• Anthes, R. A., , and T. T. Warner, 1978: Development of hydrodynamical models suitable for air pollution and other mesometerological studies. Mon. Wea. Rev., 106 , 10451078.

    • Search Google Scholar
    • Export Citation
  • Blackadar, A. K., 1979: High resolution models of the planetary boundary layer. Advances in Environmental Science and Engineering, Vol. 1, No. 1, J. R. Pffafflin and E. N. Zeigler, Eds., Gordon and Breach, 50–85.

    • Search Google Scholar
    • Export Citation
  • Castle, J. A., , J. D. Locatelli, , J. E. Martin, , and P. V. Hobbs, 1996: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part IV: The evolution of a drytrough on 8–9 March 1992. Mon. Wea. Rev., 124 , 15911595.

    • Search Google Scholar
    • Export Citation
  • Crook, N. A., , and M. W. Moncrieff, 1988: The effects of large-scale convergence on the generation and maintenance of deep moist convection. J. Atmos. Sci., 45 , 36063624.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1993: A nonhydrostatic version of the Penn State–NCAR Mesoscale Model: Validation tests and simulation of an Atlantic cyclone and cold front. Mon. Wea. Rev., 121 , 14931513.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., , M. W. Moncrieff, , and D. W. K. So, 1987: Two dimensional dynamics of West African squall lines. Quart. J. Roy. Meteor. Soc., 113 , 121146.

    • Search Google Scholar
    • Export Citation
  • Fankhauser, J. C., , G. M. Barnes, , and M. A. LeMone, 1992: Structure of a midlatitude squall line formed in strong unidirectional shear. Mon. Wea. Rev., 120 , 237260.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., , J. Dudhia, , and D. R. Stauffer, 1994: A description of the fifth generation Penn State/NCAR Mesoscale Model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 121 pp. [Available from National Center for Atmospheric Research, Boulder, CO 80301-3000.].

    • Search Google Scholar
    • Export Citation
  • Hobbs, P. V., , J. D. Locatelli, , and J. E. Martin, 1990: Cold fronts aloft and the forecasting of precipitation and severe weather east of the Rocky Mountains. Wea. Forecasting, 5 , 613626.

    • Search Google Scholar
    • Export Citation
  • Hobbs, P. V., , J. D. Locatelli, , and J. E. Martin, 1996: A new conceptual model for cyclones generated in the lee of the Rocky Mountains. Bull. Amer. Meteor. Soc., 77 , 11691178.

    • Search Google Scholar
    • Export Citation
  • Houze Jr., R. A., , S. A. Rutledge, , M. I. Biggerstaff, , and B. F. Smull, 1989: Interpretation of Doppler weather radar displays of midlatitude mesoscale convective systems. Bull. Amer. Meteor. Soc., 70 , 608619.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., , and J. M. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain–Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor. Monogr., No. 46, Amer. Meteor. Soc., 165–170.

    • Search Google Scholar
    • Export Citation
  • Koch, S. E., , and J. McCarthy, 1982: The evolution of an Oklahoma dryline. Part II: Boundary-layer forcing of meso-convective systems. J. Atmos. Sci., 39 , 237257.

    • Search Google Scholar
    • Export Citation
  • Koch, S. E., , and L. M. Siedlarz, 1999: Mesoscale gravity waves and their environment in the central United States during STORM-FEST. Mon. Wea. Rev., 127 , 28542879.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , and P. V. Hobbs, 1995: A world record rainfall rate at Holt, Missouri: Was it due to cold frontogenesis aloft? Wea. Forecasting, 10 , 779785.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , J. E. Martin, , J. A. Castle, , and P. V. Hobbs, 1995: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part III: The development of a squall line associated with weak cold-frontogenesis aloft. Mon. Wea. Rev., 123 , 26412662.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , M. T. Stoelinga, , R. D. Schwartz, , and P. V. Hobbs, 1997: Surface convergence induced by cold fronts aloft and prefrontal surges. Mon. Wea. Rev., 125 , 28082820.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , M. T. Stoelinga, , and P. V. Hobbs, 1998: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part V: Thermodynamic and dual-Doppler radar analysis of a squall line associated with a cold front aloft. Mon. Wea. Rev., 126 , 860875.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , R. D. Schwartz, , M. T. Stoelinga, , and P. V. Hobbs, 2002a: Norwegian-type and cold front aloft–type cyclones east of the Rocky Mountains. Wea. Forecasting, 17 , 6682.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , M. T. Stoelinga, , and P. V. Hobbs, 2002b: A new look at the super outbreak of tornadoes on 3–4 April 1974. Mon. Wea. Rev., 130 , 16331651.

    • Search Google Scholar
    • Export Citation
  • Maddox, R. A., 1980: Mesoscale convective complexes. Bull. Amer. Meteor. Soc., 61 , 13741387.

  • Martin, J. E., , J. D. Locatelli, , and P. V. Hobbs, 1990: Organization and structure of clouds and precipitation on the mid-Atlantic coast of the United States. Part III: The evolution of a middle-tropospheric cold front. Mon. Wea. Rev., 118 , 195217.

    • Search Google Scholar
    • Export Citation
  • Martin, J. E., , J. D. Locatelli, , P. V. Hobbs, , P-Y. Wang, , and J. A. Castle, 1995: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part I: A synoptic-scale rainband associated with a dryline and lee trough. Mon. Wea. Rev., 123 , 241264.

    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , and R. M. Wakimoto, 1999: The interaction of a Pacific cold front with shallow masses east of the Rocky Mountains. Mon. Wea. Rev., 127 , 21022127.

    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , F. M. Ralph, , M. A. Shapiro, , B. F. Smull, , and D. Johnson, 1998: An observational study of fronts and frontal mergers over the continental United States. Mon. Wea. Rev., 126 , 25212554.

    • Search Google Scholar
    • Export Citation
  • Newton, C. W., 1950: Structure and mechanisms of the prefrontal squall line. J. Meteor., 7 , 210222.

  • Ogura, Y., , H-M. Juang, , K-S. Zhang, , and S-T. Soong, 1982: Possible triggering mechanisms for severe storms in SESAME-AVE IV (9–10 May 1979). Bull. Amer. Meteor. Soc., 63 , 503515.

    • Search Google Scholar
    • Export Citation
  • Reisner, J., , R. M. Rasmussen, , and R. T. Bruintjes, 1998: Explicit forecasting of supercooled liquid water in winter storms using the MM5 mesoscale model. Quart. J. Roy. Meteor. Soc., 124 , 10711107.

    • Search Google Scholar
    • Export Citation
  • Rose, S. F., , P. V. Hobbs, , J. D. Locatelli, , and M. T. Stoelinga, 2002: Use of a mesoscale model to forecast severe weather associated with a cold front aloft. Wea. Forecasting, 17 , 755773.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , and J. B. Klemp, 1982: The influence of the shear-induced pressure gradient on thunderstorm motion. Mon. Wea. Rev., 110 , 136151.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , and J. B. Klemp, 1985: On the rotation and propagation of simulated supercell thunderstorms. J. Atmos. Sci., 42 , 271292.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , J. B. Klemp, , and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci., 45 , 463485.

  • Schaefer, J. T., 1986: The dryline. Mesoscale Meteorology and Forecasting, P. S. Ray, Ed., Amer. Meteor. Soc., 549–572.

  • Shapiro, M. A., 1982: Mesoscale weather systems of the central United States. NOAA–CIRES Tech. Rep., 78 pp. [Available from Cooperative Institute for Research in the Environmental Sciences, University of Colorado, Boulder, CO 80309.].

    • Search Google Scholar
    • Export Citation
  • Showalter, A. K., , and J. R. Fulks, 1943: Preliminary report on tornadoes. War Advisory Council on Meteorology Restricted Rep. 1110, U.S. Weather Bureau, 162 pp. [Available from NOAA Central Library, SSMC3 7th Floor, 1315 East–West Highway, Silver Springs, MD 20910.].

    • Search Google Scholar
    • Export Citation
  • Sienkiewicz, J. M., , J. D. Locatelli, , P. V. Hobbs, , and B. Geerts, 1989: Organization and structure of clouds and precipitation on the mid-Atlantic coast of the United States. Part II: The mesoscale and microscale structure of some frontal rainbands. J. Atmos. Sci., 46 , 13491364.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., , J. D. Locatelli, , and P. V. Hobbs, 2000: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part VI: A multiscale numerical modeling study of the initiation of convective rainbands. Mon. Wea. Rev., 128 , 34813500.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., , J. D. Locatelli, , and P. V. Hobbs, 2002: Warm occlusions, cold occlusions, and forward-tilting cold fronts. Bull. Amer. Meteor. Soc., 83 , 709721.

    • Search Google Scholar
    • Export Citation
  • Thorpe, A. J., , M. J. Miller, , and M. W. Moncrieff, 1982: Two-dimensional convection in nonconstant shear: A model of midlatitude squall lines. Quart. J. Roy. Meteor. Soc., 108 , 739762.

    • Search Google Scholar
    • Export Citation
  • Wang, P. Y., , J. E. Martin, , J. D. Locatelli, , and P. V. Hobbs, 1995: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part II: Arctic fronts. Mon. Wea. Rev., 123 , 13281344.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., 1992: The role of convectively generated rear-inflow jets in the evolution of long-lived mesoconvective systems. J. Atmos. Sci., 49 , 18261847.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., , J. B. Klemp, , and R. Rotunno, 1988: Structure and evolution of numerically simulated squall lines. J. Atmos. Sci., 45 , 19902013.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., , W. C. Skamarock, , and J. B. Klemp, 1997: The resolution dependence of explicitly modeled convective systems. Mon. Wea. Rev., 125 , 527548.

