 It's a pleasure to be here. Thank you for being here. So I'm going to talk today about sort of a half of the kind of the research that we do in our group and that half deals with the environmental implications of energy extraction and use. And that's the subject that we'll be talking about today. Jimmy mentioned we do a lot of other work, climate change related, drought related, and particularly ecosystems and water use is a special interest of mine. So for that I get to go caving and do other things that keep life interesting. So take measurements underground. But anyway, about six or eight, maybe close to eight or ten years ago after giving many talks about climate change and other things, but not really doing any direct energy related research. I finally realized that I wanted to incorporate some of that in my research program. So I have worked on carbon capture and storage, but about six or eight years ago became interested in unconventional energy extraction, particularly shale gas and shale oil extraction. So here's a photo. I'll start with this that I took from a helicopter. You're in the Marcellus in Pennsylvania. This is one of the larger shale gas plays in the U.S. and the world. And this is while a hydraulic fracturing operation is going on. It is not like this most of the time. This is just temporary. So you see tanks, you see compressors, you see water trucks, you see lots of things going on. And then you see that house in the back left corner, up here. And I guess what I want you to think about is how your perspective on that operation might differ depending on your economic interest and what's going on outside your door. So in some cases the people who live in those houses own the mineral rights to their land and reap the economic benefits of that activity. In other cases in the U.S. we have severed mineral rights where someone else might own the mineral rights but you own the surface rights. Or that might be well-pad on your neighbor's property. It's as close to your property as it legally can be and as far away from their house as it can be. So I think in each of these cases, and there are 20,000 or so of these hydraulic fracturing operations each year, each of these plays out in a small microcosm like that. And I think this is one of the reasons why this issue is so contentious. The philosophy of this component of our research project is how to minimize the environmental footprint of operations like this. So how can we reduce the amount of water that's used? How can we reduce any air quality impacts that might occur? And really I'm interested in, in some sense, in the exceptions. So what is it about the couple of percent, half a percent, up to five or ten percent of operations depending on what you're talking about that might have something that goes wrong? And how can we learn from those and keep them from happening elsewhere? And that's really the focus of what I'm going to talk about today. So we'll focus first on water then on air. But first, this is like a I get to know you dance. I'll just give you a little background on some other things I do. I am passionate about writing. So I've published a couple of children's poetry books. If you've ever read Highlights Magazine in the dentist's office or you have kids, they've done the books in the upper left and the upper right of mine. A trade book about the environment and a couple of textbooks. So I write and read as much as I can anything that's not peer-reviewed as often as I can. Jimmy mentioned I chair the Global Carbon Project. I have a co-chair also Naki Naki-Senevich in Austria. The GCP has been around about 15 years. Among other things we publish an annual carbon dioxide budget. So we take emissions data from companies like BP. But then we use models and inversions to attribute where the carbon dioxide goes. So every year we publish a budget how much went into the atmosphere from land use like deforestation from fossil fuel emissions from cement production and things like that. And then we track it. How much of it stays in the atmosphere. So what you actually measure in that rising CO2 trace. How much goes into the oceans for sort of free sequestration but has consequences for things like ocean acidity down the road. And how much goes back into the land in areas like Europe or the United States where because we cut our forests 100 years ago and those forests are regrowing. We have carbon sinks over large areas of the temperate of the temperate globe. So we release a global CO2 budget every year. We have a methane budget we started doing now. We have a global carbon atlas that's sort of an outreach or stakeholder tool that allows people to go in and look and see where the carbon is in different pools around the earth. Biotic fossil fuel pools and things like that. And increasingly even in the Global Carbon Project we're interested in unconventional natural gas because it's become so important. Because it's transformed energy production in the United States to the point where we now produce as much oil as Saudi Arabia does. Something that was unthinkable 10 or 20 years ago. This is a map that many of you, the kind of map that many of you would have seen. These are shale resources around the world. There are other tight sands and other unconventional resources on here. I think the main thing to take home from this slide is just how much of this there is out there. It's not all the same. It's not all the same depth. Not all the same quality. Not all the same carbon content. But what's happened in North America in particular can and might be replicated elsewhere depending on a lot of factors. Geological, human, economic and otherwise. Not much. There are other people in the room who can chime in if they feel like. Basically the North American plays none in Mexico. Argentina's drilled a few hundred wells to this point. They're one of the countries that are likely to proceed quickly. The UK's doing some extraction. China's drilling exploratory wells. Other countries are tinkering with it. But in terms of actual production, it's really the yellow blob. Because I've never been there. Russia has large unconventional resources as well. Does anyone know what that is? The name for it? I don't. So I said we'd start with water issues. And I want to, each of the steps that we take through this talk, I want to cover things that go well. And also where some opportunities are to have things go perhaps a little better. And why? And how we can