---

The Formation of a
Forward-Tilting Cold Front with Multiple Cloud Bands
During Superstorm 1993

David M. Schultz
NOAA/National Severe Storms Laboratory
Norman, Oklahoma

W. James Steenburgh
NOAA Cooperative Institute for Regional Prediction and
Department of Meteorology, University of Utah
Salt Lake City, Utah

Submitted to Monthly Weather Review December 1997
Revised May 1998, June 1998
Accepted July 1998
Page Proofs Approved March 1999

Corresponding author address:
Dr. David M. Schultz
NOAA/National Severe Storms Laboratory
1313 Halley Circle
Norman, OK 73069
E-mail: david.schultz@noaa.gov

  Abstract

1  Introduction

Forward-tilting cold fronts¹ have been noted previously in the literature. Brunt (1934, 344-345) claimed that there is ``definite observational evidence'' of cold fronts having a ``nose'' (a forward protrusion of unspecified length) as high as 500 m above ground. Brunt also argued that a surface pressure trough would be found below the nose of the cold front (a prefrontal trough). Other early views of forward-tilting cold fronts can be found in Bjerknes (1930, Fig. 8) and Flower (1931). More recent observational (Hardy et al. 1973; Bedard and Sanders 1978; Morales 1981, Fig. 9; Locatelli et al. 1989) and numerical modeling (Kuo and Reed 1988; Mass and Schultz 1993; Steenburgh and Mass 1994; Colle and Mass 1995) work has presented examples of fronts that are forward-tilting with height. Also, several authors have suggested that some midlatitude cold fronts entering the Tropics first arrive aloft and later become evident at the surface (e.g., Forsdyke 1949, p. 40; Palmer 1951, p. 867; Trewartha 1966, 46-48).

The existence of prefrontal features such as troughs, wind-shift lines, and cloud bands has also been recognized previously (e.g., Bjerknes 1930, 5-7; Tepper 1950; Newton 1950; Garratt 1988; Hanstrum et al. 1990a,b; Bluestein 1993, 258-259; Charney and Fritsch 1996; Browning et al. 1997; Neiman et al. 1998; Bryan and Fritsch 1998; Sanders 1998). Sanders (1983), Colle and Mass (1995), Sanders and Doswell (1995), and Hutchinson and Bluestein (1998) presented observational evidence that troughs and wind-shift lines can precede cold fronts in the southern and central United States. In fact, as many as 60% of the cold fronts in the lee of the Rocky Mountains may be associated with prefrontal wind shifts, often identified as lee troughs or drylines (Hutchinson and Bluestein 1998). Furthermore, prefrontal pressure troughs associated with midlatitude cold fronts entering the Tropics have been noted by several authors (e.g., Palmer 1951, p. 867; Trewartha 1966, p. 46; Fermor 1971; Hastenrath 1985, p. 241).

Schultz et al. (1997) performed an observational case study of the midlatitude cold front associated with Superstorm 1993 (12-14 March; hereafter, abbreviated SS93) that moved equatorward into the Tropics along the eastern slopes of the Sierra Madre Mountains of Mexico and Central America. They found that the structure and evolution of the cold front in eastern Mexico was nonclassical in two important aspects: the cold-frontal structure appeared to be forward tilting, and prefrontal troughs/cloud bands were present. Figure 1presents evidence for these nonclassical features. First, a cross section through the surface cold front from the European Centre for Medium-Range Weather Forecasts (ECMWF) analysis at 1800 UTC 12 March 1993 indicates that the cold advection at 850-700 hPa occurred 200-300 km ahead of the cold advection at the surface (Fig. 1a; the location of cross section EF is shown in Figs. 6a,c). Second, the change in the lower-tropospheric temperature and wind profiles at Veracruz, Mexico (VER in Fig. 1c) between 1200 UTC 12 March and 0000 UTC 13 March suggests that, because of strong afternoon boundary-layer mixing and surface heating, the strongest northerlies and largest temperature decrease arrived at 925-850 hPa before arriving at the surface (Fig. 1b), thereby suggesting a forward tilt to the front. Later (1200 UTC 13 March), the temperature and strong winds were more well-mixed through the lowest 50-150 hPa (Fig. 1b). Finally, Schultz et al. (1997, Fig. 10) described multiple cloud bands that were associated with the cold front at different times during its evolution. In particular, the cold front, initially identified by a rope cloud (the primary cloud band; cloud band 1 in Fig. 1c), began with the pressure minimum, wind shift, and temperature decrease nearly coincident, characteristics similar to a classical cold front. By 1700 UTC 12 March, a second cloud band (the prefrontal cloud band; cloud band 2 in Fig. 1c) formed in advance of the primary cloud band and was associated with a prefrontal pressure trough.² Over the next five hours, the surface wind in between the two cloud bands increased in magnitude and became increasingly northerly, as if the interstitial air were being diluted by higher-momentum northerlies from aloft (Schultz et al. 1997, 23-24). By 2200 UTC 12 March, cloud band 2 became the leading edge of the surface cold front and cloud band 1 was dissipating (Fig. 1c).

