Next Article in Journal
UAV LiDAR Based Approach for the Detection and Interpretation of Archaeological Micro Topography under Canopy—The Rediscovery of Perticara (Basilicata, Italy)
Previous Article in Journal
Failure Mechanism Analysis of Mining-Induced Landslide Based on Geophysical Investigation and Numerical Modelling Using Distinct Element Method
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 14
Issue 23
10.3390/rs14236072
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Regional Variability in Microphysical Characteristics of Precipitation Features with Lightning across China: Observations from GPM

by
Fengjiao Chen
1,2,3,*,
Mingjian Zeng
1,
Lu Yu
1,
Xiaoyong Zhuge
1 and
Hao Huang
4
1
Key Laboratory of Transportation Meteorology of China Meteorological Administration, Nanjing Joint Institute for Atmospheric Sciences, Nanjing 210041, China
2
Anhui Province Key Laboratory of Atmospheric Science and Satellite Remote Sensing, Institute of Meteorological Sciences, Hefei 230031, China
3
Shouxian National Climatology Observatory, Huaihe River Basin Typical Farm Eco-Meteorological Experiment Field of CMA, Shouxian 232200, China
4
Key Laboratory of Mesoscale Severe Weather/MOE and School of Atmospheric Sciences, Nanjing University, Nanjing 210023, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2022, 14(23), 6072; https://doi.org/10.3390/rs14236072
Submission received: 17 October 2022 / Revised: 23 November 2022 / Accepted: 26 November 2022 / Published: 30 November 2022
Browse Figures
Review Reports Versions Notes

Abstract

:
The statistical characteristics of precipitation microphysics in lightning clouds are not yet fully understood, as a result of the limitations of traditional observational methods. Using the latest observations from the dual-frequency radar and microwave imager onboard the Global Precipitation Mission (GPM) and ground-based lightning observations, the precipitation microphysics of precipitation features with and without lightning (LPFs and NLPFs) was investigated across four typical regions of China in summer during the time period of 2014–2021. The statistical results show that the LPFs are characterized by smaller concentration and larger mass-weighted mean diameter (Dm) for rain and ice hydrometeors than those of NLPFs. Below the melting layer, the radar reflectivity (Ze) for both the LPFs and NLPFs generally decreases toward the surface, indicating the evaporation or strong break-up of rain hydrometeors. Above the melting layer, the Ze values mainly increase as the altitudes decrease for both LPFs and NLPFs, indicating the rimming, aggregation, or deposition processes. However, the change in slope is much smaller for the LPFs than for the NLPFs, which suggests a more uniform distribution of large ice hydrometeors at high altitudes, probably as a result of the stronger updrafts within the LPFs. The microphysical structures of the LPFs show great regional differences among the four regions of China, which is characterized by the low concentration of large-sized rain hydrometeors over Northeast China, and a high concentration of small-sized rain hydrometeors near the surface over the Yangtze-Huaihe River basin.

Graphical Abstract

1. Introduction

Thunderstorms, also known as lightning storms, are generally accompanied by lightning, strong winds, and heavy rain or hail, and can lead to forest fires and human casualties. Cloud electrification mainly involves both micro- and macro-scale charge separation processes between graupel/hail and ice crystals by persistent vigorous updrafts at high levels [1,2]. Therefore, understanding, estimating, and predicting the occurrence and strength of lightning storms and precipitation requires knowledge of the microphysical characteristics of precipitation [3,4,5].
Numerous studies have been implemented in the past to study precipitation microphysics for thunderstorms using ground-based radar observations, although most of them focused on limited areas due to the effective detection range of radars. Cui et al. [6] found that the precipitation cloud columns with lightning over Guangdong province, China are characterized by higher cloud tops and greater maximum vertically integrated liquid path, compared to those without lightning. In terms of different lightning densities, the main differences in precipitation structures were observed in the region above the melting layer, associated with high concentrations of ice particles at high altitudes, and the presence of supercooled liquid drops above the freezing levels for the most intense lightning cases when seen from the dual-polarimetric variables [7,8]. Focusing on the convective cores, Huang, et al. [9] revealed that the liquid water content generally decreases with decreasing altitude below 4 km in altitude for the extreme rainfall convective features with intense lightning flash rates over south China.
An important method to infer the precipitation microphysics of lighting-producing clouds over the tropics and subtropics is the use of observations from Precipitation Radar (PR) onboard the Tropical Rainfall Measuring Mission (TRMM) satellite, which provides the three-dimensional radar reflectivity. Using the joint observations from Lightning Imaging Sensor (LIS), microwave imager, and PR onboard TRMM, researchers suggested that the liquid and ice condensate mixing ratio, the ice and graupel flux, and the interactions between hydrometeors in the mixed-phase regions have been considered to have important roles in determining the rates and magnitude of lightning flashes [10,11,12]. Regional variabilities show that storms with the same polarized brightness temperature (PCT) at 85 GHz, or the same reflectivity aloft, are less likely to produce lightning over the ocean than over the land, indicating different ice microphysical processes in thunderstorms over different regions [13,14,15,16]. Although the observations from PR provide a clear map for the precipitation structures for thunderstorms over the tropics and subtropics, the precipitation microphysical properties and the precipitation characteristics at midlatitudes can not be obtained due to the difficulties in the retrievals of precipitation microphysical parameters from single frequency Ku-band radar and the low orbital inclination of TRMM (35° from the equator).
The launch of the Global Precipitation Mission (GPM) in 2014 initiated the near-global-coverage (extending from 65°S to 65°N) of observations of precipitation microphysical structures [17,18]. The three-dimensional structures of precipitation and microwave signals can be measured by the GPM Microwave Imager (GMI) and the Dual-frequency Precipitation Radar (DPR). The DPR consists of a Ka and Ku band radar, which makes the retrieval of the drop-size distribution (DSD) parameters of precipitation hydrometeors possible [19]. Recent studies have validated the accuracy of these DPR retrievals [20,21,22]. The microphysical properties of both tropical and mid-latitude cyclones have been revealed using these datasets [23,24,25].
This research has two basic purposes: (1) to improve our understanding of the microphysical characteristics of lightning storm systems by considering the rain and ice hydrometeors; and (2) to understand the regional differences in precipitation microphysics between precipitating clouds with lightning and without lightning. Understanding these characteristics will give insights into the prediction of severe weather events. To achieve this goal, the GPM DPR, GMI, and lightning observational data collected during the summers (June–August) in the time period 2014 to 2021 are used to determine the precipitation microphysics of lightning precipitation clouds over four different locations across China.

