source toolbox. It's called the Invest Tool Suite and there are now 17 different service models in this toolbox. You can go and download it. It's a it's a good open community people write comments. We have forums that support the users and the service models are some of them are mentioned here. They're all different types of benefits that people get and they're all driven. The questions and the design of the models was driven by these end users and what their needs were. So here's the final set of the models that we have right now. So you can see they're really diverse. They range from terrestrial and marine and coastal and all sorts of supporting services and regulating services. And they've all been tested in some of those dots you saw on the map around the world. And at the core in all of our models is this ecological production function. So it's based on economic theory and the most important thing to remember is that all the models are really trying to estimate if you change the ecosystem due to climate or human activities either on purpose or by accident. How does that change the services that are accruing to people. So how do people's benefits change if you change the ecosystem. And what we found in developing these simple models that by design we've we've really committed to saying any model we develop needs to be able to be run anywhere in the world. So we can't test it on New York City's coastline where there's tons of data and hope that somebody in Sumatra could also develop it and apply it. We want to really be sure that anyone could use it anywhere. So it gives us interesting data and computational challenges. But even these simple models that we hope are repeatable have pretty significant data and computing challenges. So I'm going to just go through three quick examples now to give you a feel for the kinds of interactions we're having with these end users. So the first one is to look at this problem of siting wind energy in New England. So this is the first case we've looked at is off of Block Island. We've also done a bigger scale along the coast. But what they've done here in Block Island is they had already defined the leasing areas for wind energy. So where would the leases be offered. So we could look at the cost effectiveness of that solution and then look at the other benefits. This is based on the state of Rhode Island and how that cost effectiveness of wind also could be optimized by minimizing impacts on other benefits that people see from this area. And one is just views. So Block Island, a lot of summer homes there. They don't want to be looking at these big in some people's eyes ugly wind farms. And they also there's a lot of recreational visitors out there. So this was the question that they sent to us. I'm not going to talk about the wind farm analysis or the aesthetic views. I can give you more information on that later. I wanted to just tell you about one of the services they asked us about. And this is what's the recreation value of people using these coastal waters. There's a lot of people out there recreating on boats. And this presented an interesting science challenge. So natural resource economics has always had really great methods for estimating the value of different places in terms of recreation, visitor days and revenues from tourism. But if you don't have counts, if you're not at a park with someone with a clicker, then you can't estimate the value of someplace out in the middle of nowhere, like in the ocean or in the wilderness. So this led us to innovate. This is led by Spencer Wood and some others in our team to look at this general problem where you want to really get at the visitation rates and how that decision to visit is a function of a whole bunch of things. It's often the built environment, how nice are the hotels, is there roads, is there a restway, but also the natural environment. How does that change affect change in visitation rates and therefore valuable tourism dollars? So this crux issue of not being able to get counts led Spencer to look at different sources of data. And he used a whole bunch of different things initially, but he landed on using flicker data. So the flicker data set, as you know, has tons, millions and billions of geotagged photos. And what he did is he took those data and related them to empirical data sets where they did have visitation counts around the world in national parks. And he found a very strong positive correlation between the photo user days from flicker data sets and the visitation rates. So he can then use that now anywhere in the world in our recreation model and ask how do the density of flicker photos relate to natural attributes of any area just based on habitat maps? And then look at recreation value. So here's Block Island from Rhode Island. You can see these are some of the results of the recreation model. Darker areas indicate higher value. It's a higher density of photos. You can see the ferry route showing up and also a hotspot for charter fishing. So the state is now using this in addition to some of the aesthetic quality values that I mentioned earlier together with the cost effectiveness of the wind energy lease sites to try to figure out which ones they're going to propose. And this one, by most people's estimation, is going to be the first actually in the water wind farm in the US, even though Cape Wind has been permitted for a long time. They're not anywhere near as close to putting pylons in the water. So that's just one example of how an interesting question led to new science using big data like a flicker data set. And now it's in a tool that people can use and they're using it in other places outside of New England. OK, so a second example now moving a little farther south, but still in the US. This is an engagement we've had with Dow Chemical Company. This is with the TNC partner partnership of ours. And what Dow was interested in and this one of their sites, we've worked with them in Brazil as well. This is in Freeport, Texas. And they wanted to know what's the potential role of natural capital in these three things that really affect their bottom line. First is how might I use natural capital, in this case, marshes to protect my physical plant from sea level rise. So this facility that Dow runs in Freeport has 20 percent of their global chemical production, so it's the biggest facility in the United States and produces a lot of their chemicals. They're also interested in the effects of trees in regulating air quality. So trees remove ozone from the air and they can affect their regulatory risk, whether or not that is functioning in the right way. And then