Improved atmospheric and oceanic modules of coupled climate models, especially improved representation of clouds, parametrizations of boundary layer and ocean mixing, and increased grid resolution, have helped reduce and often eliminate the need for flux adjustment in some coupled climate models. This has not reduced the range of sensitivities in projection experiments.

There is a growing appreciation of the importance of the stratosphere, particularly the lower stratosphere, in the climate system. Since the mass of the stratosphere represents only about 10 to 20% of the atmospheric mass, the traditional view has been that the stratosphere can play only a limited role in climate change. However, this view is changing. The transport and distribution of radiatively active constituents, especially water vapour and ozone, are important for radiative forcing. Moreover, waves generated in the troposphere propagate into the stratosphere and are absorbed, so that stratospheric changes alter where and how they are absorbed, and effects can extend downward into the troposphere.

Observational records suggest that the atmosphere may exhibit specific regimes which characterise the climate on a regional to hemispheric scale. Climate change may thus manifest itself both as shifting means as well as changing preference of specific regimes. Examples are the North Atlantic Oscillation (NAO) index, which shows a bias toward positive values for the last 30 years, and the climate “shift” in the tropical Pacific at around 1976.

While considerable advances have been made in improving feedbacks and coupled processes and their depiction in models, the emergence of the role of natural modes of the climate system such as the El Niño-Southern Oscillation (ENSO) and NAO as key determinants of regional climate change, and possibly also shifts, has led to an increase in uncertainty in those aspects of climate change that critically depend on regional changes. It is encouraging that the most advanced models exhibit natural variability that resembles the most important modes such as ENSO and NAO.

The coupled ocean-atmosphere system contains important non-linearities which give rise to a multiplicity of states of the Atlantic thermohaline circulation (THC). Most climate models respond to global warming by a reduction of the Atlantic THC. A complete shut-down of the THC in response to continued warming cannot be excluded and would occur if certain thresholds are crossed. Models have identified the maximum strength of greenhouse gas induced forcing and the rate of increase as thresholds for the maintenance of the THC in the Atlantic ocean, an important process influencing the climate of the Northern Hemisphere. While such thresholds have been found in a variety of fundamentally different models, suggesting that their existence in the climate system is a robust result, we cannot yet determine with accuracy the values of these thresholds, because they crucially depend on the response of the atmospheric hydrological cycle to climate change.

The representation of sea-ice dynamics and sub-grid scale processes in coupled models has improved significantly, which is an important prerequisite for a better understanding of, the current variability in, and a more accurate prediction of future changes in polar sea-ice cover and atmosphere-ocean interaction in areas of deep water formation.

Recent model simulations, including new land-surface parametrizations and field observations, strongly indicate that large-scale changes in land use can lead to significant impacts on the regional climate. The terrestrial carbon and water cycles are also linked through vegetation physiology, which regulates the ratio of carbon dioxide (CO₂) uptake (photosynthesis) to water loss (evapotranspiration). As a result, vegetation water-use efficiency is likely to change with increasing atmospheric CO₂, leading to a reduction in evapotranspiration over densely vegetated areas. Tropical deforestation, in particular, is associated with local warming and drying. However, realistic land-use change scenarios for the next 50 to 100 years are not expected to give rise to global scale climate changes comparable to those resulting from greenhouse gas warming.

IPCC Homepage