1. Introduction
A hailstorm is a typical type of convective weather. Besides the hailstones, it is also usually accompanied by a large rain rate, strong wind, and active lightning discharges and therefore poses threats to agriculture, ecology, transportation, personnel safety, and so on. With population growth and increasing urbanization, economic disasters caused by hailstorms are becoming more frequent (Charlton et al. 1995; Kunz et al. 2009; Brooks 2013; Mohr and Kunz 2013; Allen 2017). Hail events (HEs) tend to occur in storms with strong updraft. A strong updraft is necessary to support the growth of hail embryos into large hailstones and can exceed 40–50 m s−1 in some convective storms producing large hail (Musil et al. 1986; Miller et al. 1988; Johns and Doswell 1992; Rosenfeld et al. 2008). The occurrence of hailstorms typically depends on favorable environmental conditions (Johnson and Mapes 2001; Brooks 2013; Allen 2017; Dennis and Kumjian 2017). For example, Miller (1972) reported that the wet bulb zero (WBZ) degree layer was a key factor for the occurrence of severe hail, and the WBZ associated with hailstorms was most likely located between 2134 and 3353 m above ground level. Most extreme hailstones were generated by supercell storms, which typically occurred in environmental conditions with strong instability and deep vertical wind shear (Johns and Doswell 1992; Johnson and Sugden 2014; Tuovinen et al. 2015; Dennis and Kumjian 2017; Suárez Molina et al. 2020).
While large hail events (LHEs) have gained more attention, small hail events (SHEs) have rarely been reported in literature. According to the definition of the China Meteorological Administration, a solid precipitation particle above 2 mm in diameter can be called “a hailstone,” and “large hail” is defined as a hailstone with a diameter of ≥20 mm, which is also the threshold to issue a severe hail warning. In practice, the SHEs (2 ≤ diameter < 20 mm) account for a large proportion of HEs, and they can also lead to severe damages (Grahame et al. 2009; Kalina et al. 2016; Kumjian et al. 2019). Crowded small-size hailstones can form “ice sheets” on the ground that can cause freezing disasters in crops and can paralyze the transportation and power system. According to Tippett et al. (2015) and Allen and Tippett (2015), in the Unites States during 1955–2014, while large-hail days seemed to change little, small-hail days showed increasing tendency. Ni et al. (2017) reported that the hail size has decreased in China since 1980: the annual mean size of 20+ mm hailstones displayed a decreasing trend of 1.7 mm decade−1; and the mean sizes of large hailstones in two periods of 1988–97 and 1998–2015 were 31.93 and 28.37 mm, respectively. Furthermore, Ni et al. (2020) found that the mean annual maximum hail size during 1980–2015 was smaller than the previous 20 years in China, especially in northern China. One of the biggest challenges to the forecasting of hail is to predict hail size. M. Li et al. (2018) investigated the spatiotemporal characteristics of HEs with different maximum hail diameter (MHD) and their associated environmental conditions, and they suggested that the environmental parameters describing the atmospheric stratification corresponding to different MHD of HE might be partly different and partly similar and could vary greatly depending on latitude and longitude. Overall, our understanding of the distinction between the environmental conditions and storm structures corresponding to the SHEs and LHEs is still relatively limited.
The Beijing–Tianjin–Hebei (BTH) region is one of the several areas with relatively frequent hailstorms in China (X. Li et al. 2018). With its large population and significant role in the agricultural industry, the BTH region is particularly sensitive to HEs. In this study, we focused on HEs in this region and investigated their associated atmospheric conditions and hailstorm structures. We divided the HEs into two classifications, namely, LHEs and SHEs, to explore the differences between them in spatial and temporal distributions, atmospheric stratification, and parent hailstorm structures. We then presented the factors potentially responsible for the size of hail.
2. Data and method
The data used in this study included HE records documented by the Chinese national meteorological stations, ERA-Interim reanalysis data, and S-band Doppler weather radar data.
a. Hail event records
HEs recorded at Chinese national meteorological stations between 2014 and 2018 were used. There was a total of 172 stations in the study area (Fig. 1). Typical information of the HEs were manually recorded, including the date, duration, location, and the maximum size of hailstone. Note that some HEs would not have been recorded because they did not impact the meteorological stations. The maximum hail size was measured or visually estimated by trained meteorological spotters. According to the Chinese specifications for surface meteorological observation, solid ice particles falling on the ground can be documented as hailstones if they feature spherical or conical shapes with diameters greater than 2 mm. From 1 January 2014, the nighttime [2000–0759 Beijing time (BT)] manual observation of hail was no longer officially required; thus, HEs at night were possibly missed. In this study, hailfall recorded at one station was reported as a single HE. If the time interval between two HEs at one station was more than 15 min, they were treated as two HEs. The MHD was retrieved by referencing the maximum-size hail in one HE.
A total of 172 Chinese national meteorological stations (red stars) and 4 referenced S-band Doppler radars (yellow triangles) in the BTH region overlaid on a terrain map. The blue circles depict the 30-km (inner circle) and 150-km (outer circle) ranges of radars. The black lines indicate provincial/city borders. “ASL (m)” indicates meters above mean sea level.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
Some hail records contained accurate hail diameters, but some contained only descriptive expressions (e.g., “beans,” “eggs,” “peanuts”). Descriptive sizes were converted to approximate numeric values by referring to a quality-controlled classification used by the National Meteorological Information Center (X. Li et al. 2018) (Table 1). For this study, LHE and SHE indicated that the MHD of the hail was ≥20 mm and 2 ≤ diameter < 20 mm, respectively.
Approximate hailstone diameter, as interpreted from hail records.
