1. Introduction
Downbursts are strong downdrafts formed in a thunderstorm, causing low-level wind shear (LLWS), which is a significant change in wind speed and/or direction that often is hazardous to aircraft safety (Fujita 1985; Wakimoto 1985). Since the late 1940s, it has been known that strong low-level outflow could occur in association with an intense downdraft in a thunderstorm (e.g., Suckstorff 1938; Müldner 1950). Extensive studies of downbursts began in the late 1970s when increasing aircraft accidents drew the attention of meteorologists (Fujita and Caracena 1977). Theoretical and observational studies have clarified the properties of downbursts and suggested the governing physical mechanisms (e.g., Fujita and Wakimoto 1981).
Downbursts are technically classified into two categories based on the horizontal dimension of outgoing wind at low altitudes: microbursts, which are smaller than 4 km; and macrobursts, which are larger than 4 km (Fujita 1981). The classification is simply made on the basis of spatial size and not by the governing physics. Downbursts are also classified by the amount of precipitation observed at the surface: dry downbursts with little or no precipitation and wet downbursts with heavy precipitation (e.g., Fujita 1985; Wakimoto 2001). Dry downbursts are common in the U.S. high plains where the boundary layer is deep and the cloud base is high (Brown et al. 1982; Wakimoto 1985), whereas wet downbursts are often seen in the U.S. Southeast and occur in a moist environment where the subcloud layer is shallow and the cloud base is warm (Atkins and Wakimoto 1991). The majority of downbursts occurring in Japan are reported as wet type (e.g., Ohno et al. 1994), and the event analyzed here exhibited properties that are typically found in wet downbursts.
Since aircraft operation is maintained by various aerodynamic balances, it is natural that a wind shear, which is a significant spatial change of wind speed/direction, threatens its safety. Low-altitude wind shear, or LLWS, is especially concerning for air traffic during landing and takeoff, when most accidents occur. To better understand the phenomena causing LLWS, the Joint Airport Weather Studies (JAWS) project was carried out in collaboration with the National Center for Atmospheric Research (NCAR) and the University of Chicago (McCarthy et al. 1982). In the JAWS project, researchers operated field measurements using multiple instruments (e.g., Rothermel et al. 1985; Hjelmfelt 1987). The project was successful in detecting and characterizing LLWS associated with downbursts and motivated researchers to design the Terminal Doppler Weather Radar (TDWR) network system (e.g., Cho and Chornoboy 2005). Compared with Weather Surveillance Radar-1988 Doppler (WSR-88D), TDWR has higher range and azimuth resolutions to observe finescale precipitation and wind field structures associated with LLWS (Vasiloff 2001; Heinselman et al. 2012b).
The Japan Meteorological Agency (JMA) also operates an airport surveillance radar system, which is referred to as Doppler Radar for Airport Weather (DRAW). DRAW is deployed at nine major airports in Japan to monitor the occurrence of LLWS and report the location of dangerous weather conditions to air traffic controllers and airlines (Ishihara and Hata 1995). The DRAW system automatically detects intense wind shear and categorizes them into shear lines (SLs) and microbursts (MBs). This information is then utilized to issue a wind shear alert (WSA) and microburst alert (MBA), which provide the time and location of the occurrence around the targeted airports. This alert system works well to help air traffic controllers and airline pilots make the correct decisions under rapidly changing meteorological conditions.
While the present DRAW system is capable of real-time monitoring of LLWS, it does not forecast their occurrence. To ensure future safety, it is essential to develop short-term forecasts of LLWS by detecting their precursor signatures in the parent storm. Roberts and Wilson (1989) analyzed 31 storms that produced microbursts and concluded that the descending precipitation core, increasing midlevel radial convergence, and reflectivity notches could be used as downburst precursors. As we will suggest, combinations of these signatures are indicative of downdraft initiation and intensification. To effectively detect the precursor signatures, which typically appear 2–6 min prior to the initiation of surface outflow (Roberts and Wilson 1989), a weather radar must observe the parent storm from low to high altitudes with a sufficiently high volume scan update rate (no more than 1 min). Since conventional radars with parabolic antennas usually take 5–10 min for a full volume scan, state-of-the-art technologies with much higher temporal resolution are required.
Heinselman et al. (2008) reported severe weather phenomena including microbursts observed by the National Weather Radar Testbed (NWRT) phased array radar (PAR). The PAR is an S-band Doppler radar with a stationary flat panel antenna comprising 4352 antenna elements (Zrnić et al. 2007). By shifting the radio signal emitted from each element, the PAR forms an agile beam to scan rapidly evolving storms without rotating the antenna. By analyzing the PAR data, Heinselman et al. (2008) showed clear precursor signatures including a rapidly descending high-reflectivity core, and midlevel convergence, which emerged ahead of strong surface outflow. Newman and Heinselman (2012) analyzed the evolution of a storm that produced microbursts and discussed the relationship between quasi-linear convective systems and microbursts in great detail. Heinselman et al. (2012a) and Cho et al. (2013) studied microburst events observed by PAR and suggested the potential use of the radar to improve the prediction capability. Bowden et al. (2015) assessed the effect of high-speed PAR data on forecaster performance for severe hail and wind cases and advocated the use of high temporal resolution radar data for more precise warnings and improved lead times. Kuster et al. (2016) analyzed PAR data and examined downburst precursor signatures, including the descending precipitation core, the magnitude and size of midlevel convergence, and descending troughs of differential reflectivity. Through the obtained results, they suggested the importance of rapid-update radar data to effectively resolve the rapidly evolving precursors.
In contrast to agile beam NWRT-PAR, the Atmospheric Imaging Radar (AIR) has employed imaging techniques to steer the beam electronically through digital beamforming (DBF) technology (Isom et al. 2013). The AIR transmits a wide beam of 20° (vertical) by 1° (horizontal) and resolves the target image by forming simultaneous multiple narrow receive beams (1° width) to realize even higher full volume update rates (e.g., 18 s) than agile beam PAR. Since AIR uses X-band beam at 9.55 GHz, the physical size of the antenna is small, which reduces fabrication, installation, and operation costs. The utilization of X-band radars were studied during the Collaborative Adaptive Sensing of the Atmosphere (CASA) project. Mclaughlin et al. (2009) suggested that a dense network system of X-band weather radars are advantageous since it provides more low-altitude data that cannot be obtained by a relatively sparse network of S- or C-band radars because of Earth’s curvature and terrain blockage.
While several types of PARs have been designed and/or developed (Heinselman et al. 2008; Bluestein et al. 2010; Zhang et al. 2011), another type of PAR was developed in Japan based on the concept of an X-band imaging PAR. The system design of the phased array weather radar (PAWR) is unique, especially in terms of the hybrid system combining the one-dimensional digital beam forming technology for an elevational scan and the mechanical antenna rotation scheme for an azimuthal scan (Yoshikawa et al. 2013). While several studies have demonstrated a relationship between lightning activity and storm structure as observed by PAWR (e.g., Wu et al. 2013), no studies have used PAWR to observe damaging wind events such as tornadoes and downbursts. The purpose of the present paper is to report the first results of a downburst observed by PAWR, examine the governing physical processes, and discuss the capability of PAWR for monitoring and short-term forecasting of LLWS associated with a downburst.
