Climate models are the best tools we have for projecting the future climate. The mathematical equations used in climate models are based on well established laws of physics, such as conservation of mass, energy and momentum.

The models have been tested against and derived from a wealth of observational data from the natural world, providing a major source of confidence in the use of models for climate projection. The other basis for confidence comes from the ability of climate models to represent current and past climates, as well as observed climate changes.

Particularly important for Australia is the ability of climate models to represent large-scale patterns of temperature, pressure and precipitation, as well as modes of variability such as the El Niño-Southern Oscillation (ENSO), the Southern Annular Mode (SAM) and the Madden-Julian Oscillation (MJO), all of which affect regional climate.

In assessing regional climate change, confidence in model projections varies with scale and variable. Highest confidence occurs at the largest scales, such as global or hemispheric, and decreases at smaller scales, such as sub-continental or regional. This is partly because the magnitude of natural variability increases as scales decrease, so that regional climate change signals are more easily masked by climate ‘noise’. Furthermore, local influences on climate (such as regional topography or processes) increase in importance at smaller scales, and resolution and parameterisation limitations in models mean they may have difficulty representing these features.

The most important limitation of current models is that a number of important physical processes occur at scales too small to be explicitly resolved by the models, and therefore have to be represented in approximate form as they interact with the larger scales.

The models give us a range of possibilities for the future to account for some of the uncertainty that exists. This uncertainty arises from three sources.

Future emissions: [brief explanation]

Climate sensitivity: [brief explanation]

Difference between models: [brief explanation]