 All right, thanks Jeff and thanks to Steve and others for the invitation to be here and thank you all for for attending So I want to talk about climate risk and Let you know, you know what I mean by climate risk and I think that that the climate climate research community is at the cusp now in terms of of big data between past in which Climate has climate research has really been very accustomed to large large data sets and Large computational resources in terms of petabytes and petaflops but in in very sort of structured Structured data sets so that being the past where we could always used to say oh big data Yeah, well, you know, we have we have big data sets. Here's our here's our terabytes and petabytes and and now On the other hand, you know looking forward. We're really not I think we're now we're now at the point We're well behind really the you know this this emerging cutting edge of big data in terms of integrating all of the unstructured new kinds of data that are that are available so I what I'd like to do is Talk a bit about climate change and why it's a big challenge and why climate risk is is a risk and and what shapes those risks and then provide some thoughts about This question whether or not whether or not big data can can help us All right, so I really have four four messages about climate change one is that the global warming has already increased the risk of high impact climate events and that going forward the risks of further change and further impacts are are much greater if we continue along a business as usual pathway in terms of Carbon intensity of energy global emissions and global warming Compared with a world in which the United Nations targets are are met and and global warming is constrained However, there's a big challenge because Really climate change is a side effect of the benefits of energy consumption and I get tremendous benefits from energy consumption the The world worldwide we get get tremendous benefits But there's a big gap in access to Energy resources and most people in the world have much less access than I do and billions of people really are in an State of energy poverty and this is this is the real challenge that creates creates momentum towards higher levels of warming and Just because I usually get told that I'm a big downer I Have the word hope here. The word hope will come up come up in a couple places So I do think there's hope I just want me out front. I do think there's hope and I think that actually we have we have the Ability to be a real engine for for innovating the solutions that are going to be needed to both Ensure access to the benefits of energy consumption to all people on planet earth while Minimizing minimizing climate change If you want to put another way maintaining something like the the climate that we're accustomed to And I think that there's a potentially a big role for for big data and I'll come back to these four towards the end and And these kind of highlight both where where a climate research has been and and where a climate research isn't yet But but I think could be so certainly, you know in terms of in terms of climate Climate modeling weather modeling Analysis Huge computational resources have been devoted in this area certainly United States and Japan and Europe So you mean just as one example the One of the one of the most recent large clusters at Oak Ridge National Laboratory, which at one point was that was the fastest fastest computer on the you know the top Top 500 charts about 35% of that computer was going to climate modeling So I mean this is a big in terms of computational demand. It's it's big produces a lot of data But again, it's very it's very structured And again as I as I alluded to I think that there are there's potential for Further innovation in in in terms of data assimilation Particularly with emerging sensors and then also kind of non climate non climate data Such as we're seeing from social and search and geolocation it gives real potential to understand climate risks from those data sources All right, so global warming is occurring. It's an observation. This is the make sure I press the right Button here to make sure I get the pointer All right. I got lucky So this is the this is the global temperature record It's expressed as an anomaly from a baseline so each year is is up or down from that just that baseline mean temperature and You we're going from 1850 on the left up to the present day on the right and you can see a lot of up and down There's variability from year to year. There's there's variability from decade to decade But they're unequivocally This noisy series is tilted on a trend that trend is global warming So global warming is happening. It's an observation You know in terms of signal to noise ratio the it's like the Higgs boson When that was when Higgs boson news was coming out We heard a lot about this five to one signal noise ratio if any of you have seen the video of how many have seen the video of the Stanford physics professor Who was Prediction he made about the big bang with you know some few decades ago. Just just was confirmed if you watch that video online It's been viewed a few million times But if you go see it when it's there's a assistant professor here at Stanford knocking on on the senior professor's door and door opens and just about the first thing the assistant professor says is It's five to one. I mean this is five to one signal noise ratio is big It's very high likelihood to be not noise And if we look at the pattern of warming in the troposphere the part of the atmosphere We live that single noise ratio is about seven to one and if we look at the stratospheric warming It's about 30 to one. So this is not the single noise In terms of the observations is really unequivocal. This is we we have we have global warming and You know we face a lot of risk in from from stresses in the climate system and and we face them now From the climate system that we have and we're facing Changes in risk as as global warming occurs. So this is this is a Conceptual diagram of the sources of risk and really the risk that we face in the climate system It is the intersection of the physical hazard, which you see on the darkest blue So that's you know heat waves and and thunderstorms and Extreme storm surges and and those kinds of hazards The the second dimension is exposure and this is really the you know, what what's in harm's way? So if you know if there's a if a tree falls in the woods and there's no one to hear it Did it make a sound if there's a tornado on the Great Plains and no one's there to get struck by it What's the risk great the risk is is not just the tornado. It's also what's exposed. This could be people infrastructure Ecosystems and then the third dimension is the vulnerability. So how how vulnerable are those those assets? and and not all not all assets are equally vulnerable and the the intersection of these three dimensions is what creates climate risks and and we are seeing evolving risks at