The purpose of this paper is to address these nonclassical aspects of the cold front associated with SS93 (the forward tilt of the cold front and its associated cloud bands) as it moved equatorward along the eastern slopes of the Sierra Madre Mountains in Mexico. The observed data presented in Schultz et al. (1997), although suggestive, was often inadequate to provide additional details about the evolution of this case and, therefore, to ascertain more confidently its structure and dynamics. Consequently, a mesoscale-model simulation is used to provide a high-resolution four-dimensional dataset for analysis and diagnosis. The mesoscale model and its configuration for this study are described in section 2. The structure and evolution of the simulated cold front is presented in section 3. In sections 4 and 5, the model simulation is further studied through trajectory and frontogenesis diagnostics, respectively, in order to examine the formation and maintenance of the forward tilt to the cold-frontal structure and the evolution of the prefrontal and primary cloud bands. Finally, in section 6, we compare and contrast the SS93 cold front to other frontal structures in the literature.

2  Mesoscale-model description

3  Synoptic evolution

To illustrate the large-scale environment in eastern Mexico shortly after the initiation of the cold surge, surface, 850-hPa, and 500-hPa analyses for 12 h into the simulation (1200 UTC 12 March 1993) are presented in Figs. 3a-c. The simulated surface cyclone was located southeast of Brownsville, Texas (BRO in Fig. 1c) and the leading edge of the surface cold front⁴ was located just south of the Texas-Mexico border with 10 m s^-1-1 northerlies yielding cold advection over south-central Texas (Fig. 3a). Note that the cyclogenesis was occurring south of the primary baroclinic zone over the Gulf of Mexico and the thermal gradient south of the analyzed cold front, laid down by prior convection in the area, was characterized by warm advection (Fig. 3a). At this time, the lower-tropospheric cold front was rearward-sloping as the leading edge of strong cold advection at the surface (near BRO) was about 200 km farther equatorward than at 850 hPa (cf. Figs. 3a,b). The rearward tilt is further illustrated by cross section AB (Fig. 3d; the location is shown in Figs. 3a,c), oriented nearly perpendicular to the surface front. Cross section AB shows the leading edge of the cold advection (identified by the thick solid line labeled 0) tilting rearward from the surface to 850 hPa, roughly coincident with the baroclinicity and wind shift from southerly-southwesterly to northerly-northeasterly. At 500 hPa (Fig. 3c), a region of strong winds and cold advection was found over northeastern Mexico in association with the upper-tropospheric disturbance responsible for the initial SS93 cyclogenesis over the Gulf of Mexico [associated with the so-called potential-vorticity anomaly C, as illustrated in Bosart et al. (1996, Fig. 3) and Dickinson et al. (1997, Fig. 3)]. Cross section CD, taken across the 500-hPa baroclinic zone (Fig. 4; the location is shown in Figs. 3a,c), illustrates that the baroclinicity extended from the tropopause through the midtroposphere down to about 800 hPa. Hereafter, we will refer to this baroclinic zone as the mid- to upper-tropospheric baroclinic zone. At this time, the mid- to upper-tropospheric baroclinic zone was further equatorward than the surface baroclinic zone, which, in cross section CD (Fig. 4), was trapped against the northeast slopes of the Sierra Madre Mountains (see also, the surface potential temperature in Fig. 3a). In cross section AB (Fig. 3d), this mid- to upper-tropospheric baroclinic zone was associated with cold advection at 700-500 hPa, which preceded that at the surface by about 200 km. Low relative-humidity air (indicative of a recent history of subsidence associated with the mid- to upper-tropospheric baroclinic zone, a point to be illustrated further in section 4) was found on the equatorward side of the mid- to upper-tropospheric baroclinic zone above 800 hPa (Fig. 3d). In general, these simulated analyses from the PSU-NCAR MM5 agree well with manual analyses for this same time (1200 UTC 12 March 1993) in Kocin et al. (1995, their Figs. 3d-f).

By 15 h (Fig. 5), cross section AB shows that the region of cold advection and drier air associated with the mid- to upper-tropospheric baroclinic zone had descended to about 850 hPa. The nascent interaction of the mid- to upper-tropospheric baroclinic zone with the surface cold front resulted in an apparent forward-tilting cold-frontal structure from the surface to 700 hPa, with 850-700-hPa cold advection located equatorward of the surface frontal position.

