1. Introduction
The development of a precipitation system is mainly supported by water vapor from its environment. A portion of the water vapor returns to the ground as precipitation (e.g., rain) through various microphysical processes (e.g., evaporation, condensation), while the remainder humidifies the atmosphere. These processes are fundamental components of the atmospheric water cycle. Precipitation efficiency (hereafter PE), defined as the ratio of the total rainfall (Rtot, kg) to the total water vapor (Qtot, kg) ingested into the cloud from the environment (Sui et al. 2005, 2007), is an important parameter resulting from these processes to characterize precipitation systems. In addition, the PE determined by control parameters (e.g., synoptic conditions, microphysical processes) plays an important role in the global water cycle by distributing the water in various forms throughout the entire atmosphere. The variation of control parameters in altering PE and consequent precipitation patterns under climate change (e.g., drought due to reduced PE) is also a key issue for numerous socioeconomic activities (Tegart et al. 1990). A better understanding of the control parameters of PE improves numerical weather prediction and reduces the damage from extreme hazards (Doswell et al. 1996).
The kinematic and microphysical structures of precipitation systems are mainly determined by the environmental conditions in different climatological regions. The precipitation systems subsequently evolve diversely with different PEs, which can vary between 10% and 110% (Market et al. 2003). Various factors influencing the PE have been studied via numerical simulations (e.g., Weisman and Klemp 1982; Ferrier et al. 1996; Li et al. 2002; Tao et al. 2004; Sui et al. 2005, 2007). Ferrier et al. (1996) indicate that the vertical tilting of the updraft, which is controlled by vertical wind shear, is important to determine the PE, while the moisture content is responsible for 10% of the PE variation. Murata (2009) found that the enhancement of PE in a heavy rainfall event was attributed to the conversion of cloud water to rainwater via accretion of cloud water by rain.
Observational studies investigated the PE (e.g., Marwitz 1972; Fankhauser 1988; Market et al. 2003; Shusse and Tsuboki 2006; Anip and Market 2007) by using the kinematic fields derived from Doppler radars and moisture fields obtained from rawinsondes. A summary of PE studies is shown in Table 1. Marwitz (1972) indicates that the PE decreases with increasing vertical wind shear. Market et al. (2003) showed that the PE is controlled by several environmental factors, namely, vertical wind shear, subcloud-base moisture, and convective inhibition. Fankhauser (1988) and Shusse and Tsuboki (2006) showed a positive correlation between PE and the size of the precipitation system.
Summary of the system type, region, and notes on previous PE studies.
Although the PE has been investigated widely from various aspects (i.e., different environmental factors), the relation between microphysical processes and PE has only been described through numerical simulations, not in past observational studies, primarily owing to the difficulty of acquiring the microphysical information. It is necessary to examine the effects of microphysical characteristics on PE in various types of precipitation systems in order to aid in the interpretation of the numerically simulated results. Moreover, most studies of PE focus on precipitation systems in the tropics (Gamache and Houze 1983; Chong and Hauser 1989; Rauber et al. 1996; Simpson et al. 1988) and higher latitudes during the warm season (Fankhauser 1988; Market et al. 2003; Shusse and Tsuboki 2006). The PE of the precipitation systems in the subtropical regions has received little attention.
During the North American Monsoon Experiment (NAME; Higgins et al. 2006), Nesbitt et al. (2008) analyzed subtropical precipitation systems in northwestern Mexico during the onset of the North American monsoon. They hypothesize that the PE of warm rain processes may increase at high elevations due to the low cloud base associated with reduced evaporation compared to low-elevation regions in the Sierra Madre Occidental (SMO). Rowe et al. (2008) concluded that the large vertical distance between the cloud base and the freezing level (4.5 km) along the coastal plain during NAME suggested a greater role for warm rain processes compared to higher terrain. Although the PE of subtropical systems during NAME was investigated, it was only in a qualitative sense.
The Southwest Monsoon Experiment/Terrain-Influenced Monsoon Rainfall Experiment (SoWMEX/TiMREX), a joint Taiwan–U.S. field project containing extensive observing facilities, was conducted from 15 May to 30 June 2008 in the northern South China Sea (SCS) and southern Taiwan (Jou et al. 2011). The project aimed at understanding the mechanisms of heavy rainfall precipitation systems during the onset of the southwest monsoon (SWM) as it interacted with the mei-yu front and the complex topography (Central Mountain Range) of Taiwan (Fig. 1). Among various observing facilities, the National Center for Atmospheric Research (NCAR) S-band dual-polarization Doppler radar (S-Pol) provided kinematic measurements and rainfall estimation. S-Pol also significantly reduces the rainfall estimation error from attenuation [e.g., X-band radars in Shusse and Tsuboki (2006)] and hail contamination (Fankhauser 1988). Most importantly, S-Pol polarimetric measurements further help interpret the microphysical characteristics of precipitation systems that were not available for previous observational PE studies. The SoWMEX/TiMREX dataset is ideal to simultaneously study PE and the associated microphysical characteristics of subtropical precipitation systems.
The configuration of the sounding stations (blue crisscrosses): Ma-Kung (46734), Ping-Dong (46750), and South Ship (99810, dots with color-coded UTC time, day/hour). Triangles show the locations of the S-Pol, RCCG, and RCKT radars. Green and red circles represent the dual-synthesis lobes of S-Pol–RCCG and S-Pol–RCKT. Solid and dashed boxes represent north (S-Pol–RCCG) and south (S-Pol–RCKT) dual-Doppler analysis areas. The distances of the baseline for S-Pol–RCCG and S-Pol–RCKT are 77.60 and 81.42 km, respectively.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
A prefrontal squall line (PFSL) on 14 June and a southwesterly monsoon mesoscale convective system (SWMCS) on 16 June during SoWMEX/TiMREX are investigated in this study. These two systems (meso α, 200–2000 km) were under distinct synoptic conditions (Wang et al. 2014a,b) but both systems produced more than 200 mm of accumulated rainfall within 24 h (Fig. 2). The diverse mesoscale characteristics of convection initiation and maintenance of these two precipitation systems have been examined in Davis and Lee (2012) and Xu et al. (2012). Yet, the role of microphysical processes in the precipitation mechanisms and the PEs of these two events have not been examined. Investigating these two heavy precipitation systems helps us to understand the relation of environmental and microphysical characteristics to the PE of subtropical heavy precipitation systems in Taiwan.