    • Search Google Scholar
    • Export Citation
  • Zhang, D. L., , and R. A. Anthes, 1982: A high-resolution model of the planetary boundary layer—Sensitivity tests and comparisons with SESAME-79 data. J. Appl. Meteor., 21 , 15941609.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Surface map for 0600 UTC 20 Jun 1979 showing the NWS analysis of sea level pressure (solid lines in hPa), surface features (conventional symbols), stations reports (conventional plotting), and regions of precipitation (shaded areas) enclosed by the lowest-level radar echo from the National Radar Summary. Selected damage reports are noted in the boxes

  • View in gallery

    (a) Sea level pressure (solid lines in hPa), 700-hPa temperature (dashed lines in °C), and lowest-level radar echo from the National Radar Summary for the CFA squall line at 1200 UTC 19 Jun 1979. (b) As in (a) but at 1200 UTC 20 Jun 1979

  • View in gallery

    (a) Sea level pressure and winds from the NWS surface map in Fig. 1. (b) Authors' surface temperature analysis (°F) and NWS winds from the surface map in Fig. 1. Shading shows regions of precipitation enclosed by the lowest-level radar echo from the National Radar Summary. (c) As in (b) but for authors' dewpoint analysis (°F). (d) Surface winds (see legend in figure) and sea level pressure (hPa) valid for 0600 UTC 20 Jun 1979 for the MM5 coarse-grid control simulation. Shading shows regions of model precipitation. (e) As in (d) but for surface temperature (°F) from the model coarse-grid control simulation. (f) As in (d) but for dewpoint (°F) from the model coarse-grid control simulation

  • View in gallery

    The 27-h MM5 coarse-grid forecast valid at 0300 UTC 20 Jun 1979: (a) surface winds, sea level pressure (heavy solid lines in hPa), and precipitation totals for model hours 26–27 (shading: see figure legend) from the coarse-grid control simulation, and (b) from the coarse-grid no-cooling simulation

  • View in gallery

    The 27-h MM5 coarse-grid forecast valid at 0300 UTC 20 Jun 1979: (a) sea level pressure (heavy solid lines in hPa), dewpoint contours (thin solid lines in °C), and precipitation totals from model hours 26–27 (shading: see figure legend) from the MM5 coarse-grid control simulation, and (b) from the coarse-grid no-cooling simulation

  • View in gallery

    As in Fig. 5 but thin solid lines are surface potential temperature (θ) contours (K)

  • View in gallery

    (a) The 27-h MM5 coarse-grid no-cooling simulation of sea level pressure valid at 0300 UTC 20 Jun 1979 minus the coarse-grid control simulation sea level pressure forecast (hPa) for the same time. Precipitation totals from model hours 26–27 (shading: see figure legend) are from the coarse-grid control simulation. (b) As in (a) but for the difference in the 950–800-hPa thickness (m)

  • View in gallery

    (a) The 27-h MM5 coarse-grid control simulation valid at 0300 UTC 20 Jun 1979 for the pressure field (hPa) produced by the atmospheric layer from 0.4 to 1.6 km. The dashed–dotted line is the axis of the highest pressure in the cold pool, and the dashed–double-dotted line is the axis of a trough of low pressure (see text for explanation). Precipitation totals from model hours 26–27 (shading: see figure legend) are from the coarse-grid control simulation. (b) As in (a) but for the atmospheric layer from 1.6 to 9.0 km. The heavy dashed line is the axis of a trough of low pressure (see text for explanation)

  • View in gallery

    As in Fig. 8 but for the MM5 coarse-grid no-cooling simulation for (a) the atmospheric layer from 0.4 to 1.2 km and (b) the atmospheric layer from 1.2 to 9.0 km

  • View in gallery

    Surface features from the MM5 high-resolution control simulation at (a) 2230 UTC 19 Jun 1979 and (b) 0300 UTC 20 Jun 1979. Shown are wind vectors at the lowest model level [vector magnitude scale in lower right of (b)], dewpoint temperature at the lowest model level (solid contours, every 2°C), and 15-min rainfall amounts (light shading, 1–5 mm; dark shading, >5 mm). Hash marks on perimeter represent intervals of 10 grid points (36 km). Heavy dashed line in (b) is the location of the surface trough–dryline. Line A–A′ is the location of the cross sections shown in Figs. 13a and 13b

  • View in gallery

    Vertical velocity and cold pool depiction from the high-resolution control simulation and the no-cooling simulation at 0100 UTC 20 Jun 1979. (a) Downdrafts (shading, scale shown at bottom) and updrafts (thin solid contours at values of 50, 100, and 200 cm s−1) at a height of 2.1 km AMSL. (b) As in (a) except for the no-cooling simulation. (c) Virtual potential temperature at the lowest model level in the control simulation minus that in the no-cooling simulation (shading, scale at bottom—note that the light gray shade outlined with a thin contour corresponds to a warm anomaly) and 15-min rainfall amounts from the control simulation (solid contours at 1- and 5-mm amounts). Hash marks on perimeter represent intervals of 10 grid points (36 km)

  • View in gallery

    As in Fig. 10 except for the no-cooling simulation at (a) 0300 UTC 20 Jun 1979 and (b) 0600 UTC 20 Jun 1979. Line A–A′ is the location of the cross sections shown in Figs. 13a and 13b, B–B′ is the location of the cross section shown in Fig. 14, and line C–C′ is the location of the cross section shown in Fig. 13c

  • View in gallery

    Vertical cross sections through the squall line produced by the high-resolution model simulations. Shown are squall-line-relative vectors in the plane of the cross section (scale at lower right or left—note dp/dt of 1 dPa s−1 corresponds to a vertical velocity of ∼1 cm s−1), virtual potential temperature (solid contours, every 2 K), equivalent potential temperature (shading, scale at bottom), an outline of the cloud and precipitation boundary (heavy solid line, corresponding to 0.025 g kg−1 for the sum of all cloud and precipitation mixing ratios), and a subjectively placed outline of the Pacific cold-frontal surface (heavy dashed line). (a) Cross section from the control simulation at 0300 UTC 20 Jun 1979 along line A–A′ in Figs. 10b and 12a. (b) As in (a) except for no-cooling simulation. (c) Cross section from the no-cooling simulation at 0600 UTC 20 Jun 1979 along line C–C′ in Fig. 12b

  • View in gallery

    (Continued)

  • View in gallery

    Vertical cross section through the squall line produced by the high-resolution no-cooling simulation at 0300 UTC 20 Jun 1979 along line B–B′ in Fig. 12a. Shown are squall-line-relative vectors in the plane of the cross section (scale at lower right—note dp/dt of 1 dPa s−1 corresponds to a vertical velocity of ∼1 cm s−1), virtual temperature perturbation (heavy gray contours, every 1°C), and pressure perturbation (thin black contours, every 1 hPa). Perturbation fields are based on a standard atmospheric temperature profile and corresponding hydrostatic pressure profile that are functions of height only. Regions where the perturbation fields are higher or lower are labeled as such to give a sense of the directions of the gradients.

  • View in gallery

    Schematic showing the differences between (a) a CFA squall line and (b) a warm-sector squall line. See text for explanation.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 57 57 5
PDF Downloads 36 36 4

Is a Cold Pool Necessary for the Maintenance of a Squall Line Produced by a Cold Front Aloft?

View More View Less
  • 1 Department of Atmospheric Sciences, University of Washington, Seattle, Washington
© Get Permissions
Full access

Abstract

On 19–20 June 1979 a cyclone moved through the central United States. This cyclone contained a squall line associated with a cold front aloft (CFA), which caused significant damage. The fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) was used to diagnose the role of the cold pool in the maintenance of the squall line. A control simulation with “full physics” was run, at 18- and 3.6-km grid spacing. Both simulations produced a squall line that was similar in location, orientation, and speed to the observed squall line, and displayed several characteristics that differed from the “leading line–trailing stratiform” paradigm for midlatitude squall lines. Sensitivity test simulations were run for both grid spacings, with diabatic cooling due to evaporation and melting of precipitation withheld to prevent the formation of a cold pool. These simulations produced a squall line similar to that in the control simulation, in terms of the location, orientation, and movement of the squall line. The simulations showed that the CFA provided the primary lifting responsible for the maintenance and movement of the simulated squall line by means of the hydrostatic surface pressure pattern it induced. The cold pool did not play a critical role in the maintenance of the simulated CFA squall line, but it did retard the progression of the synoptic-scale trough that trailed the simulated squall line, thereby increasing the forward tilt of the Pacific cold front.

Corresponding author address: Peter V. Hobbs, Dept. of Atmospheric Sciences, University of Washington, Box 351640, Seattle, WA 98195-1640. Email: phobbs@atmos.washington.edu

Abstract

On 19–20 June 1979 a cyclone moved through the central United States. This cyclone contained a squall line associated with a cold front aloft (CFA), which caused significant damage. The fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) was used to diagnose the role of the cold pool in the maintenance of the squall line. A control simulation with “full physics” was run, at 18- and 3.6-km grid spacing. Both simulations produced a squall line that was similar in location, orientation, and speed to the observed squall line, and displayed several characteristics that differed from the “leading line–trailing stratiform” paradigm for midlatitude squall lines. Sensitivity test simulations were run for both grid spacings, with diabatic cooling due to evaporation and melting of precipitation withheld to prevent the formation of a cold pool. These simulations produced a squall line similar to that in the control simulation, in terms of the location, orientation, and movement of the squall line. The simulations showed that the CFA provided the primary lifting responsible for the maintenance and movement of the simulated squall line by means of the hydrostatic surface pressure pattern it induced. The cold pool did not play a critical role in the maintenance of the simulated CFA squall line, but it did retard the progression of the synoptic-scale trough that trailed the simulated squall line, thereby increasing the forward tilt of the Pacific cold front.

Corresponding author address: Peter V. Hobbs, Dept. of Atmospheric Sciences, University of Washington, Box 351640, Seattle, WA 98195-1640. Email: phobbs@atmos.washington.edu

1. Introduction

On 19–20 June 1979 a destructive squall line moved through the central United States. Maddox (1980) described this squall line as a typical example of a “warm-sector squall line” that “produced tornadoes, large hail, and straight-line winds exceeding 75 kt.… [and it occurred] within the warm sector of an intense, large-scale cyclone. …” Figure 1 is the National Weather Service (NWS) surface map for 0600 UTC 20 June 1979. Listed in the figure are some selected reports of damage produced by the squall line, and the region of radar echoes from the National Radar Summary for the same time is superimposed and denoted by shading. The NWS analysis supports Maddox's interpretation that most of the squall line was located in a warm sector, between a cold and a warm front.

In several papers (Sienkiewicz et al. 1989; Martin et al. 1990, 1995; Hobbs et al. 1990, 1996; Locatelli and Hobbs 1995; Locatelli et al. 1995, 1997, 1998, 2002a,b; Wang et al. 1995; Castle et al. 1996; Neiman et al. 1998; Neiman and Wakimoto 1999; Koch and Siedlarz 1999; Stoelinga et al. 2000; Rose et al. 2002), cyclones have been discussed that look similar to the 19–20 June 1979 cyclone, in which the Pacific cold frontal zone is located farther forward aloft than at the surface, resulting in a cold front aloft (CFA). Frequently, an extensive squall line is aligned along the CFA. Such cyclones form when a Pacific cold front crosses the Rocky Mountains, overtakes a lee trough to the east of the Rockies, and then moves aloft over statically stable air to the east of the trough. In these cyclones, the feature that was operationally analyzed as a surface cold front was not a classical backward-tilted cold front. Instead it had the properties of a warm occluded front, in which the baroclinic zone of the cold front tilted forward with height. Hence, the conceptual model of a warm sector is not applicable in these situations.