2. Instrumentation and observation data
The PAWR was developed in collaboration between Osaka University, the National Institute of Information and Communication Technology (NICT), and Toshiba Corporation. As of February 2016, four PAWRs are under operation in Japan: Osaka (34.823°N, 135.523°E) since 2012, Okinawa (26.498°N, 127.843°E) and Kobe (34.710°N, 134.954°E) since 2014, and Tsukuba (36.053°N, 140.125°E) since 2015. These four radars have essentially the same specifications as shown in Table 1 (Yoshikawa et al. 2013; Wu et al. 2013). In contrast to conventional radars that use a parabolic antenna, PAWR has a 2 m × 2 m flat antenna in which 128 slot antenna elements are aligned in the vertical. By adjusting the phase and amplitude of electromagnetic waves radiated by each element, PAWR creates a fan-shaped transmission beam with a vertical width of 5°–10° and a horizontal width of 1.2°. When receiving the target’s scattered return signals, PAWR forms multiple narrow beams using a DBF technology. Although many adaptive DBF methods have been proposed for a sidelobe elimination (Yoshikawa et al. 2013, and references therein), the present PAWR system uses a nonadaptive DBF method. Thus, one of the future crucial improvements of PAWR is perhaps to apply an adaptive DBF method such as the minimum mean-square error (MMSE) framework (Yoshikawa et al. 2013). In the present study, we reduced sidelobe contaminations from ground clutter using conventional moving target indicator (MTI) filtering method. PAWR transmits 13 fan beams using a dual pulse repetition frequency (PRF) scheme, by steering the beam elevation from 0° to 90° in a time period of 100 ms. In the higher PRF mode, the pulse repetition time (PRT) is ~584 μs with 19 pulses per dwell for low elevation (<4.5°) observations, while PRT is ~211 μs with 19 pulses for midelevation (~45°) observations.
Specifications of phased array weather radar (PAWR) at Osaka University.
Because rapid electronic beam steering is performed for vertical scans, all that PAWR needs for a full volume scan is to mechanically rotate the antenna just once in the azimuthal direction (2–6 revolutions per minute). The hybrid scheme of the vertical electronic scan and horizontal mechanical scan provides both high performance (quick full volume scan of all azimuthal directions) and cost reduction (requires only a single flat antenna). PAWR has various modes of operation and the data presented here were obtained with a mode that covers up to 60 km in range and updates every 30 s. The range and azimuthal resolutions are constant at 100 m and 1.2°, respectively. PAWR observes three-dimensional dynamics of rapidly changing severe weather phenomena including localized heavy rain, tornadoes, and downbursts with sufficiently high spatiotemporal resolutions.
Another instrument we use in the present study is the DRAW system at Osaka airport (see Table 2). DRAW is a C-band operational radar with a 7-m parabolic antenna. In normal surveillance mode, it carries out a plan position indicator (PPI) scan for an elevation range between 0.7° and 28.5°. However, when severe storms approach the airport, DRAW automatically changes the operation mode to enhance the observation frequency at low elevations. In this mode, the update rate of the PPI scan at the lowest elevation (0.7°) is as short as 1 min, which is much faster than the normal mode operation (6 min). We also note that the DRAW observation at higher elevations becomes less frequent in exchange for more frequent observations at the lower elevations (see Table 2; “scan elevation”). The DRAW data presented here were obtained applying this special operation mode.
Specifications of Doppler Radar for Airport Weather (DRAW) at Osaka airport.
In the present study, we report the radar signatures of a downburst event observed by the PAWR deployed at Osaka University. The event we selected here is a downburst event, which was automatically detected by DRAW at Osaka airport on the night of 10 September 2014. As described in the previous section, the DRAW system is operated by JMA to instantaneously detect the occurrence of LLWS such as SLs and MBs. An MB event is detected when the divergence pattern, which has a maximum differential wind of >8 m s−1 over an area of >3 km2 is observed within 20 km from a radar site. Between 18 March 2013 and 31 December 2014, most of the MB events observed by DRAW occurred between June and October, which is consistent with previous studies of downbursts in Japan (e.g., Ohno et al. 1996; Muramatsu and Kawamura 2012). Among all of the DRAW-detected MB events, the event observed on the night of 10 September 2014 was the most suitable for analysis because the event occurred at a close distance of <10 km from PAWR and had clear divergence patterns at low altitudes. These conditions are preferable for us to analyze finescale physical processes causal to the occurrence of low-level divergence wind.
Figure 1 is a map showing the locations of PAWR at Osaka University, DRAW at Osaka airport, and the MB-type LLWS detected by DRAW. It is obvious that the LLWS was observed at very close distances from both radar sites (12–18 km from DRAW, 6–10 km from PAWR), suggesting that the radar signatures leading to the occurrence of MBs would be finely resolved. The LLWS detection by DRAW started at 2321 Japan standard time (JST) and ended at 2333 JST moving in the east-northeast direction. The DRAW-detected MBs lasted for about 12 min, which is consistent with the typical duration of a downburst (<10 min for microburst and 10–30 min for macroburst) documented in previous studies (e.g., Fujita 1985; Wolfson 1990).
Locations of X-band phased array weather radar (PAWR) at Osaka University (filled circle), C-band operational radar [Doppler Radar for Airport Weather (DRAW)] at Osaka airport (open circle), and the DRAW-detected microburst-type low-level wind shear (LLWS; crisscrosses). Large solid and dashed concentric circles indicate 10- and 20-km distances from PAWR. The LLWSs were automatically detected by DRAW from 2321 to 2333 JST with their locations moving toward the east-northeast.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Figure 2 shows PPI images obtained by DRAW. Figures 2a and 2b are the reflectivity and Doppler velocity data obtained at an elevation of 0.7°, which is equivalent to an altitude of ~147 m above ground level (AGL) at a distance of 12 km from the radar. In Figs. 2a and 2b, the red dotted ovals indicate the targeted area where a divergence couplet of downburst outflow was discernible as shown by the two black arrows. Figure 2a clearly indicates that the downburst-producing storm had a horizontal dimension of 3–8 km and was separated from other storms (located at 34.77°–34.81°N, <135.50°E) by about 5–8 km. The storm cell in question traveled toward the east-northeast at a speed of ~4.8 m s−1. The divergence couplet had the maximum radial differential wind of ~15.8 m s−1 at a horizontal distance of ~1.7 km. Judging from the spatial scale of the couplet, the event can be classified into a microburst, which is identified by an outflow pattern with a radial differential wind of 10 m s−1 or more and a distance of divergence couplet of 4 km or less (Wilson et al. 1984). We also note here that the maximum radial differential wind of 15.8 m s−1 is not in the strongest class of all the MB-type LLWS events detected by DRAW at the Osaka airport. The maximum radial differential wind of the LLWS was highly variable between events, starting from the detection criteria of ~8 up to ~41 m s−1. The arithmetic mean and median were 12.8 and 11.5 m s−1, respectively. Thus, the magnitude of this event was in a moderate class among all the DRAW-detected events. Compared to microburst events previously studied, the maximum differential wind of 15.8 m s−1 is somewhat weaker than the average. Willingham et al. (2011), for example, analyzed differential velocity of 140 microburst events and found the distribution peak at 20–25 m s−1 with a median of 24 m s−1. Therefore, we conclude that the event analyzed here is a “moderate” to “relatively weak” microburst event. This is consistent with the fact that no damage was reported with the event.
Maps of (a) reflectivity and (b) Doppler velocity obtained by DRAW at Osaka airport at 2321:08 JST 10 Sep 2014. Red dashed ovals indicate the region of interest. In the Doppler velocity data, the two black arrows heading in opposite directions represent a divergence pattern. (c)–(e) Doppler velocity data obtained at elevations of (c) 0.7°, (d) 2.1°, and (d) 3.8° during a period from 2328:12 to 2328:46 JST. Red ovals again represent the regions of interest. The approximate altitudes of the region are 192, 582, and 1042 m. (f) A schematic of an outflow structure reconstructed from PPI images at multiple elevation angles. The figure shows a vertical cross section along the axis of storm motion (the storm moves to the right in this figure).
