Mobilizing Research and Ingenuity for the World's Needs
Steven E. Koonin, Chief Scientist, BP p.l.c.

Darcy Kelley: Well, we may not be able to do something about the weather, but we can certainly come to understand it better. Thank you very much.

Our next speaker is Steven Koonin, who has a very distinguished academic background, having been on the faculty of Caltech, having served as Provost there. He's a Fellow of the American Physical Society, the American Association for the Advancement of Sciences, and the American Academy of Arts and Sciences, and he has the distinction of serving at the moment as Chief Scientist for British Petroleum, and so he brings to this dialogue that we're engaged in a new and interesting viewpoint from a concerned and very effective company in this area. Let's welcome Steven Koonin.

Steven E. Koonin: So I do have what I think is one of the most interesting jobs in science and technology at the moment, which is helping to plot the thirty-year program of technology for one of the world's largest and most progressive energy companies. And so what I want to talk about is how we go about thinking about research and technology for meeting the energy needs, and certainly sustainable is a very important consideration in thinking about that.

When you think about energy technology you quickly realize that it's not about the technology itself, but rather in fact the technologies that one develops and deploys are conditioned by the social, political and economic contexts in which they're used. And so it's important to understand a number of very simple but profound truths about the world's energy system and its likely evolution before we can talk about the technologies. And so we'll cover some of that first before turning to the technologies themselves.

There are really four drivers of the world's energy future over the next fifty years or so. The first of those is the anticipated strong growth in demand which Professor Sachs has already alluded to. That's driven largely by GDP growth, but also by population, and then I'll have something to say also about demand management. This chart for me summarizes so much about what's going on and what is likely to go on with the world's energy situation. What is plotted is energy use per capital in gigajoules against the GDP per capita in constant dollars. And what you see are several things when you look at this chart. The first is that the developed countries, most prominently the US but also countries in the EU and Japan, had a large but relatively slowly rising energy use as their GDP increases. And in fact you can argue that the US while high is simply the N member of the general trend. Secondly, that there is a broad swath of countries down at low GDP and low energy use per capita for which ontogeny recapitulates phylogeny, namely their evolution follows the general trend of the countries, and in fact their energy use will grow rather more strongly as they develop. And what is sobering to realize is that there are about a billion people in the upper right part of the graph, there are in round numbers two and a half or three billion people in the lower part of the graph, and that another two to three billion people that are not even shown on this chart. And as these countries develop it seems almost inexorable that their energy use will be rising as well. In fact, you can put that into a simulation and what you realize, this is the historical and projected energy use over the next twenty-five years, some growth in the OECD countries, but by far the greatest fraction of the growth, some 75% of it, occurring in the developing economies, and that growth ?? to about a 60% increase in the world's energy needs over the next twenty-five years. If you go out to mid-century it is just about a doubling of that energy need.

Everyone talks about energy efficiency, how important it is. It is indeed important, but one of the lessons that one learns from numerous examples is that it may not reduce demand. Efficiency through improved technology is about paying today in increased capital costs versus paying tomorrow in the expectation of reduced operating costs. For it to happen it must be perceived as being cost effective, or it must be mandated by policy. But what really matters in the end is reduced demand, not just efficiency. And efficiency in and of itself may or may not reduce demand. That's certainly true in supply-limited situations. For example, if there is now not enough electricity in China to run all of the air conditioners, making more efficient air conditioners will not reduce electrical demand, but it will result in somewhat cooler living rooms around the country. And in fact, increased efficiency may increase demand in some situations by making energy easier to use. A wonderful example is the US automobile fleet in the decade of the ‘90s where the net miles per gallon of new cars sold actually improved by about 5%. But when you open up and look at what happened to get that efficiency you discover that engine efficiency actually improved by an astounding 23% in that decade, but that was negated by an 18.4% increase in the weight and improvement in 0 to 60 performance. And so a lot of efficiency depends upon how you actually use that efficiency, whether it will result in reduced demand or not. In that same decade, you might be interested to know that annual miles driven per automobile went up by 16%, and as a result annual fuel consumption per vehicle went up by 11%. So it's not only about efficiency, it's really about reducing demand.

Let's then talk about energy supply and the headlines here are that there are significant fossil fuel resources, the world is not running out of energy anytime soon. However, there will be required to be massive investments in infrastructure and a growth in non-conventional energy resources for reasons that I'll cover in a moment.

