Finescale Radar Observations of the 22 May 2002 Dryline during the International H2O Project (IHOP)

Christopher C. Weiss Department of Geosciences, Texas Tech University, Lubbock, Texas

Search for other papers by Christopher C. Weiss in
Current site
Google Scholar
PubMed
Close
,
Howard B. Bluestein School of Meteorology, University of Oklahoma, Norman, Oklahoma

Search for other papers by Howard B. Bluestein in
Current site
Google Scholar
PubMed
Close
, and
Andrew L. Pazmany ProSensing, Inc., Amherst, Massachussetts

Search for other papers by Andrew L. Pazmany in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The dryline has long been associated with the development of severe thunderstorms in the southern plains during the spring and early summer months. The propagation and structure of the dryline are closely tied to surface processes that are neither well understood nor well resolved with current observational capabilities. As a result, there are often large errors in forecasts of dryline position and structure.

Improvements in radar technology have allowed for better observations of the dryline in recent years. Here, very finescale radar observations taken with the University of Massachusetts—Amherst (UMass) mobile W-band radar during an International H2O Project (IHOP) double-dryline event on 22 May 2002 in the Oklahoma panhandle are presented. The observations are placed in the context of the dryline secondary circulation, which describes flow in a plane normal to the dryline. The narrow, half-power beamwidth of the antenna on the W band (0.18°) permitted the measurements of channels of upward (8–9 m s−1 over a horizontal distance of 50–100 m) and downward vertical velocity, greater in absolute magnitude than that previously reported in dryline field studies.

A ground-based variational pseudo-multiple-Doppler processing technique is introduced, which is used to decompose time series of RHI velocity data into horizontal and vertical wind components. The technique is applied to a retrograding dryline from 22 May 2002. Finescale structure of the retreating dryline interface is presented.

Corresponding author address: Dr. Christopher C. Weiss, Texas Tech University—Atmospheric Science Group, Box 42101, Lubbock, TX 79409. Email: Chris.Weiss@ttu.edu

Abstract

The dryline has long been associated with the development of severe thunderstorms in the southern plains during the spring and early summer months. The propagation and structure of the dryline are closely tied to surface processes that are neither well understood nor well resolved with current observational capabilities. As a result, there are often large errors in forecasts of dryline position and structure.

Improvements in radar technology have allowed for better observations of the dryline in recent years. Here, very finescale radar observations taken with the University of Massachusetts—Amherst (UMass) mobile W-band radar during an International H2O Project (IHOP) double-dryline event on 22 May 2002 in the Oklahoma panhandle are presented. The observations are placed in the context of the dryline secondary circulation, which describes flow in a plane normal to the dryline. The narrow, half-power beamwidth of the antenna on the W band (0.18°) permitted the measurements of channels of upward (8–9 m s−1 over a horizontal distance of 50–100 m) and downward vertical velocity, greater in absolute magnitude than that previously reported in dryline field studies.

A ground-based variational pseudo-multiple-Doppler processing technique is introduced, which is used to decompose time series of RHI velocity data into horizontal and vertical wind components. The technique is applied to a retrograding dryline from 22 May 2002. Finescale structure of the retreating dryline interface is presented.

Corresponding author address: Dr. Christopher C. Weiss, Texas Tech University—Atmospheric Science Group, Box 42101, Lubbock, TX 79409. Email: Chris.Weiss@ttu.edu

1. Introduction

The dryline in the southern plains of the United States can be thought of as the intersection of the top of a surface-based layer of virtually cool, moist air originating over the Gulf of Mexico and the sloping terrain east of the Rocky Mountains. The location of the dryline represents a three-dimensional region of enhanced low-level convergence. Therefore, it also denotes a zone in which there is upward vertical motion. Owing to the strong vertical wind shear that is often characteristic of the dryline environment, storms that form on the dryline often attain supercellular attributes, thereby carrying the attendant threat of large hail, damaging winds, and tornadoes.

The dryline was first mentioned by Fujita (1958), who discussed the concept of a “dry front” and associated thunderstorm development. His analysis followed from the Tornado Research Airplane Project, which gathered the earliest observations of the dryline. During 1961–62, the National Severe Storms Project (NSSP) Staff Members (1963) carried out more detailed analyses. In these studies, dewpoint gradients on the order of 18°C over 1–10 km of horizontal extent were found. Furthermore, surface convergence values on the order of 10−3 s−1 were documented.

It is known that many of the processes controlling the propagation and structure of the dryline are intimately tied to the earth's surface. Terrain slope (e.g., Anthes et al. 1982; Sun and Wu 1992; Peckham and Wicker 2000), surface heat fluxes (e.g., Schaefer 1974; Jones and Bannon 2002), and soil-moisture gradients (e.g., Sun and Wu 1992; Ziegler et al. 1995; Grasso 2000) all have been shown to influence the development of, and advancement of, the dryline.

The moisture gradient associated with the afternoon dryline usually sharpens at a rate greater than that expected from the confluence of winds (as measured by standard surface observations). As such, additional frontogenetical processes have been suggested to play an additive role in the intensification of this moisture gradient. Ogura and Chen (1977) and Sun and Ogura (1979) introduced the “inland sea breeze” hypothesis for the intensification of the moisture gradient associated with the dryline. This hypothesis drew an analog to the coastal sea breeze, in which a diurnal variation in nonhomogeneous surface heating induces a vertical circulation in the plane normal to a sea-breeze boundary. They found in their two-dimensional planetary boundary layer (PBL) model that vertical motions along the boundary were sufficient for the development of moist convection.

Observations of the near-dryline environment have identified ancillary boundaries that form separate from, and often at an angle to, the initial dryline. These boundaries often help explain the large amount of variability noted in the along-dryline direction (Hane et al. 1997, 2002; Atkins et al. 1998).

In some cases, the ancillary boundaries had moisture gradients of comparable magnitude to that of the primary dryline, and could therefore be classified as “double drylines” (Hane et al. 1993; Crawford and Bluestein 1997).

The flow in the plane normal to the dryline defines the dryline secondary circulation (DSC). The DSC of a mature dryline (e.g., Atkins et al. 1998; Weiss and Bluestein 2002) is commonly characterized by an easterly component of ageostrophic flow near the surface to the east of the dryline, a 1–10-km-wide region of ascent in the dryline convergence zone (DCZ) which is usually located near the strongest moisture gradient, and a “return flow” of strong westerly component winds above the boundary layer to the east of the dryline. Variations in the structure of the DSC (particularly the DCZ) directly affect the likelihood of deep convection initiation (e.g., Ziegler and Rasmussen 1998). Therefore, finescale observations of this circulation are crucial to better understand the behavior of the dryline and subsequent convective initiation (e.g., Parsons et al. 1991; Hane et al. 1993, 1997).

During the spring of 2002, a multiagency field experiment, the International H2O Project (IHOP; Weckwerth et al. 2004), was conducted over the central and southern plains. The primary goal of this project was the “improved characterization of the four-dimensional distribution of water vapor and its application to improving the understanding and prediction of convection” (UCAR/ATD 2002). The convection initiation (CI) component of this project was focused on resolving the kinematics of surface boundaries, particularly heterogeneities that would yield clues to the preferential development of convection (e.g., triple points1). Multiple ground-based and aircraft-based measurement platforms were employed for this purpose.

As part of this cooperative effort, the W-band (3-mm wavelength) radar from the University of Massachusetts (UMass) (Bluestein and Pazmany 2000) gathered data on several IHOP case days. Since the quality of the data on 22 May 2002 was superior to that obtained on the rest of the operations days, this case was chosen as a focus for this study. Owing to the very fine beamwidth of the radar (0.18°), previously unresolved dryline spatial structure was observed. Considering the multitude of available measurement platforms, this dryline was one of the most intensively observed in history.

The primary scientific objective of this research is to resolve previously unseen finescale motions of the dryline. Knowing these motions helps us better visualize the contribution of each individual air mass at the dryline interface, and will ultimately lead to better conceptual models of CI in the DCZ.

An overview of the synoptic and mesoscale environment for the 22 May 2002 case is presented in section 2. Section 3 details the characteristics of the UMass W-band radar. Results from the UMass data collection, and an intercomparison with data collected by the University of Wyoming King Air, is presented in sections 4 and 5. A summary and discussion are given in section 6.

2. Synoptic and mesoscale overview

The synoptic pattern for 22 May 2002 was characteristic of many days on which a dryline was present in the southern plains. A negatively tilted longwave 500-hPa trough extended over the western United States at 1200 UTC (Fig. 1a). The trough propagated eastward through the day. A weak 500-hPa jet (35 kt) extended across central and southern Colorado in association with this wave. To the south and east of the wave, 500-hPa winds were weaker over most of Oklahoma and Texas.

At 850 hPa (near the surface), winds were south-southwesterly over the Oklahoma and Texas panhandles (hereafter, the “target region”) at 1200 UTC (Fig. 1b). Modest moisture was transported from the western Gulf of Mexico during the previous 48 h, and was present near the surface [e.g., the Amarillo, Texas (AMA), dewpoint was 10°C]. Moisture rapidly advected northward at 850 hPa over most of Texas and Oklahoma during the afternoon of 22 May.

The 1200 UTC AMA sounding (Fig. 1c) showed a strong inversion from about 840 to 770 hPa, due to a combination of nocturnal radiational cooling at the surface and advection of very warm air from the elevated terrain of New Mexico around 700 hPa. Ample potential instability for deep convection existed above this level. The convective temperature was approximately 30°C over the target region.

