Three-dimensional Cloud Model Research Articles

Evolution and Structure of Tropical Squall Line Elements within a Moderate CAPE and Strong Low-Level Jet Environment

A three-dimensional cloud model is used to simulate the early development and propagation of an arc-shaped line element similar to that found in the GARP (Global Atmospheric Research Program) Tropical Atlantic Experiment (GATE) 4‐5 September 1974 squall line system. The simulated squall line element forms in a relatively unexplored environment with moderate convective available potential energy and a strong low-level jet (bulk Richardson number 5 37) associated with an easterly wave. The simulated line element develops in a large-scale convergence region from an initial cell that splits and elongates in a manner reminiscent of some midlatitude lines. Simulation features compare favorably with observed characteristics of some of the line elements including line orientation (approximately perpendicular to the average wind shear below the low-level jet), propagation speed (11 m s21 to the southwest), length (75 km), and maximum precipitation rate (187 mm h21). In addition, the simulated line merges with cells that form ahead as observed. The simulated arc-shaped line element consists of four regions several hours after its initiation. The northern region or right flank of the line contains a long-lived cell exhibiting three-dimensional characteristics similar to midlatitude supercells including long-lasting and steady updrafts with midlevel cyclonic rotation and movement to the right of the mean winds. Although the simulated supercell characteristics cannot be confirmed because of insufficient data from the limited GATE 4‐5 September observations, updrafts with strong vertical vorticity have been observed to occur in other GATE rainbands where waterspouts have been seen. The region just to the south of this cell is the site where a cell that formed out ahead of the line segment merges with the line and also develops supercell-like characteristics. Farther to the south, convection develops with quasi twodimensional characteristics below 3‐4 km but breaking into several multicells above. A rear-inflow jet does not accompany these features and the winds in the downdraft are oblique too and nonuniform along the associated gust front. Finally, on the southernmost or left flank of the line element, cells with both long and short lifetimes develop. Sensitivity tests indicate that once the early line structure has developed, its evolution and structure are not seriously altered by the removal of large-scale forcing. Further, the formation of the cold pool and gust front between the initial separating (splitting) cells is crucial to the filling in of the line. Changes in the thermodynamic profile consistent with nonsquall observations one-half day earlier result in only modest differences. Changes in the wind profile (based on the same observations) led to significant differences, such as the lack of convection between the initial separating cells and the merger taking place to the north of the right flank creating a convective complex in this region.

The Numerical Simulation of Nonsupercell Tornadogenesis. Part II: Evolution of a Family of Tornadoes along a Weak Outflow Boundary

Nonsupercell tornadogenesis along a weak outflow boundary has been simulated using a three-dimensional moist convective cloud model. Thermodynamic conditions similar to those observed for nonsupercell tornado (NST) events of the High Plains were utilized in the model initialization. As the ensemble system of storm, outflow boundary, and leading edge vortices evolve, six distinct life cycle stages for the development and decay of NSTs are documented that span a period of about 35 min. Consistent with the results of Part I of this numerical study, vortex sheet dynamics exert considerable control over the outflow leading edge. The progression of pretornadic life cycle stages serves to concentrate vertical vorticity effectively along the outflow boundary in discrete misocyclone circulations aligned in a 3-km wavelength pattern. The organization of larger-scale misocyclones and ultimate intensification to initial tornadic intensity occurs coincident with the rapid development of deep convection overhead. The strongest members of a family of NSTs that develop in the model maintain ground-relative surface wind speeds of greater than 30 m s−1 for approximately 11 min within which wind speeds meet F1 severity criteria for 6 min. The mature vortices reach the midlevels of the moist convection and display deep, rotationally induced axial downdrafts. The rapid transition to a predominant downdraft character for the storm complex and to an outflow dominated subcloud air mass heralds the onset of NST dissipation. The NST evolution simulated here compares very favorably to observational NST studies. The misocyclones are shown to provide an asymmetric pattern of convective forcing along the outflow boundary, which supports the formation of deep moist convection directly over them. Vertical vorticity in the boundary layer misocyclones is redistributed upward into the midlevels of the moist convection (∼6 km) by developing deep updrafts. The mature vortices are maintained by vertical vorticity transported from a vortex sheet located along the outflow boundary and by vertical vorticity produced from the tilting of horizontal vorticity in the inflow region southeast of the NSTs. Low-level vortex stretching is the dominant vorticity tendency term as the vortices intensify to and maintain tornadic strength. No significant vertical tilting of baroclinically generated horizontal vorticity was indicated even after convective downdraft-associated new outflow pools formed in the environment surrounding the misocyclones. These new outflow pools play a major role in NST intensification by increasing convergence and resultant vortex stretching along the periphery of the tornadic circulations. A comparative simulation with rain production turned off, which precluded the possibility of new outflow development, revealed that NST intensity was roughly 25% greater in the baseline simulation. In an additional comparative simulation where no moist convection was allowed to develop, the resultant misocyclones markedly lacked the coherent organization and intensity of the misocyclones and NSTs of the baseline simulation and no NST-strength vortices developed. A six-stage “refined” model of NST development and decay is presented.