From 12 to 18 h (1200 UTC to 1800 UTC 12 March 1993), the simulated surface cyclone deepened 8 hPa and moved east-northeastward (cf. Figs. 3a and 6a), and by 18 h (Fig. 6a), resembled the manually produced surface analysis in Schultz et al. (1997, Fig. 13). At this time, the leading edge of the cold air at the surface was located near Tampico, Mexico (TAM in Fig. 1c) with the leading edge of cold advection at 850-700-hPa overspreading that at the surface by 200-300 km (cf. Figs. 6a,b). The horizontal potential temperature gradient within this region of forward tilt was approximately 1-2 K (100 km)^-1. Within this larger region of forward-tilting cold advection (hereafter referred to as the large-scale forward tilt of the cold front), a region of locally enhanced horizontal potential temperature gradient of 4-5 K (100 km)^-1 was embedded (hereafter, the small-scale forward tilt of the cold front). This small-scale forward tilt of strong cold advection (as evinced by the -1.5 and -3 K h^-1 contours) occurred from the surface to 850 hPa, with a forward tilt over this depth of about 100 km (Fig. 6d). Whereas the larger-scale forward tilt was due to the interaction between the mid- to upper-tropospheric baroclinic zone and the surface cold front, the embedded smaller-scale forward tilt was due to frontogenetical tilting by ascent from the surface cold front of the overlying baroclinicity (discussed further in section 5). Although much of the region immediately above the leading edge of the surface cold front became nearly isentropic from the surface to 850 hPa within this region of small-scale forward-tilt (individual isentropes in this region were nearly vertical through a depth of about 200 hPa), absolute instability did not appear to develop in the model simulation (Fig. 6d; recall footnote 1). Indeed, such superadiabatic layers are removed by the model parameterization (Grell et al. 1994, 44). As evidence of a relationship between the forward tilt to the cold front and the cloud bands discussed by Schultz et al. (1997), the locations of the observed primary cloud band (cloud band 1) and the prefrontal cloud band (cloud band 2) were similar to the leading edge of strong cold advection at the surface and 850 hPa, respectively (cf. Figs. 1c and Figs. 6a,b). This relationship between the cloud bands and the forward-tilting frontal structure will be discussed further in section 5. The 500-hPa trough and baroclinicity moved eastward (Fig. 6c) such that the most equatorward extent of the 500-hPa cold advection was 100 km poleward of the 850-hPa cold advection (Fig. 6b) and 250 km equatorward of the surface cold advection (see also Fig. 6d). In fact, the 850-hPa potential-temperature pattern was, in many ways, more similar to the 500-hPa potential-temperature pattern than the surface potential-temperature pattern, indicating that the baroclinicity at 850 hPa may have been vertically continuous with the mid- to upper-tropospheric baroclinic zone (cf. Figs. 6b,c). This hypothesis is explored further in section 4.

At 21 h, the front continued to be tilted forward on both the large and small scales as shown in cross section EJ (Fig. 7; the location of which is shown in Fig. 6a). The prefrontal surface northerlies extended about 200 km ahead of the leading edge of the surface cold advection. It may be that the mixing down of air with higher northerly momentum from aloft in the region of the small-scale forward tilt is enhanced in this case because of its arrival into Mexico during the late afternoon, a time when mixing in the planetary boundary layer would be more vigorous. A similar process was discussed by Steenburgh and Mass (1994, p. 2753).

By 24 h (0000 UTC 13 March 1993; Fig. 8), the leading edge of the surface baroclinicity, preceded by surface northerlies and the prefrontal trough, was located just poleward of the Isthmus of Tehuantepec (Fig. 8a). The prefrontal onshore surface northerlies helped push the sea-breeze front inland along the north coast of the Isthmus of Tehuantepec (Figs. 8a,d). At 850 hPa, the potential temperature gradient along the leading edge had weakened (Fig. 8b), whereas at 500 hPa, the region of strongest cold advection had moved into the west-central Gulf of Mexico (Fig. 8c). The structure of the front is illustrated in cross section GH (Fig. 8d). Amplification of mountain waves over the isthmus occurred, disrupting the frontal structure at 700-900 hPa (Fig. 8d). Nevertheless, the leading edge of the cold advection tended to tilt rearward with height, returning to a more classical cold-frontal structure (supported by trajectories, presented in section 4).

In order to validate the simulated structure of the cold front and to illustrate the surface weather patterns associated with the passage of a forward-tilting front, observed and simulated time series at VER are presented in Fig. 9. The observed time series show a minimum in altimeter setting (i.e., a pressure check) indicating the passage of a pressure trough associated with cloud band 2 at 1845 UTC and an inflection in altimeter setting (superimposed upon a larger-scale pressure rise) indicating the passage of a pressure trough associated with cloud band 1 at 2130 UTC (Fig. 9a; Schultz et al. 1997, p. 22). The model-generated time series of sea level pressure at VER (Fig. 9a) exhibits a minimum at 19 h that represented the arrival of the prefrontal trough associated with the cloud band 2. Also associated with the passage of this trough in the model were decreasing 850-hPa temperature and increasing 850-hPa wind speed (Figs. 9b,c), providing additional evidence that the passage of cloud band 2 was followed by the onset of 850-hPa cold advection. The model-generated time series at VER exhibit the passage of a second feature after 22 h when the sea level pressure began to rise at an increased rate exceeding 1 hPa h^-1, the surface temperature began to decrease sharply, and the surface wind speed increased to over 10-15 m s^-1-1 (Figs. 9a-c). These weather features represent the arrival at VER of the leading edge of the surface cold front (and cloud band 1) in the model. Except for a 1-2 hPa error in the pressure trough associated with the passage of cloud band 1 and a near-constant displacement between the observed altimeter and model pressure traces, the observed and modeled pressure traces between 1500 UTC 12 March and 0200 UTC 13 March (15 and 26 h) were broadly similar (Fig. 9a). The surface temperature and wind evolutions also seem to be well-simulated, but the intensity of the peak temperatures and winds were underforecast (Figs. 9b,c).