At 1500 UTC, the dryline was located over the western Oklahoma and Texas panhandles (Fig. 2a). With the surface winds primarily south-southwesterly to the east of the boundary, limited convergence was apparent at the dryline at this time. By 1800 UTC (Fig. 2b), the dryline had mixed eastward through the central Oklahoma and Texas panhandles and southwestern Kansas; sharp decreases in dewpoint were noted at Elkhart (EHA) and Liberal, Kansas (LBL). Hardly any wind shift occurred with the dryline passage, largely because winds had veered to the east of the boundary. Farther to the south, surface dewpoints fell gradually (e.g., at AMA), less indicative of a distinct dryline passage. This region will be shown to be in an intermediate zone between two sharp moisture gradients/drylines later in the day.

At 2100 UTC (Fig. 2c), the dryline was still located in the eastern Oklahoma panhandle, near the border of Texas and Beaver Counties (the easternmost two Oklahoma panhandle counties). The dewpoint increased by 18°C over the 60 km between Hooker and Beaver, Oklahoma (to the east). A wind shift was also evident across the dryline, from which it may be inferred that there was convergence (assuming there was limited downstream changes in wind speed). There were very gusty winds on both sides of the dryline. The dewpoint at AMA decreased in a more steady manner until 2300 UTC, when there was a sharp decrease (Fig. 2d).

Surface winds backed slightly to the east of the dryline during the late afternoon and early evening (Fig. 2d). This diurnal effect has been attributed to heating of the elevated high terrain and a trough in the lee of the Rocky Mountains, both of which increase easterly ageostrophic flow in the late afternoon and early evening hours to the east of the dryline (Benjamin and Carlson 1986). Deep convection initiated along the dryline over northwestern Kansas and eastern Nebraska by late in the afternoon. This region was more directly influenced by the western longwave trough (Fig. 1a), which likely produced vertical motion and midlevel cooling that aided in the development of the convection. Farther to the south, less midlevel cooling was apparent. According to the 0000 UTC Dodge City, Kansas (DDC), sounding (not shown), convective temperature had been achieved by this time, yet no deep convection was observed at this location.

A comparison of AMA soundings at 1200 and 0000 UTC (Figs. 1c,d) shows the effects of the (eastern) dryline passage. The sharp increase in mixing ratio at 600 hPa and decrease in mixing ratio at the surface from 1200 to 0000 UTC was indicative of the strong vertical turbulent mixing associated with dryline passage. The top of the convective boundary layer had risen to approximately 550 hPa by 0000 UTC. Horizontal momentum had also mixed vertically such that there was very little vertical wind shear in this environment.

Weather Surveillance Radar-1988 Doppler (WSR-88D) radar reflectivity data from AMA at about 2300 and 0000 UTC (Figs. 3a,b) revealed a double-fine-line structure over the northern Texas panhandle. Reflectivity data from the Shared Mobile Atmospheric Research and Teaching Radar (SMART-R) (Biggerstaff and Guynes 2000) (Fig. 3c) also show the double fine line. The passage of the westernmost fine line coincided with the sharp dewpoint decrease observed at AMA at 2300 UTC (Figs. 2c,d and 3a,b). Earlier in the afternoon, AMA was between the fine lines and experienced a gradual decrease in dewpoint.2 This behavior was in accord with that observed in other studies (e.g., Hane et al. (1997) and Crawford and Bluestein (1997)). The two fine lines were oriented such that they converged just to the north of the National Center for Atmospheric Research (NCAR) S-band dual-polarization Doppler radar (S-Pol) at Homestead, Oklahoma (http://www.atd.ucar.edu/rsf/spol/spol.html) (Fig. 4a). Visible satellite imagery (Fig. 5) clearly showed a wedge-shaped area of cumulus cloud cover in the region between the radar fine lines over the Oklahoma and northern Texas panhandles (Figs. 3a,b, 5). A time series of in situ dewpoint measurements taken aboard the University of Wyoming King Air (UWKA) supported the existence of separate moisture gradients associated with each fine line (Figs. 4a,b). It is noted that the domain of UMass operations (white box in Fig. 4a) on this day was to the south of the intersection point. Therefore, the data collection encompassed both drylines. Both dryline boundaries were beginning their diurnal retrogression by 0000 UTC (Figs. 3a,b).

3. W-band radar characteristics

The primary datasets used for this study were collected with the W-band radar from the University of Massachusetts. The characteristics of the radar are listed in Table 1.

The UMass radar has been used as a tool for investigating the finescale structure of tornadoes and their parent severe thunderstorms (e.g., Bluestein et al. 2003, 2004). However, it has demonstrated significant capability of clear-air detection as well (e.g., Bluestein and Pazmany 2000). The radar has a wavelength of 3 mm, an order of magnitude shorter than that of most operational and research radars. The shorter wavelength permits a very narrow beamwidth antenna (a half-power beamwidth of 0.18°) to be utilized. The antenna is thus relatively small (1.2 m) and can be easily mounted on a vehicle. The narrow beamwidth and short wavelength of the W-band radar both allow for greater sensitivity than that obtained with conventional mobile X-band radars. The narrow beamwidth also allows for data collection with very high spatial resolution. At a typical range of 1 km from the radar, the azimuthal/vertical resolution is 3.14 m. Since the power of return for clear-air targets is relatively low, the returned power spectra are rather noisy. Therefore, one must average multiple samples to obtain reliable reflectivity and velocity estimates.

Three different pulse lengths were used during the course of IHOP operations in 2002. For data collected on 22 May 2002, a pulse length of 228 ns was used, which translates into 30-m along-radial resolution.

The primary scatterering source for the power returned to the W-band radar was most likely insects (Wilson and Schreiber 1986; Martin 2003; Geerts and Miao 2004). Since the wavelength was comparable to the size of the targets, Mie scattering was the dominant source of returned power.3 The minimum detectable signal for the W-band radar was −35 dBZe at a range of 1 km from the radar. The average reflectivity in convergence zones at this range (shown later) was about −20 dBZe, representing a returned power over 30 times the minimum detectable signal.

During IHOP, the UMass radar gathered clear-air data on the following dates: 22 May, 3 June, 9 June, and 10 June. Different deployment strategies utilized in the data collection included the following:

  1. Velocity azimuth display (VAD) (Rabin and Zrnic 1980)—stationary collection of data with the antenna elevated at 45°. The antenna was rotated horizontally through a ∼220° portion of a cone (limited by the hardware of the positioner). The horizontal wind could be calculated as a function of height AGL assuming that the vertical profile was horizontally homogeneous over the volume of data collection.

  2. Vertical antenna—antenna pointed at 86° (maximum elevation allowed by the positioner) as the vehicle supporting it was driven across the boundary. The result was a time series of vertical velocity data, which were converted to a spatial profile using recorded GPS data.4

  3. Stationary RHI (SRHI)—stationary data collection in which the antenna was rotated from ∼0° to 86° in elevation. Multiple vertical sectors of radial velocity data were obtained in this manner. Although this collection strategy was useful for tracking reflectivity and diagnosing radial velocity, the u and w components could not be retrieved independently.

  4. Rolling RHI (RRHI)—0°–86° RHIs collected with the platform in motion (Fig. 6). The radial velocity was adjusted for platform motion. As will be discussed later, the principles of pseudo-dual-Doppler analysis (e.g., Hildebrand et al. 1996) could be applied to data collected in such a manner to retrieve the individual u and w wind components.

4. Results from vertical antenna deployment (2221–2235 UTC)

From 2221 to 2235 UTC, the UMass W-band radar executed a westward-moving vertical antenna deployment across the double dryline along U.S. Highway 270 near Elmwood, Oklahoma (Fig. 7). The objective of this deployment was to obtain a time series of vertical velocity in the near-dryline environment. The vehicle maintained a nearly constant speed of 27 m s−1 during the traverse, along which gentle undulations in the terrain were noted. Though the position of radial velocity measurements was adjusted for the pitch of the platform, it was impossible to avoid the inclusion of the horizontal wind component into the measurement. However, the effect was very small owing to the shallow terrain slope.5 It is also possible that there was a minor amount of vehicle roll (which was not measured). However, the effect, again, is expected to be very small.

The SMART-R (Fig. 3c) and S-Pol (Fig. 4a) radars both indicated a fine line associated with the eastern DCZ. This boundary was oriented north-northeast to south-southwest. Therefore, the east–west traverses were carried out at a small angle from normal to the boundary. As mentioned in section 2, a secondary fine line was evident to the west of the “primary” (i.e., targeted) dryline. Though not recognized at the time of data collection, the UMass W-band radar transected this secondary feature just before the termination of the data collection leg (Figs. 3c, 7).

The time section of reflectivity from this leg (Fig. 8a) shows clearly the eastern DCZ as an area of reflectivity in excess of −15 dBZe. The reflectivity maxima was associated with a local concentration of boundary layer scatterers, primarily insects (Wilson and Schreiber 1986; Martin 2003). To a first approximation, these insects are treated as passive tracers and are therefore representative of the wind that is transporting them. Convergence regions, like the one in Fig. 8a, therefore represent areas with a higher density of insects (assuming the insect size distribution and number concentration was homogeneous in the ambient environment). The plume of highest reflectivity was nearly vertical through the depth of the CBL, consistent with previous observations of a nearly vertical dryline interface (e.g., Crawford and Bluestein 1997; Geerts et al. 2006; Weckwerth et al. 2004). A tilted axis of reflectivity was also apparent near the same region. A mobile mesonet (Straka et al. 1996) traverse of the eastern dryline revealed a 6°C increase in dewpoint over approximately 1 km as the probe headed eastward (Fig. 8b). A very slight increase in pressure (adjusted for elevation change) was noted as the probe proceeded from west to east.