Newton (1950) proposed a mechanism for the self-propagation of midlatitude squall lines that has become widely accepted. This mechanism involves an interdependence between convective updrafts and a density-current-like cold pool. The cold pool lifts conditionally unstable air ahead of the squall line to initiate new convective updrafts, and the convective updrafts strengthen the cold pool by producing precipitation that falls into and cools the dry descending air behind the updraft region. This idea forms the basis of a theory relating the longevity of squall lines to the degree of balance between the strength of the cold pool and the environmental low-level wind shear (Thorpe et al. 1982; Rotunno et al. 1988).

Another paradigm for the maintenance of squall lines is based on the idea that an external forcing mechanism can be responsible for initiating and maintaining a squall line by providing a linearly oriented and persistent region of lifting that initiates new convective updrafts (Crook and Moncrieff 1988). In this paradigm, the external forcing serves the same role as the cold pool in the self-maintenance paradigm. Crook and Moncrieff showed that the external forcing need not be particularly strong, provided that it acts for sufficient time to lift low-level parcels to their level of free convection. The implication is that even lifting of several centimeters per second might initiate and maintain a squall line if it acts over a region of several tens of kilometers in width in appropriate environmental conditions. Crook and Moncrieff performed numerical simulations with an idealized gravity wave as the external forcing mechanism, and were able to produce and maintain a squall line whether or not diabatic cooling and precipitation loading were included in the simulations.

Several of the CFA-related studies cited above, in particular the Locatelli et al. (1998) Doppler radar study of a CFA squall line, have provided support for Crook and Moncrieff's paradigm with the CFA acting as the forcing mechanism. Squall lines associated with CFAs typically develop rapidly from their initial state into a line that is 1000 km or greater in length and is oriented in an arc-shaped band that is not connected to any apparent surface frontal feature, but is collocated with the leading edge aloft of a forward-tilted eastward-moving Pacific cold front.

Many studies have shown that convection can be initiated when a cold front overtakes a dryline, or lee trough (Koch and McCarthy 1982; Ogura et al. 1982; Shapiro 1982; Schaefer 1986; Neiman et al. 1998; Neiman and Wakimoto 1999; Stoelinga et al. 2000). However, in cases where the cold front proceeds aloft after overtaking a lee trough, and an extensive squall line develops and remains collocated with the CFA, the dynamical role of the CFA in maintaining the squall line, rather than just initiating it, has not been well established. One possibility is that the Pacific cold front is only important for initiating the linearly oriented convection when it overtakes the lee trough, and that its subsequent collocation with the squall line is simply due to similar speeds of the squall line and CFA. At the other extreme of possibilities is the idea that the CFA provides a lifting mechanism that dominates that of the cold pool, and therefore the CFA determines the speed and orientation of the squall line. Complicating the resolution of these two scenarios is the fact that such squall lines are typically collocated both with a CFA and with their own self-generated cold pool.

To address the question of the relative importance of the lifting associated with a cold pool and that associated with a CFA, we undertook a numerical modeling study of the 19–20 June 1979 case in which both a CFA squall line and cold pool were present. The observed relationship between the location of the squall line and the 700-hPa thermal pattern at two times is shown in Fig. 2. At 1200 UTC 19 June 1979 (Fig. 2a), the location of the surface trough relative to the 700-hPa baroclinic zone indicates that the cold frontal zone was nearly vertical at this time. However, the same fields from 1200 UTC 20 June 1979 (Fig. 2b) show the 700-hPa baroclinic zone well east of the surface trough, indicating a forward-tilted frontal structure and a CFA. In addition, the squall line is in advance of the surface trough and collocated with the 700-hPa baroclinic zone.

The 19–20 June 1979 squall line was numerically simulated with a mesoscale model, using two different strategies. The first strategy was to simulate the entire squall line and its synoptic-scale environment (fronts, troughs, and parent low pressure center) with an 18-km horizontal grid utilizing both a cumulus parameterization scheme (CPS) and an explicit bulk microphysical scheme. This strategy is relevant to current operational forecasting, which uses a similar modeling configuration. The second strategy was to simulate part of the squall line and its immediate environment with explicit bulk microphysics only, on a 3.6-km horizontal grid capable of resolving convective motions. The latter strategy was used to confirm the results of the first strategy, insofar as demonstrating that those results were not an artifact of the coarser model resolution or use of a CPS.

In both model configurations (which will be referred to hereafter as coarse and high resolution), sensitivity experiments were performed in which the physical processes responsible for the formation of the cold pool outflow (primarily evaporative cooling) were withheld from the simulations. As described below, a comparison of the control and “no cooling” simulations leads to a better understanding of the relative importance of the CFA versus cold-pool lifting in the maintenance of the squall line.

2. Mesoscale model configuration

The model used in this study is the fifth-generation Pennsylvania State University–National Center for Atmospheric Research (PSU–NCAR) Mesoscale Model (MM5). The MM5 is a sigma-coordinate, primitive-equation model originally developed by Anthes and Warner (1978), with more recent upgrades described by Dudhia (1993) and Grell et al. (1994). Key physical parameterizations that were chosen for this study were an explicit mixed-phase bulk microphysical scheme similar to that described by Reisner et al. (1998), with cloud water, cloud ice, rain, snow, and graupel as prognostic variables; the Kain and Fritsch (1993) CPS; and a high-resolution planetary boundary layer scheme (Blackadar 1979; Zhang and Anthes 1982). All domains were run with 27 vertical sigma layers, with vertical resolution ranging from Δσ = 0.01 (Δp ≅ 9 hPa) near the surface to Δσ = 0.50 (Δp ≅ 43 hPa) in the middle and upper troposphere.

For the coarse-grid modeling strategy, a model grid with 18-km horizontal spacing was employed, covering roughly the central two-thirds of the contiguous United States, as well as southernmost Canada and northernmost Mexico. The CPS was used in combination with the explicit bulk microphysical scheme on this domain. An outer domain with 54-km horizontal spacing, covering much of North America and parts of the eastern Pacific and the western Atlantic Oceans, was also run simultaneously with the 18-km domain to generate a two-way interactive boundary condition for the 18-km domain. Results from the 54-km domain will not be shown here, since the squall line and all relevant synoptic-scale features were entirely contained within the 18-km domain. Initial conditions for both the 18- and 54-km domains, as well as the 12-hourly boundary conditions for the 54-km domain, were generated by using a coarse-resolution gridded analysis as a first guess and refining it with all available surface and upper-air observations. Both domains were initialized at 0000 UTC 19 June 1979, and model integrations were performed for a period of 36 h.

For the high-resolution modeling strategy, a model grid with 3.6-km horizontal spacing was employed, covering the central Great Plains region. This resolution was chosen in light of the suggestion of Weisman et al. (1997) that approximately 4-km grid spacing is the minimum required to adequately resolve organized convective storms on the grid scale. The size of the model domain was made as large as the memory restrictions of the computing facility used (the NCAR Cray J924se) would allow, and was positioned to capture the initiation region of the squall line and the southern two-thirds of the 1200+ km length of the mature squall line. No CPS was used in the high-resolution simulation; instead, all cloud and precipitation processes were treated by the explicit bulk microphysical scheme. The high-resolution simulation was driven with initial and boundary conditions interpolated from the 18-km model run, starting at hour 18 (1800 UTC 19 June 1979) and run for a period of 12 h.

For both the coarse- and high-resolution model configurations, a “control simulation” and a “sensitivity test” simulation was run. In the sensitivity test simulation, the formation of a precipitation-induced cold pool was prevented in order to examine the importance of cold-pool lifting in the maintenance of the CFA squall line. In the coarse simulation, cold-pool formation was prevented by shutting off cooling produced by the CPS and diabatic cooling in the explicit bulk microphysical scheme caused by melting or evaporation of precipitation. Since the squall-line precipitation was developed mostly by the CPS in the coarse simulation, the former adjustment was significantly more important. In the high-resolution simulation, a CPS was not employed, so only diabatic cooling due to melting or evaporation of precipitation in the explicit scheme was shut off. However, with the larger precipitation mixing ratios produced in the high-resolution run, precipitation water loading may also have contributed to convective downdrafts and outflow, so precipitation water loading was also shut off in the high-resolution sensitivity test. For conciseness, we will refer to these sensitivity tests as the “no cooling” simulations. For both the coarse- and high-resolution experiments, the no-cooling simulations were initiated at hour 18 (1800 UTC 19 June 1979), which was about 1 h before the initiation of the squall line in the coarse-grid control simulation.

It should be noted that the removal of precipitation water loading also would seem to enhance the maintenance of the squall line by removing an impedance to the convective updraft. While this is true, it does not significantly counteract the intent of the experiment (which was to hinder the self-maintenance mechanism of the squall line), for two reasons. First, the effect of removal of precipitation water loading on the convective updraft can be thought of as acting to enhance the buoyancy of parcels by an amount gqL, where g is acceleration due to gravity and qL the precipitation hydrometeor mixing ratio. Integrating this quantity through the depth of the convective column in the simulated squall line yields an enhancement of CAPE of ∼230 J kg−1 in the sensitivity experiment. Although this does enhance convective vigor, it represents only a ∼10% enhancement of CAPE for low-level inflow air parcels. Furthermore, an examination of precipitation mixing ratios showed that the buoyancy enhancement due to removal of precipitation water loading occurred primarily between 4 and 12 km above ground, well removed from the near-surface region where the “help” is required for the initial lift of parcels.

3. Comparison of simulated and observed synoptic-scale features in the 19–20 June 1979 cyclone

Figures 3a–c shows the sea level pressure, surface wind, temperature, and dewpoint fields at 0600 UTC 20 June 1979 derived from the NWS surface map shown in Fig. 1. Figures 3a–c also show the NWS surface frontal analysis and radar echoes from the NWS National Radar Summary. The position of the warm front is well defined in all of the fields, but the location of the surface cold front is less certain. The location of the cold front shown by the NWS does not lie in the sea level pressure trough, or completely along the dryline (i.e., the strong gradient of surface dewpoint), or along the strongest surface temperature gradient. A strong temperature gradient associated with the cold pool from the squall line had advanced eastward into Iowa and Missouri, but the dryline remained within the sea level pressure trough, well behind the squall line, and correlated well with the wind field. In fact, the strongest signature of any significant surface boundary is in the dewpoint gradient associated with the dryline, which was reinforced by the confluence in the surface pressure trough.

