İstanbul, May 5, 2005

DAVE STAINFORTH

Oxford University

Good evening Mr. Chairman, esteemed guests, ladies and gentleman. It is a great honour for me to be able to talk to you this evening. My thanks go to the organizers of Forum Istanbul for including climate change as one of the issues which should be considered when looking to the future of Turkey.

I have never come across a meeting similar to this in the UK or elsewhere; a meeting able to look to the future over a relatively long period: almost 20 years if we consider 2023, Turkey's 100th birthday. But there is tremendous value in a country looking forward and considering its long term development strategies in open discussion. And to do so is both exciting and, I believe, serves the country well.

I believe the emphasis on the knowledge society, or the use of knowledge in the economy, or whatever term is deemed acceptable, is well placed. Access to new methods of communication is changing the world; and the availability of, and more critically the means of filtering and understanding vast quantities of information will be a significant issue in the next 20 years.

In particular I think there will be an increasing need to not only find information but also to understand its context and reliability and thus convert information into knowledge. Methods of doing this present a substantial challenge; a challenge which R&D and innovation will help tackle but which will ultimately only be solved by the development of new skills. So education is the key and along with that the availability of infrastructure which ensures that everyone, but particularly young people, have access to the full range of new information sources.

But my area of expertise is climate; and particularly climate change and that is what I'm here to talk about.

Over the next 100 years the earth's climate is going to change very substantially from that which has been experienced over the last 100 or even 1000 years. Even by 2023, when Turkey reaches its 100th birthday, the changes by comparison to 1923 will be very substantial and will be having major influences on societies worldwide and on the international social and commercial environment. Turkey is well placed to be able to use science (and by science I mean everything from the physical sciences to the social sciences and economics) to position itself well in this world of the future.

So I'm going to discuss 3 things. First I'm going to talk about climate change science in general. Then I'll talk about some exciting work going on in Oxford regarding issues of uncertainty and confidence in predictions of climate change. This is particularly important because making decisions in the context of some degree of uncertainty is an increasingly important skill in a number of areas. And finally I'll give you my views on what this might mean for nation states and particularly Turkey.

1- Is Climate Change Real?

So let me start off by addressing the issue of whether anthropogenic climate change is real? That is, is the earth's climate changing, and if it is, is it changing as a result of mankind's activities? The answer to both these questions is almost certainly yes. Let me explain why I can say that.

To start with observations of the earth's climate are already showing a wide range of changes.

Over the last century the global average temperature (over land and sea and in the surface waters of the oceans) has risen by 0.6ºC. This rise is the most dramatic rise for over 1000 years in the Northern Hemisphere. In southern hemisphere we don't have the necessary information to make the comparison. Furthermore, the 10 warmest years on record have all been since 1990. 1998 was the warmest on record and over Europe 2003 was even more extreme.

Mountain glaciers are retreating worldwide. The extent of winter snow cover in the Northern Hemisphere is reducing and the spring retreat of ice in lakes & rivers is taking place roughly 2 weeks sooner than previously. Northern Hemisphere rain and snowfall has increased by 5-10% and extreme events have become more frequent.

These changes are already affecting societies. The severe floods in Europe in 2002 had an estimated cost of $16 billion. And the 2003 European heat wave is said to have resulted in 26,000 premature deaths and cost $13.5 billion.

It is at this point I must put the brakes on what I'm saying, just for a moment. Am I claiming that the 2003 European heat wave was a direct result of mankind's emissions of greenhouse gases? No. There have always been floods and heat-waves. They occur because there are, and have always been, large natural variations in the weather from year to year. Whether one happens this year, or next year, or the year after, is down to bad luck; the roll of a dice. But mankind's activities are changing the climate. They're changing the likelihood of certain events. They're changing the way the dice is loaded.

Climate is usually defined as the long-term averages of weather. It is the weather we expect for any given season or region. One way to look at it is "climate is what you expect and weather is what you get". So you expect it to be hot in summer, but you don't expect a heatwave. Nevertheless heat-waves have always happened occasionally because of natural variability.

But mankind is loading the dice to make some things more likely and others less. Recent research has shown that more than half the risk of the 2003 heat-wave was due to human influence; mainly greenhouse gas emissions. That is to say that such an event is twice as likely to occur now, than it was a couple of hundred years ago. The climate has changed. And recent research suggests that while such a heat-wave was perhaps a one in a thousand year event in the past, by the middle of this century it is likely to happen roughly once in every two years.

It's a sobering thought and such thoughts are already substantially influencing the insurance industry. As damaging events become more frequent, premiums go up and some things become uninsurable.

This is even affecting individuals in Britain because insurance companies are beginning to refuse to insure some properties; some homes. That may not sound terribly important but that effectively means that some people can not sell their home because nobody can get a loan to buy it because banks will only lend money to buy properties which insurance companies will insure, So climate change is already affecting individuals in the UK by making their most valuable asset worthless. And thus starting to reduce mobility in the labour market.