Approximately 9 km to the west of the primary (eastern) dryline was the secondary (western) dryline. The feature, though quite subtle in plan position indicator (PPI) reflectivity imagery from S-Pol (Fig. 4a) and SMART-R1 (Fig. 3c), was very distinct in the UMass W-band cross section (Fig. 8a). The western convergence zone was nearly vertical in the lowest 1.5 km AGL, above which a considerable tilt to the east with height was evident. It is suggested that the higher altitude of the eastward tilt (relative to the eastern dryline) was correlated with the deeper moist CBL to its east. DCZs with a large downshear tilt with height have been identified as being less favorable for the development of deep convection as ascending parcels have a greater chance of advecting out of the DCZ before reaching the LCL and LFC (Ziegler and Rasmussen 1998; Peckham and Wicker 2000).

Minima in reflectivity were observed in the eastern dryline interface at approximately 1.5 km AGL. One of these areas (“D1” in Fig. 8a) was immediately to the east of the surface position of the eastern DCZ. The other position (“D2” in Fig. 8a) was about 4 km to the east of the DCZ. The vertical velocity data from the same leg (Fig. 9) showed a correlation between these low reflectivity intrusions and subsiding air motion [considering the fringe vertical velocity values (white and green) on the border of these reflectivity-void regions and others to the west of the eastern DCZ]. Since the source for the scatterers (i.e., insects) is the surface, the scatterer concentration is nearly zero at higher altitudes (e.g., above the boundary layer). Therefore, downward motion represents transport from a region where there is a dearth of insects, and is therefore associated with a lack of radar reflectivity (D. Leon, University of Wyoming, 2004, personal communication).

To confirm further the presence of downward motion in these reflectivity-void regions, observations from the Wyoming Cloud Radar (WCR) (mounted on board the UWKA) were considered. The flight leg (Fig. 7) crossed directly over the path of the UMass W-band radar, and ended approximately 10 min before the completion of the UMass data collection. The aircraft flew at a nearly constant altitude of 700 m AGL with the antenna pointed upward. Note that the flight leg was performed normal to the primary dryline, and therefore formed an angle to the UMass ground leg. The UWKA/WCR indicated downward motion (“D1” and “D2” in Fig. 10a) in the regions marked by reflectivity voids to the east of the eastern DCZ in Fig. 8a. A broad region of subsiding air was observed to the east of the dryline (“D2” in Figs. 10a and 10b). Vertical velocity in excess of −4 m s−1 was observed in this corridor. The position of the descending air at D2 was consistent with that found in the airborne radar study of Weiss and Bluestein (2002). The lowered maximum altitude of returned power at D2 suggested that the source region for this downward moving air was at least in part from above the CBL. One can therefore infer that this air had lower specific humidity. In situ measurements taken aboard the UWKA at 700 m AGL (Fig. 10c) confirm a small local decrease in dewpoint well below the level of the intrusion. Other decreases in dewpoint were observed in descending regions farther to the east (not shown). The downward transport of air from above the CBL may have assisted in the eastward propagation of the dryline through the late morning and early afternoon hours.

The UMass radar and the UWKA/WCR measured comparable values of upward vertical velocity over the width of the eastern DCZ. UMass W-band measurements indicated a maximum upward vertical velocity of 8 to 9 m s−1 (Fig. 9). This intense upward motion was evident only over a very narrow region, however, approximately 50–100 m wide. The UWKA/WCR measured a maximum w of 6.2 m s−1 (Fig. 10a). Although along-dryline variability may have been partly responsible for the discrepancy in maximum w between the platforms, the difference in beamwidth (UMass—0.18°, UWKA/WCR—0.7°) and cross-track platform motion (UMass—26.8 m s−1, UWKA—82.3 m s−1) also contributed. Both of these latter factors ultimately increased the effective size of the beam by increasing the size of the resolution volume. A crude representation of this effect was achieved by a running multiple-point average through the UMass vertical velocity data. For example, a three-point average (very roughly equivalent to a 0.6° beamwidth6) decreased maximum w in the DCZ to 7 m s−1 (not shown). For a five-point average [roughly equivalent to a 1.0° beamwidth such as that on a typical mobile X-band radar (Wurman et al. 1997)], maximum w further decreased to 6 m s−1 (not shown).

As noted above, a nearly upright dryline interface is indicated in UWKA/WCR data just prior to 2230 UTC (Fig. 10a). It is evident in S-Pol reflectivity data that there was an abrupt onset to dryline retrogression at 2230 UTC (not shown). Shortly thereafter, UWKA/WCR reflectivity data from 2241 to 2248 (along nearly the same section as that shown in Fig. 10a) show a marked decrease in the slope of the dryline boundary (Fig. 10b). Certain aspects of the intermediate UMass transect (Figs. 8a, 9) resemble a combination of the presentations in Figs. 10a and 10b. For example, both a vertical and slanted axis of ascent and high reflectivity are noted in the DCZ (Figs. 8a, 9). The correlation between the motion and structure of the dryline appears significant—the stationary dryline was much more sloped than the retreating dryline.

Both the UMass radar and the UWKA/WCR detected upward vertical velocity associated with the convergence zone of the western dryline. However, the magnitude of the upward motion again varied between the platforms. The UWKA/WCR indicated ascent of ∼3–4 m s−1 (Fig. 10a), while UMass showed a maximum w of ∼5 m s−1 in narrow regions (Fig. 10b). The larger tilt and decreased distance between the primary and secondary drylines7 (compared to observations from the UMass radar) suggest that the UWKA/WCR measurements may have been made on a distinctly different portion of the secondary dryline. Regardless, both platforms showed a wide region of descent approximately 3–4 km wide centered about 3–4 km east of the secondary dryline. A decrease in dewpoint of 1°C was present here as well in UWKA in situ data (not shown).

A boresighted video camera was mounted on the W-band dish to assist the radar operator in the proper placement of the narrow beam during field operations. At times when the antenna was pointed vertically, the video served to identify regions of cloud cover directly above the instrument. Three such regions were found during the deployment. The first (labeled “A” in Figs. 8a, 9) area was immediately to the west of the eastern DCZ and contained a very shallow cumulus cloud (Fig. 8a). The second area was likely associated with the ascending branch of a horizontal convective roll (HCR) and was found halfway between the eastern and western DCZ (labeled “B” in Figs. 8a, 9) and contained more vigorous convection. This region of cloud cover was collocated with both a narrow channel of high reflectivity (Fig. 8a) and an upward vertical velocity maximum (Fig. 9) of ∼3 m s−1. The third area of cloud cover was by far the widest (∼2 km) and was associated with the DCZ of the secondary dryline (labeled “C” in Figs. 8a, 9). No cumulus cloud cover was present above or east of the eastern DCZ, nor was any present west of the western DCZ.8 This observation was consistent with the “wedge” shape of cumulus convection shown by satellite (Fig. 5). It is possible that the zone between the two drylines represented an optimal situation for boundary layer cumulus development, with high specific humidity in the CBL (characteristic of the environment to the east of the primary dryline) and a deep CBL (characteristic of the environment to the west of the secondary dryline).

5. Results from rolling RHI deployment (0007–0036 UTC)

From 0007 to 0036 UTC, the UMass W-band radar executed a westward-moving rolling RHI deployment across the eastern dryline (Fig. 11). The geometry of the scanning strategy permitted the overlap of rays. Therefore, individual points in space received many “looks” from the radar (Fig. 12), separated in both time and look angle. Assuming stationarity for the time period between the looks (60 s, discussed below), pseudo-multiple-Doppler principles were employed to synthesize the u and w wind components in the plane approximately normal to the dryline.

A weak-constrant variational (Sasaki 1970) wind synthesis technique was developed for the rolling RHI data. The cost function to be minimized was
i1520-0493-134-1-273-e1a
i1520-0493-134-1-273-e1b
i1520-0493-134-1-273-e1c
i1520-0493-134-1-273-e1d
i1520-0493-134-1-273-e1e
where (1b) represents the contribution to the cost function from observational discrepancy and (1c) denotes the contribution to the cost function from anelastic mass continuity violation. In Eqs. (1a)(1e) u and w are the analysis values, Vr is the observed radial velocity, c1 and c2 represent geometric coefficients mapping velocities from Cartesian space to that of the radial velocity vectors, κ is the correction to mass continuity for vertical density stratification (assumed constant here), m is the total number of observations per grid point of which n is a specific observation, and α is the elevation angle for each observation. The formulation is similar to that developed by Gao et al. (1999) (less a background and smoothness constraint) and Dowell and Bluestein (2002). The technique explicitly neglects variations in υ in the y direction. Since the dryline was oriented nearly north–south, variations in the y direction (along the dryline) were expected to be much smaller than those in the x direction (across the dryline). Observation system simulation experiments using similarly oriented convergence lines showed that a highly accurate analysis could still be produced when neglecting these along-line variations.
The variations of J with respect to the analysis variables u and w were set to zero, yielding the following Euler–Lagrange equations:
i1520-0493-134-1-273-e2
i1520-0493-134-1-273-e3
Equations (2) and (3) were discretized for centered differencing to yield
i1520-0493-134-1-273-e4
and
i1520-0493-134-1-273-e5

Equations (4) and (5) were repeatedly solved in turn until the solutions for u and w over the entire domain converged. Tests were performed that demonstrated the robustness of the variational analysis technique for controlled experiments (not shown).