do something about it. These are just a handful of examples of many things that are improvements if you will from the last five or ten years. I'd like to highlight the increase of recycling and reuse of water in the Marcellus in particular in Pennsylvania. So we generate about a trillion gallons of wastewater from oil and gas operations a year in this country. The Marcellus, for a number of reasons, they are recycling and reusing most of that wastewater rather than deep injecting it or transporting it and such. So they're treating it, reusing it. And that's a good, that's a really positive advance. There's much more disclosure about the chemicals used in fracturing fluids even though many of you might hear or read about, there's no chemical disclosure or such. That's not true. There's a patchwork of states that have different regulations about how much you have to disclose, whether or not you have to disclose just the identity of chemicals or the concentrations that are used. And you can think of the label on a food product as one example that will list for you what's in a food product but doesn't tell you precisely how much is in that food product and partly for trade secret reasons. And then finally, another example would be the green completions that are happening through new EPA rules and company initiatives. The elimination of open wastewater pits in many areas around the country and things like that. So what are some things that we study? Why do we study them and what might people be concerned about? Well the water used for hydraulic fracturing is an important issue particularly in drier areas. So in the Marcellus in Pennsylvania where water is relatively abundant, it's probably not as big of an issue particularly if you don't take too much water from any one stream or river. But if you're in west Texas or North Dakota or out west in the Rockies and you're pumping groundwater, then the water supply issue can be quite large. The amount of water used for hydraulic fracturing can be 10 or 10s of percent of the previous water used in a given county for instance. So adding to additional demands on a groundwater resource. They're drinking water quality issues that I'll talk about what we found in a minority of cases. And then the disposal of this wastewater, the produced waters that I talked about which in some plays like the Bakken in North Dakota or the Marcellus in Pennsylvania and elsewhere can be 10 times saltier than seawater. They can have very high concentrations of metals, can have dilute radioactivity and other things in it. So this is why companies go to so much effort to handle this wastewater as safely as possible to not just release it into the environment. And with a few cases there are no states that allow you to do that. Reuse I've already talked about deep injection. So pumping that wastewater back underground is where most of the wastewater goes in this country more than 90 percent. But there are still half a dozen states that allow you to spray it on land. You don't have to be a genius or environmental scientist to know that's not a great idea long term. But still done in rarer cases. Now here we have again the same picture you saw before. We've done studies looking at radioactivity associated with wastewater disposal and other things which is the figure on the right. So let's look at just an overview then of water use, where hydraulic fracturing is done in the U.S. This was a paper that came out a couple of years ago. I'll talk about in this talk. Let me mention as an aside. I'll try and mention a number of colleagues as I go along. But in this talk it's more about sort of what our group is doing and is done to give you an introduction to what we do. But in a normal research seminar I'd be spending more time talking about what other groups around the world are doing. And there are a lot of people in groups in a lot of countries doing this kind of work. Today we'll focus more sort of on what we're doing. This is a paper just from a couple of months ago 44,000 wells. In this case these are data sets disclosed or reported. Excuse me in the frack focus website. So in terms of physical geographic distribution, this is what you see. The Marcellus up there on the right, the plays in Texas, the Barnett, the Eagle for the Permian, some of the western plays. Here's the Bakken, North Dakota, some hydraulic fracturing here in California generally of lower volume. Depth wise, most of it's deep. The average depth of a hydraulic fracturing operation in the U.S. is deeper than a mile. The average water use is about 2.4 million gallons per well. And then as I talked before about some of the exceptions, there are about 1% of the cases where the hydraulic fracturing is shallow, less than 3,000 feet and where the water and chemicals used are large, greater than 5 million gallons. And these are the kinds of combinations that are more of a concern to me because most of the hydraulic fracturing that's so deep, it's very, very unlikely that you're going to have any contact with that a mile or more underground and surface water and such, except through well integrity possibilities of what we'll cover later. But as that distance shrinks, and especially as high volume hydraulic fracturing is occurring near the surface, then some of the safeguards, if you will, aren't as present. So that 1% of cases is the example from the study that we highlighted. That's used, that's before the hydraulic fracturing operation itself. So that does not include things like maintenance water use, you would use in a play like the Bakken to the water use, that's during the hydraulic fracturing operation that might be a couple of weeks, a couple of days to a couple of weeks. Many of those are Colbert methane wells, but there are certain states Arkansas being one of them where they are shale wells. Good questions. I may have trouble getting through my talk, but keep them coming. So, views of water use, hydraulic fracturing is water intensive, but one of the things that we've worked a lot on is trying to place data like this in a usable metric. So what's an apples to apples comparison? So if someone tells you a well near U.S. hydraulically fractured with 2.5 million gallons of water, it sounds like a lot. And it is a lot depending on what you're comparing it to, but what you really need to know to be able to conclude or compare it to something else is how much energy you get back out of the same well. So if it takes more water than