As in other recent studies, Richard Rotunno, William Skamarock and Chris Snyder found that numerical simulations of frontogenesis with surface drag are less inclined to develop a secluded warm sector than those conducted without drag, and that drag weakens the warm front while the cold front remains strong. The different effect that friction has on the warm front with respect to the cold front in the primitive-equation simulations is fundamentally related to the tendency for the lows to be strong and narrow and the highs weak and broad, and for the warm front to form just north of, and extend eastward from the low, while the cold front extends between the high and the low (see Summary figure). Their thesis is that the Ekman pumping associated with the low, where the warm front would form in the absence of surface friction, acts to resist the formation of the warm front, while the cold front, positioned between the high and the low where Ekman pumping associated with the baroclinic wave is weak, is therefore relatively unaffected.

Much of current understanding of midlatitude dynamics in both the atmosphere and ocean arises from the quasi-geostrophic theory first proposed by Charney in 1948. David Muraki (New York University), Rotunno and Snyder have extended quasi-geostrophic theory to the next order in Rossby number through an asymptotic expansion. The key step in their approach is to re-formulate the primitive equations by including the conservation of potential vorticity (PV) as a prognostic equation and changing dependent variables (from the potential temperature and horizontal velocity components to three potential functions). Because the equations are accurate through an order beyond quasi-geostrophy remain relatively simple, they were able to derive an analytic solution for a finite-amplitude surface edge wave and to calculate numerically solutions for the evolution of an unstable baroclinic wave.

Christopher Davis (joint appointment with RAP), in collaboration with Mark Stoelinga (University of Washington) has developed a strategy for isolating numerically the effects of complex terrain on synoptic-scale weather systems. Two distinct applications were attempted. The first is in the context of a quasigeostrophic numerical model. Here they showed that the effect of topography on a baroclinic wave can be thought of as a series of contributions. The first is an asymmetric (about the center of a channel model) pattern forced by the unmodified wave interacting with the mountain; the second is a symmetric pattern forced by the interaction of the zonal flow induced by the first contribution (see figure) with the terrain. This second partition is responsible for the changes in intensity and zonal phase speed changes induced by the mountain. In general, the mountain weakens and accelerates the incoming wave.

The second application of the partitioning technique was performed in the context of MM4. An equation representing the evolution of surface potential temperature anomalies induced by terrain is integrated within the model. The result, after a finite time, is the distribution of surface potential temperature anomalies (see figure) created by the mountain.

Xiaolei Zou, Ying-Hwa Kuo, and Simon Low-Nam studied the prediction of the ERICA IOP4 storm using a 120-km hemispheric version of the MM5, ranging from 36 h to 120 h. Specifically, they examined the impact of uncertainties in the model's initial condition on the 5-day forecast of the storm. They found that the skill of the model degraded steadily as the forecast duration was lengthened. A forecast fracture occurred between the 4.5- and 5-day forecasts, and the 5-day forecast failed to predict the cyclogenesis. They then used the MM5 adjoint system to determine the "optimal" initial condition by minimizing the errors of the initial 12-h forecast. The new initial condition derived from the MM5 adjoint system dramatically improved the 5-day forecast and captured the cyclogenesis. Analysis of the model results indicated that the main deficiencies in the original analysis were (1) poor description of the upper-level potential vorticity (PV) anomaly over the Gulf of Alaska and (2) erroneous lower-troposphere temperature analysis over the southern Rockies.

A case of mesoscale explosive cyclogenesis took place over the southern U.S. in March 1984 in which the cyclone deepened 11 mb in three hours. A previous MM5 forecast was unsuccessful in capturing the rapid pressure falls associated with this land-based mesoscale cyclone. Analysis of the model results indicated that the model did not capture the mesoscale precipitation system associated with the cyclogenesis. This failure was attributed to a deficiency in the model's initial condition. In collaboration with John Gyakum (McGill University, Canada), Kuo and Yong-Run Guo performed a predictability study on this case using the MM5 adjoint system. They obtained a set of initial perturbations by minimizing the differences between the model predicted surface pressure with the subjective analysis during a 9-h period, including the rapid cyclogenesis stage. The subtle changes in the initial conditions produced significant improvements in the model prediction, in terms of surface cyclone evolution, precipitation, and the upper-level jets.