The variational analysis was applied to rolling RHI data collected on the (eastern) 22 May 2002 dryline as it retrograded toward the west in the early evening hours [(0007–0036 UTC (23 May 2002)]. The retrogression (Fig. 13) was not uniform along the length of the line, as there was evidence of wave activity along the dryline interface. Using data from various radar platforms, the retrogression speed was estimated to be between 2 and 5 m s−1 during the period of the traverse. The UMass radar platform traveled at a nearly constant velocity of 13 m s−1 toward the west as RHI sweeps were performed from the rear horizon up through ∼86° above the rear horizon (Fig. 6). The raw time series of data were postprocessed to account for truck velocity and pitch before data were analyzed.

A composite reflectivity image9 for the traverse (Fig. 14) shows the pronounced eastward tilt of the dryline interface with height during retrogression. As in the vertical antenna deployment, the DCZ appears as a maximum in reflectivity, presumably due to the local increase in insect concentration in this region. The domain chosen for analysis was the lowest 1 km AGL, where there were no data voids.

Owing to the retrogression of the dryline, the stationarity assumption was compromised. As a result, regions with large time intervals between looks produced erroneous large vertical velocity values. To mitigate the effect of dryline propagation, a cutoff time window of 60 s (from first observation) was imposed. This cutoff did decrease maximum look angle differences in the domain, but the observed improvements (by increasing the accuracy of the stationarity approximation) were deemed of greater benefit than the potentially larger error variance (due to collinearity of the radial velocity observations).

The tolerance of the iterative minimization was set to ΔJ = 1.0 m2 s−2 for the analysis, and a first guess of u = w = 0 m s−1 was applied for the entire domain. For the early iterations of the analysis, there was a very sharp decrease in the cost function. Since the cost function was quadratic, this evolution of J was expected. As the analysis approached the optimal solution, the decrease in cost function decelerated considerably. The decrease of the observation component of the cost function (1b) was somewhat offset by the increase in the mass continuity component (1c), which exhibited a substantial increase in the early iterations (expected since the first guess field of u = w = 0 satisfied mass continuity exactly over the entire domain). If the tolerance were higher, the technique would have been more cost efficient—an acceptable solution could have been found in as few as 20 iterations. For this case, however, the analysis ran for 244 iterations before ultimately converging.

The upper and lower branches of the dryline secondary circulation are quite clearly seen in the u component of the optimal analysis (Fig. 15a). The near-surface inflow to the DCZ from the east approached u = −6 m s−1 in some areas of the CBL. Above the CBL, strong westerly component winds (i.e., the return flow) were observed, a combination of air parcels from the moist CBL that had ascended in the DCZ (Hane et al. 1997) and parcels from the dry side that had advected up and over the moist CBL. Westerly winds of over 15 m s−1 were present in the upper portion of the domain. The altitude of the boundary between easterly and westerly component winds sloped eastward with height, increasing ∼1 km over ∼2 km of horizontal extent.

The DCZ appears clearly as a maximum of 8–10 m s−1 in the w-component field (Fig. 15b) (over a very narrow region of ∼100 m). The eastward tilt of the DCZ with height is again present. A small area of descent is evident at ∼500 m AGL approximately 3 km to the east of the surface position of the DCZ. The location of this descending motion is similar to that shown for the vertical antenna deployment analysis and the airborne Doppler case study of Weiss and Bluestein (2002), but lacks the vertical continuity that was present in the latter case.

UWKA/WCR data were again compared to UMass data (Fig. 16) to determine if consistent vertical velocity features were observed from both platforms. Though the flight track of the UWKA was 5–15 km to the north of the UMass W-band radar (Fig. 11) and more than 30 min had elapsed between the periods of UWKA/WCR and UMass data collection, similar areas of ascent and descent are evident in data from both platforms. The upward motion in the DCZ is again less than that measured by UMass (as in the vertical antenna deployment). Subsiding regions in UMass data valid to the east of the DCZ are very much present in the UWKA/WCR field as well. To the best of the authors' knowledge the magnitude and structure of these areas of ascent and descent have not been documented previously in a dryline environment.10

To increase the influence of continuity (and decrease the weight given to radial velocity observations) in the analysis, β was modified to scale with the number of observations:
i1520-0493-134-1-273-e6
The incorporation of (6) into (1a) effectively served as a low-pass filter, and the resulting analyses were expected to be smoother as a result.

When applied to the rolling RHI data, the new formulation provided the expected result. The u-component (Fig. 17a) and w-component wind fields (Fig. 17b) maintained their qualitative structure, though absolute magnitudes of vertical velocity were lower (e.g., maximum w in the DCZ was 6–7 m s−1). One can still see very clearly the discontinuity in the u-component (Fig. 17a) and w-component (Fig. 17b) fields at the dryline interface, and the rotor circulation on the head of the DSC. The rotor circulation and sloped nature of the shear zone mimic the type of structure one would expect to see in a propagating density current. Simpson (1969) and Parsons et al. (1991) both supported the contribution of density current dynamics to the retrograding dryline (at least at the leading edge). The observations from the UMass radar in this study also support this hypothesis, but fine-resolution boundary layer-scale thermodynamic measurements are necessary to assess more firmly the validity of density current theory. An increase in pressure (adjusted for elevation changes) was noted in an eastward mobile mesonet probe transect of the eastern dryline just prior to the UMass rolling RHI data collection (Fig. 18). Although the magnitude of the increase was small (0.5–1.0 mb), the horizontal pressure gradient was larger than that seen in a similar traverse an hour earlier (Fig. 8b). In both mobile mesonet transects (Figs. 8b, 18), a consistent eastward decrease in pressure was present immediately to the east of the strongest moisture gradient.

From Figs. 17a and 17b it is clear that the easterly component winds at the surface extended west of the area of maximum upward motion. Furthermore, the easterly component flow appeared to narrow to a point (Fig. 19), with a local upward bulge of dust/insects at the position of the retrograding dryline. The slope of the boundary had decreased compared to that earlier in the afternoon (Figs. 8a, 10a).

6. Summary and discussion

On the afternoon of 22 May 2002 during IHOP, a unique dataset was collected for a double-dryline event in the Oklahoma panhandle. This dataset was comprised of reflectivity and radial velocity measurements from the UMass W-band radar. The narrow beamwidth of the radar afforded very fine azimuthal resolution of winds at and near the dryline boundaries.

With the antenna pointed vertically, the radar was driven westward across the eastern dryline boundary. The DCZ was well resolved; a maximum upward vertical velocity of w ∼ 8–9 m s−1 was measured in a narrow channel approximately 50–100 m wide. This magnitude was larger than that reported in earlier mobile Doppler dryline studies. For example, Atkins et al. (1998) reported upward motion on the order of 1–2 m s−1 using NCAR Electra Doppler Radar (ELDORA) data with a horizontal (along-track) resolution of 600 m. Parsons et al. (1991) used lidar technology (with horizontal analysis resolution of 200 m) to measure a maximum positive vertical velocity of w ∼ 5 m s−1 on a retrograding dryline in west Texas.

The larger magnitudes of vertical velocity derived in the current study were likely in part due to the narrow beamwidth of the antenna. With a larger-beamwidth radar, samples inside the narrow channel of maximum w were averaged with neighboring samples of lesser magnitude. The narrow azimuthal and range resolution of the W band also allowed for better measurement of the convergence in the DCZ. First, this improvement was due to the narrowing of the cross-dryline width over which the velocity discontinuities were measured in the DCZ. Second, the narrower beamwidth permitted measurements to be taken much closer to the surface without contamination. Therefore the measured depth of the convergence was more accurate in this analysis. Even though a large maximum in w was found in this study, it may be possible that the maximum upward vertical velocity in the DCZ was still underrepresented due to the resistance of insects to upward transport (Achtemeier 1991), violating the assumption that insects are passive tracers of the flow. This negative vertical velocity error bias has been suggested recently by Geerts and Miao (2005), who identified a bias of −0.5 m s−1 in UWKA/WCR data.

Areas of subsidence were noted away from the DCZ. One such area was found in both UMass W-band and UWKA/WCR data approximately 3–4 km east of the DCZ. This position was consistent with a similar finding by Hane et al. (1993), in which the area of descent coincided with a moisture gradient at the surface. Unfortunately in the current case, mobile mesonets did not travel far enough east of the DCZ to confirm if similar gradients were seen at the surface. However, in situ data from the UWKA indicated a minor decrease in dewpoint (∼1°C) in the middle portion of the CBL. Other decreases in dewpoint were evident with areas of descent farther to the east.

The areas of concentrated subsidence discussed above are potentially significant for many matters related to the dryline. Double drylines, for example, may form in such a manner, similar to the observations of Hane et al. (1997) and Hane et al. (2001), and the modeled “microfronts” of Ziegler et al. (1995). The transport of dry air to the surface in the near-dryline environment can have substantial effects. In the cases presented by Hane et al. (1997), vertical mixing of westerly momentum down to the surface was shown to form a convergence line in the dry air west of the dryline. Severe thunderstorms later initiated at the intersection point of this convergence line and the dryline.