STORMTIPE-95: Results from a Convective Storm Forecast Experiment

Abstract During the spring of 1995, an operational forecast experiment using a three-dimensional cloud model was carried out for the north Texas region. Gridpoint soundings were obtained from the daily operational numerical weather prediction models run at the National Centers for Environmental Prediction, and these soundings were then used to initialize a limited-domain cloud-resolving model in an attempt to predict convective storm type and morphology in a timely manner. The results indicate that this type of convective forecast may be useful in the operational environment, despite several limitations associated with this methodology. One interesting result from the experiment is that while the gridpoint soundings obtained from the NCEP models generally overforecast instability and vertical wind shear, the resulting convective storm evolution and morphology in the cloud model was often similar to that of the observed storms. Therefore the “overforecast” of mesoscale environment’s instability and vertica...

Structure and Evolution of the 22 February 1993 TOGA COARE Squall Line: Organization Mechanisms Inferred from Numerical Simulation

Abstract Mechanisms responsible for meso- and convective-scale organization within a large tropical squall line that occurred on 22 February 1993 during the Tropical Ocean Global Atmosphere Coupled Ocean–Atmosphere Response Experiment are investigated using a three-dimensional numerical cloud model. The squall line occurred in an environment typical of fast-moving tropical squall lines, characterized by moderate convective available potential energy and moderate-to-strong vertical shear beneath a low-level jet with weak reverse vertical shear above. A well-simulated aspect of the observed squall line is the evolution of a portion of its leading convective zone from a quasi-linear to a three-dimensional bow-shaped structure over a 2-h period. This transition is accompanied by the development of both a prominent mesoscale vortex along the northern edge of the 40–60-km long bow-shaped feature and elongated bands of weaker reflectivity situated rearward and oriented transverse to the leading edge, within enha...

Evaluation of a three-dimensional cloud chemistry model

To evaluate the effectiveness of a three-dimensional dynamical cloud model in simulating cloud water chemistry, the ion concentrations in simulated clouds are compared with the results of chemical analyses of collected cloud water samples. The model is initialized with surface and aircraft measurements of clear air chemistry. The data that are used in the evaluation come from non-precipitating cumulus clouds that were sampled during the Eulerian Model Evaluation Field Study (EMEFS I) that was conducted in central Ontario in the summer of 1988. Three different cases are considered. In general, the simulated ion concentrations agree well with the observed concentrations when the ambient aerosol is assumed to be activated with 1.00% efficiency. In two of the cases the predominant contribution to the cloud water sulphate is activation of the aerosol. In the third case, in which the clear air sulphate concentration is small, oxidation of sulphur dioxide by hydrogen peroxide and ozone is the dominant source of sulphate.

Supercell simulations with simple ice parameterization

A three-dimensional cloud model derived from the Klemp-Wilhelmson model has been used to perform simulations of a right-moving supercell with the purpose of testing the effects of terminal velocity of precipitation on the storm dynamics. The model has no ice variables and computes a fall velocity for condensed water. We simulate one of the effects of ice presence, without including phase transition, by reducing the fall velocity of the precipitation in the coldest cloud layers. The reduction of precipitation terminal velocity is shown to have a strong influence on supercell dynamics and to be responsible for nearly all the improvements in the simulation of supercells that are obtained with full ice microphysics. This includes the transition to tornadic phase, which is sharper than in the case of a pure rain supercell. The vertical component of vorticity is seen to increase at all levels, as observed in real storms, instead of just at the surface as is common in simulations with no ice phase. Our results indicate the primary importance of terminal fall speed among the effects due to the presence of ice.

