Statistical Analysis of Mesovortices During The First Rainy Season In South China

Based on the Doppler radar observation and reanalysis data, the statistical characteristics of mesovortices (MVs) during the rst rainy season (April–June) in South China from 2017 to 2019 are studied, including their spatio-temporal distributions, structural features and favorable environmental conditions. The results show that the MVs usually exhibit short lifetime, among which about 70% last for less than 30 minutes. The intensity and horizontal scale of the MVs are proportional to their lifetime. Long-lived MVs have larger horizontal scales and stronger intensities than short-lived ones. The MVs are mainly observed over the Pearl River Delta region, followed by the western Guangdong province, but relatively fewer in both eastern and northern Guangdong province. The uneven spatial distribution of the MVs is closely related to the differences of environment conditions over South China. The monsoonal south-westerlies, water vapor ux, atmospheric instability and vertical wind shear over southwest Guangdong are signicantly larger than those in other regions, which are favorable for the formation of MVs. The occurrence frequencies of MVs in central and southern parts of Guangdong display similar diurnal variations, reaching the peak during the late afternoon and early evening, while dropping to the minimum overnight. However, the situation is opposite in northern Guangdong, with the peak overnight and the minimum during the late afternoon and early evening.


Introduction
South China is an area with high frequency of severe convective weather, where disastrous weather such as short-term heavy rainfall, hail, tornado and gale frequently occur (Fan and Yu, 2013;Xue et al., 2019). Severe convective weather often causes serious casualties and property losses. For example, on April 13, 2016, a strong squall line process caused gusts in most of Guangdong Province, with the wind speed in most areas reaching 13.9 m·s − 1 (even up to 41.5-46.1 m·s − 1 in some areas). Meanwhile, it was also accompanied by short-term heavy rainfall, causing great economic losses (Wang et al., 2021). On March 4, 2018, a rarely-seen strong squall line occurred over the northern and central parts of Jiangxi Province.
During this process, high wind speeds exceeding 17.2 m·s − 1 were found at 521 observational stations, among which the wind speeds at 172 stations were above 24.5 m·s − 1 . This severe squall line affected more than 267,000 people and caused great damages to a large number of houses and crops, resulting in a direct economic loss of 410 million yuan (Zhang et al., 2019 ).
The forecasting of severe convective weather is one of the key challenging tasks in current weather forecast operation due to its small horizontal scale and the feature of abruptness (Yu et al., 2012).
Studies have revealed that the severe convective weather is closely related to mesoscale convective systems (MCSs). However, previous studies mainly focused on the relationship between convective systems of meso-β-scale (with the horizontal scales ranging from 20 to 200 km, such as squall lines and bow echoes) and the strong convective weather (Evans and Doswell, 2001;Cohen et al., 2012). Slowmoving or quasi-stationary MCSs generally produce rainstorms (Schumacher and Johnson, 2005 May 11, 2014. It is suggested that the rapid splitting and reconstructing of the bow echoes in MCSs are more conducive to the occurrence of extreme precipitation. By using an objective identi cation algorithm for convection initiation based on radar mosaics, Bai et al. (2020) conducted a statistical study on the climatological characteristics of convections over the coastal regions of South China. Their results revealed that convection initiation exhibits three peaks, namely, a late-night-to-morning peak at windward coasts and offshore, a noon-tolate-afternoon peak on the coastal land, and an evening-to-early-morning peak over the northwestern highland.
With the progress of meteorological observation technology, especially the widespread application of Doppler weather radar and the continuous development of numerical models, increasing attention has been paid to the meso-γ-scale (with the horizontal scales being 2-20 km) or even smaller-scale convective systems. Several studies in recent years revealed that strong convective weather, such as tornadoes and severe straight-line winds, is directly caused by meso-γ-scale vortices within MCSs Xu et al., 2015). By using the Advanced Research Weather Research and Forecasting model, Meso-γ-scale vortices can be classi ed into two types. One is mesocyclones caused by convective storm cells (Burgess et al., 1993), and the other is the so-called mesovortices (MVs) generated in the low level of In this paper, we statistically analyze the MVs in the rst rainy season in South China during 2017-2019 according to Doppler radar observations. Combined with the reanalysis data, we further study the environmental characteristics of the MVs to nd out their favorable conditions. Furthermore, the MVs in South China are compared to those in the YHRB. The remainder of this paper is organized as follows.
Data and methods are introduced in Sect. 2. The main results are presented in Sect. 3. Finally, the conclusions and discussions are provided in Sect. 4.

