1. Introduction
Radar networks have been used for weather surveillance for nearly 70 years. During that time, X- (3 cm), C- (5 cm) and S-band (10 cm) radars were used to varying degrees in the operational weather radar networks of the day (Whiton et al. 1998a,b). In some cases, mixed-band networks of radars, or multiple, single-band networks, were deployed to serve multipurposed missions.
The first weather radar network was deployed in Panama in April 1944 during World War II (Best 1973). The network was made up of two low-power radars that had an effective range of ~48 km. A second weather radar network was located in India to cover the Assam Valley (Best 1973). The X-band AN/APQ-131 radars that made up this three-radar network would eventually become the first widely distributed ground-based radar used for weather surveillance (Whiton et al. 1998a). By using X-band radars, the radar antennas could be kept small and mobile. At the radar network’s peak, over 60 APQ-13 radars were deployed by the Air Weather Service (AWS) at military bases and post weather stations worldwide. An alternative X-band radar, known as the AN/CPS-9 Storm Detection Radar, was adopted into operation in 1954 for the AWS. About 50 CPS-9s were deployed nationally (Whiton et al. 1998a).
The U.S. Weather Bureau [later known as the National Weather Service (NWS)] deployed its own network of radars using 25 AN/APS-2F (S band) aircraft radars obtained from the navy in 1946. The S-band radars were preferred due to their decreased attenuation from rain (Atlas and Banks 1951). In later years, the AN/APS-2F radars were modified and called WSR-1s, -1As, -3s, and -4s. As a result, the United States had a mix of X- and S-band networks between the Air Weather Service and the Weather Bureau as early as the late 1940s.
A major upgrade to the Weather Bureau’s radar network occurred with the development of the WSR-57 in 1957. The WSR-57 S-band radar would be used for decades, with the last one being decommissioned in 1996. WSR-74S and WSR-74C radars, S and C bands, respectively, were developed to fill in network gaps and replace broken WSR-57s. These three types of radars remained the primary radars for the NWS until the deployment of Next Generation Weather Radar (NEXRAD; Crum and Alberty 1993) (Whiton et al. 1998a). Therefore, a mixed-band radar network existed within the NWS prior to the NEXRAD network.
In 1976, the Joint Doppler Operational Project (JDOP) was established at the National Severe Storms Laboratory (NSSL) to investigate the use of Doppler radar for identifying severe or tornadic thunderstorms (JDOP 1979). The result was a paper that described the advantages of deploying a national Doppler radar network, known as NEXRAD. A network of S-band radars was recommended to replace the WSR-57s, WSR-74Ss, and WSR-74Cs. The first Weather Surveillance Radar-1988 Doppler (WSR-88D) system was installed in 1990; over 160 operational NEXRAD radar systems are now installed throughout the United States and its territories (ROC 2013).
In 1985, the crash of Delta Flight 191 at the Dallas-Fort Worth airport led the Federal Aviation Administration (FAA) to conduct a study into how dangerous, low-level wind shear could be detected (Whiton et al. 1998b). The result was the funding of the C-band Terminal Doppler Weather Radar (TDWR) program by Congress. The FAA chose the C-band frequency for the TDWR radars because of their need for a high maximum unambiguous velocity measurement and concerns about interference from the WSR-88Ds (Whiton et al. 1998b). In addition, the FAA did not require long-range information as needed from NEXRAD. Since the product of the maximum unambiguous range and maximum unambiguous velocity is constant (i.e., the Doppler dilemma), the shorter range from the C band allowed for greater maximum unambiguous velocity measurements with the help of a pulse repetition frequency (PRF) agility scheme (Whiton et al. 1998b).
As demonstrated, the historical use of multiple-band frequencies for weather radar networks in the United States is not without precedence. However, for large networks, a single-band frequency radar was typically used, and the 10-cm (S band) frequency was the band most often employed. The exception was when WSR-74S and WSR-74C radars were used to fill in network gaps and replace broken WSR-57s. The S-band radars require a larger antenna and are usually more expensive, but incur only limited attenuation and greater Nyquist intervals. The X- (3 cm) and C-band (5 cm) radars require smaller antennas and are cheaper to operate, but suffer greater attenuation in heavy precipitation and have smaller Nyquist intervals.
In recent years, mixed-band radar networks have become common in research field programs. The Engineering Research Center for Collaborative Adaptive Sensing of the Atmosphere (CASA; McLaughlin et al. 2009) operated a network of four X-band radars (Junyent et al. 2010) located in southwest Oklahoma within the range of two S-band WSR-88Ds. The X-band Polarimetric Radar Network in the Tokyo Metropolitan Area (X-NET; Maki et al. 2008) operated a network of five X-band radars, and was complemented with a network of four C-band radars. In this network, dual-Doppler analysis provided key information for the detection of significant wind events (Maesaka et al. 2007). In the recent Second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2; Wurman et al. 2012), a suite of mobile C-, X-, Ka- (1 cm), and W-band (3 mm) radars were used to collect radar data at multiple frequencies to analyze more accurately the storm, mesocyclone, and tornado scales. In addition, data collection from multiple radars enabled continuous dual-Doppler coverage of the storms.
With the NEXRAD network aging, the National Oceanic and Atmospheric Administration (NOAA) is evaluating its options for augmentation or replacement of the WSR-88Ds (Crum et al. 2013). With a diversity of radar bands now being sold commercially, a mixed-band network may be an optimal low-cost solution for NOAA to consider. The X- or C-band radars could improve current gaps in low-level coverage and temporal sampling. At the present time, there are significant gaps in low-level coverage, especially in regions of high terrain across the western United States (Fig. 1) (Westrick et al. 1999; Maddox et al. 2002; Beck and Bousquet 2013). Beck and Bousquet (2013) found gaps in the existing French Application Radar à la Météorologie Infrasynoptique (ARAMIS) operational radar network could be supplemented by X-band radars to improve wind-flow retrieval and the accuracy of reflectivity measurements. The temporal volume sampling of NEXRAD ranges from 4.5 to 10 min depending on the volume coverage patterns (VCPs; Office of the Federal Coordinator for Meteorological Services and Supporting Research 2013), which can result in missing critical information in rapidly changing events such as downbursts or tornadoes (e.g., Bluestein et al. 2003). The X- and C-band radars could provide critical low-level coverage and more frequent updates through the use of adaptive scanning.
Low-level coverage for NEXRAD across the contiguous United States. [From ROC (2013).]
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The goal of this paper is to demonstrate qualitatively the advantages of a mixed-band radar network for use in severe weather analysis. Using the experimental CASA network, together with the operational WSR-88Ds, a tornadic event from 13 May 2009 provided an excellent case study to demonstrate the value of a multiband radar network and its potential for use in severe weather operations. The WSR-88D and CASA radar networks are described in more detail in section 2. This section is followed by descriptions of the evolution of the event, including the genesis of the severe convective complex, the enhanced Fujita scale (EF)-2 tornado, the associated tornadic debris signature, and the development of the straight-line wind event with an emphasis on the advantages provided by each measurement system. Discussion and conclusions for the study are discussed in the last section.
2. Data and tools
Between 2006 and 2011, the WSR-88D and CASA radar networks in southwest Oklahoma provided an example configuration of a regional mixed-band radar network that could be used in severe weather operations. In this model, the CASA network was a gap-filling network located midway between two WSR-88D radars, those in Oklahoma City (KTLX) and Frederick (KFDR).