Whereas the model-generated VER time series (Fig. 9) suggest that the previously discussed weather features associated with the passage of cloud bands 1 and 2 were reasonably simulated, the model-generated cloud-water fields (not shown) do not indicate the formation of these multiple cloud bands. As will be seen in section 5, however, regions of mesoscale ascent do occur in the simulation in nearly the correct locations as the expected cloud bands. Consequently, the appearance of these regions of ascent indicates that the dynamics to generate the cloud bands were present in the model.

4  Trajectory analysis

Nine-hour trajectories beginning at 3 h (0300 UTC 12 March 1993) and ending on s = 0.995 (approximately 40 m above ground level) at 12 h (1200 UTC 12 March 1993) (Fig. 10a; trajectory ending locations are shown in Fig. 3d) show two airstreams: those poleward of the surface cold front moving southwestward (trajectories A1-A11) and those equatorward of the front moving northwestward (trajectories A14-A23). Trajectories A12-A13 started in the prefrontal air and were ingested into the postfrontal air in a manner similar to that described by Schultz and Mass (1993, p. 929). Trajectories ending on s = 0.885 (approximately 850-914 hPa; note that trajectories labeled Ann and Bnn end at the same horizontal position) (Fig. 10b) show that the boundary between the two airstreams occurred between trajectories B11 and B13, with trajectory B12 having been ingested into the frontal zone, consistent with the cross section at this time (Fig. 3d). Therefore, at 12 h, trajectories in the lower troposphere indicate that the airstream boundary comprising the front, consistent with the thermal structure from the cross section (Fig. 3d), tilted rearward with height.

By 21 h (2100 UTC 12 March 1993), the frontal interaction had occurred with the cold front developing its large-scale and small-scale forward-tilting structure (Fig. 7). For the 9-h trajectories beginning at 12 h and ending on s = 0.995 at 21 h (trajectory ending locations are shown in Fig. 7), trajectories C1-C8 originated and remained within the baroclinicity associated with the cold front (Fig. 11a). Trajectory C9 (see also Table 1) originated in the warm poleward flow ahead of the cold front, but between 17 and 18 h, reached its most poleward position. After 18 h and the passage of the prefrontal trough, trajectory C9 turned equatorward, embedded in the prefrontal northerlies near the surface. After 19 h, the potential temperature of trajectory C9 began to drop (Table 1), indicating ingestion by the frontal zone. In contrast, trajectories C10-C13 appeared to reach their most poleward position around 19 h before turning and heading equatorward (Fig. 11a) without experiencing a decrease in potential temperature (not shown), indicating that trajectories C10-C13 were not ingested by the surface cold front (see also Fig. 7, where trajectories C10-C13 are found ahead of the surface position of the front). Therefore, the onset of equatorward motion of trajectories C9-C13 illustrated the passage of the prefrontal trough. Trajectories C14-C17 remained equatorward of the prefrontal trough.

Further aloft, 9-h trajectories D1-D17 ending on s = 0.885 at 21 h are displayed in Fig. 11b (trajectory ending locations are shown in Fig. 7). Trajectories D1-D9 traveled within the baroclinicity behind the equatorward-moving cold front over southern Texas, as at 12 h, whereas trajectories D14-D17 were found in the warm air equatorward of the front. Trajectories D10 and D11, however, both originated from the west in the baroclinicity associated with the mid- to upper-tropospheric baroclinic zone. The approximate positions of these two trajectories at 12 h are shown in Fig. 4. These two trajectories descended down the east side of the Sierra Madre Mountains and ended up in the forward-tilting portion of the cold front (Fig. 7). For example, trajectory D11 descended from 747 hPa to 901 hPa in 3 h (an average vertical velocity of 14 mb s^-1). In contrast, trajectory D12, farther equatorward and not in the midtropospheric baroclinicity, underwent comparatively less descent (17 hPa compared to 154 hPa) over the same 3-h period. These descending trajectories help confirm that the forward-tilting front was formed by the interaction of the baroclinicity associated with the mid- to upper-tropospheric baroclinic zone arriving across the Sierra Madre Mountains and the surface cold front.

5  Frontogenesis analysis

The sum of the horizontal frontogenesis terms 1 and 2 in (2), F₁₊₂, also can be expressed in the form developed by Petterssen (1936):

As illustrated previously, the formation of the large-scale forward-tilting structure was initiated by the interaction of the surface cold front with the mid- to upper-tropospheric baroclinic zone. In this section, the tilting frontogenesis demonstrates how enhancements of the mid-tropospheric horizontal potential temperature gradient can occur in association with the lower- to mid-tropospheric ascent along the leading edge of the surface front in order to support the small-scale forward-tilted structure.