Some recent airborne studies of the dryline (e.g., Atkins et al. 1998; Weiss and Bluestein 2002) have shown descending motion immediately to the west of the dryline. The results from the UMass vertical antenna and the UWKA/WCR traverses in this case also indicate weak areas of descent in this region. These observations are consistent with previous observations of westerly-component acceleration immediately to the west of the dryline (e.g., Danielsen 1974; Doswell 1976; Ogura and Chen 1977; McCarthy and Koch 1982). Atkins et al. (1998) documented a similar increase from 0 to 1.5 km AGL west of the dryline with WSR-88D VAD data. The effects of this downward momentum transport are similar to that discussed in the previous paragraph. More data need to be gathered to assess how ubiquitous this downward motion is, but the effect on convective initiation (a precursor?) could be significant.

As we have increased our operational ability for clear-air detection through the years, more “boundaries of unknown origin” have been identified in low-level reflectivity scans near, but seemingly separate from, drylines. Some of these boundaries may well be vertical mixing lines/ancillary drylines. The resultant maxima in surface convergence, particularly at the intersections of these boundaries, could make conditions more favorable for local convective initiation.

Acknowledgments

This work was supported by National Science Foundation (NSF) Grants ATM-9912097 and ATM-0241037 (to the University of Oklahoma), and ATM-0129374 (to the University of Wyoming), and was part of the first author's doctoral dissertation at the University of Oklahoma. We are in debt to Evgeni Fedorovich, Bart Geerts, Carl Hane, Alan Shapiro, David Stensrud, Baxter Vieux, Conrad Ziegler, and two anonymous reviewers for useful input that has improved this work. Bart Geerts and Dave Leon were instrumental in the UWKA/WCR intercomparison efforts. Rick Damiani created the single- and dual-Doppler processing algorithms for the UWKA/WCR data. Thanks also to David Dowell for his useful insight into variational radar data processing. Curtis Alexander, Nettie Arnott, Mike Buban, Yvette Richardson, and Josh Wurman all assisted in providing mobile radar locations and orientations for the 22 May 2002 case. Paul Markowski provided guidance on GIS applications used in the truck pitch correction. Ming Xue assisted in the use of ZXPLOT, which was used to generate the analysis figures. Brendan Fennell drove the UMass vehicle for the 22 May 2002 data collection. We also appreciate the computer support from Mark Laufersweiler at the School of Meteorology.

REFERENCES

  • Achtemeier, G. L., 1991: The use of insects as tracers for “clear-air” boundary-layer studies by Doppler radar. J. Atmos. Oceanic Technol, 8 , 746765.

    • Search Google Scholar
    • Export Citation
  • Anthes, R. A., Y-H. Kuo, S. G. Benjamin, and Y-F. Li, 1982: The evolution of the mesoscale environment of severe local storms: Preliminary modeling results. Mon. Wea. Rev, 110 , 11851213.

    • Search Google Scholar
    • Export Citation
  • Atkins, N. T., R. M. Wakimoto, and C. L. Ziegler, 1998: Observations of the finescale structure of a dryline during VORTEX 95. Mon. Wea. Rev, 126 , 525550.

    • Search Google Scholar
    • Export Citation
  • Benjamin, S. G., and T. N. Carlson, 1986: Some effects of surface heating and topography on the regional severe storms environment. Part I: Three-dimensional simulations. Mon. Wea. Rev, 114 , 307329.

    • Search Google Scholar
    • Export Citation
  • Biggerstaff, M. I., and J. Guynes, 2000: A new tool for atmospheric research. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., 277–280.

  • Bluestein, H. B., and A. L. Pazmany, 2000: Observations of tornadoes and other convective phenomena with a mobile, 3-mm wavelength, Doppler radar: The spring 1999 field experiment. Bull. Amer. Meteor. Soc, 81 , 29392952.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2003: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part I: Tornadogenesis. Mon. Wea. Rev, 131 , 29542967.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2004: The vertical structure of a tornado near Happy, Texas, on 5 May 2002: High-resolution, mobile, W-band, Doppler radar observations. Mon. Wea. Rev, 132 , 23252337.

    • Search Google Scholar
    • Export Citation
  • Crawford, T. M., and H. B. Bluestein, 1997: Characteristics of dryline passage. Mon. Wea. Rev, 125 , 463477.

  • Danielsen, E., 1974: The relationship between severe weather, major duststorms, and rapid large-scale cyclogenesis. Subsynoptic Extratropical Weather Systems, Part I, M. Shapiro, Ed., National Center for Atmospheric Research, 215–241.

    • Search Google Scholar
    • Export Citation
  • Doswell III, C. A., 1976: Subsynoptic scale dynamics as revealed by use of filtered surface data. NOAA Tech. Memo. ERL NSSL-79, 40 pp.

  • Dowell, D. C., and H. B. Bluestein, 2002: The 8 June 1995 McLean, Texas, storm. Part I: Observations of cyclic tornadogenesis. Mon. Wea. Rev, 130 , 26262648.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1958: Structure and movement of a dry front. Bull. Amer. Meteor. Soc, 39 , 574582.

  • Gao, J., M. Xue, A. Shapiro, and K. K. Droegemeier, 1999: A variational method for the analysis of three-dimensional wind fields from two Doppler radars. Mon. Wea. Rev, 127 , 21282142.

    • Search Google Scholar
    • Export Citation
  • Geerts, B., and Q. Miao, 2005: The use of millimeter Doppler radar echoes to estimate vertical air velocities in the fair-weather convective boundary layer. J. Atmos. Oceanic Technol, 22 , 225246.

    • Search Google Scholar
    • Export Citation
  • Geerts, B., R. Damiani, and S. Haimov, 2006: Finescale vertical structure of a cold front as revealed by airborne Doppler radar. Mon. Wea. Rev, 134 , 151271.

    • Search Google Scholar
    • Export Citation
  • Grasso, L. D., 2000: A numerical simulation of dryline sensitivity to soil moisture. Mon. Wea. Rev, 128 , 28162834.

  • Hane, C. E., C. L. Ziegler, and H. B. Bluestein, 1993: Investigation of the dryline and convective storms initiated along the dryline: Field experiments during COPS-91. Bull. Amer. Meteor. Soc, 74 , 21332145.

    • Search Google Scholar
    • Export Citation
  • Hane, C. E., H. B. Bluestein, T. M. Crawford, M. E. Baldwin, and R. M. Rabin, 1997: Severe thunderstorm development in relation to along-dryline variability: A case study. Mon. Wea. Rev, 125 , 231251.

    • Search Google Scholar
    • Export Citation
  • Hane, C. E., M. E. Baldwin, H. B. Bluestein, T. M. Crawford, and R. M. Rabin, 2001: A case study of severe storm development along a dryline within a synoptically active environment. Part I: Dryline motion and an Eta Model forecast. Mon. Wea. Rev, 129 , 21832204.

    • Search Google Scholar
    • Export Citation
  • Hane, C. E., R. M. Rabin, T. M. Crawford, H. B. Bluestein, and M. E. Baldwin, 2002: A case study of severe storm development along a dryline within a synoptically active environment. Part II: Multiple boundaries and convective initiation. Mon. Wea. Rev, 130 , 900920.

    • Search Google Scholar
    • Export Citation
  • Hildebrand, P. H., and Coauthors, 1996: The ELDORA/ASTRAIA airborne Doppler weather radar: High resolution observations from TOGA COARE. Bull. Amer. Meteor. Soc, 77 , 213232.

    • Search Google Scholar
    • Export Citation
  • Jones, P. A., and P. R. Bannon, 2002: A mixed-layer model of the diurnal dryline. J. Atmos. Sci, 59 , 25822593.

  • Martin, W. J., 2003: Measurements and modeling of the Great Plains low-level jet. Ph.D. dissertation, University of Oklahoma, 242 pp. [Available online at ftp://caps.ou.edu/wmartin.].

  • McCarthy, J., and S. E. Koch, 1982: The evolution of an Oklahoma dryline. Part I: A meso- and subsynoptic-scale analysis. J. Atmos. Sci, 39 , 225236.

    • Search Google Scholar
    • Export Citation
  • National Severe Storms Project (NSSP) Staff Members, 1963: Environmental and thunderstorm structures as shown by National Severe Storms Project observations in spring 1960 and 1961. Mon. Wea. Rev, 91 , 271292.

    • Search Google Scholar
    • Export Citation
  • Ogura, Y., and Y-L. Chen, 1977: A life history of an intense mesoscale convective storm in Oklahoma. J. Atmos. Sci, 34 , 14581476.

  • Parsons, D. B., M. A. Shapiro, R. M. Hardesty, R. J. Zamora, and J. M. Intrieri, 1991: The finescale structure of a west Texas dryline. Mon. Wea. Rev, 119 , 12831292.

    • Search Google Scholar
    • Export Citation
  • Peckham, S. E., and L. J. Wicker, 2000: The influence of topography and lower-tropospheric winds on dryline morphology. Mon. Wea. Rev, 128 , 21652189.

    • Search Google Scholar
    • Export Citation
  • Rabin, R. M., and D. Zrnic, 1980: Subsynoptic-scale vertical wind revealed by dual Doppler radar and VAD analysis. J. Atmos. Sci, 37 , 644654.

    • Search Google Scholar
    • Export Citation
  • Sasaki, Y., 1970: Some basic formalisms in numerical variational analysis. Mon. Wea. Rev, 98 , 875883.

  • Schaefer, J. T., 1974: The life cycle of the dryline. J. Appl. Meteor, 13 , 444449.