The WSR-88D (Table 1) has several set VCPs for its scanning strategy (Office of the Federal Coordinator for Meteorological Services and Supporting Research 2013). For clear air, VCP 31 and 32 are used to provide enhanced sensitivity to detect low-reflectivity echoes (Office of the Federal Coordinator for Meteorological Services and Supporting Research 2013). These VCPs are useful for detecting the early formation of precipitation and help in identifying airmass boundaries (e.g., cold fronts and the dryline) by detecting the concentration of biological scatterers from the associated convergence along these boundaries. Both of these VCPs have a scanning cycle of 10 min. A 360° scan is taken at 0.5°-, 1.5°-, 2.5°-, 3.5°-, and 4.5°-elevation angles. For severe convective events, VCPs 11, 211, 12, and 212 are primarily used (Office of the Federal Coordinator for Meteorological Services and Supporting Research 2013). In VCP 11 and VCP 211, each scanning cycle takes 5 min and includes 360° scans collected at 0.5°-, 1.5°-, 2.4°-, 3.4°-, 4.3°-, 5.3°-, 6.2°-, 7.5°-, 8.7°-, 10.0°-, 12.0°-, 14.0°-, 16.7°-, and 19.5°-elevation angles. In VCP 12 and 212, each scanning cycle takes 4.5 min. These VCPs are used to increase the low-level vertical resolution over those provided by VCP 11 and 211. For these VCPs, 360° scans are taken at 0.5°-, 0.9°-, 1.3°-, 1.8°-, 2.4°-, 3.1°-, 4.0°-, 5.1°-, 6.4°-, 8.0°-, 10.0°-, 12.5°-, 15.6°-, and 19.5°-elevation angles. The difference between VCPs 11/12 and VCPs 211/212 is that the latter use the Sachidananda–Zrnić algorithm (SZ-2) to mitigate range–velocity ambiguity (Office of the Federal Coordinator for Meteorological Services and Supporting Research 2013).
The WSR-88D network has had recent changes to help in the detection of severe weather. The implementation of superresolution provided an increase of sampling resolution from 1° to 0.5° at the lowest elevation angles. In addition, the gate spacing for reflectivity data decreased from 1 km to 250 m (Torres and Curtis 2007). Brown et al. (2002) found that mesocyclones and tornado signatures can be detected at 50% greater ranges because of superresolution radar data. Superresolution WSR-88D data were available for this event.
More recently, a new adaptive scanning technique called the Automated Volume Scan Evaluation and Termination (AVSET) was implemented into the WSR-88D network (Office of the Federal Coordinator for Meteorological Services and Supporting Research 2013). If enabled, AVSET terminates the current volume scan after the radar has scanned all elevations with significant returns. In other words, the radar skips the higher elevations that do not have significant reflectivity to speed up the volume scan time interval. Therefore, AVSET allows for shorter-timed VCPs for storms at greater distances from the radar. Also, a two-dimensional velocity dealiasing algorithm (2DVDA) was implemented to replace the legacy velocity dealiasing algorithm (VDA) in the WSR-88D network (Zittel and Jing 2012). In this algorithm, a least squares method is used to dealias velocities. Neither AVSET nor 2DVDA were operational at the time of this case study, but both are now available.
Four CASA polarimetric radars (Table 1) made up the Oklahoma testbed; radars were spaced about 30 km apart (Fig. 2). The CASA radar network used an adaptive scanning strategy (Brotzge et al. 2010). Data mining was used to identify significant weather features from the data in real time. Scanning tasks were then optimized over the network to scan these features of interest. Each scanning cycle took 1 min. A 360° surveillance scan was completed at 2.0°-elevation angle, while sector scans were collected at 1.0°, 3.0°, 5.3°, 7.4°, 9.5°, 11.6°, 13.6°, and 15.9°. The surveillance scan took 20 s, and the remaining 40 s were used for targeted sector scanning beginning at 1.0° elevation. Sector scans were completed until the end of the 60 s; thus, not all the elevation angles were completed when wider sector scans were used. A spectral-based clutter suppression was used by the CASA radars (McLaughlin et al. 2009). In this method, spectral decompositions of the dual-polarized measurements are used to discriminate between weather and nonweather signals (Moisseev and Chandrasekar 2009). Moisseev and Chandrasekar (2009) and Moisseev et al. (2008) provide more details on this method of clutter suppression.
The WSR-88D radar scan at 0.5°-elevation angle from KVNX at (a) 2145, (b) 2230, (c) 0000, and (d) 0100 UTC. Note that, at 2145 UTC, KVNX is operating in clear-air mode. Surface wind barbs (full barb ≡ 5 m s−1; half barb ≡ 2.5 m s−1) are shown from the Oklahoma Mesonet. Range rings of 40 km are shown around each CASA radar: KSAO (yellow), KCYR (red), Lawton (KLWE, green), and Rush Springs (KRSP, blue). The location of the fineline is shown by the dotted line in (a). The locations of Eakly (E), Gracemont (G), Lookeba (L), and Mino (M) are indicated in (d).
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The CASA radar data were processed using National Center for Atmospheric Research’s (NCAR) software package REORDER. REORDER uses objective analysis to transform radar data from the radar coordinate system into a Cartesian coordinate system (Oye and Case 1995). Once the objective analysis was complete, dual-Doppler analysis of the CASA data was performed using NCAR’s Custom Editing and Display of Reduced Information in Cartesian Space software package (CEDRIC). CEDRIC is designed for the reduction and analysis of single and multiple Doppler radar volumes in a Cartesian or coplane coordinate system (NCAR 1998).
Attenuation-corrected radar reflectivity data were included on the dual-Doppler analyses; the objective analysis scheme smoothed out any errors or discontinuities in the attenuation-corrected reflectivity. Attenuation correction was performed using the measured differential propagation phase shift in a self-consistent method (Bringi et al. 2001; Liu and Bringi 2006).
The storm-relative wind and vertical vorticity fields were calculated to gain further insights into the evolution of the storm morphology. The storm motion was estimated to be from 350° at 15 m s−1 for the storm-relative wind; the storm motion was estimated by the distance and direction the storm moved over several WSR-88D scans. Bilinear interpretation was used once the calculations were complete. In this method, the parameter values for each cell were determined by bilinear interpolation from the values at its four vertices (Mathworks 2012).
3. Genesis of the convective complex: Clear air and broad overview
An S-band network, such as NEXRAD, can give a broad overview in severe weather warning operations until finer convective details are needed to be analyzed. On 13 May 2009, NEXRAD provided a mechanism to detect the initiation of convection and track the convection until it approached the CASA domain. Once in the CASA domain, X-band radars provided high spatial and temporal observations of tornadogenesis and evolution.
The closest representation of the atmospheric conditions near the tornado was a sounding taken at Norman, Oklahoma (KOUN), at 0000 UTC on 14 May 2009 (Fig. 3a). Surface-based convective available potential energy (CAPE) was ~4600 J kg−1, while mixed layer CAPE (MLCAPE) was ~4900 J kg−1. Surface-based convective inhibition was ~−10 J kg−1. Overall, the atmosphere could be classified as highly unstable with very weak inhibition to prevent convection.