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Figures 2c–e are PPI images of Doppler velocity data at several elevations obtained during a period of 34 s. It is clear from Figs. 2c and 2d that the divergence couplet, indicated by the red solid ovals, was larger in size at a lower elevation than at a higher elevation. The horizontal dimensions were ~5 km × ~9 km at 0.7° (equivalent to ~192 m AGL) and ~4 km × 7 km at 2.1° (~582 m AGL). At an elevation of 3.8° (Fig. 2e; ~1042 m AGL), it was significantly more difficult to identify a pattern of divergence. Figure 2f is a schematic of the low-level outflow vertical structure reconstructed from PPI images at different elevations. Strong outflow occurred below 400–500 m AGL, which is consistent with the typical morphology of the outflow associated with a downburst as calculated by numerical simulations (e.g., Proctor 1989).
3. Results obtained by PAWR
Between 2310 and 2340 JST, PAWR observed the parent storm cell that produced the LLWS detected by DRAW. Figure 3 shows successive images of the three-dimensional structure of the storm cell observed by PAWR, which was visualized using Unidata’s Integrated Data Viewer software. Here, the three-dimensional structure was produced by converting the radar data in the original polar coordinate to the Cartesian coordinate using the Cressman interpolation method with a radius of influence of 0.3 km and a gridding size of 0.1 km along both horizontal and vertical axes. These interpolation parameters were optimized in order to resolve the structures with fine detail without violating the radar spatial resolution as large as 1.2° (azimuthal), which corresponds to 0.1–0.2 km at a distance of 6–10 km where the storm cell was observed.
High-speed volumetric images of a storm cell producing low-level outflow observed at 30-s intervals by PAWR on 10 Sep 2014. Each single image shows the three-dimensional structure viewed from the south. The storm moved toward the east-northeast corresponding to the right in each panel. The precipitation cores of the cell are represented by white (40 dBZ) and gray (35 dBZ) isosurfaces and the low-level outflow with a radial divergence of 4 × 10−3 s−1 is represented by red isosurfaces.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
It is clear in Fig. 3 that the PAWR-observed storm cell had a vertical extent of ~5 km and a horizontal dimension of 5–10 km, which is consistent with the spatial scale observed by DRAW as shown in Fig. 2a. At around 2310 JST, a precipitation core with a radar reflectivity >40 dBZ appeared at 3–5 km AGL (white volumetric region in Fig. 3). The core then rapidly grew in both vertical extent and reflectivity intensity, and then rapidly fell down toward the ground. The first core (denoted as “core 1”) arrived at the near-surface region, which we define here as the altitude range below 1 km AGL, at ~2316 JST. The core was associated with a weak low-level outflow with a maximum wind speed difference of ~7.8 m s−1 (not shown here), or below the detection criterion of LLWS by DRAW. Approximately 3 min later, a large volume of precipitation core (denoted as “core 2”) descended to the near-surface region and a strong low-level outflow appeared, which was detected by DRAW at 2321 JST. PAWR observed an increase in the magnitude of the low-level outflow, which eventually reached the maximum radial differential wind of ~13 m s−1 at a distance of 1.5–2.0 km at 2328 JST. Since the angle between the radar beam and the storm motion was approximately 30°–45° and the outflow pattern is most evident along the axis of storm motion (west-southwest to east-northeast; see Fig. 9), the actual maximum differential wind projected onto the axis would be in the range of 15–18 m s−1, showing remarkable consistency with the DRAW-observed differential wind. We also calculated the radial divergence based on the method described in Roberts and Wilson (1989). The red isosurfaces in Fig. 3 show the region with a radial divergence of 4 × 10−3 s−1. The significant divergence of low-level outflow lasted for more than 10 min between 2320 and 2330 JST while precipitation cores descended toward the ground continuously. After 2333 JST, the divergence disappeared with the dissipation of the parent storm cell. Similarly, the DRAW system automatically detected MB-type LLWS between 2321 and 2333 JST, which agrees well with the time period that PAWR observed the strong radial divergence shown by the red isosurfaces in Fig. 3. Given the discrepancies in the source–observer geometry and instrumental specifications, this consistency is a remarkable finding, suggesting that DRAW is a highly efficient surveillance system for LLWS. However, a more important result in this study is that PAWR observed precursor signatures such as descending precipitation cores, in addition to LLWS. As proposed by Roberts and Wilson (1989), such signatures could be utilized for short-term LLWS forecasting technology.
To better understand the physical processes causing the formation of strong downdrafts and the resulting low-level outflow, we further investigate the structure of the precipitation core and the associated wind field in more detail. Figure 4 is an enlarged snapshot of the precipitation core and the low-level wind divergence observed at 2320:32 JST. This snapshot was visualized in a geometry looking at the north-northwest from the south-southeast, and the storm moved toward the east-northeast corresponding to the right. The white and gray isosurfaces represent echo intensities of 40 and 35 dBZ, respectively, and the red isosurface represents LLWS with a radial divergence of 4 × 10−3 s−1. Also shown in this figure by the black lines are the 40-dBZ contours of the precipitation core at several altitudes. The black arrows indicate the regions of interest where the outlines are curved inward significantly, referred to here as a “notch” structure. As will be discussed in the next section, the notch is a real structure since C-band DRAW also observed a similar pattern (see Fig. 9). In Fig. 4, it is clear that the reflectivity notch consistently existed from high to low altitudes and was located just above the divergence of LLWS given by the red isosurface.
A snapshot of the precipitation core and the divergence associated with low-level outflow observed by PAWR at 2320:32 JST. The figure format is as in Fig. 3, but that the viewing angle is from the south-southeast to the north-northwest. Also shown in this figure are black lines that represent contours of the 40-dBZ core at several altitudes. As shown by black arrows, the 40-dBZ lines are significantly curved inside toward the center of the core, which is referred to as notch.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Figure 5 shows the evolution of reflectivity notch depicted by 40 dBZ (light gray) and 47 dBZ (dark gray) contours at altitudes of 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 km AGL. At 2316:02 JST, a reflectivity notch, which we define here as a concave shape of a 40-dBZ outline with an angle of approximately less than 100°, was found at 4.0 km AGL as indicated by the red arrow. The notch also appeared at lower altitudes; down to 2.5 km AGL at 2317:02 JST and to 1.0 km AGL at 2319:02 JST. This temporal development suggests a downward phase propagation of notch formation from the top of the precipitation core to the bottom. Although not identified as a notch by the present definition, one might recognize weak inward curvatures of the 40-dBZ contours even at lower altitudes; down to 3.0 km AGL at 2316:02 JST, 2.0 km at 2317:32 JST, and 1.0 km AGL at 2318:32 JST. The lowest altitude of notch formation, hence, depends on the definition. However, it is clear that the notch pattern descended from 4.0 to 1.0 km AGL, and the low-level outflow with a radial divergence of 4 × 10−3 s−1 (orange) appeared around the same time that the notch formed at the near-surface region. This finding suggests that the descending reflectivity notch was relevant to the production of LLWS and might be a good precursor signature for short-term LLWS forecasting.