2. Material and Methods

2.1. Datasets

Three primary datasets are used: the GPM DPR official dataset (version 7), the GMI/1C dataset (version 7) [26], and the Advanced Time of arrival and Direction system (ADTD) lightning “flash” dataset. The GPM DPR dataset provides the three-dimensional radar reflectivity (Ze), phase (solid, mixed, and liquid) and DSD retrieval products, the near-surface rain rate (RR), and the rain type (convective, stratiform, and other) [19]. The phase of hydrometeors and rain types are retrieved using the information of changes in the profiles of the dual frequency ratio [26]. The DSD parameter Dm (units: mm) and the generalized intercept parameter (Nw: units mm−1 m−3) are provided in the official product; note that the liquid-equivalent DSDs are retrieved for mixed-phase and solid-phase hydrometeors using the dual-frequency algorithm and single-frequency algorithm. Due to the spatial and temporal variations of precipitation and relevant microphysical variables; the consistency between DPR retrievals and ground-based observations has been validated in several representative regions. For example, using the observations from polarimetric radars, Huang et al. [22] found that there is relatively good consistency for the reflectivity factor and microphysical variables between the two instruments in south China. However, there might be some underestimations in liquid water content, rainfall rate, and raindrop number concentration, especially for intense echoes [22]. The DPR version 7 algorithm can be referred to in Liao and Meneghini [27]. Further detailed information about the GPM spacecraft, instruments, and retrieval methods has been reported previously by numerous studies [18,19,24,25,26,28,29]. The 7 km height level was selected to study the properties of ice particles in the previous study [10], which may introduce errors due to the different climatic characteristics over different regions. In the present study, using the “phase” in the official DPR product, the solid, mixed, or liquid hydrometeors are quantitatively identified. Furthermore, to reduce the uncertainties caused by the classifications of the phases in the official GPM DPR dataset, we examined the characteristics of ice hydrometeors above 1 km in altitude from SH and rain hydrometeors below 3 km altitude from LH, respectively.
The presence of sufficient ice hydrometeors (e.g., graupel, or hail) inside precipitating clouds can lead to scattering of the upwelling microwave radiation at a frequency of 85 GHz, leading to a colder brightness temperature [30]. A few studies have shown a good correlation between the lightning flash rate and the brightness temperature at wavelengths of 85 and 37 GHz [31,32]. Therefore, the brightness temperature at 89 GHz (PCT89) provided in the GMI 1C dataset is used to denote the ice hydrometeors inside precipitating clouds. PCT89 is defined as 1.818 × T89v − 0.818 × T89h, where T89v and T89h are the vertically and horizontally polarized brightness temperature at 89 GHz, respectively [33].
The ADTD lightning location network, which provides information about the location, polarity, and current of lightning, was established by the China Meteorological Administration and is designed to detect intra-cloud and cloud-to-ground lightning flash densities. The detection efficiency is up to 90%, with the effective detection range reaching about 300 km.

2.2. Methods

The DPR, GMI, and lightning flash datasets are merged to investigate the precipitation microphysics of thunderstorms. The footprint size of the DPR is ~5 km at nadir, while 4.4 km along-scan direction and 7.3 km cross-scan direction for the footprint size of GMI at 89 GHz channel. As a result of the different footprint sizes, the PCT89 data from the GMI dataset are spatially interpolated to the DPR measurements within the DPR orbital swath using nearest-neighbor methods. The lightning flash counts are then spatially (temporally) collocated to each DPR pixel within a 5 km (5 min) radius of the DPR pixel. Rather than investigating precipitation microphysics of individual pixels, the contiguous rainy pixels (near-surface reflectivity > 20 dBZ) are grouped as a precipitation feature (PF) to characterize a precipitation event [15,34]. PFs with less than four rainy pixels are not included in this study. PFs with at least one lightning flash are defined as lightning PFs (short as LPFs) and PFs with no detected lightning flashes are classified as non-lightning PFs (NLPFs). By focusing on the PFs, the overall precipitation and microphysical characteristics of precipitating clouds with and without lightning can be revealed. As a result of the different scan modes of the Ku band precipitation radar—the KuPR (normal scan) and KaPR (matched, high-sensitivity scan) modes, only the central 25 beams of the KuPR mode in the inner swath of the normal scan have information about the reflectivity in both the KuPR and KaPR modes before May 2018 [26]. Since May 2018, the high-sensitivity beams by KaPR have been redirected to the outer swath to obtain the dual frequency information over the full scan. Therefore, the merged data in the inner swath of the normal scan before 2018 May and in the full swath after 2018 May are used to guarantee the reliability and consistency of the statistical analysis.
LPFs and NLPFs over four typical regions of China, including South China (SC, 18–24°N, 110–123°E), Yangtze-Huaihe River basin (YHRB, 24–35°N, 110–123°E), North China (NC, 35–42°N, 110–123°E), and Northeast China (NEC, 42–55°N, 110–135°E) [35,36] are studied by taking into account the availability of lightning observations and different climatic characteristics. SC impacted by the summer monsoon generally has sufficient rainfall and frequent lightning activities, whereas NEC generally has less precipitation and fewer lightning events [36,37]. During the summers between 2014 and 2021, a total of 2737 LPFs and 71,989 NLPFs were observed by GPM DPR over four study regions.
Figure 1 illustrates an example of LPFs and NLPFs over YHRB and the corresponding near-surface RR, lightning flash counts, rain type, storm-top height, DSDs, and PCT89 values. The storm-top height is calculated as the maximum height at which the radar reflectivity exceeded 20 dBZ [34,38], which may be an indicator of the updraft [39,40]. Generally, the stronger the updraft, the deeper the thunderstorm will be (the larger the storm-top height will be). There are six contiguous areas of precipitation over YHRB (Figure 1c), indicating six PFs as defined in this study. The two PFs producing lightning flashes (Figure 1b) over the southeastern part are classified as LPFs and the other four PFs with no lightning detected are defined as NLPFs (Figure 1c). The LPFs have much lower PCT89 and higher storm-top heights than the NLPFs (Figure 1e,f), probably indicating a stronger updraft and more ice hydrometeors within these clouds. More precipitation is classified as convection for the LPFs, with a more intense near-surface RR, larger Dm, and smaller Nw values at 2 km above the mean sea level (MSL).