  • Simpson, J. E., 1969: A comparison between laboratory and atmospheric density currents. Quart. J. Roy. Meteor. Soc, 95 , 758765.

  • Straka, J. M., E. N. Rasmussen, and S. E. Fredrickson, 1996: A mobile mesonet for finescale meteorological observations. J. Atmos. Oceanic Technol, 13 , 921936.

    • Search Google Scholar
    • Export Citation
  • Sun, W. Y., and Y. Ogura, 1979: Boundary layer forcing as a possible trigger to a squall line formation. J. Atmos. Sci, 36 , 235254.

  • Sun, W. Y., and C. C. Wu, 1992: Formation and diurnal variation of the dryline. J. Atmos. Sci, 49 , 16061619.

  • UCAR/ATD, 2002: International H2O Project (IHOP_2002): Operations plan. 160 pp. [Available online at http://www.atd.ucar.edu/dir_off/projects/2002/IHOPdocs/opsplan.doc.].

  • Weckwerth, T. M., and Coauthors, 2004: An overview of the International H2O Project (IHOP_2002) and some preliminary highlights. Bull. Amer. Meteor. Soc, 85 , 253277.

    • Search Google Scholar
    • Export Citation
  • Weiss, C. C., and H. B. Bluestein, 2002: Airborne pseudo-dual Doppler analysis of a dryline-outflow boundary intersection. Mon. Wea. Rev, 130 , 12071226.

    • Search Google Scholar
    • Export Citation
  • Wilson, J. W., and W. E. Schreiber, 1986: Initiation of convective storms by radar-observed boundary layer convergence lines. Mon. Wea. Rev, 114 , 25162536.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., J. M. Straka, E. N. Rasmussen, M. Randall, and A. Zahrai, 1997: Design and deployment of a portable, pencil-beam, pulsed, 3-cm Doppler radar. J. Atmos. Oceanic Technol, 14 , 15021512.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., and C. E. Hane, 1993: An observational study of the dryline. Mon. Wea. Rev, 121 , 11341151.

  • Ziegler, C. L., and E. N. Rasmussen, 1998: The initiation of moist convection at the dryline: Forecasting issues from a case study perspective. Wea. Forecasting, 13 , 11061131.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., W. J. Martin, R. A. Pielke, and R. L. Walko, 1995: A modeling study of the dryline. J. Atmos. Sci, 52 , 263285.

Fig. 1.
Fig. 1.

(a) The 500- and (b) 850-hPa analyses valid on 22 May 2002 at 1200 UTC. Solid contours denote height [m, contoured every 40 m in (a) and every 30 m in (b)]. Dashed lines denote temperature [°C, contoured every 2°C in (a) and every 3°C in (b)]. Skew T diagrams from rawinsondes released at Amarillo, TX, valid on (c) 22 May 2002 at 1200 UTC and (d) 23 May 2002 at 0000 UTC. Full (half) wind barbs represent 5 (2.5) m s−1

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 1.
Fig. 1.

(Continued)

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 2.
Fig. 2.

High plains surface map valid at (a) 1500, (b) 1800, (c) 2100, and (d) 2300 UTC on 22 May 2002. Temperature and dewpoint for each station reported in °C. Full (half) wind barbs denote 5 (2.5) m s−1. The scalloped lines indicate the position of (known) drylines. The labels “EHA,” “LBL,” “AMA,” “H,” and “B” denote the locations of Elkhart, KS; Liberal, KS; Amarillo, TX; Hooker, OK; and Beaver, OK, respectively

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 3.
Fig. 3.

WSR-88D 0.5° reflectivity at Amarillo, TX, valid (a) 2308 UTC (22 May 2002) and (b) 0007 UTC (23 May 2002). Reflectivity scale (dBZ) provided to the left. The reflectivity maxima associated with the eastern and western drylines are noted in (a). (c) The 0.5° reflectivity from the SMART-R at 2254 UTC on 22 May 2002. The black line indicates the path of the UMass vertical antenna deployment. Note that these data were taken approximately 20 min after the termination of the UMass vertical antenna deployment. The locations of the eastern and western drylines are shown

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 4.
Fig. 4.

(a) Reflectivity from S-Pol (location indicated in Fig. 5). Reflectivity scale (dBZ) indicated to the right. The straight white lines indicate the axes of two separate drylines. The white box represents the domain of operations for the UMass W-band radar. The red line denotes the flight track of the UWKA from 2233 to 2240 UTC. (b) Traces of in situ specific humidity (g kg−1, solid trace) and u-component wind (m s−1, dashed trace) taken aboard the UWKA for the flight leg indicated in (a). Time (UTC) is indicated along the bottom axis, scales for mixing ratio (g kg−1) and u-component wind (m s−1) are indicated on the left and right axes, respectively. The two regions of sharp moisture gradient are circled in green

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 5.
Fig. 5.

Geostationary Operational Environmental Satellite-8 (GOES-8) visible satellite image at 2233 UTC on 22 May 2002. Black boundaries denote the state borders. The black dot indicates the location of S-Pol

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 6.
Fig. 6.

Schematic of RRHI method of data collection

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 7.
Fig. 7.

A map of the vertical antenna UMass W-band deployment (2221–2235 UTC). Distance scale located in the lower-right-hand corner. The thick lines denote the path of the UMass vertical antenna deployment, from right to left (east to west) and the path of the UWKA radar from right to left (east-southeast to west-northwest) (2218–2224 UTC). The scalloped lines indicate the position of known drylines. The dot labeled SPOL denotes the position of the S-Pol radar near Homestead, OK

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 8.
Fig. 8.

(a) East–west cross section of reflectivity from the vertical antenna deployment (2221–2235 UTC). Equivalent reflectivity scale (dBZe) is shown at the top. One-kilometer scales for the horizontal and vertical direction are shown in the upper-right-hand corner. Domain size is approximately 18 km wide (east–west) by 3.4 km high. Letters “A,” “B,” and “C” are the locations of cloud cover discussed in the text. The UMass vehicle was in motion toward the west (left). Labels “D1” and “D2” are referred to in the text. Images of video from the W-band boresighted video camera are shown. The blue bar at the base of the reflectivity image is the approximate track of the mobile mesonet probe in (b). (b) Temperature, dewpoint (°C, scale on left), and elevation-adjusted pressure (hPa, scale on right) measured by a mobile mesonet probe from 2237 to 2245 UTC. Approximate track denoted by blue line in (a)

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 9.
Fig. 9.

As in Fig. 8a, except colors denote vertical velocity (m s−1). Orange colors indicate upward motion, and green colors indicate downward motion. Velocity scale is indicated at the top. Labels “D1” and “D2” are referred to in the text

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 10.
Fig. 10.

UWKA/WCR vertical velocity (m s−1, scale to left) from (a) 2218 to 2224 and (b) 2241 to 2248 UTC. (c) Trace of in situ dewpoint measurements at flight level (700 m AGL) aboard the UWKA (°C, scale indicated to left) from 2218 to 2224 UTC. The horizontal distance scale in all three images represents the distance (in km) east (positive) or west (negative) of the eastern dryline (positioned at 0 km). The vertical distance scale is in km AGL [note the different flight altitudes in (a) and (b)]. Labels “D1” and “D2” are referred to in the text

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 11.
Fig. 11.

A map of the rolling RHI UMass W-band deployment (0007–0036 UTC). Distance scale located in the lower-right-hand corner. The thick lines denote the path of the UMass rolling RHI deployment from right to left (east to west), and the path of the UWKA radar (2345–2351 UTC) from right to left (east-southeast to west-northwest). The dot labeled SPOL denotes the position of the S-Pol radar at Homestead, OK

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 12.
Fig. 12.

Look count composite from the UMass rolling RHI (0007–0036 UTC). The distance scale (km) is indicated at the bottom of the figure and is the same as in Figs. 15 and 17

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 13.
Fig. 13.

SMART-R 0.5° reflectivity valid at (a) 0012 and (b) 0030 UTC. The north radial is highlighted in black to show retrogression more clearly. The approximate path of the UMass rolling RHI (right to left) is shown in red

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 14.
Fig. 14.

An east–west display of composite reflectivity from the UMass rolling RHI (2007–2036 UTC).Horizontal and vertical distance scales are indicated. The black box denotes the domain for analysis

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 15.
Fig. 15.

(a) Ground-relative u-component wind (m s−1, contoured) and (b) w-component wind (m s−1, contoured) from the variational analysis of the rolling RHI. Cool colors indicate negative component, and warm colors indicate positive component. The arrows in (a) and (b) represent u/w wind vectors. Horizontal and vertical distance scales (km AGL) are indicated. Grid points with less than 10 radial velocity observations have been omitted. The areas of strong ascent “A” and descent “D” are identified for the purposes of intercomparison with the UWKA/WCR data in Fig. 16

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 16.
Fig. 16.

UWKA/WCR vertical velocity (m s−1, scale to right) from 2345 to 2351 UTC. The distance scale represents along-track distance (km), with the east–west dryline-relative position approximately the same as in Fig. 15 (centered on the surface dryline position at 1 km). The areas of strong ascent “A” and descent “D” in Fig. 15 are identified for the purposes of intercomparison

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 17.
Fig. 17.

As in Fig. 15, except for analyses with β = m(Δx)2

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 18.
Fig. 18.

Temperature, dewpoint (°C, scale on left), and elevation-adjusted pressure (hPa, scale on right) measured by a mobile mesonet probe from 2331 to 2343 UTC. The probe crossed the dryline approximately 3.5 km to the north of the UMass radar

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Fig. 19.
Fig. 19.