Data
In this paper, observations from nine Doppler weather radars in Guangdong from April to June in 2017-2019 are used. The distribution of these radar stations is shown in Fig. 1, covering almost the whole Guangdong Province. As in T20, the 88D2ARPS program of the Advanced Regional Prediction System (ARPS) from the Center for Analysis and Prediction of Storms (CAPS) at the University of Oklahoma (Xue et al., 2000) is used for the raw radar data quality control, which removes non-meteorological echoes and performs the radial velocity dealiasing (Brewster et al., 2005). The processed radar data is further used to identify and statistically analyze the MVs. Finally, the ERA5 reanalysis data (Hersbach and Dee, 2016) with a high horizontal resolution of 0.25° × 0.25° from the European Centre for Medium-Range Weather Forecasts (ECMWF) is utilized to study the environmental conditions for MVs in Guangdong from April to June.

Identi cation of MVs
Since MVs are mainly generated in organized MCSs, it is necessary to identify the MCSs in advance. Firstly, the composite re ectivity from the nine Doppler radars ( Figure 1) is used to identify convective zones (CZs) of >35 dBZ. Then, the area of each CZ is calculated, and the CZ with an area greater than 1000 km 2 is considered as a MCS. More details about the detection of MCSs can be found in T20.
In the past, the identi cation of meso--scale vortices (e.g., mesocyclones) or even smaller-scale circulations (e.g., tornadoes) mainly depends on the appearance of positive-negative velocity pair in the Doppler radar radial velocity eld . The rotational strength of a vortex is measured by the difference between the maximum inbound and outbound radial velocities which is, however, susceptible to the velocity noises. When the environmental wind speed is strong, the positive-negative velocity cannot be identi ed because it is easily submerged in the background wind eld. Moreover, when the vortex is far from the radar station, the azimuthal resolution becomes too low to well resolve the velocity pair. In this paper, a more advanced linear least square derivative method (Smith and Elmore, 2004) is used to calculate the azimuthal shear of the MV. This method synthetically considers the contribution of all radial velocities in a given range to the azimuthal shear at the center point, thus reducing the noise errors caused by the radial velocities.
In this study, we only calculate the azimuthal shear at the 0.5° elevation within 150 km of the radar because the MVs are mainly generated in the lower troposphere below 3 km. As in T20, a lower limit of 10 −3 s −1 is adopted for the azimuthal shear of MVs, which is one order greater than the Coriolis parameter in mid latitudes. The duration of the MV must exceed 18 minutes, i.e., three volume scans of the radar. On this basis, a storm cell identi cation technology similar to WSR-88D is adopted for the forward and backward tracking of MVs to obtain their life cycles (Johnson et al., 1998).
Because the radar radial velocity cannot be composed, the identi cation of MVs can only be performed for each individual radar. Despite the dense distribution of Doppler radars in Guangdong Province, only four radars covering almost the whole province are selected to identify the MVs. These four radars include Shaoguan (9751), Shantou (9754), Yangjiang (9662) and Guangzhou (9200) radar stations, representing the northern, eastern, western Guangdong Province and the Pearl River Delta region, respectively.