At 12 h (Fig. 12a; the location of cross section AB is found in Fig. 3a), much of the cold-frontal region near the surface was associated with deformation/divergence frontogenesis (thick solid lines). Tilting (medium solid lines) contributed to frontogenesis at the leading edge of the front from 950-550 hPa, along the equatorward side of the ascent maximum (shaded; presumably associated, in part, with a thermally direct secondary circulation). By 15 h (Fig. 12b), the deformation/divergence frontogenesis along the surface cold front reached its maximum magnitude. Also, the cross-frontal scales of the ascent at the leading edge of the front and, correspondingly, the regions of tilting frontogenesis have contracted. The arrival of the mid- to upper-tropospheric baroclinic zone into the plane of the cross section is indicated by weak prefrontal ascent and tilting frontogenesis at 850-550 hPa near the leading edge of the cold advection (Fig. 12b). Also, the depth of the circulation associated with cloud band 1 decreased. These changes to the cold-frontal structure result in the regions of strongest horizontal potential temperature gradient (i.e., strongly sloping isentropes when viewed in cross section) colocated with the regions of ascent above the surface cold front and along the mid- to upper-tropospheric baroclinic zone, thereby indicating the importance of the tilting frontogenesis to maintaining and enhancing the small-scale forward tilt of the cold front. At 18 h (Fig. 12c; note the change in horizontal scale from Figs. 12a,b; the location of cross section EF is shown in Fig. 6a), the front became tilted forward even further on the small scale. Two regions of ascent were apparent: one near the leading edge of the surface front (presumably associated with cloud band 1) and one at 850-750 hPa (presumably associated with cloud band 2). The tilting frontogenesis, likewise, evolved from a single large maximum to two regions at the leading edges of the ascent regions. Note that descent overtop ascent near the surface frontal position, implies 850-900-hPa horizontal divergence. Divergence results in frontolysis [the second term on the right-hand side of (3)] in this area (not shown), weakening the horizontal potential temperature gradient, and the secondary circulation and forcing for ascent for cloud band 1. This weakening of the horizontal potential temperature gradient and dissipation of cloud band 1 led to the dominance of cloud band 2, in agreement with the observations (Fig. 1c).

6  Discussion

The forward-tilting cold-frontal structure in the present case is unusual, but not unique; similar structures have been reported previously in the literature (e.g., section 1). Similar structures, however, do not necessarily imply similar evolutions or dynamics. Therefore, the results from the SS93 case are compared to other forward-tilting cold fronts in section 6a, then placed in a general context for frontal interaction in section 6b.

6a  Relationship to other forward-tilting cold fronts

In contrast, the forward-tilting fronts in Shapiro (1984) and Colle and Mass (1995, Fig. 17) appear to be affected by friction, as the nose occurs only a few hundred meters above the Earth's surface. Also, there does not appear to be a mid- to upper-tropospheric baroclinic zone involved, and so these cases are not likely to be directly comparable to the SS93 case. For a case of a forward-tilting cold front over the eastern United States, Bedard and Sanders (1978) claimed that the cold front moved faster over a prefrontal inversion than near the surface. This explanation also has been offered for Hardy et al.'s (1973) case of a forward-tilting front (R. Reed 1989, personal communication). Forsdyke (1949, p. 40) argued that subsidence (and concomitant adiabatic warming) in the cold air and diabatic surface processes acting on midlatitude cold fronts entering the Tropics weaken and slow the equatorward advance of the front at the surface compared to that aloft. Although surface friction, elevated inversions, and diabatic processes were present in the SS93 case, the trajectory and frontogenesis diagnostics used in this study help to affirm that these processes do not appear to play a primary role in the formation of the forward tilt to the cold-frontal structure in the SS93 case. Indeed, an adiabatic simulation of the SS93 cold front (not shown) also produced a forward-tilting front with a structure and evolution comparable to that in the present simulation, indicating that the contribution of diabatic processes to the structure and evolution of this forward-tilting cold front were of secondary importance. Whereas cold fronts can evolve towards a forward-tilting state in a variety of ways, this paper describes only one way in which this can happen (through the interaction of mid- to upper-tropospheric- and lower-level baroclinic zones). Evidence for alternative mechanisms that produce forward-tilting fronts has not been as forthcoming in the literature.

6b  A general context for frontal interaction

The mid- to upper-tropospheric baroclinic zone originating over the Sierra Madre Mountains in the SS93 case appears to resemble what has been termed a cold front aloft (e.g., Hobbs et al. 1990; Martin et al. 1990), where a mid- to upper-tropospheric baroclinic zone, associated with cold advection and frontogenesis (the cold front aloft), moves over the Rocky Mountains in association with a 500-hPa trough. Hobbs et al. (1990, p. 615) suggest that quasigeostrophic ascent associated with the cold front aloft in a region of potential instability downstream of the 500-hPa trough and in advance (equatorward) of a surface cold front (see, e.g., Hobbs et al. 1990, their Fig. 16) can lead to deep moist convection and severe weather [analogous to the vertically coupled jet-front system of Shapiro (1982, Fig. 23)]. In SS93, the mid- to upper-tropospheric baroclinic zone arrived in eastern Mexico with the passage of the surface cold front underneath, a situation unlikely to spawn deep moist convection because the stable lower troposphere and large-scale subsidence in the midtroposphere in advance of the front at the surface (e.g., Fig. 6d) would have inhibited strong upward vertical motion. As a result, only the modest cloud bands 1 and 2 occurred [analogous to the vertically uncoupled jet-front system of Shapiro (1982, Fig. 22)], rather than deep moist convection. This evolution may explain the relative lack of precipitation in eastern Mexico associated with the passage of this cold front (Schultz et al. 1997, Figs. 11b-d and 18), when compared to other Mexican cold-frontal cases (T. Bals-Elsholz 1997, personal communication). Therefore, different frontal structures and circulations may result, depending on the timing of the surface and mid- to upper-level baroclinic zones.

