Robert G. Fovell

Numerical Simulation of a Midlatitude Squall Line in Two Dimensions

Abstract A two-dimensional, anelastic cloud model was used in attempts to numerically replicate the observed structure of a midlatitude squall line. Initial conditions were adapted from observations of the 22 May 1976 Oklahoma line. Model simulations were made with and without considering the ice phase of water. These model storms have, within the constraints of the models geometry, replicated the basic multicellular character and general line-normal airflow typical of these lines. In addition, the structure and intensity of the subcloud cold air pool and the propagation speeds developed by the storms appear to be reasonable. Further, it was found that the initial conditions chosen resulted in model storms which were not only long-lasting but also essentially repetitive, indicating that a state of quasi-equilibrium had been attained. The storms did not decay because the environmental conditions ahead of the storms were favorable and essentially unchanged during the course of the simulations. The inclusio...

Western US streamflow and atmospheric circulation patterns during El Niño-Southern Oscillation

Abstract Using principal component analysis (PCA), cluster analysis, and jackknife analysis, we investigated the spatial and temporal modes that dominate streamflow variability in the western US in response to El Nino-Southern Oscillation (ENSO) events. Spatial variability was investigated with data only from ENSO years and with rotated PCA on 79 streamflow stations in the western United States. Eight regions, or clusters, were thus pinpointed as areas where streamflow tends to co-vary similarly following ENSO events; traditional cluster analysis confirmed the identification of these regions. The ENSO response in streamflow was then further evaluated by forming an aggregate ENSO composite for each region. Temporal variability of western US streamflow in the PCA-identified regions was evaluated with a ‘T-mode’ PCA that isolated the different responses in streamflow following ENSO events. The T-mode PCA breaks the 13 ENSO events that occurred from 1932 to 1993 into five subsets. It is interesting to note that the events in the dominant mode, PC1(+), occurred before 1976, and next mode, PC2(+), included events prior to 1976. Finally, we investigated the atmospheric circulation patterns over the North Pacific Ocean and much of North America that are associated with the various US streamflow responses. The circulation patterns vary according to the prescribed ENSO forcing. The results of this study contribute to a better understanding of the varied ENSO-streamflow relationship in the western US and the use of ENSO for long-range streamflow forecasting.

Effect of Vertical Wind Shear on Numerically Simulated Multicell Storm Structure

Abstract A strictly two-dimensional cloud model was used to gauge the effect of vertical wind shear on the mature phase behavior of model-simulated multicellular storms, extending the previous work of the authors. We specifically examined the propagation speed, quasi-equilibrium behavior, storm scale and updraft orientation of the model storms as a function of shear intensity. We also considered the precipitation efficiencies of our. model storms and applied density current and Rotunno–Klemp–Weisman theories to our results. Our previous work revealed that model storms could achieve a mature phase consisting of repetitive multicellular development when certain numerical obstacles were overcome. This was referred to as a “quasi-equilibrium state.” We found herein that this state was also reached by model storms even when subjected to a very wide range of low-level wind shear intensities, although the temporal behavior during this stage was clearly dependent on the shear. We also found a very systematic rela...

The Temporal Behavior of Numerically Simulated Multicell-Type Storms. Part II: The Convective Cell Life Cycle and Cell Regeneration

The authors study herein the convective cell life cycle and the cell generation process in mature, multicellular squall-line storms possessing well-developed subcloud cold pools using two- and three-dimensional models. The multicellular storm establishes new cells on its forward side, in the vicinity of the forced updraft formed at the pool boundary, that first intensify and then decay as they travel rearward within the storm’s upward sloping front-to-rear airflow. The principal effort is expended on the two-dimensional case owing to the strong similarity in basic behavior seen in the two geometries. The cell life cycle is examined in several complementary fashions. The cells are shown to be convectively active entities that induce local circulations that alternately enhance and suppress the forced updraft, modulating the influx of the potentially warm inflow. This transient circulation also drives the episodic mixing of stable air into the inflow that results in the cell’s ultimate dissipation. The timing of cell regeneration is also examined; an explanation involving two separate and successive phases, each with their own timescales, is proposed. The second of these phases can be shortened if a ‘‘convective trigger,’’ another by-product of the cell’s circulation, is present in the storm’s inflow environment. Sensitivity of the results to strictly numerical model details is also discussed.

Numerical Simulation of the Interaction between the Sea-Breeze Front and Horizontal Convective Rolls. Part I: Offshore Ambient Flow

Abstract A three-dimensional, cloud-resolving model is used to investigate the interaction between the sea-breeze circulation and boundary layer roll convection. Horizontal convective rolls (HCRs) develop over land in response to strong daytime surface heating and tend to become aligned parallel to the vertical wind shear vector, whereas the land–sea heating contrast causes the formation of the sea-breeze front (SBF) along the coastline. The ability of HCRs to modulate the along-frontal structure of the SBF is examined, complementing and extending previous observational and numerical studies. Three simulations are discussed, the first two demonstrating that the model can produce both phenomena independently. The third is initialized with offshore mean flow and vertical shear perpendicular to the coastline, and results in a sharply defined, inland-propagating SBF that encounters HCRs aligned perpendicular to it. Before the interaction takes place, the SBF is nearly two-dimensional and devoid of along-front...