3. Results

3.1. Precipitation and Microphysical Characteristics near the Surface

Table 1 summarizes the general statistical characteristics of the LPFs and NLPFs over four typical regions in the summer during the time period 2014–2021. The samples were calculated by summing all the DPR pixel samples inside the LPFs and NLPFs. It can be seen that the pixel samples are sufficient to guarantee relatively reliable statistical results. The samples within LPFs are much smaller than those within the NLPFs, corresponding to the less frequent LPFs. The convective (stratiform) frequency is computed using the ratio of convection (stratiform) precipitating samples to the total precipitating samples. The mean height when the hydrometeors change from solid or mixed phase to liquid phase (the height of liquid phase, shortened as LH) is also calculated from the variable “phase”. The LPFs in the four-sub regions of China are associated with a higher convective frequency, a more intense RR, a higher reflectivity, and a relatively lower LH than the NLPFs. The lower LH and higher storm-top heights for the LPFs (Table 2) suggest a deeper vertical extent of solid- and mixed-phase hydrometeors, consistent with the strong correlation between the number of lightning flashes and the mass of ice particles [11]. Stratiform precipitation accounts for a large portion of both the LPFs and NLPFs. The large fraction of stratiform precipitation within the LPFs is because stratiform precipitation is often attached to the main convective core. Based on the mean DSDs (labeled as texts in Figure 2) at 2 km above the MSL, the precipitation hydrometeors near the surface of LPFs have smaller mean Nw and larger Dm values than the NLPFs, indicating the existence of large hydrometeors near the surface, probably as a result of the melting of ice-phase hydrometeors with large sizes and high densities (hail or graupel) above the melting layer. Previous studies have found that when the ice-based growth is dominant, the DSDs skew towards a low concentration of large-sizes hydrometeors [41].
Regional variations can be seen for both the LPFs and NLPFs. For the NLPFs, the precipitation over SC has a higher mean RR (2.75 mm h−1), reflectivity (34.49 dBZ), LH (5.04 km), Dm (1.42 mm), convective frequency (19.99%), and a smaller stratiform frequency (54.4%) than the other three regions. The mean Dm is similar to that of the convective rain band of typhoon Matmo (Dm 1.41 mm) over the land [42], while much smaller than that for precipitation in the monsoon season (Dm 1.75 mm) over the South China Sea [43]. These results further confirm that there are great variations of precipitation and relevant microphysical parameters in different precipitation types, climatic regions, and seasons. About 26% of shallow convective precipitation occurred in SC, consistent with the previous study [44]. The NLPFs over NEC exhibit the smallest mean near-surface RR (2.09 mm h−1), reflectivity (32.65 dBZ), LH (3.52 km), Nw (35.52), convective frequency (12.41%), and the largest stratiform precipitation frequency (83.64%). This is consistent with the higher atmospheric instability and more sufficient water vapor over SC than those over NEC [45]. These differences also suggest that convective precipitation plays an important part in the formation of precipitation over SC and that larger hydrometeors exist near the surface than in the other three regions. The relatively higher (lower) LH values over SC (NEC) are probably determined by the higher (lower) ambient air temperature. Over YHRB, the rain hydrometeors near the surface are generally characterized by the highest concentration (Nw 37.42) of relatively small-sized hydrometeors (Dm 1.32 mm), in contrast to relatively lower concentration (Nw 36.74) of large-sized hydrometeors (Dm 1.42 mm) over SC. The differences of DSDs for NLPFs between over YHRB and over SC are statistically significant with a confidence level of 0.95 using a t-test method.
Convective cores play a critical role in the lightning production processes and there are clear increases in convective frequency for the LPFs in all four regions, with the greatest increase in the amplitude of 16.58% over NEC and the smallest increase of 11.3% over SC. Similar to NLPFs, the rain hydrometeors near the surface of LPFs over YHRB also show the smallest Dm (1.56 mm) and the highest Nw (36.62) among the four regions. By contrast, the LPFs over NEC are characterized by the highest Dm (1.66 mm) and the lowest Nw (34.29) among the four regions.
To illustrate the characteristics of the DSDs at low altitudes, the frequency patterns of the DSDs at 2 km above the MSL are investigated for the LPFs and NLPFs (Figure 2). All the DSDs of the LPFs and NLPFs show an overall pattern of large concentrations of small hydrometeors and small concentrations of large hydrometeors, which is in good agreement with the previous results revealed by ground-based radars or space-borne radars [24,25,42]. For the NLPFs, the DSDs over SC (Figure 2e) tend to have a broader distribution (5% contour line) than those in the other three regions, with a larger portion of large hydrometeors (Dm > 1.5 mm). The DSDs are shifted to smaller Dm and Nw values over NEC (Figure 2h, 5% contour line). Only a small fraction of the NLPFs over NEC has Nw values > 40 (5% contour line), suggesting a smaller concentration of hydrometeors. This is in good agreement with the weaker rain intensity during summer NEC [35]. When seen from the 50% contour lines, the Dm and Nw values over SC and YHRB are centered at about 0.8 mm and 40, respectively, indicating the prevalence of small rain hydrometeors. Note that the DSDs over YHRB have another maximum frequency center at larger Dm and smaller Nw values, with Nw values ranging between 32 and 38, and Dm values between 1.0 and 1.5 mm. This is also the maximum center for the DSDs over NC and NEC.
Over a given region, the LPFs (Figure 2a–d) tend to have more larger raindrops at 2 km above the MSL (Dm > 1.5 mm) than the NLPFs (Figure 2e–h). Previous studies have also found abundant larger drops for precipitation with lightning than those without lightning for a total of 48 events in Kolkata [46]. These large raindrops within LPFs probably result from the melting of ice-phase hydrometeors with large sizes and high densities (hail or graupel) above the melting layer. The DSDs of LPFs over SC and YHRB show a much wider range of Dm and Nw values (5% contour line) than over NC and NEC. Similar to the NLPFs over YHRB, the LPFs over this region also show the smallest mean Dm and the largest mean Nw values and are associated with larger variations in Nw values for a given Dm value. The DSDs of LPFs over NC and NEC are much narrower in their Nw values, whereas the corresponding Dm values are similar to those over SC. These results indicate the different microphysical characteristics near the surface within lightning precipitating clouds in different regions.
This analysis has shown the different regional characteristics of near-surface DSDs between LPFs and NLPFs. As suggested in many previous studies, changes in the polarimetric and space-borne radar variables (i.e., the differential reflectivity factor ZDR, Dm, and Ze) over a 2 km layer (1 and 3 km above the MSL) can serve as signatures to indicate the microphysical processes (collisional coalescence, break-up, or evaporation) [47,48]. To guarantee that the hydrometeors are rain rather than ice, the LH as the reference level are used instead of sea level. Similarly, the slope of the Ze profiles is calculated using the linear regression of the Ze profiles from 1 to 3 km below the LH [49]. A positive (negative) slope indicates a decrease (increase) in Ze values as the altitude decreases, possibly signifying the governing role of break-up (collision–coalescence). Figure 3 shows the two-dimensional frequency pattern diagrams of the slope and Ze at 2.5 km above the MSL for LPFs and NLPFs.
For the LPFs (Figure 3a–d), the peak frequency over the four regions mainly occurs when the slope is concentrated at positive values ranging from 0 to 2.5 dBZ km−1. This indicates a decrease in Ze when the rain hydrometeors fall to the lower altitudes, possibly due to the break-up of rain hydrometeors toward the surface. Some profiles also show negative slopes, especially when Ze at 2 km in altitude is <30 dBZ. The fractions of the negative slope are more frequent over relatively humid regions (SC and YHRB), suggesting an increase in Ze towards the surface. This increase may result from the collection of cloud droplets, the coalescence of rain hydrometeors or the presence of sufficient water vapor transported by the summer monsoon [50]. As the increases in the Ze at 2 km above the MSL, more precipitation profiles show a decrease in Ze towards the surface, indicating stronger break-up of rain hydrometeors, especially over NC and NEC (dry region).
Compared to LPFs, the Ze at 2 km above the MSL show a narrower range for NLPFs (Figure 3e–h), besides, the peak frequency is centered on smaller Ze values, indicating more intense precipitation within LPFs. The peak frequency of NLPFs over the four regions also occurs when the slopes are concentrated at positive values, mainly ranging from 0 to 2.5 dBZ km−1. Liu and Zipser [49] have also found that the radar reflectivity below the freezing level usually decreases toward the surface over land. Additionally, there are similar negative slopes of NLPFs when the Ze at 2.5 km above the MSL is low. While the fraction of negative slopes for the LPFs over a given region is more frequent than for the LPFs, besides, the amplitude of the negative slope is much larger than for LPFs. The stronger updrafts within the LPFs may facilitate these identical distributions. When the Ze at 2 km above the MSL exceeds 40 dBZ, the profiles mainly exhibit a positive slope towards the surface indicating the significant break-up of rain hydrometeors even for intense rain, which is consistent with the decreases in liquid water content for extreme rainfall and intense lightning convective features [9].