From 2014 to 2018, a total of 286 HEs were recorded at 119 national meteorological stations in the BTH region, including 21 LHEs recorded by 19 stations and 265 SHEs recorded by 112 stations. The SHEs accounted for 92.7% of the documented HEs.
b. ERA-Interim reanalysis data
ERA-Interim is a global atmospheric reanalysis that is available from 1 January 1979 to 31 August 2019. ERA-Interim data, obtained from the European Centre for Medium-Range Weather Forecasts, were used to obtain the atmospheric environment associated with the HEs. The ERA-Interim gridded data products herein included 6-hourly (0000, 0600, 1200, and 1800 UTC) surface and upper-air parameters, with a horizontal spatial resolution of 0.125° and 60 vertical levels from the surface up to 0.1 hPa. Detailed descriptions of the ERA-Interim products archive and reanalysis system have been published by Berrisford et al. (2009) and Dee et al. (2011). For the analysis of the atmospheric stratification conditions associated with HEs, we chose the data at the grid nearest to the recorded HE station, from the period immediately before HE occurrence.
The surface dataset used in this study included the 2-m temperature T, 2-m dewpoint temperature Td, 10-m eastward wind component u, and 10-m northward wind component υ. The upper-air parameters included T, relative humidity (RH), u, and υ at pressure levels of 1000, 925, 850, 700, 500, 400, 300, 250, 200, 150, and 100 hPa.
c. Radar data
The structures of 39 hailstorms responsible for 19 LHEs and 20 SHEs were investigated in section 5, based on the data of the S-band Doppler radars located in Beijing (116°28′19″E, 39°48′32″N), Tanggu (117°43′1″E, 39°2′ 38″N), Shijiazhuang (114°42′42″E, 38°21′5″N), and Qinhuangdao (118°52′51″E, 39°52′52″N) (Fig. 1). Each HE was determined to correspond to a hailstorm process by first finding the cell at the position and time of the HE, and then tracking the development of this cell before and after the HE. For the hailstorms corresponding to 19 of the 21 LHEs, the cells producing the hailfall were located in a position approximately 30–150 km from the radars (Fig. 1) during the period before and when the hail fell, avoiding the impacts of the silence cone near the radar and low detection quality far away from the radar. The hailstorms corresponding to 20 SHEs were more strictly selected, by requiring that their records of hail size were precise (it means the numerical hail diameter was given), and their positions during the entire lifetime were located within 30–150 km of the radars. The radars provided precipitation volume coverage patterns and completes one volume scan at nine elevation angles from 0.5° to 19.5° and 360 azimuths (1° interval) in 6 min with a radial resolution of 1 km. The referenced radar products in this study included mesoscale rotation and vertically integrated liquid (VIL) water (Amburn and Wolf 1997). These data were obtained from the Radar Product Generator (RPG) and the Principal User Processor (PUP), which are the software applications used by the Chinese Meteorological Administration to perform quality control on raw radar data and generate radar products with the same algorithm as that for WSR-88D (Belville 2002).
Mesoscale rotation is a vortex of air with a horizontal extension of approximately 2–15 km in the updraft area of a convective storm. Long-lived mesoscale rotation with significant vertical extension generally produces a typical rotation dipole signature in the radar radial velocity image because half of the rotation moves toward the radar (negative radial velocity), whereas the other half of the rotation moves away (positive radial velocity). We manually calculated the maximum rotational velocity, which presents the azimuthal shear in the hailstorm, by reading off the maximum inbound and outbound velocities, adding their absolute values, and then dividing by 2. Their distance was required to be > 1 and ≤ 15 km in the dipole (Zrnić et al. 1985). The mesoscale rotation height (above ground level) was determined from the center of the maximum rotational velocity. In this study, the investigated mesoscale rotations were all cyclonic.
The VIL algorithm is based on theoretical studies of drop size distribution and empirical studies of the relationship between the reflectivity factor and liquid water content. It is given by the equation
3. Spatial and temporal distributions
a. Spatial distribution
Figure 2 shows the spatial distribution of HEs at meteorological stations. The three stations recording the most HEs were Guyuan (11 events), Chongli (8 events), and Pingquan (8 events) stations (they are labeled in Fig. 2). The majority of stations with more than five HEs were located in mountainous areas or near the foot of the mountains, whereas the stations with no more than five HEs were predominantly located in the plain areas. This result agrees with X. Li et al. (2018), who reported that HEs occurred more frequently in mountainous areas than in the plains in their investigation of HE activity throughout China. On the other hand, among the 21 LHEs, 13 LHEs (61.9%) were located in the hinterland of the plain areas, 6 LHEs (28.6%) were located near the boundary between the mountain and plain areas, and only 2 LHEs (9.5%) occurred in the mountains (Fig. 3). This suggests that although more HEs were associated with the meteorological stations in mountainous areas, they tended to be dominated by SHEs.
The spatial distribution of hail events embedded along the contours of a terrain map of the BTH region.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
The spatial distribution of maximum hail diameter overlaid on the contours of a terrain map of the BTH region. The red triangles are the stations where LHEs were recorded, and the blue triangles are the stations where SHEs were recorded. The red triangle with a 3 inside means there were three LHEs recorded in this station during the study period.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
Complex boundaries formed by hills and mountains have a significant impact on clouds and precipitation (Houze 2014). In many regions of the world, the origin of convective systems can be traced back to upstream mountains. The BTH region is located in complex terrain: the Yanshan and Taihang Mountain ranges are in the north and west, respectively, and the Bohai Sea is located in the east (see Fig. 1). Thunderstorms in the BTH region are frequently initiated over the mountains and move toward the southeast (Wilson et al. 2010). Large hailstones are typically associated with severe storms featuring strong dynamics and microphysical processes (also see section 5). The plains in the BTH region have conditions conducive to severe storms. First, there are two main water vapor origins of BTH region; one is the influxes from the Bohai Sea lying to the east (Wilson et al. 2010), the other is the south or southwest airflow influenced by the extension of the west Pacific subtropical high (Liang et al. 2011). During the summer, moist airflow converges in the windward direction, resulting in water vapor gathering over the plains and foot of the mountains. This provides the large volume of water vapor required for severe storms and the growth of hailstones. Second, a variety of boundary layer convergence lines, including synoptic fronts, sea breezes generated by the Bohai Sea, horizontal convection rolls, etc., frequently reach the plains where they can initiate and advance the storm development (Wilson et al. 2010). Third, the urban heat island effect caused by metropolitan areas also appears to occasionally contribute to triggers of severe storms in this region (Dixon and Mote 2003). Thus, the plains provide a favorable environment for intense convection storms and LHEs.