The official daily accumulation rain map (mm) from 0000 to 2400 LST (LST = UTC + 8 h) of 14–16 Jun 2008 from the Central Weather Bureau of Taiwan are shown in shaded colors. The rainfall estimations from RCWF, RCHL, RCCG, and RCKT (Taiwan S-band network) are merged and adjusted by rain gauge to generate official daily accumulation rain maps.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
The major goals of this paper are to examine the microphysical characteristics and quantify PEs of these two subtropical heavy precipitation systems. Moreover, the dependence of PEs on their environmental and microphysical factors using observational data will be investigated for the first time. The authors first examine the kinematic and microphysical characteristics of these systems under distinct synoptic conditions and, then, investigate the corresponding PEs and their relations to kinematic and microphysical characteristics. The methodology used in this research is presented in section 2. The synoptic conditions, environmental instability, and the spatiotemporal structures of these precipitation systems are reviewed in section 3. Section 4 portrays the mesoscale characteristics and kinematic fields from dual-Doppler analyses. The microphysical characteristics are presented in section 5. The PE is investigated and related to synoptic, thermodynamic, kinematic, and microphysical characteristics in section 6. Finally, a summary is provided in section 7.
2. Methodology
a. Kinematic field retrieval from three-dimensional wind field synthesis
The three-dimensional wind fields were derived from the variational method from Liou and Chang (2009) every 7.5 min across two regions (Fig. 1). The variational method derives the vertical motion via dynamical constrains and numerical smoothing terms including the anelastic continuity equation, vertical vorticity equation, background wind, and spatial smoothness terms. The advantage of this algorithm is that it negates the need to artificially prescribe the top or bottom boundary conditions for the vertical velocities in the traditional sense and estimates more accurate vertical velocities (Mewes and Shapiro 2002), which is essential for PE calculation. The horizontal and vertical resolutions of retrieved wind fields are 1.0 and 0.5 km, respectively, and the top boundary is 15.0 km to cover the tops of the precipitating systems. The available volumes of dual-Doppler analysis are summarized in Table 2.
The total number of available dual-Doppler analysis volumes for the PFSL and SWMCS. The total hours are listed in parentheses. The spatial structure characteristics, vertical wind shear, lightning activity, and instability (S) of the precipitation systems are summarized correspondingly.
b. Microphysical retrieval from S-Pol
1) Rain and ice water content estimation
2) Rainfall-rate estimation
c. Classification of precipitation type
To investigate the characteristics of the PE of convective and stratiform systems, a robust algorithm (Steiner et al. 1995) separating convective and stratiform region was applied. The precipitation types were identified from low-level radar reflectivity (i.e., 2-km height) by examining the spatial uniformity and intensity. The visual examination of the results shows the algorithm separates the convective and stratiform region fairly well.
d. Precipitation efficiency
The calculation of PE requires information about the kinematic, moisture, and rainfall-rate fields, as shown in (6)–(8). These fields were determined by using a sophisticated wind retrieval technique and available observational data (sounding network and S-Pol). The uncertainties of the calculated PE from choosing different reference bottom boundaries (1.0–2.5 km) are investigated. The results suggest that about one-twentieth of the uncertainty could be introduced into the calculation of PE due to the choice of the bottom boundary conditions and can therefore be assumed to be negligible in this study.
3. Environmental conditions and evolution of the precipitation systems
The environmental conditions and evolution of the precipitation systems have been documented in previous research (Davis and Lee 2012; Xu et al. 2012; Wang et al. 2014a,b). A brief summary is provided in this section.
a. Synoptic conditions
Between 14 and 17 June 2008, Taiwan was in a prefrontal environment, while the mei-yu front extended east-northeast (southeast of Japan) to west-southwest (southeast of the China coast) and remained to the north of Taiwan (Wang et al. 2014a,b). Cyclonic wind shear was located over eastern China (Figs. 3a–c) associated with the mei-yu front on 14 June and weakened on 15 June.
The mixing ratio of water vapor (g kg−1), wind fields (m s−1), and geopotential height (m) at 800, 700, and 500 hPa from the National Centers for Environmental Prediction’s (NCEP) Final (FNL) operational global analysis data at 0000 UTC 14–17 Jun are shown as shaded colors, vectors, and contours, respectively. The reference wind vectors (10 m s−1) are shown at lower-right corner.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
As shown in Figs. 3a,b; 3d,e; and 3g,h, the southwesterly low-level jet (LLJ) from 800 to 700 hPa flowed around the region near the north SCS and southwest Taiwan from 14 to 16 June (Wang et al. 2014a,b). The time–height cross section of the wind fields from Ping-Dong station (Fig. 4) shows the wind intensified around 0000 UTC 14 June from 850 to 600 hPa as the PFSL passed by. The low-level (900–600 hPa) winds were mainly southwesterly with westerly winds above 400 hPa during the PFSL passage. The high wind speed (>10 m s−1) that accompanied the PFSL passage was observed throughout most of the troposphere (900–100 hPa). The PFSL fully passed southern Taiwan at 2230 UTC, the wind speed then decreased, and the wind direction had a more southerly component.
Time–height cross section of wind fields (the long and short wind barbs represent wind speeds of 10 and 2 m s−1, respectively), isotachs (the solid and dashed lines represent the 10 and 15 m s−1 wind speeds, respectively), and relative humidity (% by shaded colors) from the Ping-Dong sounding station from 13 to 17 Jun.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
During SWMCS, the southwesterly wind was restored at 2100 UTC 15 June (Fig. 4) as the MCS associated with SWM developed near southern Taiwan from the north SCS. The low-level warm, moist, and unstable air was found over the north SCS on 16 June (Xu et al. 2012; Wang et al. 2014a). The SWMCS had a uniform wind speed profile. The wind direction turned southerly at 1200 UTC 16 June (Fig. 4) as part of the circulation of the Pacific subtropical high located over eastern Taiwan (Figs. 3j–l). While the Pacific subtropical high gradually extended westward (5880-m contour at 500 hPa in Figs. 3c,f,i,l), the moist air over southern Taiwan shifted northward around 1200 UTC 16 June (not shown) and the air mass over southern Taiwan became relatively drier (Ruppert et al. 2013). The MCS subsequently decayed due to the relatively dry air (Chen et al. 2007).
The low-level moisture field suggests that southern Taiwan remained moist (mixing ratio of water vapor at 800 hPa >12 g m−3; Figs. 3a,d,g,j). The relative humidity remained high (>70%) from the low to upper levels (1000–300 hPa) during 14–17 June (Fig. 4). This high relative humidity provided a favorable environment for efficient rainfall production by preventing the loss of raindrops from evaporation (Rauber et al. 1996).
b. Environmental instability
The environmental instability S is defined as the vertical gradients of equivalent potential temperature (
Time series of the environmental instability S is defined by the vertical gradients of equivalent potential temperature [
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
c. Spatiotemporal evolution of the precipitation systems
The synoptic conditions of 14 and 16 June had pronounced differences, and consequently, the precipitation systems had dissimilar spatiotemporal evolutions. A northeast–southwest-oriented PFSL triggered by the mei-yu front developed over the southeast China coast on late 13 June. Satellite IR images show that the PFSL propagated southeasterly toward Taiwan on 14 June (Fig. 6a). The PFSL was subsequently altered by the complex terrain (Figs. 6b,c). The northern part of the PFSL became less organized and the southern part changed to an east–west orientation. The entire PFSL passed through Taiwan at 2230 UTC 14 June (Fig. 6d). After the PFSL passage, scattered precipitation from the northern SCS associated with the SWM propagated northeastward into southwest Taiwan (Figs. 6e,f). The system gradually weakened around 0900 UTC 15 June and was followed by a break with less intense convection (Figs. 6g,h).