present Just you know in the US we again we experience, you know, we know we're not optimized to the climate system that we have now so we've More than 70 billion dollar weather and climate disasters in the last decade So we're not we're not optimized To the current climate and we are We're experiencing a changing climate and we know from observations that the likelihood of many kinds of extremes has already changed We see this with severe heat. We see it with heavy rainfall. We see it with extreme storm surge flooding couple of examples work that we've done in in my lab has You know the quantification that we've done suggests that the severe heat that occurred in the Midwest and Northeast in the summer of 2012 when this picture was taken Huge huge impacts on crop yields that went all the way through in the food prices Our quantification is that the That level of severe heat is about four times more likely in the current climate Than without the global warming that's already been observed again increasing risk of high impact events similar work by others analyzing the storm surge that occurred during sandy in New York has Suggested that the sea level rise that's already occurred almost doubled the likelihood of that level that sandy level of flooding in at Battery Park Just for exactly the same storm But just that the sea level rise has doubled the risk of extreme flooding So we're already seeing just in that hazard dimension I never mind vulnerability exposure which are moving in different directions in different areas just from the hazard Perspective of hazard that we're seeing increased risk already We also know that That the global warming that we've observed is due primarily to the emission of greenhouse gases from human activities particularly carbon dioxide emissions and Here in the the black dots each of these Each black dot is the fossil fuel CO2 emissions in a given year and if we look back historically Since the start of the industrial age about a quarter of the emissions have been from the United States about a quarter of been from Europe and what we're seeing in recent years is Not only increasing total emissions in each year, but an acceleration in the rate of increase in the rate of emissions and the annual emissions have about doubled over the last 25 years and We know from the from just the energy balance of the of the planet the basic physics of Energy input and output That if if the world continues along this business as usual Pathway, we're likely to see about four degrees of global warming relative to the pre-industrial and That's about twice the level of warming that's been identified by the United Nations Which has a two-degree warming target and the international community is moving forward with binding commitments To achieve that two-degree target. That's not a prediction. I'm not making a prediction whether or not they will We will achieve that to be target, but that's that that's the the position of international negotiations Now achieving that two-degree target is really going to require Just in terms of the basic arithmetic of How temperature responds to To co2 in the atmosphere achieving to the target is you know given that we've already we're already almost at one degree It's really going to require a Real change in the global pathway of emissions, and that's not I'm not arguing that that has to happen I'm simply declaring that based on the arithmetic you That the pathway to two degrees is a different pathway than what what is currently empirically seen in the real world But it is possible. It is the windows the window remains open But what I what I want to give you a few examples of now is the difference between this four-degree world As you'll hear me call it that's basically this business as usual the trajectory We're on and the two-degree world that that the UN has identified and so one example is is hot extremes and We experienced a lot of stress year-to-year around the world from hot extremes and What I'm showing you are two maps here one for the four-degree world and one for the two-degree world And what we've done is we've looked at each Each area of the globe and we've looked back at the late 20th century at the recent past and we've asked what was the hottest season that occurred in each area During that that recent historical period And then we've looked out using using a large suite of climate models again this you know big data in terms of These climate model experiments So in in total those it's about three petabytes of data although That's for the more than just the temperature field, but we're analyzing it's randomized a large a large climate model experiment and We're just asking how often does that? Hottest temperature from the from the recent historical period how often does it occur in each of these worlds the four-degree world And the two-degree world and you can see on the right in the four-degree world is pretty much all dark red so 80 90 100% of the years exceeding what used to be the hottest season that The risk of of intensification of extreme heat is substantially reduced in the two-degree world Similarly for extremely low snow years so here in in California about 30% of our water resources are dependent on Sierra snowpack And this is generally true in in the Western United States and Western North America that are water resources are heavily dependent on on the snowpack and We're in the midst of an extremely low Snow year as was pointed out by Governor Brown right there pointing and Just going to show you an animation that's a very similar calculation what I just showed you for temperature But in this case, it's the the percentage of years that are below What used to be the lowest? spring snowpack and you can see that I just replay the animation if I can all right So out by the time by the end of this business's usual pathway Depending on where you're looking in the in Western North America somewhere between 40 and 80 percent of years Fall below the level that used to be the lowest level. So this is when when you hear a new normal This is what people are talking about. It's literally statistically a new normal where what used to be extreme becomes common So there are some other other examples I'll just show one really quick another example from our from our group looking at premium wine grape suitability in in the Western United States when we look out over the next three decades again using a suite of climate model simulations because Wine grapes have a relatively narrow climate Tolerance what we see is that The the global warming in the regional warming it's associated with that ends up really restricting the area in in a number of the High-value growing region Napa for example greater than 40 percent reduction over the next three decades And that's in that area that's that falls in what's the current suitability So these are these are the kinds of impacts that that we're facing the kinds of risks that we're facing We see it with again with severe heat and snow also in