3.2. Precipitation and Microphysical Characteristics of Ice Hydrometeors

Previous studies have shown the importance of the collisions between graupel and smaller ice particles on the charging of thunderstorms [51,52]. The variations in the microphysical characteristics of ice hydrometeors are presented for the LPFs and NLPFs. The liquid-equivalent DSDs were retrieved for mixed- and solid-phase hydrometeors from the official DPR datasets [53]. Similarly, the height when the hydrometeors change from mixed or liquid phase to solid phase is calculated (the height of solid phase, shorten as SH). Specifically, the ice hydrometeors in the layer 1 km above the SH were studied to minimize the uncertainties of the retrieval algorithms. Figure 4 demonstrates the probability distribution functions (PDFs) of storm-top height, PCT89, DSDs, and Ze at 1 km above SH for LPFs and NLPFs during the study period. The mean values of these variables are summarized in Table 2 and labeled as texts in Figure 5.
Generally, the LPFs have larger storm-top heights, Ze, Dm values, smaller Nw and PCT89 values than NLPFs. When compared with the NLPFs, the PDFs of storm-top height (Figure 4a) for LPFs are centered at much larger values, with the mean storm-top height reaching 7.8 km over SC, 7.6 km over YHRB, 7.5 km over NC, and 6.9 km over NEC. There are more fractions of PCT89 < 260 K for LPFs than NLPFs (Figure 4b) and the mean PCT89 values are 250 K over SC, 251 K over YHRB, 246 K over NC, and 248 K over NEC, respectively. The ice hydrometeors for LPFs at 1 km above the SH have larger mean Dm values and smaller mean Nw values than those of the NLPFs over all four regions, indicating more active rimming processes. More fractions of large Dm values (>1.5 mm) and small Nw values (<30) can be seen for the LPFs (Figure 4d,e). All of these results indicate the much deeper vertical extent and the prevalence of large-sized ice hydrometeors within the LPFs.
The regional differences show that the ice hydrometeors for LPFs over NEC have the lowest concentrations (Nw 32.79) of the largest hydrometeor (Dm 1.44 mm), relatively larger Ze values (32.29 dBZ), and the lowest mean storm-top height (6.9 km) than those over other regions. The storm-top height generally denotes the vertical depth of precipitating clouds, and is positively correlated with the convective intensity. This seems to contradict the smaller mean PCT89 values (248 K for LPFs and 265 K for NLPFs) over NEC, with more fractions of PCT89 < 260 K. The PDFs of storm-top height for NLPFs show bimodal distribution over SC and YHRB (Figure 4a), with two frequency peaks at 3 and 6 km, respectively. The peak frequency of storm-top height at 3 km disappears for LFPs over SC and YHRB. This suggests that there is shallow precipitation for the NLPFs over SC and YHRB, which mainly come from warm rain showing relatively high PCT89 values [44]. While over NEC, precipitation mainly forms when ice hydrometeors dominate within clouds, scattering the upwelling radiation and reducing the brightness temperature.
Similar to the near-surface DSDs, the ice hydrometeors over YHRB are characterized by the largest concentration (Nw 33.87) of the smallest hydrometeor (Dm 1.31 mm), and the smallest mean Ze value (30.64 dBZ). Among the four regions, the most evident increase in Dm values of ice hydrometeors for LPFs (1.44 mm for LPFs and 1.19 mm for NLPFs, respectively) occurs over NEC. This difference is statistically significant with a confidence level of 0.95. The apparent differences in the sizes of ice hydrometeors between the LPFs and NLPFs over NEC indicate the existence of more active rimming or aggregation of ice hydrometeors for lightning production. This phenomenon is not that evident over SC, because there is no clear increase in the concentration or size of hydrometeors within LPFs.
The joint distribution of DSDs at 1 km above the SH is analyzed for the LPFs and NLPFs over the four regions (Figure 5). Similar to the near-surface DSD distributions (Figure 2), there are large concentrations of small hydrometeors and small concentrations of large hydrometeors for ice particles. However, these DSDs are shifted to smaller values for the LPFs and NLPFs, with the 50% contour line lying at Nw values between 28 and 36 and Dm values between 0.8 and 1.4 mm, compared to the near-surface DSDs. Similar to the relatively smaller near-surface Dm–Nw pairs over NEC, the DSDs of ice hydrometeors for the NLPFs are also concentrated at small Dm–Nw pairs over NEC, with few DSDs with large Dm (>1.8 mm) and Nw (>42) values (Figure 5h). The ice hydrometeors of the NLPFs over SC tend to have a broader distribution (5% contour line) of Dm and Nw values than the three other regions (Figure 4e). Compared with the NLPFs, there are more ice hydrometeors of LPFs with larger Dm values (seen from the 50% contour line) over YHRB, NC, and NEC, resulting in much larger mean Dm values. While, there is only a slight increase in Dm values for LPFs over SC, which is different from the apparent increase in Dm values below the LH (Figure 2a,e).
Figure 6 shows the two-dimensional frequency pattern of the slope and Ze values at 2 km above the MSL. The slope indicates the regression coefficient of the Ze profiles from 3 to 1 km above the SH. A negative (positive) slope above the melting layer indicated an increase (decrease) in Ze values as the altitude decreases. The larger the amplitude of the slope, the larger the increase or decrease in the Ze values.
For both the LPFs and NLPFs, the peak frequency over the four regions mainly occurred when the slope become concentrated at negative values, ranging from −2.5 to 0 dBZ km−1, which indicates that the Ze values mainly increased as the altitudes decrease above the melting layer, indicating the rimming, aggregation, or deposition processes. While the amplitude of the negative slope for the LPFs is centered on smaller values than the NLPFs. Large reflectivity (Table 2) and a small lapse late in reflectivity above the SH thus indicate vigorous convection, because such reflectivity features occur when large hydrometeors are suspended in the solid phase layer by a strong updraft within the LPFs. Furthermore, the slope shows stronger dependency on the Ze values at 2 km above the MSL for NLPFs than LPFs. For example, when Ze is smaller than 30 dBZ, the positive slope accounts for a great portion, suggesting a decrease in Ze values towards the melting layer. As the Ze increases, the negative slope becomes more dominant and the amplitude of the negative slope increases, showing an increase in Ze values toward the melting layer. This suggests that for producing relatively intense rain near the surface, the rimming, aggregation, and deposition of ice hydrometeors should be more active. Among the four regions, the dependency on Ze for NLPFs is the most evident over NEC and least evident over SC. This may indicate the presence of different ice microphysics over SC and NEC. Over SC, where convection is generally strong, the ice hydrometeors are lofted to higher levels, and therefore there is no sharp increase in the Ze values, compared with the other regions. By contrast, over NEC, where the convection is less intense, only a sharp increase in Ze values (probably rimming, aggregation, or deposition) could lead to relatively large Ze values at 2 km above the MSL. For LPFs, this dependency becomes much smaller especially over NEC and NC even for the profiles with intense Ze values at 2 km above the MSL, suggesting deeper convections within LPFs.

