• Auer, A. H., 1987: An aid to forecasting heavy snowfall episodes. Natl. Wea. Dig., 12 , 1114.

  • Barnes, S. L., 1973: Mesoscale objective map analysis using weighted time-series observations. NOAA Tech. Memo. ERL NSSL-62, National Severe Storms Laboratory, Norman, OK, 60 pp.

    • Search Google Scholar
    • Export Citation
  • Bennetts, D. A., , and Hoskins B. J. , 1979: Conditional symmetric instability—A possible explanation for frontal rainbands. Quart. J. Roy. Meteor. Soc., 105 , 945962.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bennetts, D. A., , and Sharp J. C. , 1982: The relevance of conditional symmetric instability to the prediction of mesoscale frontal rainbands. Quart. J. Roy. Meteor. Soc., 108 , 595602.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., 1993a: Principles of Kinematics and Dynamics. Synoptic–Dynamic Meteorology in Midlatitudes, Vol. I, Oxford University Press, 431 pp.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., . 1993b: Observations and Theory of Weather Systems. Synoptic–Dynamic Meteorology in Midlatitudes, Vol. II, Oxford University Press, 594 pp.

    • Search Google Scholar
    • Export Citation
  • Browne, R. F., , and Younkin R. J. , 1970: Some relationships between 850-millibar lows and heavy snow occurrences over the central and eastern United States. Mon. Wea. Rev., 98 , 399401.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Carlson, T. N., 1980: Airflow through midlatitude cyclones and the comma cloud pattern. Mon. Wea. Rev., 108 , 14981509.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Crocker, A. M., , Godson W. L. , , and Penner C. M. , 1947: Frontal contour charts. J. Meteor., 4 , 9599.

  • Emanuel, K. A., 1983: On assessing local CSI from atmospheric soundings. Mon. Wea. Rev., 111 , 20162033.

  • Emanuel, K. A., . 1985: Frontal circulations in the presence of small moist symmetric stability. J. Atmos. Sci., 42 , 10621071.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Godson, W. L., 1951: Synoptic properties of frontal surfaces. Quart. J. Roy. Meteor. Soc., 77 , 633653.

  • Keyser, D., , and Shapiro M. A. , 1986: A review of the structure and dynamics of upper-level frontal zones. Mon. Wea. Rev., 114 , 452499.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Koch, S. E., , DesJardins M. , , and Kocin P. J. , 1983: An interactive Barnes objective map analysis scheme to use with satellite and conventional data. J. Climate Appl. Meteor., 22 , 14871503.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martin, J. E., 1998: The structure and evolution of a continental winter cyclone. Part II: Frontal forcing for an extreme snow event. Mon. Wea. Rev., 126 , 329348.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martin, J. E., . 1999a: Quasigeostropic forcing of ascent in the occluded sector of cyclones and the trowal airstream. Mon. Wea. Rev., 127 , 7088.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martin, J. E., . 1999b: The separate roles of geostropic vorticity and deformation in the midlatitude occlusion process. Mon. Wea. Rev., 127 , 24042418.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , and Blakely P. D. , 1988: The role of frontogenetical forcing and conditional symmetric instability in the Midwest snowstorm of 30–31 January 1982. Mon. Wea. Rev., 116 , 21552171.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , and VanKnowe G. , 1992: The effect of jet-streak curvature on kinematic fields. Mon. Wea. Rev., 120 , 24292441.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , and Lambert T. E. , 1993: The use of equivalent potential vorticity to diagnose regions of conditional symmetric instability. Wea. Forecasting, 8 , 301308.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , Czarnetzki A. C. , , and Market P. S. , 1998: Heavy precipitation associated with elevated thunderstorms formed in a convectively unstable layer aloft. Meteor. Appsl., 5 , 373384.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Murray, R., , and Daniels S. M. , 1953: Transverse flow at entrance and exit to jet streams. Quart. J. Roy. Meteor. Soc., 79 , 236241.

  • Namias, J., , and Clapp P. F. , 1949: Confluence theory of the high tropospheric jet stream. J. Meteor., 6 , 330336.

  • Nicosia, D. J., , and Grumm R. H. , 1999: Mesoscale band formation in three major northeastern United States snowstorms. Wea. Forecasting, 14 , 346368.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • O'Brien, J. J., 1970: Alternative solutions to the classical vertical velocity problem. J. Appl. Meteor., 9 , 197203.

  • Penner, C. M., 1955: A three-front model for synoptic analyses. Quart. J. Roy. Meteor. Soc., 81 , 8991.

  • Sanders, F., , and Bosart L. F. , 1985: Mesoscale structure in the megalopolitan snowstorm of 11–12 February 1983. Part I: Frontogenetical forcing and symmetric instability. J. Atmos. Sci., 42 , 10501061.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saucier, W. J., 1955: Principles of Meteorological Analysis. University of Chicago Press, 438 pp.

  • Schultz, D. M., , and Schumacher P. N. , 1999: The use and misuse of conditional symmetric instability. Mon. Wea. Rev., 127 , 27092732.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Seltzer, M. A., , Passarelli R. E. , , and Emanuel K. A. , 1985: The possible role of symmetric instability in the formation of precipitation bands. J. Atmos. Sci., 42 , 22072219.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Steigerwald, H., 1986: Deformation zones and heavy precipitation. Natl. Wea. Dig., 11 , 1520.

  • Thorpe, A. J., , and Emanuel K. A. , 1985: Frontogenesis in the presence of small stability to slantwise convection. J. Atmos. Sci., 42 , 18091824.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Uccellini, L. W., 1976: Operational diagnostic applications of isentropic analysis. Natl. Wea. Dig., 1 , 412.

  • View in gallery

    Storm total snowfall (in.), 13–14 Mar 1999, for MO, AR, and KS. Isopleths of snowfall of 5 (2) 15 (6), and 30.5 cm (12 in.) are analyzed

  • View in gallery

    Synoptic analysis valid at 0000 UTC 13 Mar 1999. (a) Standard station models of observed surface data are presented with objectively analyzed mean sea level pressure, which is contoured every 4 mb. (b) Analysis of observed 850-mb heights (30 gpm; solid), divergence (10−5 s−1; thin solid; shaded light gray <−2 and dark gray >+2), temperature (4°C; dashed), and ageostrophic winds (arrows; reference vector at lower left). Frontal zone location based upon the observed wind field (not shown), the isotherm pattern, and the ageostrophic wind field. (c) A 500-mb analysis of observed data; standard station models of observed data are presented with objectively analyzed geopotential height (solid) and 1000–500-mb thickness (dashed), both of which are contoured every 60 gpm. Thin lines denote cross-section line used in Figs. 7 and 8 (easternmost line) and in Fig. 10 (westernmost line). (d) Analysis of observed 300-mb heights (120 gpm; solid), divergence (10−5 s−1; thin solid; shaded light gray <−2 and dark gray >+2), isotachs (kt; dashed), and ageostrophic winds (arrows; reference vector at lower left)

  • View in gallery

    As in Fig. 2 but at 1200 UTC 13 Mar 1999. Cross-section lines are absent

  • View in gallery

    As in Fig. 2 but at 0000 UTC 14 Mar 1999. International Falls, MN (INL), to Biloxi, MS (BIX), cross-section line for Figs. 7 and 8 is shown again

  • View in gallery

    As in Fig. 2 but at 1200 UTC 14 Mar 1999. Cross-section lines are absent

  • View in gallery

    Plots of surface observations of sky cover and present weather at (a) 0000 (b) 0600, and (c) 1200 UTC 14 Mar 1999