The brightness temperature (K) from the Magyar Satellite-1 (MASAT-1) IR channel over East Asia every 6 h from (a) 0030 UTC 14 Jun to (l) 1830 UTC 16 Jun. The prefrontal squall line on 14 Jun is marked as a dashed line in (a). The MCS on 16 Jun is marked as a dashed circle in (i).
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
Another intense precipitation system approached the southwest coast of Taiwan around 1930 UTC 15 June as it moved from the north SCS toward the northeast (Fig. 6i). The SWMCS (Figs. 6i,j) developed offshore and remained quasi stationary over the upstream ocean and southwest coast of Taiwan (Xu et al. 2012). The precipitation system later transformed to scattered convective cells at 1200 UTC 16 June (Figs. 6k,l).
4. Kinematic characteristics
The aforementioned evolving synoptic conditions between 14 and 16 June resulted in distinct spatiotemporal structures in PFSL and SWMCS (summarized in Table 2). Considering the organizational/structural modes were different in PFSL and SWMCS, only the data from available dual-Doppler analysis periods were included for the investigation to ensure the consistency of kinematic, microphysical, and PE analyses. The horizontal winds (no vertical velocity) at 2.5 km were obtained by a conventional approach (Armijo 1969; Miller and Strauch 1974) from three radar (RCCG, RCKT, and S-Pol) syntheses (Fig. 7) to better illustrate the general flow patterns of these two events. The detailed kinematic structures of these two systems are portrayed using radar reflectivity and variational-derived system-relative wind analyses in the dual-Doppler analysis areas (black boxes in Fig. 7).
(a) The mosaic reflectivity at 1000 UTC 14 Jun from S-Pol. The vectors represent winds at 2.5 km derived from three radar syntheses (RCCG, RCKT, and S-Pol). The box represents the dual-Doppler analysis region of RCCG–S-Pol. The topography of 0.5 km is shown as a black contour line. (b) As in (a), but at 0507 UTC 16 Jun. The box represents the dual-Doppler analysis area of RCKT–S-Pol.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
a. Prefrontal squall line
The S-Pol mosaic reflectivity at 1000 UTC 14 June (Fig. 7a) illustrates that the PFSL was composed of several northeast–southwest-oriented rainbands associated with both convective and stratiform precipitation. The individual convective cells moved from the southwest to the northeast while the synoptic low-level flow was southwesterly (Fig. 7a). The detailed kinematic structures are illustrated by the reflectivity and system-relative wind fields as shown in Fig. 8.1 As the rainband moved toward the southeast, the low-level front-to-rear flow was observed along the eastern side of the rainband with northeasterly system-relative winds at 2 km (Fig. 8a). The low-level front-to-rear flow decelerated toward the leading edge of the rainband. A midlevel rear-to-front flow with westerly winds was found on the western side of the rain band at 5 km (Fig. 8b).
The interpolated reflectivity (color shaded) and retrieved system-relative winds (vectors) at (a) 2.0 and (b) 5.0 km from S-Pol and RCCG at 1000 UTC 14 Jun. The blue (downdraft) and red (updraft) solid lines represent the vertical velocity every 2 m s−1. The vertical cross section between A and B (dashed line) is shown in Fig. 9.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
The vertical cross sections of reflectivity and divergence overlaid by the system-relative wind field of the most vigorous cell (i.e., most intense vertical velocity) at y = −14 km (A–B in Fig. 8) are shown in Fig. 9. The convective precipitation represented by the intense reflectivity field at x = −60 km was accompanied by a relatively strong updraft (maximum retrieved updraft ~10.0 m s−1) located at ~7-km altitude in the convective core (Fig. 9a). The convergence field of the convective region can be found up to 6 km (Fig. 9b). The low-level front-to-rear flow transported moist and unstable pre-squall-line air into the convective cells (Fig. 9b). The updraft further transported the water vapor deeper into the upper levels of the troposphere. The midlevel rear-to-front flow with system-relative wind speed of about 4.0 m s−1 is observed at 6.0 km (Fig. 9b).
Vertical cross section of the (a) reflectivity (dBZ) and (b) retrieved divergence (s−1) fields at Y = −14.0 km in Fig. 8. The vertical (w) and system relative horizontal (u) velocities are shown as contours in (a) and (b), respectively.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
Unlike the tropical convection often associated with maximum low-level convergence and an updraft in a narrow ribbon within 1–2 km of the leading edge (LeMone et al. 1984, 1998; Jorgensen et al. 1997), the convective system in this case (at x = −60 km in Fig. 9a) was located about 15 km behind the leading edge (reflectivity = 0 dB, x = −45 km in Fig. 9a). This feature is similar to a squall line in the SCS described by Wang and Carey (2005). The trailing stratiform region (x = −90 to −75 km) with a pronounced bright band was located behind the convective region with a transition region (x = −75 to −65 km) between them. The transition region was characterized by weaker reflectivity without a bright band. The upper-level (>7.0 km) westerly flow ahead of the convective line transported particles downstream, forming a pre-squall-line stratiform region. As the PFSL formed over southeast China and propagated toward Taiwan, the kinematic structure of the evolving PFSL was similar to the midlatitude squall line summarized in Houze (1993).
b. Southwesterly monsoon mesoscale convective system
Convective characteristics in SWMCS at 0507 UTC 16 June (Fig. 7b) were distinctly different from those in the PFSL portrayed in Fig. 7a. The SWMCS was composed of scattered cells generated repeatedly offshore that moved onshore (Xu et al. 2012; Wang et al. 2014a,b). These cells were at the convergence zone of westerly winds at the northwest part of the SWMCS and southwesterly flow upstream. This westerly wind was part of a disturbance over the coastal area in southern China and can be seen up to 3 km from the Ma-Kung sounding (not shown). The MCS gradually weakened and dissipated at 0600 UTC 16 June while the westerly winds slowly withdrew toward the north and were replaced with predominant southerly winds from the Pacific subtropical high over southern Taiwan.