4. Discussion

Vertical Structures

Observations of the vertical characteristics of precipitation microphysics are essential for understanding cloud development and electrification. Contoured frequency by altitude diagrams (CFADs) have been widely used to display the binned reflectivity values [54,55], which can provide insights into the vertical dynamical and microphysical properties of LPFs. The CFADs of Ze and Dm over the four typical regions were calculated on the basis of all the observations of LPFs and NLPFs during the summer in the time period of 2014–2021 and are presented in Figure 7 and Figure 8, respectively. The CFADs were normalized by the maximum samples into 1 dBZ bins and every 0.25 km of altitude.
There are some common differences between the LPFs and NLPFs over the four typical regions. The maximum frequency center of Ze and Dm for the LPFs generally occurs above the melting layer (see reference line), whereas this occurs below the melting layer for the NLPFs. All of the LPFs over the four regions have deeper radar echo-tops (see the 10% contour line) than the NLPFs, indicating stronger convective intensity within the LPFs. There is a much smaller amplitude of increase in Ze and Dm values above the melting layer for the LPFs than for the NLPFs, which is consistent with the results in Figure 6, and there are much more hydrometeors with larger Ze and Dm values for the LPFs throughout the column. The intense convection within the LPFs also results in a more uniform distribution of relatively large Dm and Ze values above the melting layer. The stronger updraft within the LPFs extends the water vapor to higher altitudes, which facilitates the aggregation or rimming of ice hydrometeors above the melting layer, forming large-sized ice hydrometeors. These processes promote bigger drops to precipitate within LPFs, therefore, relatively large drops are observed below the melting layer within LPFs. By contrast, the water vapor within the NLPFs is mainly concentrated at lower levels due to relatively weaker updrafts, resulting in the maximum frequency occurring below the melting layer.
These regional differences show that below the melting layer, the Ze and Dm values for the LPFs and NLPFs over NEC are concentrated at smaller values and decrease sharply toward the surface. This suggests the evaporation or break up of hydrometeors below the melting layer (Figure 7d,h and Figure 8d,h). By contrast, the Ze and Dm values over SC have a wider range of values below the melting layer, showing a large frequency of Ze > 30 dBZ and Dm > 1.8 mm (Figure 7a,e and Figure 8a,e). There are large numbers of NLPF hydrometeors with Dm < 1.0 mm in the layer between 1 and 3 km altitude over SC and YHRB (Figure 8e,f), whereas these small hydrometeors have a relatively low frequency for LPFs (Figure 8a,b). As suggested in Figure 2, these small hydrometeors could result from shallow precipitation.
The LPFs and NLPFs over SC have larger Ze and Dm values above the melting layer than the other three regions. For example, Ze and Dm could reach 30 dBZ and 1.8 mm at 9 km altitude, respectively. In addition, the LPFs and NLPFs over SC (NEC) have the deepest (shallowest) reflectivity echo-top of the three other regions, indicating stronger (weaker) development in the vertical direction. The NLPFs over NC show the clearest maximum frequency center of radar echo above the melting layer (Figure 7h and Figure 8h), which suggests a larger amount of ice hydrometeors in precipitating clouds over NC.

5. Conclusions

Based on seven years of joint observations from the DPR and GMI instruments onboard the GPM and ground-based lightning observations, the precipitation microphysics of LPFs and NLPFs were investigated over four typical regions of China (SC, YHRB, NC, and NEC) during the summers between 2014 and 2021. Our results can be summarized as follows.
In general, the LPFs are embedded with more convective precipitation and are associated with a more intense RR, a larger reflectivity near the surface, lower LH, and higher storm-top height than the NLPFs. The deeper vertical extent of solid- and mixed-phase hydrometeors provides the favorable condition for micro-scale charge separation processes. The precipitation hydrometeors near the surface of the LPFs have smaller Nw and larger Dm values than the NLPFs. The mean Dm (Nw) values at 2 km altitude above the MSL for the LPFs over SC, YHRB, NC, and NEC are 1.61, 1.56, 1.62, and 1.66 mm (36.57, 36.62, 35.76, and 34.16), respectively. The large raindrops near the surface within the LPFs are in good agreement with relatively large-sized ice hydrometeors at higher altitudes. The regional differences show that the LPFs over NEC have the largest mean Dm (1.66 mm) and the lowest mean Nw (34.29) near the surface among the four regions, also with the largest increases in convective frequency compared to NLPFs (16.58%), probably suggesting stronger updraft within LPFs. The LPFs over YHRB are characterized by the highest concentrations (Nw 36.62) of the smallest hydrometeors (Dm 1.56 mm) near the surface among the four regions. Below the melting layer, the Ze for LPFs and NLPFs generally decreases toward the surface over all four regions, with a positive slope ranging from 0 to 2.5 dBZ km−1, suggesting the dominance of the break up of rain hydrometeors towards the surface. There are also negligible profiles showing increases in Ze towards the surface, especially for NLPFs over SC and YHRB, which indicates the existence of the rain-drop coalescence processes. The relatively warm and humid environmental conditions (sufficient water vapor transported by the summer monsoon) over SC and YHRB, may favor the collection of cloud droplets and the coalescence of rain hydrometeors.
As for the ice hydrometeors, the Dm–Nw pairs are shifted to smaller values than the near-surface DSDs for the LPFs and NLPFs, and are mainly concentrated for Nw values of 28 and 36, and the Dm values of 0.8 and 1.4 mm. There is an apparent growth in the mean size and decrease in Nw values of ice hydrometeors within LPFs and NLPFs, indicating more active rimming processes. The regional differences show that the DSDs of ice hydrometeors for the LPFs are characterized by large concentrations (Nw 33.87) of small hydrometeors (Dm 1.31 mm) over YHRB, and relatively small concentrations (Nw 32.79) of the largest hydrometeors (Dm 1.44 mm) over NEC, similar to the near-surface DSDs. The slopes of Ze from 3 and 1 km above the SH indicate that the Ze values mainly increase as the altitudes decrease for both LPFs and NLPFs, indicating the rimming, aggregation, or deposition processes. However, this increase in amplitude is much smaller for the LPFs than for the NLPFs. One possible explanation is that the stronger updrafts within LPFs transport large ice hydrometeors to higher altitudes, which makes the slopes smaller. Additionally, the increases in amplitude are much larger for more intense rain for NLPFs (large Ze values at 2 km above the MSL), suggesting more active riming, aggregation, and deposition of ice hydrometeors.
The CFADs show that the maximum frequency center of Ze and Dm occurs above the melting layer for the LPFs, but below the melting layer for the NLPFs. There are much more hydrometeors with larger Ze and Dm values for the LPFs throughout the column. One possible explanation is that the stronger updraft within the LPFs extends the water vapor to higher altitudes, which facilitates the aggregation or rimming of ice hydrometeors above the melting layer, forming large-sized ice hydrometeors. These processes promote bigger drops to precipitate within LPFs, therefore, relatively large drops are observed below the melting layer within LPFs. The LPFs over SC (NEC) have the deepest (shallowest) reflectivity echo-top of the three other regions, indicating stronger (weaker) development in the vertical direction. The most evident evaporation or break up of hydrometeors of LPFs below the melting layer occurs over NEC, followed by NC.
It is also important to study the regional variations of microphysics for convective cores in the four regions, e.g., the DSD characteristics. It was found that the mean Dm (Nw) values for the convective precipitation within the PFs are generally larger (smaller) than those for all types of precipitation. The mean Dm (Nw) values for the convective precipitation within LPFs are much larger (smaller) than those for NLPFs (figures not shown). Meanwhile, the convective precipitation has the highest concentrations of the smallest hydrometeors over YHRB, and the lowest concentration of the largest hydrometeors over NEC near the surface. These conclusions are similar to the results for all types of precipitation within the PFs presented above.
This study provides new evidence of the statistical characteristics of the microphysics of rain and ice hydrometeors within LPFs and NLPFs based on a relatively long period of GPM observations, and may improve future numerical models. However, it is also important to address the potential limitations of the study. Firstly, there are still some uncertainties in the DSD estimates in DPR official datasets, which may result from measurement errors, the assumptions of DSD model, the uncertainty of attenuation correction, and so on [21,22]. The improvement in the retrieval algorithms of the official DPR datasets would be helpful for the study on the precipitation microphysics in the future. Secondly, as revealed by Huang et al. [22], the stronger low-level wind shear and higher terrain may favor the ice processes above the melting layer over the Pearl River Delta Region, South China. Besides these, the complex environmental conditions, such as temperature, humidity, and water vapor responsible for the different microphysical processes (like evaporation, breakup, and coalescence) are worthy of further studies. Furthermore, a longer period of observations will be required to conduct a more robust analysis of the microphysical characteristics of precipitating clouds producing strong or frequent lightning flashes.

Author Contributions

Conceptualization and methodology, F.C. and H.H.; formal analysis, M.Z.; writing—original draft preparation, F.C.; writing—review and editing, L.Y. and X.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This work has been jointly supported by National Natural Science Foundation of China (grant 41805023), Basic Research Fund of CAMS (2020R002), and the innovation team construction plan by Anhui Meteorological Bureau.

