Dr Geoff Jenkins is a climatologist working at the Hadley Centre for Climate Prediction and Research, part of the Met Office. Dr Jenkins is an expert on the models used to predict future climate.

Could you explain the recent trends in temperature change? Is what we find unusual?
Over the last 140 years or so since good global measurements have been made we’ve seen a temperature rise of something like 0.6 or 0.7 degrees Celsius. Some of that occurred during the earlier part of the century and we don’t necessarily believe that much of that is due to human activity, but also there’s been a large surge in temperature since the mid-1970s and we believe that a substantial part of that may be manmade. Going back over the last 1000 years or so using proxy records from ice-cools and tree rings and so on, that sort of temp rise hasn’t been seen before over that period so it does appear to be exceptional in the context of the last 1000 years at least.

What makes you believe the recent rise is due to human activity?
Feeding in the different agents that cause climate change into our models – like greenhouse gases, output from the Sun, volcanoes - we’ve looked at the patterns of change they cause across the surface of the Earth and through the atmosphere. We compare them to what’s actually been observed and find the best match between computer simulations and the observations. This has indicated to us that over the past 30 or 40 years that most of the warming has been due to human activities.

Do you therefore believe temperature rise to be directly linked to CO2 emitted by human activity?
Some people maintain that the reason that CO2 has gone up in the atmosphere in the past 200 years or so is because the Earth has warmed, due to solar changes for example, and this warming has acted to put more CO2 from the natural carbon cycle back into the atmosphere. However, by looking in detail at the isotopic composition of the CO2 we can determine that that has come from human activities and isn’t natural. We believe that the simple explanation is the correct one, that it is CO2 that’s been increasing due to human activities, and that has led to the greenhouse effect, to an increase in temperature.

Can you explain the ‘greenhouse effect’?
There are natural gases in the Earth’s atmosphere, particularly water vapour, which trap heat and make the Earth warmer than it would otherwise be – this is known as the ‘greenhouse effect’. So these natural gases already keep us something like 33 degrees warmer than we would be and indeed make life on earth possible. The concern comes from increasing amounts of these gases and new greenhouse gases which will exacerbate the problem. When the amount of CO2 rises, as it has done due to fossil fuel burning and deforestation, it causes warming. As the temperature of the atmosphere rises it can hold more water vapour, which is a greenhouse gas, so you get positive feedback and the Earth gets warmer and warmer.

How do you respond to accusations of inaccurate temperature measurements?
We’ve made temperature measurements across the surface of the Earth, good temperature measurements, for about the past 140 years or so. We’ve got 3 independent measurements for temperature - we’ve got the temperature over land; we’ve got the temperature of the sea surface; and also we’ve got the temperature of the air over the sea that are made by thermometers on ships bridges. We’re pretty confident the global average temperature record over this period is pretty sound because we have these different types of measurements which all in the long term fall into line, although there are differences over years and decades where they don’t agree. We don’t fully understand the difference for that short period disagreement, but I stress that over the long period they all pretty much fall into line.

What about recent work that questions the link between CO2 and warming?
We’ve from very good records from French and Russian work going back about half a million years that there’s a very close correlation between CO2 and temp. If we go back very much further than that and we’re talking now about hundreds of millions of years, then what we have to remember is that the Earth was in a very different condition from what it is now, the land masses were in different places, the ocean currents were very different, and it’s very difficult to apply any of the knowledge that’s been gained over that very long term to the situation that obtains today, because the conditions were so very different.

Is global warming causing more extreme weather conditions – for example the floods in Britain in winter 2000?
We can’t point to one event and say that’s due to global warming. However, what we can say is that we are beginning to see more and more events, and this is line with what the climate models will predict and we would expect to see more of them in the future. So it’s cautious, but I think it’s not unreasonable to link the two together. We don’t have any robust predictions of increases in major catastrophic events like hurricanes and cyclones and so on in the future, that varies from model to model. In principle because the atmosphere has got more energy in it we might expect to see more of those, but getting a robust conclusion from models hasn’t been straightforward.

Everyone agrees climate is very complex. How can it be accurately represented in models?
The mathematical models that we use to try to simulate the complex climate system have to make some what we call prioritisations or representations of processes that go on at very small scales, in clouds and so on. And therefore there is some simplification involved. But we believe that we can represent the most important processes and indeed when we compare the model results over the last 150 years with the observations, we can see that we can do a relatively good job of simulating that. We can also test the model against data from many thousands of years ago when the Earth’s climate was in a different state and again the models do a reasonably good job of simulating those.

There have also been some views that the strength of the water vapour feedback that we have in the climate models is wrong, that the water vapour feedback is quite an important enhancement of the greenhouse effect, the basic greenhouse effect due to CO2 is probably doubled by the effect of water vapour. And there are uncertainties about, but we believe again from the tests we can do looking at the way the model responds and looking at satellite information about water vapour that we can say that the model roughly gets it right. Nonetheless, there are large uncertainties and we may find that further work in the atmosphere and so on does lead to a modification of some of these views.

How can policy makers use the models if there are uncertainties in them?
We have a diversity of predictions from these climate models. But it sometimes can be quite difficult for people to deal with this, people for example who want to design flood defences will say, ‘Which model is right, what’s actually going to happen, is it going to be x or y?’ We’re trying to address that by instead of giving them single predictions, of giving them a distribution of possible outcomes, a probability distribution, and we’re going to be doing that by building perhaps 10 000 or 100 000 climate models, all of which have slightly different ways of representing processes in the atmosphere, all of which are plausible, but are each slightly different. We then run those models to produce a climate prediction and from that we can build up a probability distribution of one particular type of climate change. So if someone says to us ‘what’s the change in rainfall over southern England going to be by 2100’, we can say not ‘it’s going to be x%’, but we can say that there’s an x% probability that it’s going to be 5mm or 10mm or whatever it is. And that’s very much what users want in doing risk assessments and deciding how much to build in the way of flood walls or coastal defences.

We believe this is the way forward because of course the answer isn’t a black or white one, we don’t have one figure with zero uncertainty, we never will have zero uncertainty in our climate predictions. So in some ways it’s more important to know what the uncertainties are and what the distribution of probabilities are, rather than try to keep homing in on one figure which we’re never going to achieve.

  < previous   next > Page 3 of 7