An intense southerly to southwesterly low-level rear-to-front flow was found in the SWMCS (Fig. 10a).2 The low-level rear-to-front flow provided continuous moist and unstable air from the north SCS to support the quasi-stationary MCS. The north–south-oriented convective line was located at x = −50 km. The convergence of the wind field in the convective cell can be clearly seen from low to midlevels (2–5 km; Fig. 10b). From the vertical cross sections of the most intense cell at y = −60 km (C–D in Fig. 10), a convective cell with strong reflectivity and maximum updraft speeds of 8.0 m s−1 was located at x = −58 km (Fig. 11a). The depth of the low-level rear-to-front flow (with u-component wind speed of about 2.0–6.0 m s−1) was about 4.0 km (Fig. 11b), much deeper than the low-level front-to-rear inflow in PFSL. This convective cell was less intense and less elevated compared to the PFSL in terms of reflectivity. The corresponding convergence field was also less elevated (up to 5 km) compared to PFSL (Fig. 11b). Uniform easterly winds above 6 km carried the particles upstream and formed stratiform precipitation.
As in Fig. 8, but from S-Pol and RCKT at 0507 UTC 16 Jun. The vertical cross section between C–D (dashed line) is shown in Fig. 11.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
As in Fig. 9, but for Y = −60.0 km at 0507 UTC 16 Jun.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
c. Composite vertical structure of kinematic fields
The kinematic fields of PFSL and SWMCS have been characterized by the analysis of representative dual-Doppler analysis and show distinct features. To examine the general features of the kinematic fields statistically, the contoured frequency by altitude diagram (CFAD; Yuter and Houze 1995) was applied. The CFADs of horizontal velocity, upward motion, and divergence fields of convection including all available dual-Doppler analysis periods are shown in Fig. 12. Because of the higher westerly wind speed between 900 and 400 hPa during the PFSL period (Fig. 4), the PFSL (Fig. 12a) had stronger u-component winds than did the SWMCS (Fig. 12b). Below 2 km, the SWMCS had a stronger υ than u component of horizontal velocity consistent with sounding data (Fig. 4). The mean u component of horizontal velocity (Fig. 12c) shows that the PFSL had pronounced vertical wind shear, while the SWMCS exhibited more uniform profiles. The vertical wind shears from the mean u component of the horizontal velocity between 0.5 and 12 km were 0.684 and −0.258 (m s−1 km−1) for PFSL and SWMCS periods, respectively. In the meantime, the υ components show more uniform vertical profiles.
(a),(b) The CFAD of the u component of the horizontal wind (U, m s−1) of convective regions for PFSL and SWMCS periods. The CFAD contours are shaded at 0.01%, 0.1%, 1%, 5%, and 10%. The corresponding mean values of the vertical profiles of the U (solid line) and V (dashed line) components of the horizontal wind of the convective regions are shown in (c). (d),(e) As in (a),(b), but for the updraft (W, m s−1). (f) The mean values of the updraft (m s−1). (g),(h) As in (a),(b), but for the divergence field (DIV, s−1). (i) The mean values of DIV.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
As the pronounced vertical wind shear supported deep convection by separating the updraft and downdraft (Emanuel 1994), as well as a higher environmental instability (Fig. 6), the maximum up- and downdrafts during the PFSL (Fig. 12d, 0.01% contour line) were consequently stronger than those of the SWMCS (Fig. 12e). The divergence field of the PFSL (Fig. 12g) shows larger variances than the SWMCS (Fig. 12h), consistent with stronger up- and downdrafts. Shallower mean convergence fields of SWMCS (Fig. 12i) are consistent with weaker updrafts. Contrary to the SWMCS, the PFSL had a deeper mean convergence (up to 6 km); therefore, the altitude of maximum mean updraft of the PFSL was higher than the SWMCS.
5. Microphysical characteristics
a. Polarimetric measurements
The corresponding S-Pol polarimetric measurements of Figs. 9 and 11 associated with the most intensive convective cells were investigated to illustrate the microphysical characteristics of PFSL and SWMCS. The convection within the PFSL was characterized by strong ZHH (~50 dB) associated with a column of high values of differential reflectivity (ZDR > 1.5 dB) and specific differential phase shift (KDP > 2.2° km−1) at about x = −60 km in Figs. 13a,b. Note that ZDR provides information about the oblateness of particles and KDP depends on liquid water content (Bringi and Chandrasekar 2001). Therefore, the positive ZDR and KDP column suggests a rain mass existed up to 7-km altitude, well above the freezing level (about 4.7 km according to rawinsonde data). Relatively high values of ZDR (1.5–1.75 dB) and linear depolarization ratio (LDR, −22 to −24 dB; Fig. 13c) associated with lower values of zero-lag cross-correlation coefficient (ρHV, 0.96–0.97; Fig. 13d) at the freezing level corresponded to the presence of a rain–ice mixture (e.g., melting aggregates). Both LDR and 
Vertical cross section at 1000 UTC 14 Jun of S-Pol polarimetric measurements of (a) differential reflectivity (ZDR), (b) specific differential phase (KDP), (c) LDR, and (d) correlation coefficient (ρHV). The system-relative wind and vertical velocity are shown as black vectors and contours, respectively. The 10 m s−1 reference vectors are shown in the top-right corner. The blue lines are contours of 30-, 40-, and 50-dBZ reflectivity. The black dashed line represents the 0°C level (4.7 km) according to sounding data. (e)–(h) As in (a)–(d), but at 0507 UTC 16 Jun and the 0°C level is 5.0 km.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
The distinct microphysical characteristics between the PFSL and SWMCS convective regions were further illustrated by PIDs in Fig. 14. Within the PFSL, the graupel–rain category was located above the melting level and graupel–small hail mixtures were evident at higher altitudes (Fig. 14a). These microphysical features were similar to the Amazon squall line that formed in low-level easterly flow (Cifelli et al. 2002) and deep convection that initiated over the SMO during NAME (Rowe et al. 2011). During the SWMCS, most of the mixed phases near the melting level were wet snow (Fig. 14b). Only limited graupel and rain–graupel mixtures were found in SWMCS. The SWMCS was similar to the convection during the onset of the SCS monsoon from SCSMEX (Wang and Carey 2005), MCSs that formed in the Amazon within the low-level westerly flow regime (Cifelli et al. 2002) and the organized convections over the Gulf of California and had longer residence over water during NAME (Rowe et al. 2012). The warm rain processes play a more significant role than the mixed-phased processes. In the stratiform regions of both the PFSL and SWMCS, the PIDs identified wet snow in the bright band due to water-coated snow or coexisting rain and snow.