(a) (left) Upper-air sounding and (right) hodograph from KOUN at 0000 UTC 14 May 2009. Courtesy of Storm Prediction Center (SPC). (b) Modified hodograph using 10-m wind measurements from the Minco mesonet station at 0145 UTC. The red circle indicates storm motion.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
Storm-relative environmental helicity (SREH) is one preferred method of assessing environmental wind conditions because of its theoretical link with updraft rotation, sensitivity to storm motion, Galilean invariance, and differentiation between speed and directional shear in the storm’s reference frame (Davies-Jones et al. 1990). The KOUN hodograph was modified to account for backed surface winds near the tornadic storm (Fig. 3b). The mesonet site within the CASA domain the cluster of severe storms impacted first was located near Minco, Oklahoma (Fig. 2), so the modified hodograph was created using data from this radar. At 0145 UTC, just prior to convective contamination, the wind direction backed at the Minco site to 110° at 4.3 m s−1. Using the modified hodograph and the estimated storm motion for the tornadic supercell (from 350° at 15 m s−1), the 0–3-km SREH was ~630 m2 s−2 and the 0–1 km SREH was ~145 m2 s−2. Supercells that can produce tornadoes tend to be found in environments where 0–3-km SREH is at least 150 m2 s−2 (Davies-Jones et al. 1990).
Among the greatest benefits of the S-band radar is its enhanced sensitivity in clear air. Before the initiation of convection, the Vance Air Force Base, Oklahoma (KVNX), WSR-88D was using the VCP 32 scan strategy—clear-air mode (Fig. 2a). The enhanced sensitivity from the clear-air mode detected a frontal boundary (a cold front), which is indicated by a fine line of higher radar reflectivity. The data provided an area of interest to watch for convective initiation. The ability to determine boundary locations is beneficial for areas without a dense observing network.
Convection began to initiate along the cold front in northwestern Oklahoma around 2230 UTC on 13 May 2009 (Fig. 2b), as seen by the KVNX radar. At this time, the WSR-88D was using the VCP 12 (convective) scan strategy. The convection organized over the next 2 h and almost evolved into a quasi-linear convective system (QLCS) by 0000 UTC on 14 May 2009; however, individual convective cells within the line still had supercell characteristics (Fig. 2c). The line of convection quickly broke down into three distinct clusters by 0100 UTC (Fig. 2d); only two of three clusters are shown in the Fig. 2d.
The southernmost convective cluster moved almost due south toward the CASA testbed (recall the estimated storm motion was ~350° at about 15 m s−1). Hail in excess of 3 cm in diameter was reported in Lookeba, Oklahoma (Fig. 2d), at around 0145 UTC, and hail in excess of 4.5 cm in diameter was reported at 0155 UTC in Eakly, Oklahoma (National Weather Service Norman Forecast Office 2013). The leading edge of this complex eventually entered the CASA domain just prior to 0200 UTC.
The S-band radar provided broad-scale surveillance. By 0208 UTC, two very broad circulation patterns were located within the convective complex, as seen from the KFDR WSR-88D scan at 0.5°-elevation angle (Fig. 4a). The maximum inbound ground-relative Doppler wind speed was ~19 m s−1 and the maximum outbound was ~13 m s−1 in the eastern circulation. In the western circulation, the maximum inbound velocity was ~25 m s−1 and the maximum outbound was ~17 m s−1.
The WSR-88D radar scan at 0.5°-elevation angle from KFDR at (a) 0208 and (b) 0218 UTC showing (left) reflectivity (dBZ) and (right) radial velocity (m s−1). Range rings are in kilometers. Two broad circulation patterns are circled in white in the velocity panels. The left circulation is ~1.7–1.8 km ARL, and the right circulation is ~2.0–2.1 km ARL.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The eastern circulation pattern was not well organized until around 0218 UTC, as the western circulation was weakening (Fig. 4b). At this time, a mesocyclone was identified by a velocity couplet at ~2.1 km above radar level (ARL) and was located now within the CASA domain near Gracemont, Oklahoma (Fig. 2d). Radial winds had increased substantially over the 10-min period by 0218 UTC. The maximum inbound ground-relative wind speed was ~38 m s−1, and the maximum outbound was ~16 m s−1. The inbound velocity would be expected to be higher since the storm motion was toward the radar. The maximum reflectivity at this time was still ~71 dBZ. At 0220 UTC, the mesocyclone entered the dual-Doppler region within the CASA domain.
4. EF2 tornado
The X-band radar network provided much greater detail of the storm in time and space. Dual-Doppler analyses using data from the CASA radars near Chickasha (KSAO) and Cyril (KCYR), Oklahoma, were performed for the period from 0220 to 0230 UTC, with horizontal grid spacing of 0.65 km and at time intervals of 1 min. The analyses were all conducted at ~0.6 km ARL.
The dual-Doppler analysis at 0220 UTC (Fig. 5a) identifies supercellular features within the convective storm: a hook echo is depicted in the reflectivity from KSAO, while the storm-relative horizontal wind vectors depict a well-defined rear-flank gust front (RFGF; e.g., Markowski 2002) and storm-relative inflow from the south. The RFGF was located along the maximum convergence (denoted by the dotted black line). No forward-flank gust front (FFGF; e.g., Shabbott and Markowski 2006) is clearly evident in the analysis domain. A broad area of vertical vorticity with two distinct maxima is present in the analysis; the maximum vertical vorticity is ~0.018 s−1, which is on the order of the vertical cyclonic vorticity defined in a mesocyclone (≥10−2 s−1) (Glickman 2013). Therefore, the broad area of vorticity represents the mesocyclone in the supercell. The RFGF is to the east and southeast of the mesocyclone; while it is difficult to deduce if an occlusion is present, one may surmise an occlusion is in process since the RFGF is about to separate from the mesyclone to the east of the broad mesocyclone. An occlusion occurs in a supercell when the rear-flank downdraft (RFD) reaches the ground and curves around the mesocyclone, cutting off the warm, moist air from entering the updraft near the mesocyclone (Bluestein 2007), similar to what is seen in synoptic-scale extratropical cyclones (Bluestein 1993). This particular analysis fits the Lemon and Doswell (1979) conceptual model with one exception—there was no clear FFGF present in the analysis domain.
Storm-relative dual-Doppler analyses (1 km MSL or ~0.6 km ARL) at (a) 0220, (b) 0221, and (c) 0222 UTC. Horizontal storm-relative wind vectors with (top) KSAO attenuation-corrected equivalent radar reflectivity factor (dBZe), (middle) vertical vorticity (s−1), and (bottom) storm-relative horizontal wind speed (m s−1). The convergence boundary is depicted by the dotted blank line. The mesocyclone is circled in black.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
In the 0221 UTC analysis (Fig. 5b), higher storm-relative wind is present on the north side of the mesocyclone. This region of enhanced storm-relative wind that results in an asymmetric wind pattern around the mesocyclone will be referred to as a surge. The vorticity maximum has strengthened to ~0.024 s−1. The RFGF is well defined by the increased storm-relative wind speed and wind shift located south and southeast of the mesoyclone; regions of higher vertical vorticity are analyzed ahead of the RFGF. At 0222 UTC (Fig. 5c), the analysis indicates that the surge continues to intensify just north of a strengthening vertical vorticity maximum (~0.030 s−1). The broad mesocyclone has consolidated to a single vorticity maximum. The RFGF continues to move southward and eastward; the RFGF appears to be completely separated from the mesocyclone in this analysis. Higher vertical vorticity values indicate that a new mesocyclone is attempting to develop on the eastern edge of the elliptical pattern on the northern tip of the RFGF as in cyclic mesocyclogenesis (e.g., Burgess et al. 1982); however, the analyses suggest that a closed circulation did not develop.