a conventional gas well, which it does, but you get a lot more energy back out of that conventional gas well, it can be better than a conventional gas well. Or surprisingly, something you don't hear about very often is that in a paper we published last year we showed how hydraulic fracturing is less water intensive than many other fuels we're used to using like coal, nuclear, and tar sands. And that'll idea we have done some good work in this area too. So it's even though it takes a lot of water to get the oil and gas out of the ground, you get a lot of energy back out. Now it's not as good as wind or solar PV, which require no water effectively. But it depends the way you compare it to. We need to be careful about the metrics we use. And then finally, just in terms of the intensity of hydraulic fracturing, that intensity is increasing as wells have longer horizontal lateral legs. They're using more water and the amount of wastewater being generated is generally increasing through time across the U.S. as well. And the lower photos are not mine, those are from the AP. Actually the one on the right is mine from the Marcellus. The one on the left is from the dry area of South Texas. And just thinking about water resources again, it really depends on where you are. So that number there, a great question, that number there is for extraction and processing for sort of handling the fuel, but does not include the cooling water demands of the power plant. And natural gas is actually, as many of you know, natural gas is better in many ways of the power plant stage 2. Again, depending on what you're comparing it to, better than coal, better than nuclear in terms of cooling water. Alright, that was water quantity. So let me dig in a little bit to water quality. So this is the work that's probably been the most controversial that my group's done. This has been in collaboration with Abner Vengosh, my former colleague at Duke. This is an operation where you can think about where might something go wrong. So my first job was with the Dow Chemical Company. I know something about handling chemicals. I know something about corporate safety and other things. I could tell you that the corporate safety of Dow was a lot more stringent than the corporate safety of any university I'd ever been to when I walked into a laboratory. What people are most worried about is what's going on way down here. Here's the hydraulic fracturing operation down through the bottom of the floor, a mile down, two miles down if you're in the Bakken. What we've argued and what many people have argued is what we really ought to be concentrating on is what's going on up here. So the surface operations, where the water comes from here, perhaps if there's a spill occurs does it reach that waterway? Is there an open wastewater pit? Is that pit line? Is it unlined? Is that liner intact? Is it's integrity intact? Might you have a spill at an operation at the surface as in any chemical industrial operation that's a possibility? And then here we've looked a lot at well integrity issues. So not so much what's happening a mile or two mile down, but what's happening a few hundred feet down with the integrity of your well. And that's not a new idea to shale gas or to shale oil. Alright so starting about six or seven years ago we published the first study asking a simple question. We asked is a homeowner's water any different if he or she lives near an unconventional oil or gas well compared to if that person lives far away? So is there any distance effect? One of the first plays we worked in with the USGS was in Arkansas in Fayetteville. In this play we found no evidence for any changes in water quality whatsoever. Didn't matter whether somebody was living right next to a well or was very far away from that well. So no evidence for any changes in water quality whatsoever. In the Marcellus earlier on we found something that was a little bit different. So we've been working now in the Marcellus for six or seven years. We started in the Marcellus and in the northeast corner of the Marcellus for a couple of reasons. One this was an area despite Pennsylvania's history of oil and gas extraction that did not have a long history of oil and gas extraction. There were not many wells in this area compared especially to central and western PA where some of the earliest wells in the country were. It was also of interest to us because this is the state boundary with New York. New York had and still has a ban on high volume hydraulic fracturing. Pennsylvania did not. But the geology is the same. So this state border to us is like a fence line comparison working across a landscape. We can learn things about the water here by looking at the water here even if we can't necessarily go back in time to ask what was a particular water quality like in a well before drilling if we didn't have a pre-drilling sample. So what did we find? I'll just give you a couple of examples and we can talk all day about this. First let me start with what we didn't find. So we didn't find any evidence in the Marcellus for many things that people are concerned about. Salts, metals have never seen evidence for radioactivity associated with the drilling. In rare cases associated with wastewater treatment and streams and such, yes. What we did find evidence for in a minority of cases was stray gas contamination. So let me tell you what I mean by that. So here's a figure of methane concentration. This is distance to nearest gas well and each of these dots is somebody's water. So these are the 40 or 50 million people in the US in that subset of people who were revived at water wells. So they don't have city water wells, they don't have testing or anything like that to monitor the quality of their water. So most of the people are down here. You see background variation in methane. I'm happy to talk about that in detail. What you see in a subset of cases within about a kilometer distance though is a small subset. It doesn't look like it's small but there are a lot of people down here. So there's a small subset of people with very high methane concentrations in water. These are the people that we think have potentially been affected by drilling and we'll talk about next about some of the sourcing, how we might determine or know whether that's natural or not. Ethane is up here in the right. Now ethane is even more useful to us and has become particularly useful to me as a researcher because there aren't biological