A NUMERICAL INVESTIGATION OF THE LAS VEGAS CONVERGENCE ZONE

Kim J. Runk    National Weather Service, Las Vegas, NV

1. Introduction

   Understanding how deep, moist convection is initiated and organized is fundamental to improving National Weather Service (NWS) short-term forecasts and warnings. Key processes associated with convective initiation and organization have been the subject of numerous investigations. These processes have been explored in a variety of geographical settings and across a broad spectrum of initiating/organizing mechanisms. A number of studies have focused on the sensitivity of cumulus convection to local variations in buoyancy and vertical wind shear (Weisman and Klemp 1982; Weisman and Klemp 1984; Crook 1996). Research and observations have demonstrated that thunderstorm initiation and evolution are commonly connected to, and modulated by, boundary-layer convergence zones (e.g., Szoke et al. 1984; Wilson and Schreiber 1986; Wilson et al. 1988; Wilczak and Christian 1990; Koch and Ray 1997; Wilson et al. 1998).

   Commissioning WSR-88D Doppler radars, associated with the NWS modernization program, and establishing local networks of automated surface observation platforms, primarily through interagency agreements, have greatly improved the capability to detect boundary-layer structures near developing convection nationwide. These observing systems offer capabilities to observe and study features which were not readily identifiable prior to the emergence of such technology.

   Within the Las Vegas valley of southern Nevada, an arc-shaped convergence zone has been repeatedly observed to form in the low-level wind field. The development and structure of this convergence zone has been documented previously for a specific case (Runk 1996), in which it was shown to have played a key role in explosive convective growth. The purpose of this paper is to expand the original work on the Las Vegas Convergence Zone (LVCZ) in order to isolate the most important processes associated with its development. This was accomplished by comparing multiple numerical modeling simulations with surface observations and radar imagery from actual LVCZ cases, using version 3b of the Aster Corporation Regional Atmospheric Modeling System (RAMS, Pielke et al. 1992). By varying the initialization, radiation, and microphysics, we have tested the hypothesis that the convergence zone is chiefly related to a complex interaction between the background wind field and a mountain-valley solenoid generated by differential heating, such as those studied by Banta (1984); and Schreiber-Abshire and Rodi (1991).

   We will demonstrate that relatively weak mid-level flow from the southwest coupled with an elevated region of low static stability in the presence of strong solar insolation is responsible for the formation of the classic LVCZ circulation. In contrast, strong mid-level winds (i.e., higher Froude number flow) produces a circulation over the orographic barriers with a different structure than that which is observed during a LVCZ event. Moreover, even if all other parameters are ideal, the direction of the mean wind in the 1-4 km layer plays a crucial role in determining the formation and structure of the boundary-layer convergence zone.

2. Background

   The Las Vegas metropolitan area is situated in a northwest-southeast valley, the floor of which is approximately 2200 ft MSL (Fig. 1). To the west are the Spring Mountains, which average about 6000 ft MSL, but include two peaks which are significantly higher: Mt. Charleston at 11,910 ft and Mt. Potosi, at 8800 ft. At the north end of the valley lies the Sheep Range, topped by Hayford Peak at 9912 ft. To the south, the McCollough Range varies from 3000 to 7000 ft. Farther east, the terrain slopes downward rapidly into the Colorado River basin and Lake Mead, then ascends again into the high mesa country of northwest Arizona. This complex terrain has an enormous influence on local circulations.

   The classic LVCZ circulation is illustrated in Fig. 2. In this figure, a radar reflectivity image, valid 2039 UTC 30 July 1997, is superimposed with corresponding surface winds from the local mesonet. A region of confluence extends northwest-southeast along the lee slopes of the Spring Mountains, then veers to an east-west orientation across the southern Las Vegas valley to Lake Mead. This configuration typifies the early afternoon stage of LVCZ development. The real-time RAMS simulation of this particular event failed to capture this signature because convection in the model overwhelmed the base state wind field by early afternoon. In addition, the real-time simulation was conducted at 10-km horizontal grid spacing on the inner nest, a resolution which may have been too coarse to accurately depict the CZ, regardless of whether deep convection occurred in the model. For this study, all simulations were conducted with a fine-nest horizontal grid spacing of 4 km and convective precipitation processes were turned off.

   Four cases were initially selected for base-line simulations, each representing a day in which the classic LVCZ circulation was clearly observed in the mesonet wind field. Each of these cases was run on a two-way interactive nest, initialized with real data, using the NCEP 29-km Eta for lateral boundary conditions. Time steps of 30 sec were computed with hourly output from 1200-2400 UTC. The Mahrer-Pielke radiation scheme was utilized, with the exception of one dry run with no active radiation at all. The model terrain was a 1-km NCAR grid. In each case where a diurnal radiation cycle was included, the finer resolution coupled with lack of convection resulted in successful simulation of the LVCZ circulation (not shown).

