Figure 1 (click on figure to view larger version): Supercell Thunderstorm Simulation. Height coordinate model, (dx = dy = 2 km, dz = 500 m, dt = 12 s, 160 x 160 x 20 km domain ); Surface temperature, surface winds and cloud field at 2 hours

Figure 2 (click on figure to view larger version): Vertical cross section of reflectivity (top), ZDR (middle), and particle identification (bottom) through a tornadic supercell, as observed by SPOL on 29 June 2000 at 23:36 UTC during the STEPS field program. Surface hailfall, as suggested by the yellow region on the bottom panel, was confirmed by mobile ground units. This storm was also sampled by the T28 aircraft and mobile electric ballooning units, and was characterized by a dominance of positive cloud-to-ground lightning strikes.

Figure 3 (click on figure to view larger version): Hövmoller (space-time) diagram of the surface precipitation rate for the 2D cloud-resolving simulation using a planetary-scale (20,000 km) computational domain. Areas with surface precipitation rate larger than 0.5mm hr^-1 are shaded.  The lines show propagation speed of mesoscale convective systems (-12 ms^-1, solid line) and the convection envelope (8 ms^-1, dashed line) relative to the earth-stationary observer. Note that surface precipitation from a well-resolved mesoscale convective system is dot-sized in this figure.

Figure 4 (click on graphic to view larger version): The vertical velocity on a spherical surface in the middle of the solar convection zone. Red shades denote upflows, while blue shades denote downflows.

Figure 5 (click on graphic to view larger version): This figure illustrates how the particle size distributions broaden from cloud top downward to approximately 4.5. km.  This growth of the larger particles was through the aggregation process, whereby the larger particles collected the smaller particles in the 100 to 600 micron size range; this resulted in reduced concentrations of the particles < 600 microns.  From about 4.5 to 4.3 km, the particles melted to become rain.  Marshall-Palmer type exponential curves were fitted to the measured size distributions to develop a means of retrieving particle size distributions from radar data, either from the TRMM radar in space or from ground-based radars.  What was found from the 10 cases these authors observed was a systematic variation in the fit parameters with height below cloud top.  Within the melting layer, these parameters also varied in a systematic way.

Figure 6 (click on graphic to view larger version): Light curves show the ice crystal area ratio as a function of crystal size for different datasets, and the bold curve is an average of the datasets.  Small crystals are round and compact, becoming more elongated and open as they grow, affecting crystal optical properties, fallspeeds, and calculations of crystal mass.

Figure 7 (click on figure to view a larger version): Frequency distribution of cloud droplet concentrations from the C130 for all flights during INDOEX in clean regions (where condensation nucleus (CN) concentrations were <500 cm-3), moderately polluted regime (CN = 500 to 1500 cm-3), and polluted regimes (CN > 1500 cm-3).

Figure 8 (click on figure to see larger version): Photos from the NCAR C130 aircraft during research flight in the Indian Ocean Experiment (INDOEX) showing dirty regime at 0.1 deg N latitude (top) and 4.9 deg S latitude (bottom).

Figure 9 (click on figure to view movie): The time evolution of a horizontal (x-y) slice of water vapor mixing ratio at ten meters.  The two hundred frames in the movie span approximately one half hour of simulated time.  Purple corresponds with high mixing ratio and red with low mixing ratio.

Figure 10 (click on figure to see larger version): Instantaneous y-averaged velocity vectors in an x-z plane from two coupled LES-LSM simulations responding to two different values of incoming solar radiation (700, 350) W/m^-2.  Colors and vector length depict velocity magnitude.  Red: large magnitude, Black: small magnitude.  Heavy black line is the instantaneous y-averaged boundary layer height.

Jordan Powers and Christopher Davis conducted a high-resolution modeling investigation of the genesis and organization of tropical cyclones. Figure 11 at the left shows MM5 output from a 1.2-km simulation of Hurricane Diana (1984).  The simulation of this case is unique in that it prescribes only the synoptic-scale baroclinic wave, yet captures the subsequent development of mesoscale convective structures that lead to tropical cyclogenesis, and does it at cloud-resolving resolution (1.2 km). This is the first successful simulation of the genesis phase of a tropical cyclone wherein the deep convection is well resolved and no cumulus parameterization or artificial implant of an initial vortex is required.

Figure 11 (right) (click on figure to see larger version): View is of hour 48 of the run (00 UTC 10 Sept 1984) with system at tropical storm stage and shows sea level pressure (dark blue contours) and surface rainwater (shaded).  Both individual convective cells and aggregated convective bands are evident.  Sea level pressure contour interval= 2 mb; rainwater shading scale at bottom (shading begins at .02 g/kg).

Figure 12 (click on figure to see larger version) This figure illustrates the response of the MM5 3DVAR system to a hypothetical single total precipitable water (TPW) observation situated over Taiwan. Particular isosurfaces of specific humidity (blue), temperature (yellow) and pressure (purple) are shown, as is the coupled wind response (cyclonic circulation aloft, anticyclonic below) to the TPW observation. The mass/wind coupling is specified via linearised geostrophic/cyclostrophic balance within 3DVAR.