Temporal development of notch structure and low-level divergence observed between 2316:02 and 2320:32 JST. Gray hatched contours indicate the morphology of radar reflectivity areas at 40 dBZ (light gray) and 47 dBZ (dark gray) depicted at altitudes of 1.0–4.0 km AGL. Red arrows indicate the location of the reflectivity notch. Low-level divergence at 0.3 km AGL with values of 4 × 10−3 s−1 (orange) and 5 × 10−3 s−1 (red) are also shown.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Figure 6 shows the midlevel convergence associated with the notch structure. The figure is basically identical to Fig. 5 except that the midlevel convergence with a value of −2 × 10−3 s−1 (light blue) is additionally drawn. Blue arrows indicate the regions of interest where the midlevel convergence appeared around the notch structure. The convergences associated with the notch at 4.0 km AGL were observed between 2316:32 and 2320:32 JST, whereas those at 3.5 km AGL were found between 2317:32 and 2318:32 JST. The horizontal dimensions of the convergence at 4.0 and 3.5 km AGL maximized at 2317:32 and 2318:32 JST, respectively. Judging from the vertical extent and horizontal dimension, it is possible that the midlevel convergence associated with the reflectivity notch reached a maximum at around 2317–2318 JST, which is 3–4 min earlier than the local peak time of the low-level divergence at 2320:32 JST.
Midlevel convergence associated with the reflectivity notch. The data are as in Fig. 5, but that the midlevel convergence with a value of −2 × 10−3 s−1 (light blue) is also drawn. Arrows indicate convergences located around the notch.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Figure 7 shows the horizontal structure of the reflectivity notch, midlevel convergence, and low-level divergence sliced at different altitudes observed at 2318:32 JST. The lowest right panel is a summary plot of crucial structures. It is clear that the deepest point of the notch structure, shown by black crosses, is located to the east at higher altitudes. Also, the notch-associated midlevel convergences (blue squares) is located at 2–2.5 km east of a low-level divergence (red square). These findings suggest that the reflectivity notch, midlevel convergence, and low-level divergence occurred consistently with an east–west inclination of 25°–30° from the vertical axis (top east and bottom west).
Horizontal cross section sliced at each altitude observed at 2318:32 JST. Color contours are the same as those in Figs. 5 and 6. Black crosses indicate the location of the deepest point of notch. Blue squares represent the approximate center of the midlevel convergence associated with notch, while a red square represents the approximate center of the low-level divergence. Horizontal relationships of these crucial signatures are summarized in the bottom-right panel where numbers show altitudes in kilometers.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Figure 8 shows the relationship between the reflectivity notch and the surrounding wind field. The vertical plane is the east–west cross section of the Doppler velocity data, which passes through the deepest point of the notch, while the horizontal plane shows the cross section at 0.5 km AGL. The storm cell in question is shown by a semitransparent white isosurface (35 dBZ) and, for easy recognition, the location of the reflectivity notch is also represented by a slant column of black dotted lines. Since PAWR is located to the north-northwest as shown by a light blue arrow, the cold color region, which represents hydrometeors moving toward the radar, corresponds to an easterly/southerly wind. Conversely, a warm color region indicates a westerly/northerly wind. We note that the vertical plane does not represent a radial section of a range–height indicator (RHI) scan and thereby the Doppler velocity data are significantly affected by environmental winds. The JMA objective analysis data showed south-southwest to west winds of 5–20 m s−1 flowing at altitudes below 6 km AGL. These environmental winds are visible on the vertical plain of this figure as a background color gradation: colder color at the west side (left) than at the east side (right). To the east of the reflectivity notch, one can recognize the region of a localized easterly/southerly wind (cold color), which indicates the existence of an inflow impinging on the precipitation core. As shown by several white arrows, the notch-associated midlevel inflow was located just above the divergence couplet of low-level outflow. These results suggest that the midlevel inflow was involved in the formation of the reflectivity notch, which was then later associated with low-level outflow.
A snapshot showing the relationship between the notch structure and radial wind field observed by PAWR at 2320:32 JST. The precipitation core is represented by a semitransparent white region and the location of the notch structure is shown by a black dotted slant column. The vertical cross section is the east–west plane, which passes through the deepest point of the reflectivity notch and the horizontal cross section is at 0.5 km AGL. In both vertical and horizontal cross sections, color shades represent the Doppler velocity where a positive value (warm color) indicates receding wind and a negative value (cold color) indicates approaching wind. PAWR is located to the west-northwest of this figure as shown by a light blue arrow.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
4. Discussion
Downburst is a strong downdraft occurring in a thunderstorm that results in the formation of horizontal divergent wind at low altitudes. Therefore, one of the most important issues for downburst investigation is to elucidate the physical processes that play crucial roles in producing strong downdrafts. Past studies have suggested several important physical processes: dynamic effects such as nonhydrostatic vertical pressure gradients and thermodynamic effects such as precipitation loading and latent cooling (Markowski and Richardson 2010 and references therein). Although the effect of a vertical pressure gradient can be identified by Doppler radar measurements as a signature of increasing vorticity with decreasing height (Roberts and Wilson 1989), we do not find such a clear signature in the present analysis. We, therefore, focus on the thermodynamic effects.
Clark and List (1971) performed two-dimensional numerical calculations and suggested that descending hydrometeors can create a strong downdraft by dragging the surrounding air downward. Roberts and Wilson (1989) suggested that descending precipitation cores associated with midlevel convergence are a useful signature for radar measurements to assess the role of hydrometeor loading. If large mass hydrometeors such as hailstones exist, the loading effect works efficiently. Proctor (1988) demonstrated in a two-dimensional axisymmetric simulation that the mass loading acts to initiate the development of a downdraft. Atkins and Wakimoto (1991) reported that the main precipitation core of a microburst-producing storm had a radar reflectivity of >55 dBZ, which was mainly composed of frozen particles. Heinselman et al. (2008) showed the phased array radar data of a microburst event that had a maximum reflectivity of parent storm exceeding 60 dBZ. Since a radar reflectivity associated with hailstones is typically more than 55 dBZ (e.g., Kunz and Kugel 2015), these results suggest a role of large-mass hydrometeors in driving strong downdraft. Markowski and Richardson (2010) pointed out that hydrometeor loading plays a crucial role especially in initiating downdraft, but other mechanisms are also important afterward.
Another important process in producing strong downdraft is the phase change of frozen particles and water droplets (e.g., Srivastava 1987; Hjelmfelt et al. 1989; Straka and Anderson 1993; Atlas et al. 2004). Theoretical and observational studies have clarified that rain droplet evaporation, ice crystal sublimation, and melting of small hailstones effectively absorb latent heat and cool the surrounding air to create an intensified downdraft. Evaporation of water droplets primarily works at altitudes lower than the 0°C isotherm (the freezing level), whereas the sublimation of ice particles works at higher altitudes. In particular, such phase changes are much more active when drier air entrains into the precipitation region. Emanuel (1981) carried out a theoretical investigation of downdraft formation and suggested that the penetrative downdraft entrainment mechanism is of great importance for a downburst. When environmental dry air entrains into the precipitation core, the region is recognizable by radar as a reflectivity notch (Roberts and Wilson 1989). The melting of frozen droplets also occurs below the freezing level and is thought to play another significant role in cooling the air and producing strong downdraft (e.g., Rasmussen et al. 1984). Wakimoto and Bringi (1988) found from the MIST Project that melting hailstones most likely existed in a microburst-producing downdraft. Strong radial convergence just below the 0°C isotherm is a radar signature, which is suggestive of a downdraft that might be intensifying because of melting (Roberts and Wilson 1989).