b. Temporal distribution
Figure 4 shows the monthly distribution of HEs and MHD, respectively. The HEs were observed to occur between March and September, and peak in June (121 samples) (Fig. 4a). The average MHD was found to be comparatively larger from May to July (with median MHDs of 7.0 mm and mean MHDs between 7.8 and 9.4 mm) than in any other month (Fig. 4b). Among the recorded LHEs, approximately 57.1% (12 of 21) of samples happened in June (Fig. 4c), possibly suggesting the preference of LHE for this month. The LHEs had a mean MHD of 32.2 mm in June (Fig. 4d). SHE also tended to peak in June (109 samples) and had relatively larger MHD from May to July (Figs. 4e,f). In the BTH region, the East Asian summer monsoon carrying the warm and moist air invades in June; it combines with frequent intrusions of midtropospheric cold and dry air to provide favorable moisture and dynamic forcing conditions for the occurrence of hailstorms (Zhang et al. 2013). Yihui and Chan (2005) reported that the monsoon rainfall during the warm season in the BTH region peaks in June. M. Li et al. (2018) also found that the seasonal variations of HEs were associated with the summer monsoon.
Monthly variations of HEs (bar charts) and MHD (box-and-whisker plots) in the BTH region. Whiskers in the box-and-whisker plots indicate the 5th–95th-percentile values; the bottom and top of the boxes are the first (25th percentile) and third (75th percentile) quartiles, respectively; the orange line inside the box is the median value; and the green triangle is the mean value. The table below each diagram lists the specific medians and means of MHD.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
Figure 5 shows the distribution of HEs and MHD in different periods of the day. The day was divided into five periods: morning (0800–1100 BT), noon (1100–1400 BT), afternoon (1400–1700 BT), evening (1700–2000 BT), and night (2000–0800 BT). Note that, because the manual observation of hail at night has not been mandatory since 1 January 2014, the samples at night are assumed to be incomplete. Because of this, we will ignore the nighttime samples in the analysis of HE quantity, although they are also shown in Fig. 5. The HEs predominantly occurred in the afternoon and evening, with 109 HEs in each period and together counting for approximately 76.2% of the samples (Fig. 5a). In an additional investigation, the HE peaked in the interval between 1600 and 1700 (the figure is omitted). The average MHD during the evening and night periods (9.5 and 9.2 mm) were larger than that during other periods (Fig. 5b). The LHEs were more frequent and featured larger MHD (mean of 32.5 mm) in the evening (Figs. 5c,d). Most of SHEs occurred in the afternoon and evening, with a total of 200 samples in these periods (Fig. 5e). On average, the MHD of SHEs in the periods from afternoon to night seemed to be larger than that in the morning and noon periods (Fig. 5f).
Diurnal variations of HEs (bar charts) and MHD (box-and-whisker plots) in the BTH region. Each day was divided into five periods morning (0800–1100 BT), noon (1100–1400 BT), afternoon (1400–1700 BT), evening (1700–2000 BT), and night (2000–0800 BT). The box-and-plot plot is interpreted as shown in the caption of Fig. 4. The median and mean values for MHD are listed in the tables below the diagrams, where “nan” means no value. Because the manual observation of hail at night has not been mandatory since 1 Jan 2014, the samples at night are assumed to be incomplete (the “night” bar graph is shown in gray).
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
Hailstorms in the study area occurred more often in the afternoon and evening than in morning or midday. For high amounts of moisture to be drawn into the hailstorm along with warm rising air, there must be instability in the atmosphere. As a result of daytime heating, atmospheric instability tends to be at its strongest during the late afternoon and evening, and its release can cause intense convection and severe storms (He and Zhang 2010; Chen et al. 2012).
4. Atmospheric environment associated with HEs
In this section, we compare the atmospheric environment conditions associated with LHEs and SHEs by investigating the thermodynamic parameters contained in the ERA-Interim data. We demanded that the HEs and environment parameters must be matched in both time and space. The environmental parameters at the position nearest to and the time just prior to the reported HE were associated with the HEs. The environmental parameters corresponding to 21 LHEs and 265 SHEs were analyzed and compared. The results are shown in Fig. 6, in the form of violin plots.