  • View in gallery

    Cross-section analyses of rawinsonde data from INL to BIX, valid at 0000 UTC 14 Mar 1999. (a) Petterssen frontogenesis [10−1 K (100 km)−1 (3 h)−1; solid] and kinematic vertical velocity (μb s−1; dashed) are shown. (b) Shown are θe (K; bold solid), relative humidity (%; thin solid, shaded >60%, 80% is bold), and Mg (m s−1; dashed). Regions of conditional symmetric instability (CSI), potential symmetric instability (PSI), and potential instability (PI) are outlined in the cross section. In both, bold X near the surface denotes the extreme southeast Missouri boot-heel region. The bold line and snow symbol (**) in both frames denote the northern extent of the precipitation field

  • View in gallery

    Cross-section analyses of rawinsonde data from INL to BIX, valid at 1200 UTC 14 Mar 1999. (a) Petterssen frontogenesis [10−1 K (100 km)−1 (3 h)−1; solid] and kinematic vertical velocity (μb s−1; dashed) are shown. (b) Shown are θe (K; bold solid) and relative humidity (%; thin solid, shaded >60%). In both, bold X near the surface denotes the extreme southeast Missouri boot-heel region. The bold lines and snow symbol (**) in both frames denote the extent of the precipitation field

  • View in gallery

    Plan-view analyses of Petterssen frontogenesis [10−1 K (100 km)−1 (3 h)−1] at 700 mb valid at 0000 (solid) and 1200 UTC 14 Mar 1999 (dashed)

  • View in gallery

    Cross-section analysis from GGW to SIL, valid at 0000 UTC 14 Mar 1999. Potential temperature is analyzed every 2 K (solid), with relative humidity contoured at 20%, 40%, 70%, and 90%, and shaded light (40%–70%), medium (70%–90%), and dark (>90%) gray. Observed winds are plotted (kt), with half barbs (5 kt), full barbs (10 kt), and flags (50 kt) on each shaft. Bold lines at the base of the figure denote the extent of precipitation in the figure at this time

  • View in gallery

    Radar summary valid at (a) 0000 and (b) 1200 UTC 14 Mar 1999. (From National Climatic Data Center online archives)

  • View in gallery

    (a) Objectively analyzed streamlines of the observed wind, pressure (mb; dashed), and relative humidity (%; gray shades) with legend on left, valid on the 298-K surface at 1200 UTC 13 Mar 1999. (b) Resultant deformation of the observed wind (10−5 s−1; solid) valid on the 298-K surface at 1200 UTC 13 Mar 1999

  • View in gallery

    Objectively analyzed streamlines of storm-relative flow (C of uC = +3.42 m s−1 and υC = −6.69 m s−1), pressure (mb; dashed), and relative humidity (%; gray shades) with legend on left, valid on the 298-K surface at 0000 UTC 14 Mar 1999. (b) Resultant deformation of the observed wind (10−5 s−1; solid) valid on the 298-K surface at 0000 UTC 14 Mar 1999. (c) Isobars (every 50 mb; solid) depicting the 310-K θe surface. (d) Pressure advection (−V · ∇p) on the 298-K isentropic surface (μb s−1; thin solid, shaded <−2), and mixing ratio advection (−V · ∇r) on the 298-K isentropic surface (10−8 kg kg−1 s−1; dashed)

  • View in gallery

    (a) Objectively analyzed streamlines of storm-relative flow (C of uC = +13.42 m s−1 and υC = +2.83 m s−1), pressure (mb; dashed), and relative humidity (%; gray shades) with legend on left, valid on the 298-K surface at 1200 UTC 14 Mar 1999. (b) Resultant deformation of the observed wind (10−5 s−1; solid) valid on the 298-K surface at 1200 UTC 14 Mar 1999. (c) Pressure advection (−V · ∇p) on the 298-K isentopic surface (μb s−1; thin solid, shaded <−2), and mixing ratio advection (−V · ∇r) on the 298-K isentropic surface (10−8 kg kg−1 s−1; dashed)

  • View in gallery

    Cross-section analysis of storm-relative winds (using C of uC = +13.42 m s−1 and υC = +2.83 m s−1) shown plotted (short barb, 5 kt; long barb, 10 kt; flag, 50 kt) and contours of the plane-normal storm-relative wind (kt; solid) valid at 1200 UTC 14 Mar 1999. The bold lines and snow symbol (**) at the base of the figure denote the extent of the precipitation field

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 105 105 4
PDF Downloads 50 50 2

Formation of a Sharp Snow Gradient in a Midwestern Heavy Snow Event

View More View Less
  • 1 University of Missouri—Columbia, Department of Soil and Atmospheric Sciences, Columbia, Missouri
  • | 2 National Weather Service Office, Springfield, Missouri
© Get Permissions
Full access

Abstract

A case study of the 13–14 March 1999 heavy snow event across southern Missouri and neighboring states is presented. Of the many features that made this storm notable, the very sharp gradient on the northern periphery of the snowfall field was most intriguing. Moreover, that the snowfall field was confined to the southern half of the state resulted in snow-free regions across central Missouri where significant accumulations had been predicted. The focus of this study was thus to reveal the cause of such large snowfall gradients. Little evidence exists of convective snowfall over Missouri through 1200 UTC 14 March 1999, when this study ends. Analyses confirm that the release of neither convective instability nor conditional symmetric instability was responsible for the large snowfall gradient on the northern boundary. Instead, the juxtaposition of dry and moist airstreams from the north and south, respectively, as components of a deformation zone ultimately defined the large snowfall gradient across southern Missouri.

Corresponding author address: Dr. Patrick S. Market, Dept. of Soil and Atmospheric Sciences, University of Missouri—Columbia, 203 Gentry Hall, Columbia, MO 65201. Email: marketp@missouri.edu

Abstract

A case study of the 13–14 March 1999 heavy snow event across southern Missouri and neighboring states is presented. Of the many features that made this storm notable, the very sharp gradient on the northern periphery of the snowfall field was most intriguing. Moreover, that the snowfall field was confined to the southern half of the state resulted in snow-free regions across central Missouri where significant accumulations had been predicted. The focus of this study was thus to reveal the cause of such large snowfall gradients. Little evidence exists of convective snowfall over Missouri through 1200 UTC 14 March 1999, when this study ends. Analyses confirm that the release of neither convective instability nor conditional symmetric instability was responsible for the large snowfall gradient on the northern boundary. Instead, the juxtaposition of dry and moist airstreams from the north and south, respectively, as components of a deformation zone ultimately defined the large snowfall gradient across southern Missouri.

Corresponding author address: Dr. Patrick S. Market, Dept. of Soil and Atmospheric Sciences, University of Missouri—Columbia, 203 Gentry Hall, Columbia, MO 65201. Email: marketp@missouri.edu

1. Introduction

On 13–14 March 1999, southern Missouri was visited by a heavy snowfall event that produced well over 30 cm (12 in.) of snowfall in 18 h over many counties (Fig. 1). This storm system was notable in several respects. First, it presented a difficult forecasting problem in terms of predicting the onset of snowfall. Model output failed to demonstrate consistently a consensus between different model solutions or continuity among subsequent simulations from the same model. With systems that generate such locally heavy snowfall amounts [58 cm (23 in.)], timing is as crucial in a forecast as accumulations and areal extent. Second, although the southern edge of the snow field was diffuse and the gradient of snow accumulations relatively small, the northern boundary of the snow field was quite sharp, with large gradients in the snow totals. Indeed, population centers such as St. Louis and Columbia, Missouri, received no snowfall although forecasts for those locations had consistently warned of snowfall for up to 48 h in advance of the (local) nonevent. It is the latter feature, the sharp northern snowfall boundary, that is the focus of this paper. Last, although conditional symmetric instability (CSI) and weak conditional symmetric stability (CSS) in the presence of frontogenesis were diagnosed as far north as northern Arkansas, the atmosphere over most of Missouri will be shown to be stable to both vertical and slantwise parcel displacements.