This is the business as usual energy supply situation. You can see that currently about 85% of the world's energy is provided by fossil fuels, coal, oil and gas, and that's not going to be changing very much over the next twenty-five years, despite the stronger growth in renewables. Even renewables growing at 3.3% per year do not shift the balance very much over the next twenty-five years.

Fortunately, there are substantial fossil fuel resources available to meet that demand. Right now the world has proven forty years of oil at current production rates, and plausibly about another forty years to find conventional crude oil. For gas there are seventy years of proven reserves, and plausibly another seventy years or so of reserves yet to find. Coal is the big gorilla in this discussion. There are right now about 160 years' worth of coal known to be in the ground, and plausibly as much as six times that, since no one has really gone looking for coal yet.

The existence of oil reserves is a much discussed and debated topic, as Professor Sachs alluded to at the beginning. This is a very interesting graph that shows how much oil is available at what price. And you can see that the Middle East has another trillion barrels of oil. The world has already consumed a trillion barrels of oil. Other conventional oil is another trillion at somewhat higher cost. And then one could go to more exotic situations, heavy oils, oil shales and so on, amounting to about another four and a half trillion barrels of oil at prices that are relatively economic compared to today's high price. Note that the world is projected to need about another trillion barrels over the next twenty-five years. This bar corresponds to consumption at about a hundred million barrels a day. Right now the world is consuming eighty-five million barrels a day and growing rather rapidly. And so the world is not going to run out of oil anytime soon, certainly at today's prices.

It's also interesting to note that through large scale chemical processes one can convert coal or natural gas or biomass into the liquid fuels that we need to power transportation. For gas that can be done at about $25 a barrel right now. For coal it can be done at about $45 a barrel. And you start to see these technologies being deployed in today's high oil price regime.

Let's talk about security of supply, which is another very important issue. The headlines here are growing import dependence and growing competition for limited free resources around the world. This chart compares the consumption versus reserves for the three fossil fuels. Shown here are the three largest energy markets, North America, Europe and Asia Pacific, and the rest of the world. And you can see that those three largest markets consume 77% of the world's oil production, but have only 10% of the world's oil reserves. The gas, the situation is a little bit more well balanced but not really. 60% of consumption in the largest markets, 15% of the reserves. For coal it's a much more even split. To say it a little more succinctly, the oil is where the people aren't and the coal is where the people are. And you can then imagine as security of supply becomes an increasing consideration that the world is going to be turning increasingly to coal for its energy needs, as in fact it is.

Finally I want to discuss the last driver which is environmental issues. Local pollution I won't say much about, but I do want to focus some on the issue of climate change, which is a major subject for this get-together. What is shown in these two graphs are the carbon emissions and the atmospheric concentration of carbon dioxide, both historically up until the present, and then projected going forward for the next 150 years or so. The black line for the emissions is what is history, up until 2004, and then projections going forward. Similarly over here, concentrations which are historical, and then projected going forward. Focus for a moment on the black line which is, if you like, the business as usual scenario. Emissions are rising at one and a half percent a year, and as the result concentrations are rising. They have in fact risen from 280 parts per million pre-industrial to the present 380-381, and are projected to hit 550 by about mid-century or slightly thereafter.

What is important to realize is that because the lifetime of carbon dioxide in the atmosphere is a couple of hundred years, effectively the atmosphere just accumulates the emissions that we're putting out over the next century or so. If one wants to stabilize concentrations at 550 ppm, which is considered by many to be a point beyond which we will start to dangerously influence the climate system, then one must stabilize emissions at their current value for the next fifty years, and then see a decrease thereafter. These scenarios, the solid green or the dashed green, or the blue or the red, are different emissions trajectories that lead to stabilization of…

…that will only delay crossing any particular danger line. So if you manage to take this black curve and bend it down slightly, the corresponding curve in the concentration will also rise, but will just be delayed in its trajectory. A good rule of thumb is that a 10% reduction in emissions, which the world has certainly not managed to achieve yet, buys you only seven years of time before you cross a given concentration threshold.

Also important in discussing CO2 emissions is the great heterogeneity of CO2 emissions around the world. This is a chart that shows CO2 emissions against GDP per capita for various countries. It looks very much like the energy chart, except for two outstanding exceptions, which I'll come back to in a moment. So again, developed countries, high per capita emissions, developing countries, low per capita emissions but increasingly rapidly.

