User:William M. Connolley/Climate change feedbacks



Climate change feedbacks are processes in the climate system that either augment or diminish the systems response to a climate forcing. The increase in atmospheric greenhouse gases is an anthropogenic forcing that increases the temperature of the surface. This recent increase in surface temperatures and its projected continuation is called global warming. The International Panel on Climate Change in its fourth assessment reported a climate sensitivity to be likely in the range of 2 to 4.5 °C for a doubling of atmospheric carbon dioxide. The uncertainty is largely due to differences in the how different global climate models incorporate and parameterize each of the climate feedbacks.

Feedbacks can be positive or negative, but it is important to remember that a feedback subsystem never acts in isolation, but is always embedded within the overall climate system. The climate system is always subject to one very powerful negative feedback, the temperature-radiation feedback: emitted radiation rises with the fourth power of temperature. Hence, on earth the long-term gain of the overall system is always less than one, i.e. there can be no unlimited runaway effects.

Climate sensitivity
Equilibrium climate sensitivity refers to the equilibrium change in global mean near-surface air temperature that would result from a sustained change in radiative forcing. This can be expressed mathematically as $$\Delta T_{eq} = \lambda \Delta Q$$, where &Delta;Teq is the equilibrium temperature resultant from a change in radiative forcing &Delta;Q. The proportionality constant relating the two, &lambda;, is called the climate sensitivity parameter.

The fundamental climate feedback is how radiative emission changes with temperature. $$Q = \epsilon \sigma T^4$$, where &epsilon; is the emissivity and &sigma; is the Stefan–Boltzmann constant. The resultant temperature change for only this feedback can be expressed as $$\Delta T_0 = \lambda_0 \Delta Q$$. The climate sensitivity parameter from this feedback can be found by differentiation:$$\lambda_0 = (4\epsilon\sigma{T_e}^3)^{-1}$$.

This strong negative feedback is central to the changing temperature of the climate system. Therefore, when diagnosing other climate feedbacks, it is useful to express them in terms of a gain factor that accounts for this. The gain factor, g, is the change in equilibrium temperature associated with a particular feedback mechanism accounting for the fundamental blackbody feedback.

$$g = \frac{\Delta T_{eq} - \Delta T_0}{\Delta T_{eq}} = \frac{\Delta T_{feedbacks}}{\Delta T_{eq}} $$

When all the feedbacks are taken into account, the total equilibrium temperature change can be found by the temperature change with respect to the blackbody feedback and the cumulative gain.

$$\Delta T_{eq} = \frac{\Delta T_0}{1-\sum g}$$

Effects on magnitude of climate response
There are a variety of feedbacks that affect the climate system.

Cloud feedbacks
Warming is expected to change the distribution and type of clouds. Seen from below, clouds emit infrared radiation back to the surface, and so exert a warming effect; seen from above, clouds reflect sunlight and emit infrared radiation to space, and so exert a cooling effect. Cloud representations vary among global climate models, and small changes in cloud cover have a large impact on the climate. Differences in boundary layer cloud modeling schemes can lead to large differences in derived values of climate sensitivity. A model that decreases boundary layer clouds in response to global warming has a climate sensitivity twice that of a model that does not include this feedback. However, satellite data show that cloud optical thickness actually increases with increasing temperature. Whether the net effect is warming or cooling depends on details such as the type and altitude of the cloud, details that are difficult to represent in climate models.

In addition to how clouds themselves will respond to increased temperatures, there exist other feedbacks that will affect clouds properties and formation. The amount and vertical distribution of water vapor is closely links to the formation of clouds. Ice crystals have been shown to largely influence the amount of water vapor. Water vapor in the subtropical upper troposphere has been linked to the convection of water vapor and ice. Changes in subtropical humidity could provide a negative feedback that decreases the amount of water vapor which would act to mediate global climate transitions.

Changes in cloud cover are closely coupled with other feedback, including the water vapor feedback and ice albedo feedback. Changing climate is expected to alter the relationship between cloud ice and supercooled cloud water, which in turn would influence the microphysics of the cloud which would result in changes in the radiative properties of the cloud. Climate models suggest that a warming will increase fractional cloudiness. More clouds cools the climate, resulting in a negative feedback. Increasing temperatures in the polar regions is expected in increase the amount of low-level clouds, whose stratification prevents the convection of moisture to upper levels. This feedback would partically cancel the increased surface warming due to the cloudiness.