sources of methane. When we're talking about methane, it's not just about the methane from the soil itself. There's natural geologically derived methane that's moved up through many, many years. There are biological sources microbes and aquifers and things like that. You have to tease all that apart in attributing the cause or the source of the methane that you see. But ethane is different. Ethane has no biological sources. So the ethane signal is much cleaner and the propane signal is like this. The ethane signal is flat and then bam, you get within a kilometer and you see a small subset of these homes in the same way. It's hard to say. I'm going to say small. I'm going to avoid that question for now. I'm happy to talk to you offline. You can't even do that. Okay, evidence for at least a relationship with distance doesn't prove causality and I will point out that for those points we do not have pre-drilling samples. So that's what you would have in a perfect operation, in a perfect case. That's what you would want to have. We do not have that. So how can we say some things about the sourcing of that methane anyway, even though you can't go back in time. This is one way of many ways that we can do it. This is work with noble gases. Tom Derra was a postdoc who did a lot of this work. We had a follow-up paper that came out last year. This is a plot of helium over methane. On the y-axis, noble gases are useful to us because they're inert. They don't react. So this is helium, neon, argon, the gases that you're often familiar with. Same kind of plot, distance to nearest gas wells. So for people who are across this range, any distance from a gas well, you get sort of a natural variation in helium to methane that looks like that. So why helium-4? Well, helium-4 is a decay product. It only comes essentially from underground. From decay, it moves upwards. So it's not strictly derived, you know, moving downward. So we use noble gases to differentiate sources that are rained down into water and sources that are derived in the crust itself. Alright, so natural range here. Here's a salt spring. Nothing to do with oil and gas drilling. Methane, that's saturation naturally. Just water bubbles. Very salty water. It's at a value here. Now here are source gases. Marcellus gas itself. These are actual gas samples from production gas. Intermediate samples found a thousand, a few thousand feet underground. There are other pockets of gas that used to be extracted before the target was the Marcellus, a mile or more underground. Within a kilometer of homes, we see the same variation. Within a kilometer of the gas wells, we see the same variation up here. Then you see the subset of houses. Some of them look just like the intermediate gas. Some of them look like Marcellus gas. You don't see those at all beyond that distance. So what's going on? Well, we believe the upper Devonian gases represent issues with cement. So if you have either cement absence or cement that's cracked, you have an intermediate pocket of gas that can move into the annulus of that gas well, move up the annulus and then move out into the aquifer. So that gas should not look like Marcellus gas and that gas will never have the chemicals or anything else associated with hydraulic fracturing in it because there's no breach in the safety of the casing. Down here, we're not suggesting in any sense that the gas is moving up through thousands of feet of bedrock. We think these are casing leaks. Another thing that's not a new idea. So in this case where you have a casing breach, perhaps the threads that are broken, a crack, corrosion, you can have leakage of that gas into the surrounding aquifer. Those cases are a little bit more of a concern because where you have methane leaking, you can have other things that might leak as well. So that's just one example. We spend years and years looking at sourcing and how to say something with confidence about where these gases come from. So well integrity is the key for the stray gas issue. Again, not a new idea to anybody in the oil and gas industry, well known. This is a paper that Tom led last year. Lots of different scenarios, some natural, some not. You might have in the worst case scenario, a hydraulic fracturing operation that connects to a natural fissure and if you had enough pressure, which is also unlikely, you could have movement upwards. If you're in a fractured geological area like the Marcellus, like some areas of Texas, this kind of movement is a little more plausible even at surface layers because there are more natural pathways in those rocks for things to move around. So if something goes wrong, you're more likely to see it in water. But that's not what we think is happening here. There was one case documented and it was already documented of an abandoned well that was intersected but we think is happening in these two cases. So here's a pocket of gas, intermediate level gas, not the target layer, where if you had either no cement or poor cement, that gas could move up into and out into the aquifer. That's the intermediate gas, I believe, that I showed you before, and then here's a casing breach issue. The gas is moving up through the issue but if there's a break in that casing, you can then move it into the aquifer. So it's all about well integrity. This was a paper from last year Richard Davies in the UK led, but it was an analysis we did of well integrity globally going back in time. It's apples and oranges, it's old, it's new, it's conventional, it's unconventional, it's very hard to compare all these data together in a quantitative way. But this is sort of a compendium, if you will, of different data sets and different evidence. And then sort of finishing up this thread, I do care a lot about making a difference. So like you, I want people to use what I do and I want it to make a positive difference out there where possible. So here's one example, Pennsylvania passed a law after our first paper came out in 2012. They passed several laws in the last five, six years to enhance casing integrity. In this case, when our first paper came out that was led by Stephen Osborn, we saw this relationship with a one kilometer distance, again a minority of people. We said the setback distance in that time for presumptive liability testing for pre-drilling water testing was 1,000 feet, it ought to be 3,000 feet. So the law was upgraded to 2,500 feet and now 2,500 feet tends to be the standard that most states