In the present analysis, a strong low-level outflow appeared after the main body of the precipitation core (core 2 in Fig. 3) reached the near-surface region. Therefore, it is most likely that the hydrometeor loading played a major role in initiating downdraft as suggested by Roberts and Wilson (1989) and Markowski and Richardson (2010). The PAWR-observed peak reflectivity of the precipitation core was 54 dBZ, which is almost comparable, but slightly lower than the typical echo maximum of >55 dBZ for microburst-producing storms (e.g., Isaminger 1988; Atkins and Wakimoto 1991; Heinselman et al. 2008). Since PAWR transmits X-band radio waves, we must consider the effect of rain attenuation, which leads to an underestimate in the measured echo intensity. The degree of rain attenuation is highly variable depending on the structure of the target rainfall and the observation geometry (Delrieu et al. 2000). Snyder et al. (2010) have reported that the rain attenuation at the X band is at least an order of magnitude larger than that at the S band and is often several times larger than that at the C band. To correct rain attenuation, dual-polarization radar data are used (e.g., Bringi et al. 1990; Testud et al. 2000; Gorgucci and Chandrasekar 2005; Snyder et al. 2010). There are several techniques to compensate for the rain attenuation effect by using, for example, the differential phase parameterization method (Bringi et al. 1990), the ZPHI method (Testud et al. 2000), the self-consistent with constraints method (Bringi et al. 2001; Park et al. 2005), and the pseudo-dual-frequency method (Zhang et al. 2004). All of these studies require dual-polarization data, and therefore, single-polarization data of PAWR limits the compensation capability of rain attenuation.
Since we do not have any dual-polarization radar data in this case study, the concurrent C-band DRAW data are the best information to assess the effect of rain attenuation in the X-band PAWR measurements. Figure 9 shows a comparison of PPI images obtained by DRAW and PAWR at close times (2317:40 and 2317:32 JST at ~1.5 km AGL; 2320:14 and 2320:02 JST at 0.2–0.3 km AGL). In Figs. 9a and 9d, the maximum reflectivity of the DRAW data is ~56 dBZ, whereas that of the PAWR data is ~49 dBZ. Also, the reflectivity of the echo region located to the southeast of the arrow with the striped pattern (Figs. 9b and 9e) is ~43 dBZ in the DRAW data and ~33 dBZ in the PAWR data. These results suggest a significant rain attenuation effect in the X-band PAWR data at a deficit of 5–10 dBZ. Consequently, the PAWR-measured maximum echo intensity of 54 dBZ obviously is an underestimate and we expect the actual intensity of 60 dBZ or more. This implication supports the idea that hydrometeor loading is probably one of the major processes in this case.
Comparison between PPI images observed by (top) DRAW and (bottom) PAWR. (a) DRAW data at 2317:40 JST (elevation = 6.9°, ~1.54 km AGL at the notch) and (d) PAWR data at 2317:32 JST (elevation = 13.1°, ~1.48 km AGL). (b),(c) Reflectivity and Doppler velocity data observed by DRAW at 2320:14 JST (elevation = 1.5°; ~0.33 km AGL). (e) Reflectivity, (f) Doppler velocity, and (g) velocity width data observed by PAWR at 2320:02 JST (elevation = 2.0°, ~0.22 km AGL). In the reflectivity data in (a),(b),(d),(e), large arrows with black and white striped pattern represent the location of reflectivity notch, which is visible as an inward curvature of the echo reflectivity pattern (black dotted lines). In the Doppler velocity and velocity width data, star symbols represent the center of outflow (black thin arrows) and the black dotted lines represent the approximate location of gust front. A pink arrow in the velocity width map shows a local maximum. The width data in (g) are smoothed by a radius of 0.2 km to reduce the background noise level. White color pixels indicate the missing areas, which were removed through quality control process.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
Another important finding in this study is the formation of a reflectivity notch associated with midlevel inflow occurring right above the low-level outflow. As previously discussed, the notch was formed first at the top of the precipitation core with significant midlevel convergence, and later at lower altitudes (see Figs. 5 and 6). Close to the notch, a midlevel inflow impinged on the precipitation core from the southeast side (Fig. 8). Roberts and Wilson (1989) discussed that a notch is a clear signature of the entrainment of dry air, which effectively intensifies the downdraft by evaporative cooling and causes strong low-level outflow. The results obtained in the present study clearly support this idea, suggesting that the reflectivity notch is a useful signature for the prediction of LLWS. Nonetheless, detecting the structure is not straightforward using a conventional radar with a parabolic antenna, since the temporal development of the notch is faster than the radar volume scan speed (Vasiloff and Howard 2009). Indeed, the notch in Fig. 5 initially appeared at 2316:02 JST at 4.0 km AGL, which is only several minutes earlier than the initiation of low-level divergence at 2318:32 JST. To detect a reflectivity notch as a downburst precursor, a high-speed volumetric observation using rapid-scan PAR at a rate of 30 s is ideal.
One might consider that the notch in Fig. 4 could be a pseudopattern as a consequence of attenuation caused by the high reflectivity regions that are located closer to the radar. However, we have found in Fig. 9 that a reflectivity map obtained by the C-band DRAW measurement shows a similar notch pattern both at 2317 JST (~1.5 km AGL) and 2320 JST (0.2–0.3 km AGL). Furthermore, the center of outflow represented by the star symbols in Fig. 9 is located around the deepest point of the notch and is displaced horizontally from the highest reflectivity region. Figure 9g shows that the local maximum of velocity width (pink arrow) is located close to the center of the outflow pattern (star symbol). Since enhancements of Doppler velocity width represent perturbations of the radial velocity field, it is natural that the velocity width maximizes near the center of low-level outflow. Therefore, the divergence point associated with a significant downdraft most likely occurred somewhere around the star symbol and pink arrow, which is close to the location of the notch structure. We note here that a C-band radar also suffers from rain attenuation (e.g., Hildebrand 1978; Snyder et al. 2010) and compensations using dual-polarization data are an important issue. However, given the fact that the reflectivity map observed by two radars at different viewing angles (from east-northeast by DRAW; from northeast by PAWR) consistently exhibits a similar structure, we conclude that the notch observed by PAWR is a real structure and is associated with the formation of intense downdrafts causing LLWS. As discussed above, one must examine the possible effect of rain attenuation in making a pseudonotch pattern. For future operations, polarimetric capability of PAWR would be one of the ideal solutions.
Past studies have demonstrated that downbursts frequently occur in relation to a bow echo, which is a bow-shaped radar reflectivity pattern discovered by Fujita (1978). A rear-inflow jet (RIJ) moves a part of a line echo forward to form a bow shape and, at the ends of the bow, cyclonic and anticyclonic motions drive the entrainment of relatively dry air into the rear side. The systematically driven dry air then produces reflectivity notches and strong downdrafts on the rear side through evaporative cooling (Businger et al. 1998). Bow echoes are formed under certain meteorological conditions such as a unidirectional environmental shear with a magnitude of >20 m s−1 over a 0–2.5-km AGL layer (Weisman and Trapp 2003; Newman and Heinselman 2012). In the present study, we neither observed a bow-shaped reflectivity pattern nor a systematically driven RIJ. Rather than that created by RIJ, the notch shown here is associated with a front-to-side inflow. Moreover, the JMA mesoscale objective analysis data have indicated an environmental wind shear of 5–7 m s−1 over a 0–2.5-km AGL layer, which is significantly lower than that typically found in bow echo events (>20 m s−1). Judging from the reflectivity pattern and the environmental wind shear, the event analyzed here is not associated with the storm dynamics producing a bow echo. Lee et al. (1992) mentioned that, in addition to bow echo events, notches can also be formed by a simple inflow of environmental dry air, as we have found in this study.