Typically, sharp snowfall gradients suggest the release of CSI or the presence of locally strong frontogenesis (e.g., Moore and Blakely 1988; Moore and Lambert 1993). Such was not the case here. Instead, transports of moist and dry air along neighboring airstreams within a deformation zone led to the sharp snow accumulation gradient, an interaction discussed by Steigerwaldt (1986). To establish this, this paper will follow this format: section 2 will present a brief overview of our analysis methods of observed data and model output; section 3 provides a synopsis of the event, emphasizing the interaction of the short-wave systems and the creation of the deformation zone; section 4 elucidates mesoscale processes, especially the origins of dry air north of the primary storm system; and in section 5 we offer some concluding remarks.

2. Methodology

a. Observed data

Both surface (METAR, from a French phrase meaning aviation routine weather report) and upper-air observed data form the foundation of this study. In addition to subjective, hand-drawn analyses of plotted data, these data were objectively analyzed to a representative grid using the Barnes (1973) technique as employed in the General Meteorological Package (GEMPAK; Koch et al. 1983). Specifically, the upper-air data were fitted to a 28 × 28 grid with an average grid spacing of 150 km. Diagnoses of pressure, deformation, and storm-relative flow on an isentropic surface from observed data will be based upon these objectively analyzed fields. Cross-section analyses of frontogenesis, relative humidity, equivalent potential temperature (θe), and absolute geostrophic momentum (Mg) are also based upon these objectively analyzed fields. Calculations of kinematic vertical motions shown are based upon the O'Brien (1970) scheme also found in GEMPAK. Surface data, being more densely distributed, were fitted to a 55 × 55 grid with a smaller average grid spacing of 75 km.

3. Synopsis

a. 0000 UTC 13 March 1999

At this time, the surface analysis clearly suggests a developing storm system, with an anticyclone (1033 mb; not shown) centered over the upper peninsula of Michigan and a developing cyclone (998 mb) over the Rio Grande valley of west Texas (Fig. 2a). In between these features, easterly to northeasterly winds and falling pressures cover most of Missouri in response to the retreating high pressure to the north. The clear skies, dewpoint depressions of 5.6°–8.3°C (10°–15°F) over northern Illinois, and easterly surface flow regime suggest a persistent supply of relatively cold dry surface air into northern Missouri overnight. However, the influence of the developing storm is felt already as far east as eastern Missouri where the sky has already become overcast with cirrus. Precipitation is already under way at this hour, with light snow falling as far north as central Nebraska and rain as far east as Arkansas. In the latter case, the state of Arkansas is bracketed by light freezing rain in the northwest and elevated convection in the southeast part of the state.

The 850-mb analysis valid at 0000 UTC 13 March 1999 (Fig. 2b) reveals a warm front extending from a low (1400 gpm) centered in southern Texas, northeastward across southern Arkansas. The 0°C isotherm is oriented east–west, across southern Missouri, placing much of the state below freezing at 850 mb. The location of the 850-mb 0°C isotherm has been shown to be an effective snowfall forecasting tool (Browne and Younkin 1970), because the underlying layer is relatively shallow and usually features temperatures close to freezing. A large region of convergence is indicated at this level along and north of the 850-mb warm front; meanwhile, the southeasterly ageostrophic wind field across extreme eastern Texas, northern Louisiana, and southwestern Arkansas is oriented nearly perpendicular to the 300-mb jet streak, to be discussed later. The observed winds (not shown) place the core of the southerly low-level jet over northwestern Louisiana, with a speed of 25 m s−1.

At 500 mb (Fig. 2c), a closed low (5510 gpm) is found over eastern New Mexico, indicating a well-organized baroclinic position northwest of the surface low. Although significant geopotential and thermal gradients are present downstream of the trough axis [with a local 40 m s−1 (80 kt) wind maximum at Midland, Texas (MAF)], the coldest air is still upstream of the axis, a favorable factor for further cyclogenesis at that level. Of equal interest in this case is the weak short-wave trough over North Dakota. As this disturbance propagates eastward in the northern stream, it will provide crucial kinematic support, as well as a source of dry air over Missouri, which will be revealed in later analyses.

The analysis at 300 mb valid at 0000 UTC 13 March 1999 (Fig. 2d) shows one distinct jet core just east of the trough axis at that level. The best divergence values are found downstream (following the geostrophic flow) concurrent with the diffluent region centered over Oklahoma. Although the ageostrophic wind pattern and the observed low-level jet over northern Louisiana at 850 mb (Fig. 2b) suggest the lower-tropospheric arm of the indirect thermal circulation associated with the 300-mb jet streak, the ageostrophic flow regime at 300 mb suggests divergence born of along-stream rather than cross-stream variation of the ageostrophic wind. Clearly, the classic indirect thermal circulation expected with linear jet streaks (e.g., Namias and Clapp 1949; Murray and Daniels 1953; Uccellini 1976; Moore and VanKnowe 1992) is not yet present in this case. This is due in large part to the immaturity of the developing cyclone system as well as the proximity of the jet core to the trough axis at this level, and the effects of curvature and gradient flow on the resulting ageostrophic wind pattern (Keyser and Shapiro 1986).

b. 1200 UTC 13 March 1999

At this hour, the surface low (1004 mb) is weaker, but is propagating rapidly eastward, with centers on the Gulf Coast south of Lake Charles, Louisiana, and over western Louisiana (Fig. 3a). To the north, the surface anticyclone (1032 mb) has neither weakened, nor has it moved much. However, surface pressure rises of 5–10 mb over the preceding 12 h across Utah and Colorado culminate in a distinct ridge over these states. It is also noteworthy that skies cleared over Nebraska and Colorado during the 12 h prior, concurrent with the development and elongation of the surface ridge, signaling the establishment of a deep layer of dry air.

By 1200 UTC 13 March, the 850-mb low (Fig. 3b) is positioned over extreme northeast Texas (1400 gpm), with the associated warm front moving northward. The divergence field reveals an area of significant convergence centered over Tennessee and northern Mississippi, along the 850-mb warm frontal position. While the ageostrophic wind field at 850 mb reveals little of value with respect to jet streak thermal circulations at this time, it is valuable for diagnosing the development of the warm front at 850 mb. In particular, note the confluent nature in the vector field, signaling further strengthening of the frontal zone at this level.

While not substantially deeper at this hour, the 500-mb (Fig. 3c) closed low is exhibiting development consistent with typical midlatitude cyclone evolution. Cold air arriving in the base of the trough and height rises of 50–100 gpm over Arizona and Utah help to decrease the length of the wave. To the northeast of the 500-mb low, a well-defined deformation zone centered over northern Iowa (calculated and shown explicitly in section 4) is now present, developed and enhanced by the trough over Minnesota propagating eastward in the northern stream. While the atmosphere is quite moist (dewpoint depression <5°C) over most of the central United States, dewpoint depressions are ≥10°C over Minnesota and the Dakotas associated with the northern stream trough.

The 300-mb analysis at 1200 UTC 13 March 1999 reveals the primary jet core centered over Oklahoma with evidence of a second jet core to the northeast over Ohio. Divergence is diagnosed in the right entrance and left exit regions of the primary jet core, with convergence identified in the left entrance region of the streak. The ageostrophic flow pattern clearly indicates a growing dominance of the cross-stream component in shaping the 300-mb divergence field at this time. Thus the classic model for the straight jet streak is beginning to emerge. Still, divergence is also calculated over northern Mississippi, in the right exit region of the Oklahoma jet core. The broad region of southerly ageostrophic flow over the Ohio River valley region suggests that the extended region of divergence (which includes the southeastern one-third of Missouri) is forced largely by the cross-stream contribution in the entrance region of the northern jet core.

c. 0000 UTC 14 March 1999

A slight deepening of the surface low (Fig. 4a) is evident at this hour, as the pressure (1000 mb) was some 4 mb lower than 12 h previously. A distinct closed low circulation has developed, with 3-h pressure changes in eastern Mississippi of greater than 4 mb as the low approached and deepened. A broad precipitation shield attends this system, with heavy rainfall and thunderstorms along the Gulf Coast and a broad region of light rain farther north across Tennessee and Arkansas. Snow is falling on the northwest periphery of the precipitation shield, heaviest at Springfield, Missouri, at this time. To the north, the anticyclone over Ontario and Quebec remains quasi-stationary, while the associated ridge to its southwest is becoming sharper, with pronounced cross-isobar flow across Kansas, Oklahoma, and Texas toward the surface cyclone. Note also the abrupt change to stations reporting clear skies over Kansas and Iowa. These surface observations suggest that appreciably drier air resides immediately beside far more moist air, an assertion that will be supported and confirmed shortly.