use, not every state. So they increase the setbacks to public water supplies and other things. And then finally, I'm interested in what happens long term. So here's a picture of, let's see, in this case this is Mary Kang's experiment. She's a postdoc in our group. She's interested in methane emissions from legacy wells, if you will. So what are we doing now perhaps in 25 or 50 years we might regret that we're doing, the way we regret acid mine drainage in Pennsylvania now. So Mary's interested in these old and abandoned wells that are out there. There are millions of these wells across the landscape in the United States and a small subset of them. She's trying to figure out how many people have issues. Here's one from Pennsylvania. So that's methane and other gases bubbling out that old. Pre-drilling baselines are important. We just published this analysis for North Carolina. Now North Carolina is not going to see any oil and gas drilling anytime soon if ever. But the oil price, the way it is, and the resource, the way it is in that state. However, we were asked to provide a drinking water quality, a groundwater quality baseline, if that extraction ever happened. So that baseline is now publicly available. Okay some recommendations for water and we'll finish up with air. Gathers much pre- and post-drilling data as possible and make as much of it as possible publicly available. Keep track of everything. So what's used in the fluids, where the water comes from, where the wastewater goes, how wells are cased and cemented. And of course states and companies do a lot of this and in some states are better at making such information available than others. Where are the old and abandoned wells? How much is a state setting aside for legacy issues to fix problems that might occur in 25 or 50 years? And most of them, in my opinion, aren't setting aside enough. And then to protect landowners and consumers through bus management practices and rules and especially good communication. So that's sort of the first half or two thirds, if you will. Let me finish with a couple of air examples and we'll break for the social. So on the energy side, research in our group is split roughly evenly between water issues which you've heard just a very small snippet about. And then air issues particularly emissions of methane and other gases from operations both upstream from well pads and elsewhere and downstream to the end users say in city streets, through natural gas lines. And I'll give you some examples of these as we go along and what kinds of inferences we can make from them. So first of all just as in water there's some good things to say about natural gas and air quality too, particularly when you compare it to coal. So not only is it better in terms of the combustion of carbon dioxide, it results in about half the CO2 use for the same amount of energy or heat generated as coal does, but it also doesn't emit anywhere near as much sulfur dioxide and mercury. So less than 90%, less NOx and particulates, these are things that kill people still in the U.S. today. So I think sometimes the air quality issues get forgotten when there's all this discussion about the CO2 emissions. And once again there are other sources renewables, in this case nuclear, when it's solar that also have zero emissions. So it depends what you compare it to. At the EPA they estimate that methane emissions occur about equally upstream and downstream. So if we're going to make an investment in natural gas at which we are making for power, phasing out coal to some extent we want to minimize leakage of methane and other gases from oil and gas operations as much as we can. Everybody has that in their best interest. It's both environmental and economic sense. EPA estimates that leak is about equal upstream and downstream. So about 1.5% total of the natural gas that's produced. EPA estimates leaks to the atmosphere. This is an analysis of Adam Brantz that came out last year. So we'll spend a few minutes talking about upstream, but then we'll focus on the downstream part of this and we'll finish. I'll just give you a few more examples. So over the past year and working with the Environmental Defense Fund we've sampled about several thousand operations across different shale gas plays. So to do something like this you can take an infrared camera. So if you see if there's an emission from or a leak from a well pad like a tank for instance, that emission is not visible with the naked eye obviously. But it is visible in the infrared. So you can have an infrared camera that allows you to film these and then to start to ask questions like what percentage of operations leak in a particular play. Does it matter where you are in terms of geology? Does it matter whether it's oil or natural gas? Does it matter who's doing the work? Does it matter what state you're in because of the regulations? There are all sorts of questions that you can ask. These are photos I took a couple weeks ago in the Bakken. This was one of the last places that we've been to. So this is an example of the kind of thing you can see. Now this is not from the Bakken this is from Texas but this would just be an image that you're seeing from the helicopter with an infrared camera. You're flying over the well pad. You'll see in a second. That's the emission that's invisible to the naked eye. Now that's not normal. Alright, depending where you are, that might be 1 in 200 well pads. It might be as many as 1 in 10 which is the kind of range that we see. But if you can identify those quickly and cheaply and particularly if it's the same well pads and the same operations that leak through time then you can really make a big difference relatively cheaply. So that's one of the reasons for doing this kind of work and for why so much of this is going on now. So in this audience he's doing a study on intermittency. When you find a leak does that operation, does that tank, does that well keep leaking for hours, days, weeks or months. You can't yet see it in satellites infrared. Now there have been satellite studies in the most well-known one is Eric Court's study that's more at a regional level. He was the first author on a paper that targeted or highlighted the four corners area of Colorado and Utah as a hotspot of methane leakage. Not to the point where you can use a satellite to do attribution at a particular well pad. That day's coming though in the not too distant future. And drones offer another possibility