Click on image for full size
UCAR
Coupled Ocean-Atmosphere Models
Scientists use computer models to help them understand and predict the behaviors of Earth's atmosphere and oceans. These models, which run on supercomputers, are extremely complex. Some models only cover the atmosphere, while others only deal with the oceans. A third type of model combines ocean and atmospheric models into a single integrated framework in which interactions between the oceans and atmosphere are a key feature. These combined models are referred to as "coupled models".
Let's look at an example to clarify the distinction between coupled and uncoupled models. Atmospheric models that are used to assess climate change are very much concerned with the amount of carbon dioxide (CO2) in the air. Some processes, such as fossil fuel burning, add carbon dioxide to the atmosphere. Other processes remove carbon dioxide from the air. The world's oceans absorb large amounts of CO2, for instance. In an uncoupled model of the atmosphere, the portion of the model that portrays the oceans always stays the same, no matter how much CO2 the seas absorb. In a coupled model, on the other hand, the ocean model would change over time as the levels of carbon dioxide in the oceans increased. The oceans would gradually become saturated with CO2 and would thus be able to absorb less and less from the atmosphere. This change would, in turn, alter the behavior of the atmosphere part of the model.
It probably seems obvious to you that a coupled model is more realistic. Why don't scientists always use coupled models? Coupled models are much more complex, so it takes a lot more effort to make sure they are working correctly. The greater level of complexity also means that coupled models take much longer to run, even on supercomputers. Some scenarios can be modeled pretty well with uncoupled models. Other situations demand the more complex coupled models.
