Climate models attempt to simulate the behaviour of the climate. The ultimate objective is to understand the key physical, chemical and biological processes which govern climate. Through understanding the climate system, it is possible to obtain a clearer picture of past climates by comparison with observational data, and to predict future climate change. Models can be used to simulate climate on a variety of spatial and temporal scales. The basic laws and other relationships necessary to model the climate system are expressed as a series of equations. Because modelling the climate is so complex, supercomputers are used for the task.
Modelling the Climatic Response
Climate models first use mathematical equations to determine a "base" climate. For models estimating future climate change, the "base" climate would normally be the present climate conditions of the Earth. Estimates of future increases in greenhouse gases, for example, are then inputted into the model, which then calculates how the climate might "evolve" or respond in the future. Climate models can also be used to model the climate of past ages, for which we have very little observational records.
Simplifying the Climate System
All models must simplify what is a very complex climate system. This is in part due to the limited understanding that exists of the climate system, and partly the result of computational restraints. The simplest models are zero-dimensional. This means that the state of the climate system is defined by a single global average. Other models include an ever-increasing dimensional complexity, from 1-D, 2-D and finally to 3-D models. With these models, further simplification takes place in terms of geographical resolution. For example, there will be a limited number of latitude bands in a 1-D model, and a limited number of gridpoints in a 2-D model, since the computer is not powerful enough to determine the climate at every point on the Earth. Similarly, the time resolution of climate models varies substantially, since the computer could not calculate the state of the climate every second. Typical models which forecast future changes in the global climate have a time resolution of months to years.
Given their stage of development, and the limitations imposed by incomplete understanding of the climate system and computational constraints, climate models cannot yet be considered as predictive tools of future climate change. They can, however, offer a valuable window on the workings of the climate system, and of the processes which have influenced both past and present climate.
The Climate Models
It is often convenient to regard climate models as belonging to one of four main categories:
1) energy balance models (EBMs);
2) one dimensional radiative-convective models (RCMs);
3) two-dimensional statistical-dynamical models (SDMs);
4) three-dimensional general circulation models (GCMs).
These models increase in complexity, from first to last, in the degree to which they simulate the particular processes and in their temporal and spatial resolution.
It is not always necessary or beneficial to the analysis, however, to invariably choose the more sophisticated models. The choice of model depends upon the nature of the analysis. For example, the simpler models, unlike the 3-D GCMs, may be run many times in sensitivity studies, which test the influence of modelling assumptions. For simulation experiments, which require complex modelling of the physical, chemical and biological processes inherent in the climate system, more sophisticated models may indeed be more appropriate. Computational cost is always an important factor to consider when choosing a climate model.
Confidence and Validation
Although climate models should aid understanding in the processes which govern the climate, the confidence placed in such models should always be questioned. Critically, it should be remembered that all climate models represent a simplification of the climate system, a system that indeed may ultimately prove to be too complex to model accurately. Model performance can be tested through the simulations of shorter time scale processes, but short-term performance may not necessarily reflect long-range accuracy.
Climate models must therefore be used with care and their results interpreted with due caution. Margins of uncertainty must be attached to any model projection. Uncertainty margins can be derived by the comparison of the results of different model experiments or through sensitivity studies, in which key assumptions are altered to determine the role they play in influencing the final climatic response. Validation of climate models (testing against real-world data) provides the only objective test of model performance.