The whole of climate science is beset by these issues of probability and statistics. I said earlier that mountain glaciers were retreating worldwide. That is true. It doesn't mean that every mountain glacier is retreating but the vast majority are. Those that aren't tend to be in regions which are seeing other changes in the form of increased rain and snowfall. I'll come back to this issue in a moment because it pertains to how the issue is communicated and understood.

Anthropogenic Climate Change:

But first, it is clear that our climate is changing in an unprecedented fashion but why do we think that mankind is to blame?

Well fundamentally the science is quite simple. The greenhouse effect is so called because it behaves like a greenhouse. Increasing the levels of CO2 and other greenhouse gases simply traps more heat and makes the temperatures rise. This is what we would expect at a very fundamental level so in one sense the onus is on those who don't believe we have a problem, the so called "climate sceptics", to show that there are feedbacks, that is to say other consequential changes, which counteract this effect.

It is certainly true that increasing the level of atmospheric greenhouse gases causes a large number of complex, interacting effects. Since the industrial revolution atmospheric levels of CO2 lave increased by about 35% above the level they were at for many thousands of years. This has direct effects on climatic processes such as cloud and rain formation and it has indirect effects via the increasing temperatures. These consequences may either worsen or lessen the impacts. To study them we use highly complex computer models of the climate system which lave been developed over a decade or more.

A handful of these complex models have been developed worldwide. They represent our understanding of the many different physical and chemical interactions in the climate system, and they demand substantial computing resources. Indeed they tend to be run on the most powerful supercomputers such as the Earth Simulator in Japan. This is not surprising since amongst many other calculation, these models must solve 6 interacting equations at each of at least 200,000 points around the world just to work out how things change in half an hour. The whole process has then to be repeated over a million times to simulate a century.

Over the last 5 years these models have developed to such complexity that they can do a pretty good job at simulating present day weather; and some of them are also used as weather forecasting tools.

They have also been used to simulate the climate of the last century. What they show is that it is only possible to reproduce the changing temperatures of the last 100 years if we include both natural effects and mankind's emissions of greenhouse gases. If either is left out then the comparison with observations fails. This has enabled scientists to separate the effects of natural factors such as volcanic eruptions and solar activity from anthropogenic factors. The robust conclusion is that a moderate warming at the beginning of the century is largely due to natural factors but the dramatic changes seen since 1950 is largely due to mankind's influence.

Climate Sceptics:

Now I'd like to say a few words about the confidence we have that climate is changing and that mankind is the principle culprit. I've just described the scientific basis for this confidence.

But often when climate change is in the news it is presented as a balance between some scientists who think anthropogenic climate change is real and a serious problem, and others who don't. It is important to realise that the vast majority of scientists, as represented by the InterGovernmental Panel on Climate Change, the IPCC, are clear that it is real and a problem. They concluded that "most of this warning over the last 50 years is due to human activities". Over 500 scientists from all around the world were involved in that report.

The number of scientists who are specialists in this field and disagree is very very small. There is almost scientific consensus on this general issue.

There is however a problem that scientists don't tend to talk about what they agree on. And there are many many issues in climate which are still not well understood e.g. how can we give probabilities for future climate, just how large will the warming be, what will happen in specific regions and localities, what will happen to different species of wildlife and industrial sectors, and many many others?

So there is lots of disagreement and that unfortunately can lead to confusion for the general public and for decision makers. Furthermore, the probabilistic nature of climate means it is easy to find examples which seem to contradict the claims of climate change. For instance, as I mentioned earlier, not all glaciers are retreating so it is easy for the non-specialist to pick one of the few that isn't and claim there is nothing to worry about.

This is a major problem in trying to tackle climate change. The consequence of some misrepresentations is that the whole issue of whether there is actually any cause for concern is sometimes seen as uncertain. While in fact there is very widespread scientific confidence in the cause for concern if not in many details of what will actually happen.

2- Climateprediction.net

So now I'd like to take a couple of minutes telling you about the climateprediction.net project. This is a project which I set-up with a colleague of mine, Dr. Myles Allen, and the help and support of many other individuals.

I have emphasized that predicting climate is a matter of probabilities rather than absolutes. Unfortunately to study those probabilities using complex climate models requires vastly more computing power than is available to any climate science project anywhere in the world. What we need to do is run 10s of thousands of slightly different versions of these models.

That would be impractical on even the largest supercomputer. But personal computers are now so powerful, largely due to the games industry, that these models can be run on the humble PC. Okay they take rather longer, perhaps 3 weeks rather than a few days to simulate 50 years. But this is a long term problem. I think we can wait a few more months for the results.

So climateprediction.net was set up to encourage businesses and the general public worldwide to download one of these models and run it in the background on their PC while they are using the machine for other things such as writing presentations, developing spreadsheets or even playing games. When they've simulated 50 years they return the data to us.

That way we can do the experiment we need. We can explore the range of future possibilities consistent with the observations of the past. And at the same time we can engage people in the science o climate change.

The response has been phenomenal. Since the launch of the software in September 2003, over 100,000 people from 150 countries have participated. More than 8000 years of computing time have been donated. People have not only been generous but tremendously interested with very active web-based discussions. And colleagues have developed complementary materials for schools and higher education courses.