Violin plots showing the distributions of the environmental parameters including (a) 2-m temperature T, (b) 2-m dewpoint Td, (c) average RH below 850 hPa, (d) RH (925–500 hPa), (e) CAPE, and (f) wind shear (500–1000 hPa). The width of the violin shape is proportional to the ratio of HEs at a given environmental parameter value to all HEs. Black boxplots are overlaid on the violins. The bottom of the black box indicates the 25th percentile, the white dotted line indicates the median value of the distribution, and the top of the black box indicates the 75th percentile. The table below each diagram lists the corresponding medians and means.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
We will mainly focus on the comparison of mean and median environmental parameters and the environmental parameters associated with the most samples (corresponding to the widest positions of the violin plots) between the LHEs and SHEs. Although the sample size of LHE was relatively small, these mentioned parameters still indicated the difference in atmospheric stratification between LHE and SHE. In general, the LHEs tended to be associated with the atmospheric stratification featuring greater T (Fig. 6a), Td (Fig. 6b), RH below 850 hPa (Fig. 6c), difference in RH between 925 and 500 hPa (Fig. 6d), CAPE (Fig. 6e), and wind shear between 500 and 1000 hPa (Fig. 6f). This result is in line with expectations. Agreeing with previous studies (Weisman and Klemp 1982, 1984; Kaltenböck et al. 2009; Johnson and Sugden 2014; Tuovinen et al. 2015; Dennis and Kumjian 2017; Kahraman et al. 2017; Zheng et al. 2009), it is indicated that the high 2-m temperature, wet lower layer, dry middle layer, a large amount of unstable energy, and organized ambient wind are conducive to the stronger convection, therefore more likely to correspond to LHEs. On the other hand, although there were some differences between the environmental parameters associated with LHEs and SHEs, the significant overlaps of the distributions of the parameters between these two categories also hinted that the factors determining the hail size are comprehensive and complicated, and the environmental atmospheric stratification is a relatively indirect influencing factor.
Some previous studies also investigated the relationship between hail size and environmental parameters. Taszarek et al. (2020) suggested that increasing moisture content generally favors increasing hail size in Europe, especially the increase of moisture at lower levels. Rasmussen and Blanchard (1998) found that the supercells yielding large hail tended to be associated with greater CAPE (median: 1152 J kg−1) than the nonsupercell thunderstorms (median: 537 J kg−1). Johnson and Sugden (2014) indicated that CAPE displayed a slight increasing tendency as hail size becomes larger. Taszarek et al. (2020) also revealed that the HEs with diameter no less than 5 cm notably corresponded with larger CAPE than those with diameter less than 2 cm. Rasmussen and Blanchard (1998) investigated the wind shear vector between 0–500 m and 6 km, and found that the nonsupercell thunderstorms had values between 5 and 15 m s−1, generally smaller than the supercells yielding large hail, which had values between 11 and 21 m s−1. Taszarek et al. (2020) presented that effective shear (Thompson et al. 2007) had the best skill in discriminating between the nonsevere thunderstorms and those producing ≥5-cm hail events. Kumjian and Lombardo (2020) suggested that increased shear and broader updraft could lead to the increased residence time of hail, permitting longer trajectories and thus larger sizes. Gensini et al. (2021) also proposed that the 0–6-km wind shear was one of the top three important variables in predicting severe hail.
5. Structures of storm responsible for HEs
The radar observations of 19 hailstorms associated with LHEs and 20 hailstorms associated with SHEs were analyzed to compare their structures including mesoscale rotation and VIL.
a. Mesoscale rotation in hailstorms
In severe storms producing large hailstones, hail embryos can typically be found in regions close to the rotating updraft (Miller et al. 1988). The hail embryos would limit each other’s growth by depleting the cloud water. Rotating updrafts introduce “unfair competition,” which makes bigger hail embryos get access to undepleted cloud water and enables the growth of large hail (Browning and Foote 1976; Browning 1977; Knight and Knight 2001). Rotunno and Klemp (1985) demonstrated that storm rotation maintains supercell storm structure and updraft propagation through a vertical pressure perturbation associated with the updraft region. In this way, vertical pressure gradient forces induced by rotation enhance vertical accelerations within the updraft, well beyond those accelerations associated with buoyancy alone (Weisman and Klemp 1982; Rotunno and Klemp 1985; McCaul and Weisman 2001).
The maximum rotational velocity and its height were obtained by the method described in section 2. Mesoscale rotations were found in 38 of 39 hailstorms, except for the hailstorm associated with the LHE occurred on 7 July 2017, which was embedded in an intense squall line and did not show discernable mesoscale rotation. Lee and White (1998) provided a mesocyclone strength nomogram for measuring the maximum rotational velocity; this nomogram comprehensively considers radar beamwidth, rotational velocity, circulation size, and range from radar. According to this nomogram, the rotation strength can be divided from weak to strong into four levels: “weak shear,” “minimal mesocyclone,” “moderate mesocyclone,” and “strong mesocyclone.” Figure 7 gives the information on the mesoscale rotations of the hailstorms.
Distributions of (a) mesoscale rotation strength, (b) maximum rotational velocity, and (c) mesoscale rotation height according to 19 hailstorms associated with LHEs and 20 hailstorms associated with SHEs. The explanations for (b) and (c) are the same as in Fig. 6.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
Figure 7a shows that, while the rotation strength of 70.0% (14 of 20) hailstorms associated with SHEs was classified as weak shear, about 89.5% (17 of 19) of hailstorms associated with LHEs were accompanied by mesocyclones. Furthermore, the classification of strong mesocyclone contained the largest number of hailstorms associated with LHEs (8 of 19). According to the general definition of supercell suggested by Burgess and Lemon (1990) and Doswell and Burgess (1993), most of the hailstorms associated with LHEs in present study can actually be classified as supercells. The close link between mesocyclones and LHEs has also been suggested by previous studies (Miller et al. 1988; Witt et al. 1998; Blair et al. 2011; Wapler 2017). Figure 7b shows that the hailstorms associated with LHEs tended to have greater maximum rotational velocity than those with SHEs, with the former and latter having mean of 18.9 and 10.9 m s−1, respectively, and median of 20.5 and 9.5 m s−1, respectively. The violin plots further suggest that the highest occurrence possibility (corresponding to the widest position of the violin plots) of the maximum rotational velocity for the hailstorms associated with LHEs and SHEs were 22.5 and 9.0 m s−1, respectively. The difference in the mesoscale rotation height between the hailstorms associated with LHEs and SHEs was not significant according to their means, medians, and distribution patterns (Fig. 7c).
b. Vertically integrated liquid
In some studies, VIL information was used to discriminate between giant hail and large hail (Amburn and Wolf 1997; Lahiff 2005; Blair et al. 2011). However, correspondence between VIL magnitude and size of hailstones is weak (Edwards and Thompson 1998), indicating that the relationship between hail size and VIL may be more complicated.