At 0000 UTC 14 March 1999, the 850-mb low (to 1350 gpm) resides over central Arkansas (Fig. 4b). The 0°C isotherm is much farther south than even 12 h ago, indicating that the layer beneath 850 mb that may be above freezing is becoming more shallow; this motion also indicates significant upward vertical motion capable of offsetting the horizontal warm air advection at this level. Indeed this contention is supported by surface observations of snowfall as far south as northern Arkansas (Fig. 4a). The strongest convergence persists along the warm front, maximized over south-central Kentucky. Meanwhile, most of Missouri features a very weak convergence field, although divergence is actually diagnosed over southeastern Missouri. While the ageostrophic flow pattern lends no clues to the secondary transverse ageostrophic circulations at this time, there is still a weak confluent pattern apparent over the Kentucky–Tennessee border suggesting that the 850-mb warm front was still in a frontogenesis phase.

Both the northern stream trough and the southern stream closed low have become well established at 500 mb at this time (Fig. 4c). The southern closed low continues to deepen, with central geopotential heights 30 gpm lower than 12 h prior and affecting a height fall of 130 gpm over northeast Texas. More important, the trough in the northern stream, having propagated eastward more rapidly than its closed southern neighbor, is now in a position to contribute better to the well-defined deformation zone over Missouri and southern Illinois. Note the pronounced easterly wind components at Springfield, Missouri (SGF), and Topeka, Kansas (TOP), while farther east, Lincoln, Illinois (ILX), and Nashville, Tennessee (BNA), have large westerly contributions to the total wind. This abrupt longitudinal wind gradient has an analog in the latitudinal moisture gradient, as we note dewpoint depressions of 21° and 22°C at TOP and ILX, respectively, while farther south dewpoint depressions of only 2° and 7°C are reported at SGF and BNA, respectively. The establishment of the deformation zone due to the greater eastward propagation speed of the northern short-wave trough is a critical component in the distribution and orientation of the final surface snowfall field.

At 300 mb, significant divergence covers most of Missouri at 0000 UTC 14 March 1999. The fields of divergence and ageostrophic wind at this level are more complex at this time because of the deepening nature of the cyclone (now closed), and the maturing col and divergence maximum over northeastern Missouri. Still, Missouri resides between the jet streak cores over northwestern Louisiana and the eastern Great Lakes, with the associated divergence extending over a broad area and featuring the aforementioned maximum centered over northeastern Missouri. The surface analysis for this hour (Fig. 4a) indicates precipitation, at times heavy, over southwestern Missouri and northeastern Oklahoma, collocated with an area of weak divergence. However, no precipitation is observed with the divergence maximum over Missouri where the cross-stream ageostrophic wind dominates, because insufficient moisture exists to benefit from the vertical motion (demonstrated in Figs. 4a and 4c, and section 4). Of additional note, however, is the classic swirl pattern in the ageostrophic wind field at this level, indicative of convergence production (following Bluestein 1993a, 173–174). The swirl pattern is centered directly on the 300-mb cyclone, where convergence values of −2 × 10−5 s−1 are diagnosed.

d. 1200 UTC 14 March 1999

By this time, surface-observed precipitation has largely ended over southwest and southern Missouri, although light snow persists over southeast Missouri and the boot-heel. In addition to the occlusion and northeast motion of the surface low (Fig. 5a), the surface ridge, while weaker, has elongated, orienting itself parallel to the deformation zone farther aloft. Rising pressures and backing winds overnight across Missouri portend the improving conditions across the state.

The 850-mb analysis for this time reveals a 1340-gpm low over extreme western Tennessee, with a warm front extending northeast, to eastern Kentucky (Fig. 5b). Easterly flow dominates Missouri with observed dewpoint depressions ranging from 1°C over southern Missouri to at least 21°C over northern Missouri (not shown). Convergence and a confluent pattern in the ageostrophic wind field continue to support the location of the warm front. In addition, the ageostrophic wind field over Georgia is oriented nearly perpendicular to and toward the southern jet core at 300 mb; the observed low-level jet [southerly at 25 m s−1 (50 kt)] also resides over Georgia (not shown).

Though still a closed low, the 500-mb cyclone appears less cut off from the northern stream now (Fig. 5c). The northern disturbance now over New York and Pennsylvania is well downstream of the southern low at this time, but the northerly flow over Illinois and Missouri had introduced cooler temperatures (at 500 mb and below) and thus falling heights in those areas. With the coldest temperatures in the base of the trough, the highest wind speeds by far found downstream of the low, and the previously noted occlusion process at the surface, further development seemed unlikely. In addition, the expanding southern cyclonic circulation and the progression of the northern stream trough downstream have reoriented the east–west-aligned deformation zone north of the low, and now it stretches from southeastern Missouri toward northwestern Ohio. Clearly, this deformation zone played a role in suppressing precipitation development farther east as well (Fig. 5a).

Analysis of the 300-mb level at this time reveals divergent flow over most of Missouri as well as Tennessee and Kentucky where significant precipitation is also occurring; the strongest divergence (8 × 10−5 s−1) is found to the left of the southern jet core over southeastern Missouri where moisture is still sufficient to produce snowfall (Fig. 5a). Note the ageostrophic flow field concurrent with the divergence maximum over southeastern Missouri. This flow indicates decelerating parcels emerging from the trough and into a region of weaker geopotential height gradient. With the pressure gradient force reduced, parcels turn to the right (in the direction of the ageostrophic wind vectors) to satisfy the temporarily dominant Coriolis force. This behavior is a product of the diffluent geopotential height pattern as well as the deformation zone in the observed wind field, both of which reside over southeastern Missouri at this time. In the ensuing section we will explore the vertical motions suggested by this divergence field.

4. The precipitation gradient

Although radar summaries that depict the evolution of the precipitation field are provided later in this section, we shall first examine the surface observations of precipitation and sky condition. A band of heavy snowfall stretching from Stillwater, Oklahoma (SWO), to Springfield, Missouri (SGF), is suggested by the surface observations at 0000 UTC 14 March 1999 (Fig. 6a). Notice also the prevalence of overcast skies over most of the analysis domain, while reports of clear skies at Kirksville, Missouri (IRK); St. Joseph, Missouri (STJ); and over northern Missouri as well as other stations to the east and west are found on the northern periphery of the region. At 0600 UTC 14 March 1999, the number of stations reporting clear to mostly clear skies has grown as has the area influenced by snowfall (Fig. 6b). The snow at Rolla, Missouri (VIH), began near 0400 UTC (Table 1) and represents the northernmost surface station to report snow over mid-Missouri during this event. Meanwhile the snow at SGF is still falling, though weaker at this hour (Table 2). By 1200 UTC the snowfall persists (though lightly) at SGF (Fig. 6c) and is found as far southwest as McAlester, Oklahoma (MLC). These reports suggest the presence of a contiguous band of precipitation. The sky condition reports continue to show clearing from north to south with scattered clouds at Joplin, Missouri (JLN), and clear skies at SWO and Bartlesville, Oklahoma (BVO). All three of these locations experienced heavy snow no more than 12 h prior to this time. Analysis in this section will demonstrate the roles of midlevel deformation, frontogenesis, and instability in shaping this band of heavy precipitation.

a. Frontogenesis and instability assessment

Cross sections oriented perpendicular to the precipitation band and the 1000–500-mb thickness field (locally over most of Missouri), as well as across the warm frontal zone at 0000 UTC 14 March 1999, were used to assess the forcing mechanisms (and gradients therein) for the precipitation band. A low-to-middle-level upward vertical motion maximum (−2 μb s−1) is identified in cross section, centered over Missouri in conjunction with a sloping area of frontogenesis at 0000 UTC 14 March 1999 (Fig. 7a). Also of note is the small region of downward motion immediately to the south that corresponds to a frontolysis minimum. Such a distribution corresponds to a local break in the precipitation field over southern Missouri and northern Arkansas. Yet to the south is where the greater supply of moisture exists (Fig. 7b). Thus the frontogenetical forcing and attendant ascent are maximized over southern Missouri on the fringe of the moisture necessary to capitalize on the lifting. That heavy snow was observed in the region after 0000 UTC 14 March 1999 demonstrates the availability of moisture that could benefit from the forcing and ascent.