If you look at that what you realize is that in the 21st Century emissions from the developing world will be much more important than those from the industrial world. The developing world emissions are rising at 2.8% a year commensurate with their development, whereas the industrial world emissions are rising at only 1.2% a year, and they become equal sometime within the next ten to fifteen years or so. This is what the curve looks like, emissions versus time, for both the developing and industrialized world. Two sobering facts then come out from that analysis. Every 10% reduction that the industrial world can make in its emissions is compensated by less than four years of growth in the developing world. And second, if China or India's per capita emissions were equal to those of Japan, which is among the most emissions-like of the developed countries, then global emissions would be up by 40%. And remember, we're trying to stabilize and then get down by a factor of two in the face of dramatically growing energy demand.

You can conclude from this that this is an enormous and complex challenge, and technology will be central to meeting it. So let's talk about technologies then, and how do you evaluate various technology options. One that is very important is the technology headroom. Can we see significant improvements in any given technology? Second, very important, are budgets for the three Es, what is the economic budget, cost relative to other options, the energy budget, how many times do you get energy out versus what you put in, and then the emissions budget, how much CO2 are you putting out? Also very important that people tend to forget is the materiality. As I showed you in dealing with CO2 emissions, you have to make substantial changes in energy technologies, and so materiality is very important. And then there are other costs, reliability, intermittency, and so on. There are issues with social and political acceptability, not only for nuclear most obviously, but also for wind increasingly. And then you have to know also what problem are you trying to solve, because there are at least two problems here. One of the issues is the concern over supply, and are you very concerned about the availability of energy, or are you less concerned about it? Read that, if you like, as the price of oil. The other dimension of concern is the carbon dimension, and how constrained does one want to be on carbon emissions? And by combining these two axes together you can then talk about various technologies in these two different but equally important dimensions.

Let's turn first to the transport sector, which is really about liquid hydrocarbons and being able to move people and things around. You can see that absent concern about climate emissions there is coal to liquids, going after heavy oil resources, producing oil in the Arctic, and other exotic environments which one can go after and provide the required energy resources. In the upper right, where one would like to be, there are certainly some demand-side options, hybridization of vehicles, light-weighting of vehicles, and so on, but in terms of the supply it is only advanced biofuels as contrasted with conventional biofuels that can address simultaneously the security and CO2 emissions concerns.

In the stationary sources, the power sector, it's good to look at where the world gets its energy, electricity, currently. You can see that about 40% of the world's electricity currently comes from coal, 20% from gas, 16% each from nuclear power and hydropower, and then the renewable options account for about 2% of it, and the solar and wind account for respectively about a third of a percent and 10-4 of the world's electricity right now. So going forward, the big sectors on this chart are likely to be the big sectors over the next twenty-five or thirty years. Fortunately there are some options. If you don't care about CO2 emissions it's coal, coal, coal. With increasing concern about CO2 emissions the big options are hydrogen for power, nuclear, and to some extent wind as well.

Let me make a remark about the hydrogen for power, also called carbon capture for sequestration. This is a project, the first demonstration at a commercial scale that BP has announced in southern Scotland. Gas from the North Sea is burned in such a way so that the hydrogen that is produced can be used to make electrical power, the CO2 that is produced can be pumped back under the ground where one expects that it will stay for many thousands of years, and incidentally help in enhancing oil recovery, which is an important part of the economics. This can also be practiced of course for coal. Very importantly BP has announced a demonstration plant in southern California that will do with solid carbon starting in about 2010. 90% of the emissions can be removed by these technologies while still using fossil fuels and generating electricity.

The cost of electricity is very important in thinking about alternative technologies. This shows the cost of electricity for gas and coal-fired plants. You can see that sequestering of carbon or hydrogen for power and nuclear are each about 30 to 35% more expensive. As you go to wind and then solar and biomass and so on, the costs go up significantly. Finally you can ask how much does it cost to save a ton of carbon? The lessons here are that the power options are much less expensive than the transport, which transport accounts for only 20% of carbon emissions.

There are many demand-side options which unfortunately I don't have time to cover. This is the list, starting with efficiency in each case for heat and power and transport, and then moving to various clean uses of hydrocarbons. And then what does it take to make this happen? Education of the public, education of the policymakers who often do things that are directly contrary to the facts, not only the current administration in this country, but previous administrations, and administrations in other countries, let me assure you, are equally ill informed of what the technology facts are. Business needs a reasonable expectation of the price of carbon, and then universities and laboratories have to recognize the importance of both technology and policy research for energy matters.

Thank you for your attention.