Figures 3d–f shows the same fields as Figs. 3a–c but for the 30-h control coarse-grid simulation of the MM5 model valid at 0600 UTC 20 June 1979. The shaded region in Figs. 3d–f comprises those areas where the model produced greater than 0.10 mm in accumulated precipitation (total convective and explicit) over the previous hour. The modeled temperature and dewpoint fields are very similar to the observed fields. Indeed, the position of the enhanced dewpoint gradient in the control simulation is in almost exactly the same location as the corresponding gradient in the actual cyclone. The wind fields are also very similar, although the low pressure center in the control simulation is 4 hPa too low and displaced 250 km northward of its actual position. The position of the surface warm front in the control simulation, which is marked in Figs. 3d–f in the traditional way, is readily discerned from the wind, pressure, temperature, and dewpoint fields. The location of the surface warm front is almost exactly the same as that of the warm front shown in Figs. 3a–c.

The lee trough is shown as a dashed line in Figs. 3d–f. North of approximately the Texas–Oklahoma border, the lee trough has already been overtaken by the Pacific cold front, with the leading edge of the Pacific cold front advancing aloft as a CFA over the stable layer east of the lee trough. The farthest forward advance of the Pacific cold-frontal zone aloft is indicated by the CFA symbol (open arrow heads) in Figs. 3d–f. South of approximately the Texas–Oklahoma border, the Pacific cold front has not overtaken the lee trough, and it is indicated as a surface cold front. Along much of its southern extent, the lee trough has the moisture characteristics of a dryline, so the dashed line could be replaced by the symbols for a dryline. In Fig. 3d there is a pressure trough and cyclonic wind shift over the Kansas–Nebraska border that is approximately 200 km west of the trough. This is a secondary trough, behind the primary trough shown in Fig. 1. Although the pressure signal of the primary trough is not evident in that particular location at this time, it is marked as a trough for temporal continuity, and to be consistent with the well-established dewpoint gradient and cyclonic wind shift at its location. The primary trough can be seen more clearly 3 h earlier in Figs. 4a and 5a.

The control simulation shows a precipitation band in essentially the same location (relative to the surface features and low pressure center) as the observed squall line. The rainfall shown in Figs. 3d–f was produced primarily by the CPS, although at isolated locations along the squall line the explicit bulk microphysical scheme produced a significant fraction of the precipitation. We refer to this simulated precipitation feature as a squall line because it is a convective precipitation band (albeit primarily parameterized convection) that matches the observed squall line in terms of location, orientation, and speed of movement normal to the line (15 m s−1 for the simulated line versus 13 m s−1 for the observed).

4. Comparisons of the model control simulations with the “no cooling” simulations

a. Coarse-grid horizontal fields

Figures 4–6 show a smaller part of the coarse-grid model domain that contains the model-produced squall line. In these six panels we compare fields of sea level pressure, dewpoint, potential temperature (θ), and precipitation from the model control simulation with the same fields for the coarse-grid no-cooling simulation. All the panels are for the coarse-grid model simulations at 27 h valid for 0300 UTC 20 June 1979. There are several important differences between these two model simulations:

  • The central sea level pressure of the low pressure center in the no-cooling simulation is lower than that of the control simulation, and the pressure gradients around the low pressure center are greater (Fig. 4).
  • The position of the enhanced dewpoint gradient is farther east in the no-cooling simulation and closer to the position of the squall line (Fig. 5).
  • The no-cooling simulation provides no indication of a cold pool starting at the leading edge of the squall line, although the cold pool can be seen clearly in the control simulation (Fig. 6).
  • The leading edge of the squall line in the control simulation is no more than 50 km ahead of the position of the squall line in the no-cooling simulation (Fig. 6).
  • The squall line is wider and not as well defined in the no-cooling simulation, but it contains regions of higher precipitation rates than in the control simulation (Fig. 6).

When the sea level pressure distribution for the control simulation is subtracted from that of the no-cooling simulation (Fig. 7a), most of the difference is centered in regions where precipitation was present in both model simulations. These regions are where cooling due to evaporation of precipitation falling into the lower troposphere could take place in the control simulation, but not in the no-cooling simulation. Consistent with this explanation is the difference between the thickness of the 950–800-hPa layer in the no-cooling simulation and that in the control simulation (Fig. 7b). Figure 7b shows lower-tropospheric cooling (lower thickness) due to evaporation over approximately the same regions where precipitation occurred and where there were significant pressure differences.

To differentiate between the pressure effects of diabatic cooling and those of the atmosphere above the cold pool, we examined the pressure field produced by specific layers of the atmosphere for both model simulations. Figure 8a shows the contribution to the surface pressure from the layer of the atmosphere that is located approximately between the surface and the mean height of the CFA (i.e., the mean height of the farthest forward extent of the Pacific cold-frontal zone aloft), or between 0.4 and 1.6 km above MSL. Figure 8b shows the contribution to the surface pressure by the layer of the atmosphere that is approximately between the mean height of the CFA and the tropopause, or between 1.6 and 9.0 km above MSL. There are striking differences between these two pressure patterns. The layer of the atmosphere closest to the ground produces a region of high pressure coincident with the cold pool, which starts at the leading edge of the squall line. The ridge of highest pressure in the cold pool is marked by the heavy dashed–dotted line in Fig. 8a. The lower layer also produces a trough in the pressure pattern, which is located behind the region of high pressure (Fig. 8a) and to the rear of the surface cold pool (Fig. 6a). This trough is marked in Fig. 8a by a heavy, dashed–double-dotted line. This trough is coincident with the position of the surface trough and enhanced surface dewpoint gradient for the control simulation shown in Fig. 5a. In contrast, neither the high pressure region nor the dewpoint gradient–trough coincidence is seen in the pressure pattern produced by the upper layer (Fig. 8b). Instead, the pattern for the upper layer is dominated by a pressure trough that is coincident with the squall line, and by a region of rapidly rising pressure to the west of the squall line. This pressure pattern is similar to that of a surface cold front; it simply reflects the fact that the Pacific cold front moved aloft in advance of the surface trough as a CFA, and that the CFA is coincident with the squall line. The moving pressure pattern in Fig. 8b could add to the forcing of convergence and vertical velocity in the lower troposphere in the location of the squall line, as described by Locatelli et al. (1997).

Shown in Fig. 9 are the corresponding pressure patterns from the no-cooling simulation, where the lower layer is taken to be between 0.4 and 1.2 km and the upper layer between 1.2 and 9.0 km. The boundary between the upper and lower layers is 0.4 km lower in the no-cooling simulation than in the control, because the CFA was, on average, 0.4 km lower in the no-cooling simulation. The pressure pattern for the lower layer of the no-cooling simulation (Fig. 9a) does not show a high pressure region associated with the squall line, as expected since diabatic cooling was not present in the no-cooling simulation. However, it does show a trough, slightly to the rear of the squall-line precipitation that is coincident with the location of the weak surface trough and the enhanced dewpoint gradient in the no-cooling simulation (Fig. 5b). The trough that is evident in the contribution to the surface pressure by the upper layer (Fig. 9b) is similar and nearly collocated with the trough produced by the upper layer in the control simulation. This indicates that in both the control and no-cooling simulations the CFA is in virtually the same location at model hour 27.

b. High-resolution results

As discussed in section 2, the high-resolution simulation was initialized at 1800 UTC 19 June 1979, 18 h into the coarse-grid simulation. This time was 1 h prior to initiation of the squall line in the coarse-grid control simulation. As will be shown below, the high-resolution simulation produced the essential features of the squall line seen in the observations and in the coarse-grid simulation.

Approximately 2.5 h after the start of the high-resolution simulation (2030 UTC 19 June), and as the Pacific cold front occluded with the lee trough and dryline, small areas of convection appeared along the simulated dryline. Just 2 h later, the convection expanded into an extensive convective line 800 km in length (Fig. 10a). About this time (0000 UTC 20 June 1979), a sounding from Monette, Missouri, approximately 220 km ahead of the southern end of the squall line, showed “most unstable” convective available potential energy (MUCAPE) values of 2600 J kg−1. The model atmosphere at the same time, latitude, and distance ahead of the squall line (the model squall line was somewhat slower than observed) showed MUCAPE of 1700 J kg−1, although values closer to the squall line ranged from 2000 to 3000 J kg−1. This grid-resolved squall line moved eastward and, at its most mature stage (0100–0200 UTC 20 June), it contained updrafts of up to 15 m s−1 and downdrafts of up to 3 m s−1. At hour 27 (Fig. 10b; 0300 UTC 20 June, the time at which most of the coarse-grid results were shown), the squall line occupied a position similar to that seen in the coarse-grid simulation.

Several similarities and differences between the high-resolution (Fig. 10b) and the coarse-grid simulations of the squall line are noteworthy. The pressure, dewpoint, and wind shift patterns associated with the surface trough (marked with a dashed line in Fig. 10b) indicate that the surface trough had advanced farther eastward in the high-resolution simulation than in the coarse-grid simulation, whereas, the squall line was essentially in the same place. Therefore, the advance of the squall line ahead of the surface trough, while clearly evident in the high-resolution simulation, is less than in the coarse-grid simulation. This is likely due to the fact that the CPS produced a stronger cold pool in the coarse-grid simulation than the explicit bulk microphysical scheme produced in the high-resolution simulation. Another difference is that the squall line in the high-resolution simulation extended only as far south as the Kansas–Oklahoma border, and developed a ∼100 km gap near the Kansas–Missouri border, whereas, the coarse-grid simulation maintained a continuous squall line well into Oklahoma (Fig. 4a). However, aside from that difference, the high-resolution simulation developed the squall line in essentially the same location as the coarse-grid simulation, and maintained it throughout the 12-h high-resolution model run.