At this point, a diagnosis of conditional symmetric instability becomes valid. Seltzer et al. (1985) state that if the precipitation bands are oriented along the thermal wind shear vector, are quasi-two-dimensional, and are in an environment of strong vertical wind shear, then CSI may be present. On a vertical cross section, CSI is diagnosed by comparing the slopes of absolute geostrophic momentum (Mg) and equivalent potential temperature (θe) isopleths. In a saturated environment, the region with steeper θe contours than Mg contours indicates the presence of CSI (Emanuel 1983). Discussion of the interaction of frontogenesis and small conditional symmetric stability is also found throughout the literature (e.g., Sanders and Bosart 1985; Thorpe and Emanuel 1985; Bluestein 1993b). It has been shown that frontogenesis in a region of conditional symmetric neutrality or small conditional symmetric stability results in a narrow, sloping updraft. Emanuel (1985) states: “An atmosphere with small conditional symmetric stability will cause a strong, concentrated sloping updraft ahead of the maximum frontogenesis.” The release of midlevel potential instability (PI) has also been shown to enhance the secondary indirect vertical circulation pattern, positively contributing to ascent for major snow swaths (Martin 1998).

Indeed, the instability regime at 0000 UTC 14 March 1999 (Fig. 7b) comprises large regions of PI and potential symmetric instability (PSI) as well as smaller regions of CSI and conditional instability (CI). This terminology is in keeping with Schultz and Schumacher (1999), although the use of θe is retained, in keeping with the larger body of literature (e.g., Bennetts and Hoskins 1979; Sanders and Bosart 1985; Moore and Lambert 1993). In this work, only those regions of a cross section where the slope of θe contours exceeds that of Mg and where the relative humidity exceeds 80% are termed as CSI. This approach follows the landmark work of Bennetts and Sharp (1982) who assumed saturation for relative humidity >80% to account for observational errors and also the smoothing introduced by objective analysis. In cases where the assumption of saturation is valid, then θe is identical to the saturation equivalent potential temperature (θes) and a diagnosis of CSI can indeed be made. Moreover, the retention of relative humidity in the cross sections presented here serves to discriminate between those regions where a diagnosis of PSI or CSI is operationally worthwhile and those where it is of mere academic interest. Indeed, the cross section for 0000 UTC 14 March 1999 reveals a small area of CSI over extreme southern Missouri at ∼650 mb, the northern portion of which is weakly frontogenetical and experiencing upward vertical velocities of up to −1 μb s−1; weak CSS is found still farther north at the same level. A slantwise parcel displacement from this CSI region involves parcels that are saturated, as opposed to the much larger area of PSI above. Not only would parcels from the PSI region have to be lifted to achieve saturation, but in this case, much of the PSI region is frontolytic, and experiencing only weak upward vertical motion, with a pocket of descent analyzed just above the PSI region. Especially in the CSI region, ascent and the release of instability could have produced precipitation (especially snow crystals) to seed saturated layers beneath, closer to the ground over northern Arkansas. Moreover, lightning data (from the National Lightning Detection Network) indicated several cloud-to-ground strokes in extreme northeast Arkansas, extreme southeast Missouri, and over southern Illinois from the afternoon of the 13 March 1999 to the early morning hours of 14 March 1999; clearly, vertical motions of a magnitude necessary to achieve charge separation were achieved in localized areas. While these fields are maximized just south of the region of heaviest snowfall, we acknowledge the likelihood that important mesoscale gradients in derived parameters (e.g., frontogenesis, deformation) may be broadened and deemphasized because of the wide grid spacing (150 km) chosen for the objective analysis. Frontogenesis in the presence of weak CSS likely contributed to the enhanced snowfall totals over southwestern Missouri.

To maintain consistency, the same cross-section line is retained for analyses of 1200 UTC 14 March 1999. Furthermore, use of this same cross section is warranted, as precipitation persists over the southeastern third of Missouri, a region through which this cross section passes. Interestingly, much of Missouri and locations farther south are overlain by a deep layer of negative lower- to middle-tropospheric frontogenesis (Fig. 8a). A couplet of positive–negative frontogenesis exists at ∼400 mb over Missouri and Arkansas; yet there is little moisture at that level to benefit from the forcing. More important, the calculated vertical motion field yields an ascent maximum just north of and below the negative frontogenesis maximum while a descent maximum is diagnosed below and farther to the north of the positive frontogenesis maximum. Last, the analysis for moisture and stability (Fig. 8b) clearly indicates an environment that is stable for parcels perturbed in a vertical or slantwise fashion. That the upward vertical motion maximum occurs over southern Missouri in concordance with the precipitation band at this time indicates the presence of a forcing mechanism for ascent other than frontogenesis.

Indeed, plan-view analyses of frontogenesis at 700 mb (Fig. 9) bear this out. The 700-mb level was selected because it exhibits the best signal of frontogenesis, especially at 0000 UTC 14 March 1999. Although at that time a weak axis of frontogenesis [∼0.2 K (3 h)−1 (100 km)−1] is oriented east to west across Missouri, by 1200 UTC 14 March 1999 the positive frontogenesis pattern has vanished, with the eastern two-thirds of Missouri under negative values, thus mirroring the behavior observed in the cross-section analyses. That upward vertical motion is enhanced at 0000 UTC 14 March 1999 over southwestern Missouri because of, in part, frontogenesis is not in question. This time coincides most closely with the heaviest snowfall. Again, the relatively coarse 150-km grid spacing likely diminishes the field of frontogenetical forcing. Yet, with the ascent and its forcing explained, further analysis is required to establish why the snowfall gradient is so large.

b. Deformation and the origins of dry air

We begin our analysis of the moisture gradient with a cross section through the storm from Glasgow, Montana (GGW), to Slidell, Louisiana (SIL), valid at 0000 UTC 14 March 1999 (Fig. 10). This analysis depicts the broad deep moisture layer attending the cyclone south of SGF and below about 500 mb. However, significant relative humidities (>70%) extend the depth of the troposphere as high as 300 mb at SGF. This concurs well with the precipitation observed at this same time at SGF and Little Rock, Arkansas (LZK; Fig. 11a). Note the large gradient in the relative humidity field between SGF and TOP, below 500 mb. While the relative humidity is >80% at SGF up to 400 mb, the relative humidity at TOP over the same depth exceeds 70% only once in a thin layer near 650 mb. Of interest, too, is the deformation zone in the middle to lower troposphere across Missouri suggested by the wind profiles at TOP and LZK and even between TOP and SGF. Indeed, the former depicts northeasterly flow up to the 400-mb level, while the latter exhibits southwesterly flow above ∼600 mb.