Spatio-temporal distribution
During the rst rainy season (April-June) in South China from 2017 to 2019, a total of 7,965 MVs are identi ed using the four radars in Guangdong Province (Table 1). Figure (Figure 2c), which together account for less than half of the total. In particular, the MVs are sparsely distributed near the Nanling mountains in northern Guangdong. The uneven spatial distribution of the MVs can be, at least, attributed to the following two reasons. The rst is due to the different topographies features over Guangdong. Laing and Fritsch (1997) have pointed out that MVs are prone to form over at regions, such as the America Great Plains. The formation of MVs is therefore affected by the topography of the Nanling mountains in northern Guangdong and the Lianhua, Luofu and Jiulian Mountains in eastern Guangdong. In contrast, the relatively at terrain in western Guangdong and Pearl River Delta region is conducive to the formation of the MVs. Secondly, the uneven spatial distribution of MVs is intimately related to the differences of environmental conditions in different regions of Guangdong, which will be discussed in section 3.3. In terms of lifetime, most of the MVs (about 67%) last less than 30 minutes, about 30% last more than half an hour, while the ones lasting more than one hour only account for about 5% ( Table 1). The statistical characteristics of MVs' lifetime in Guangdong are similar to those in the YHRB.      (Table 3). As noted above, the spatial distribution of the MVs during the rst rainy season in South China is affected by their environmental conditions. During this period, the South Asia high center is located in the upper troposphere (200 hPa) over the Indochina Peninsula and adjacent areas (Figure 6a). South China is on the southern side of the upper-level jet axis at its entrance region. In the middle troposphere (500 hPa), the subtropical high lies between 10°N and 20°N, and South China lies on the northern side of the subtropical high and in the westerlies at the bottom of the East Asia trough (Figure 6b). In the lower troposphere (850 hPa), with the onset of the southwest monsoon, the equatorial westerlies from the Indian Ocean and the cross-equatorial ow from near Kalimantan Island entering the South China Sea converge and blow over South China (Figs. 6c and 6d). Meanwhile, the low-level southwesterly jets transport abundant warm and moist air to South China, especially to the west of the Pearl River Estuary (the Pearl River Delta region and western Guangdong), resulting in the higher potential pseudo-equivalent temperature and water vapor ux in this region than in northern and eastern Guangdong.
Due to the warm and moist air ow transported by the southwest monsoon, the mean low-level instability during April-June is relatively higher across the whole Guangdong Province (Figure 7a). The potential pseudo-equivalent temperature in southwest Guangdong (30°C-32°C) is signi cantly higher than that in other regions (28°C-30°C). In addition, the existence of the low-level jet causes stronger low-level wind shear over southwest Guangdong than in other regions (Figure 7b). In short, the relatively higher instability and stronger low-level vertical wind shear are conducive to the formation of MVs, which is similar to the environmental conditions for the formation of MVs in the YHRB. Weisman and Trapp

Conclusions And Discussion
In this paper, the mesovortices (MVs) occurred during the rst rainy season (April to June) in South China during 2017-2019 are analyzed using Doppler radar observations as well as ERA5 reanalysis. The spatio-temporal distributions and structural features of MVs are examined, as well as their favorable environmental conditions. The main conclusions are as follows.
The MVs are mainly observed in the Pearl River Delta region, followed by western Guangdong, and relatively fewer in both eastern and northern Guangdong. Similar to the Yangtze-Huaihe River Basin (YHRB), the MVs in South China present very short lifetime, with about 70% lasting less than 30 minutes. The intensity and horizontal scale of the MVs are proportional to their lifetime, i.e., long-lived MVs have larger horizontal scales and stronger intensities than short-lived ones. Different from the MVs in the YHRB, the MVs in South China display more signi cant diurnal variations, which occur most frequently from late morning toward evening (1100 BJT to 1700 BJT) but drop to the minimum overnight (2000 BJT to 0200 BJT on the next day). In contrast, the MVs in the YRHB present slight diurnal variations, with two weak peaks at night (1800 BJT to 2100 BJT) and in the morning (0600 BJC to 0900 BJT) as well as a minimum in the early morning (0300 BJT to 0600 BJT).
The diurnal variations are quite similar for the MVs in the Pearl River Delta region, western and eastern Guangdong, with an afternoon-to-evening peak and a night-to-early-morning minimum. On the contrary, the diurnal variation of MVs in northern Guangdong is opposite to that mentioned above. This is because during the rst rainy season in South China, under the combined in uence of dry and cold air from the north and the warm and moist air from the sea, there is usually an MCS accompanied by a cold front, which thereby affects South China from north to south. The MCS tends to be triggered in northern Guangdong from night to morning, and then moves to the southern coastal areas from afternoon to evening. As a result, MVs generated within MCSs have a similar diurnal variations to their parent systems.
The uneven spatial distribution of MVs is closely related to the environmental differences in South China. Affected by the southwest monsoon, the water vapor ux, low-level instability and vertical wind shear in southwest Guangdong are signi cantly greater than those in other regions during the rst rainy season, which are favorable to the formation of MVs.
This study contributes to the understanding of MVs in South China and thus has a great signi cance for the operational nowcasting and warning of MVs. However, the relationship between the MVs and severe weather (such as gales, heavy rainfall) in South China is still unclear, which will be studied in the future.         territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.