3. Analysis

   After demonstrating the LVCZ could be reproduced when initialized with real data, follow-up simulations were then conducted with horizontally homogeneous initialization from a single proximity sounding. The sounding profile was initially based on the rawinsonde observation from Desert Rock, NV, modified at the surface for conditions in the Las Vegas valley. The surface parameters were taken from NWS observations. In each case, a surface-based inversion existed at initialization time (1200 UTC) with a deep, well-mixed nearly adiabatic layer extending from the inversion top to at least 4 km above ground level (AGL). This well-mixed layer will hereafter be referred to as the elevated boundary layer (EBL).

   In subsequent simulations, we systematically varied the following conditions for each sounding:

   >    wind direction in the EBL

   >    wind speed in the EBL

   >    lapse rate in the EBL

   >    strength and depth of the low-level inversion layer

   Results from these data revealed important characteristics with respect to processes governing boundary-layer convergence zone development and structure.

   The first set of horizontally homogeneous runs utilized a 5 K surface-based inversion extending to ~500 m AGL. Surface winds were from the northeast at 3 ms^-1, veering through the inversion layer and becoming southwest (230º) at 1 km. The 1-4 km mean wind was 230º at 7 ms^-1.

   Between 1500 and 1600 UTC, the inversion was eliminated by diurnal heating, resulting in a pressure gradient favorable to upslope acceleration. As air drawn from the valley interior impinged on southwest flow aloft, the convergence zone formed and continued to deepen throughout the morning hours. The early stages of the circulation were clearly visible in both the low-level wind field and convergence contours by 1700 UTC (Fig. 3). The relatively cool valley air was apparently replaced by warmer air from above, completing a thermally forced mountain-valley solenoid (Fig. 4). Once the mature CZ was in place, moisture pooling was observed to occur along the interface prior to the inversion being eroded. This acted to pre-condition the local environment for deep convection by destabilizing the air within the convergence zone.

   As heating continued, the primary convergence zone advanced downwind (i.e., to the northeast) and a secondary convergence zone developed along the I-15 corridor, orthogonal to the original CZ (Fig. 5). This secondary region is attributed primarily to gap flow between the Spring Mountains and the McCollough Range.

   In the vertical, both the rising and descending branches of the mountain-valley circulation strengthened as the afternoon progessed. The intensifying downward-branch may have played an important role in inhibiting convective initiation during the afternoon on the east side of the valley. This is consistent with observations of actual LVCZ events. Strong thunderstorms over the east side of the valley are typically associated with advection into the area, but the region is much less prone to thunderstorm initiation compared to western or southern valley locations.

   Storms that form along the eastern slopes of the Spring Mountains or the northern slopes of the McCollough Range have frequently been observed to intensify rapidly as they move off the mountains toward the valley. Figure 6 provides a clue to the mechanism responsible for this common evolution. Local horizontal relative vorticity values on the order of 10^-3 s^-1 are generated within the convergence zone as the circulation matures. If vorticity of this magnitude is tilted and stretched into a convective updraft, the attendant instantaneous surge of moist lift and rotation could easily be envisioned to take control of the upward-branch of the mountain-valley solenoid and deepen it to tropopause level within minutes. This is precisely the evolution documented in the original LVCZ case of a devastating hailstorm and convective downburst event on 23 August 1995.

   Using the same thermal profile as the southwest flow case, additional simulations were run, varying the 1-4 km mean wind direction and speed. With some variation in orientation, the LVCZ forms, as long as the mean wind direction lies between 200º and 280º and the mean wind speed does not exceed 10 ms^-1.

   This latter limitation roughly equates to a 0.3 Froude number flow. As the Froude number approaches this threshold, a greater percentage of the ambient wind flows over and around the mountain barriers and no discrete convergence zone is formed.

   If the 1-4 km mean wind is less than 10 ms^-1, but the direction is outside the range given above, smaller regions of boundary-layer convergence develop, but with substantially different shape (and generally weaker intensity) than the classic LVCZ circulation. Boundaries such as these undoubtedly also play a role in regulating the local wind field, and interact with circulations on other scales. However, since these flow directions are less frequently associated with situations in which deep moisture and convective instability are present, their specific contribution toward modulating thunderstorm development was not pursued in the course of this study.

   Vertical stability also affects the character of CZ development, albeit less significantly than the mean wind. In our numerical investigation, the stability profile was varied in two ways. First, the lapse rate in the 1-4 km layer was adjusted. As the mid-tropospheric lapse rate decreases, the evolving CZ circulation weakens. This is presumably because deep, rapid mixing is more limited under such conditions. Second, modifications were made in the strength and depth of the surface-based inversion layer. As expected, if the inversion is too strong to be eroded by solar insolation, the surface winds remain decoupled from the winds in the EBL, and the CZ does not form. Furthermore, if the inversion is on the order of 20-25 K, the time it takes to realize the amount of heating required to break the cap appears to inhibit the CZ from achieving maturity. However, the number of test cases run thus far is insufficient to determine a reasonable parameter for how deep or strong the inversion must be to preclude CZ development altogether.

   When no low-level inversion is present, the CZ develops quickly, but the strength of convergence is relatively weak and the boundary begins to move out across the valley during the early morning hours. This suggests that the cooler air below the inversion and the attendant east-to-northeast winds play an important role in the timing, strength and location of CZ formation, as well as its subsequent motion away from the mountains. In addition, the convergence zone that does form in a null inversion case is less prone to convective initiation. This is because the inversion itself is essential for local pre-moistening of the environment in the vicinity of the CZ boundary.