With a few exceptions, past studies have reported that most downburst events occur under wet conditions in Japan (e.g., Ohno et al. 1996). Muramatsu and Kawamura (2012) has performed the statistical analysis of a downburst-producing environment and suggested that a relatively high vertical temperature gradient (vertical totals = 25.6° ± 2.6°C) and a relatively low relative humidity (RH) at low altitudes (74.6% ± 14.4% at 925 hPa) are useful indices for the prediction of a downburst in Japan. In the present study, the JMA objective analysis data at the closest geographical grid point (~6 km away from the event) and at the nearest time (2400 JST, or ~30 min after the event) have indicated a typical wet environment, with a most humid layer (RH = 92%–96%) located at 3–4.5 km AGL (700–600 hPa). The humidity slightly decreases at lower altitudes and the estimated RH is 78% at 925 hPa. On the other hand, the vertical temperature gradient is relatively high (vertical totals = 26.7°C). These results are clearly consistent with the findings by Ohno et al. (1996) and Muramatsu and Kawamura (2012), leading to the conclusion that the event has occurred in a typical downburst-producing environment in Japan. The successive volumetric images obtained by PAWR have indicated that the precipitation core initially appeared at 3–5 km AGL, which corresponds to the most humid layer. Since the 0°C isotherm is located at ~3.8 km, it is most likely that hydrometeors within the precipitation core were at least partially frozen. Such frozen particles have perhaps played an additional role in intensifying the downdraft through the melting process below the freezing level. We, however, note here that a more detailed discussion on microphysical processes cannot be made since PAWR does not provide polarimetric data. Past studies have demonstrated the usage of dual-polarization data on hydrometeor classification (e.g., Snyder et al. 2010; Kumjian 2012). Snyder et al. (2010) has classified meteorological echoes into big drops, light to moderate rain, heavy rain, and a rain–hail mixture, using X-band polarimetric radar data. Results from hydrometeor classifications play an essential role in advancing our understanding of the microphysical processes causing a downburst.
In summary, the 10 September 2014 case is a moderate to relatively weak wet microburst event occurring in a typical downburst-producing environment in Japan. The results obtained by the PAWR data analysis have suggested that hydrometeor loading and evaporative cooling played major roles in producing a strong downdraft in the parent storm. Figure 10 is a schematic showing the physical mechanism in this case. Precipitation cores initially appeared in a humid layer at 3–5 km AGL, increased in both the maximum reflectivity and vertical extent, and rapidly descended toward the ground. Descending precipitation cores observed by PAWR suggest that hydrometeor loading probably played a major role in initiating a localized downdraft, which then resulted in the formation of a weak low-level outflow (~2316–2317 JST). During this time, relatively dry environmental air impinged on the upper part of the precipitation core and created a low-reflectivity notch through evaporation and possible sublimation (~2316–2317 JST). This process effectively cooled the air to intensify the magnitude of the downdraft and formed a strong outflow at low levels (~2320–2321 JST), which was eventually detected by the DRAW system at the Osaka airport. While DRAW detected LLWS very accurately (Fig. 3), PAWR additionally observed the precursor signatures in the parent storm. Since precursor signatures such as descending precipitation core, reflectivity notch, and midlevel convergence happen in a significantly short time scale (a few minutes), fast volumetric observations by PAWR with an update rate of 30 s are indeed useful for future shot-term forecasting of LLWS in contrast to normal-speed conventional weather radars that update every 5–10 min.
A schematic showing the generation mechanism of wet microburst analyzed in the present study. Light and dark gray colors represent reflectivity regions of 35 and 40 dBZ, respectively.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-16-0125.1
5. Conclusions
We have analyzed high-speed volumetric data obtained by an X-band phased array weather radar (PAWR) to elucidate the physical mechanism of a wet microburst event occurring at midnight on 10 September 2014 in Japan. Precipitation cores appeared in a humid layer at 3–5 km AGL and rapidly descended to the ground. Simultaneously, a notch structure was observed in association with midlevel inflow and convergence, and a strong low-level outflow was subsequently observed just below them. These results suggest that, in addition to the hydrometeor loading, the evaporative cooling due to the midlevel inflow of relatively dry environmental air played an additional role in intensifying a localized downdraft. Since the preceding signatures such as descending precipitations cores and reflectivity notches occur in a short time scale (2–5 min), it is hard for a conventional weather radar with mechanically rotating parabolic antenna to observe their dynamical features. In contrast, PAWR is capable of detecting such signatures in great detail by a fast full volume scanning scheme at sufficient spatial resolutions. Our findings suggest the efficiency of PAWR for future short-term prediction of microburst-type LLWS by detecting precursor signatures in the parent storm such as the descending precipitation core, reflectivity notch, and midlevel convergence. Although the coverage of X-band PAWR is limited to 60 km in range, which is far shorter than that of C-band operational radars (~400 km), it is highly useful for surveillance of severe weather events occurring around an airport, which is essential for aircraft safety during takeoff and landing.
Acknowledgments
The PAWR data are archived in the National Institute of Information and Communications Technology (NICT) Science Cloud and are available online to the public (http://pawr.nict.go.jp/). The DRAW-detected LLWS data were provided by the Kansai Aviation Weather Service Center and the Osaka Aviation Weather Station under a collaborative research program between the Meteorological Research Institute and Osaka, Kyoto, and Kobe Local Meteorological Offices as well as the Kansai Aviation Weather Service Center. The authors thank the anonymous reviewers for providing helpful comments. This work was supported by JSPS KAKENHI Grant-in-Aid for Scientific Research (B) 15H03728.
REFERENCES
Atkins, N. T., and R. M. Wakimoto, 1991: Wet microburst activity over the southeastern United States: Implications for forecasting. Wea. Forecasting, 6, 470–482, doi:10.1175/1520-0434(1991)006<0470:WMAOTS>2.0.CO;2.
Atlas, D., C. W. Ulbrich, and C. R. Williams, 2004: Physical origin of a wet microburst: Observations and theory. J. Atmos. Sci., 61, 1186–1196, doi:10.1175/1520-0469(2004)061<1186:POOAWM>2.0.CO;2.
Bluestein, H. B., M. M. French, I. PopStefanija, R. T. Bluth, and J. B. Knorr, 2010: A mobile phased-array Doppler radar for the study of severe convective storms: The MWR-05XP. Bull. Amer. Meteor. Soc., 91, 579–600, doi:10.1175/2009BAMS2914.1.
Bowden, K. A., P. L. Heinselman, D. M. Kingfield, and D. P. Thomas, 2015: Impacts of phased-array radar data on forecaster performance during severe hail and wind events. Wea. Forecasting, 30, 389–404, doi:10.1175/WAF-D-14-00101.1.
Bringi, V. N., V. Chandrasekar, N. Balakrishnan, and D. S. Zrnić, 1990: An examination of propagation effects in rainfall on radar measurements at microwave frequencies. J. Atmos. Oceanic Technol., 7, 829–840, doi:10.1175/1520-0426(1990)007<0829:AEOPEI>2.0.CO;2.
Bringi, V. N., T. D. Keenan, and V. Chandrasekar, 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, doi:10.1109/36.951081.
Brown, J., K. R. Knupp, and F. Caracena, 1982: Destructive winds from shallow, high-based cumulonimbi. Preprints, 12th Conf. on Severe Local Storms, San Antonio, TX, Amer. Meteor. Soc., 272–275.