In the high-resolution, no-cooling simulation, the initial squall line formed identically to that in the high-resolution control simulation, but the absence of diabatic cooling and precipitation loading quickly led to a different squall-line structure. A comparison of squall-line updrafts and downdrafts in the control and no-cooling simulations at its most vigorous stage (0100 UTC 20 June) shows that the control squall line (Fig. 11a) was characterized by intermixed cores of updrafts and downdrafts, with a slight tendency for downdrafts to be located rearward of updrafts. In contrast, the squall line in the no-cooling simulation (Fig. 11b) was more two-dimensional, with more continuous line segments of updrafts and virtually no downdrafts (aside from a few bands of weak subsidence that were likely associated with gravity waves excited by the vigorous convective activity). The diabatic cooling and convective downdrafts in the control squall line led to the development of a surface cold pool, as seen when the surface virtual potential temperature field from the no-cooling simulation is subtracted from that of the control simulation (Fig. 11c). Immediately behind the precipitation band in the control simulation is a cold perturbation band of mostly −1 to −3 K, which represents the cold pool. The cold pool spreads out in both the forward and rearward directions. In CFA-type squall lines, there is typically a small region of warm air that is confined between the surface trough and the squall line. The rearward spreading cold pool retards the forward advance of this region of warm air and of the trough itself. This results in the slight warm perturbation (about 1 K) behind the cold pool in Fig. 11c. Note that the high-resolution simulation produced a weaker surface cold pool (−1 to −3 K) than the coarse-grid–CPS simulation (−4 to −6 K, not shown), or the observed squall line (also roughly −4 to −6 K, as can be seen by examining the thermal gradient at the leading edge of the cold pool in Fig. 3b). In the high-resolution simulation, the weaker cold pool would likely decrease the retarding effect of the cold pool on the movement of the trailing surface trough, and might explain why there was greater separation between the trough and squall line in the observations and coarse-grid simulation than in the high-resolution simulation.

In spite of the absence of a cold pool in the high-resolution, no-cooling simulation, the squall line in that simulation mimicked the control squall line in terms of duration, movement, and spatial extent, verifying similar results from the coarse-grid simulation. This can be seen by comparing the horizontal coverage of precipitation for the control squall line (Fig. 10b) and the squall line without a cold pool (Fig. 12a). As was seen in the coarse-grid simulations, the trailing trough (the wind shift and dewpoint gradient marked with a dashed line in Figs. 10b and 12a) is much farther forward in the no-cooling simulation, almost coincident with the squall line at this time. This confirms that the cold convective outflow contributes to the separation between the trough and the squall line. However, it is not the only process that leads to that separation, since even in the no-cooling simulation, the squall line can be seen to separate from the surface trough in northern Missouri and Iowa 3 h later (Fig. 12b).

Several insights into the structure of the squall line and its relationship to the synoptic-scale structure are gained by examining a vertical cross section through the grid-resolved squall line at hour 27 (Fig. 13, 0300 UTC 20 June, the time at which most of the coarse-grid results were shown). The location of the cross section is shown in Figs. 10b and 12a. Due to the cellularity of the convection along the simulated squall line (cf. Fig. 11a), results shown in the cross section are averaged normal to the cross section over a distance covering three to four updraft and downdraft cells (we used 65 km into and out of the page), so that the two-dimensional aspects of the squall line can be elucidated.

The synoptic-scale thermal structure in the cross section through the squall line at 0300 UTC 20 June is quite clear from the equivalent potential temperature (θe) and virtual potential temperature (θυ) fields (Fig. 13a). The θυ field is shown here instead of θ because, in the presence of very strong moisture gradients, virtual temperature effects are significant at this scale. The Pacific cold air mass is identified by a region of low θe and θυ in the left half of the cross section. The forward boundary of this air mass (i.e., the Pacific cold frontal surface) is subjectively identified by the heavy dashed line in Fig. 13a. As is often the case over the Great Plains, the air mass into which the Pacific cold front is moving is characterized by strong moisture stratification (dry air overlying moist air), resulting in a potentially unstable configuration of low-θe air above high-θe air. Surface-based CAPE values (in the model simulation) were in the range of 1200–2500 J kg−1.

The squall line itself can be seen in both the cloud/precipitation outline and in the airflow relative to the squall line in the plane of the cross section of Fig. 13a. The speed of the simulated squall line (calculated by measuring the movement of the line of hourly precipitation, normal to its orientation, for a centered period of 6 h) was 15 m s−1, compared to the observed speed of 13 m s−1. Moist low-level air flows toward the squall line from the right. Most of this inflow enters the narrow convective updraft, but the air nearest the surface does not. Rather, it decelerates and rises only slightly as it continues to move toward the rear of the squall line. The updraft itself rises vigorously in a vertical column up to 200 hPa. There is evidence of a net downdraft immediately behind the updraft and below 500 hPa. Below 600 hPa, the entire squall line (updrafts, downdrafts, cloud, and precipitation) is confined to a narrow column only 65 km wide; there is no trailing stratiform region. Above 600 hPa, the cloud and precipitation outline, as well as the airflow vectors, show that the upper-level outflow and anvil cloud are almost entirely forward of the squall line.

The cold pool is not particularly strong at the time and location shown in Fig. 13a, but it can be identified, and its relationship to the surface trough can be seen, by following the 306-K θυ contour in this cross section. In the warm air to the right of the squall line, the 306-K contour is relatively flat and lies just beneath 900 hPa. Beneath the squall line, the 306-K contour rises slightly, indicating the leading edge of the cold pool. To the left of the squall line, the 306-K contour approximately caps the dome of the weak cold pool back to the surface trough, where it dips down in the low-level warm pocket above the remnant lee trough (marked as “wind shift, temperature maximum, and enhanced dewpoint gradient” in Fig. 13a).

When viewed in the same cross section, the squall line produced by the high-resolution no-cooling simulation (Fig. 13b) differs from that of the control (Fig. 13a) in two ways. Of less importance is the fact that, with precipitation loading shut off everywhere, the updraft is more vigorous in the no-cooling simulation, resulting in a broader squall line, deeper penetration of the cloud top into the stratosphere, and a more extensive anvil in both the forward and rearward directions. The more important difference is the lack of development of downdrafts and a cold pool. The lack of a cold pool can be seen by following the 306-K θυ contour in Fig. 13b. Note that the 306-K contour does not rise up beneath the squall line. Instead, it dips down almost immediately behind the squall line, reflecting the fact discussed previously that the surface trough in the no-cooling simulation remains much closer to the squall line because its eastward progress is not retarded by the rearward spreading cold outflow of the squall line. The Pacific cold frontal boundary, which intersects the ground at the location of the trough, is only slightly tipped forward, perhaps better described in this cross section as nearly vertical. However, at a later time and in a cross section somewhat farther north (Fig. 13c), the Pacific cold frontal boundary tips forward significantly, even without a squall-line cold pool to hinder the eastward movement of the surface trough. Reasons for this will be discussed in the next section.

5. Discussion

a. Structure and evolution of the squall line

The squall line produced in the high-resolution control simulation had several features that are characteristic of many squall lines associated with CFAs, and which differ from the “leading line–trailing stratiform” paradigm for midlatitude squall lines discussed by Houze et al. (1989). The primarily forward outflow anvil in the upper troposphere and lack of any trailing stratiform region are seen clearly in Fig. 13a, and agree with what was observed in satellite imagery and the radar summary. Also, the high-resolution simulation did not show any significant mesoscale descending rear-to-front inflow, a feature that is often observed in leading line–trailing stratiform type systems. Previous case studies have shown that CFA-related squall lines can be of either type—forward anvil only and no trailing stratiform region (Locatelli et al. 1998) or rearward anvil and trailing stratiform region (Rose et al. 2002). From our inspections of the radar patterns associated with these systems over the past 10 years, the former type seems to be more common for CFA squall lines. It is likely that with the squall line developing in the immediate presence of a midlevel baroclinic zone, which typically has some southward-directed thermal gradient in addition to a strong eastward-directed gradient, there is often strong upper-level forward shear that precludes the development of a rearward anvil or trailing stratiform region.

Although it is not shown here, the simulated squall line did undergo a life cycle of initiation, maturity, and decay, and, as in the case of many observed and simulated squall lines, the strength and extent of the cold pool changed through the life cycle of the simulated squall line. However, an examination of the squall line at several times throughout its life cycle did not show any significant structural evolution. The strong low-level inflow from the east, the deep, narrow, vertical convective plume, the forward anvil, and the lack of a trailing stratiform region persisted throughout the life cycle of the simulated squall line.

b. Lifting mechanisms for the squall line

As mentioned previously, the coarse-grid simulation produced a stronger cold pool, which, in fact, agreed more closely with the strength of the observed cold pool. However, in both simulations, the cold pool was not strong enough to counteract the low-level shear and have its leading edge act as a material barrier to the inflow. All of the inflow decelerated when it encountered the cold pool, but some of it continued to pass through the precipitation curtain near the surface. This type of non-density-current-like cold pool was termed a gravity wave without stagnation by Crook and Moncrieff (1988), and has been seen in both numerically simulated squall lines (Dudhia et al. 1987; Crook and Moncrieff 1988) and in at least one observed squall line (Fankhauser et al. 1992). Although this type of cold pool is capable of lifting inflow air and initiating convection, it differs somewhat from the paradigm of the squall-line cold pool acting as a density current.

In both the coarse-grid and high-resolution no-cooling simulations of the 19–20 June 1979 squall line, the key difference from the control simulation was the lack of formation of a cold pool. However, the key features that did not change were the development of an extensive, long-lived squall line, and its collocation with a CFA. After being active for 6 h and traveling 270 km, the high-resolution squall line without a cold pool and its attendant CFA were only 20 km west of the squall line and CFA with a cold pool (Figs. 13a,b). Therefore, we conclude that the CFA was more important in determining the shape, orientation, speed, and longevity of the squall line than was the interaction of the cold pool with the low-level shear.

One could postulate that if the cold pool had been strong enough it might have formed a density-current-like structure that could have moved faster than the CFA and caused the squall line to propagate on its own away from the CFA and into the warm sector. To test this possibility, a coarse-grid model simulation (not shown) was carried out with the rate of evaporational cooling increased by a factor of 4 over that of the control simulation. In this case, the squall line moved ahead of the CFA and remained at the leading edge of an intense cold pool. This result suggests that, depending on the strength of the cold pool and the speed of the CFA, a CFA squall line might move away from its associated CFA and become self-propagating. However, this did not occur for the squall line studied in this paper.

