Cloud Feedbacks in Climate Models and Their Influence on Climate Sensitivity

The energy balance model of Earth's climate system is useful to assess the influences of particular processes on global mean surface temperature (T) and to compare these influences across different climate models. Within this energy balance framework, the time-dependent change in heat content of the climate system, AQ, is related to radiative forcing, AF, and the change in T, AT, as:

dt a where A is the equilibrium climate sensitivity, as is readily seen by considering a system in a new equilibrium in response to a forcing AF, for which d(AQ)/dt = 0. Hence,

where ATeq is the temperature difference between two equilibrium states. At present, climate sensitivity is not well constrained in climate models or in empirical analyses. According to the IPCC's Fourth Assessment Report (IPCC 2007), the likely range of global equilibrium temperature increase for the doubling of CO2, AT2x, lies between 2.0 and 4.5 K; values below 1.5 K are considered very unlikely. This sensitivity range is in agreement with values exhibited by current climate models (Figure 23.6). Since a doubling of CO2 causes direct radiative forcing of about 3.7 W m-2 (Forster et al. 2007), the range of 2.0-4.5 K for a doubling of CO2 corresponds to climate sensitivity between 0.54 and 1.22 K/(W m-2). Roe and Baker (2007) point out that the upper range of the climate sensitivity is relatively insensitive to decreases in uncertainties associated with the underlying climate processes. In this context, we note that some recent experiments with CRMs embedded into climate models (Miura et al. 2005; Wyant et al. 2006) suggest a lower climate sensitivity, with values of 0.44 and 0.41 K/(W m-2), respectively (AT2x 1.6 and 1.5 K). These findings point to the strong influence of the treatment of clouds on modeled climate sensitivity.

Cloud feedback is the response of CRF to a change in global temperature. Key questions involve the nature and extent of CRF changes in a greenhouse-warmed world: As the climate warms, will longwave CRF increase (positive feedback) or decrease (negative feedback)? Similarly, will shortwave CRF increase (negative feedback) or decrease (positive feedback)? Current climate models produce a wide variety of cloud feedbacks ranging from weakly negative to strongly positive, depending on the relative magnitude of different cloud feedback mechanisms (Bony et al. 2006; Webb et al. 2006). As seen in Figure 23.6, much of the model-to-model difference in climate sensitivity results from differences in cloud feedback.

CRF depends largely on the spatial and temporal distribution of clouds and their radiative properties, which are determined by their microphysical characteristics such as size distribution of droplets and ice crystals. These cloud properties are controlled by cloud-forming processes (e.g., cooling through rising air or radiative cooling) and by cloud-dissipating processes (e.g., precipitation, sinking motion, and mixing with dry air). Thus, cloud feedbacks involve changes in the spatial distribution of clouds and their microphysical properties that result from alterations in processes that form and dissipate

Feedback factor

Feedback factor

Sensitivity

UIUC BMRC GFDL HadSM3 HadSM4 HadGSMI MIROC lo IPSL

MIROC hi

Cloud

I I Longwave

Cloud

I I Longwave

UIUC BMRC GFDL HadSM3 HadSM4 HadGSMI MIROC lo IPSL

MIROC hi

UIUC

GFDL

HadSM4

MIROC lo

IPSL

MIROC hi

Total

BMRC

HadSM3

HadGSM1

Sensitivity

Figure 23.6 Influence of cloud feedback on total feedback and sensitivity of current GCMs. Left panel shows total cloud feedback and long- and shortwave components in nine current climate models. Right panel shows total feedback factor and sensitivity expressed as the inverse of the total feedback factor, in units of K/(W m 2), and as the equilibrium increase in global mean surface temperature that would result from a doubling of CO2, AT2x, evaluated as -3.7 W m 2 upon the total feedback factor. Modified from Webb et al. (2006).

clouds. Uncertainty arises because it is not clear how cloud microphysical properties and cloud horizontal/vertical distributions respond to alterations in controlling variables as the climate changes. In contrast to CRF, cloud feedback cannot be measured directly; it can only be determined from GCM simulations. Consequently, confidence in estimates of cloud feedback can only be assessed by using observations to evaluate simulations of cloud microphysical and radiative properties, cloud distribution, and radiative forcing in a variety of conditions that span the range expected under climate change. Therefore, atmospheric process research on cloud feedbacks focuses on how these cloud features depend on processes that form and dissipate clouds under a variety of conditions.

Closely related to cloud feedback is water-vapor feedback. Water vapor is the most important greenhouse gas in Earth's atmosphere. Consistent with basic physics, all climate models show an increase in the amount of atmospheric water vapor with rising global mean surface temperature. However, the amount and spatial distribution of the resultant radiative forcing differ considerably. This water-vapor feedback is strongly connected to the cloud feedback because water vapor is the source of condensed phase water in clouds, and clouds remove water from the atmosphere when they precipitate. Clouds also influence evapotranspiration, which is the source of water vapor to the atmosphere.

0 0

Post a comment