Human activities which contribute to an increase in greenhouse gases in the atmosphere: However, it is difficult to model such complex systems as global climate change. This leads to simplified models, speculation and opinions presented in the media that may be based on only parts of the evidence and which may be biased.

The idea

Global climate is a complex system: the atmosphere, oceans, ice, land and living things interact, and changes feed back on one another. Because so many variables interact over long timescales, computer models must simplify reality, for example by using grid boxes of hundreds of kilometres or by averaging ocean mixing. Such simplifications make predictions uncertain, so scientists compare model outputs with observations and with each other. When this science reaches the media, reports may quote only one study, ignore uncertainty ranges, or select evidence supporting a particular viewpoint, producing speculation or bias. A useful method is to ask: what evidence is used, what is left out, who benefits from the claim, and does the source cite peer-reviewed data?

What earns the marks

• Complex systems such as global climate involve many interacting components, including the atmosphere, oceans, ice sheets, land surfaces and living organisms, so a single simple cause-and-effect chain is inadequate.
• Models are simplified representations: they divide the world into grid boxes, use averaged values and approximate processes, so they cannot reproduce every local detail or every feedback.
• Simplification creates uncertainty in predictions, which is why scientists test models against past and present observations and report ranges rather than single definite numbers.
• Media reports may present only part of the evidence, for example one dramatic study, and may omit uncertainty, contrary evidence or the wider scientific context.
• Speculation and opinion can be presented as established fact, and bias can arise from commercial, political or campaigning interests, so the same evidence may be framed in different ways.
• Evaluating a claim involves checking the evidence base, the source and its interests, whether other scientists agree, and whether the language matches the strength of the data.

Common mistakes and how to put them right

• Saying that because models are simplified they are worthless.

  Put it right: Simplified models are still useful because they are tested against observations and can project broad trends, even though local details are uncertain.

• Treating a media headline as if it were the scientific evidence itself.

  Put it right: The headline is a summary; the underlying peer-reviewed data and its uncertainty must be examined separately.

• Assuming bias always means the evidence is false.

  Put it right: Bias means evidence is selected or framed in a particular way; the underlying data may still be valid but incomplete.

Exam tips

• When asked why climate models are difficult, name at least two specific complexities, such as ocean heat uptake or cloud feedback, rather than writing only that climate is complicated.
• Use the phrase 'simplified model' and explain one simplification, such as dividing the atmosphere into grid boxes, to show understanding rather than repeating the statement.
• For media questions, structure your answer around evidence used, evidence omitted and possible source bias, and avoid claiming that all media reporting is wrong.