If a cold pool did not initiate new convection and maintain the squall line in the no-cooling simulations, what lifting mechanism did? One possibility we examined, which is unrelated to the CFA, is that supercell dynamics may have been active within the squall line. Although a cold pool is often active and contributes to lifting of environmental inflow air in supercell thunderstorms, it is not required because lifting is also forced by a vertical pressure gradient that is dynamically induced by a midlevel mesoscale vortex within the updraft (Rotunno and Klemp 1982, 1985). Thus, supercells can be identified by a strong correlation between updrafts and vertical vorticity (Weisman et al. 1988). We examined the cellular structure of the high-resolution squall line in both the control and no-cooling simulations. Although there was rotation in some of the stronger updrafts, there was generally not a strong correspondence between the vertical velocity and vorticity patterns at midlevels along the entire squall line. This suggests that supercell dynamics were not the primary forcing mechanism for the convection.

The clear connection between the squall line and the CFA suggests that the CFA itself provided lifting that was able to maintain convection, playing the role of the external lifting mechanism discussed by Crook and Moncrieff (1988). Locatelli et al. (1997) postulated that a CFA induces a moving pressure pattern at the surface that may provide sufficient lifting to continually initiate convection. There is clear evidence that such lifting was active in the no-cooling simulation. This can best be illustrated by examining perturbation fields of pressure and virtual temperature in a cross section through the squall line at 0300 UTC 20 June 1979, as shown in Fig. 14. Several cross sections at various times were examined and found to show essentially the same process, but the one shown here illustrates it robustly. The perturbation fields of pressure and virtual temperature were calculated by subtracting from the full model fields a standard temperature profile, and corresponding hydrostatic pressure profile, that vary only with height. In the perturbation virtual temperature field, a strong baroclinic zone can be seen emanating from near the top-left corner of the cross section and extending down and to the right, toward the squall line, which can clearly be seen as a narrow column of strong upward motion at the 215-km position along the cross section in Fig. 14. The forwardmost “nose” of the baroclinic zone is the CFA, confined between ∼550 and 800 hPa. Above the CFA (around the 500-hPa level), an examination of the pressure perturbation field out to a few hundred kilometers on either side of the squall line shows that the along-cross-section component of the horizontal pressure gradient is fairly uniform and continuous at this level. However, if one moves vertically downward at the squall-line location, there is a strong temperature gradient to the left of the squall line (i.e., the CFA baroclinic zone), but very little temperature gradient to the right of the squall line. This results in a more rapid increase in pressure with descending height on the left side of the CFA than on the right, which can be seen as a kink in the pressure perturbation contours that increases with descent, down to the 800-hPa level. From this level down to the surface, the temperature gradient is weak on both sides of the squall line, so that the pressure pattern at 850 hPa is preserved down to the surface. Therefore, as was shown on a larger scale in the coarse-grid simulation, the sharp change in pressure gradient at the surface is due to the sharp structure of the CFA between 550 and 800 hPa, rather than to any thermal structure between the surface and 850 hPa. Furthermore, we compared the total pressure field with the hydrostatic pressure field and found that the nonhydrostatic pressure perturbation associated with the strong vertical acceleration in the updraft did not significantly contribute to the pressure perturbation pattern seen in Fig. 14.

As discussed by Locatelli et al. (1997), the low-level inflow ahead of the pressure perturbation induced by the CFA is typically in geostrophic–frictional balance with an essentially uniform pressure pattern. This is the case here—note in Fig. 14 the fairly uniform pressure perturbation lines to the right of 220 km and below 700 hPa, and also the fairly uniform inflow vectors in that region. When that low-level inflow encounters the sudden change in pressure gradient beneath the CFA, it decelerates rapidly throughout the lowest 1 km, resulting in a column of large horizontal convergence, and strong vertical velocity within a short distance above the surface. This vertical velocity is sufficient to lift inflow parcels to their level of free convection and maintain the squall line. Therefore, our no-cooling simulation demonstrates that the CFA-induced surface convergence discussed by Locatelli et al. (1997) can act as the externally applied forcing mechanism for maintaining a squall line in the paradigm presented by Crook and Moncrieff (1988).

c. Relationship of the squall line to the forward-tilted frontal structure

Figure 15 shows schematic vertical cross sections that illustrate some of the important differences between a genuinely self-propagating squall line that might develop in a true warm sector of a classical cyclone, and a CFA squall line of the type simulated here. The first panel shows a vertical cross section through a warm-sector squall line, or, more specifically, a squall line that is ahead of a backward-tilted Pacific cold front in the central United States. The structure and airflow of the squall line is highly oversimplified, but is intended to represent the structure of self-propagating midlatitude squall lines as embodied in the observational studies of Newton (1950) and Houze et al. (1989), and in idealized modeling studies in horizontally uniform environments such as those of Rotunno et al. (1988), Weisman et al. (1988), and Weisman (1992).

The large solid arrows in Fig. 15a show the airflow relative to the Pacific cold front, and the dashed arrows the airflow relative to the warm-sector squall line. The cold front and its associated circulation are far removed from the mesoscale structure and kinematics of the squall line itself. Within the squall line, the leading edge of the cold pool acts analogously to a density current, where the air approaching the cold pool from ahead of the warm-sector squall line does not penetrate the cold pool. In some warm-sector squall lines the air may flow in a relative sense from ahead of the squall line through the cold pool from front to rear (Crook and Moncrieff 1988; Fankhauser et al. 1992), but the former type is more common. Regardless, the cold pool is an essential part of the self-propagating process, and the movement of the Pacific cold front is not directly related to the movement of the squall line. The cloud outline, shown as a thin black line, depicts a primarily rearward extension of the upper-level outflow and anvil, and a trailing stratiform rain region.

In the schematic cross section through a CFA squall line depicted in Fig. 15b, the Pacific cold front is tipped forward in the lower troposphere ahead of the surface trough, and the squall line is coincident with the farthest forward extension of the Pacific cold frontal zone aloft (i.e., coincident with the CFA). The dashed arrows in Fig. 15b show the airflow relative to the movement of the CFA or the CFA squall line. Cold postfrontal Pacific air moves toward (and sinks relative to) the motion of the CFA and the CFA squall line. This air mass, which is synoptic in scale and is typically characterized by low equivalent potential temperature, may feed into a descending rear inflow of the squall line. Air ahead of the CFA moves in a relative sense toward the CFA and the CFA squall line. Low-level lifting of inflow air beneath the CFA is forced both by the surface pressure pattern induced by the CFA and by the cold pool. However, as shown in this paper, the cold pool was not essential for the maintenance and forward movement of the CFA squall line of 19–20 June 1979. The cloud outline depicts a primarily forward extension of the upper-level outflow and anvil, and no trailing stratiform rain region.

The cause of the forward tilt of the Pacific cold front was found to be due, in part at least, to the rearward spreading of the cold convective outflow, which tended to inhibit the eastward progression of the surface trough and its associated potential temperature maximum, as well as the leading portion of the Pacific cold air mass at the surface that lay immediately to the west of the surface trough. In effect, the rearward spreading convective cold outflow formed a pool of surface air that was colder and more stable than that at the front of the Pacific cold air mass; this caused the Pacific cold front to advance aloft in a warm-occluded-like manner. This process was also seen in a mesoscale model simulation of the superoutbreak storm of 3–4 April 1974 (Locatelli et al. 2002b). However, the formation of the cold pool was not the only process that led to the forward tilt of the Pacific cold front, because even without a cold pool, the Pacific cold front still developed a forward tilt that ranged from slight (Fig. 13b) to significant (Fig. 13c), a result that was also found in the Locatelli et al. (2002b) study. The tendency toward a forward tilt is likely due to the inherent static stability contrast between the two air masses, as discussed by Stoelinga et al. (2002). In all three panels of Fig. 13 there is an important synoptic-scale transition in dry static stability that occurs at approximately 650 hPa in both the prefrontal and postfrontal air masses. Below 650 hPa, the postfrontal air is characterized by a deep well-mixed layer (excluding the near-surface nocturnal inversion that is forming well behind the front in Fig. 13c), and so is statically less stable than the prefrontal air below 650 hPa, which includes a capping stable layer between 700 and 850 hPa. This static stability contrast is consistent with a forward-tilting cold front below 650 hPa. Above 650 hPa, the upper part of the Pacific cold-frontal baroclinic zone is characterized by stronger static stability than the prefrontal air at that level, which is consistent with a conventional rearward-tilting frontal surface. This static stability contrast is common when a Pacific cold air mass traverses the Rockies and encounters the low-level capping stable layer east of the lee trough. Therefore, there is a natural tendency for Pacific cold fronts to tilt forward in these situations, even without the aid of a convectively generated cold pool.

d. Cold fronts aloft and surface analysis

Finally, we would like to make some suggestions concerning the identification and location of CFAs. The NWS analyzed the 19–20 June 1979 cyclone in a manner consistent with the Norwegian cyclone model, that is, by marking the surface trough as a cold front and, by implication, classifying the convective line as a “warm-sector squall line” unassociated with any frontal features. This study has shown that the squall line was coincident with a CFA and was, therefore, not a conventional warm-sector squall line. This type of frontal system is often operationally analyzed with a surface cold front in the trough position, conveying an incorrect picture of the relationship of the squall line to the attendant frontal structure.

When a Rocky Mountain lee trough and/or dryline are overtaken by a Pacific cold front, a CFA may or may not form, depending on the relative stability of the air masses behind the Pacific cold front and ahead of the lee trough (Locatelli et al. 2002a; Stoelinga et al. 2002). As shown by Locatelli et al. (1997), Neiman et al. (1998), and Neiman and Wakimoto (1999), the pressure field can provide indications of a CFA. This is illustrated in the model-simulated sea level pressure pattern at hour 27, with the model precipitation field overlaid (Fig. 4a). The CFA is located at the leading edge of the precipitation band. Aside from the obvious and expected mesoscale pressure perturbations associated with the squall line itself, the synoptic-scale pressure pattern associated with the CFA is readily identifiable. On the synoptic scale, there is a strong eastward-oriented pressure gradient ahead of the squall line, whereas, behind the squall line the east–west component of the pressure gradient nearly vanishes. The fact that a synoptic-scale change in the orientation of isobars can be used to help determine from a surface map whether a CFA is located ahead of a surface trough was noted many years ago by Showalter and Fulks (1943) who wrote: “A simple rule to check the analysis of the warm sector is the following: if straight and parallel isobars cannot be drawn in the warm sector without ignoring the reported data, the main cold front should be advanced and called an upper cold front if need be.” It should be emphasized that this rule applies to a synoptic-scale isobaric analysis, not a mesoanalysis that will contain strong pressure perturbations associated with the squall line itself.