Microwave and Infrared Simulations of an Intense Convective System and Comparison with Aircraft Observations

A three-dimensional cloud model, radiative transfer model-based simulation system is tested and validated against the aircraft-based radiance observations of an intense convective system in southeastern Virginia on 29 June 1986 during the Cooperative Huntsville Meteorological Experiment. NASA's ER-2, a high-altitude researchaircraft with a complement of radiometers operating at 1 I-pm infrared channel and IS-, 37-, 92-, and 183-GHz microwave channels provided data for this study. The cloud model successfully simulated the cloud systemwith regard to aircraft- and radar-observed cloud-top heights and diameters and with regard to radar-observed reflectivity structure. For the simulation time found to correspond best with the aircraft- and radar-observed structure, brightness temperatures Tb are simulated and compared with observations for all the microwave frequencies along with the 1 1 -pm infrared channel. Radiance calculations at the various frequencies correspond well with the aircraft observations in the areas of deep convection. The clustering of 37-174-GHz Tb observationsand the isolation of the 18-GHz values over the convective cores are well simulated by the model. The radiative transfer model, in general, is able to simulate the observations reasonably well from 18 GHz through 174 GHz within all convective areas of the cloud system. When the aircraft-observed 18- and 37-GHz, and 90- and 174-GHz 7's are plotted against each other, the relationships have a gradual difference in the slope due to the differences in the ice particle size in the convective and more stratiform areas of the cloud. The model is ableto capture these differences observed by the aircraft. Brightness temperature-rain rate relationships compare reasonably well with the aircraft observations in terms of the slope of the relationship.The model calculations are also extended to select high-frequency channels at 220, 340, and 400 GHz to simulate the Millimeter-wave Imaging Radiometer aircraft instrument to be flown in the near future. All three of these frequencies are able to discriminate the convective and anvil portions of the system, providing useful information similar to that from the frequencies below 183 GHz but with potentially enhanced spatial resolution from a satellite platform. In thin clouds, the dominant effect of water vapor is seen at 174, 340, and 400 GHz.In thick cloudy areas, the scattering effect is dominant at 90 and 220 GHz, while the overlying water vapor can attenuate at 174, 340, and 400 GHz. All frequencies (90-400 GHz) show strong signatures in the core.

Updraft Dynamics within a Numerically Simulated Subtropical Rainband

Abstract In this study a three-dimensional numerical cloud model is used to examine the early evolution of deep convective rainbands that occur in an environment of weak to moderate buoyancy and directionally varying lower-tropospheric vertical wind shear. A simulation based on a case observed on 8 June 1987 during the Taiwan Area Mesoscale Experiment produced a narrow bow-shaped rainband that comprised 1) short-lived updrafts along the downshear portion of the weak rain-induced cold pool and 2) more persistent updrafts along its southern flank, which were highly correlated with vertical vorticity. Trajectory calculations and an analysis of the dynamic portion of the perturbation pressure field are presented to illustrate the hybrid dynamical character of the simulated rainband. The shorter-lived updrafts were associated with weak upward-directed pressure gradient forces at the leading edge of the surface-based cold pool. The more persistent updrafts exhibited much stronger upward-directed pressure gradie...

Heat, Moisture, and Momentum Budgets of Isolated Deep Midlatitude and Tropical Convective Clouds as Diagnosed from Three-Dimensional Model Output. Part I: Control Experiments

Abstract This project uses a three-dimensional anelastic cloud model with a simple ice phase parameterization to evaluate the feedback between isolated deep convective clouds and their near surroundings. The horizontal Reynolds averaging approach of Anthes is adopted to diagnose the vertical profiles of the individual budget terms for heat, moisture, and horizontal momentum, as well as the resultant effects of each budget as defined by apparent sources or sinks. The averaging area, 33.75 km on a side, is comparable to one grid cell for typical mesoscale numerical weather prediction models. Two comparative simulations are run, one for a severe Oklahoma thunderstorm in strong vertical wind shear and the other for a tropical Atlantic cumulonimbus in much weaker shear. The midlatitude cloud evolves to a vigorous quasi-steady mature stage with several supercell characteristics including an erect large-diameter updraft, a strong and vertically extensive mesolow, and a well-developed highly asymmetric cold pool ...

Balanced Dynamics of Mesoscale Vortices Produced in Simulated Convective Systems

Abstract Long-lived, mesoscale convective systems are known to occasionally produce mesoscale convective vortices (MCVs) in the lower to middle troposphere with horizontal scales averaging 100–200 km. The formation of MCVs is investigated using fully three-dimensional cloud model simulations of idealized, mesoscale convective systems (MCSs), initialized with a finite length line of unstable perturbations. In agreement with observations, the authors find that environmental conditions favoring MCV formation exhibit weak vertical shear confined to roughly the lowest 3 km, provided the Coriolis parameter (f) is chosen appropriate for midlatitudes. With f = 0, counterrotating vortices form on the line ends, positive to the north and negative to the south with westerly environmental shear. The MCV and end vortices are synonymous with anomalies of potential vorticity (PV). Using PV inversion techniques, the authors show that the vortices are nearly balanced, even with f = 0. However, the formation of mesoscale v...