In this study, we considered the variation in the VIL of hailstorms and its relationship with LHEs and SHEs. All the hailstorms showed an unambiguous increase in VIL before the hailstones were observed. Here, the rapid-increase part of VIL was called “VIL rapid increase (RI).” Some VIL-based parameters are defined in Fig. 8, referring to the temporal variation of the VIL in the hailstorm associated with the LHE on 29 June 2016. We proposed the following procedure for the identification of VIL RI. When the increase of VIL between two adjacent radar volume scans (6-min interval) was first greater than a certain value (called δVIL here), the time of the previous radar volume scan was considered to be the beginning of the VIL RI. If the VIL increase corresponding to one radar volume scan was less than δVIL, the VIL RI was considered to continue only when the VIL increase in the next radar volume was greater than 2δVIL. The VIL RI was determined to end when the increases in VIL for two consecutive radar volume scans were both less than δVIL. The end time of the VIL RI was decided to be the time of the last radar volume that met the abovementioned VIL increase criteria. If more than one VIL RI period was recorded, the one with the largest VIL RI variation would be chosen (This was not the case in current samples). After testing, we defined δVIL to be equal to 10 kg m−2. VIL RI rate was calculated by dividing RI variation by RI time. The distributions of VIL-related parameters and their comparisons between the hailstorms associated with LHEs and SHEs are shown in Fig. 9 in the form of violin diagrams.
Explanation of the parameters related to the vertically integrated liquid as based on the VIL variation (blue line) of the hailstorm yielding LHE on 29 Jun 2016. The yellow triangle indicates the time when the hailfall occurred.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
Distributions of the VIL-related parameters including (a) VIL before the rapid increase, (b) variation of RI, (c) VIL RI rate, (d) maximum VIL increase rate during RI, (e) VIL after RI, (f) time from RI end to hailfall, and (g) mean VIL from RI end to hailfall. The explanation for the plots is as in Fig. 6.
Citation: Journal of Applied Meteorology and Climatology 62, 5; 10.1175/JAMC-D-22-0124.1
According to Fig. 9, the hailstorms associated with LHEs and SHEs did not show a significant difference in most of the VIL-related parameters, referring to their comparison in mean, median, and distribution ranges. In contrast, two parameters including the VIL before RI (Fig. 9a) and the time from end of RI to hailfall occurrence (Fig. 9f) exhibited relatively distinct differences between the hailstorms associated with LHEs and SHEs, especially in terms of their means and medians. Before the VIL RI, the hailstorms associated with LHEs and SHEs had the mean VIL of 23.6 and 15.7 kg m−2, respectively, and median VIL of 21.5 and 14.0 kg m−2, respectively, with, for both mean and median values, the LHE being approximately 1.5 times the SHE. What is more remarkable is that the hailstorms associated with LHEs exhibited much longer interval between the end of RI and the hailfall occurrence than those associated with SHEs, with the former having mean and median of 52.0 and 45.0 min, respectively, and the latter having mean and median of 26.5 and 18.0 min, respectively.
The comparison of the VIL-related parameters implies some interesting things about the difference of hailstone growth process in the hailstorms associated with LHEs and SHEs. In the early stage, although the hailstorms associated with LHEs and SHEs both had a relatively low VIL, the former, with greater mean VIL, tended to have relatively superior conditions for availability of liquid water, which might be consistent with the fact that the LHEs tended to be associated with the wetter lower level in atmospheric stratification (section 4). Thereafter, the hailstorms associated with LHEs and SHEs both experienced VIL RI with analogous magnitude, according to the statistics and distributions of variations of RI (Fig. 9b), RI rates (Fig. 9c) and maximum VIL increase rates (Fig. 9d), which means that strong updrafts transported large amounts of water vapor into the cloud, causing a great increase in number of the liquid water droplets. This process would be accompanied by the generation and growth of hail, simultaneously consuming liquid water. Interestingly, after the VIL RI, there was a significantly longer duration before the hailfall for the hailstorms associated with LHEs, relative to the hailstorms associated with SHEs (Fig. 9f), while the mean VIL during this period for the hailstorms associated with LHEs was only slightly greater than that for the hailstorms associated with SHEs (Fig. 9g). It is deduced that the stronger updraft in the hailstorms associated with LHEs tended to lift the hail for a longer period; this would cause the hail to continue to grow as water droplets are “captured.” This process would continue until the updraft was unable to support the large hailstones. In contrast, the hailstorm associated with SHEs tended to feature a relatively weak updraft, and the hail would therefore fall while it was still at a relatively small size. This would be responsible for the shorter period from the end of VIL RI to the hailfall.
6. Summary
Using hailfall records in national meteorological stations from 2014 to 2018, combined with the ERA-Interim reanalysis data and Doppler weather radar data, we explored the spatiotemporal distribution of HEs in the BTH region, and aimed to understand the role of the environmental conditions and hailstorm structures in deciding the size of hail by comparing them in two classifications: LHEs and SHEs. The conclusions are outlined below.
The meteorological stations in the mountainous area recorded more HEs than those in the plains over the BTH region. Most of LHEs occurred in the plains and near the foot of the mountains, and the HEs in the mountainous areas were predominantly SHEs. The recorded HEs started in March, ended in September, and peaked in June. The average hailstone size was found to be larger from May to June. From observations during the daytime hours, HEs occur predominantly in the afternoon and evening, as compared with morning and midday, whereas LHEs tended to be more in the evening. The average MHD of HEs seemed to be larger during the evening and night periods.