Businger, S., T. Birchard Jr., K. Kodama, P. A. Jendrowski, and J.-J. Wang, 1998: A bow echo and severe weather associated with a kona low in Hawaii. Wea. Forecasting, 13, 576–579, doi:10.1175/1520-0434(1998)013<0576:ABEASW>2.0.CO;2.
Cho, J. Y. N., and E. S. Chornoboy, 2005: Multi-PRI signal processing for the terminal Doppler weather radar. Part I: Clutter filtering. J. Atmos. Oceanic Technol., 22, 575–582, doi:10.1175/JTECH1730.1.
Cho, J. Y. N., R. S. Frankel, M. F. Donovan, M. S. Veillette, and P. L. Heinselman, 2013: Wind-shear detection performance analysis for MPAR risk reduction. 36th Conf. on Radar Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 3B.3. [Available online at https://ams.confex.com/ams/36Radar/webprogram/Paper228623.html.]
Clark, T. L., and R. List, 1971: Dynamics of a falling particle zone. J. Atmos. Sci., 28, 718–727, doi:10.1175/1520-0469(1971)028<0718:DOAFPZ>2.0.CO;2.
Delrieu, G., H. Andrieu, and J. D. Creutin, 2000: Quantification of path-integrated attenuation for X- and C-band weather radar systems operating in Mediterranean heavy rainfall. J. Appl. Meteor., 39, 840–850, doi:10.1175/1520-0450(2000)039<0840:QOPIAF>2.0.CO;2.
Emanuel, K. A., 1981: A similarity theory for unsaturated downdrafts within clouds. J. Atmos. Sci., 38, 1541–1557, doi:10.1175/1520-0469(1981)038<1541:ASTFUD>2.0.CO;2.
Fujita, T., 1978: Manual of downburst identification for Project NIMROD. Satellite and mesometeorology Research Paper 156, Dept. of Geophysical Sciences, University of Chicago, 104 pp.
Fujita, T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38, 1511–1534, doi:10.1175/1520-0469(1981)038<1511:TADITC>2.0.CO;2.
Fujita, T., 1985: The downburst, microburst and macroburst. SMRP Research Paper 210, University of Chicago, 122 pp.
Fujita, T., and F. Caracena, 1977: An analysis of three weather-related aircraft accidents. Bull. Amer. Meteor. Soc., 58, 1164–1181, doi:10.1175/1520-0477(1977)058<1164:AAOTWR>2.0.CO;2.
Fujita, T., and R. Wakimoto, 1981: Five scales of airflow associated with a series of downbursts on 16 July 1980. Mon. Wea. Rev., 109, 1438–1456, doi:10.1175/1520-0493(1981)109<1438:FSOAAW>2.0.CO;2.
Gorgucci, E., and V. Chandrasekar, 2005: Evaluation of attenuation correction methodology for dual-polarization radars: Application to X-band systems. J. Atmos. Oceanic Technol., 22, 1195–1206, doi:10.1175/JTECH1763.1.
Heinselman, P. L., D. L. Priegnitz, K. L. Manross, T. M. Smith, and R. W. Adams, 2008: Rapid sampling of severe storms by the National Weather Radar Testbed Phased Array Radar. Wea. Forecasting, 23, 808–824, doi:10.1175/2008WAF2007071.1.
Heinselman, P. L., D. S. LaDue, and H. Lazrus, 2012a: Exploring impacts of rapid-scan radar data on NWS warning decisions. Wea. Forecasting, 27, 1031–1044, doi:10.1175/WAF-D-11-00145.1.
Heinselman, P. L., J. Cho, B. L. Cheong, M. Weadon, M. Emanuel, T. Crum, and P. Byron, 2012b: Multifunction Phased Array Radar Wind Shear Experiment. 26th Conf. on Severe Local Storms, Nashville, TN, Amer. Meteor. Soc., P155. [Available online at https://ams.confex.com/ams/26SLS/webprogram/Manuscript/Paper212593/Heinselman-etal-MPAR-WSE-ConfPaper.pdf.]
Hildebrand, P. H., 1978: Iterative correction for attenuation of 5 cm radar in rain. J. Appl. Meteor., 17, 508–514, doi:10.1175/1520-0450(1978)017<0508:ICFAOC>2.0.CO;2.
Hjelmfelt, M. R., 1987: The microbursts of 22 June 1982 in JAWS. J. Atmos. Sci., 44, 1646–1665, doi:10.1175/1520-0469(1987)044<1646:TMOJIJ>2.0.CO;2.
Hjelmfelt, M. R., H. D. Orville, R. D. Roberts, J. P. Chen, and F. J. Kopp, 1989: Observational and numerical sturdy of a microburst line-producing storm. J. Atmos. Sci., 46, 2731–2744, doi:10.1175/1520-0469(1989)046<2731:OANSOA>2.0.CO;2.
Isaminger, M. A., 1988: A preliminary study of precursors to Huntsville microbursts. MIT Lincoln Laboratory Project Rep. ATC-153, MIT Lincoln Laboratory, 21 pp.
Ishihara, M., and K. Hata, 1995: Operational Doppler weather radar for airport in Japan. Preprints, 27th Conf. on Radar Meteorology, Vail, CO, Amer. Meteor. Soc., 723–724.
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, doi:10.1175/JTECH-D-12-00063.1.
Kumjian, M. R., 2012: The impact of precipitation physical processes on the polarimetric radar variables. Ph.D. dissertation, University of Oklahoma, 327 pp. [Available online at http://gradworks.umi.com/35/07/3507433.html.]
Kunz, M., and P. I. S. Kugel, 2015: Detection of hail signatures from single-polarization C-band radar reflectivity. Atmos. Res., 153, 565–577, doi:10.1016/j.atmosres.2014.09.010.
Kuster, C. M., P. L. Heinselman, and T. J. Schuur, 2016: Rapid-update radar observations of downbursts occurring within an intense multicell thunderstorm on 14 June 2011. Wea. Forecasting, 31, 827–851, doi:10.1175/WAF-D-15-0081.1.
Lee, W. C., R. M. Wakimoto, and R. E. Carbone, 1992: The evolution and structure of a “bow-echo-microburst” event. Part II: The bow echo. Mon. Wea. Rev., 120, 2211–2225, doi:10.1175/1520-0493(1992)120<2211:TEASOA>2.0.CO;2.
Markowski, P., and Y. Richardson, 2010: Mesoscale Meteorology in Midlatitudes. Wiley-Blackwell, 430 pp.
McCarthy, J., J. W. Wilson, and T. T. Fujita, 1982: The Joint Airport Weather Studies (JAWS) project. Bull. Amer. Meteor. Soc., 63, 15–22, doi:10.1175/1520-0477(1982)063<0015:TJAWSP>2.0.CO;2.
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, doi:10.1175/2009BAMS2507.1.
Müldner, W., 1950: Die Windbruchschäden des 22.7.1948 im Reichswald bei Nürnberg, ein Beispiel für ein Wirbelfeld als Teilerscheinung einer Böenfront (The wind damage of 22 July 1948 in the Reichswald forest near Nuremberg, an example of a vortex field as a local phenomenon of a gust front). Ber. Dtsch. Wetterdienstes, 19, 3–29.
Muramatsu, T., and R. Kawamura, 2012: The environment and potential predictability of downbursts in Japan. Tenki, 59 (9), 827–845.