One may also envision scenarios where either the surface or the mid- to upper-level baroclinic zone is absent or interaction does not occur. For example, Neiman et al. (1998) analyze a case in which a mid- to upper-tropospheric baroclinic zone (what they term ``the Pacific front'') arrived over the United States Central Plains, but there was no surface frontal structure (only a near-surface inversion) with which the mid- to upper-tropospheric baroclinic zone could interact. Elevated convection resulted in a squall line over the southeastern United States along the leading edge of the baroclinic zone aloft. Alternatively, the surface cold front could occur without an interaction with a mid- to upper-tropospheric baroclinic zone; this scenario has been discussed by, for example, Colle and Mass (1995, Fig. 17).

Keyser (1986, 253) advocated further research on the interaction between surface and mid- to upper-tropospheric baroclinic zones and this investigation represents a contribution to that effort. We envision a spectrum of possible interactions between surface frontal features and mobile mid- to upper-tropospheric baroclinic zones, ranging from a mid- to upper-tropospheric baroclinic zone with no surface front (Neiman et al. 1998) to a surface front with no baroclinic zone aloft (Colle and Mass 1995). In between, the possibility exists for interactions between upper- and lower-level phenomena (e.g., cold fronts aloft, the SS93 cold front) in different environments which may inhibit or favor deep moist convection and coupling between upper- and lower-tropospheric circulations (e.g., Shapiro 1982, Figs. 22-23). In these instances, the mountains initially maintain the separation between the surface frontal features in the Central Plains and eastern Mexico and the eastward-moving mid- to upper-tropospheric baroclinic zone until interaction. Therefore, depending on the prior weather systems over the Central Plains and eastern Mexico, a variety of structural evolutions and weather patterns can result when mid- to upper-tropospheric baroclinic zones arrive from the west over the Rocky and Sierra Madre Mountains.

  Acknowledgments

We are grateful to ECMWF, the Data Support Section of the Scientific Computing Division of NCAR, and the University of Utah Center for High Performance Computing for providing data and a portion of the computing resources used in this study. Research on this project was partially conducted while the first author was a National Research Council Postdoctoral Research Associate at the National Severe Storms Laboratory. This research was partially supported by the National Science Foundation through grant ATM-9634191 and NOAA grant OGP-526404 to the second author through the University of Utah.

Bedard, A. J., Jr., and M. J. Sanders, Jr., 1978: Thunderstorm-related wind shear detected at Dulles international airport using a Doppler acoustic/microwave radar, a monostatic sounder and arrays of surface sensors. Proc. Conf. on Weather Forecasting and Analysis and Aviation Meteorology, Silver Spring, MD, Amer. Meteor. Soc., 347-352.

Berry, F. A., Jr., E. Bollay, and N. R. Beers, 1945: Handbook of Meteorology. McGraw-Hill, 1068 pp.

Bjerknes, J., 1930: Practical examples of polar-front analysis over the British Isles in 1925-6. Geophys. Mem., 5(10), 1-21 and 28 pp. of figs.

Bluestein, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Volume II: Observations and Theory of Weather Systems. Oxford University Press, 594 pp.

Bond, N. A., and R. G. Fleagle, 1985: Structure of a cold front over the ocean. Quart. J. Roy. Meteor. Soc., 111, 739-759.

Bond, N. A., and M. A. Shapiro, 1991: Research aircraft observations of the mesoscale and microscale structure of a cold front over the eastern Pacific Ocean. Mon. Wea. Rev., 119, 3080-3094.

Bosart, L. F., G. J. Hakim, K. R. Tyle, M. A. Bedrick, W. E. Bracken, M. J. Dickinson, and D. M. Schultz, 1996: Large-scale antecedent conditions associated with the 12-14 March 1993 cyclone (``Superstorm `93") over eastern North America. Mon. Wea. Rev., 124, 1865-1891.

Browning, K. A., 1990: Organization of clouds and precipitation in extratropical cyclones. Extratropical Cyclones, The Erik Palmén Memorial Volume, C. W. Newton and E. O. Holopainen, Eds., Amer. Meteor. Soc., 129-153.

Browning, K. A., and T. W. Harrold, 1970: Air motion and precipitation growth at a cold front. Quart. J. Roy. Meteor. Soc., 96, 369-389.

Browning, K. A., N. M. Roberts, and A. J. Illingworth, 1997: Mesoscale analysis of the activation of a cold front during cyclogenesis. Quart. J. Roy. Meteor. Soc., 123, 2349-2375.

Brunt, D., 1934: Physical and Dynamical Meteorology. Cambridge University Press, 411 pp.