Other parameters on a surface map can provide indications of a CFA situated ahead of a surface trough. For example, in the 19–20 June 1979 cyclone the surface trough had a weak signal in temperature and pressure but a stronger signal in the dewpoint field (Fig. 3), indicating the possible presence of a CFA. This is not meant to imply that every dryline indicates the presence of a CFA. The important qualifiers are that if a dryline or lee trough has been overtaken by a Pacific cold front, and if the remaining surface trough retains the characteristics of a dryline (much stronger signature in dewpoint than in temperature), then a Pacific cold front has likely proceeded aloft to form a CFA. If, on the other hand, the surface trough retains the character of a surface cold front (with a significant temperature gradient), then this is an indication that the Pacific cold front has likely retained its standard backward tilt and its leading edge is at the surface.

We have found that the most useful method for identifying and locating a CFA is to examine fields of potential temperature, equivalent potential temperature, and winds produced by real-time mesoscale forecast models, similar to the coarse-grid MM5 model configuration used in the present study. These fields should be examined at a variety of lower-tropospheric pressure levels, as well as in vertical cross sections such as those shown in Fig. 10. Furthermore, in the present study, the successful high-resolution simulation of the squall line showed that the squall line produced by the coarse-grid model was not an artifact of the grid resolution or use of a CPS. This result improves confidence in the ability of a mesoscale model to capture not only the synoptic-scale structure that leads to the formation of CFA squall lines but also the timing and location of the squall line, even when the model is run at coarse resolution in conjunction with a CPS such as the Kain and Fritsch (1993) scheme. However, this improved confidence should be applied only to model-forecasted squall lines for which an associated CFA is clearly indicated. It is well known that coarse-grid models with a CPS are often unreliable for forecasting squall lines or other mesoscale convective systems when there is weak or nonexistent frontal forcing.

6. Conclusions

The cyclone that occurred in the upper midwest of the United States on 19–20 June 1979 was well reproduced by an MM5 simulation. The model also produced a squall line with location, orientation, and speed similar to that of the observed squall line, in both the control simulation with “full physics” (which developed a cold pool), and in a sensitivity test with diabatic cooling due to melting and evaporation of precipitation shut off (which prevented the formation of a cold pool). Furthermore, these results apply to both a coarse-grid (18 km) model configuration that utilized a cumulus parameterization scheme and a high-resolution (3.6 km) model configuration that did not. The actual cyclone and the simulated cyclone had a nonclassical, forward-tilted Pacific cold front, in which the squall line was coincident with the leading edge aloft of the Pacific cold-frontal zone (i.e., with the cold front aloft, or CFA) in advance of the surface trough. The simulated squall line exhibited characteristics that differed from the conventional leading line–trailing stratiform paradigm for midlatitude squall lines, namely, a consistently vertical convective updraft with primarily forward upper-level anvil, no trailing stratiform region, and a non-density-current-like leading edge to the cold pool. In the model simulations, the CFA provided sufficient lifting on its own to maintain the squall line and, thereby, determine its longevity, location, and movement. The CFA maintained the squall line by means of the pressure pattern it induced at the surface, which decelerated and lifted low-level inflow air. The cold pool, which was stronger in the observed case than in the high-resolution control simulation, might also have been sufficient on its own to maintain a similarly initiated squall line in a similar environment (but without a CFA). However, the cold pool was not a necessary component for maintaining the observed longevity, location, and movement of the squall line, as shown by the similarity between the control and no-cooling model simulations. The cold pool did contribute to the forward-tilted structure of the Pacific cold front by retarding the eastward movement of the surface trough at the rear flank of the cold pool. This effect increased the distance between the CFA and the surface trough, and thereby increased the forward tilt of the Pacific cold frontal surface.

Acknowledgments

This research was supported by Grant ATM-9106235 from the Mesoscale Dynamic Meteorology Program, Atmospheric Research Division, National Science Foundation. Computing facilities were provided by National Center for Atmospheric Research, which is supported by the National Science Foundation.

REFERENCES

  • Anthes, R. A., , and T. T. Warner, 1978: Development of hydrodynamical models suitable for air pollution and other mesometerological studies. Mon. Wea. Rev., 106 , 10451078.

    • Search Google Scholar
    • Export Citation
  • Blackadar, A. K., 1979: High resolution models of the planetary boundary layer. Advances in Environmental Science and Engineering, Vol. 1, No. 1, J. R. Pffafflin and E. N. Zeigler, Eds., Gordon and Breach, 50–85.

    • Search Google Scholar
    • Export Citation
  • Castle, J. A., , J. D. Locatelli, , J. E. Martin, , and P. V. Hobbs, 1996: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part IV: The evolution of a drytrough on 8–9 March 1992. Mon. Wea. Rev., 124 , 15911595.

    • Search Google Scholar
    • Export Citation
  • Crook, N. A., , and M. W. Moncrieff, 1988: The effects of large-scale convergence on the generation and maintenance of deep moist convection. J. Atmos. Sci., 45 , 36063624.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1993: A nonhydrostatic version of the Penn State–NCAR Mesoscale Model: Validation tests and simulation of an Atlantic cyclone and cold front. Mon. Wea. Rev., 121 , 14931513.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., , M. W. Moncrieff, , and D. W. K. So, 1987: Two dimensional dynamics of West African squall lines. Quart. J. Roy. Meteor. Soc., 113 , 121146.

    • Search Google Scholar
    • Export Citation
  • Fankhauser, J. C., , G. M. Barnes, , and M. A. LeMone, 1992: Structure of a midlatitude squall line formed in strong unidirectional shear. Mon. Wea. Rev., 120 , 237260.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., , J. Dudhia, , and D. R. Stauffer, 1994: A description of the fifth generation Penn State/NCAR Mesoscale Model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 121 pp. [Available from National Center for Atmospheric Research, Boulder, CO 80301-3000.].

    • Search Google Scholar
    • Export Citation
  • Hobbs, P. V., , J. D. Locatelli, , and J. E. Martin, 1990: Cold fronts aloft and the forecasting of precipitation and severe weather east of the Rocky Mountains. Wea. Forecasting, 5 , 613626.

    • Search Google Scholar
    • Export Citation
  • Hobbs, P. V., , J. D. Locatelli, , and J. E. Martin, 1996: A new conceptual model for cyclones generated in the lee of the Rocky Mountains. Bull. Amer. Meteor. Soc., 77 , 11691178.

    • Search Google Scholar
    • Export Citation
  • Houze Jr., R. A., , S. A. Rutledge, , M. I. Biggerstaff, , and B. F. Smull, 1989: Interpretation of Doppler weather radar displays of midlatitude mesoscale convective systems. Bull. Amer. Meteor. Soc., 70 , 608619.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., , and J. M. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain–Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor. Monogr., No. 46, Amer. Meteor. Soc., 165–170.

    • Search Google Scholar
    • Export Citation
  • Koch, S. E., , and J. McCarthy, 1982: The evolution of an Oklahoma dryline. Part II: Boundary-layer forcing of meso-convective systems. J. Atmos. Sci., 39 , 237257.

    • Search Google Scholar
    • Export Citation
  • Koch, S. E., , and L. M. Siedlarz, 1999: Mesoscale gravity waves and their environment in the central United States during STORM-FEST. Mon. Wea. Rev., 127 , 28542879.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , and P. V. Hobbs, 1995: A world record rainfall rate at Holt, Missouri: Was it due to cold frontogenesis aloft? Wea. Forecasting, 10 , 779785.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , J. E. Martin, , J. A. Castle, , and P. V. Hobbs, 1995: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part III: The development of a squall line associated with weak cold-frontogenesis aloft. Mon. Wea. Rev., 123 , 26412662.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , M. T. Stoelinga, , R. D. Schwartz, , and P. V. Hobbs, 1997: Surface convergence induced by cold fronts aloft and prefrontal surges. Mon. Wea. Rev., 125 , 28082820.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , M. T. Stoelinga, , and P. V. Hobbs, 1998: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part V: Thermodynamic and dual-Doppler radar analysis of a squall line associated with a cold front aloft. Mon. Wea. Rev., 126 , 860875.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , R. D. Schwartz, , M. T. Stoelinga, , and P. V. Hobbs, 2002a: Norwegian-type and cold front aloft–type cyclones east of the Rocky Mountains. Wea. Forecasting, 17 , 6682.

    • Search Google Scholar
    • Export Citation
  • Locatelli, J. D., , M. T. Stoelinga, , and P. V. Hobbs, 2002b: A new look at the super outbreak of tornadoes on 3–4 April 1974. Mon. Wea. Rev., 130 , 16331651.

    • Search Google Scholar
    • Export Citation
  • Maddox, R. A., 1980: Mesoscale convective complexes. Bull. Amer. Meteor. Soc., 61 , 13741387.

  • Martin, J. E., , J. D. Locatelli, , and P. V. Hobbs, 1990: Organization and structure of clouds and precipitation on the mid-Atlantic coast of the United States. Part III: The evolution of a middle-tropospheric cold front. Mon. Wea. Rev., 118 , 195217.

    • Search Google Scholar
    • Export Citation
  • Martin, J. E., , J. D. Locatelli, , P. V. Hobbs, , P-Y. Wang, , and J. A. Castle, 1995: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part I: A synoptic-scale rainband associated with a dryline and lee trough. Mon. Wea. Rev., 123 , 241264.

    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , and R. M. Wakimoto, 1999: The interaction of a Pacific cold front with shallow masses east of the Rocky Mountains. Mon. Wea. Rev., 127 , 21022127.

    • Search Google Scholar
    • Export Citation
  • Neiman, P. J., , F. M. Ralph, , M. A. Shapiro, , B. F. Smull, , and D. Johnson, 1998: An observational study of fronts and frontal mergers over the continental United States. Mon. Wea. Rev., 126 , 25212554.

    • Search Google Scholar
    • Export Citation
  • Newton, C. W., 1950: Structure and mechanisms of the prefrontal squall line. J. Meteor., 7 , 210222.

  • Ogura, Y., , H-M. Juang, , K-S. Zhang, , and S-T. Soong, 1982: Possible triggering mechanisms for severe storms in SESAME-AVE IV (9–10 May 1979). Bull. Amer. Meteor. Soc., 63 , 503515.

    • Search Google Scholar
    • Export Citation
  • Reisner, J., , R. M. Rasmussen, , and R. T. Bruintjes, 1998: Explicit forecasting of supercooled liquid water in winter storms using the MM5 mesoscale model. Quart. J. Roy. Meteor. Soc., 124 , 10711107.