(a) Vertical cross section at 1000 UTC 14 Jun of the PID results from NCAR’s fuzzy-logic method (Vivekanandan et al. 1999). The particle types for each abbreviation are as follows: CL, cloud; DZ, drizzle; LR, light rain; MR, moderate rain; HR, heavy rain; HA, hail; RH, rain–hail mixture; GSH, graupel–small hail mixture; GR, graupel–rain mixture; DS, dry snow; WS, wet snow; IC, ice crystals; IIC, irregular ice crystals; SLD, supercooled liquid droplets; BGS, biological signals; ST, second trip; and GC, ground clutter. The system-relative wind and vertical velocity are shown as black vectors and contours, respectively. The 10 m s−1 reference vectors are shown in the middle right of the figure. The blue lines are contours of 30-, 40-, and 50-dBZ reflectivity. The black dashed line represents the 0°C level (4.7 km) according to the sounding data. (b) As in (a), but at 0507 UTC 16 Jun. The 0°C level is 5.0 km.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
b. Composite vertical structure of polarimetric measurements
As the convective (stratiform) regions of PFSL and SWMCS show distinct (similar) features, only the microphysical characteristics of the convective regions are described using CFADs of ZHH, ZDR, and KDP. The PFSL with more intense, deeper convection had 1% of 30-dBZ ZHH up to 8 km (Fig. 15a). The less intense, shallower convection during the SWMCS periods had 1% of 30-dB ZHH extending through 7 km (Fig. 15b). The mean ZHH profile (Fig. 15c) indicates higher values in PFSL than SWMCS throughout the low to upper levels. Above the freezing level (h ≈ 4.7–5.0 km), the majority of ZDR values were in the range between 0.0 and 1.0 dB for PFSL and SWMCS (Figs. 15d,e). The downward increase of ZDR due to the melting of wet-snow/graupel was pronounced near the freezing level in both systems. Below the freezing level (h < 4.0 km), the 1% frequency of ZDR was toward higher values in PFSL, suggesting larger raindrops. Also, the mean values of ZDR (Fig. 15f) below the freezing level show different slopes (dB km−1) between 0.5 and 3.5 km. They were about −0.032 and 0.002 for PFSL and SWMCS, respectively.
(a),(b) The CFAD of reflectivity (ZHH, dBZ) of convective regions for PFSL and SWMCS periods. The CFAD contours are shaded at 0.01%, 0.1%, 1%, 5%, and 10%. (c) The corresponding mean values of the reflectivity (dB) of the convective region. (d),(e) As in (a),(b), but for differential reflectivity (ZDR, dB). (f) The mean values of the differential reflectivity of the convective region. The corresponding values of the mass-weighted diameter (mm) are shown in parentheses. (g),(h) As in (a),(b), but for the specific differential phase (KDP, ° km−1). (i) The mean values of the specific differential phases (KDP, ° km−1).
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
The KDP of PFSL and SWMCS (Figs. 15g,h) remained low (0.0°–0.5° km−1) above the freezing level, indicating limited rainwater content. The 1% frequency of 1.0° km−1 KDP can be found at 3.0 km during PFSL, but only at 1.8 km during SWMCS, suggesting that the stronger updraft in PFSL could carry liquid water to a deeper level. It is noticed that the SWMCS had higher values of mean ZDR and KDP above 8 km (Figs. 15f,i). It is postulated that the lower (higher) values of mean ZDR and KDP from PFSL (SWMCS) were due to more irregular ice crystals (oriented ice crystals) according to PIDs (not shown).
The mean KDP (Fig. 15i) between 3 km and the freezing level was greater for PFSL than SWMCS, suggesting more liquid water present in the midlevels for PFSL. It is noticed that the higher slope of mean KDP (between 0.5 and 3.5 km) in SWMCS (−0.053° km−1) than in PFSL (−0.036° km−1) suggests that the rainwater content also increased downward at a greater rate in the SWMCS. Hence, the SWMCS had higher mean KDP below 2 km, indicating more rainwater content than PFSL. The low lifting condensation levels (LCLs) during 14–16 June (about 100 m) and the high freezing level (4.7–5.0 km), resulting in deep warm cloud depths of about 4.6–4.9 km, allows for accretional processes to produce rainwater at low levels efficiently, which would explain the increasing KDP values toward cloud base. Similar results are also found in Rowe et al. (2012).
c. Vertical structural of water content
The vertical profiles of mean liquid water content of rain (LWC) and ice water contents (IWC) of graupel, snow, and ice in convective regions of PFSL and SWMCS, estimated from S-Pol, are summarized in Fig. 16. With stronger updrafts, the graupel IWC of PFSL was a factor of 3 higher than SWMCS (Fig. 16a). The snow IWC in the PFSL above 7 km was slightly higher than in the SWMCS (Fig. 16b) as well. On the other hand, the SWMCS had more wet snow near the freezing level (Fig. 16b) and heavy aggregation of snow above than did the PFSL. In terms of LWC, greater LWC in the SWMCS below 2 km (Fig. 16c) was consistent with observations of higher ZHH and KDP with lower ZDR.
Mean values of water content (g m−3) of (a) graupel and (b) snow. The water content of graupel includes the ice portion of the graupel–rain and the graupel–small hail categories via the ZDP–Zrain relation. (c) The mean values of total IWC from the sums of graupel, snow, and ice, as well as LWC.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
d. Lightning data analysis3
Greater IWC of graupel and snow with stronger updrafts in convective precipitation systems of the PFSL is found to be associated with higher lightning frequency than SWMCS, as shown in Fig. 17 (summarized in Table 2). Deierling et al. (2008) indicated that particle collision of graupel and ice crystals within the updraft associated with gravitational forces led to separating charges and an increase in the occurrence of lightning flashes. Contrarily, less graupel and weaker updrafts during the SWMCS corresponded to low lightning frequency.
The lightning frequency [h−1 (10 km)−2] from the north and south dual-Doppler synthesis areas (Fig. 1). The size of the dual-Doppler synthesis area is 15 488 km2.
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
6. Precipitation efficiency
The analyses of synoptic conditions and kinematic and microphysical features from the PFSL and the SWMCS have shown diverse characteristics. The corresponding Rtot, Qtot, and PEs were computed in this section to understand the dependence of PE on environmental and microphysical factors. As shown in Fig. 18, the Rtot and Qtot of convective precipitation regions could vary by a factor of 3 every 7.5 min. The instantaneous PEs (i.e., the ratios between Rtot and Qtot) of PFSL and SWMCS range from 30% to 80%, and the variations are presumably due mainly to including different life stages of precipitation systems from radar observations in analysis domains (Doswell et al. 1996). Market et al. (2003) suggest that Rtot, Qtot, and PE are most meaningful when averaged over the lifetime of a system. Subsequently, the PE, mean total rainfall flux (
The total rainfall flux (
Citation: Monthly Weather Review 143, 1; 10.1175/MWR-D-14-00081.1
As summarized in Table 3, the convective region in SWMCS had higher 
The mean total rainfall flux (
In terms of the 
In terms of the PE in convective precipitation, SWMCS (53.1%) was higher than PFSL (45.0%). Despite lower 
The stratiform regions of PFSL and SWMCS had consistently low values of PE (9.0% and 20.9%). The low PE is presumably due to evaporation in the stratiform region. Overall, the value of PE for the convective regions (ALL) of 47.9% was similar to the results of previous studies in tropical regions (e.g., 45%–59% in Table 1). However, the overall PE (ALL) including both convective and stratiform precipitation was 33.0% and higher than the study in eastern China [less than 10% in Shusse and Tsuboki (2006), Table 1]. A strong positive correlation between the PE and the size of the precipitation systems was found in Fankhauser (1988) and Shusse and Tsuboki (2006). The dilution rate of the air within the clouds by entrainment is generally inversely proportional to the size of the clouds. The moist environment (relative humidity >70%) reduced the effect of the entrainment of dry air, and the larger size (104 km2) of the precipitation systems with lower dilution rates act to increase the PE of the precipitation systems from 14 to 16 June compared to Shusse and Tsuboki (2006) with a drier environment (relative humidity ~70%) and a smaller-sized (102 km2) precipitation system.