of relatively cheap detection too. I'm going to run out of time here. Go ahead. That's in Texas. That's in the Barnett. Well the coordinates are at the beginning but she wasn't writing fast enough so we're not going back there. So let's talk about downstream. This is my last example and we'll break. I want to spend a little more time on this because in the last four or five years we spent a lot of time trying to understand the contributing factors to emissions and leaks in cities. This is work with Nathan Phillips at Boston University, a close colleague. So we took new laser technology. In fact technology that was developed here at Stanford many years ago, commercialized in the Bay by several companies, Pacaro, whose instruments we use here but also Los Gatos and there may be some folks from those companies in this room. So these are newly commercially available. We took these brand new laser based methane instruments and put them in cars and started driving around block by block in cities and along pipelines and such to ask how often do you see a leak? And if you see a leak what predicts you? What allows you to predict the presence of that leak or at least the probability that you'll see a leak? So you're seeing here the operations, the guts if you will, in the car here and here the analyzers and the sampling tube will be in front of the car and then once in a while when we see a big leak we'll hop out of the car and try and figure out where it's coming from in the street or from a building or somewhere else. That's John Carr in the lower right he used to run the mass spec lab for me at Duke. What would be fine? Well first study and Nathan was the first author on this paper, this was Boston so it's the first publicly available map if you will of a city. That's 800 road miles block by block literally across Boston 3400 leaks. We define a leak as an emission that's well as a concentration that's 25% above background so 2.5 ppm is our threshold. The background is about 1.9 or 2 ppm methane. It's a conservative threshold. So each of these yellow spikes is a leak in a particular place or an emission source. Number one predicting factor for those leaks old cast iron piping. So if you look at the percentage if you will cast iron piping and old unprotected steel piping in a neighborhood it's that old piping that is correlated with the presence of those leaks. So it's not socioeconomic it's not the poor neighborhoods leak more than rich neighborhoods or anything else. No cows on the street. We won't do it today we do a lot of things though so attribution is always an important question right? How do you know that these are all from natural gas pipelines? Well they're not all but the vast majority are and we know that from isotope use. So we also measure the isotopes of the methane the C13 values and we know it from the ethane that we observe as well. So this study we'll get back to attribution in a second. This study got some quick responses within a day. The mayor had commented on it at that time Representative Markey now Senator Markey had commented on it last summer Massachusetts passed an accelerated pipeline replacement bill for the state. So it's a bill that doesn't tell people what they can't do it's a bill that among other things does have some requirements in it about reporting and such but it allows the companies to recover costs to fix pipes earlier than they otherwise would have. So they can front load those repairs long term I think that's a great solution so if I were a guru which I'm obviously not the two things I'd like to see happen are greater cost recoveries from the public utility commissions that's the lever that keeps companies from being able to spend more on pipeline repair and replacement. Another disincentive if you will is that rate payers pay for the gas that goes missing from the pipelines. So the companies don't pay for that gas the people do. So my argument is let them front load the cost for the repairs wean the companies from being paid for gas that they lose long term. Our attribution how much in total leaks or is emitted from a place like the Boston Metroplex. So to do that you can either try and do it 3,400 sources at a time which is inaccurate and a pain or you can try and do it from top down instead of bottom up and the way we did it in this case this was led by Catherine McCain the grad student at Harvard now in Steve Wolfsey's lab the way we did it here was to put sensors on a couple of the skyscrapers in the city to put sensors outside the city so that when the wind was blowing into Boston from different directions you knew what the concentration of methane in the air was what the concentration of ethane in the air was then it moves into the city you watch that concentration build up so you can see how much extra methane is in the air and then you can ask is it really coming from natural gas or might it be coming from capped landfills or the sewer system or something else and always to be asking how do we know what we know the way you know that is if it's coming from the natural gas system then the methane that's in the air of the Boston Metroplex ought to have the ratio of methane to ethane that's the same ratio found in the pipelines and that is in fact what we see so through time the blue line is the average of the pipeline to ethane ratio so about 2% of the gas of the natural gas running through Boston system is ethane you ought to see that 2% ethane in the atmosphere as well and we do so each of these dots is an observation point if it were 100% natural gas you'd be along the blue line we're down here at the red right so in winter it's about 90% of the methane that flows to the atmosphere in summer it's about 60% the amounts about the same but in summer you've got extra emissions from wetlands and landfills and biological activity because the temperature is warmer so we know that most of this is dry from the natural gas system because it matches the ethane signature as well as the methane DC was the next city we did we published that last year another old city with lots of old infrastructure not typical of every city in the US not typical of young cities so 6,000 leaks approximately in Washington DC Capitol Hill one of the oldest neighborhoods in the city each of these is a spike this is leakier than Boston was Washington posted an app that people could plug in their address and see what was around them based on our data set all of you both and actually we