SOLO display of a vertical sweep during the UMass rolling RHI. Pictured are reflectivity (dBZe) and ground-relative radial velocity (m s−1) of the retrograding dryline. Scales are located at the bottom of each figure. Vertical and horizontal distance scales are shown. Approximate dryline position indicated by the scalloped line. Here, “R” denotes the (stationary) radar location

Citation: Monthly Weather Review 134, 1; 10.1175/MWR3068.1

Table 1.

Characteristics of the University of Massachusetts W-band radar

Table 1.

1

The term “triple point” is colloquially used to denote the intersection point of three distinctly different air masses. The intersection of a dryline and a baroclinic boundary (front, outflow boundary) is often referred to in such a manner.

2

No evidence is available to the authors that makes possible the precise determination of the onset of the double-dryline structure. Clear-air reflectivity data were unavailable from Amarillo, Texas, between 1600 and 2100 UTC.

3

Equivalent reflectivity (dBZe) takes into account the possibility of Mie scattering, where reflectivity is related to the square of the target diameter.

4

According to the manufacturer, the position of the Garmin GPS receiver is accurate to within 15 m on average.

5

As a worst-case scenario, a 20 m s−1 wind with a 2.0° platform pitch would erroneously add 0.7 m s−1 to the vertical velocity measurement. The roll of the vehicle was assumed to be negligible.

6

To find the exact equivalency, the gain pattern of the respective antennas must be considered.

7

Recall that the UWKA intercepted the secondary dryline north of the UMass intercept (Fig. 7).

8

A narrow band of cirrus clouds was seen, however.

9

The reflectivity depicted at each point is that of the most recent data collected by the radar at each point.

10

Stepped aircraft traverses have shown broad descending areas previously (e.g., Ziegler and Hane 1993; Hane et al. 1997; Hane et al. 2001), but this study provides a more detailed look at the structure and magnitude of these areas.

Save
  • Achtemeier, G. L., 1991: The use of insects as tracers for “clear-air” boundary-layer studies by Doppler radar. J. Atmos. Oceanic Technol, 8 , 746765.

    • Search Google Scholar
    • Export Citation
  • Anthes, R. A., Y-H. Kuo, S. G. Benjamin, and Y-F. Li, 1982: The evolution of the mesoscale environment of severe local storms: Preliminary modeling results. Mon. Wea. Rev, 110 , 11851213.

    • Search Google Scholar
    • Export Citation
  • Atkins, N. T., R. M. Wakimoto, and C. L. Ziegler, 1998: Observations of the finescale structure of a dryline during VORTEX 95. Mon. Wea. Rev, 126 , 525550.

    • Search Google Scholar
    • Export Citation
  • Benjamin, S. G., and T. N. Carlson, 1986: Some effects of surface heating and topography on the regional severe storms environment. Part I: Three-dimensional simulations. Mon. Wea. Rev, 114 , 307329.

    • Search Google Scholar
    • Export Citation
  • Biggerstaff, M. I., and J. Guynes, 2000: A new tool for atmospheric research. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., 277–280.

  • Bluestein, H. B., and A. L. Pazmany, 2000: Observations of tornadoes and other convective phenomena with a mobile, 3-mm wavelength, Doppler radar: The spring 1999 field experiment. Bull. Amer. Meteor. Soc, 81 , 29392952.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2003: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part I: Tornadogenesis. Mon. Wea. Rev, 131 , 29542967.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2004: The vertical structure of a tornado near Happy, Texas, on 5 May 2002: High-resolution, mobile, W-band, Doppler radar observations. Mon. Wea. Rev, 132 , 23252337.

    • Search Google Scholar
    • Export Citation
  • Crawford, T. M., and H. B. Bluestein, 1997: Characteristics of dryline passage. Mon. Wea. Rev, 125 , 463477.

  • Danielsen, E., 1974: The relationship between severe weather, major duststorms, and rapid large-scale cyclogenesis. Subsynoptic Extratropical Weather Systems, Part I, M. Shapiro, Ed., National Center for Atmospheric Research, 215–241.

    • Search Google Scholar
    • Export Citation
  • Doswell III, C. A., 1976: Subsynoptic scale dynamics as revealed by use of filtered surface data. NOAA Tech. Memo. ERL NSSL-79, 40 pp.

  • Dowell, D. C., and H. B. Bluestein, 2002: The 8 June 1995 McLean, Texas, storm. Part I: Observations of cyclic tornadogenesis. Mon. Wea. Rev, 130 , 26262648.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1958: Structure and movement of a dry front. Bull. Amer. Meteor. Soc, 39 , 574582.

  • Gao, J., M. Xue, A. Shapiro, and K. K. Droegemeier, 1999: A variational method for the analysis of three-dimensional wind fields from two Doppler radars. Mon. Wea. Rev, 127 , 21282142.

    • Search Google Scholar
    • Export Citation
  • Geerts, B., and Q. Miao, 2005: The use of millimeter Doppler radar echoes to estimate vertical air velocities in the fair-weather convective boundary layer. J. Atmos. Oceanic Technol, 22 , 225246.

    • Search Google Scholar
    • Export Citation
  • Geerts, B., R. Damiani, and S. Haimov, 2006: Finescale vertical structure of a cold front as revealed by airborne Doppler radar. Mon. Wea. Rev, 134 , 151271.

    • Search Google Scholar
    • Export Citation
  • Grasso, L. D., 2000: A numerical simulation of dryline sensitivity to soil moisture. Mon. Wea. Rev, 128 , 28162834.

  • Hane, C. E., C. L. Ziegler, and H. B. Bluestein, 1993: Investigation of the dryline and convective storms initiated along the dryline: Field experiments during COPS-91. Bull. Amer. Meteor. Soc, 74 , 21332145.

    • Search Google Scholar
    • Export Citation
  • Hane, C. E., H. B. Bluestein, T. M. Crawford, M. E. Baldwin, and R. M. Rabin, 1997: Severe thunderstorm development in relation to along-dryline variability: A case study. Mon. Wea. Rev, 125 , 231251.

    • Search Google Scholar
    • Export Citation
  • Hane, C. E., M. E. Baldwin, H. B. Bluestein, T. M. Crawford, and R. M. Rabin, 2001: A case study of severe storm development along a dryline within a synoptically active environment. Part I: Dryline motion and an Eta Model forecast. Mon. Wea. Rev, 129 , 21832204.

    • Search Google Scholar
    • Export Citation
  • Hane, C. E., R. M. Rabin, T. M. Crawford, H. B. Bluestein, and M. E. Baldwin, 2002: A case study of severe storm development along a dryline within a synoptically active environment. Part II: Multiple boundaries and convective initiation. Mon. Wea. Rev, 130 , 900920.

    • Search Google Scholar
    • Export Citation
  • Hildebrand, P. H., and Coauthors, 1996: The ELDORA/ASTRAIA airborne Doppler weather radar: High resolution observations from TOGA COARE. Bull. Amer. Meteor. Soc, 77 , 213232.

    • Search Google Scholar
    • Export Citation
  • Jones, P. A., and P. R. Bannon, 2002: A mixed-layer model of the diurnal dryline. J. Atmos. Sci, 59 , 25822593.

  • Martin, W. J., 2003: Measurements and modeling of the Great Plains low-level jet. Ph.D. dissertation, University of Oklahoma, 242 pp. [Available online at ftp://caps.ou.edu/wmartin.].

  • McCarthy, J., and S. E. Koch, 1982: The evolution of an Oklahoma dryline. Part I: A meso- and subsynoptic-scale analysis. J. Atmos. Sci, 39 , 225236.

    • Search Google Scholar
    • Export Citation
  • National Severe Storms Project (NSSP) Staff Members, 1963: Environmental and thunderstorm structures as shown by National Severe Storms Project observations in spring 1960 and 1961. Mon. Wea. Rev, 91 , 271292.

    • Search Google Scholar
    • Export Citation
  • Ogura, Y., and Y-L. Chen, 1977: A life history of an intense mesoscale convective storm in Oklahoma. J. Atmos. Sci, 34 , 14581476.

  • Parsons, D. B., M. A. Shapiro, R. M. Hardesty, R. J. Zamora, and J. M. Intrieri, 1991: The finescale structure of a west Texas dryline. Mon. Wea. Rev, 119 , 12831292.

    • Search Google Scholar
    • Export Citation
  • Peckham, S. E., and L. J. Wicker, 2000: The influence of topography and lower-tropospheric winds on dryline morphology. Mon. Wea. Rev, 128 , 21652189.

    • Search Google Scholar
    • Export Citation
  • Rabin, R. M., and D. Zrnic, 1980: Subsynoptic-scale vertical wind revealed by dual Doppler radar and VAD analysis. J. Atmos. Sci, 37 , 644654.

    • Search Google Scholar
    • Export Citation
  • Sasaki, Y., 1970: Some basic formalisms in numerical variational analysis. Mon. Wea. Rev, 98 , 875883.

  • Schaefer, J. T., 1974: The life cycle of the dryline. J. Appl. Meteor, 13 , 444449.

  • Simpson, J. E., 1969: A comparison between laboratory and atmospheric density currents. Quart. J. Roy. Meteor. Soc, 95 , 758765.

  • Straka, J. M., E. N. Rasmussen, and S. E. Fredrickson, 1996: A mobile mesonet for finescale meteorological observations. J. Atmos. Oceanic Technol, 13 , 921936.