7. Summary
Two diverse subtropical heavy precipitation systems 1) prefrontal squall line (PFSL) and 2) southwesterly monsoon mesoscale convective system (SWMCS) under distinct synoptic conditions from 14 to 16 June during SoWMEX/TiMREX in 2008 over southern Taiwan were investigated. The PFSL was triggered by the mei-yu front over east China in an unstable environment [S ≈ 6.5 K (100 hPa)−1], propagated southeasterly toward Taiwan on 14 June, and featured a northeast–southwest-oriented rainband. Conversely, the quasi-stationary SWMCS developed over the coastal area of southern Taiwan on 16 June and was maintained by a less unstable SWM [S ≈ 3–4 K (100 hPa)−1] from the north SCS.
For the first time, the PE of subtropical precipitation systems is investigated by observational data according to these two events. The PE of the convective precipitation region (47.9%) from 14 to 16 June was similar to the findings of past studies of convective precipitation in tropical regions (e.g., 45%–59%). However, the PE of these entire precipitation systems (33%) was much higher than the cumulonimbus clouds (10%) in eastern China (Shusse and Tsuboki 2006). It is postulated that the reduced effect of the entrainment of dry air due to the relatively larger size of the precipitation systems and moister environment during 14–16 June increased the PE.
Furthermore, the dependences of PE of subtropical precipitation systems on environmental and microphysical factors were investigated simultaneously via observational data by examining these two events in detail for the first time. Despite the SWMCS having lower 
The diverse PEs of PFSL and SWMCS were associated with distinct synoptic, kinematic, and microphysical characteristics. The PFSL had deeper convection facilitated by pronounced vertical wind shear separating the up- and downdrafts. Subsequently, the PFSL associated with strong updrafts and frequent lightning activity featured more frequent graupel–rain mixing above the freezing level and graupel–small hail extending throughout the deep level. Instead, the SWMCS had shallower convection associated with a uniform wind profile. The SWMCS had weaker updrafts and contained more snow and less graupel water content with reduced lightning frequency.
It is postulated that the higher wind shear of the PFSL reduces the PE by increasing the contact with drier environmental air and aiding evaporation of potential precipitation (Marwitz 1972). Moreover, the deeper convection with evaporative loss of rain that was transported upstream and that formed larger stratiform regions also reduced the PE of PFSL (Ferrier et al. 1996). In contrast, the SWMCS exhibited a higher downward increasing rate of liquid water content compared to PFSL. The deep warm cloud depth provided favorable conditions for accretional growth of rain. It is postulated that the SWMCS had higher PE as the higher liquid water content at low levels enhanced the accretional growth of rain.
During SWMCS, the long duration of rainfall due to the quasi-stationary MCSs is the main reason for the high accumulations of precipitation over the coastal area of southern Taiwan (Xu et al. 2012). Nevertheless, from the studies of synoptic, kinematic, and microphysical characteristics, the higher PE of SWMCS due to the aforementioned environmental factors and favorable microphysical processes also played critical roles in producing heavy rainfall. Despite the convective precipitation of the SWMCS having lower mean total water vapor flux (
This study provides the preliminary investigation of the subtropical PEs (i.e., those over southern Taiwan) during SoWMEX/TiMREX from observational data. Extending the PE study to various precipitation systems (e.g., mei-yu events, typhoons) is necessary as future work. Moreover, further detailed study of PE using numerical simulation with sophisticated microphysical parameterization schemes is required. By comparing the PE from numerical weather simulations and observational studies of various precipitation systems, quantitative precipitation forecasts (QPFs) can be improved by better describing the evolution of the precipitation processes.
Acknowledgments
We thank the SoWMEX/TiMREX science team and office sponsored by the National Science Foundation and the Taiwan Central Weather Bureau and National Science Council for providing the observational and analysis data used in this paper. This work benefited from the internal reviews by Dr. Angela K. Rowe (University of Washington) and Dr. Scott Ellis (NCAR). The authors also thank three anonymous reviewers for constructive comments.
REFERENCES
Anip, M. H. M., and P. S. Market, 2007: Dominant factors influencing precipitation efficiency in a continental mid-latitude location. Tellus, 59A, 122–126, doi:10.1111/j.1600-0870.2006.00208.x.
Armijo, L., 1969: A theory for the determination of wind and precipitation velocities with Doppler radars. J. Atmos. Sci., 26, 570–575, doi:10.1175/1520-0469(1969)026<0570:ATFTDO>2.0.CO;2.
Bringi, V. N., and V. Chandrasekar, 2001: Polarimetric Doppler Weather Radar—Principles and Applications. Cambridge University Press, 636 pp.
Carey, L. D., and S. A. Rutledge, 2000: The relationship between precipitation and lightning in tropical island convection: A C-band polarimetric radar study. Mon. Wea. Rev., 128, 2687–2710, doi:10.1175/1520-0493(2000)128<2687:TRBPAL>2.0.CO;2.
Chen, C.-S., Y.-L. Chen, C.-L. Liu, P.-L. Lin, and W.-C. Chen, 2007: Statistics of heavy rainfall occurrences in Taiwan. Wea. Forecasting, 22, 981–1002, doi:10.1175/WAF1033.1.
Chong, M., and D. Hauser, 1989: A tropical squall line observed during the COPT 81 experiment in West Africa. Part II: Water budget. Mon. Wea. Rev., 117, 728–744, doi:10.1175/1520-0493(1989)117<0728:ATSLOD>2.0.CO;2.
Ciesielski, P. E., and Coauthors, 2010: Quality-controlled upper-air sounding dataset for TiMREX/SoWMEX: Development and corrections. J. Atmos. Oceanic Technol., 27, 1802–1821, doi:10.1175/2010JTECHA1481.1.