The single radar analysis from the X-band radars was also helpful. The 0223 UTC KSAO radar scan depicts a well-defined hook echo on radar reflectivity (Fig. 6b). The wraparound of the reflectivity around the mesocyclone is significantly more defined in the KSAO data than in the KTLX data at 0224 UTC (Fig. 6a). This better-defined structure is primarily a result of the CASA radar’s closer range (29 km) to the storm compared to KTLX’s range (88 km). The CASA radar’s closer gate spacing was a secondary factor in the better resolution. The wraparound reflectivity structure on this single-radar image would give a forecaster cause to investigate additional radar products. A benefit of the WSR-88D radar also is seen because of the significant attenuation on KSAO; this attenuation is not present on KTLX.
(a) Radar reflectivity (dBZe) from KSAO at 2.0°-elevation angle and (b) radar reflectivity (dBZ) from KTLX at 0.5°-elevation angle. The hook echo is located ~29 km to the northwest of KSAO (~1.1 km ARL) and ~88 km to the west of KTLX (~1.4 km ARL). These images are depicted using Warning Decision Support System–Integrated Information (WDSS-II) software (Lakshmanan et al. 2007).
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The dual-Doppler analysis at 0223 UTC (Fig. 7a) shows the mesocyclone is continuing to move southward with a well-defined surge to the north and west of the vertical vorticity maximum (~0.031 s−1). A second surge of enhanced storm-relative wind was present on the southern side of the mesocyclone. The vorticity maximum near the tip of the RFGF remains constant and ill-defined. At 0224 UTC (Fig. 7b), the analysis indicates the intensification process of the mesocyclone has slowed down, as suggested by nearly constant vertical vorticity (~0.031 s−1) between 0223 and 0224 UTC. The southern extent of the RFGF is more difficult to see because it has expanded outward into areas of weaker reflectivity. The surge on the northern side of the mesocyclone has rotated counterclockwise around the mesocyclone; the strongest storm-relative winds are now located on the northwestern and western sides of the mesocyclone. By the 0225 UTC (Fig. 7c) analysis, the western surge has increased in intensity as it rotated around the mesocyclone (~0.031 s−1); the western surge has now connected with the second surge on the southern side of the mesocyclone. As a result, the mesocyclone’s wind pattern is transitioning to become more symmetric. The NWS damage survey indicated that tornado damage began at 0226 UTC (National Weather Service Norman Forecast Office 2013), so an analysis at 0225 UTC represents the storm structure just prior to tornadogenesis.
As in Fig. 5, but at (a) 0223, (b) 0224, and (c) 0225 UTC.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The 0226 UTC (Fig. 8a) analysis indicates the low-level mesocyclone is at its peak intensity at ~0.033 s−1 of vorticity. While a tornado cannot be resolved in this analysis due to the relatively course grid size, a tornado was present according to the damage survey. Note that the tornado developed while the RFGF was well ahead of and detached from the mesocyclone. The wind pattern around the mesocyclone continues to look more symmetric. Note that the mesocyclone appears to be embedded in precipitation. By the 0227 UTC (Fig. 8b) analysis, the vorticity maximum has fallen slightly to ~0.031 s−1. The mesocyclone weakens even further at 0228 UTC (Fig. 8c) with a magnitude of vertical vorticity at ~0.028 s−1. Any surges have become ill-defined as a result of the mesocyclone becoming more symmetric. The leading edge of the RFGF is several kilometers to the east of the mesocyclone. The wind speed is beginning to increase to the southwest of the mesocyclone; this region of stronger winds is a separate entity from the RFD of the embedded supercell.
As in Fig. 5, but at (a) 0226, (b) 0227, and (c) 0228 UTC. The X denotes the center of the velocity couplet in the KCYR single-Doppler radar data. This represents the approximate tornado location.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The greater detail provided by the X-band radars may improve our ability to discern potential tornado precursors. In addition to the dual-Doppler analysis, 1-min KCYR scans from 0224 to 0227 UTC indicate the rapid contraction of the mesocyclone, coinciding with the period of tornadogenesis (Fig. 9). Note that the clutter suppression algorithm results in somewhat noisy reflectivity near the velocity couplet. As in the dual-Doppler analyses, higher velocities are initially collocated with the surge on the western side of the mesocyclone. As the contraction occurs, the outbound radial velocities increase quickly and the mesocyclone intensifies; the gate-to-gate distance between the maximum outbound and inbound radial velocities decreases substantially. The presence of a smaller-scale couplet (within the broader couplet) at 0225 UTC indicates that a failed tornadogenesis may have occurred between 0224 and 0225 UTC. The maximum outbound and inbound radial velocities in the smaller-scale couplet were weaker than the maximum outbound and inbound radial velocities in the broader couplet. This smaller-scale couplet dissipates at 0226 UTC as the broader couplet contracts. The contraction of the broader couplet is complete at 0227 UTC, as indicated by the nearly gate-to-gate distance between the maximum outbound and inbound radial velocities. This contraction appears to be similar to the evolution in Figs. 10 and 11 of Snyder et al. (2013) and Fig. 3 in Wakimoto et al. (2011). The radial velocity pattern becomes more symmetric between 0224 and 0227 UTC, consistent with the behavior of the mesocyclone in the dual-Doppler analysis. It is unknown whether the surge and the mesocyclone contraction are related or a coincidence. The contraction of the mesocyclone is somewhat evident between the two KFDR scans at 0222 and 0227 UTC (Fig. 10); KFDR was scanning the mesocyclone at ~2.1 km ARL. However, it is much easier to resolve the mesocyclone contraction in the KCYR radar data because of the radar’s higher spatial and temporal resolutions. Also, as in the dual-Doppler analyses, the mesocyclone is well behind the subjectively identified RFGF (Fig. 9; black lines).
KCYR radar scans from 0224 to 0225 UTC at 2.0°-elevation angle showing (top) reflectivity (dBZe) and (bottom) radial velocity (m s−1). The maximum outbound radial velocity is indicated by a plus sign, and the maximum inbound radial velocity is indicated by a minus sign. The radius of the distance between the two points is indicated by a white circle, except in the last panel. The primary RFGF is depicted by the dotted blank line.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
KFDR radar scans from 0224 and 0227 UTC at 0.5°-elevation angle showing (top) reflectivity (dBZ) and (bottom) radial velocity (m s−1).
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
Overall, dual- and single-Doppler analyses indicate that a surge of wind propagated counterclockwise on the western side of the mesocyclone just prior to tornadogenesis. During this period, the surge moved inward and the mesocyclone contracted and increased in intensity. It cannot be deduced whether the surge and the mesocyclone contraction are related; however, the contraction of enhanced velocity and the coincident mesocyclone intensification have been seen in radar observations (e.g., Snyder et al. 2013). Over time, the mesocyclone transitioned to a more symmetric wind pattern as the surge wrapped around the mesocyclone. Recent data analyses from VORTEX2 have documented the wrapping surges preceding near-surface vortex intensification (Kosiba et al. 2013). Kosiba et al. (2013) attributed this to a secondary RFD, which has been observed in other studies using various terminology (e.g., Marquis et al. 2008; Wurman et al. 2010; Skinner et al. 2011; Lee et al. 2012; Marquis et al. 2012); however, without any reliable vertical velocity estimates, the source of the surge can only be speculated upon as a downdraft in this case study.
An X-band network, similar to CASA, could provide real-time dual-Doppler analyses and high-resolution single-Doppler analyses, allowing for the ability to detect and track surges and mesocyclone contraction. In this example, the dual-Doppler analysis allowed for a more accurate measurement of the wind velocity, and not just the component parallel to the radar beam. This allowed for the easy detection of the wind shift associated with the RFGF. One-minute sampling also provided additional information on the evolving storm dynamics at low levels. In this particular event, 1-min updates provided the temporal resolution to detect the evolution of a surge and the coincident contraction of the mesocyclone. As the mesocyclone contracted, tornadogenesis occurs in the supercell.