In order to examine the moisture gradient and the air masses that generated it, we select the 298-K isentropic surface to render the plan-view motion and moisture fields. Selection of the 298-K surface is also useful in that the −13° to −17°C thermal layer where dendritic crystal growth is preferred (Auer 1987) is resolved in this case. The 600-mb level on a 298-K surface corresponds to a temperature of −15.6°C, which is precisely the preferred 600-mb temperature for the 12–24-in. snowfall events studied by Auer (1987).

Because of the intersection of the 298-K surface with the terrain in the western United States at 0000 UTC 13 March 1999, observed streamlines are offered at 1200 UTC 13 March 1999 (Fig. 12). However, storm-relative streamlines are presented for both 0000 UTC (Fig. 13) and 1200 UTC (Fig. 14) 14 March 1999. These storm-relative motions are given by
VSRVobservedC
where C is the calculated storm motion (Saucier 1955; Moore et al. 1998), based upon the 12-h motion of the absolute vorticity maximum on the 298-K surface. The storm-relative velocities at 0000 UTC 14 March 1999 are based upon components of C of uC = +3.42 m s−1 and υC = −6.69 m s−1, calculated from the 12-h absolute vorticity maximum motion between 1200 UTC 13 March 1999 and 0000 UTC 14 March 1999; the 12-h period ending at 1200 UTC 14 March 1999 revealed a 12-h C of uC = +13.42 m s−1 and υC = +2.83 m s−1. It should be noted that neither of these velocities corresponds well to the surface low, which is generally toward the northeast throughout the 24-h period ending at 1200 UTC 14 March 1999. Still, we retain the motion of the absolute vorticity center to define C because parcels will, to a first approximation, follow the isentropic surface defined by their potential temperatures. As such, we are exploring motions relative to the storm center on the level we have chosen to examine, and not relative to the storm motion on some other, arbitrary level (i.e., the surface).

We return to the 298-K analysis for 1200 UTC 13 March 1999 (Fig. 12a) and note a well-developed, closed circulation over northern Texas. To the east, a pressure (thermal) ridge is present over the lower Mississippi River valley as well as an airstream of the like described by Carlson (1980), which entered on the Gulf Coast and followed the Mississippi River before splitting into an eastward-moving branch over northern Illinois and a westward-moving flow across southern Nebraska and northern Kansas. The pressure (thermal) ridge constitutes the trowal (Crocker et al. 1947; Godson 1951; Penner 1955), along with its attendant airstream that has received renewed attention of late (e.g., Martin 1999a). The diffluent region found in the warm conveyor belt coupled with the one over South Dakota constituted a significant deformation zone. A maximum in resultant deformation (Fig. 12b) was calculated over northern Iowa and southern Minnesota. The trough in the northern stream, and its interaction with the closed cyclonic system to the south, produced a classically appearing deformation zone and a maximum in resultant deformation (Fig. 12b) over northern Iowa and southern Minnesota. Of great importance here is the moisture field with large values of relative humidity attending the warm airstream and a sharp gradient of relative humidity across Missouri, southern Illinois, Kentucky, and Tennessee. The southerly flow south of this gradient suggests a northward advection of moisture.

Indeed, the moisture field does progress northward by 0000 UTC 14 March 1999 (Fig. 13a), but across Missouri, the gradient becomes oriented almost east–west, with more moistening at 298 K over southern than northern Missouri. The heavy snowfall over Oklahoma at this time (Fig. 6c) corresponds to the 650–600-mb layer over northeast Oklahoma. Also, although we employ storm-relative streamlines at this analysis time, Fig. 13a reveals the expected eastward motion of all key features. Chief among these features is the deformation zone centered over Wisconsin. The western half of the axis of dilatation across northwestern Iowa and Nebraska not only represents descent manifested by isentropic downglide, but a dry flow as well, especially with relative humidity values along and north of the axis less than 40%. The trowal is more pronounced at this time, here shown as a series of θe contours (Fig. 13c) at successively lower pressures as suggested by Martin (1998); this analysis depicts a pressure trough that generally agrees with the pressure pattern identified on the 298-K potential temperature surface in Fig. 13a. The stream splitting and diffluence of the warm conveyor belt are more pronounced at this hour as well (Fig. 13a). Thus, it is the deformation zone centered over Wisconsin (Fig. 13b) forced by the differences in translation speeds between the southern cyclone and the northern stream trough and the dry descending air behind the latter that compressed the moisture gradient across Missouri.

The precipitation field over southern Missouri was becoming better organized (Fig. 11a) but clearly remained over southern Missouri and northern Arkansas. Indeed, ample moisture advection exists across the Deep South, with a distinct gradient along the Ohio River valley and westward across mid-Missouri (Fig. 13d). Moreover, vertical motion estimated through the use of the storm-relative advection of pressure on an isentropic surface indicates ascent over the Mississippi and Ohio River valleys, with a local maximum over northwestern Arkansas (Fig. 13d). While most of Missouri resides beneath upward vertical motion on the 298-K surface at this hour, the values drop off as we look farther north (and aloft) where moisture is far less plentiful (Fig. 13a). The results do not seem supportive of strong clearing at this time. Still, we note the deep moisture gradient north of SGF at this time (Fig. 10). Moreover, descent and negative moisture advection are diagnosed over northern Missouri (Fig. 12b). This area expanded southward as the storm system propagated eastward. We note the sequence of surface observations at VIH (Table 1) where snow began before 0500 UTC and ended (after 0900 UTC) relatively late. In particular, we note that the end of the snowfall coincided with a 700-m (2300 ft) increase in the ceiling, a 13.7-km (8.5 mi) increase in visibility, and a 3°C drop in the dewpoint temperature. We reemphasize here our view of the deformation zone and its axis of dilatation not so much as a region to focus ascent and inclement weather (Martin 1999b), but as a physical, kinematic barrier that acted to retard the northward progression of moisture.

By 1200 UTC 14 March 1999 (Fig. 14a), dramatic changes had occurred on the 298-K surface. A broad easterly storm-relative flow regime had become established over Missouri, north of the broadening cyclone center. The center of the deformation zone (Fig. 14b) was found oriented southwest–northeast and maximized over Indiana, a testament to the greater translation speed of the northern stream trough. Moreover, the axis of dilatation extended far west across central Missouri. To the north of the axis, very dry air was found (<10% over Iowa) following a 150–200-mb descent from Ontario across the Great Lakes and into Missouri. Meanwhile, to the south of the axis, a tongue of moist air was present across southern Missouri, with an even stronger gradient of moisture advection across southeastern Missouri (Fig. 14c). Indeed, this analysis reflects well the location of the precipitation field at this time (Fig. 11b). Concurrent with the moisture advection maximum is an isentropic vertical motion maximum (Fig. 14c) over southeastern Missouri. This maximum exists at ∼650 mb (Fig. 14a) within the trowal airstream and is of a value consistent with the kinematic vertical velocity at ∼650 mb in Fig. 8.

At this time, we also examine a cross section of storm-relative winds and the component of the storm-relative wind that is normal to the cross section (Fig. 15). An analysis of the layer around ∼450 mb where the vertical motions were diagnosed (Fig. 8) reveals a storm-relative flow from a southeasterly direction at and south of the ascent maximum. North of the ascent maximum, the storm-relative flow is northeasterly. These flow regimes depict the trowal airstream and the dry descending airstream behind the northern stream trough, respectively. These analyses demonstrate that, with time, the precipitation was increasingly forced directly by the isentropic upglide of relatively warm, moist air in the trowal airstream.

5. Summary

The cause of the snowfall in this case seems simple enough. Isentropic upglide was present across the Deep South, along and just to the east of the Mississippi River valley. The focus of this ascent was the trowal and the airstream found therewith. In addition, relative humidity in excess of 90% stretched from the Gulf of Mexico northward into southern Missouri. Figure 13 shows clearly the cross-isobaric flow with the trowal airstream that persists as far as southern Kansas, in spite of a westward curl around the northern rim of the cyclone. Moreover, we may gather from Fig. 14 that the lack of ascent (isobar-parallel flow) over and north of central Missouri effectively limits precipitation north of that region. Still, a more complete examination was needed to explain the lack of snowfall beyond the snowfield's northern edge.