Recent Advances in the Understanding of Near-Cloud Turbulence

Anyone who has flown in a commercial aircraft is familiar with turbulence. Unexpected encounters with turbulence pose a safety risk to airline passengers and crew, can occasionally damage aircraft, and indirectly increase the cost of air travel. Deep convective clouds are one of the most important sources of turbulence. Cloud-induced turbulence can occur both within clouds and in the surrounding clear air. Turbulence associated with but outside of clouds is of particular concern because it is more difficult to discern using standard hazard identification technologies (e.g., satellite and radar) and thus is often the source of unexpected turbulence encounters. Although operational guidelines for avoiding near-cloud turbulence exist, they are in many ways inadequate because they were developed before the governing dynamical processes were understood. Recently, there have been significant advances in the understanding of the dynamics of near-cloud turbulence. Using examples, this article demonstrates how the...

Discrete Propagation in Numerically Simulated Nocturnal Squall Lines

Abstract Simulations of a typical midlatitude squall line were used to investigate a mechanism for discrete propagation, defined as convective initiation ahead of an existing squall line leading to a faster propagation speed for the storm complex. Radar imagery often shows new cells appearing in advance of squall lines, suggesting a causal relationship and prompting the search for an “action-at-a-distance” mechanism to explain the phenomenon. In the simulations presented, the identified mechanism involves gravity waves of both low and high frequency generated in response to the latent heating, which subsequently propagate out ahead of the storm. The net result of the low-frequency response, combined with surface fluxes and radiative processes, was a cooler and more moist lower troposphere, establishing a shallow cloud deck extending ahead of the storm. High-frequency gravity waves, excited in response to fluctuations in convective activity in the main storm, were subsequently ducted by the storm’s own upp...

Cloud Microphysics Impact on Hurricane Track as Revealed in Idealized Experiments

Analyses of tropical cyclones created in an idealized environment reveal how and why cloud microphysical assumptions can influencestorm motion, including speed and direction. It is well known that in the absence of a mean flow, a leading factor in storm propagation is the establishment of ‘‘beta gyres’’ owing to planetary vorticity advection by the storm’s circulation. Previous research demonstrated that tangential winds well beyond the core influence storm motion by helping to determine the gyres’ orientation and intensity. Microphysical assumptions, especially involving average particle fall speeds, can strongly influence the winds at outer radius. More specifically, microphysics modulates the radial distribution of column-average virtual temperature, which largely determines the radial surface pressure gradient and therefore the winds because they tend to be in gradient balance beyond the core. Microphysics schemes can differ markedly with respect to average fall speed, depending on the complexity of the scheme and how interactions among condensation types are handled. Average fall speed controls the outward movement of particles produced in the eyewall into the anvil, where they can influence the environment through cloud‐radiative interactions and phase changes. With the assistance of some special sensitivity tests, the influence of microphysics and fall speed on radial temperature gradients, leading to different outer wind strengths and tracks, is shown. Among other things, this work demonstrates that the treatment of outer rainbands in operational models can potentially influence how simulated storms move, thus affecting position forecasts.

Convective Initiation ahead of the Sea-Breeze Front

Abstract In earlier work, a three-dimensional cloud model was used to simulate the interaction between the sea-breeze front (SBF) and front-parallel horizontal convective rolls (HCRs), resulting in the SBF systematically encountering roll updrafts and downdrafts as it progressed inland. Interestingly, deep convection was spawned above an HCR updraft ahead of the SBF as the front approached, well before the inevitable front–roll merger. Ostensibly, both the sea-breeze and roll circulations were required for deep convection to be present in this case at all because convection was entirely absent when either phenomenon was removed. Further analysis reveals why both circulations were necessary yet not sufficient for the excitation of deep convection in this case. The sea-breeze circulation (SBC) made its upstream (inland) environment more favorable for convection by bringing about persistent if gentle lifting over an extended region stretching well ahead of the SBF. This persistent ascent established a moist ...

Cirrus cloud simulations using WRF with improved radiation parameterization and increased vertical resolution

[1] The capability of Weather Research and Forecasting (WRF) model in the simulation of cirrus clouds has been examined, with a focus on the effects of radiative processes and vertical model resolution. We incorporate in WRF a new radiation module, referred to as the Fu‐Liou‐Gu scheme, which is an improvement and refinement based on the Fu‐Liou scheme, particularly in reference to parameterization of the single‐scattering properties of ice crystal size and shape. We conducted a number of real‐time WRF simulations for cirrus cases that were observed in the coastal and western United States on 29–30 March 2007, and we compared these with available observations from Moderate Resolution Imaging Spectroradiometer (MODIS) and GOES visible and IR images over the same areas. Simulation results show that WRF is capable of generating reasonable cirrus cloud fields and their movement and dissipation processes, especially those associated with the large‐scale frontal system. Radiative processes are important in cirrus cloud simulations by affecting the vertical thermal structure and hence convection. The newly implemented radiation module, the Fu‐Liou‐Gu scheme, has been demonstrated to work well in WRF and can be effectively used for studies related to cirrus cloud formation and evolution and aerosol‐cloud‐radiation interactions. With the newly implemented radiation scheme, the simulations of cloud cover and cloud and ice water path (CWP and IWP) have been improved for cirrus clouds, with a more consistent comparison with the corresponding MODIS observations in terms of CWP and IWP means and CWP frequency distribution, especially for optically thin cirrus with an improvement of about 20% in simulated mean IWP. The model‐simulated ice crystal sizes have also been shown to be comparable to those determined from MODIS cloud products. Finally, we have demonstrated that model vertical resolution plays a significant role in cirrus cloud simulation in terms of altering vertical velocity field and the associated regional circulation.