Richard Carbone, James Wilson (ATD), Thomas Keenan (BMRC), Peter Hacker (Flinders University), N. Andrew Crook (joint appointment with RAP), and Mitchell Moncrieff investigated the evolution of convection over the Tiwi Islands north of the Australian continent as part of the Maritime Continent Thunderstorm Experiment (MCTEX). Observations from aircraft, radar, mesonet, flux stations, profilers, and satellite and Clark-Hall model simulations are involved. Examination of the sea breeze forcing indicates two scales and phases of evolution prior to the development of deep organized convection. Colliding peninsula breezes (10–20 km scale) initiate early convection in favored areas, but these are not directly related to the organized deep convection that follows some hours later. Island scale breezes (50–140 km) and the evolution of the island boundary layer (IBL) are instrumental in the initiation of convection that leads to "Hector," a mesoscale convective system (MCS) that occurs on approximately 70 percent of all days during the transition and wet seasons or the region. On more suppressed days, island scale breezes may converge or even collide near the islands’ center, directly initiating deep convection. More commonly, a free convective state is achieved along the downwind sea breeze front where the IBL is deepest and the density discontinuity with the marine boundary layer (MBL) is greatest (typically 0.3 – 1 percent). The breeze-forced convection achieves a free state when the convectively generated cold pool exceeds the breeze density discontinuity. Thereafter, cold pool dynamics overtake the organization of convection and one or more small tropical squalls develop according to ambient horizontal vorticity and IBL thermodynamic considerations. Hector typically evolves through the interaction of one such squall and one of the zonally oriented sea breeze fronts, reaching a mature MCS status by about 1530 Local Standard Time.

James Bresch (visitor, University of Washington) in cooperation with Morris Weisman and Jimy Dudhia examined the ability of the PSU-NCAR Mesoscale Model (MM5) to simulate the 7 May 1995 VORTEX squall line and the impact of using fine horizontal resolution and parameterizations upon the simulations. Experiments were conducted with horizontal grid spacings of 27, 9, and 3 km, covering the range from hydrostatic scales requiring convective parameterization to the fully explicit, non-hydrostatic scale. Because of the strong synoptic-scale forcing in this case, all three resolutions were able to produce a squall line in about the right time and place. As grid spacing decreased, smaller scale features became more plentiful, with the 3-km simulation able to correctly separate a leading mesoscale convective system from the squall line itself. However, the 3-km simulation overpredicted the precipitation amount and all resolutions were too fast with the squall line and overpredicted the areal rainfall coverage. For this case, increased horizontal resolution alone was insufficient to produce a significant improvement in the forecast. Because of the supplemental VORTEX datasets, this case is ideal for testing the potential benefits of 4-dimensional variational data assimilation (4DVAR) techniques on the mesoscale and future efforts should be directed in this area.

Rossella Ferretti (visitor, University of L'Aquilla, Italy), in collaboration with Low-Nam and Rotunno, studied the Piedmont flood of 6 November 1994 using the MM5 model over one of the Mesoscale Alpine Program (MAP) domains. The model results showed that the interaction of the large-scale flow with the mountains, in the presence of moisture, were the main factors contributing to the Piedmont flood. The Alps acted to slow the propagation of an eastward moving cold front from the west, as well as to block the moist unstable air from the south over the Mediterranean. The moist unstable flow from the south, meeting the drier air being channeled from the east, converged to produce the flooding over the Piedmont area. This study is in support of the international MAP program and supported by the Italian MAP.

Weisman and Davis completed an investigation into the mechanisms for the generation of mesoscale vortices at the ends of finite length convective lines. Through the use of idealized non-hydrostatic numerical cloud model simulations, they demonstrated that such vortices originate primarily via the tilting of system-generated horizontal vorticity associated with the cold pool/updraft interface, leading to the development of both cyclonic and anticyclonic line-end vortices to the left and right of the mean low-level vertical wind shear vector, respectively. For strongly sheared environments, such mesoscale vortices can also originate via the tilting of ambient horizontal vorticity within supercellular updraft/downdraft couplets, although the tilting of system-generated horizontal vorticity dominates in the later stages for these vortices as well. In either case, the convergence of Coriolis rotation over the longer term leads to the development of a preferred cyclonic vortex, which is frequently observed in asymmetric convective systems. The strength and scale of the resultant vortices was also found to depend on the strength of the ambient vertical wind shear, with the stronger, smaller vortices produced only in the more strongly sheared environments.

Rajul Pandya (ASP) and Weisman demonstrated that these three-dimensional flow features can be reproduced in a much simplified model by merely imposing the diabatic heating and cooling rates observed in the full simulations on an ambient sheared flow. These results emphasize that the cold pool processes must especially be represented in simplified numerical or theoretical models that are designed to study or forecast the generation of such features.

Much confusion has arisen recently in both the forecast and research communities as to the role of vertical wind shear versus storm-relative environmental helicity in controlling supercell dynamics for straight versus curved hodograph environments. In an attempt to remedy this confusion, Weisman and Rotunno analyzed a set of idealized numerical simulations of supercell storms to more clearly document the dynamical character of supercells in various shear regimes. This analysis clearly showed that the dynamical forcing of the updrafts for both the straight and curved hodographs were fundamentally the same, resulting primarily from the non-linear interactions between the updraft and environmental shear, which result from the development of updraft rotation, etc. This is in contrast to the helicity approach, which considers only linear updraft/shear interactions, which may bias cyclonic or anticyclonic updraft development in curved hodograph environments, but which do not contribute significantly to storm dynamics for straight hodographs. The present results confirm the universal character of the updraft-vertical wind shear interactions that lead to the production of long-lived, rotating convective storms, and reemphasize that supercell potential is related to the overall strength of the environmental vertical wind shear rather than just hodograph shape.