5. Tornadic debris signature
Many studies have shown that a tornadic debris signature (TDS) can be detected from copolar correlation coefficient (ρhv) measurements (e.g., Ryzhkov et al. 2005; Bluestein et al. 2007; Kumjian and Ryzhkov 2008; Snyder et al. 2010; Palmer et al. 2011; Tanamachi et al. 2012). A TDS is produced when a tornado lofts debris to heights observable by the radar. The high density and random shapes of the debris cause the ρhv to drop. The ρhv minimum must be collocated with the velocity couplet and of sufficiently high reflectivity (i.e., reflectivity several dB greater than the signal-to-noise ratio).
The higher time and space resolutions provided by the X-band system aided in the identification of polarimetric features. At 0231 UTC (Fig. 11a), a ρhv minimum began to develop at the center of the velocity couplet as detected by the KSAO CASA radar. These low values may not be significant at this time since there were similar ρhv values to the west of the vortex; however, this signature became better defined at 0232 UTC (Fig. 11b). By 0233 UTC (Fig. 12a), the ρhv minimum could be easily recognized at the center of the velocity couplet (~0.98 km ARL). The signature could be seen simultaneously from the KCYR CASA radar as well (~1.00 km ARL) (Fig. 12b). Because debris (i.e., nonmeteorological targets) was present, some of the reflectivity may have been suppressed within parts of the velocity couplet by the spectral-based clutter filter (Moisseev and Chandrasekar 2009), which demonstrates a disadvantage of a spectral-based clutter filter. However, the equivalent radar reflectivities measured by both radars were ≥20 dBZe in the lowest ρhv measurements at 0233 UTC (greater than the signal-to-noise ratio). In this particular event, a TDS developed between 0230 and 0233 UTC; the TDS became undetectable after 0237 UTC by both radars (not shown).
The CASA radar scan at 2.0°-elevation angle from KSAO at (a) 0231 and (b) 0232 UTC showing (left) reflectivity (dBZe), (middle) radial velocity (m s−1), and (right) ρhv. Range rings are in kilometers. The area of low ρhv, possibly damage associated with a tornado, is circled in white in all the panels.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The CASA radar scan at 2.0°-elevation angle at 0233 UTC from (a) KSAO and (b) KCYR. (left) reflectivity (dBZe), (middle) radial velocity (m s−1), and (right) ρhv. Range rings are in kilometers. The area of low ρhv, possibly damage associated with a tornado, is circled in white in all the panels.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The scenario is instructive because two radars with overlapping coverage indicated the presence of a TDS. This should increase forecaster confidence that there was a tornado lofting debris, demonstrating one advantage to having overlapping, low-level scanning, dual-polarized radars.
6. Straight-line wind event
The geometric juxtaposition of multiple radars within range of an event improves the odds that the maximum wind velocities will be measured. The 0231 UTC KFDR scan at 0.5°-elevation angle continued to depict a strong mesocyclone. In addition, an area of enhanced inbound radial velocities was present to the southwest of the mesocyclone (Fig. 13a). Initially, cyclonic shear may have partially contributed to this area of enhanced velocity (Fig. 13b); however, the shear decreased even as the inbound velocity increased (Fig. 13c). The inbound radial velocity increased at 0241 UTC with values just over 30 m s−1 at ~1.7 km ARL from KFDR data (Fig. 14a). In comparison, KCYR depicted ~43 m s−1 inbound radial velocity at ~0.25 km ARL (Fig. 14b). KSAO did not scan this region at a 1.0°-elevation angle at this particular time, which was a limitation of the automated adaptive scanning; however, an advantage of having a networked approach with overlapping coverage is the increased likelihood of not missing an event. Even though KSAO did not scan the region of interest, KCYR did. The period between 0231 and 0241 UTC was the time frame of greatest intensification of these straight-line winds to the southwest of the mesocyclone. In addition, the mesocyclone became much broader and weaker at KFDR by 0241 UTC. KCYR continued to observe a low-level mesocyclone, but it had weakened. The weakening of the low-level mesocyclone is in accord with the damage survey that indicated the tornado had dissipated by 0240 UTC.
(right) KFDR WSR-88D cross-sectional analyses of radar reflectivity (dBZ) at (a) 0231, (b) 0236, and (c) 0241 UTC. The descending reflected core is located within the white ovals. (left) Cross sections relative to the Doppler velocity (m s−1) field (white line).
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
The radar scans at 0241 UTC from (a) KFDR at 0.5°-elevation angle and from (b) KCYR at 1.0°-elevation angle. The region of strong straight-line winds is circled in white. The height of the strongest winds is ~1.7 km ARL at KFDR and ~0.25 km ARL at KCYR.
Citation: Weather and Forecasting 29, 1; 10.1175/WAF-D-13-00024.1
Cross sectional analysis from the KFDR radar of volume scans with starting times at 0231, 0236 and 0241 UTC indicated that these stronger winds coincided with the descent of a reflectivity core aloft down to the near surface (Fig. 13). This analysis suggests these strong winds may have been associated with a downburst (e.g., Wakimoto and Bringi 1988). Downbursts have been classified by length and precipitation amount. Microbursts are <4 km in length and usually have winds that last 2–5 min. Macrobursts are >4 km in length, and are especially common in bow echoes (Glickman 2013). The spatial and temporal scales of the strong winds suggest that these straight-line winds could be classified as a macroburst. KFDR measured >30 m s−1 (>67 mi h−1) radial velocity on its lowest elevation angle with the downburst through 0255 UTC (not shown), though the peak was around 0241 UTC. KCYR radar measurements became significantly attenuated by heavy precipitation as the macroburst moved over the radar site. The advantage of a mixed-band radar network is that additional radars, such as NEXRAD in this case, can provide less attenuated measurements.
This analysis shows an advantage of the S-band network, which provides volumetric coverage in conjunction with an X-band network that provides low-level coverage. In this particular event, the NEXRAD radar detected a descending reflectivity core and strong winds aloft associated with a downburst. Once the descending reflectivity core and strong winds were detected, the CASA radars could be utilized to measure more accurately the speed and direction of the winds near the surface associated with the downburst. The CASA radars measured ~43 m s−1 (96 mi h−1) inbound radial velocity very near the surface—significantly stronger than the radial velocity measurements by KFDR. CASA provided a low-level scan with an adequate sampling frequency and viewing angle to detect the strong winds near the surface.
7. Conclusions and discussion
The 13 May 2009 case is an example event in which a mixed-band radar network proves advantageous for severe weather analysis and provides an example of what could be implemented into operations. In this case study, WSR-88D (S band) radars were used in conjunction with the CASA (X band) testbed radars.
Overall, each type of radar contributed to an improved analysis. The WSR-88D radars gave a broad overview and excellent upper-level coverage; the CASA radars provided higher-resolution analyses with rapid, low-level scans. Several storm features were identified with the X-band radars that would normally have been missed when using only the WSR-88Ds.
For this event, the following specific advantages of the NEXRAD are identified:
NEXRAD provided varied VCPs that evolved with the situation. In this event, VCP 32 (clear air) and VCP 12 (convective) were used. The high-sensitivity of the clear-air VCP allowed for the detection and tracking of the cold front, where convective initiation occurred. Once initiation was imminent, the VCP was changed to 12.