    • Search Google Scholar
    • Export Citation
  • Sun, W. Y., and Y. Ogura, 1979: Boundary layer forcing as a possible trigger to a squall line formation. J. Atmos. Sci, 36 , 235254.

  • Sun, W. Y., and C. C. Wu, 1992: Formation and diurnal variation of the dryline. J. Atmos. Sci, 49 , 16061619.

  • UCAR/ATD, 2002: International H2O Project (IHOP_2002): Operations plan. 160 pp. [Available online at http://www.atd.ucar.edu/dir_off/projects/2002/IHOPdocs/opsplan.doc.].

  • Weckwerth, T. M., and Coauthors, 2004: An overview of the International H2O Project (IHOP_2002) and some preliminary highlights. Bull. Amer. Meteor. Soc, 85 , 253277.

    • Search Google Scholar
    • Export Citation
  • Weiss, C. C., and H. B. Bluestein, 2002: Airborne pseudo-dual Doppler analysis of a dryline-outflow boundary intersection. Mon. Wea. Rev, 130 , 12071226.

    • Search Google Scholar
    • Export Citation
  • Wilson, J. W., and W. E. Schreiber, 1986: Initiation of convective storms by radar-observed boundary layer convergence lines. Mon. Wea. Rev, 114 , 25162536.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., J. M. Straka, E. N. Rasmussen, M. Randall, and A. Zahrai, 1997: Design and deployment of a portable, pencil-beam, pulsed, 3-cm Doppler radar. J. Atmos. Oceanic Technol, 14 , 15021512.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., and C. E. Hane, 1993: An observational study of the dryline. Mon. Wea. Rev, 121 , 11341151.

  • Ziegler, C. L., and E. N. Rasmussen, 1998: The initiation of moist convection at the dryline: Forecasting issues from a case study perspective. Wea. Forecasting, 13 , 11061131.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., W. J. Martin, R. A. Pielke, and R. L. Walko, 1995: A modeling study of the dryline. J. Atmos. Sci, 52 , 263285.

  • Fig. 1.

    (a) The 500- and (b) 850-hPa analyses valid on 22 May 2002 at 1200 UTC. Solid contours denote height [m, contoured every 40 m in (a) and every 30 m in (b)]. Dashed lines denote temperature [°C, contoured every 2°C in (a) and every 3°C in (b)]. Skew T diagrams from rawinsondes released at Amarillo, TX, valid on (c) 22 May 2002 at 1200 UTC and (d) 23 May 2002 at 0000 UTC. Full (half) wind barbs represent 5 (2.5) m s−1

  • Fig. 1.

    (Continued)

  • Fig. 2.

    High plains surface map valid at (a) 1500, (b) 1800, (c) 2100, and (d) 2300 UTC on 22 May 2002. Temperature and dewpoint for each station reported in °C. Full (half) wind barbs denote 5 (2.5) m s−1. The scalloped lines indicate the position of (known) drylines. The labels “EHA,” “LBL,” “AMA,” “H,” and “B” denote the locations of Elkhart, KS; Liberal, KS; Amarillo, TX; Hooker, OK; and Beaver, OK, respectively

  • Fig. 3.

    WSR-88D 0.5° reflectivity at Amarillo, TX, valid (a) 2308 UTC (22 May 2002) and (b) 0007 UTC (23 May 2002). Reflectivity scale (dBZ) provided to the left. The reflectivity maxima associated with the eastern and western drylines are noted in (a). (c) The 0.5° reflectivity from the SMART-R at 2254 UTC on 22 May 2002. The black line indicates the path of the UMass vertical antenna deployment. Note that these data were taken approximately 20 min after the termination of the UMass vertical antenna deployment. The locations of the eastern and western drylines are shown

  • Fig. 4.

    (a) Reflectivity from S-Pol (location indicated in Fig. 5). Reflectivity scale (dBZ) indicated to the right. The straight white lines indicate the axes of two separate drylines. The white box represents the domain of operations for the UMass W-band radar. The red line denotes the flight track of the UWKA from 2233 to 2240 UTC. (b) Traces of in situ specific humidity (g kg−1, solid trace) and u-component wind (m s−1, dashed trace) taken aboard the UWKA for the flight leg indicated in (a). Time (UTC) is indicated along the bottom axis, scales for mixing ratio (g kg−1) and u-component wind (m s−1) are indicated on the left and right axes, respectively. The two regions of sharp moisture gradient are circled in green

  • Fig. 5.

    Geostationary Operational Environmental Satellite-8 (GOES-8) visible satellite image at 2233 UTC on 22 May 2002. Black boundaries denote the state borders. The black dot indicates the location of S-Pol

  • Fig. 6.

    Schematic of RRHI method of data collection

  • Fig. 7.

    A map of the vertical antenna UMass W-band deployment (2221–2235 UTC). Distance scale located in the lower-right-hand corner. The thick lines denote the path of the UMass vertical antenna deployment, from right to left (east to west) and the path of the UWKA radar from right to left (east-southeast to west-northwest) (2218–2224 UTC). The scalloped lines indicate the position of known drylines. The dot labeled SPOL denotes the position of the S-Pol radar near Homestead, OK

  • Fig. 8.

    (a) East–west cross section of reflectivity from the vertical antenna deployment (2221–2235 UTC). Equivalent reflectivity scale (dBZe) is shown at the top. One-kilometer scales for the horizontal and vertical direction are shown in the upper-right-hand corner. Domain size is approximately 18 km wide (east–west) by 3.4 km high. Letters “A,” “B,” and “C” are the locations of cloud cover discussed in the text. The UMass vehicle was in motion toward the west (left). Labels “D1” and “D2” are referred to in the text. Images of video from the W-band boresighted video camera are shown. The blue bar at the base of the reflectivity image is the approximate track of the mobile mesonet probe in (b). (b) Temperature, dewpoint (°C, scale on left), and elevation-adjusted pressure (hPa, scale on right) measured by a mobile mesonet probe from 2237 to 2245 UTC. Approximate track denoted by blue line in (a)

  • Fig. 9.

    As in Fig. 8a, except colors denote vertical velocity (m s−1). Orange colors indicate upward motion, and green colors indicate downward motion. Velocity scale is indicated at the top. Labels “D1” and “D2” are referred to in the text

  • Fig. 10.

    UWKA/WCR vertical velocity (m s−1, scale to left) from (a) 2218 to 2224 and (b) 2241 to 2248 UTC. (c) Trace of in situ dewpoint measurements at flight level (700 m AGL) aboard the UWKA (°C, scale indicated to left) from 2218 to 2224 UTC. The horizontal distance scale in all three images represents the distance (in km) east (positive) or west (negative) of the eastern dryline (positioned at 0 km). The vertical distance scale is in km AGL [note the different flight altitudes in (a) and (b)]. Labels “D1” and “D2” are referred to in the text

  • Fig. 11.

    A map of the rolling RHI UMass W-band deployment (0007–0036 UTC). Distance scale located in the lower-right-hand corner. The thick lines denote the path of the UMass rolling RHI deployment from right to left (east to west), and the path of the UWKA radar (2345–2351 UTC) from right to left (east-southeast to west-northwest). The dot labeled SPOL denotes the position of the S-Pol radar at Homestead, OK

  • Fig. 12.

    Look count composite from the UMass rolling RHI (0007–0036 UTC). The distance scale (km) is indicated at the bottom of the figure and is the same as in Figs. 15 and 17

  • Fig. 13.

    SMART-R 0.5° reflectivity valid at (a) 0012 and (b) 0030 UTC. The north radial is highlighted in black to show retrogression more clearly. The approximate path of the UMass rolling RHI (right to left) is shown in red

  • Fig. 14.

    An east–west display of composite reflectivity from the UMass rolling RHI (2007–2036 UTC).Horizontal and vertical distance scales are indicated. The black box denotes the domain for analysis

  • Fig. 15.

    (a) Ground-relative u-component wind (m s−1, contoured) and (b) w-component wind (m s−1, contoured) from the variational analysis of the rolling RHI. Cool colors indicate negative component, and warm colors indicate positive component. The arrows in (a) and (b) represent u/w wind vectors. Horizontal and vertical distance scales (km AGL) are indicated. Grid points with less than 10 radial velocity observations have been omitted. The areas of strong ascent “A” and descent “D” are identified for the purposes of intercomparison with the UWKA/WCR data in Fig. 16

  • Fig. 16.

    UWKA/WCR vertical velocity (m s−1, scale to right) from 2345 to 2351 UTC. The distance scale represents along-track distance (km), with the east–west dryline-relative position approximately the same as in Fig. 15 (centered on the surface dryline position at 1 km). The areas of strong ascent “A” and descent “D” in Fig. 15 are identified for the purposes of intercomparison

  • Fig. 17.

    As in Fig. 15, except for analyses with β = m(Δx)2

  • Fig. 18.

    Temperature, dewpoint (°C, scale on left), and elevation-adjusted pressure (hPa, scale on right) measured by a mobile mesonet probe from 2331 to 2343 UTC. The probe crossed the dryline approximately 3.5 km to the north of the UMass radar

  • Fig. 19.

    SOLO display of a vertical sweep during the UMass rolling RHI. Pictured are reflectivity (dBZe) and ground-relative radial velocity (m s−1) of the retrograding dryline. Scales are located at the bottom of each figure. Vertical and horizontal distance scales are shown. Approximate dryline position indicated by the scalloped line. Here, “R” denotes the (stationary) radar location