Search

This paper presents a hybrid approach to model warm-rain processes, merging the diverse schemes of bulk and detailed (bin) microphysics. In the bulk scheme, the key assumption is that the exact saturation is maintained inside a cloud. In contrast, the supersaturation inside a cloud is predicted in the bin scheme and is applied to calculate the diffusional growth of cloud droplets. Predicting the supersaturation is numerically cumbersome, however, and typically requires spatial and temporal resolutions that are significantly higher than those that can be applied in the bulk scheme. At the same time, supersaturations inside clouds are small, and the condensate amounts in bulk and bin schemes differ insignificantly. This critical observation forms a starting point for the hybrid bulk-bin approach. In this approach, when the cloud water first appears, the activation scheme inserts cloud droplets at the low end of the bin representation. Subsequent diffusional and eventually accretional growth shift the spectrum toward larger sizes so that the saturation inside a cloud is maintained. Details of the hybrid approach are discussed in this paper, and the validation against the traditional bin scheme in a framework of the adiabatic rising parcel is presented. Before the scheme can be applied to the multidimensional cloud model, a ID advection-condensation problem of Grabowski and Smolarkiewicz is used to address the issue of the numerical difficulties that finite- difference schemes experience near cloud edges. In the bulk case, these are in the form of condensation rate overshoots and undershoots; and this aspect requires special attention in the hybrid scheme. A novel approach is developed that provides a physically consistent solution near cloud edges using the hybrid bulk-bin scheme. The key is to allow grid boxes near the edges to be partly cloudy and to include spectral changes of cloud droplets that take this into account. Application of the hybrid scheme to an idealized 2D problem of moist thermal rising from rest and producing rain illustrates the application of the scheme to practical problems of cloud dynamics and warm-rain microphysics. DOI: 10.1175/2009JAS3155.1

The simulations of five midlatitude precipitating events by the nonhydrostatic mesoscale model Meso-NH are analyzed. These cases cover contrasted precipitation situations from 30[degrees] to 60[degrees]N, which are typical of midlatitudes. They include a frontal case with light precipitation over the Rhine River area (10 February 2000), a long-lasting precipitation event at Hoek van Holland, Netherlands (19 September 2001), a moderate rain case over the Elbe (12 August 2002), an intense rain case over Algiers (10 November 2001), and the 'millennium storm' in the United Kingdom (30 October 2000). The physically consistent hydrometeor and thermodynamic outputs are used to generate a database for cloud and precipitation retrievals. The hydrometeor vertical profiles that were generated vary mostly with the 0[degrees]C isotherm, located between 1 and 3 km in height depending on the case. The characteristics of this midlatitude database are complementary to the GPROF database, which mostly concentrates on tropical situations. The realism of the simulations is evaluated against satellite observations by comparing synthetic brightness temperatures (BTs) with Advanced Microwave Sounding Unit (AMSU), Special Sensor Microwave Imager (SSM/I), and Meteosat observations. The good reproduction of the BT distributions by the model is exploited by calculating categorical scores for verification purposes. The comparison with 3-hourly Meteosat observations demonstrates the ability of the model to forecast the time evolution of the cloud cover, the latter being better predicted for the stratiform cases than for others. The comparison with AMSU-B measurements shows the skill of the model to predict rainfall at the correct location.

Real midlatitude meteorological cases are simulated over western Europe with the cloud mesoseale model Meso-NH, and the outputs are used to calculate brightness temperatures at microwave frequencies with the Atmospheric Transmission at Microwave (ATM) radiative transfer model. Satellite-observed brightness temperatures (TBs) from the Advanced Microwave Scanning Unit B (AMSU-B) and the Special Sensor Microwave Imager (SSM/I) are compared to the simulated ones. In this paper, one specific situation is examined in detail. The infrared responses have also been calculated and compared to the Meteosat coincident observations. Overall agreement is obtained between the simulated and the observed brightness temperatures in the microwave and in the infrared. The large-scale dynamical structure of the cloud system is well captured by Meso-NH. However, in regions with large quantities of frozen hydrometeors, the comparison shows that the simulated microwave TBs are higher than the measured ones in the window channels at high frequencies, indicating that the calculation does not predict enough scattering. The factors responsible for the scattering (frozen particle distribution, calculation of particle dielectric properties, and nonsphericity of the particles) are analyzed. To assess the quality of the cloud and precipitation simulations by Meso-NH, the microphysical fields predicted by the German Lokal-Modell are also considered. Results show that in these midlatitude situations, the treatment of the snow category has a high impact on the simulated brightness temperatures. The snow scattering parameters are tuned to match the discrete dipole approximation calculations and to obtain a good agreement between simulations and observations even in the areas with significant frozen particles. Analysis of the other meteorological simulations confirms these results. Comparing simulations and observations in the microwave provides a powerful evaluation of resolved clouds in mesoscale models, especially the precipitating ice phase.

A one-dimensional version of a multilevel mesoscale model is used to represent the cloud-topped boundary layer (CTBL). Turbulent exchanges are parameterized with a prognostic equation for the turbulent kinetic energy and an improved length-scale formulation. Furthermore, the scheme is extended to give a statistical description of the subgrid-scale condensation with a one-and-a-half-order closure. Several observed reference cases are simulated in order to test the model against observational data and results obtained with a higher-order turbulence model. The latter one is used as a powerful approach for testing the closure of the second-order moments involved in the partial cloudiness scheme. Two of the reference cases are extracted from stratocumulus (Sc) observations off the coast of the United Kingdom with a purely buoyancy-driven and a purely shear-driven CTBL, respectively. The third experiment tries to reproduce a case of Californian Sc clouds where both turbulent effects are important. Finally, the last numerical experiment concentrates on a cloudiness transition case observed during FIRE (First International Satellite Cloud Climatology Project Regional Experiment) with a series of soundings documenting the cloudiness transition from a solid Sc cloud deck over a partly covered region to a clear-sky region. The model results are shown to be in reasonable agreement with both observational data and numerical outputs from the higher-order turbulence model. Finally, it is shown that the partial cloudiness scheme does not only produce more realistic cloudiness and cloud water content than a simple 'all or nothing' condensation scheme, but that it also assures model stability by producing a smooth onset of condensation.