We published the first results in Nature just 3 months ago. And those results, though not surprising, are shocking. We found no evidence that previous predictions overestimated the likely consequences of greenhouse gas emissions in terms of average global temperature. But we found that we could not rule out the possibility that the long term warming could be much greater than was previously thought likely. In fact this reinforced various other studies over recent years but by using complex models we give further credibility to the result and can begin to study the range of possible futures at the regional as well as the global level.

And what are the numbers? Well most specialists believe that concentrations of atmospheric CO2 will reach twice pre-industrial levels by the middle of this century. We found that even if we manage to stop concentrations going above this level but instead maintain them at this level, the long term change in global average temperature still has a large range of uncertainty. Anywhere from 2oC right up to 11oC.

It would be wrong to take from this that because there is a large range of uncertainty, that the science is not trustworthy. If I role a dice I know I will get a number between 1 and 6. And if the dice is not loaded then I know the probability of each number is about 17%, one sixth. The science is clear and absolutely understood. But when I role the dice it could still give a number anywhere between I and 6, a large range of uncertainty. We have the same situation here. The issue, however, is what are the consequences of different outcomes? If rolling a I means I loose half of everything I own I may decide not to role the dice, even if that costs me $100.

Future research may help make more accurate climate forecasts but even the 2-11 degree range gives some important and policy relevant information. Firstly we have no evidence that the response to reaching double pre-industrial levels will be much less than 2oC. So the imperative to take action is absolutely clear. Secondly there is a risk of a very extreme response. Societies need to decide what risks they are willing to accept. It is however clear that the states are high.

Let me give you some details from climateprediction.net for this region. For the Mediterranean basin the long term warming is also likely to be in the 2-11 range as an annual average but there are indications that the balance might be skewed to greater warming in summer. I conjunction with this the summer is likely to get drier by anywhere from 10 to 70% but we can't yet say whether winters will get drier or wetter.

3- Implications for Turkey:

To finish up let me give you my view of the implications of climate change for Turkey and for Forum Istanbul.

Education and Leapfrogging:

There has been some discussion of the benefits of leapfrogging. That is there is no need for emerging or developing economics to go through the same intermediate stages as those which western economies went through. This is a tremendously important concept which others know far more about than I.

I would simply emphasize that global society in 50 or even 20 years time will look very different today. It has changed massively in the last 50 years and there's no reason not to expect it to change similarly in the next. Climate change provides additional evidence that this change is unavoidable.

So when examining the opportunities for leapfrogging it is important to concentrate on developing systems appropriate to 20 years in the future, not today.

There is much scientific, engineering and economic expertise available to help do this. It is not just crystal ball gazing; particularly when considering climate. But it is critical to have wide-ranging expertise to provide inputs to local and national governments and to industry and commerce. Consequently education and R&D in both the private and the public sectors are fundamental to the development of robust, long-term strategies.

Indeed expertise may well turn out to be the corner stone of the societies of the future. Understanding and processing complex and to some extent uncertain information to make robust and sound judgements is the challenge for us all. It influences societies at all levels from governmental policy to individual choices, in subjects as diverse as agriculture, security and health. And in climate change it is fundamental.

However, if science is to be used to inform and guide policy decisions then in democratic societies the general public as well as decision makers need to be able to understand that science and its relevance to them. Education, including life long education, and an informed media is therefore a basic requirement.

This is particularly true when addressing the issue of climate charge because climate change requires planning decisions based on timescales much longer the typical business, governmental or political cycle.

The Reasons for Action:

What are the reasons to act on a national level?

Climate will affect the world of the future in 2 ways. Firstly, attempts to limit the degree of climate change, so called "mitigation", will lead to different commercial frameworks. Secondly, the actual changes in climate will require people, nations and businesses to adapt.

We already have an unprecedented international agreement in the form of the Kyoto protocol. Whether the form of that agreement is right or wrong is not important for us here tonight. But as time goes by and evidence of climate change becomes clearer and clearer it seems almost certain to me that in 20 years time there will be further agreements in place influencing nations and/or industry and trade. My guess is that the path to such agreements is likely to be a difficult one and it may at times look as though none will come through. But at some stage or another they are likely to appear.

Preparing a nation to comply with such agreements, and to develop the many opportunities afforded by such agreements, is likely to bring many benefits. And let's be clear, there will be many opportunities. A nation with industry and expertise to respond to the new requirements will be well placed to meet such agreements and to export.

We are already committed to a certain level of climate change because it takes some time for the climate to respond to changed levels of greenhouse gases. So there will also be a need for societies to adapt. Forecasts of how climate change will affect regions and commercial sectors could be extremely valuable for that. Over the next 5 years I believe there will be substantial developments in the ability of science to provide such forecast.

A third potential, and much more uncertain, driver of the need to take action on climate change is the possibility of legal action against organisations which cause climate change. There are now a number of cases being pursued in the United States and the science required to make such a case is developing quickly.