Indeed, the case presented featured a sharp snowfall gradient on its northern periphery, while the southern boundary of the snowfall band was more broad and diffuse. In effect, the half bandwidth on the northern side was more narrow than that on the southern side. This kind of measure is important when determining the nature of a mesoscale snowband (e.g., Bennetts and Hoskins 1979), and the narrowness of the northern half in this case originally suggested the presence of moist symmetric instability with this event. Previously, investigators have examined cases of banding with more or less equal areas on either side of the axis of maximum snowfall (e.g., Moore and Blakely 1988; Moore and Lambert 1993; Nicosia and Grumm 1999).

This case of snowfall across southern Missouri during 14 March 1999 invited closer examination as a case of symmetric instability. The analyses presented here demonstrate that potential instability, which might lead to upright convection, and potential symmetric instability which, with appropriate forcing, would lead to saturation and slantwise convection, both existed over northern Arkansas and extreme southern Missouri. However, the absence of sufficient moisture and/or frontogenetical forcing for ascent over Missouri seriously curtailed convection there. This assertion is confirmed by the lack of surface stations reporting thunder as well as the lightning strikes (<5) identified by the National Lightning Detection Network being confined to a region well south of the northern edge of the snow field.

Additional analysis was undertaken to determine a cause for the strength of the northern snowfall gradient. Failing the existence of significant moist symmetric instability to generate the snowfall gradient, an examination of the frontogenesis field was produced. Frontogenetical forcing was identified with this event, along a sloping line that paralleled the warm front; yet the values were not as strong as one might expect and occasionally failed to agree with the calculated vertical motion field.

On the meso-α scale, the moisture field was controlled and shaped by the deformation zone north of the closed low in the southern stream. This deformation zone was created by the interaction of the southern stream system and the short-wave trough in the northern stream. More important, the speed of the northern stream short-wave trough was crucial in generating the deformation zone in this case. Because of its faster propagation speed toward the east, it arrived in a position to supply the closed southern system with ample dry descending air from aloft in the northerly flow upstream of the northern disturbance's trough axis. In short, the moisture gradient was strengthened by the same deformation process that contracted the potential temperature gradient. In the case of the latter, the attendant forcing for vertical motion would only have been important where there was sufficient moisture to generate deep cloudiness and precipitation. The extreme moisture gradient across southern Missouri prohibited such cloud and snow development over the northern half of the state.

Acknowledgments

This work was funded from a subaward (S00-19120) under a cooperative agreement between the University of Missouri and the University Corporation for Atmospheric Research (UCAR). The views expressed herein are those of the authors and do not necessarily reflect the views of NOAA, its subagencies, or UCAR. Additional support was received through a grant from the University of Missouri Research Council, Award URC 00-018. The authors thank Drs. James Moore and Scott Rochette; Messrs. Ron Przybylinski, Chris Halcomb, Lou Hull, and Philip Schumacher; and two other anonymous reviewers for their helpful discussions and comments.

REFERENCES

  • Auer, A. H., 1987: An aid to forecasting heavy snowfall episodes. Natl. Wea. Dig., 12 , 1114.

  • Barnes, S. L., 1973: Mesoscale objective map analysis using weighted time-series observations. NOAA Tech. Memo. ERL NSSL-62, National Severe Storms Laboratory, Norman, OK, 60 pp.

    • Search Google Scholar
    • Export Citation
  • Bennetts, D. A., , and Hoskins B. J. , 1979: Conditional symmetric instability—A possible explanation for frontal rainbands. Quart. J. Roy. Meteor. Soc., 105 , 945962.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bennetts, D. A., , and Sharp J. C. , 1982: The relevance of conditional symmetric instability to the prediction of mesoscale frontal rainbands. Quart. J. Roy. Meteor. Soc., 108 , 595602.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., 1993a: Principles of Kinematics and Dynamics. Synoptic–Dynamic Meteorology in Midlatitudes, Vol. I, Oxford University Press, 431 pp.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., . 1993b: Observations and Theory of Weather Systems. Synoptic–Dynamic Meteorology in Midlatitudes, Vol. II, Oxford University Press, 594 pp.

    • Search Google Scholar
    • Export Citation
  • Browne, R. F., , and Younkin R. J. , 1970: Some relationships between 850-millibar lows and heavy snow occurrences over the central and eastern United States. Mon. Wea. Rev., 98 , 399401.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Carlson, T. N., 1980: Airflow through midlatitude cyclones and the comma cloud pattern. Mon. Wea. Rev., 108 , 14981509.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Crocker, A. M., , Godson W. L. , , and Penner C. M. , 1947: Frontal contour charts. J. Meteor., 4 , 9599.

  • Emanuel, K. A., 1983: On assessing local CSI from atmospheric soundings. Mon. Wea. Rev., 111 , 20162033.

  • Emanuel, K. A., . 1985: Frontal circulations in the presence of small moist symmetric stability. J. Atmos. Sci., 42 , 10621071.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Godson, W. L., 1951: Synoptic properties of frontal surfaces. Quart. J. Roy. Meteor. Soc., 77 , 633653.

  • Keyser, D., , and Shapiro M. A. , 1986: A review of the structure and dynamics of upper-level frontal zones. Mon. Wea. Rev., 114 , 452499.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Koch, S. E., , DesJardins M. , , and Kocin P. J. , 1983: An interactive Barnes objective map analysis scheme to use with satellite and conventional data. J. Climate Appl. Meteor., 22 , 14871503.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martin, J. E., 1998: The structure and evolution of a continental winter cyclone. Part II: Frontal forcing for an extreme snow event. Mon. Wea. Rev., 126 , 329348.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martin, J. E., . 1999a: Quasigeostropic forcing of ascent in the occluded sector of cyclones and the trowal airstream. Mon. Wea. Rev., 127 , 7088.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Martin, J. E., . 1999b: The separate roles of geostropic vorticity and deformation in the midlatitude occlusion process. Mon. Wea. Rev., 127 , 24042418.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , and Blakely P. D. , 1988: The role of frontogenetical forcing and conditional symmetric instability in the Midwest snowstorm of 30–31 January 1982. Mon. Wea. Rev., 116 , 21552171.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , and VanKnowe G. , 1992: The effect of jet-streak curvature on kinematic fields. Mon. Wea. Rev., 120 , 24292441.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , and Lambert T. E. , 1993: The use of equivalent potential vorticity to diagnose regions of conditional symmetric instability. Wea. Forecasting, 8 , 301308.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Moore, J. T., , Czarnetzki A. C. , , and Market P. S. , 1998: Heavy precipitation associated with elevated thunderstorms formed in a convectively unstable layer aloft. Meteor. Appsl., 5 , 373384.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Murray, R., , and Daniels S. M. , 1953: Transverse flow at entrance and exit to jet streams. Quart. J. Roy. Meteor. Soc., 79 , 236241.

  • Namias, J., , and Clapp P. F. , 1949: Confluence theory of the high tropospheric jet stream. J. Meteor., 6 , 330336.

  • Nicosia, D. J., , and Grumm R. H. , 1999: Mesoscale band formation in three major northeastern United States snowstorms. Wea. Forecasting, 14 , 346368.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • O'Brien, J. J., 1970: Alternative solutions to the classical vertical velocity problem. J. Appl. Meteor., 9 , 197203.

  • Penner, C. M., 1955: A three-front model for synoptic analyses. Quart. J. Roy. Meteor. Soc., 81 , 8991.

  • Sanders, F., , and Bosart L. F. , 1985: Mesoscale structure in the megalopolitan snowstorm of 11–12 February 1983. Part I: Frontogenetical forcing and symmetric instability. J. Atmos. Sci., 42 , 10501061.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Saucier, W. J., 1955: Principles of Meteorological Analysis. University of Chicago Press, 438 pp.