Parameterization of Bulk Condensation in Numerical Cloud Models

The accuracy of the moist saturation adjustment scheme has been evaluated using a three-dimensional explicit microphysical cloud model. It was found that the error in saturation adjustment depends strongly on the CCN concentration in the ambient atmosphere. The scheme provides rather accurate results in the case where a sufficiently large number of CCN (on the order of several hundred per cubic centimeter) is available. However, under conditions typical of marine stratocumulus cloud layers with low CCN concentration, the error in the amounts of condensed water vapor and released latent heat may be as large as 40%–50%. A revision of the saturation adjustment scheme is devised that employs the CCN concentration, dynamical supersaturation, and cloud water content as additional variables in the calculation of the condensation rate. The revised condensation model reduced the error in maximum updraft and cloud water content in the climatically significant case of marine stratocumulus cloud layers by an order of magnitude.

A comparative study on hailstone trajectories using different motion equations, drag coefficients and wind fields

The growth of hydrometeors to hailstone sizes is simulated by computer. The trajectories are calculated using the results of a three-dimensional cloud model as input. The model simulates the time evolution of the wind, the liquid water content and the air temperature fields in a cubic region of 16 km per side. These fields are recorded and later used in another program that calculates the hail growth and the aerodynamical forces involved. Embryos of 0.25, 0.5 and 1.00 mm of initial radii are released at 676 locations uniformly distributed in a region of 6×6×4 km 3 around the updraft center. The trajectories of all the particles are followed all the way to the 0°C altitude. The final radii and residence times in the cloud of all particles are also stored. A systematic study on the changes brought about by the use of different expressions for the drag coefficient and for the equation of motion on both the computed trajectories and on the final size of hailstones is performed. The study includes the location of the most favored regions for the initiation of hail in each case. The effect of using stationary and time dependent fields is also considered in detail. A comparison of the results obtained using a realistic equation of motion and those employing an approximation frequently used by others show the respective final radii differing by up to 30%. It is observed that expressions predicting values for the drag coefficient that are not nearly as scattered as those found in nature can lead to final radii differing by as much as 50%. The use of stationary wind fields can produce an important overestimation of the maximum sizes achieved by hail inside the cloud.

Comparison of microphysics parameterizations in a three-dimensional dynamic cloud model

Abstract A three-dimensional, time-dependent flow model developed to simulate dynamics and cloud microphysics for warm-cloud processes is used to compare the moisture fields simulated with different cloud microphysics parameterization. Our parameterizations are based on detailed simulations of the physics of aerosols, clouds, and precipitation. The moisture content in the system is differentiated into water vapor, cloud water, and rain water. Advection and diffusion terms for momentum, potential temperature, and mixing ratios of water vapaor and cloud water are calculated by using finite-volume methods, while rain water mixing ratio fields are calculated by a Lagrangian method. The model simulations are examined through sensitivity analysis, interpreted in terms of the physical processes occurring, and are compared with simulations using parameterizations from the literature. Calculated precipitation totals are shown to vary by 10–30% among simulations with alternate parameterization.

The Influence of Helicity on Numerically Simulated Convective Storms

Abstract A three-dimensional numerical cloud model is used to investigate the influence of storm-relative environmental helicity (SREH) on convective storm structure and evolution, with a particular emphasis on the identification of ambient shear profiles that are conducive to the development of long-lived, strongly rotating storms. Eleven numerical simulations are made in which the depth and turning angle of the ambient vertical shear vector are varied systematically while maintaining a constant magnitude of the shear in the shear layer. In this manner, an attempt is made to isolate the effects of different environmental Felicities on storm morphology and show that the SREH and bulk Richardson number, rather than the mean shear in the low levels, determine the rotational characteristics and morphology of deep convection. The results demonstrate that storms forming in environments characterized by large SREH are longer-lived than those in less helical surroundings. Further, it appears that the storm-relat...

Numerical Simulations of Microburst-producing Storms: Some Results from Storms Observed during COHMEX

In the first part of this paper, the characteristics of microburst-producing storms are examined with a three-dimensional cloud model using soundings from the Cooperative Huntsville Meteorological Experiment (COHMEX). With a grid resolution of 500 m, it is shown that the general characteristics of observed vertical velocities, vertical draft sizes, water contents, radar reflectivities, and surface outflow strengths can be simulated. In addition, observed microburst precursors such as midlevel convergence and descending precipitation cores can also be simulated. Using a grid resolution of 250 m, the observed structure of a particularly well-documented storm on 20 July 1986 during COHMEX is simulated, including a hail shaft 1–2 km wide that descended to the ground. In the second part of this paper, the influence of microphysical processes in the production of low-level downdrafts in simulated COHMEX storms is investigated. It is shown that low-level downdrafts are in some cases stronger and deeper in simulations made with the ice phase than in simulations made without the ice phase. These differences are due, in part, to the additional cooling associated with the melting of ice, and are consistent with findings of several other recent studies of low-level downdraft production in deep convective storms.