Bryan, G. H., and J. M. Fritsch, 1998: Discrete frontal propagation induced by convection. Preprints, 16th Conf. on Weather Analysis and Forecasting, Phoenix, AZ, Amer. Meteor. Soc., 152-154.

Charney, J. J., and J. M. Fritsch, 1996: The formation of mesohighs in non-convective environments. Preprints, 15th Conf. on Weather Analysis and Forecasting, Norfolk, VA, Amer. Meteor. Soc., 242-244.

Colle, B. A., and C. F. Mass, 1995: The structure and evolution of cold surges east of the Rocky Mountains. Mon. Wea. Rev., 123, 2577-2610.

Dickinson, M. J., L. F. Bosart, W. E. Bracken, G. J. Hakim, D. M. Schultz, M. A. Bedrick, and K. R. Tyle, 1997: The March 1993 Superstorm cyclogenesis: Incipient phase synoptic- and convective-scale flow interaction and model performance. Mon. Wea. Rev., 125, 3041-3072.

Dudhia, J., 1989: Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 3077-3107.

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, 1493-1513.

Fermor, J. H., 1971: The weather during northers at Kingston, Jamaica. J. Trop. Geogr., 32, 31-37.

Flower, W. D., 1931: An analysis of the cold front over Egypt on March 7, 1929. Quart. J. Roy. Meteor. Soc., 57, 275-287.

Forsdyke, A. G., 1949: Weather forecasting in tropical regions. Geophys. Mem. No. 82, 47 pp.

Garratt, J. R., 1988: Summertime cold fronts in southeast Australia-Behavior and low-level structure of main frontal types. Mon. Wea. Rev., 116, 636-649.

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, 138 pp. [Available from NCAR, P.O. Box 3000, Boulder, CO 80307-3000.]

Hardy, K. R., R. J. Reed, and G. K. Mather, 1973: Observation of Kelvin-Helmholtz billows and their mesoscale environment by radar, instrumented aircraft, and a dense radiosonde network. Quart. J. Roy. Meteor. Soc., 99, 279-293.

Hanstrum, B. N., K. J. Wilson, and S. L. Barrell, 1990a: Prefrontal troughs over Southern Australia. Part I: A climatology. Wea. Forecasting, 5, 22-31.

Hanstrum, B. N., K. J. Wilson, and S. L. Barrell, 1990b: Prefrontal troughs over Southern Australia. Part II: A case study of frontogenesis. Wea. Forecasting, 5, 32-46.

Hastenrath, S., 1985: Climate and Circulation of the Tropics. D. Reidel Publishing Co., 455 pp.

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, 613-626.

Hutchinson, T. A., and H. B. Bluestein, 1998: Prefrontal wind-shift lines in the Plains of the United States. Mon. Wea. Rev., 126, 141-166.

Janes, S. A., H. W. Brandli, and J. W. Orndorff, 1976: ``The blue line'' depicted on satellite imagery. Mon. Wea. Rev., 104, 1178-1181.

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. 24, Amer. Meteor. Soc., 165-170.

Keyser, D., 1986: Atmospheric fronts: An observational perspective. Mesoscale Meteorology and Forecasting, P. S. Ray, Ed., Amer. Meteor. Soc., 216-258.

Klemp, J. B., and D. R. Durran, 1983: An upper boundary condition permitting internal gravity wave radiation in numerical mesoscale models. Mon. Wea. Rev., 111, 430-444.

Knight, D. J., and P. V. Hobbs, 1988: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. Part XV: A numerical modeling study of frontogenesis and cold-frontal rainbands. J. Atmos. Sci., 45, 915-930.

Kocin, P. J., P. N. Schumacher, R. F. Morales Jr., and L. W. Uccellini, 1995: Overview of the 12-14 March 1993 Superstorm. Bull. Amer. Meteor. Soc., 76, 165-182.

Kuo, Y.-H., and R. J. Reed, 1988: Numerical simulation of an explosively deepening cyclone in the eastern Pacific. Mon. Wea. Rev., 116, 2081-2105.

Locatelli, J. D., J. M. Sienkiewicz, and P. V. Hobbs, 1989: Organization and structure of clouds and precipitation on the mid-Atlantic coast of the United States. Part I: Synoptic evolution of a frontal system from the Rockies to the Atlantic Coast. J. Atmos. Sci., 46, 1327-1348.

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, 195-217.

Mass, C. F., and D. M. Schultz, 1993: The structure and evolution of a simulated midlatitude cyclone over land. Mon. Wea. Rev., 121, 889-917.

Miller, J. E., 1948: On the concept of frontogenesis. J. Meteor., 5, 169-171.

Morales, H. C., 1981: A case study of a dust storm weather situation in the Sudan in April 1973. Pure Appl. Geophys., 119, 658-676.

Neiman, P. J., P. T. May, B. B. Stankov, and M. A. Shapiro, 1991: Radio acoustic sounding system observations of an arctic front. J. Appl. Meteor., 30, 881-892.

Neiman, P. J., M. A. Shapiro, F. M. Ralph, B. F. Smull, and D. Johnson, 1998: An observational study of fronts and frontal mergers during the phasing of arctic and Pacific-polar synoptic-scale waves over the continental United States. Mon. Wea. Rev., 125, in press.