Water vapor feedback
From the Clausius–Clapeyron relation, it is known that the saturation vapor pressure over a flat surface of water increases exponentially with temperature. An increase in temperature of 3°C (5.4°F) above present conditions would lead to an increase of 20% in the saturation vapor pressure. Since water vapor is a greenhouse gas, the changes in atmospheric water vapor content are important for projecting future temperatures. Climate sensitivity is doubled in the presence of a water vapor feedback if the relative humidity is held constant. Climate models and observations agree that the relative humidity is constant at interannual time scales. Recent modeling results suggests that although this feedback process causes an increase in the absolute moisture content of the air, the relative humidity may actually decreases slightly because the warmer air.

It is not known if relative humidity will stay constant with the changing mean climate. Hypotheses suggesting that the relative humidity will be altered have been suggested. Using data from the Earth Radiation Budget Experiment, a comparison was made between the surface temperature and the outgoing longwave radiation at the top of the atmosphere, and found to be consistent with the hypothesis of fixed relative humidity.

The lack of reliable instrumentation to measure the amount of water vapor and its spatial variation is the leading reason why there is uncertainty in the water vapor feedback. Efforts have been made to use the current radiosonde network, but it is unable to produce accurate and precise measurements of water vapor in the upper troposphere needed to diagnose the water vapor feedback. An observing network with an accuracy to measure decadal trends, and to test whether the increase in water vapor content is consistent with models has been proposed, but not implemented.

Lapse rate feedback
The atmospheric lapse rate refers to the decrease in temperature above the surface of the Earth. The numerical value is a function of convective, radiative, and large-scale dynamical processes. Convection and dynamics generally transport heat from the surface upwards to space, while radiation warms the surface and cools the atmosphere. The lapse rate in the tropics is observed to be near the moist adiabatic lapse rate due to the high moisture content. This is the rate that saturated parcels will cool when lifted adiabatically. Models and observations show that with an increased surface temperature, the moist adiabatic lapse rate will decrease. By itself, theory says the lapse rate feedback is negative in the tropics.

It has been shown that if the relative humidity remains constant in a warming atmosphere, then the lapse rate feedback and the water vapor feedback partially cancel each other. The changes in vertical temperature structure are one of the tools that climate scientists use to attribute climate change to greenhouse forcing or natural variability. Climate models which model convection and radiation generally produce the observed lapse rate changes. Temperatures in the upper atmosphere are difficult to accurately measure, and there have been minor inconsistencies between observations and the models. The radiosonde data has been shown to contain biases that are difficult to remove.

The United States National Research Council has recommended that radiosonde observations must be sustained and improved with the objective to remove biases that will allow them to be used for the monitoring of long-term climate.

Sea ice feedbacks
At high latitudes, temperatures get low enough that sea water freezes into ice. Sea ice strongly impacts the climate through two different positive feedbacks. First is the sea ice-albedo feedback (or ice-albedo feedback). As temperatures increase, the sea ice melts, resulting in a lower surface albedo, and increasing the amount of solar radiation absorbed, which reinforces the initial warming. Reviews of various models show that the magnitude of this feedback is uncertain due to differences in ice dynamics, and in their treatment of clouds. The albedo has been shown to change in two ways: changes in the areal extent of the ice, and changes in the multi-year ice such as ice thickness, and melt pond characteristics. Sea ice models that incorporate the multi-year ice feedback are more sensitive to initial changes in temperature.

Sea ice also acts as an insulator between the ocean and the atmosphere. As the ice melts, it allows more liquid water to evaporate into the air. This latent heat flux from the ocean to the atmosphere results in additional warming of the boundary layer. Melting of sea ice also results in an increase in the poleward moisture flux from the tropical oceans.

Several climate models suggest that there may be a negative feedback between sea ice and the meridional overturning circulation. The meridional overturning in the Southern Hemisphere westerlies may delay the response to the radiative forcing from greenhouse gases by upwelling deep, cold, unmodified water. In another experiment, the higher sea surface temperatures resulting from greenhouse warming lead to more evaporation than precipitation, resulting in a higher ocean salinity. The changes in salinity cancel the effects of changes in sea surface temperatures so that there is no net change in the circulation. The Antarctic subpolar gyre is an important process in determining the thickness of Antarctic sea ice. Changes in the meridional overturning circulation will feedback into changing the deep ocean heat content and sea ice thickness.