A Numerical Study of a Rotating Downburst

Abstract Previous studies have revealed that convective storms often contain intense small-scale downdrafts, termed “downbursts,” that are a significant hazard to aviation. These downbursts sometimes possess strong rotation about their vertical axis in the lower and middle levels of the storm, but studies of how this rotation is produced and how it impacts downdraft strength are lacking. In this study a three-dimensional cloud model was used to simulate a rotating downburst based on conditions observed on a day that produced rotating downbursts. It was found that rotating downbursts may occur when the direction of the wind shear vector in the middle levels of the troposphere varies with height. In the early stages of the convective system, vertical vorticity is generated from tilting of the ambient vertical shear by the updraft, resulting in a vertical vorticity couplet on the flanks of the updraft. Later, the negative buoyancy associated with precipitation loading causes the updraft to collapse and to be...

A three-dimensional large-scale cloud model: testing the role of radiative heating and ice phase processes

A three-dimensional large-scale cloud model: testing the role of radiative heating and ice phase processes

3次元浅い雲モデルでの"CoolRain"

Simulations using a three-dimensional shallow cloud model with detailed microphysics showed that even in maritime clouds, an ice-phase enhanced precipitation particle production compared to that occurring solely from warm rain processes. Rain fell earlier, was heavier, and lasted longer from maritime ice clouds than from warm clouds. The most remarkable result was a strong interaction between precipitation and wind shear as ice grew in the cloud. Rain fell in such a way that the low-level convergence was intensified. Modelled ice clouds with wind shear lasted much longer and produced much heavier rainfall than did similar clouds without ice, or clouds with ice but no wind shear.

A three-dimensional large-scale cloud model: testing the role of radiative heating and ice phase processes

A time-dependent, three-dimensional, large-scale cloud model has been developed for the prediction of cloud cover, cloud liquid/ice water content (LWC/IWC), precipitation, specific humidity and temperature. Partial cloudiness is allowed to form when large-scale relative.humidity is less than 100%. Both liquid and ice phases are included in the model. The liquid phase processes consist of evaporation, condensation, autoconversion and precipitation. The ice phase processes include heterogeneous nucleation to generate ice crystals, depositional growth to simulate Bergeron-Findeisen's process, sublimation to deplete ice crystals, and gravitational settling of ice crystals. The radiative transfer parameterization scheme is based on a broadband method and involves the transfer of infrared and solar radiation in clear and cloudy regions. The broadband infrared emissivity, reflectivity, and transmissivity for cirrus clouds, as well as the broadband solar absorption, reflection, and transmission values for low, middle and high clouds are computed based on the cloud LWC and IWC interactively generated by the cloud model. Large amounts of satellite data, including cloud cover climatology derived from the US Air Force three-dimensional nephanalysis (3DNEPH) and earth radiation budget (ERB), have been processed into formats appropriate for verification. The 96-h model simulations of cloud cover, outgoing long-wave radiation (OLR), and cloud LWC have been verified against data analyzed from 3DNEPH, ERB and the Nimbus-7 Scanning Multichannel Microwave Radiometer (SMMR) satellite observations, respectively. The predicted cloud IWC is compared to in situ observations as well as to results from other studies. The predicted cloud and radiation results compare well with those analyzed from satellite data. Numerical experiments are carried out with and without radiative heating and ice phase processes in the cloud formation scheme. The inclusion of radiative heating produces a significant change in temperature, cloud cover and total cloud water content, while the inclusion of ice phase processes generates a substantial change only in total cloud water content. If the cloud LWC is sufficiently large to initiate effective Bergeron-Findeisen's processes, the total cloud water content decreases, indicating that gravitational settling is an efficient mechanism in reducing the cloud IWC. DOI: 10.1034/j.1600-0870.1992.t01-1-00001.x

A comparison of 3D model results with observations for an isolated CCOPE thunderstorm

This paper is concerned with the simulation of deep convection for the CCOPE 19 July 1981 case study. Clark's three-dimensional (3D) cloud model modified to use the bulk water parameterization scheme of Lin et al. has been used in the simulation of the CCOPE 19 July 1981 case in coarse mesh, fine mesh, and interactive grid nested schemes, respectively. Comparisons with observations show this 3D grid nested cloud model is capable of both capturing both the dynamic and microphysical properties of the cloud.