NEXRAD provided a broad overview of the situation, as the WSR-88D is less susceptible to attenuation and has a large maximum unambiguous range. Once the mesocyclone was detected by NEXRAD, the CASA radars were used for further interrogation.
NEXRAD provided upper-level scans up to 19.5°. Mid- to upper-level features may act as precursors to surface events. In this particular case, a descending reflectivity core and strong winds aloft indicated that a downburst was occurring. As a result, further investigation could be done using the CASA radars.
CASA provided 1-min updates. High temporal (60 s) resolution data provide useful information in highly unsteady, dynamic events, such as supercells and straight-line wind events. In this particular event, 1-min updates provided adequate sampling frequency to detect a surge in enhanced velocity and the contraction of the mesocyclone. Further study on the relationship between the surge and the mesocyclone contraction is needed. As the mesocyclone contracted inward, the tornado developed. Neither of these features were adequately sampled by NEXRAD.
CASA provided overlapping coverage. The overlapping coverage allowed for dual-Doppler analysis, which permits a more accurate estimate of the true wind speed and derived quantities such as vertical vorticity. The RFGF was easily tracked using dual-Doppler analyses. Overlapping coverage also provides redundant information that may increase forecaster confidence. In this case, a TDS was detected simultaneously from two radars.
CASA provided low-level coverage. The goal of severe weather operations is to warn for weather impacts at the surface. In this event, rapid, low-level coverage from CASA allowed for a more accurate estimate of the surface winds associated with a downburst than could be estimated from NEXRAD. The downburst’s winds were nearly parallel to the CASA radar’s beam, allowing the radial component to measure nearly the full wind speed. The difference in this case was 13 m s−1 (29 mi h−1) or a 43% increase in the maximum radial velocity measurement.
The value of rapid-scan data has been demonstrated from a multitude of large fixed-location radars such as S-band phased-array radar (e.g., Heinselman et al. 2008; Heinselman and Torres 2011; Heinselman et al. 2012), WSR-88Ds (Kumjian et al. 2010), and the C-band University of Oklahoma Polarimetric Radar for Innovations in Meteorology and Engineering (OU-PRIME; Palmer et al. 2011). The advantages of the rapid-scan approach have also been demonstrated by mobile radar systems (e.g., Bluestein et al. 2010; Pazmany et al. 2013; Isom et al. 2013; French et al. 2013). Mobile radar systems have also been used to demonstrate the value of high-resolution, low-level, and dual-Doppler capabilities (e.g., Tanamachi et al. 2012, Kosiba et al. 2013). However, it is the combination of these advantages made available from a fixed network of X-band radars that allows for the overlapping, low-level, and rapid scanning necessary for the routine measurement of finescale, transient storm features.
A mixed-band network could become operational today. For the 13 May 2009 event, the X-band radar data and assimilated products were made available to the local NWS Weather Forecasting Office (WFO) through the National Oceanic and Atmospheric Administration’s (NOAA) Hazardous Weather Testbed (HWT; Clark et al. 2012). Data from this case aided the NWS local WFO in its issuance of the NWS tornado warning; a simultaneous, nonoperational experiment in the HWT issued the tornado warning 3 min earlier than did the NWS based solely upon CASA radar data and products (Philips et al. 2010). As a test project, the Dallas–Fort Worth (DFW) Urban Demonstration Network (Philips et al. 2012) is being developed across the DFW Metroplex. Once complete, this system will utilize data from eight X-band radars, two C-band Terminal Doppler Weather Radars (TDWRs), and one WSR-88D. Dual-Doppler analysis is provided in real time at a resolution of 100 m (see Fig. 17 in Junyent et al. 2010). Assimilating data from radar, satellite, aircraft, surface, and sounding systems, a three-dimensional variational data assimilation analysis (3DVAR; Gao et al. 2013) is provided every 30 min at a resolution of 400 m (Gao et al. 2010), and 2-h forecasts are provided every 30 min at a resolution of 1 km (Brewster et al. 2010). All of these products are made available to local NWS forecasters, media outlets, and emergency managers for use in operations.
A mixed network of radars, including X-, C-, or S-band radars, may present some new challenges. One consideration is data overload (Heinselman et al. 2012). Real-time data collected every 60 s from dozens of radars have the potential to overwhelm a forecaster. Another issue is the variation in quality control across radar systems, including differences in attenuation correction, clutter mitigation, dealiasing, sensitivity, resolution, and receiver noise. Solutions will be needed to integrate and visualize these data to improve the forecaster’s situational awareness and warning capability. Despite these caveats, a mixed-band radar network appears to offer several key advantages for severe weather analysis over a single-frequency network. Such an approach may provide a cost-effective, optimal solution for planning the next-generation radar networks.
Acknowledgments
The authors thank the anonymous reviewers, who provided constructive feedback on the manuscript. The authors also thank Drs. Michael Biggerstaff and Frederick Carr of the School of Meteorology for providing feedback while this work was in the form of the first author’s M.S. thesis. This work is supported by the Engineering Research Centers Program of the National Science Foundation under NSF Award 0313747. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Science Foundation.
REFERENCES
Atlas, D., and Banks H. C. , 1951: The interpretation of microwave reflections from rainfall. J. Meteor., 8, 271–282.
Beck, J., and Bousquet O. , 2013: Using gap-filling radars in mountainous regions to complement a national radar network: Improvements in multiple-Doppler wind syntheses. J. Appl. Meteor. Climatol., 52, 1836–1850.
Best, W. H., Jr., 1973: Radars over the hump—Recollections of the first weather radar network. Bull. Amer. Meteor. Soc., 54, 205–208.
Bluestein, H. B., 1993: Observations and Theory of Weather Systems. Vol. II, Synoptic–Dynamic Meteorology in Midlatitudes, Oxford University Press, 594 pp.
Bluestein, H. B., 2007: Advances in applications of the physics of fluids to severe weather systems. Rep. Prog. Phys., 70, 1259–1323.
Bluestein, H. B., Lee W.-C. , Bell M. , Weiss C. C. , and Pazmany A. L. , 2003: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part II: Tornado-vortex structure. Mon. Wea. Rev., 131, 2968–2984.
Bluestein, H. B., French M. M. , Tanamachi R. L. , Frasier S. , Hardwick K. , Junyent F. , and Pazmany A. , 2007: Close-range observations of tornadoes in supercells made with a dual-polarization, X-band, mobile Doppler radar. Mon. Wea. Rev., 135, 1522–1543.
Bluestein, H. B., French M. M. , PopStefanija I. , Bluth R. T. , and Knorr J. B. , 2010: A mobile, phased-array Doppler radar for the study of severe convective storms. Bull. Amer. Meteor. Soc., 91, 579–600.
Brewster, K., Thomas K. W. , Gao J. , Brotzge J. , Xue M. , and Wang Y. , 2010: A nowcasting system using full physics numerical weather prediction initialized with CASA and NEXRAD radar data. Preprints, 25th Conf. on Severe Local Storms, Denver, CO, Amer. Meteor. Soc., 9.4. [Available online at https://ams.confex.com/ams/pdfpapers/176053.pdf.]
Bringi, V. N., Keenan T. D. , and Chandrasekar V. , 2001: Correcting C-band radar reflectivity and differential reflectivity data for rain attenuation: A self-consistent method with constraints. IEEE Trans. Geosci. Remote Sens., 39, 1906–1915.
Brotzge, J., Hondl K. , Phillips B. , Lemon L. , Bass E. J. , Rude D. , and Andra D. L. Jr., 2010: Evaluation of distributed collaborative adaptive sensing for the detection of low-level circulation and implications for severe weather warning operations. Wea. Forecasting, 25, 173–189.