The comparison of environmental parameters of LHEs and SHEs suggested that the LHEs tended to correspond to higher 2-m temperature, wetter lower layer, larger difference in RH between 925 and 500 hPa, greater CAPE, and stronger wind shear than the SHEs, mainly referring to the means and medians of the environmental parameters. More hail observations at night are needed to achieve a robust statistical sample of these parameters.
Mesoscale rotations were found in 38 of 39 hailstorms. Whereas 70.0% (14 of 20) of hailstorms associated with SHEs featured weak shear, approximately 89.5% (17 of 19) of hailstorms associated with LHEs were accompanied by mesocyclones, of which 8 hailstorms belonged to the classification of strong mesocyclone. With regard to the maximum rotational velocity, the hailstorms associated with LHEs had mean and median of 18.9 and 20.5 m s−1, respectively, distinctly greater than those associated with SHEs, which had mean and median of 10.9 and 9.5 m s−1, respectively.
Hailstorms exhibited obvious VIL RI before hailfall occurrence. It was found that the most significant difference in VIL-related parameters between the hailstorms associated with LHEs and SHEs were the time interval between the end of VIL RI and the occurrence of hailfall, with the former having mean of 52.0 min and median of 45.0 min and the latter having mean of 26.5 min and median of 18.0 min. This indicates that, supported by strong updrafts, the rapidly increasing and long-duration liquid water supply provides favorable conditions for the growth of hailstone size.
We emphasize that the sample size of this study is relatively small, especially the sample size of LHEs, because 1) an LHE is inherently relatively rare, 2) only records at national meteorological stations were selected so as to ensure the accuracy of data, and 3) the manual recording of hailfall at night has been no longer mandatory, which likely affects the robustness of the results presented here. We look forward to future testing of these results with additional data.
Acknowledgments.
This work was supported by the National Natural Science Foundation of China (42175090), Special Fund for Forecaster of China Meteorological Administration (CMAYBY2019–154), and Basic Research Fund of CAMS (2020Z009).
Data availability statement.
The data associated with this paper can be accessed online (https://doi.org/10.5281/zenodo.7503269) or can be obtained by contacting the corresponding author.
REFERENCES
Allen, J. T., 2017: Hail potential heating up. Nat. Climate Change, 7, 474–475, https://doi.org/10.1038/nclimate3327.
Allen, J. T., and M. K. Tippett, 2015: The characteristics of United States hail reports: 1955–2014. Electron. J. Severe Storms Meteor., 10 (3), https://ejssm.com/ojs/index.php/site/article/view/60.
Amburn, S. A., and P. L. Wolf, 1997: VIL density as a hail indicator. Wea. Forecasting, 12, 473–478, https://doi.org/10.1175/1520-0434(1997)012<0473:VDAAHI>2.0.CO;2.
Belville, J. D., 2002: Abbreviation for Weather Surveillance Radar–1988 Doppler (WSR-88D) integrated logistics support plan. NWS Tech. Rep. R400-IS301C, 89 pp., https://www.roc.noaa.gov/wsr88d/publicdocs/ilspfinal.pdf.
Berrisford, P., D. Dee, K. Fielding, M. Fuentes, P. Kallberg, S. Kobayashi, and S. Uppala, 2009: The ERA-Interim archive. ERA Rep. Series 1, 16 pp., https://www.ecmwf.int/en/elibrary/73681-era-interim-archive.
Blair, S. F., D. R. Deroche, J. M. Boustead, J. W. Leighton, B. L. Barjenbruch, and W. P. Gargan, 2011: A radar-based assessment of the detectability of giant hail. Electron. J. Severe Storms Meteor., 6 (7), https://ejssm.com/ojs/index.php/site/article/view/34.
Brooks, H. E., 2013: Severe thunderstorms and climate change. Atmos. Res., 123, 129–138, https://doi.org/10.1016/j.atmosres.2012.04.002.
Browning, K. A., 1977: The structure and mechanisms of hailstorms. Hail: A Review of Hail Science and Hail Suppression, Meteor. Monogr., No. 16, Amer. Meteor. Soc., 1–43, https://doi.org/10.1007/978-1-935704-30-0_1.
Browning, K. A., and G. B. Foote, 1976: Airflow and hail growth in supercell storms and some implications for hail suppression. Quart. J. Roy. Meteor. Soc., 102, 499–533, https://doi.org/10.1002/qj.49710243303.
Burgess, D. W., and L. R. Lemon, 1990: Severe thunderstorm detection by radar. Radar in Meteorology, D. Atlas, Ed., Amer. Meteor. Soc., 619–647, https://doi.org/10.1007/978-1-935704-15-7_41.
Charlton, R. B., B. M. Kachman, and L. Wojtiw, 1995: Urban hailstorms: A view from Alberta. Nat. Hazards, 12, 29–75, https://doi.org/10.1007/BF00605280.
Chen, M., Y. Wang, F. Gao, and X. Xiao, 2012: Diurnal variations in convective storm activity over contiguous North China during the warm season based on radar mosaic climatology. J. Geophys. Res., 117, D20115, https://doi.org/10.1029/2012JD018158.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, https://doi.org/10.1002/qj.828.
Dennis, E. J., and M. R. Kumjian, 2017: The impact of vertical wind shear on hail growth in simulated supercells. J. Atmos. Sci., 74, 641–663, https://doi.org/10.1175/JAS-D-16-0066.1.
Dixon, P. G., and T. L. Mote, 2003: Patterns and causes of Atlanta’s urban heat island–initiated precipitation. J. Appl. Meteor., 42, 1273–1284, https://doi.org/10.1175/1520-0450(2003)042<1273:PACOAU>2.0.CO;2.
Doswell, C. A., III, and D. W. Burgess, 1993: Tornadoes and tornadic storms: A review of conceptual models. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, Geophys. Monogr., Vol. 79, Amer. Geophys. Union, 161–172, https://doi.org/10.1029/GM079p0161.