Cifelli, R., W. A. Petersen, L. D. Carey, S. A. Rutledge, and M. A. F. da Silva Dias, 2002: Radar observations of the kinematic, microphysical, and precipitation characteristics of two MCSs in TRMM LBA. J. Geophys. Res., 107, 8077, doi:10.1029/2000JD000264.
Davis, C. A., and W.-C. Lee, 2012: Mesoscale analysis of heavy rainfall episodes from SoWMEX/TiMREX. J. Atmos. Sci., 69, 521–537, doi:10.1175/JAS-D-11-0120.1.
Deierling, W., W. A. Petersen, J. Latham, S. Ellis, and H. J. Christian, 2008: The relationship between lightning activity and ice fluxes in thunderstorms. J. Geophys. Res., 113, D15210, doi:10.1029/2007JD009700.
Doswell, C. A., III, H. E. Brooks, and R. A. Maddox, 1996: Flash flood forecasting: An ingredients-based methodology. Wea. Forecasting, 11, 560–581, doi:10.1175/1520-0434(1996)011<0560:FFFAIB>2.0.CO;2.
Emanuel, K. A., 1994: Atmospheric Convection. Oxford University Press, 580 pp.
Fankhauser, J. C., 1988: Estimates of thunderstorm precipitation efficiency from field measurements in CCOPE. Mon. Wea. Rev., 116, 663–684, doi:10.1175/1520-0493(1988)116<0663:EOTPEF>2.0.CO;2.
Ferrier, B. S., J. Simpson, and W.-K. Tao, 1996: Factors responsible for precipitation efficiencies in midlatitude and tropical squall simulations. Mon. Wea. Rev., 124, 2100–2125, doi:10.1175/1520-0493(1996)124<2100:FRFPEI>2.0.CO;2.
Gamache, J. F., and R. A. Houze Jr., 1983: Water budget of a mesoscale convective system in the Tropics. J. Atmos. Sci., 40, 1835–1850, doi:10.1175/1520-0469(1983)040<1835:WBOAMC>2.0.CO;2.
Golestani, Y., V. Chandrasekar, and V. N. Bringi, 1989: Intercomparison of multiparameter radar measurements. Preprints, 24th Conf. on Radar Meteorology, Tallahassee, FL, Amer. Meteor. Soc., 309–314.
Higgins, R. W., and Coauthors, 2006: The NAME 2004 field campaign and modeling strategy. Bull. Amer. Meteor. Soc., 87, 79–94, doi:10.1175/BAMS-87-1-79.
Houze, R. A., Jr., 1993: Cloud Dynamics. Academic Press, 573 pp.
Jorgensen, D. P., M. A. LeMone, and S. B. Trier, 1997: Structure and evolution of the 22 February 1993 TOGA COARE squall line: Aircraft observations of precipitation, circulation, and surface energy fluxes. J. Atmos. Sci., 54, 1961–1985, doi:10.1175/1520-0469(1997)054<1961:SAEOTF>2.0.CO;2.
Jou, B. J.-D., W.-C. Lee, and R. H. Johnson, 2011: An overview of SoWMEX/TiMREX and its operation. The Global Monsoon System: Research and Forecast, 2nd ed., C.-P. Chang, Ed., World Scientific, 303–318.
Khairoutdinov, M., and Y. Kogan, 2000: A new cloud physics parameterization in a large-eddy simulation model of marine stratocumulus. Mon. Wea. Rev., 128, 229–243, doi:10.1175/1520-0493(2000)128<0229:ANCPPI>2.0.CO;2.
LeMone, M. A., G. M. Barnes, E. J. Szoke, and E. J. Zipser, 1984: The tilt of the leading edge of mesoscale tropical convective lines. Mon. Wea. Rev., 112, 510–519, doi:10.1175/1520-0493(1984)112<0510:TTOTLE>2.0.CO;2.
LeMone, M. A., E. J. Zipser, and S. B. Trier, 1998: The role of environmental shear and thermodynamic conditions in determining the structure and evolution of mesoscale convective systems during TOGA COARE. J. Atmos. Sci., 55, 3493–3518, doi:10.1175/1520-0469(1998)055<3493:TROESA>2.0.CO;2.
Li, X., C.-H. Sui, and K.-M. Lau, 2002: Precipitation efficiency in the tropical deep convective regime: A 2-D cloud resolving modeling study. J. Meteor. Soc. Japan, 80, 205–212, doi:10.2151/jmsj.80.205.
Liou, Y.-C., and Y.-J. Chang, 2009: A variational multiple–Doppler radar three-dimensional wind synthesis method and its impacts on thermodynamic retrieval. Mon. Wea. Rev., 137, 3992–4010, doi:10.1175/2009MWR2980.1.
Market, P., S. Allen, R. Scofield, R. Kuligowski, and A. Gruber, 2003: Precipitation efficiency of warm-season midwestern mesoscale convective systems. Wea. Forecasting, 18, 1273–1285, doi:10.1175/1520-0434(2003)018<1273:PEOWMM>2.0.CO;2.
Marwitz, J. D., 1972: Precipitation efficiency of thunderstorms on the High Plains. J. Res. Atmos, 6, 367–370.
Mewes, J. J., and A. Shapiro, 2002: Use of the vorticity equation in dual-Doppler analysis of the vertical velocity field. J. Atmos. Oceanic Technol., 19, 543–567, doi:10.1175/1520-0426(2002)019<0543:UOTVEI>2.0.CO;2.
Miller, L. J., and R. G. Strauch, 1974: A dual Doppler radar method for the determination of wind velocities within precipitating weather systems. Remote Sens. Environ., 3, 219–235, doi:10.1016/0034-4257(74)90044-3.
Murata, A., 2009: A mechanism for heavy precipitation over the Kii Peninsula accompanying Typhoon Meari (2004). J. Meteor. Soc. Japan, 87, 101–117, doi:10.2151/jmsj.87.101.
Nesbitt, S. W., D. J. Gochis, and T. J. Lang, 2008: The diurnal cycle of clouds and precipitation along the Sierra Madre Occidental observed during NAME-2004: Implications for warm season precipitation estimation in complex terrain. J. Hydrometeor., 9, 728–743, doi:10.1175/2008JHM939.1.
Rauber, R. M., N. F. Laird, and H. T. Ochs, 1996: Precipitation efficiency of trade wind clouds over the north central tropical Pacific Ocean. J. Geophys. Res., 101, 26 247–26 253, doi:10.1029/96JD01383.
Rowe, A. K., S. A. Rutledge, T. J. Lang, P. E. Ciesielski, and S. M. Saleeby, 2008: Elevation-dependent trends in precipitation observed during NAME. Mon. Wea. Rev., 136, 4962–4979, doi:10.1175/2008MWR2397.1.