Data Availability Statement

Data is contained within the article. The GPM DPR and GMI data can be downloaded from NASA Goddard Space Flight Center’s Mesoscale Atmospheric Processes Laboratory and Precipitation Processing System (PPS) providing (https://pmm.nasa.gov/data-access/downloads/gpm (accessed on 1 January 2022)).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Williams, E.R.; Weber, M.E.; Orville, R.E. The relationship between lightning type and convective state of thunderclouds. J. Geophys. Res. Atmos. 1989, 94, 13213–13220. [Google Scholar] [CrossRef] [Green Version]
  2. MacGorman, D.R.; Rust, W.D. The Electrical Nature of Storms; York, N., Ed.; Oxford Univ. Press: New York, NY, USA, 1999; 422p. [Google Scholar]
  3. Sun, J. Initialization and Numerical Forecasting of a Supercell Storm Observed during STEPS. Mon. Weather Rev. 2005, 133, 793–813. [Google Scholar] [CrossRef]
  4. Huang, H.; Zhao, K.; Zhang, G.; Hu, D.; Yang, Z. Optimized raindrop size distribution retrieval and quantitative rainfall estimation from polarimetric radar. J. Hydrol. 2020, 580, 124248. [Google Scholar] [CrossRef]
  5. Lang, T.J.; Miller, L.J.; Weisman, M.; Rutledge, S.A.; Barker, L.J.; Bringi, V.N.; Chandrasekar, V.; Detwiler, A.; Doesken, N.; Helsdon, J.; et al. The Severe Thunderstorm Electrification and Precipitation Study. Bull. Am. Meteorol. Soc. 2004, 85, 1107–1126. [Google Scholar] [CrossRef] [Green Version]
  6. Cui, Y.; Zheng, D.; Zhang, Y.; Ruan, Z.; Li, F.; Yao, W.; Meng, Q.; Zhao, C. Association of lightning occurrence with precipitation cloud column structure at a fixed position. Atmos. Res. 2022, 267, 105989. [Google Scholar] [CrossRef]
  7. Vendrasco, E.P.; Machado, L.A.T.; Araujo, C.S.; Ribaud, J.-F.; Ferreira, R.C. Potential use of the GLM for nowcasting and data assimilation. Atmos. Res. 2020, 242, 105019. [Google Scholar] [CrossRef]
  8. Mattos, E.V.; Machado, L.A.T.; Williams, E.R.; Albrecht, R.I. Polarimetric radar characteristics of storms with and without lightning activity. J. Geophys. Res. Atmos. 2016, 121, 201–214,220. [Google Scholar] [CrossRef]
  9. Huang, H.; Zhao, K.; Chan, J.C.L.; Hu, D. Microphysical Characteristics of Extreme-Rainfall Convection over the Pearl River Delta Region, South China from Polarimetric Radar Data during the Pre-summer Rainy Season. Adv. Atmos. Sci. 2022. [Google Scholar] [CrossRef]
  10. Petersen, W.A.; Rutledge, S.A. Regional Variability in Tropical Convection: Observations from TRMM. J. Clim. 2001, 14, 3566–3586. [Google Scholar] [CrossRef]
  11. Petersen, W.A.; Christian, H.J.; Rutledge, S.A. TRMM observations of the global relationship between ice water content and lightning. Geophys. Res. Lett. 2005, 32, L14819. [Google Scholar] [CrossRef]
  12. Xu, W.; Zipser, E.J. Convective intensity, vertical precipitation structures, and microphysics of two contrasting convective regimes during the 2008 TiMREX. J. Geophys. Res. Atmos. 2015, 120, 4000–4016. [Google Scholar] [CrossRef]
  13. Cecil, D.J.; Goodman, S.J.; Boccippio, D.J.; Zipser, E.J.; Nesbitt, S.W. Three Years of TRMM Precipitation Features. Part I: Radar, Radiometric, and Lightning Characteristics. Mon. Weather Rev. 2005, 133, 543–566. [Google Scholar] [CrossRef] [Green Version]
  14. Toracinta, E.R.; Zipser, E.J. Lightning and SSM/I-Ice-Scattering Mesoscale Convective Systems in the Global Tropics. J. Appl. Meteorol. 2001, 40, 983–1002. [Google Scholar] [CrossRef]
  15. Nesbitt, S.W.; Zipser, E.J.; Cecil, D.J. A Census of Precipitation Features in the Tropics Using TRMM: Radar, Ice Scattering, and Lightning Observations. J. Clim. 2000, 13, 4087–4106. [Google Scholar] [CrossRef]
  16. Xu, W.; Zipser, E.J.; Liu, C.; Jiang, H. On the relationships between lightning frequency and thundercloud parameters of regional precipitation systems. J. Geophys. Res. Atmos. 2010, 115, D12203. [Google Scholar] [CrossRef] [Green Version]
  17. Iguchi, T.; Oki, R.; Smith, E.; Furuhama, Y. Global Precipitation Measurement program and the development of dual-frequency precipitation radar. J.-Commun. Res. Lab. 2002, 49, 37–46. [Google Scholar]
  18. Smith, E.A.; Asrar, G.; Furuhama, Y.; Ginati, A.; Mugnai, A.; Nakamura, K.; Adler, R.F.; Chou, M.-D.; Desbois, M.; Durning, J.F. International global precipitation measurement (GPM) program and mission: An overview. In Measuring Precipitation from Space; Springer: Berlin/Heidelberg, Germany, 2007; pp. 611–653. [Google Scholar]
  19. Le, M.; Chandresekar, V. A Potential DSD Retrieval Process for Dual-Frequency Precipitation Radar (DPR) on Board GPM. Asia-Pac. Environ. Remote Sens. 2012, 8523. [Google Scholar] [CrossRef]
  20. D’Adderio, L.P.; Vulpiani, G.; Porcù, F.; Tokay, A.; Meneghini, R. Comparison of GPM Core Observatory and Ground-Based Radar Retrieval of Mass-Weighted Mean Raindrop Diameter at Midlatitude. J. Hydrometeorol. 2018, 19, 1583–1598. [Google Scholar] [CrossRef]
  21. Radhakrishna, B.; Satheesh, S.; Narayana Rao, T.; Saikranthi, K.; Sunilkumar, K. Assessment of DSDs of GPM-DPR with ground-based disdrometer at seasonal scale over Gadanki, India. J. Geophys. Res. Atmos. 2016, 121, 11792–11802. [Google Scholar] [CrossRef]
  22. Huang, H.; Zhao, K.; Fu, P.; Chen, H.; Chen, G.; Zhang, Y. Validation of Precipitation Measurements from the Dual-Frequency Precipitation Radar Onboard the GPM Core Observatory Using a Polarimetric Radar in South China. IEEE Trans. Geosci. Remote Sens. 2021, 1–16. [Google Scholar] [CrossRef]
  23. Zhang, A.; Fu, Y. The structural characteristics of precipitating cases detected by GPM Dual-frequency Radar. Chin. J. Atmos. Sci. 2018, 42, 33–51. [Google Scholar]
  24. Huang, H.; Chen, F. Precipitation Microphysics of Tropical Cyclones Over the Western North Pacific Based on GPM DPR Observations: A Preliminary Analysis. J. Geophys. Res. Atmos. 2019, 124, 3124–3142. [Google Scholar] [CrossRef]
  25. Chen, F.; Fu, Y.; Yang, Y. Regional Variability of Precipitation in Tropical Cyclones Over the Western North Pacific Revealed by the GPM Dual-Frequency Precipitation Radar and Microwave Imager. J. Geophys. Res. Atmos. 2019, 124, 11281–11296. [Google Scholar] [CrossRef]
  26. Iguchi, T.; Seto, S.; Meneghini, R.; Yoshida, N.; Awaka, J.; Kubota, T. GPM/DPR Level-2 Algorithm Theoretical Basis Document; NASA Goddard Space Flight Center: Greenbelt, MD, USA, 2017.
  27. Liao, L.; Meneghini, R. GPM DPR Retrievals: Algorithm, Evaluation, and Validation. Remote Sens. 2022, 14, 843. [Google Scholar] [CrossRef]