Edwards, R., and R. L. Thompson, 1998: Nationwide comparisons of hail size with WSR-88D vertically integrated liquid water and derived thermodynamic sounding data. Wea. Forecasting, 13, 277–285, https://doi.org/10.1175/1520-0434(1998)013<0277:NCOHSW>2.0.CO;2.
Gensini, V. A., C. Converse, W. S. Ashley, and M. Taszarek, 2021: Machine learning classification of significant tornadoes and hail in the United States using ERA5 proximity soundings. Wea. Forecasting, 36, 2143–2160, https://doi.org/10.1175/WAF-D-21-0056.1.
Grahame, N., B. Riddaway, A. Eadie, B. Hall, and E. McCallum, 2009: Exceptional hailstorm hits Ottery St Mary on 30 October 2008. Weather, 64, 255–263, https://doi.org/10.1002/wea.458.
He, H., and F. Zhang, 2010: Diurnal variations of warm-season precipitation over Northern China. Mon. Wea. Rev., 138, 1017–1025, https://doi.org/10.1175/2010MWR3356.1.
Houze, R. A., Jr., 2014: Cloud Dynamics. Academic Press, 496 pp.
Johns, R. H., and C. A. Doswell III, 1992: Severe local storms forecasting. Wea. Forecasting, 7, 588–612, https://doi.org/10.1175/1520-0434(1992)007<0588:SLSF>2.0.CO;2.
Johnson, A. W., and K. E. Sugden, 2014: Evaluation of sounding-derived thermodynamic and wind-related parameters associated with large hail events. Electron. J. Severe Storms Meteor., 9 (5), https://ejssm.org/archives/2014/vol-9-5-2014/.
Johnson, R. H., and B. E. Mapes, 2001: Mesoscale processes and severe convective weather. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 71–122, https://doi.org/10.1007/978-1-935704-06-5_3.
Kahraman, A., M. Kadioglu, and P. M. Markowski, 2017: Severe convective storm environments in Turkey. Mon. Wea. Rev., 145, 4711–4725, https://doi.org/10.1175/MWR-D-16-0338.1.
Kalina, E. A., K. Friedrich, B. C. Motta, W. Deierling, G. T. Stano, and N. N. Rydell, 2016: Colorado plowable hailstorms: Synoptic weather, radar, and lightning characteristics. Wea. Forecasting, 31, 663–693, https://doi.org/10.1175/WAF-D-15-0037.1.
Kaltenböck, R., G. Diendorfer, and N. Dotzek, 2009: Evaluation of thunderstorm indices from ECMWF analyses, lightning data and severe storm reports. Atmos. Res., 93, 381–396, https://doi.org/10.1016/j.atmosres.2008.11.005.
Knight, C. A., and N. C. Knight, 2001: Hailstorms. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 223–254, https://doi.org/10.1007/978-1-935704-06-5_6.
Kumjian, M. R., and K. Lombardo, 2020: A hail growth trajectory model for exploring the environmental controls on hail size: Model physics and idealized tests. J. Atmos. Sci., 77, 2765–2791, https://doi.org/10.1175/JAS-D-20-0016.1.
Kumjian, M. R., Z. J. Lebo, and A. M. Ward, 2019: Storms producing large accumulations of small hail. J. Appl. Meteor. Climatol., 58, 341–364, https://doi.org/10.1175/JAMC-D-18-0073.1.
Kunz, M., J. Sander, and C. Kottmeier, 2009: Recent trends of thunderstorm and hailstorm frequency and their relation to atmospheric characteristics in southwest Germany. Int. J. Climatol., 29, 2283–2297, https://doi.org/10.1002/joc.1865.
Lahiff, C. T., 2005: Vertically integrated liquid density and its associated hail size range across the Burlington, Vermont county warning area. NWS Eastern Region Tech. Attachment 2005-01, 20 pp., https://repository.library.noaa.gov/view/noaa/6653.
Lee, R. R., and A. White, 1998: Improvement of the WSR-88D mesocyclone algorithm. Wea. Forecasting, 13, 341–351, https://doi.org/10.1175/1520-0434(1998)013<0341:IOTWMA>2.0.CO;2.
Li, M., D.-L. Zhang, J. Sun, and Q. Zhang, 2018: A statistical analysis of hail events and their environmental conditions in China during 2008–15. J. Appl. Meteor. Climatol., 57, 2817–2833, https://doi.org/10.1175/JAMC-D-18-0109.1.
Li, X., Q. Zhang, T. Zou, J. Lin, H. Kong, and Z. Ren, 2018: Climatology of hail frequency and size in China, 1980–2015. J. Appl. Meteor. Climatol., 57, 875–887, https://doi.org/10.1175/JAMC-D-17-0208.1.
Liang, F., S. Tao, J. Wei, and C. Bueh, 2011: Variation in summer rainfall in North China during the period 1956–2007 and links with atmospheric circulation. Adv. Atmos. Sci., 28, 363–374, https://doi.org/10.1007/s00376-010-9220-2.
McCaul, E. W., Jr., and M. L. Weisman, 2001: The sensitivity of simulated supercell structure and intensity to variations in the shapes of environmental buoyancy and shear profiles. Mon. Wea. Rev., 129, 664–687, https://doi.org/10.1175/1520-0493(2001)129<0664:TSOSSS>2.0.CO;2.
Miller, L. J., J. D. Tuttle, and C. A. Knight, 1988: Airflow and hail growth in a severe Northern High Plains supercell. J. Atmos. Sci., 45, 736–762, https://doi.org/10.1175/1520-0469(1988)045<0736:AAHGIA>2.0.CO;2.