Newman, J. F., and P. L. Heinselman, 2012: Evolution of a quasi-linear convective system sampled by phased array radar. Mon. Wea. Rev., 140, 3467–3486, doi:10.1175/MWR-D-12-00003.1.
Ohno, H., O. Suzuki, H. Nirasawa, M. Yoshizaki, N. Hasegawa, Y. Tanaka, Y. Muramatsu, and Y. Ogura, 1994: Okayama downburst on 27 June 1991: Downburst identifications and environmental conditions. J. Meteor. Soc. Japan, 72, 197–222.
Ohno, H., O. Suzuki, and K. Kusunoki, 1996: Climatology of downburst occurrence in Japan. Tenki, 43 (2), 101–112.
Park, S. G., M. Maki, K. Iwanami, V. N. Bringi, and V. Chandrasekar, 2005: Correction of radar reflectivity and differential reflectivity for rain attenuation at X band. Part II: Evaluation and application. J. Atmos. Oceanic Technol., 22, 1633–1655, doi:10.1175/JTECH1804.1.
Proctor, F. H., 1988: Numerical simulations of an isolated microburst. Part I: Dynamics and structure. J. Atmos. Sci., 45, 3137–3160, doi:10.1175/1520-0469(1988)045<3137:NSOAIM>2.0.CO;2.
Proctor, F. H., 1989: Numerical simulations of an isolated microburst. Part II: Sensitivity experiments. J. Atmos. Sci., 46, 2143–2165, doi:10.1175/1520-0469(1989)046<2143:NSOAIM>2.0.CO;2.
Rasmussen, R. M., V. Levizzani, and H. R. Pruppacher, 1984: A wind tunnel and theoretical sturdy on the melting behavior of atmospheric ice particles: III. Experiment and theory for spherical ice particles of radius > 500 μm. J. Atmos. Sci., 41, 381–388, doi:10.1175/1520-0469(1984)041<0381:AWTATS>2.0.CO;2.
Roberts, R. D., and J. W. Wilson, 1989: A proposed microburst nowcasting procedure using single-Doppler radar. J. Appl. Meteor., 28, 285–303, doi:10.1175/1520-0450(1989)028<0285:APMNPU>2.0.CO;2.
Rothermel, J., C. Kessinger, and D. L. Davis, 1985: Dual-Doppler lidar measurement of winds in the JAWS Experiment. J. Atmos. Oceanic Technol., 2, 138–147, doi:10.1175/1520-0426(1985)002<0138:DDLMOW>2.0.CO;2.
Snyder, J. C., H. B. Bluestein, G. Zhanz, and S. J. Frasier, 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, doi:10.1175/2010JTECHA1356.1.
Srivastava, R. C., 1987: A model of intense downdrafts driven by the melting and evaporation of precipitation. J. Atmos. Sci., 44, 1752–1773, doi:10.1175/1520-0469(1987)044<1752:AMOIDD>2.0.CO;2.
Straka, J. M., and J. R. Anderson, 1993: Numerical simulations of microburst-producing storms: Some results from storms observed during COHMEX. J. Atmos. Sci., 50, 1329–1348, doi:10.1175/1520-0469(1993)050<1329:NSOMPS>2.0.CO;2.
Suckstorff, G. A., 1938: Kaltlufterzeugung durch Niederschlag (Cold air generation by precipitation). Meteor. Z., 55, 287–292.
Testud, J., E. Le Bouar, E. Obligis, and M. Ali-Mehenni, 2000: The rain profiling algorithm applied to polarimetric weather radar. J. Atmos. Oceanic Technol., 17, 332–356, doi:10.1175/1520-0426(2000)017<0332:TRPAAT>2.0.CO;2.
Vasiloff, S. V., 2001: Improving tornado warnings with the Federal Aviation Administration’s Terminal Doppler Weather Radar. Bull. Amer. Meteor. Soc., 82, 861–874, doi:10.1175/1520-0477(2001)082<0861:ITWWTF>2.3.CO;2.
Vasiloff, S. V., and K. W. Howard, 2009: Investigation of a severe downburst storm near Phoenix, Arizona, as seen by a mobile Doppler radar and the KIWA WSR-88D. Wea. Forecasting, 24, 856–867, doi:10.1175/2008WAF2222117.1.
Wakimoto, R. M., 1985: Forecasting dry microburst activity over the High Plains. Mon. Wea. Rev., 113, 1131–1143, doi:10.1175/1520-0493(1985)113<1131:FDMAOT>2.0.CO;2.
Wakimoto, R. M., 2001: Convectively driven high wind events. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 255–298, doi:10.1175/0065-9401-28.50.255.
Wakimoto, R. M., and V. N. Bringi, 1988: Operational detection of microbursts associated with intense convection: The 20 July case during the MIST project. Mon. Wea. Rev., 116, 1521–1539, doi:10.1175/1520-0493(1988)116<1521:DPOOMA>2.0.CO;2.
Weisman, M. L., and R. J. Trapp, 2003: Low-level mesovortices within squall lines and bow echoes. Part I: Overview and dependence on environmental shear. Mon. Wea. Rev., 131, 2779–2803, doi:10.1175/1520-0493(2003)131<2779:LMWSLA>2.0.CO;2.
Willingham, K. M., E. J. Thompson, K. W. Howard, and C. L. Dempsey, 2011: Characteristics of Sonoran Desert microbursts. Wea. Forecasting, 26, 94–108, doi:10.1175/2010WAF2222388.1.
Wilson, J. W., R. D. Roberts, C. Kessinger, and J. McCarthy, 1984: Microburst wind structure and evaluation of Doppler radar for airport wind shear detection. J. Climate Appl. Meteor., 23, 898–915, doi:10.1175/1520-0450(1984)023<0898:MWSAEO>2.0.CO;2.
Wolfson, M. M., 1990: Understanding and predicting microbursts. Preprints, 16th Conf. on Severe Local Storms, Kananaskis Park, AB, Canada, Amer. Meteor. Soc., 340–351.
Wu, T., Y. Takayanagi, S. Yoshida, T. Funaki, T. Ushio, and Z.-I. Kawasaki, 2013: Spatial relationship between lightning narrow bipolar events and parent thunderstorms as revealed by phased array radar. Geophys. Res. Lett., 40, 618–623, doi:10.1002/grl.50112.
Yoshikawa, E., T. Ushio, Z.-I. Kawasaki, S. Yoshida, T. Morimoto, F. Mizutani, and M. Wada, 2013: MMSE beam forming on fast-scanning phased array weather radar. IEEE Trans. Geosci. Remote Sens., 51, 3077–3088, doi:10.1109/TGRS.2012.2211607.
Zhang, G., J. Vivekanandan, and M. K. Politovich, 2004: Radar/radiometer combination to retrieve cloud characteristics for icing detection. Preprints, 11th Conf. on Aviation, Range, and Aerospace, Hyannis, MA, Amer. Meteor. Soc., P6.6. [Available online at https://ams.confex.com/ams/11aram22sls/techprogram/paper_81556.htm.]
Zhang, G., R. J. Doviak, D. S. Zrnić, R. Palmer, L. Lei, and Y. Al-Rashid, 2011: Polarimetric phased-array radar for weather measurement: A planar or cylindrical configuration? J. Atmos. Oceanic Technol., 28, 63–73, doi:10.1175/2010JTECHA1470.1.
Zrnić, D. S., and Coauthors, 2007: Agile-beam phased array radar for weather observations. Bull. Amer. Meteor. Soc., 88, 1753–1766, doi:10.1175/BAMS-88-11-1753.