also how can natural capital help them secure their water supply? There's a terrible drought in Texas, as you know, and they're very curious about that. I'm just going to talk about this first one for today just to give you a feel for some of these models again. So here's the coastal site. The blue areas are parts of their facilities that are currently, they think adequately protected by the current system of levees and dykes and marshes. The orange areas are not protected and projected sea level rise in the case that these will be inundated and exposed to greater storms. As you know, there's a lot of hurricane tracks that come right through this part of the Gulf Coast. So their question they asked us was, how might different levees and marsh restoration and the combination of the two, so could they get protection value from levees and or marshes together to reduce some of these impacts? And then also looking at scenarios of sea level rise and storm intensity. So things they can control like levee and marsh restoration and building and also things they can't control like storms and sea level rise. And the basic structure of this model is that production function I mentioned earlier. So you have these physical forcing factors like storms and sea level rise. Then land cover in this case, both the bathymetry and the topography on the shore, which can affect flooding and inundation and also erosion. And then you can value that based on the avoided damage that these habitats might offer in terms of ameliorating the flooding. They also asked us to look at recreation value and fishery values because these are externalities, as economists call them, or extra benefits to the public that would accrue if there were extra marsh habitat in the area. So again, this was another new science demand and a real meeting of the minds. As you can imagine, Dow has a lot of really smart engineers and really smart economists. So we had to make sure that our models could match with theirs. So we wanted to add natural capital effects on the biophysical side and also on the valuation side. So it was a great, great collaboration. And here's just a simple schematic of a model led by Greg Gwinell in our group that's now in press. And just to give you a feel for what this model shows. So this is a profile of the shore and it shows you offshore and onshore. And you can add or subtract a different area of a marsh or a mangrove or a dune or whatever biophysical habitat feature you want to look at. And then you can look at different physical responses to the presence of that habitat. This is showing scour. So in the absence of vegetation in this model run, you can see that there's much more scour, so much more erosion along the marsh edge than in the presence of vegetation in this green line. Oh, and I just put this in here to show you that we do a lot of sensitivity analysis and looking at how parameter values changes in those and the ranges that we introduce change the overall outcomes of the models because people really want to know how certain are you in these estimates that you're providing. And in short, what we found for the Dow free port example was in this case salt marsh restoration did not in fact by itself reduce damages in this area. But if you add marsh plus levees, then you could actually have you had a marsh in front of a levee that the height of the levee required to provide protection was much lower and the cost effectiveness of that levee was much higher. And so that a combination of marsh and levee system actually turned out to be a really good solution for them. It's also interesting that again this lesson that we keep coming back to that place matters just around the corner in Galveston, we did a similar analysis for the city and marshes were there. Marshes alone were more than able to protect city and people and property and infrastructure in Galveston from future sea level rise and storm. So it really depends on what the bathymetry is and the topography on land and how big of an area of these natural biotic habitats are there to in terms of what kind of answer you get for the protection value. And then we looked at some of these other values that they were interested in and so far what Dow has decided is that they're really looking at a deeper local site analysis of this combined marsh levee system to protect their physical plant. And they're thinking about using these sorts of social benefits as a way to talk about with the community what the benefits of a combined marsh levee system will be. Okay, and the final example I'll tell you is a newer one we have with Unilever. So the most important and exciting thing to me is that Unilever owns Ben and Jerry's, but they own a lot of other things as you know. They have a huge global reach in terms of the agricultural products that they source around the world. And Becky Chaplin Kramer who's here has led this work with us. So these are just three of the questions that we've been just beginning to start to do analysis and help them think about. So the first is how dependent is their commodity supply chain on ecosystem services? And second, where are the safest sourcing regions around the world to minimize harm to ecosystems or people? And then how can natural capital protect our water infrastructure investments? So I'll just show you a little bit about this supply chain sourcing question. And what's interesting with companies like Unilever is Paul Pullman, the CEO has been very bold and a very strong leader in making statements that they call their sustainable living plan in growing. So doubling their overall growth by also sourcing 100% of their agricultural raw materials sustainably by 2020. So that's a big bold leadership statement. And now they want to know what does that mean? Where should I source? And how will those changes in their decisions actually change the impacts on the planet? And here's just an idea about how big their reach is. Just as an example, they source 12% of the black tea in the world, in the market, that's Lipton. So they have a lot of questions about where to either sustain agriculture in an intensified way or to extensify it to spread it and what those impacts might be. So just as one example, again, here to give you a feel for how our contribution is adding to the way they think about it. So supply chain analyses now will often do things like what is shown here, where this was an analysis, very interesting and cutting edge analysis done to look at alternatives for margarine production. So what are different ways to source margarine and process it? And these bar graphs, you really just think about them as stacking