Brown, R. A., Wood V. T. , and Sirmans D. , 2002: Improved tornado detection using simulated and actual WSR-88D data with enhanced resolution. J. Atmos. Oceanic Technol., 19, 1759–1771.
Burgess, D. W., Wood V. T. , and Brown R. A. , 1982: Mesocyclone evolution statistics. Preprints, 12th Conf. on Severe Local Storms, San Antonio, TX, Amer. Meteor. Soc., 422–424.
Clark, A. J., and Coauthors, 2012: An overview of the 2010 Hazardous Weather Testbed Experimental Forecast Program Spring Experiment. Bull. Amer. Meteor. Soc., 93, 55–74.
Crum, T. D., and Alberty R. L. , 1993: The WSR-88D and the WSR-88D Operational Support Facility. Bull. Amer. Meteor. Soc., 74, 1669–1687.
Crum, T. D., Chrisman J. N. , Smith S. D. , Istok M. J. , Hall R. W. , and Saffle R. , 2013: WSR-88D radar projects: 2013 update. Preprints, 29th Conf. on Environmental Information Processing Technologies, Austin, TX, Amer. Meteor. Soc., 6B.1. [Available online at https://ams.confex.com/ams/93Annual/webprogram/Paper221461.html.]
Davies-Jones, R. P., Burgess D. W. , and Foster M. , 1990: Test of helicity as a forecast parameter. Preprints, 16th Conf. on Severe Local Storms, Kananaskis Park, AB, Canada, Amer. Meteor. Soc., 588–592.
French, M. M., Bluestein H. B. , PopStefanija I. , Baldi C. A. , and Bluth R. T. , 2013: Reexamining the vertical development of tornadic vortex signatures in supercells. Mon. Wea. Rev., 141, 4576–4601.
Gao, J., Brewster K. , Xue M. , Brotzge J. , Thomas K. , and Wang Y. , 2010: Real-time, low-level wind analysis including CASA and WSR-88D radar data using the ARPS 3DVAR. Preprints, 25th Conf. Severe Local Storms, Denver, CO, Amer. Meteor. Soc., 7B.4. [Available online at https://ams.confex.com/ams/pdfpapers/176005.pdf.]
Gao, J., and Coauthors, 2013: A real-time weather-adaptive 3DVAR analysis system for severe weather detections and warnings. Wea. Forecasting, 28, 727–745.
Glickman, T., Ed., 2000: Glossary of Meteorology. 2nd ed. Amer. Meteor. Soc., 855 pp.
Heinselman, P. L., and Torres S. M. , 2011: High-temporal-resolution capabilities of the National Weather Radar testbed phased-array radar. J. Appl. Meteor. Climatol., 50, 579–593.
Heinselman, P. L., Priegnitz D. L. , Manross K. L. , Smith T. M. , and Adams R. W. , 2008: Rapid sampling of severe storms by the National Weather Radar testbed phased array radar. Wea. Forecasting, 23, 808–824.
Heinselman, P., LaDue D. , and Lazrus H. , 2012: Exploring impacts of rapid-scan radar data on NWS warning decisions. Wea. Forecasting, 27, 1031–1044.
Isom, B., and Coauthors, 2013: The atmospheric imaging radar: Simultaneous volumetric observations using a phased array weather radar. J. Atmos. Oceanic Technol., 30, 655–675.
JDOP, 1979: Final report on the Joint Doppler Operational Project. NOAA Tech. Memo. ERL NSSL-86, Norman, OK, 84 pp.
Junyent, F., Chandrasekar V. , McLaughlin D. , Insanic E. , and Bharadwaj N. , 2010: The CASA Integrated Project 1 networked radar system. J. Atmos. Oceanic Technol., 27, 61–78.
Kosiba, K., Wurman J. , Richardson Y. , Markowski P. , Robinson P. , and Marquis J. , 2013: Genesis of the Goshen County, Wyoming, tornado on 5 June 2009 during VORTEX2. Mon. Wea. Rev., 141, 1157–1181.
Kumjian, M. R., and Ryzhkov A. V. , 2008: Polarimetric signatures in supercell thunderstorms. J. Appl. Meteor. Climatol., 47, 1940–1961.
Kumjian, M. R., Ryzhkov A. V. , Melnikov V. M. , and Schuur T. J. , 2010: Rapid-scan superresolution observations of a cyclic supercell with a dual-polarization WSR-88D. Mon. Wea. Rev., 138, 3762–3786.
Lakshmanan, V., Smith T. , Stumpf G. J. , and Hondl K. , 2007: The Warning Decision Support System–Integrated Information. Wea. Forecasting, 22, 596–612.
Lee, B. D., Finley C. A. , and Karstens C. D. , 2012: The Bowdle, South Dakota, cyclic tornadic supercell of 22 May 2010: Surface analysis of rear-flank downdraft evolution and multiple internal surges. Mon. Wea. Rev., 140, 3419–3441.
Lemon, L. R., and Doswell C. A. III, 1979: Severe thunderstorm evolution and mesocyclone structure as related to tornadogenesis. Mon. Wea. Rev., 107, 1184–1197.
Liu, Y., and Bringi V. N. , 2006: Improved rain attenuation correction algorithms for radar reflectivity and differential reflectivity with adaptation to drop shape model variation. Proc. 26th Int. Geoscience and Remote Sensing Symp., Denver, CO, Institute of Electrical and Electronics Engineers, 1910–1913.
Maddox, R. A., Zhang J. , Gourley J. J. , and Howard K. W. , 2002: Weather radar coverage over the contiguous United States. Wea. Forecasting, 17, 927–934.
Maesaka, T., Maki M. , Iwanami K. , Misumi R. , and Shimizu S. , 2007: Real-time wind field retrieval system by using X-band radar network around Tokyo metropolitan area Preprints, 33rd Conf. on Radar Meteorology, Cairns, QLD, Australia, Amer. Meteor. Soc., P8B.11. [Available online at https://ams.confex.com/ams/pdfpapers/123537.pdf.]
Maki, M., and Coauthors, 2008: X-band polarimetric radar network in the Tokyo metropolitan area—X-NET. Proc. Fifth European Conf. on Radar Meteorology, Helsinki, Finland, ERAD, 3.7. [Available online at http://www.erad2010.org/pdf/POSTER/Thursday/02_Xband/11_ERAD2010_0354_extended.pdf.]
Markowski, P. A., 2002: Hook echoes and rear-flank downdrafts: A review. Mon. Wea. Rev., 130, 852–876.
Marquis, J., Richardson Y. , Wurman J. , and Markowski P. M. , 2008: Single- and dual-Doppler analysis of a tornadic vortex and surrounding storm-scale flow in the Crowell, Texas, supercell of 30 April 2000. Mon. Wea. Rev., 136, 5017–5043.
Marquis, J., Richardson Y. , Markowski P. , Dowell D. , and Wurman J. , 2012: Tornado maintenance investigated with high-resolution dual-Doppler and EnKF analysis. Mon. Wea. Rev., 140, 3–27.
Mathworks, cited2012: Product documentation: pcolor. [Available online at http://www.mathworks.com/help/techdoc/ref/pcolor.html.]
McLaughlin, D., and Coauthors, 2009: Short-wavelength technology and the potential for distributed networks of small radar systems. Bull. Amer. Meteor. Soc., 90, 1797–1817.
Moisseev, D. N., and Chandrasekar V. , 2009: Polarimetric spectral filter for adaptive clutter and noise suppression. J. Atmos. Oceanic Technol., 26, 215–228.