Recent observational studies highlight the fact that tornadogenesis within supercell storms may be significantly enhanced through interactions of a convective cell with a pre-existing, shallow surface boundary. In order to test this hypothesis, Nolan Atkins (ASP) and Weisman completed a set of simulations of the 16 May 1995 tornadic supercell storm observed during VORTEX, which developed along such a boundary. By analyzing simulations with and without various boundaries included, they documented the impact of such boundaries on the evolution of the low-level mesocyclone within the simulated storm, and preliminarily confirmed that such boundaries can enhance low-level mesocyclogenesis, but only for a narrow range of boundary strengths and orientations relative to the supercell. In general, the best interactions are found when the storm's independent motion is nearly parallel to the preexisting boundary, such that the horizontal vorticity associated with the boundary can feed in a streamwise sense into the low-level supercellular updraft.

Carbone, John Tuttle (joint appointment with RAP), and Roger Wakimoto (UCLA) began analysis of ELDORA (Electra Doppler Radar) data for the 31 May 1995 Sweetwater case, which occurred during VORTEX where an F1 tornado was spawned from a non-classical supercell storm. Two mesoscylones were observed during the Electra aircraft flight. Results from this study will be compared to other VORTEX cases in an attempt to associate environmental differences with factors instrumental in tornado genesis.

Weisman continued collaborations with Howard Bluestein (University of Oklahoma) on the study of supercell interactions within lines of convective storms. Simulations demonstrate that the ability to generate and sustain supercell storms within convective lines depends strongly on the orientation of the vertical wind shear profile relative to the line. For vertical wind shears parallel to a north-south oriented line, only the most southern or northern cells are able to develop supercell characteristics. However, for certain shear profiles oriented at significant angles to the line, a line of supercells may evolve. These simulations are being compared to observed supercell line cases to determine the applicability of these results to forecasting real supercell behavior in such situations.

Low-Nam, using the MM5 model in collaboration with Kuo and Alexis Lau (visitor, Hong Kong University of Science and Technology), studied the interaction of a mesoscale convective system with the tropical cyclone Gary, which formed over the South China Sea on 29 July 1995. The results from a 2-km MM5 simulation showed that the predicted cyclone track was strongly influenced by the mutual interaction between the MCS, which formed along the monsoon trough, and the approaching cyclone itself. The interaction is analogous to the Fujiwara effect, in that the two systems rotate about each other once they are within a radius of influence of each other and eventually merge. The mutual interaction causes a meandering effect in the resulting cyclone path.

An extreme heavy-rainfall event took place over the Central Mountain Range (CMR) of Taiwan as super-typhoon Herb 1996 (with an estimated central pressure of 935 mb), landed and passed through northern Taiwan. A maximum 24-h rainfall of 1749 mm was recorded at a mountain station. The extremely heavy rainfall in the mountain area triggered disastrous flooding, debris flow and land sliding, and took 73 lives. Kuo and Wei Wang performed a high-resolution simulation of this extreme heavy-rainfall event using the MM5 model at a grid resolution of 6.7 km. The model successfully simulated the track of the storm and the detailed mesoscale rainfall distribution and predicted a maximum 24-h rainfall of 936 mm. Additional model sensitivity experiments showed that the intensity of the rainfall amount and its distribution are highly sensitive to the model’s ability in resolving the detailed topography of the CMR. Analysis of the model results showed that the extremely heavy rainfall over the CMR was produced by the steady uplifting of a deep layer of moisture-laden typhoon circulation over the mountains.

In collaboration with Shih-Ting Wang (Central Weather Bureau, Taiwan), Kuo and Wang performed a study of Typhoon Dot (1990) and its interaction with the CMR of Taiwan. Typhoon Dot landed at the southeastern coast of Taiwan on 7 September 1990. The low-level circulation was destroyed by the CMR, while the upper-level circulation continued to move across the mountain. Rapid surface cyclogenesis then took place after the upper-level circulation moved over the mountains. The MM5 model, with a grid resolution of 6.7 km, successfully simulated the dissipation of the low-level cyclone and its "re-birth" after the upper-level vortex moved to the west side of the mountains. Analysis of the model simulation showed that the re-developed surface circulation follows a track very close to that of the upper-level circulation. Although a number of topographically induced mesolows formed on the lee side of the mountain as a result of subsidence warming, these systems are generally shallow and tied to the topography. The circulation aloft drove the formation and development of a secondary low (or the re-birth of the low-level circulation).

NCAR | UCAR | NSF | NCAR FY96 ASR | NCAR FY97 ASR