    • Search Google Scholar
    • Export Citation
  • Rose, S. F., , P. V. Hobbs, , J. D. Locatelli, , and M. T. Stoelinga, 2002: Use of a mesoscale model to forecast severe weather associated with a cold front aloft. Wea. Forecasting, 17 , 755773.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , and J. B. Klemp, 1982: The influence of the shear-induced pressure gradient on thunderstorm motion. Mon. Wea. Rev., 110 , 136151.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , and J. B. Klemp, 1985: On the rotation and propagation of simulated supercell thunderstorms. J. Atmos. Sci., 42 , 271292.

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , J. B. Klemp, , and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci., 45 , 463485.

  • Schaefer, J. T., 1986: The dryline. Mesoscale Meteorology and Forecasting, P. S. Ray, Ed., Amer. Meteor. Soc., 549–572.

  • Shapiro, M. A., 1982: Mesoscale weather systems of the central United States. NOAA–CIRES Tech. Rep., 78 pp. [Available from Cooperative Institute for Research in the Environmental Sciences, University of Colorado, Boulder, CO 80309.].

    • Search Google Scholar
    • Export Citation
  • Showalter, A. K., , and J. R. Fulks, 1943: Preliminary report on tornadoes. War Advisory Council on Meteorology Restricted Rep. 1110, U.S. Weather Bureau, 162 pp. [Available from NOAA Central Library, SSMC3 7th Floor, 1315 East–West Highway, Silver Springs, MD 20910.].

    • Search Google Scholar
    • Export Citation
  • Sienkiewicz, J. M., , J. D. Locatelli, , P. V. Hobbs, , and B. Geerts, 1989: Organization and structure of clouds and precipitation on the mid-Atlantic coast of the United States. Part II: The mesoscale and microscale structure of some frontal rainbands. J. Atmos. Sci., 46 , 13491364.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., , J. D. Locatelli, , and P. V. Hobbs, 2000: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part VI: A multiscale numerical modeling study of the initiation of convective rainbands. Mon. Wea. Rev., 128 , 34813500.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., , J. D. Locatelli, , and P. V. Hobbs, 2002: Warm occlusions, cold occlusions, and forward-tilting cold fronts. Bull. Amer. Meteor. Soc., 83 , 709721.

    • Search Google Scholar
    • Export Citation
  • Thorpe, A. J., , M. J. Miller, , and M. W. Moncrieff, 1982: Two-dimensional convection in nonconstant shear: A model of midlatitude squall lines. Quart. J. Roy. Meteor. Soc., 108 , 739762.

    • Search Google Scholar
    • Export Citation
  • Wang, P. Y., , J. E. Martin, , J. D. Locatelli, , and P. V. Hobbs, 1995: Structure and evolution of winter cyclones in the central United States and their effects on the distribution of precipitation. Part II: Arctic fronts. Mon. Wea. Rev., 123 , 13281344.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., 1992: The role of convectively generated rear-inflow jets in the evolution of long-lived mesoconvective systems. J. Atmos. Sci., 49 , 18261847.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., , J. B. Klemp, , and R. Rotunno, 1988: Structure and evolution of numerically simulated squall lines. J. Atmos. Sci., 45 , 19902013.

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., , W. C. Skamarock, , and J. B. Klemp, 1997: The resolution dependence of explicitly modeled convective systems. Mon. Wea. Rev., 125 , 527548.

    • Search Google Scholar
    • Export Citation
  • Zhang, D. L., , and R. A. Anthes, 1982: A high-resolution model of the planetary boundary layer—Sensitivity tests and comparisons with SESAME-79 data. J. Appl. Meteor., 21 , 15941609.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Surface map for 0600 UTC 20 Jun 1979 showing the NWS analysis of sea level pressure (solid lines in hPa), surface features (conventional symbols), stations reports (conventional plotting), and regions of precipitation (shaded areas) enclosed by the lowest-level radar echo from the National Radar Summary. Selected damage reports are noted in the boxes

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 2.
Fig. 2.

(a) Sea level pressure (solid lines in hPa), 700-hPa temperature (dashed lines in °C), and lowest-level radar echo from the National Radar Summary for the CFA squall line at 1200 UTC 19 Jun 1979. (b) As in (a) but at 1200 UTC 20 Jun 1979

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 3.
Fig. 3.

(a) Sea level pressure and winds from the NWS surface map in Fig. 1. (b) Authors' surface temperature analysis (°F) and NWS winds from the surface map in Fig. 1. Shading shows regions of precipitation enclosed by the lowest-level radar echo from the National Radar Summary. (c) As in (b) but for authors' dewpoint analysis (°F). (d) Surface winds (see legend in figure) and sea level pressure (hPa) valid for 0600 UTC 20 Jun 1979 for the MM5 coarse-grid control simulation. Shading shows regions of model precipitation. (e) As in (d) but for surface temperature (°F) from the model coarse-grid control simulation. (f) As in (d) but for dewpoint (°F) from the model coarse-grid control simulation

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 4.
Fig. 4.

The 27-h MM5 coarse-grid forecast valid at 0300 UTC 20 Jun 1979: (a) surface winds, sea level pressure (heavy solid lines in hPa), and precipitation totals for model hours 26–27 (shading: see figure legend) from the coarse-grid control simulation, and (b) from the coarse-grid no-cooling simulation

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 5.
Fig. 5.

The 27-h MM5 coarse-grid forecast valid at 0300 UTC 20 Jun 1979: (a) sea level pressure (heavy solid lines in hPa), dewpoint contours (thin solid lines in °C), and precipitation totals from model hours 26–27 (shading: see figure legend) from the MM5 coarse-grid control simulation, and (b) from the coarse-grid no-cooling simulation

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 6.
Fig. 6.

As in Fig. 5 but thin solid lines are surface potential temperature (θ) contours (K)

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 7.
Fig. 7.

(a) The 27-h MM5 coarse-grid no-cooling simulation of sea level pressure valid at 0300 UTC 20 Jun 1979 minus the coarse-grid control simulation sea level pressure forecast (hPa) for the same time. Precipitation totals from model hours 26–27 (shading: see figure legend) are from the coarse-grid control simulation. (b) As in (a) but for the difference in the 950–800-hPa thickness (m)

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 8.
Fig. 8.

(a) The 27-h MM5 coarse-grid control simulation valid at 0300 UTC 20 Jun 1979 for the pressure field (hPa) produced by the atmospheric layer from 0.4 to 1.6 km. The dashed–dotted line is the axis of the highest pressure in the cold pool, and the dashed–double-dotted line is the axis of a trough of low pressure (see text for explanation). Precipitation totals from model hours 26–27 (shading: see figure legend) are from the coarse-grid control simulation. (b) As in (a) but for the atmospheric layer from 1.6 to 9.0 km. The heavy dashed line is the axis of a trough of low pressure (see text for explanation)

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 9.
Fig. 9.

As in Fig. 8 but for the MM5 coarse-grid no-cooling simulation for (a) the atmospheric layer from 0.4 to 1.2 km and (b) the atmospheric layer from 1.2 to 9.0 km

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 10.
Fig. 10.

Surface features from the MM5 high-resolution control simulation at (a) 2230 UTC 19 Jun 1979 and (b) 0300 UTC 20 Jun 1979. Shown are wind vectors at the lowest model level [vector magnitude scale in lower right of (b)], dewpoint temperature at the lowest model level (solid contours, every 2°C), and 15-min rainfall amounts (light shading, 1–5 mm; dark shading, >5 mm). Hash marks on perimeter represent intervals of 10 grid points (36 km). Heavy dashed line in (b) is the location of the surface trough–dryline. Line A–A′ is the location of the cross sections shown in Figs. 13a and 13b

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 11.
Fig. 11.

Vertical velocity and cold pool depiction from the high-resolution control simulation and the no-cooling simulation at 0100 UTC 20 Jun 1979. (a) Downdrafts (shading, scale shown at bottom) and updrafts (thin solid contours at values of 50, 100, and 200 cm s−1) at a height of 2.1 km AMSL. (b) As in (a) except for the no-cooling simulation. (c) Virtual potential temperature at the lowest model level in the control simulation minus that in the no-cooling simulation (shading, scale at bottom—note that the light gray shade outlined with a thin contour corresponds to a warm anomaly) and 15-min rainfall amounts from the control simulation (solid contours at 1- and 5-mm amounts). Hash marks on perimeter represent intervals of 10 grid points (36 km)

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 12.
Fig. 12.

As in Fig. 10 except for the no-cooling simulation at (a) 0300 UTC 20 Jun 1979 and (b) 0600 UTC 20 Jun 1979. Line A–A′ is the location of the cross sections shown in Figs. 13a and 13b, B–B′ is the location of the cross section shown in Fig. 14, and line C–C′ is the location of the cross section shown in Fig. 13c

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 13.
Fig. 13.

Vertical cross sections through the squall line produced by the high-resolution model simulations. Shown are squall-line-relative vectors in the plane of the cross section (scale at lower right or left—note dp/dt of 1 dPa s−1 corresponds to a vertical velocity of ∼1 cm s−1), virtual potential temperature (solid contours, every 2 K), equivalent potential temperature (shading, scale at bottom), an outline of the cloud and precipitation boundary (heavy solid line, corresponding to 0.025 g kg−1 for the sum of all cloud and precipitation mixing ratios), and a subjectively placed outline of the Pacific cold-frontal surface (heavy dashed line). (a) Cross section from the control simulation at 0300 UTC 20 Jun 1979 along line A–A′ in Figs. 10b and 12a. (b) As in (a) except for no-cooling simulation. (c) Cross section from the no-cooling simulation at 0600 UTC 20 Jun 1979 along line C–C′ in Fig. 12b

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 14.
Fig. 14.

Vertical cross section through the squall line produced by the high-resolution no-cooling simulation at 0300 UTC 20 Jun 1979 along line B–B′ in Fig. 12a. Shown are squall-line-relative vectors in the plane of the cross section (scale at lower right—note dp/dt of 1 dPa s−1 corresponds to a vertical velocity of ∼1 cm s−1), virtual temperature perturbation (heavy gray contours, every 1°C), and pressure perturbation (thin black contours, every 1 hPa). Perturbation fields are based on a standard atmospheric temperature profile and corresponding hydrostatic pressure profile that are functions of height only. Regions where the perturbation fields are higher or lower are labeled as such to give a sense of the directions of the gradients.

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Fig. 15.
Fig. 15.

Schematic showing the differences between (a) a CFA squall line and (b) a warm-sector squall line. See text for explanation.

Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0095:IACPNF>2.0.CO;2

Save