Introduction

There is an apparent paradox in cloud physics: clouds are very diverse, in terms of morphology, depth, cloud base and top heights, horizontal extent, dynamics and microphysics; they exert very different radiative impacts, in terms of the balance between their albedo and their greenhouse effect, as well as very distinct dynamic impacts, in terms of how they contribute to the redistribution of water vapor, sensible, and latent heat in the atmosphere. Cloud spatial and temporal distributions are very heterogeneous; large regions may be devoid of clouds, whereas others can be overcast almost all year round. The lifetimes of cloud systems vary enormously from those that are almost stationary for days to others that are fleeting. Nevertheless—and herein lies the paradox—together they manage to maintain a somewhat constant Earth albedo, close to 0.3, and a fairly constant global balance between their albedo and greenhouse contributions to the climate system.

Such a large-scale equilibrium calls for observational approaches that begin on a global scale and progress down to the microscale to capture the processes responsible for the regulation of the hydrological cycle. However, to date, most observational studies of clouds have focused on the microscale, progressing up to the mesoscale. Such studies have served as our fundamental tools for constructing, from the bottom-up, the current cloud models we use to simulate the climate system. However, model comparison exercises, such as those by the IPCC (2007), clearly show that some key feedback processes are still poorly represented. Moreover, although the models predict a mean cloud radiative forcing in agreement with observations, they also exhibit noticeable biases, with a consistent overprediction of optically thick clouds and an underpredic-tion of optically thin low and middle-top clouds (Zhang et al. 2005). This led the IPCC to conclude that "differences in cloud response are the primary source of inter-model differences in climate sensitivity" (Randall et al. 2007, p. 633).

Better understanding of how clouds react to anthropogenic forcings is therefore a priority for improving the accuracy of climate change projections. Anthropogenic forcings, however, can produce very diverse impacts on the general circulation. The primary greenhouse gases have a long residence time (centuries) and are homogeneously distributed. Aerosols, by contrast, have a short residence time (days to weeks) and, consequently, their spatial distribution is heterogeneous, mainly concentrated in the vicinity of the sources. Greenhouse gases interact primarily with longwave radiation. The net radiative impact of aerosols depends on their chemical composition and the balance between their light scattering and absorbing contributions. Impacts of greenhouse gases on clouds can thus be explored from a global perspective to examine, for instance, how the increase in the column water vapor might be compensated by a damping of the convective mass flux to reduce the fractional increase in precipitation (Held and Soden 2006). Aerosol impacts should also be considered at the regional scale, where they are concentrated. The hypothesized "elevated heat pump" effect (Lau et al. 2006) provides an example of a plausible, but currently unverified, regional-scale aerosol effect which may impact the Indian monsoon at the Himalaya foothills.

This greenhouse perspective suggests that anthropogenic forcings might only perturb clouds by modifying the general circulation (i.e., from the global scale down to the cloud scale). However, aerosols impact also cloud micro-physics, since some act as droplet or ice crystal nuclei. Such microphysical changes can propagate up to the cloud scale. The first order response to changing nucleus concentration is an effect upon the cloud albedo by changing the surface area of the droplets, but feedbacks on cloud-scale dynamics must also be considered to understand the response fully. By modifying cloud microphys-ical and optical properties, aerosol particles thus perturb clouds not only from the global/regional scale downward, via their direct effect, but also upward from the microscale (the aerosol indirect effect). Parameterizations of these processes in global climate models (GCMs) are very crude, partly because they involve nonlinear processes at scales that are not accessible to such models, but also because our knowledge of the various feedbacks of cloud microphysics on cloud dynamics is still limited. For example, high-resolution cloud models suggest that the aerosol impacts result in nontrivial effects, inducing either a decrease or an increase of the cloud liquid water path (LWP), with the sign of the response depending upon poorly understood factors. Hence, the aerosol impacts on clouds remain the most uncertain of the climate forcings in terms of efficacy (Forster et al. 2007, Fig. 2.19).

To improve climate change projections, it is crucial to understand how climate change might impact the spatiotemporal distribution and hence radiative forcing of clouds (from the global scale down). In addition, we must also find out whether clouds respond only to changes in large-scale dynamic forcings, or if microphysical processes might also modulate their response, thereby impacting the hydrological cycle and general circulation (i.e., from the bottom up).

Because of the very large range of scales involved, it has been difficult to connect large-scale observational studies of the hydrological cycle with micro-and mesoscale observations of cloud physics. Using examples from existing and future field studies, we will show how effective the micro- to mesoscale approach was for understanding cloud dynamics and microphysics, and will suggest that it now needs to evolve to progressively larger scales. This mandates a greater degree of multidisciplinarity in the design of future observational studies, in an effort to clarify the interactions between aerosol physics and chemistry, small-scale turbulent dynamics, radiation, and the hydrological cycle at the global scale.

0 0

Post a comment