BBC HomeExplore the BBC
This page has been archived and is no longer updated. Find out more about page archiving.

24 September 2014

BBC Homepage

In Hot Topics:


Contact Us

You are here: BBC > Science & Nature >

Distributed computing and climate prediction

The BBC teamed up with Oxford University to conduct the world's most ambitious climate modelling experiment. The aim? To match or exceed the power of a supercomputer, using a technique known as distributed computing.

How does distributed computing work?

Modern home computers can perform billions of calculations a second. Most of the time, that's far more power than the average user needs – so even though you're working hard, your computer is just ticking over. Distributed computing projects make use of this spare computer potential.

Distributed computing is a particularly valuable tool for scientists who have large amounts of data to analyse, or who are modelling very complex systems like the Earth's climate.

Why is climate prediction so complicated?

Greenhouse gases affect how much of the Sun's energy the Earth loses back to space.

Predicting global temperature change is hard, even though the principle sounds easy. In simple terms, energy reaches Earth from the Sun. Some of it is immediately reflected. Some is absorbed and then re-emitted. If the amount of energy that leaves the Earth is the same as the energy that arrives, then temperature stays the same. If not, then the Earth's temperature changes.

A huge number of factors affect how much energy the Earth reflects and absorbs. How much of the planet is covered in clouds – and what kind of clouds are they? How much ice is there at the poles? And of course, the amounts of so-called greenhouse gases like carbon dioxide play a role too.

All these factors make for an incredibly complex calculation. However, there's something else that makes climate prediction even harder: feedback mechanisms.

What are feedback mechanisms?

In relation to climate change, a feedback mechanism is something affected by climate change, which itself makes climate change happen more or less quickly.

Melting ice caps could accelerate global warming as less light is reflected back to space.

For example, heating the Earth could make the ice caps melt, which could mean that less of the Sun's light is reflected back into space, which could in turn cause the Earth's temperature to rise even faster.

Some feedback mechanisms could slow climate change – some could accelerate it. Either way, they make prediction harder and mean that scientists need to run many more models to get a feel for what is likely to happen.

Why so many people?

This experiment uses a computer model to try to calculate what the climate will be like in the future. However, small changes to the model can have large effects on the predictions that result.

There's only one way to get around this problem. The results of a single model cannot be gauged for accuracy. From many thousands of models, patterns emerge. Some results will be wildly inaccurate and predict warming or cooling outside what the Earth is likely to see. But if a significant percentage of results fall within a smaller range, the researchers can get a feel for how the climate might be changing.

Of course, it's not just about the state of the planet now. The biggest question of all is the effect that mankind is having on the climate. What happens if we continue to pump out greenhouse gases at the same rate as we are today? What happens if emissions grow? And by how much do we need to cut emissions to have an impact on global warming?

The model shows variables such as temperature, pressure, precipitation and cloud cover.

What does each computer do?

In the BBC experiment, each computer ran one individual climate model. The model calculated the climate, year by year, from 1920 right through to 2080.

Calculating the climate for the 20th century might sound like an odd thing to do. It's a check of the validity of the model's parameters. If the model's prediction for 2006 is very inaccurate (for example, if the whole world turned to ice), then the model is rejected. But if not, the model continues into the 21st century.

What next?

By the end of 2006, thousands of people had finished running their model up to the year 2080 and their data contributed to a BBC One programme in January 2007. You can watch online video, see UK climate prediction graphs or analyse the results yourself at bbc.co.uk/sn/climateexperiment.

The BBC and the Oxford University research team are no longer running the joint climate change experiment. It has closed to new participants.

If you have already installed the BBC climate experiment on your computer, it will continue to run to completion. There is no deadline and every day that your computer is running adds to the research. Whenever your individual model finishes, its results will be sent to Oxford and will contribute to the climate science.

The Oxford team is still welcoming volunteers to their own experiments but – please note – theirs are aimed at experienced computer users. They are available for download from climateprediction.net.



Science Homepage | Nature Homepage
Wildlife Finder | Prehistoric Life | Human Body & Mind | Space
Go to top



About the BBC | Help | Terms of Use | Privacy & Cookies Policy