  28. Awaka, J.; Le, M.; Chandrasekar, V.; Yoshida, N.; Higashiuwatoko, T.; Kubota, T.; Iguchi, T. Rain Type Classification Algorithm Module for GPM Dual-Frequency Precipitation Radar. J. Atmos. Ocean. Technol. 2016, 33, 1887–1898. [Google Scholar] [CrossRef]
  29. Chen, F.; Zheng, X.; Wen, H.; Yuan, Y. Microphysics of Convective and Stratiform Precipitation during the Summer Monsoon Season over the Yangtze–Huaihe River Valley, China. J. Hydrometeorol. 2022, 23, 239–252. [Google Scholar] [CrossRef]
  30. Draper, D.W.; Newell, D.A.; Wentz, F.J.; Krimchansky, S.; Skofronick-Jackson, G.M. The global precipitation measurement (GPM) microwave imager (GMI): Instrument overview and early on-orbit performance. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2015, 8, 3452–3462. [Google Scholar] [CrossRef]
  31. Toracinta, E.R.; Cecil, D.J.; Zipser, E.J.; Nesbitt, S.W. Radar, Passive Microwave, and Lightning Characteristics of Precipitating Systems in the Tropics. Mon. Weather Rev. 2002, 130, 802–824. [Google Scholar] [CrossRef]
  32. Liu, C.; Cecil, D.; Zipser, E.J. Relationships between lightning flash rates and passive microwave brightness temperatures at 85 and 37 GHz over the tropics and subtropics. J. Geophys. Res. Atmos. 2011, 116, D23108. [Google Scholar] [CrossRef] [Green Version]
  33. Spencer, R.W.; Goodman, H.M.; Hood, R.E. Precipitation retrieval over land and ocean with the SSM/I: Identification and characteristics of the scattering signal. J. Atmos. Ocean. Technol. 1989, 6, 254–273. [Google Scholar] [CrossRef]
  34. Liu, C.; Zipser, E.J.; Nesbitt, S.W. Global distribution of tropical deep convection: Different perspectives from TRMM infrared and radar data. J. Clim. 2007, 20, 489–503. [Google Scholar] [CrossRef]
  35. Ding, Y.; Wang, Z.; Sun, Y. Inter-decadal variation of the summer precipitation in East China and its association with decreasing Asian summer monsoon. Part I: Observed evidences. Int. J. Climatol. 2008, 28, 1139–1161. [Google Scholar] [CrossRef]
  36. Ma, M.; Tao, S.; Zhu, B.; LÜ, W. Climatological distribution of lightning density observed by satellites in China and its circumjacent regions. Sci. China Ser. D Earth Sci. 2005, 48, 219–229. [Google Scholar] [CrossRef]
  37. Yang, X.; Sun, J.; Li, W. An Analysis of Cloud-to-Ground Lightning in China during 2010–13. Weather Forecast. 2015, 30, 1537–1550. [Google Scholar] [CrossRef]
  38. Chen, F.; Fu, Y.; Liu, P.; Yang, Y. Seasonal Variability of Storm Top Altitudes in the Tropics and Subtropics Observed by TRMM PR. Atmos. Res. 2016, 169, 113–126. [Google Scholar] [CrossRef]
  39. Yoshida, S.; Adachi, T.; Kusunoki, K.; Hayashi, S.; Wu, T.; Ushio, T.; Yoshikawa, E. Relationship between thunderstorm electrification and storm kinetics revealed by phased array weather radar. J. Geophys. Res. Atmos. 2017, 122, 3821–3836. [Google Scholar] [CrossRef]
  40. Yoshida, S.; Morimoto, T.; Ushio, T.; Kawasaki, Z. A fifth-power relationship for lightning activity from Tropical Rainfall Measuring Mission satellite observations. J. Geophys. Res. Atmos. 2009, 114, D09104. [Google Scholar] [CrossRef]
  41. Bringi, V.N.; Chandrasekar, V.; Hubbert, J.; Gorgucci, E.; Randeu, W.L.; Schoenhuber, M. Raindrop Size Distribution in Different Climatic Regimes from Disdrometer and Dual-Polarized Radar Analysis. J. Atmos. Sci. 2003, 60, 354–365. [Google Scholar] [CrossRef]
  42. Wang, M.; Zhao, K.; Xue, M.; Zhang, G.; Liu, S.; Wen, L.; Chen, G. Precipitation microphysics characteristics of a Typhoon Matmo (2014) rainband after landfall over eastern China based on polarimetric radar observations. J. Geophys. Res. Atmos. 2016, 121, 12415–12433. [Google Scholar] [CrossRef]
  43. Zeng, Q.; Zhang, Y.; Lei, H.; Xie, Y.; Gao, T.; Zhang, L.; Wang, C.; Huang, Y. Microphysical Characteristics of Precipitation during Pre-monsoon, Monsoon, and Post-monsoon Periods over the South China Sea. Adv. Atmos. Sci. 2019, 36, 1103–1120. [Google Scholar] [CrossRef]
  44. Yang, Y.; Wang, R.; Chen, F.; Liu, C.; Bi, X.; Huang, M. Synoptic weather patterns modulate the frequency, type and vertical structure of summer precipitation over Eastern China: A perspective from GPM observations. Atmos. Res. 2021, 249, 105342. [Google Scholar] [CrossRef]
  45. Dong, W.; Lin, Y.; Wright, J.S.; Xie, Y.; Yin, X.; Guo, J. Precipitable water and CAPE dependence of rainfall intensities in China. Clim. Dyn. 2019, 52, 3357–3368. [Google Scholar] [CrossRef]
  46. Chatterjee, C.; Das, S. On the association between lightning and precipitation microphysics. J. Atmos. Sol.-Terr. Phys. 2020, 207, 105350. [Google Scholar] [CrossRef]
  47. Chen, F.; Huang, H. Comparisons of Gauge, TMPA and IMERG Products for Monsoon and Tropical Cyclone Precipitation in Southern China. Pure Appl. Geophys. 2019, 176, 1767–1784. [Google Scholar] [CrossRef]
  48. Kumjian, M.R.; Prat, O.P. The Impact of Raindrop Collisional Processes on the Polarimetric Radar Variables. J. Atmos. Sci. 2014, 71, 3052–3067. [Google Scholar] [CrossRef]
  49. Liu, C.; Zipser, E.J. Why does radar reflectivity tend to increase downward toward the ocean surface, but decrease downward toward the land surface? J. Geophys. Res. Atmos. 2013, 118, 135–148. [Google Scholar] [CrossRef] [Green Version]
  50. Liu, G.; Fu, Y. The Characteristics of Tropical Precipitation Profiles as Inferred From Satellite Radar Measurements. J. Meteorol. Soc. Jpn. Ser. II 2001, 79, 131–143. [Google Scholar] [CrossRef] [Green Version]
  51. Berdeklis, P.; List, R. The Ice Crystal–Graupel Collision Charging Mechanism of Thunderstorm Electrification. J. Atmos. Sci. 2001, 58, 2751–2770. [Google Scholar] [CrossRef]
  52. Dash, J.G.; Mason, B.L.; Wettlaufer, J.S. Theory of charge and mass transfer in ice-ice collisions. J. Geophys. Res. Atmos. 2001, 106, 20395–20402. [Google Scholar] [CrossRef]
  53. Seto, S.; Iguchi, T.; Oki, T. The Basic Performance of a Precipitation Retrieval Algorithm for the Global Precipitation Measurement Mission’s Single/Dual-Frequency Radar Measurements. IEEE Trans. Geosci. Remote Sens. 2013, 51, 5239–5251. [Google Scholar] [CrossRef] [Green Version]
  54. Yuter, S.E.; Houze, R.A. Three-Dimensional Kinematic and Microphysical Evolution of Florida Cumulonimbus. Part II: Frequency Distributions of Vertical Velocity, Reflectivity, and Differential Reflectivity. Mon. Weather Rev. 1995, 123, 1941–1963. [Google Scholar] [CrossRef]