up different potential sources, like rapeseed oil and sunflower oil, and their impacts on really important environmental services like reducing erosion potential or reducing variability in freshwater supplies. And these are really important cutting edge analyses, as I said. But what they don't do is they don't take into account where on the landscape these certain crops might be grown or sourced from. So these stacked kind of analyses really just take an average. So they say, if you're growing rapeseed oil in this area, this is going to be the impact on erosion or water quality or water supply. And the kind of models that we build really help them see where on the landscape this shows one of our models with the erosion filtration and nutrient filtration happening nearest to the stream network. So the impacts of sourcing from different areas is really going to depend on how close you are to the stream and what the slope is. And a lot of other biophysical attributes that the models that we have are going to help unilever see where in space might the sourcing decisions have bigger or lesser impacts. And just to show you another way, we're visualizing uncertainty for them. This is another service they asked us about which is carbon storage and sequestration. This shows in Brazil now some of the effects of carbon in one of these agricultural extension, extensification scenarios. Green shows increased carbon storage and sequestration. Red shows loss. And yellow means basically we're not sure or relatively in between. And if you look instead at the 90% confidence interval, the map looks like this. So black is outside the area of the confidence interval. So it changes the kind of information we give them. And here's the 95% confidence interval. So we're figuring different ways to visualize and show not only the answers that they're asking us for but also how certain or uncertain we are. Okay, so I just want to, that's the end of the examples. I want to just quickly segue to show you what we're thinking about to try to really scale up. So we've got about two dozen cases around the world. A lot of our colleagues have done similar projects in other places, but we're frustrated at the pace at which this is easily replicable. It takes people time and energy in each place. So some of the impediments that we're seeing are really transparent, accessible and updated databases and analytical tools. So just making these much more open and accessible to people so that they can use them. And then having the capacity and the training in people around the world so that they can actually run the models and interpret the results so that decision makers can start incorporating them much more rapidly. We see a lot of demand. We actually, unfortunately it kills us. We turn away a lot of people who come to us and want to work with us because we can't be doing more than about a dozen projects any given year. So we really want to try to build capacity so that more people can be doing this. And then communicating examples of this at the beginning just to show that this isn't that hard and that if people are interested which we're hearing a lot more about then it's not bad to try to do it on your own. So this leads me to this vision for this development platform that we're calling the Earth Genome Project. And it's named similarly to the Human Genome Project because we really see a potential to greatly reduce the time and cost associated with amassing and collating and processing data to get more into the right hand part of this system. So this whole scheme is what we envision to be the Earth Genome Project and it's a system map. And the main thing is that on the right hand side here is what we all want to do and that is to transform decisions so that people and the environment can both continue to prosper in the future. And to do that we think we need to get rid of some of these impediments on the left side of the system map. So all of the data aggregating new data but also a lot of existing data and processing and really getting it into a format so that it's easily taken up by models like many of you in this room probably run and really get this to be open and very interoperable and accessible. Then the platform itself is really gonna be, I'll show you in a minute, but it's a way to allow lots of different disparate data sets to plug in and lots of models or apps on this end to plug in so that there's really just much more open and easy access. Okay, I'm gonna go very quickly now. So what we have in mind for this architecture is first just to build off of what Rich Sharpe who's in this room and his team have shown already is how to build what we already do in NatCap. So we have our software tools and that is very actively run off of a geo-processing pipeline and there are databases and algorithms and other types of software that feed into that. We have a whole group of model developers about 20 to 30 of them. Then there's applications for GIS that sit on top of the software suite and the data inputs and outputs go beyond those and then there are end users that end up visualizing and gathering data, et cetera. So that's the current system and now these here show what some of the impediments are. So how do we handle big data? Can we run global data simulations? Can we really customize, invest, and other applications for end users so that we really grow this and not have it be so slow? And then how do we share data and the results and new applications? So this leads to the earth genome platform design that Rich can tell you much more about and we think that with a newer design, it really this database could be online. Some of it will be sourced locally. Some of it will be in the cloud. There'll be an API connecting the database to this overall database and third party generators can also be called in. Then there'll be a geo-processing pipeline probably in the cloud. Some of it here depending on what the functions that are needed. Then there'll be customizing the apps for lots of users. I'm just gonna run through these. Those boxes are really just showing the different suites of users that we envision taking advantage of these apps that could then be developed all open, all very open and transparent so that people can see what's in there. And I'm going to stop there because Steve is standing up. I'm just gonna close by saying we think this software sharing and storing platform will help but it also is gonna take a village because you can see there would be many, many partners and coalition players in that whole system. So I'll look forward to talking more with you during the panel. Thank you very much.