Newton, C. W., 1950: Structure and mechanism of the prefrontal squall line. J. Meteor., 7, 210-222.

Newton, C. W., 1963: Dynamics of severe convective storms. Severe Local Storms. Meteor. Monogr., No. 5(27), Amer. Meteor. Soc., 33-58.

Palmer, C. E., 1951: Tropical meteorology. Compendium of Meteorology, Amer. Meteor. Soc., 859-880.

Petterssen, S., 1936: Contribution to the theory of frontogenesis. Geophys. Publ., 11(6), 1-27.

Sanders, F., 1955: An investigation of the structure and dynamics of an intense surface frontal zone. J. Meteor., 12, 542-552.

Sanders, F., 1983: Observations of fronts. Mesoscale Meteorology-Theories, Observations and Models, D. K. Lilly and T. Gal-Chen, Eds., Reidel, 175-203.

Sanders, F., 1998: A proposed method of surface map analysis. Preprints, 16th Conf. on Weather Analysis and Forecasting, Phoenix, AZ, Amer. Meteor. Soc., 214-217.

Sanders, F., and C. A. Doswell III, 1995: A case for detailed surface analysis. Bull. Amer. Meteor. Soc., 76, 505-521.

Schultz, D. M., and C. F. Mass, 1993: The occlusion process in a midlatitude cyclone over land. Mon. Wea. Rev., 121, 918-940.

Schultz, D. M., W. E. Bracken, L. F. Bosart, G. J. Hakim, M. A. Bedrick, M. J. Dickinson, and K. R. Tyle, 1997: The 1993 Superstorm cold surge: Frontal structure, gap flow, and tropical impact. Mon. Wea. Rev., 125, 5-39; Corrigendum, 125, 662.

Seaman, N. L., 1987: Program TRAJEC: Documentation and user's guide. PSU-NWPLIB-0010-87, 95 pp. [Available from Department of Meteorology, The Pennsylvania State University, University Park, PA 16802.]

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

Shapiro, M. A., 1984: Meteorological tower measurements of a surface cold front. Mon. Wea. Rev., 112, 1634-1639.

Shapiro, M. A., and D. Keyser, 1990: Fronts, jet streams and the tropopause. Extratropical Cyclones, The Erik Palmén Memorial Volume, C. W. Newton and E. O. Holopainen, Eds., Amer. Meteor. Soc., 167-191.

Smith, R. B., 1982: A differential advection model of orographic rain. Mon. Wea. Rev., 110, 306-309.

Steenburgh, W. J., and C. F. Mass, 1994: The structure and evolution of a simulated Rocky Mountain lee trough. Mon. Wea. Rev., 122, 2740-2761.

Steenburgh, W. J., D. M. Schultz, and B. A. Colle, 1998: The structure and dynamics of gap outflow over the Gulf of Tehuantepec, Mexico. Mon. Wea. Rev., 126, in press.

Tepper, M., 1950: A proposed mechanism of squall lines: The pressure jump line. J. Meteor., 7, 21-29.

Trewartha, G. T., 1966: The Earth's Problem Climates. Univ. of Wisconsin Press, 334 pp.

Zhang, D., 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, 1594-1609.

  Table

  Table Caption

  Figure Captions

Footnotes:

¹ In this paper, the term forward-tilting cold front is used to describe a cold front where the onset of cold advection in the lower- to midtroposphere precedes that at the surface. A possible interpretation of this definition, and one that appeared to be in favor among meteorologists during the 1940s and 1950s (see, e.g., Berry et al. 1945, p. 653; Newton 1963, p. 36), is that a forward-tilting front may be associated with regions of the atmosphere that are absolutely unstable (i.e., ¶q/¶z < 0), but this scenario is not validated by the case studies in the following discussion (section 1) and later in this paper (e.g., Fig. 6d and section 6).

² Cloud bands 3-5 are not discussed in this paper. The dynamics associated with cloud band 5 over the Gulf of Tehuantepec is examined by Steenburgh et al. (1998).

³ Specifically, the half-s levels were located at s=0.995, 0.985, 0.970, 0.945, 0.915, 0.885, 0.855, 0.825, 0.795, 0.765, 0.735, 0.705, 0.675, 0.645, 0.615, 0.585, 0.550, 0.510, 0.470, 0.430, 0.390, 0.350, 0.310, 0.270, 0.230, 0.190, 0.150, 0.110, 0.070, 0.025.

⁴ To distinguish between the leading edge of the surface cold front and the prefrontal pressure trough, the cold front is defined in this paper as the leading edge of strong northerlies and strong cold advection, consistent with the definition employed by Colle and Mass (1995, p. 2577) in their study of cold surges east of the Rocky Mountains. Because a purpose of this paper is to distinguish the prefrontal trough from the cold front, the definition employed presently is not consistent with the more liberal definition of Steenburgh et al. (1998), who placed the cold front coincident with the pressure trough and northerly wind shift.

Return to Current Research Interests.
Return to David Schultz's homepage.