Introduction Clouds and Earths Radiation Budget

Clouds greatly impact the Earth's radiation budget (ERB) because they reflect solar or shortwave (SW) radiation back to space and restrict the emission of thermal or longwave (LW) radiation to space. One conventional measure of this radiative effect is "cloud radiative forcing" (CRF); that is, the difference between actual radiation flux and what it would be if clouds were absent (Ramanathan et al. 1989). A more proper term might be "cloud radiative effect," since "forcing" is usually reserved for external perturbations to the climate system rather than internal components. Regardless of etymology, we will follow common usage and employ CRF in this chapter. Shortwave cloud radiative forcing (SWCRF) is almost always negative, since reflection by clouds usually causes a loss of energy from the climate system, and LWCRF is almost always positive, since clouds usually reduce the amount of radiation emitted to space and thus cause a gain of energy to the climate system.

Several factors influence SWCRF, including the incident solar flux, the horizontal extent of clouds, the vertically integrated amount of cloud condensate (cloud water path), and cloud particle size, phase, and shape. Cloud water path and cloud particle characteristics contribute both to cloud optical thickness and cloud albedo, with the water path being the dominant factor. Variations in cloud cover generally affect SWCRF much more than do variations in cloud albedo. The magnitude of SWCRF varies linearly with Insolation such that cloud changes occurring during daytime, the summer season, and at low latitude have the greatest impact on ERB. The dominance of clouds in the SW portion of the spectrum is illustrated by the fact that if clouds were instantaneously removed, the reflectivity of the planet would be approximately halved.

In addition, cloud condensate and cloud particle characteristics affect cloud emissivity in the LW part of the spectrum, and little upwelling radiation passes through clouds with high emissivity (low transmissivity). Usually, clouds emit less radiation than does the surface due to their colder temperatures, and extensive coverage by clouds with tops high in the atmosphere reduces substantially the amount of LW flux escaping to space. Variability in the horizontal extent of high-level clouds generally has a greater influence on LWCRF than does variability in the specific height or emissivity of high-level clouds. Outgoing LW radiation would increase by about 15% if clouds were instantaneously removed. The loss of radiation by SWCRF is greater than the retention of radiation by LWCRF in the global average, and clouds consequently produce a net cooling effect in the current climate.

The basic geographical distributions and seasonal variations of reflected SW and outgoing LW radiation are well observed and well understood. Information is also available on interannual variability, particularly the large perturbations to ERB resulting from the El Niño/Southern Oscillation (ENSO) phenomenon (Figure 2.1). These perturbations are closely related to shifts in the spatial distribution of optically thick clouds with high tops, which are primarily generated by deep convection. One interesting feature, which may have far-reaching consequences for climate stability, is the tendency for a cancellation between the SWCRF and LWCRF over tropical deep convection regions. It

CERES minus ERBS radiative anomalies

January 1998 minus January 1985-1989

Shortwave

Shortwave

90°E 180° 90°W

Longwave

Longwave

-100 -80 -60 -40 -20 0 20 40 60 80 100 Radiative flux anomaly (Wm-2)

Figure 2.1 Differences in outgoing shortwave (top) and longwave (bottom) radiation fluxes between CERES data for January 1998 (the peak of a very strong ENSO event) and the January average of ERBE data for 1985-1989.

has been argued that this is a fundamental property of the tropical atmosphere, which ensures that the net cloud radiative feedback from deep tropical convection during climate change must be close to zero (Hartmann et al. 2001; cf. also Bretherton and Hartmann, this volume).

0 0

Post a comment