don't you can't get information about pipeline materials or aging out publicly after 9-11 that information was removed from the public realm if you will I knew someone was going to say that those are some pretty high ones so what can you do with those data well one of the things you can do is to try and say well in what circumstances there are very very rare instances of explosions if you look at the data set for the last decade pipeline safety in this country has gotten markedly better over the last 10 or 20 years there are half the incidents per year today than there were a decade or two ago and these are doing rules and other things however rare incidents still occur so one of the things we did in DC was when we mapped our top 20 concentrations on the street we hopped out of the car found out where that gas was coming from 12 out of those 20 cases there was an explosive level of methane in the manhole under the street we called those in called those 12 in went back 4 months later and tested them again 9 of the 12 were still explosive and that should not happen we can also use this kind of concentration information to prioritize where to do perhaps repair so we present maps of the top 50 or 100 concentrations and sources so that you can target where to go that's a mix of economics of cost of safety of other factors it's not practical to say go fix 6000 leaks in DC or it takes decades to do that but we can try and use the information to help prioritize those leaks being fixed and the spirit of prioritizing my last example we had a paper that came out a couple weeks ago some of you may have seen and the spirit of this analysis was different so we wanted to say how much difference does it make when companies and the public utility commissions and the municipalities get together and partner in these decadal to multi-decadal replacement programs so for this study we drove most of the roads in Cincinnati, Durham, North Carolina and Manhattan so why Cincinnati and Durham, North Carolina those are well I was living in Durham, North Carolina which made a convenience but that wasn't the only reason Cincinnati and Durham have finished or are almost finished their pipeline replacement programs so Cincinnati for instance began in 2000 and for the last 15 years they've been marching along to get rid of all that old piping that we've been talking about Manhattan has done a lot over the last decade or so but is still much farther behind it's more like the rate that Boston DC and other cities take even longer like Baltimore so we mapped these three cities and then compared them to what we saw in Boston and DC what we found was that in Cincinnati and Durham we found 90 or 95 percent fewer leaks in these cities where these pipeline replacements have been done so on the one hand you can say well duh you spend money to fix the pipes and the number of emission sources or leaks goes down but still a remarkable decrease in the amount it's beneficial from both an economic standpoint in terms of the loss value of the gas recognizing that it costs more to fix a pipe in Manhattan than it does in Durham or Cincinnati so it's tougher for the companies working there but it's an improvement cost wise and improvement safety wise, improvement air quality wise and in the Boston Massachusetts bill that they passed last year the cost of the extra charge for the acceleration was about a dollar a month for the average rate payer and they were expected to get most of that back if they reduced the emissions, the leakage of methane and natural gas to the atmosphere so it was almost a wash for consumers not to mention the safety benefits alright so I'll finish justifications for fixing the leaks, money so that's almost two billion dollars a year on average but the rate payers pay for gas that leaks into the atmosphere from these pipelines, job creation, consumer safety, rare instances you know a dozen or so people a year are killed in accidents property damage, hydrocarbons catalyze ozone formation one thing I've also had in a slide earlier that didn't say is one of the reasons we target methane in this kind of work is because where a site is emitting methane it's often emitting other gases and sometimes those gases might just be heavier ethane and propane for instance but in well operations and such you also have benzene and toluene and xylene things that are much stronger health issues if you will so by targeting methane it's relatively easy to measure you can also get quite a bang for our buck on some of these other gases and then finally climate change and greenhouse gas emissions too so we have a natural gas initiative here on campus that's led by Mark Zobac who's in Australia now working on it as part of that so the natural gas initiative Adam Branson and I have a proposal for a mobile methane laboratory if you will a set of technologies including a Stanford helicopter so if anybody wants to give us a Stanford helicopter I'm all ears and then finally I'll finish there just with the thought of where will our energy come from it's an amazing time for me to be working in this area I think for all of you we're seeing so much happening right now the shale oil and gas revolution that started a decade or more ago has been absolutely transformational coal use is dropping at least in the US on the other hand in the last 5 to 10 years the rise in solar and wind has also been equally incredible when you look at the decline in the cost curves and reaching cost parity in many markets between renewables and fossil fuels depending on how you define levelized costs and parity and subsidies and all of that so at the same time there's been this amazing rise in North America for unconventional oil and gas around the world there's an absolute revolution in a boom in solar and wind happening right now and so that's a race if you will it can be a complementary race if we use our natural gas well if we reduce emissions, reduce issues for air and water phase out coal plants and old coal plants in particular ultimately we want to be in renewable space but natural gas can provide bridge fuel if it's done well and that is how long will we use it for time for a few questions and then there's wine calling that is improved sources of energy occur that the old sources continue to increase exponentially so very good observations first of all let me say this was the global energy assessment this was not work from our group I was not involved in this assessment my co-chair