  • Schultz, D. M., , and Schumacher P. N. , 1999: The use and misuse of conditional symmetric instability. Mon. Wea. Rev., 127 , 27092732.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Seltzer, M. A., , Passarelli R. E. , , and Emanuel K. A. , 1985: The possible role of symmetric instability in the formation of precipitation bands. J. Atmos. Sci., 42 , 22072219.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Steigerwald, H., 1986: Deformation zones and heavy precipitation. Natl. Wea. Dig., 11 , 1520.

  • Thorpe, A. J., , and Emanuel K. A. , 1985: Frontogenesis in the presence of small stability to slantwise convection. J. Atmos. Sci., 42 , 18091824.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Uccellini, L. W., 1976: Operational diagnostic applications of isentropic analysis. Natl. Wea. Dig., 1 , 412.

Fig. 1.
Fig. 1.

Storm total snowfall (in.), 13–14 Mar 1999, for MO, AR, and KS. Isopleths of snowfall of 5 (2) 15 (6), and 30.5 cm (12 in.) are analyzed

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 2.
Fig. 2.

Synoptic analysis valid at 0000 UTC 13 Mar 1999. (a) Standard station models of observed surface data are presented with objectively analyzed mean sea level pressure, which is contoured every 4 mb. (b) Analysis of observed 850-mb heights (30 gpm; solid), divergence (10−5 s−1; thin solid; shaded light gray <−2 and dark gray >+2), temperature (4°C; dashed), and ageostrophic winds (arrows; reference vector at lower left). Frontal zone location based upon the observed wind field (not shown), the isotherm pattern, and the ageostrophic wind field. (c) A 500-mb analysis of observed data; standard station models of observed data are presented with objectively analyzed geopotential height (solid) and 1000–500-mb thickness (dashed), both of which are contoured every 60 gpm. Thin lines denote cross-section line used in Figs. 7 and 8 (easternmost line) and in Fig. 10 (westernmost line). (d) Analysis of observed 300-mb heights (120 gpm; solid), divergence (10−5 s−1; thin solid; shaded light gray <−2 and dark gray >+2), isotachs (kt; dashed), and ageostrophic winds (arrows; reference vector at lower left)

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 3.
Fig. 3.

As in Fig. 2 but at 1200 UTC 13 Mar 1999. Cross-section lines are absent

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 4.
Fig. 4.

As in Fig. 2 but at 0000 UTC 14 Mar 1999. International Falls, MN (INL), to Biloxi, MS (BIX), cross-section line for Figs. 7 and 8 is shown again

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 5.
Fig. 5.

As in Fig. 2 but at 1200 UTC 14 Mar 1999. Cross-section lines are absent

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 6.
Fig. 6.

Plots of surface observations of sky cover and present weather at (a) 0000 (b) 0600, and (c) 1200 UTC 14 Mar 1999

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 7.
Fig. 7.

Cross-section analyses of rawinsonde data from INL to BIX, valid at 0000 UTC 14 Mar 1999. (a) Petterssen frontogenesis [10−1 K (100 km)−1 (3 h)−1; solid] and kinematic vertical velocity (μb s−1; dashed) are shown. (b) Shown are θe (K; bold solid), relative humidity (%; thin solid, shaded >60%, 80% is bold), and Mg (m s−1; dashed). Regions of conditional symmetric instability (CSI), potential symmetric instability (PSI), and potential instability (PI) are outlined in the cross section. In both, bold X near the surface denotes the extreme southeast Missouri boot-heel region. The bold line and snow symbol (**) in both frames denote the northern extent of the precipitation field

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 8.
Fig. 8.

Cross-section analyses of rawinsonde data from INL to BIX, valid at 1200 UTC 14 Mar 1999. (a) Petterssen frontogenesis [10−1 K (100 km)−1 (3 h)−1; solid] and kinematic vertical velocity (μb s−1; dashed) are shown. (b) Shown are θe (K; bold solid) and relative humidity (%; thin solid, shaded >60%). In both, bold X near the surface denotes the extreme southeast Missouri boot-heel region. The bold lines and snow symbol (**) in both frames denote the extent of the precipitation field

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 9.
Fig. 9.

Plan-view analyses of Petterssen frontogenesis [10−1 K (100 km)−1 (3 h)−1] at 700 mb valid at 0000 (solid) and 1200 UTC 14 Mar 1999 (dashed)

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 10.
Fig. 10.

Cross-section analysis from GGW to SIL, valid at 0000 UTC 14 Mar 1999. Potential temperature is analyzed every 2 K (solid), with relative humidity contoured at 20%, 40%, 70%, and 90%, and shaded light (40%–70%), medium (70%–90%), and dark (>90%) gray. Observed winds are plotted (kt), with half barbs (5 kt), full barbs (10 kt), and flags (50 kt) on each shaft. Bold lines at the base of the figure denote the extent of precipitation in the figure at this time

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 11.
Fig. 11.

Radar summary valid at (a) 0000 and (b) 1200 UTC 14 Mar 1999. (From National Climatic Data Center online archives)

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 12.
Fig. 12.

(a) Objectively analyzed streamlines of the observed wind, pressure (mb; dashed), and relative humidity (%; gray shades) with legend on left, valid on the 298-K surface at 1200 UTC 13 Mar 1999. (b) Resultant deformation of the observed wind (10−5 s−1; solid) valid on the 298-K surface at 1200 UTC 13 Mar 1999

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 13.
Fig. 13.

Objectively analyzed streamlines of storm-relative flow (C of uC = +3.42 m s−1 and υC = −6.69 m s−1), pressure (mb; dashed), and relative humidity (%; gray shades) with legend on left, valid on the 298-K surface at 0000 UTC 14 Mar 1999. (b) Resultant deformation of the observed wind (10−5 s−1; solid) valid on the 298-K surface at 0000 UTC 14 Mar 1999. (c) Isobars (every 50 mb; solid) depicting the 310-K θe surface. (d) Pressure advection (−V · ∇p) on the 298-K isentropic surface (μb s−1; thin solid, shaded <−2), and mixing ratio advection (−V · ∇r) on the 298-K isentropic surface (10−8 kg kg−1 s−1; dashed)

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 14.
Fig. 14.

(a) Objectively analyzed streamlines of storm-relative flow (C of uC = +13.42 m s−1 and υC = +2.83 m s−1), pressure (mb; dashed), and relative humidity (%; gray shades) with legend on left, valid on the 298-K surface at 1200 UTC 14 Mar 1999. (b) Resultant deformation of the observed wind (10−5 s−1; solid) valid on the 298-K surface at 1200 UTC 14 Mar 1999. (c) Pressure advection (−V · ∇p) on the 298-K isentopic surface (μb s−1; thin solid, shaded <−2), and mixing ratio advection (−V · ∇r) on the 298-K isentropic surface (10−8 kg kg−1 s−1; dashed)

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Fig. 15.
Fig. 15.

Cross-section analysis of storm-relative winds (using C of uC = +13.42 m s−1 and υC = +2.83 m s−1) shown plotted (short barb, 5 kt; long barb, 10 kt; flag, 50 kt) and contours of the plane-normal storm-relative wind (kt; solid) valid at 1200 UTC 14 Mar 1999. The bold lines and snow symbol (**) at the base of the figure denote the extent of the precipitation field

Citation: Weather and Forecasting 17, 4; 10.1175/1520-0434(2002)017<0723:FOASSG>2.0.CO;2

Table 1. 

Abbreviated METAR surface observations from Rolla, MO (VIH), on 14 Mar 1999

Table 1. 
Table 2. 

Abbreviated METAR surface observations from Springfield, MO (SGF), on 14 Mar 1999

Table 2. 
Save