Rowe, A. K., S. A. Rutledge, and T. J. Lang, 2011: Investigation of microphysical processes occurring in isolated convection during NAME. Mon. Wea. Rev., 139, 424–443, doi:10.1175/2010MWR3494.1.
Rowe, A. K., S. A. Rutledge, and T. J. Lang, 2012: Investigation of microphysical processes occurring in organized convection during NAME. Mon. Wea. Rev., 140, 2168–2187, doi:10.1175/MWR-D-11-00124.1.
Ruppert, J. H., R. H. Johnson, and A. K. Rowe, 2013: Diurnal circulations and rainfall in Taiwan during SoWMEX/TiMREX (2008). Mon. Wea. Rev., 141, 3851–3872, doi:10.1175/MWR-D-12-00301.1.
Shusse, Y., and K. Tsuboki, 2006: Dimension characteristics and precipitation efficiency of cumulonimbus clouds in the region far south from the mei-yu front over the eastern Asian continent. Mon. Wea. Rev., 134, 1942–1953, doi:10.1175/MWR3159.1.
Simpson, J., R. F. Adler, and G. R. North, 1988: A proposed Tropical Rainfall Measuring Mission (TRMM) satellite. Bull. Amer. Meteor. Soc., 69, 278–295, doi:10.1175/1520-0477(1988)069<0278:APTRMM>2.0.CO;2.
Steiner, M., R. A. Houze Jr., and S. E. Yuter, 1995: Climatological characteristics of three-dimensional storm structure from operational radar and rain gauge data. J. Appl. Meteor., 34, 1978–2007, doi:10.1175/1520-0450(1995)034<1978:CCOTDS>2.0.CO;2.
Sui, C.-H., X. Li, M.-J. Yang, and H.-L. Huang, 2005: Estimation of oceanic precipitation efficiency in cloud models. J. Atmos. Sci., 62, 4358–4370, doi:10.1175/JAS3587.1.
Sui, C.-H., X. Li, and M.-J. Yang, 2007: On the definition of precipitation efficiency. J. Atmos. Sci., 64, 4506–4513, doi:10.1175/2007JAS2332.1.
Tao, W. K., D. Johnson, C.-L. Shie, and J. Simpson, 2004: The atmospheric energy budget and large-scale precipitation efficiency of convective systems during TOGA COARE, GATE, SCSMEX, and ARM: Cloud-resolving model simulations. J. Atmos. Sci., 61, 2405–2423, doi:10.1175/1520-0469(2004)061<2405:TAEBAL>2.0.CO;2.
Tegart, W. J. McG., G. W. Sheldon, and D. C. Griffiths, Eds., 1990: Climate Change: The IPCC Impacts Assessment. Australian Government Publishing Service, 294 pp.
Tokay, A., and Coauthors, 2001: On the role of drop size distribution in TRMM rain profiling algorithm. Preprints, 30th Int. Conf. on Radar Meteorology, Munich, Germany, Amer. Meteor. Soc., 6.6. [Available online at https://ams.confex.com/ams/pdfpapers/21466.pdf.]
Vivekanandan, J., S. M. Ellis, R. Oye, D. S. Zrnic, A. V. Ryzhkov, and J. Straka, 1999: Cloud microphysics retrieval using S-band dual-polarization radar measurements. Bull. Amer. Meteor. Soc., 80, 381–388, doi:10.1175/1520-0477(1999)080<0381:CMRUSB>2.0.CO;2.
Vivekanandan, J., G. Zhang, and E. Brandes, 2004: Polarimetric radar estimators based on a constrained gamma drop size distribution model. J. Appl. Meteor., 43, 217–230, doi:10.1175/1520-0450(2004)043<0217:PREBOA>2.0.CO;2.
Wang, C., J. Hsu, G. Chen, and D. Lee, 2014a: A study of two propagating heavy-rainfall episodes near Taiwan during SoWMEX/TiMREX IOP-8 in June 2008. Part I: Synoptic evolution, episode propagation, and model control simulation. Mon. Wea. Rev., 142, 2619–2643, doi:10.1175/MWR-D-13-00331.1.
Wang, C., J. Hsu, G. Chen, and D. Lee, 2014b: A study of two propagating heavy-rainfall episodes near Taiwan during SoWMEX/TiMREX IOP-8 in June 2008. Part II: Sensitivity tests on the roles of synoptic conditions and topographic effects. Mon. Wea. Rev., 142, 2644–2664, doi:10.1175/MWR-D-13-00330.1.
Wang, J.-J., 2004: Evolution and structure of the mesoscale convection and its environment: A case study during the early onset of the southeast Asian summer monsoon. Mon. Wea. Rev., 132, 1104–1120, doi:10.1175/1520-0493(2004)132<1104:EASOTM>2.0.CO;2.
Wang, J.-J., and L. D. Carey, 2005: The development and structure of an oceanic squall-line system during the South China Sea Monsoon Experiment. Mon. Wea. Rev., 133, 1544–1561, doi:10.1175/MWR2933.1.
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, doi:10.1175/1520-0493(1982)110<0504:TDONSC>2.0.CO;2.
Xu, W., E. J. Zipser, Y. L. Chen, Ch. Liu, Y. C. Liou, W. C. Lee, and B. J. D. Jou, 2012: An orography-associated extreme rainfall event during TiMREX: Initiation, storm evolution, and maintenance. Mon. Wea. Rev., 140, 2555–2574, doi:10.1175/MWR-D-11-00208.1.
Yuter, S. E., and R. A. Houze Jr., 1995: Three-dimensional kinematic and microphysical evolution of Florida cumulonimbus. Part III: Vertical mass transport, mass divergence, and synthesis. Mon. Wea. Rev., 123, 1964–1983, doi:10.1175/1520-0493(1995)123<1964:TDKAME>2.0.CO;2.
Zhang, G., J. Vivekanandan, and E. Brandes, 2001: A method for estimating rain rate and drop size distribution from polarimetric radar measurements. IEEE Trans. Geosci. Remote Sens., 39, 830–841, doi:10.1109/36.917906.
Zhang, P., D. Zrnić, and A. Ryzhkov, 2013: Partial beam blockage correction using polarimetric radar measurements. J. Atmos. Oceanic Technol., 30, 861–872, doi:10.1175/JTECH-D-12-00075.1.
The system-relative wind fields were obtained by subtracting the system motions that were determined by examining two consecutive scans of convective cells. The motions of the PFSL were about 10.3 and 12.0 m s−1 in the x and y directions, respectively. The system motion derived in this research is considered to be the motion of convective cells.
The system motions of the SWMCS were about 6.6 and 8.8 m s−1 for the x and y directions, respectively.
The lightning data were collected by the Taiwan Power Company. The data include the in-cloud (IC) and cloud-to-ground (CG) lightning.