in the global carbon project was one of the leads so at the same time there's been this incredible rise in renewables so this sort of darker gold wedge you see the rise in gas that's somewhat attributable to unconventionals so here's coal still going up right so even as coal use has declined in the US if you look at our exports the coal extraction has been about the same so coal use in China in other countries is going up rapidly well but what's happening globally I think in the last couple years you started to see coal use decline or level offer decline globally too we can talk about that it'll be part of our new budget but you're absolutely right if we keep burning coal more and more globally then all this stuff helps but it's not going to be enough to slow climate change there was one more in back and then we'll move to the front my name is Craig Lewis I work in energy policy and one of the things that seems like a big factor for natural gas in general but fracking in particular is just the phenomenal number of points of failure that could happen from the casing of the wells to the leaks along the way to apartment buildings blowing up in Manhattan and neighborhoods in San Bruno so I'm just curious from your point of view seems that natural gas has had somewhat of a free pass from clean air, clean water rules and it wouldn't take much to have policy put in place a few accidents happen here and there especially in high profile neighborhoods and all of a sudden you have a whole different dynamic in Congress I would think for putting natural gas drilling back under kind of normal environmental regulation what do you kind of what's your take on that risk factor for gas industry in general? I suspect some of the people in the room might have a different perspective on the industry getting a free pass that's not my view of the industry there are particular policies that hydraulic fracturing has been some people use the word exempted some people don't like that word but the circular super fund bill some aspects of the clean water bill so particularly for water there are some areas where I think there are some issues on the airside I don't think the industry has been given a free pass at all there are stronger rules coming out now released by EPA there are new rules being applied on BLM on federal lands there are best practices that are improving independent of rules and regulations so I don't know how about turn it over to somebody else on the floor and give us your perspective let's move to the front I don't see hydrogen on there that's because effectively there isn't okay you're absolutely correct this is 2050 so there isn't hydrogen here because we don't use hydrogen globally out here there are hydrogen experts here in the audience which I'm not it's a matter of cost and many other things I'm not sure take the California example in Governor Schwarzenegger but you're right you don't see hydrogen on there if it is it's buried in this renewables she was next I believe right here two she's right this she so you mentioned before you moved away from this graph that coal globally is reducing you see that as well and what do you think will take to really just kind of flatten that out and of course this is a projection not there's not facts out here obviously but in here it's still increasing I think it's going to be a mixture of different things I really really really meant what I said about the price parity issue for wind and solar you know you could argue that we're not quite there literally depends what market you're in but that is a game changer in terms of a long-term energy supply I believe when it becomes just as cost effective to deploy to deploy wind and solar when that wind and solar at least for solar PV has no water requirements concentrated thermal solar is a different issue so you have that issue there are you have issues in China obviously with their quality their estimates are a million people are dying because of the air pollution a year in China so China is working breakneck they're still building coal plants but they're working breakneck to advance renewables they built they built more renewables last year than anybody on the planet so I think it's a combination of pricing as always policies to incentivize use away from coal as you're seeing in this country now and consumer choice here first so last one the wine calls hi methane is a far more potent greenhouse gas than CO2 what is it like four times as effective or more so more than that it depends what time so methane is shorter lived in CO2 it's about 10 years on average in the atmosphere instead of 100 more than 100 years for CO2 so if you take a short time scale it's more like forget the numbers 80 or so you take a longer sort of 100 year time scale it's more like 36 times as potent so molecule for molecule it's much more potent but you have to remember it's 100 times less abundant in the atmosphere than CO2 is right so it's 2 ppm versus now 400 ppm 200 times less abundant so while we're struggling to get our CO2 parts per billion down do you think the methane leaks globally are significant in terms of greenhouse warming yeah of course they are I mean methane is the after CO2 methane is the second most important greenhouse gas that we affect in our activities in the United States the number one source of anthropogenic human derived methane is oil and gas operations according to the EPA and then a close second is livestock and ruminants so absolutely it's an important source and if we there's been a lot of controversy over the last five years many of you know about about whether emissions from leakage and emissions from operations might be enough to offset that CO2 benefit you get at the combustion stage power plant that burns natural gas uses about it generates about half the CO2 that a coal plant does per unit of energy but what's the what are the sources of methane emissions from the coal operation and from the natural gas extraction use my take on the literature is that the work over the last five years suggests that the EPA is underestimating methane emissions from activities but not to the extent that you're going to see it offset that CO2 combustion benefit for coal not to mention the other air quality benefits I mentioned so this is why from my perspective it's all about what can we do to reduce those emissions those emissions those leaks intentional unintentional that's one of the motivations for this work thanks for your time