 The big problems with knowledge that we have are a gap in terms of the quantity. We don't know pretty much where water is, either underground or flowing. We have an idea that a farmer might divert water. We don't necessarily know the return flows. We have the idea that a household owner might irrigate with water or flush the toilet. We don't know where the quality of that water is or how much is actually going back into the sewer system. I think one of the gaps that we have is gaps in water data. We don't actually know the situation in water quality. We don't survey very often or very widely. Even for the regulatory effort and for the subsidies that we're spending, I think we don't really know what we're really buying. In fact, it's even difficult to track in the water quality aspects what are the extent of the problem and what sort of are we buying in terms of reductions. There's another gap. There's a disclosure gap. Companies, at least some companies, disclose their water use and water pollution. But they do it in a way that hardly resonates with investors at all. They don't read these glossy sustainability reports that are full of beautiful happy stories and numbers, but there's no context so there's no way to compare or make sense of those numbers. And this is the key point that, you know, we've got to go local if we are really going to get our minds around water risk. We don't understand exactly how, as people referred to, how to factor in risk. Not just risk, but catastrophic risk into this equation. We talk about 100-year floods or 50-year floods. Those 50-year floods in many parts of the world are not in 20-year cycles. $15 billion have just been spent or will be spent as of next year in New Orleans generating a system that will supposedly defend New Orleans against a 1% chance surge event. $15 billion. And we know that 1% surge event is not correct. My second point about knowledge gaps is, yes, there is lack of scientific information in some cases, but again, it is how we are framing that information that I think we need greatest work on. So, for instance, in the context of climate change, what is the time frame? Other speakers have referred to this. How will disruptions present themselves? How do we take data and integrate it into an interpretive and explanatory system for policymakers? How do we present to such a way that it's actionable, as Pete referred to, for policymakers to make long-term decisions? The basic idea is this, that we track these different indicators and we aggregate them. You can aggregate or disaggregate them in a tool online that we've built. We've done a prototype for the Yellow River in China looking at the power sector. So, there's all these different risk drivers that you can either group in either physical, cost or disruption risks. You can aggregate them, as I said, and we've done so for 132 different geographic units within the Yellow River Basin. So, the best way to think about it is a water risk atlas, but a pretty granular one. And we believe that is the key. We've got to go local with our information and our knowledge. And so, it's really important that these robust risk management tools that integrate the water and energy intersection, as well as land use planning, that they get stronger, that they incorporate the tools that Pete talked about, things that WRI is doing, things that Jemi is doing, things that the World Business Council for Sustainable Development are developing. These are tools that allow you to look multidimensional. You look at the local level in terms of water risk management. You can look at the regional level, and then World Business Council allows you to look at a much higher level. To be able to evaluate, if I look at this series of water risk, energy risk, land use, ecosystem risk kinds of questions, I can make a better educated analysis of where my footprint is going to impact, and then what I can do to potentially mitigate those impacts. It's important for experts to work together. It's important to integrate water and energy planning. But we all know that none of that is easy. It just won't happen because we want it to happen. I mean, several of us have been around in the water space at least for decades, and we've been talking about integrated water resources management for a long, long time already. We've never really pulled it off because maybe it's too complicated. So I think there's a need, certainly a big need for a change in the way industry and government and the private sector work to solve these problems. But I think there's a big role to be played by the government agencies. We've got to figure out how to do that role. We're not doing it now. We need development of better decision support tools, like was mentioned earlier. We need integrated projects that work at the community or the watershed level to solve problems, to really get something done. We need synthesis projects and translational science. I would say in some cases we need just pure outreach, and this was talked about, projects that drive the knowledge that we have out into adaptation. And then last also mentioned we need some good old public education projects. And so to me the actual topic, it'll change, our priorities will move and shift. But the way that we approach these topics I think is something that we need to think about. So regarding the university role, I think it's obvious to everybody here that our primary role is that of training the next generation of managers, scientists, lawyers, engineers. That's our job. But hopefully also to train these students to work at the intersection of food, water, work across these sectors to think critically, to think broadly and to be lifeline long learners. And that's what it really takes to work across sectors. And I would submit to you that the problem of getting students to think across these sectors is just as difficult as getting agencies to do it. I mean it's a vexing problem, but it's certainly one that we have to address. I think number one, biggest issue that we're faced with is a policy issue, not a scientific knowledge issue. Our policies have been developed in very narrow silos. They typically do not address multiple sectors like water and energy and agriculture or food and environment. And I think if there was one thing I could fix with my magic wand, if I were Harry Potter, it would be policy. And it's very true that sometimes you do need to have someone in charge, you do need to have a rule, you do need to have a law, but you also want to try and disaggregate the responsibility for information and decision making as low a level as possible. And unfortunately that means away from a central federal role. This means towards more local control and I think the most appropriate scale of control for water is the watershed level. Regulators face a knowledge problem. This is basically the idea that they don't know what's going on and they don't know what people want. You need signals and signals of supply and demand which are based on prices. This matters in water because as I mentioned water is not priced. We have no idea of the value of water, right? So sometimes it's appropriate to have a price of water. I'm not saying universally we should have prices everywhere, but I will point out that there is a huge knowledge problem in terms of the value of water. What we need to do is set up a robust system where individuals managing their own behavior or their own access to water, their own supply and their own demand can coordinate with each other. And some people might have thought about this idea already. It's called a market, right? And I mean a market in terms of signals and information exchange. So at the tap water level, whether it's Washington D.C. or Chicago or Los Angeles, you want to have a certain allocation per person, something like 50 gallons a day, I don't know, for free or for cheap, and then additional water costs more. This is similar to increasing block rates, but the very important difference is twice. The first one is that you do this per person, not per water meter. We're selling water to people, not to meters.