the literature heavy precipitation extreme storm surge flooding Likely increases in the volatility of crop yields and and as a result crop prices changes in the location of Suitable climate not only for crops, but also for for natural ecosystems, and we see this not only in Not only in managed systems, but also on land, but also in in coastal ocean habitats and part of why this is a big challenge is really that we're we're what the the the Rate of change in this four-degree world You know four degrees the global warming in a century and the associated regional and local Climate changes that come with that are unprecedented in the last 65 million years since since the extinction of the dinosaurs and we know already that The level of carbon dioxide in the atmosphere is higher than it's been in the last 800,000 years Looking back at geologic proxies We know that that for most of the last 20 million years carbon dioxide concentrations were below The level that they are now 400 parts per million and if we Again if we look at the geologic record of global scale climate transitions what we're what we're facing in in This business is usual World and really even in two degrees two degrees of warming in a century is really Substantially faster for global scale change than what the world has has seen in the recent geologic past So there's a lot of inertia towards towards further climate change Given given our current energy system our current energy economy again the tremendous benefits that I get from consuming energy And there are a lot of a lot of sources of inertia But the one I want to focus on for just a couple minutes is human well-being and this has come up a number of times today because We climate change is a side effect of of the benefits of energy consumption. There's no whether without being normative it's not No one's no one's as far as I know no one's trying to alter the climate. It's just people are people are trying to improve their well-being and and fossil fuel carbon-based energy sources are conferred tremendous benefits and I have this picture up here. These are these are college students in guinea and There this is the airport and they're all lined up each on a ballard at the airport underneath the lights outside in the outside the airport Studying their their homes don't have electricity. They have to literally migrate to the airport to study. So the most fundamental Kind of drive to improve human well-being education it requires requires energy and and Large populations lack access to that to those benefits and we see that We see that with with lack of access to clean water lack of access to clean Cooking fuels electricity as I've mentioned transportation and in fact the lack of energy access Actually creates tremendous vulnerability to climate stresses and so one of the one of the paradoxes of Of climate is and climate change is that you know for billions of people in the world It could actually decrease their vulnerable vulnerability to climate stresses by increasing their energy consumption and in the current energy system That means emitting more greenhouse gases and creating further climate change. So that's really that's really the challenge is If we look at all the countries in the world Each of these are at dot the horizontal axis is Is the per capita energy consumption the vertical axis is that is the human development index And you know these all the countries that are above 0.9 There's 28 countries that are above this you know in this top Top 0.9 and above this 50 percent of the global population and 85 percent of the global co2 emissions And so the real question is how is what what's the energy mix that's going to enable The human development of of this other you know 85 percent of the global population and if that looks something like me Like my energy consumption then that's the equivalent of adding 15 united states to the global emissions total And that gives us a more than a 50 probability of exceeding the four degree target or four degree the four degree level of global warming It doesn't have to be the u.s. I I picked the u.s. First because that's that's me. That's what I look like There are other countries that have a you know still have a high level of human development have a lower lower level of energy consumption And maybe a maybe a less carbon intensive fuel mix But even something like japan still gives More than 90 probability of exceeding the two degree target. So the real challenge is how can we Again, both ensure the Ensure the well-being that requires energy while while also limiting global warming and this is where I think we can we in the bay area and and in the u.s. can really be be the engine of innovation in Figuring out the energy technology the behavioral technologies, which you've heard a lot about today already Infrastructure design a whole suite of of areas where you know We don't have an example of a country in the world where we can say if everybody just looked like that Then we'd solve this this problem. We need we need innovation. And so Um, it's the last I want to come back to the big data here right at the end And I think that there are there's tremendous potential in In each of these areas for for innovation Uh, certainly we're you know that the uh The first of these is you know in many ways is a um You know has been a access to computational resources Question, but at this point. I think we're actually lagging behind intellectually And we're not you know, we heard about GPUs earlier today We're not the climate community isn't really taking advantage of those yet and other other sort of So there is potential to take the lid off. Um, just on the computational availability I think there are a lot of ways we can improve our analysis of these these massive climate data sets And then just the last thing I want to mention really quick Is What you know something that I don't I don't know the answer to but I think in terms of this risk framing We have a lot of opportunities to understand Exposure and vulnerability through these through these new unstructured data streams and just one One example really quick here is of something that's just You know, it's really new not it actually not even published yet. Um, I found this on the internet. Um It's not involved in this work But you know in this case similar as we saw this morning with the tweets during the during the earthquakes this is analysis of tweets during hurricane sandy and You know, this is a this is a their number of of big data Big data challenges here In this case, they're trying to correlate the tweets with the location of the hurricane and the forecast the hurricane forecast through time Um, this is a this is a short term application of this kind of New social network sensor integration into in the climate weather research I think where there's real potential is to look over the long term And try to understand vulnerability and exposure using these using these new data streams. So with that Thank you for your attention