Miller, R. C., 1972: Notes on analysis and severe storm forecasting procedures of the Air Force Global Weather Central. Air Weather Service Tech. Rep. 200(R), 190 pp., https://apps.dtic.mil/sti/citations/AD0744042.
Mohr, S., and M. Kunz, 2013: Recent trends and variabilities of convective parameters relevant for hail events in Germany and Europe. Atmos. Res., 123, 211–228, https://doi.org/10.1016/j.atmosres.2012.05.016.
Musil, D. J., A. J. Heymsfield, and P. L. Smith, 1986: Microphysical characteristics of a well-developed weak echo region in a High Plains supercell thunderstorm. J. Climate Appl. Meteor., 25, 1037–1051, https://doi.org/10.1175/1520-0450(1986)025<1037:MCOAWD>2.0.CO;2.
Ni, X., Q. Zhang, C. Liu, X. Li, T. Zou, J. Lin, H. Kong, and Z. Ren, 2017: Decreased hail size in China since 1980. Sci. Rep., 7, 10913, https://doi.org/10.1038/s41598-017-11395-7.
Ni, X., A. Muehlbauer, J. T. Allen, Q. Zhang, and J. Fan, 2020: A climatology and extreme value analysis of large hail in China. Mon. Wea. Rev., 148, 1431–1447, https://doi.org/10.1175/MWR-D-19-0276.1.
Rasmussen, E. N., and D. O. Blanchard, 1998: A baseline climatology of sounding-derived supercell and tornado forecast parameters. Wea. Forecasting, 13, 1148–1164, https://doi.org/10.1175/1520-0434(1998)013<1148:ABCOSD>2.0.CO;2.
Rosenfeld, D., W. L. Woodley, A. Lerner, G. Kelman, and D. T. Lindsey, 2008: Satellite detection of severe convective storms by their retrieved vertical profiles of cloud particle effective radius and thermodynamic phase. J. Geophys. Res., 113, D04208, https://doi.org/10.1029/2007JD008600.
Rotunno, R., and J. Klemp, 1985: On the rotation and propagation of simulated supercell thunderstorms. J. Atmos. Sci., 42, 271–292, https://doi.org/10.1175/1520-0469(1985)042<0271:OTRAPO>2.0.CO;2.
Suárez Molina, D., S. Fernández-González, J. C. Suárez González, and A. Oliver, 2020: Analysis of sounding derived parameters and application to severe weather events in the Canary Islands. Atmos. Res., 237, 104865, https://doi.org/10.1016/j.atmosres.2020.104865.
Taszarek, M., J. T. Allen, T. Púčik, K. A. Hoogewind, and H. E. Brooks, 2020: Severe convective storms across Europe and the United States. Part II: ERA5 environments associated with lightning, large hail, severe wind, and tornadoes. J. Climate, 33, 10 263–10 286, https://doi.org/10.1175/JCLI-D-20-0346.1.
Thompson, R. L., C. M. Mead, and R. Edwards, 2007: Effective storm-relative helicity and bulk shear in supercell thunderstorm environments. Wea. Forecasting, 22, 102–115, https://doi.org/10.1175/WAF969.1.
Tippett, M. K., J. T. Allen, V. A. Gensini, and H. E. Brooks, 2015: Climate and hazardous convective weather. Curr. Climate Change Rep., 1, 60–73, https://doi.org/10.1007/s40641-015-0006-6.
Tuovinen, J.-P., J. Rauhala, and D. M. Schultz, 2015: Significant-hail-producing storms in Finland: Convective-storm environment and mode. Wea. Forecasting, 30, 1064–1076, https://doi.org/10.1175/WAF-D-14-00159.1.
Wapler, K., 2017: The life-cycle of hailstorms: Lightning, radar reflectivity and rotation characteristics. Atmos. Res., 193, 60–72, https://doi.org/10.1016/j.atmosres.2017.04.009.
Weisman, M. L., and J. B. Klemp, 1982: The dependence of numerically simulated convective storms on vertical wind shear and buoyancy. Mon. Wea. Rev., 110, 504–520, https://doi.org/10.1175/1520-0493(1982)110<0504:TDONSC>2.0.CO;2.
Weisman, M. L., and J. B. Klemp, 1984: The structure and classification of numerically simulated convective storms in directionally varying wind shears. Mon. Wea. Rev., 112, 2479–2498, https://doi.org/10.1175/1520-0493(1984)112<2479:TSACON>2.0.CO;2.
Wilson, J. W., Y. Feng, M. Chen, and R. D. Roberts, 2010: Nowcasting challenges during the Beijing Olympics: Successes, failures, and implications for future nowcasting systems. Wea. Forecasting, 25, 1691–1714, https://doi.org/10.1175/2010WAF2222417.1.
Witt, A., M. D. Eilts, G. J. Stumpf, J. T. Johnson, E. D. W. Mitchell, and K. W. Thomas, 1998: An enhanced hail detection algorithm for the WSR-88D. Wea. Forecasting, 13, 286–303, https://doi.org/10.1175/1520-0434(1998)013<0286:AEHDAF>2.0.CO;2.
Yihui, D., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117–142, https://doi.org/10.1007/s00703-005-0125-z.
Zhang, L., R. Guo, N. He, and X. Liao, 2013: Characteristic analysis of hail weather in Beijing (in Chinese). Qixiang Keji, 41, 114–120.
Zheng, D., Y. Zhang, Q. Meng, W. Lu, and X. Yi, 2009: Total lightning characteristics and electric structure evolution in a hailstorm. J. Meteor. Res., 23, 233–249.
Zrnić, D. S., D. W. Burgess, and L. D. Hennington, 1985: Automatic detection of mesocyclonic shear with Doppler radar. J. Atmos. Oceanic Technol., 2, 425–438, https://doi.org/10.1175/1520-0426(1985)002<0425:ADOMSW>2.0.CO;2.