Moisseev, D. N., Nguygen C. M. , and Chandrasekar V. , 2008: Clutter suppression for staggered PRT waveforms. J. Atmos. Oceanic Technol., 25, 2209–2218.
National Weather Service Norman Forecast Office, cited 2013: Storm data and unusual weather phenomena - May 2009. [Available online at http://www.srh.noaa.gov/media/oun/stormdata/oun200905.pdf.]
NCAR, 1998: CEDRIC: Custom Editing and Display of Reduced Information in Cartesian Space. National Center for Atmospheric Research, 130 pp. [Available online at http://www.eol.ucar.edu/content/cedric-reference-manual.]
Office of the Federal Coordinator for Meteorological Services and Supporting Research, 2013: Doppler radar meteorological observations. Part A: System concepts, responsibilities, and procedures. Federal Meteorological Handbook No. 11, FCM-H11A-2013, 49 pp. [Available online at http://www.ofcm.gov/fmh11/fmh11.htm.]
Oye, D., and Case M. , 1995: REORDER: A program for gridding radar data installation and use manual for the Unix version. National Center for Atmospheric Research, 44 pp. [Available online at http://www.eol.ucar.edu/content/reorder-reference-manual.]
Palmer, R. D., and Coauthors, 2011: Observations of the 10 May 2010 tornado outbreak using OU-PRIME: Potential for new science with high-resolution polarimetric radar. Bull. Amer. Meteor. Soc., 92, 871–891.
Pazmany, A. L., Mead J. B. , Bluestein H. B. , Snyder J. C. , and Houser J. B. , 2013: A mobile rapid-scanning X-band polarimetric (RaXPol) Doppler radar system. J. Atmos. Oceanic Technol.,30, 1398–1413.
Philips, B., Chandrasekar V. , Brotzge J. , Zink M. , Rodriguez H. , League C. , and Diaz W. , 2010: Performance of the CASA radar network during the May 13, 2009 Anadarko tornado. Preprints, 15th Symp. on Meteorological Observation Instrumentation, Atlanta, GA, Amer. Meteor. Soc., 9.3. [Available online at https://ams.confex.com/ams/pdfpapers/165864.pdf.]
Philips, B., Chandrasekar V. , Carr F. , Brotzge J. , Rees P. , League C. , and Bajaj A. , 2012: Dallas Fort Worth Urban Demonstration Network. Preprints, Second Conf. on Transition of Research to Operations: Successes, Plans, and Challenges, New Orleans, LA, Amer. Meteor. Soc., TJ25.5. [Available online at https://ams.confex.com/ams/92Annual/webprogram/Paper203251.html.]
ROC, cited 2013: NEXRAD radar locations. NOAA/Radar Operations Center. [Available online at http://www.roc.noaa.gov/wsr88d/maps.aspx.]
Ryzhkov, A. V., Schuur T. J. , Burgess D. W. , and Zrnić D. S. , 2005: Polarimetric tornado detection. J. Appl. Meteor., 44, 557–570.
Shabbott, C. J., and Markowski P. M. , 2006: Surface in situ observations within the outflow of forward-flank downdrafts of supercell thunderstorms. Mon. Wea. Rev., 134, 1422–1441.
Skinner, P. S., Weiss C. C. , Schroeder J. L. , Wicker L. J. , and Biggerstaff M. I. , 2011: Observations of the surface boundary structure within the 23 May 2007 Perryton, Texas, supercell. Mon. Wea. Rev., 139, 3730–3749.
Snyder, J. C., Bluestein H. B. , Zhang G. , and Frasier S. J. , 2010: Attenuation correction and hydrometeor classification of high-resolution, X-band, dual-polarized mobile radar measurements in severe convective storms. J. Atmos. Oceanic Technol., 27, 1979–2001.
Snyder, J. C., Bluestein H. B. , Venkatesh V. , and Frasier S. J. , 2013: Observations of polarimetric signatures in supercells by an X-band mobile Doppler radar. Mon. Wea. Rev., 141, 3–29.
Tanamachi, R. L., Bluestein H. , Houser J. , Frasier S. , and Hardwick K. , 2012: Mobile, X-band, polarimetric Doppler radar observations of the 4 May 2007 Greensburg, Kansas, tornadic supercell. Mon. Wea. Rev., 140, 2103–2125.
Torres, S. M., and Curtis C. D. , 2007: Initial implementation of super resolution data on the NEXRAD network. Preprints, 23rd Conf. on International Interactive Information and Processing Systems (IIPS) for Meteorology, Oceanography, and Hydrology, San Antonio, TX, Amer. Meteor. Soc., 5B.10. [Available online at https://ams.confex.com/ams/pdfpapers/116240.pdf.]
Wakimoto, R. M., and Bringi V. N. , 1988: Dual-polarization observations of microbursts associated with intense convection: The 20 July storm during the MIST project. Mon. Wea. Rev., 116, 1521–1539.
Wakimoto, R. M., Atkins N. T. , and Wurman J. , 2011: The LaGrange tornado during VORTEX2. Part I: Photogrammetry analysis of the tornado combined with single-Doppler radar data. Mon. Wea. Rev., 139, 2233–2258.
Westrick, K. J., Mass C. F. , and Colle B. A. , 1999: The limitations of the WSR-88D radar network for quantitative precipitation measurement over the coastal western United States. Bull. Amer. Meteor. Soc., 80, 2289–2298.
Whiton, R. C., Smith P. L. , Bigler S. G. , Wilk K. E. , and Harbuck A. C. , 1998a: History of operational use of weather radar by U.S. weather services. Part I: The pre-NEXRAD era. Wea. Forecasting, 13, 219–243.
Whiton, R. C., Smith P. L. , Bigler S. G. , Wilk K. E. , and Harbuck A. C. , 1998b: History of operational use of weather radar by U.S. weather services. Part II: Development of operational Doppler weather radars. Wea. Forecasting, 13, 244–252.
Wurman, J., Kosiba K. A. , Markowski P. , Richardson Y. , Dowell D. , and Robinson P. , 2010: Finescale single- and dual-Doppler analysis of tornado intensification, maintenance, and dissipation in the Orleans, Nebraska, supercell. Mon. Wea. Rev., 138, 4439–4455.
Wurman, J., Dowell D. , Richardson Y. , Markowski P. , Rasmussen E. , Burgess D. , Wicker L. , and Bluestein H. B. , 2012: The Second Verification of the Origins of Rotation in Tornadoes Experiment: VORTEX2. Bull. Amer. Meteor. Soc., 93, 1147–1170.
Zittel, W. D., and Jing Z. , 2012: Comparison of a 2-D velocity dealiasing algorithm to the legacy WSR-88D velocity dealiasing algorithm during Hurricane Irene. Preprints, 30th Conf. on Hurricanes, Ponte Vedra Beach, FL, Amer. Meteor. Soc., 7C.7. [Available online at https://ams.confex.com/ams/30Hurricane/webprogram/Paper205076.html.]
The two letters before the slash indicate U.S. Army, Air Force, or Navy. The first letter after the slash indicates the class of installation or platform (A, airborne; C, air transportable; F, ground fixed; T, ground transportable). The second letter indicates the type of equipment (M, meteorological, nonradar; P, radar). The third letter indicates the purpose of the system (N, navigational aids; Q, special or combination of several purposes; R, receiving such as passive detection systems; S, detecting or establishing azimuth and range; T, transmitting) (Whiton et al. 1998a).
