Representation of Convective Radiative Processes in Climate Models

The research results discussed in this review represent a one-way interaction approach for process models to study the physical processes interacting with climate oscillations. The acquired knowledge forms a basis for improving the representation of these physical processes in climate models, and also motivates the next-step approach to examining the radiative-convective (two-way) interaction with climate dynamics in large-scale models. Indeed, since the original attempt by Sundqvist (1978) to include prognostic cloud water content for parametrizing noncon-vective condensation, there has been increasing research attempts to incorporate detailed micro-physical processes in weather and climate models. The trend is to treat cloud water and/or ice content as a prognostic variable which is governed by microphysical processes (e.g. Tiedtke, 1993; Del Genio et al., 1996; Fowler et al., 1996; Sud and Walker, 1999, 2003; Zhao and Carr, 1997). This approach allows the storage and full life cycle of cloud water, and a better cloud-radiation linkage through interactive cloud optical properties. While this approach provides a more physically based framework for representing the physical processes, it also introduces a number of microphysical parameters absent from the simpler approaches. Thus, a great deal of effort is required to implement a prognostic cloud parametrization scheme in climate models, and to validate the scheme against various observational data sets.

Another approach to incorporating detailed convective processes in weather and climate models is through directly resolving convective dynamics in the GCM, such as "super-parametrization" (Grabowski and Smolarkiewicz, 1999; Grabowski, 2001;

Khairoutdinov and Randall, 2001). This approach is to implement a cloud-resolving model (CRM) inside each grid box of a global model. The CRM (so far two-dimensional) does not fill the global model's grid box. Instead, it occupies only part of the grid box. The advec-tion terms computed at each grid of the global model is imposed in the CRM as heat and moisture source/sink. In return, the CRM computes cloud ensemble statistics. In this way, the super-parametrization provides a framework for coupling convective-radiative processes with large-scale dynamics all at the physical time and space scales of the convective process. But this approach also introduces many problems different from those of simpler cumulus parametriza-tion. See Randall et al. (2003) for details.

At the GSFC, continual efforts have been made to improve the representation of convective-radiative processes on a multi-model framework. One approach is to run the finite-volume general circulation model (fvGCM; Lin, 2004) at 1/8 of a degree (about 12 km) to resolve convective vortices in the global model context. The model is capable of simulating realistic tropical cyclones in the weather prediction model (Altas et al., 2005). An attempt at super-parametrization in the fvGCM is also being examined.

Another important effort at the GSFC is to utilize satellite measurements to advance cloud-climate feedback study in climate models. Lau and Wu (2003) performed an analysis of satellite data from the Tropical Rainfall Measuring Mission (TRMM; Simpson et al., 1988). They found that warm rain accounts for 31% of the total rain amount and 72% of the total rain area in the tropics, and that there is a substantial increase in the precipitation efficiency of light warm rain as the sea surface temperature increases, but the precipitation efficiency of heavy rain associated with deep convection is independent of the sea surface temperature. This implies that in a warmer climate, there may be more warm rain, at the expense of less cloud water available for middle and high level clouds. The study points out a possible need to pay attention to resolving the melting/freezing zone in convection to simulate and better understand the role of cumulus congestus in tropical convection, and its sensitivity to SST, and global warming. Lau et al. (2005) performed a sensitivity test of GCM dynamics to the microphysics parameter of autoconversion. The result shows that a faster autoconversion rate leads to enhanced deep convection, more warm rain but less cloud over oceanic regions, and an increased convective-to-stratiform rain ratio over the entire tropics. The resultant vertical differential heating destabilizes the tropical atmosphere, producing a positive feedback, resulting in more rain and an enhanced atmospheric water cycle over the tropics. The feedback is maintained via secondary circulations between convective tower and anvil regions (cold rain), and adjacent middle-to-low cloud (warm rain) regions. The lower cell is capped by horizontal divergence and maximum cloud detrainment near the freezing/melting (0°C) level, with rising motion (relative to the vertical mean) in the warm rain region connected to sinking motion in the cold rain region. The upper cell is found above the 0° C level, with induced subsidence in the warm rain and dry regions, coupled to forced ascent in the deep convection region. The above result reveals that warm rain plays an important role in regulating the time scales of convective cycles, and in altering the tropical large-scale circulation through radiative dynamic interactions. Reduced cloud-radiation feedback by a faster autoconversion rate results in intermittent but more energetic eastward-propagating Madden and Julian oscillations (MJO's). Conversely, a slower autconversion rate, with increased cloud radiation produces MJO's with more realistic westward-propagating transients embedded in eastward-propagating supercloud clusters.

Super-parametrization is an intermediate approach to representing convective-radiative processes in a global model, between a prognostic cloud scheme and a global cloud-resolving model. If computational resources allow, it is most straightforward to develop an ultrahighresolution nonhydrostatic climate model that can resolve clouds with explicit microphysics. Global nonhydrostatic models are being developed at many institutions. In particular, a global nonhydrostatic grid model with icosahedral structure is being developed in Japan (Satoh et al., 2005). This model is intended for highresolution climate simulation, so the numerical scheme is designed for conserving mass and energy. Global simulations with cloud-resolving physical processes (cloud microphysics, radiation, and boundary layer processes) have been performed on an aqua planet setup with grid intervals of 7 km and 3.5 km. The model simulates reasonable features in the tropics, like the diurnal cycle of precipitation, hierarchical structure of clouds, and intraseasonal oscillations (Tomita et al., 2005). The model's response to SST warming has also been investigated by Miura et al. (2005).

0 0

Post a comment