4. Conclusions

   While this study is not exhaustive, we have been able to deduce several key factors which are both necessary and sufficient to form the classic LVCZ circulation.

   These factors include:

   >    Mean wind direction in the 1-4 km layer AGL between 200º and 280º (within 40º of normal to the Spring Mountains)

   >    Mean wind speed in the 1-4 km layer AGL less than 10 ms^-1 (optimum ~5-7 ms^-1)

   >    A deep, well-mixed unstable boundary layer elevated above a shallow, surface-based inversion in the early morning hours

   >    Froude number less than 0.30 (optimum ~0.10)

   The classic LVCZ is the result of an interaction between these prerequisite conditions and a thermally forced mountain-valley circulation. By mid-afternoon, the mature LVCZ develops horizontal relative vorticity values on the order of 10^-3 s^-1 (Fig. 6). This vorticity represents a concentrated source of available potential energy which has been shown to enhance convective growth as mountain-generated storms intersect the convergence zone. The corresponding descending branch of the mountain-valley solenoid is a likely inhibiting factor for convective development in that vicinity.

   A conceptual diagram of the typical LVCZ life cycle is provided in Figure 7. The CZ forms during the late morning and early afternoon along the Spring Mountain foothills, then extends east-west across the southern Las Vegas valley to Lake Mead. The boundary then drifts northeastward with the mean wind during the afternoon and early evening hours. Gap flow between the Spring Mountains and the McCollough Range frequently generates a secondary convergence zone southwest of Las Vegas along I-15. This boundary occasionally acts as a focusing mechanism for new convective initiation late in the day.

5. References

   Banta, R.M., 1984: Daytime boundary layer evolution over mountainous terrain. Part I: Observations of the dry circulations. Mon. Wea. Rev., 112, 340-356.

   Crook, N.A., 1996: Sensitivity of moist convection forced by boundary layer processes to low-level thermodynamic fields. Mon. Wea. Rev., 124, 1767-1785.

   Koch, S.E., and C.A. Ray, 1997: Mesoanalysis of summertime convergence zones in central and eastern North Carolina. Wea. Forecasting, 12, 56-77.

   Pielke, R.A., W.R. Cotton, R.L. Walko, C.J. Trembeck, W.A. Lyons, L.D. Grasso, M.E. Nicholls, M.D. Moran, D.A. Wesley, T.J. Lee, and J.H. Copeland, 1992: A comprehensive meteorological modeling system - RAMS. Meteor. Atmos. Phys., 69-91.

   Runk, K.J., 1996: The Las Vegas convergence zone: Its development, structure and implications for forecasting. NWS Western Region Tech. Attach. 96-18, 21p.

   Schreiber-Abshire, W., and A. Rodi, 1991: Mesoscale convergence zone development in northeastern Colorado under southwest flow. Mon. Wea. Rev., 119, 2926-2977.

   Szoke, E.J., M.L. Weisman, J.M. Brown, F. Caracena, and T.W. Schalatter, 1984: A sub-synoptic analysis of the Denver tornado of 3 June 1981. Mon. Wea. Rev., 112, 790-808.

   Weisman, M.L., and J.B. Klemp, 1984: The structure and classification of numerically simulated convective storms in directionally varying wind shears. Mon. Wea. Rev., 112, 2479-2498.

   Weisman, M.L., and J.B. Klemp, 1982: The dependence of numerically simulated convective storms on vertical wind shear and buoyancy. Mon. Wea. Rev., 110, 504-520.

   Wilczak, J., and T. Christian, 1990: Case study of an orographically Induced mesoscale vortex (Denver cyclone). Mon. Wea. Rev., 118, 1082-1102.

   Wilson, J.W., and W.E. Schreiber, 1986: Initiation of convective storms at radar-observed boundary layer convergence lines. Mon. Wea. Rev., 114, 2515-2536.

   Wilson, J.W., J.A. Moore, G.B. Foote, B. Martner, A.R. Rodi, T. Uttal, and J.M. Wilczak, 1988: Convection initiation and downburst experiment (CINDE). Bull. Amer. Meteor. Soc., 69, 1328-1348.

   Wilson, J.W., N.A. Crook, C.K. Mueller, J. Sun, and M. Dixon, 1998: Nowcasting thunderstorms: A status report. Bull. Amer. Meteor. Soc., 79, 2079-2100.

6. Acknowledgments

   The authors wish to extend their appreciation to Drs. Doug Wesley and Greg Byrd for their helpful editorial suggestions, and to NWS Western Region Headquarters for its support and encouragement in this project. This paper is funded in part by cooperative agreement #NA67WD0097 from the National Oceanic and Atmospheric Administration (NOAA). The views expressed herein are those of the authors and do not necessarily reflect the views of NOAA or any of its sub-agencies. The Cooperative Program for Operational Meteorology and Training (COMET^®) is primarily funded by the NWS, with additional funding from NMOC and AFWA.