  55. Hence, D.A.; Houze, R.A. Vertical structure of hurricane eyewalls as seen by the TRMM Precipitation Radar. J. Atmos. Sci. 2011, 68, 1637–1652. [Google Scholar] [CrossRef]
Remotesensing 14 06072 g001 550
Figure 1. An example of precipitation features observed by GPM satellite. (a) Near-surface RR, (b) lightning flashes, (c) classifications of LPF and NLPF areas with blue (red) contiguous rainy pixels representing LPFs (NLPFs), (d) rain type, (e) storm-top height, (f) PCT89, (g) Dm at 2 km altitude and (h) Nw at 2 km above the MSL.
Figure 1. An example of precipitation features observed by GPM satellite. (a) Near-surface RR, (b) lightning flashes, (c) classifications of LPF and NLPF areas with blue (red) contiguous rainy pixels representing LPFs (NLPFs), (d) rain type, (e) storm-top height, (f) PCT89, (g) Dm at 2 km altitude and (h) Nw at 2 km above the MSL.
Remotesensing 14 06072 g001
Remotesensing 14 06072 g002 550
Figure 2. Frequency pattern in two-dimensional space of Dm and Nw at 2 km above the MSL over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021. D m ¯ and N w ¯ denote the mean values of Dm and Nw, respectively. The 5% and 50% contours are indicated by black and white solid lines, respectively.
Figure 2. Frequency pattern in two-dimensional space of Dm and Nw at 2 km above the MSL over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021. D m ¯ and N w ¯ denote the mean values of Dm and Nw, respectively. The 5% and 50% contours are indicated by black and white solid lines, respectively.
Remotesensing 14 06072 g002
Remotesensing 14 06072 g003 550
Figure 3. Frequency pattern in two-dimensional space of the slope of Ze from 1 km to 3 km below the LH and Ze at 2 km above the MSL over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for the LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021.
Figure 3. Frequency pattern in two-dimensional space of the slope of Ze from 1 km to 3 km below the LH and Ze at 2 km above the MSL over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for the LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021.
Remotesensing 14 06072 g003
Remotesensing 14 06072 g004 550
Figure 4. Probability distributions (%) of (a) storm-top height, (b) PCT89, (c) Ze at 1 km above the SH, (d) Dm at 1 km above the SH and (e) Nw at 1 km above the SH over SC, YHRB, NC and NEC for LPFs (solid lines) and NLPFs (dashed lines) during the summer for the time period 2014–2021. The bin sizes for storm-top height, PCT89, Ze, Dm and Nw are 0.5 km, 10 K, 1 dBZ, 0.1 mm and 1.0, respectively.
Figure 4. Probability distributions (%) of (a) storm-top height, (b) PCT89, (c) Ze at 1 km above the SH, (d) Dm at 1 km above the SH and (e) Nw at 1 km above the SH over SC, YHRB, NC and NEC for LPFs (solid lines) and NLPFs (dashed lines) during the summer for the time period 2014–2021. The bin sizes for storm-top height, PCT89, Ze, Dm and Nw are 0.5 km, 10 K, 1 dBZ, 0.1 mm and 1.0, respectively.
Remotesensing 14 06072 g004
Remotesensing 14 06072 g005 550
Figure 5. Frequency pattern in two-dimensional space of Dm and Nw at 1 km above the SH over (a,e) SC, (b,f) YRHB, (c,g) NC and (d,h) NEC for LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021. D m ¯   and   N w ¯   denote the mean Dm and Nw values at 1 km above the SH, respectively. The 5% and 50% contours are indicated by black and white solid lines, respectively.
Figure 5. Frequency pattern in two-dimensional space of Dm and Nw at 1 km above the SH over (a,e) SC, (b,f) YRHB, (c,g) NC and (d,h) NEC for LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021. D m ¯   and   N w ¯   denote the mean Dm and Nw values at 1 km above the SH, respectively. The 5% and 50% contours are indicated by black and white solid lines, respectively.
Remotesensing 14 06072 g005
Remotesensing 14 06072 g006 550
Figure 6. Frequency pattern in two-dimensional space of the slope of Ze from 3 km to 1 km above the SH and Ze at 2 km above the MSL over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for the LPFs (upper panels) and NLPFs (lower panels) during summer in the time period 2014–2021.
Figure 6. Frequency pattern in two-dimensional space of the slope of Ze from 3 km to 1 km above the SH and Ze at 2 km above the MSL over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for the LPFs (upper panels) and NLPFs (lower panels) during summer in the time period 2014–2021.
Remotesensing 14 06072 g006
Remotesensing 14 06072 g007 550
Figure 7. CFADs of Ze over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for the LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021. The 10% and 50% contours are indicated by black and white solid lines, respectively. The bin sizes for Ze and height are 1 dBZ and 250 m, respectively. Purple lines denote the mean heights of the LH over four regions.
Figure 7. CFADs of Ze over (a,e) SC, (b,f) YHRB, (c,g) NC and (d,h) NEC for the LPFs (upper panel) and NLPFs (lower panel) during summer in the time period 2014–2021. The 10% and 50% contours are indicated by black and white solid lines, respectively. The bin sizes for Ze and height are 1 dBZ and 250 m, respectively. Purple lines denote the mean heights of the LH over four regions.
Remotesensing 14 06072 g007
Remotesensing 14 06072 g008 550
Figure 8. Similar to Figure 7, but for Dm. The bin sizes for Dm and height are 0.1 mm, 250 m, respectively.
Figure 8. Similar to Figure 7, but for Dm. The bin sizes for Dm and height are 0.1 mm, 250 m, respectively.
Remotesensing 14 06072 g008
Table
Table 1. Total number of DPR pixels, the mean near-surface RR, the LH, the convective and stratiform frequency, and the reflectivity at 2 km altitude above the MSL for LPFs and NLPFs over SC, YHRB, NC and NEC during summer in the time period 2014–2021.
Table 1. Total number of DPR pixels, the mean near-surface RR, the LH, the convective and stratiform frequency, and the reflectivity at 2 km altitude above the MSL for LPFs and NLPFs over SC, YHRB, NC and NEC during summer in the time period 2014–2021.
SCYHRBNCNEC
LPFsNLPFsLPFsNLPFsLPFsNLPFsLPFsNLPFs
No. of samples21,17444,296122,161258,43063,559134,025106,982325,955
RR (mm h−1)4.332.754.122.564.212.233.572.09
LH (km)4.955.044.874.964.124.333.483.52
Convective frequency (%)31.0219.9925.2812.8128.7113.2528.9912.41
Stratiform frequency (%)64.0254.470.9666.0770.2778.2570.583.64
Ze (dBZ)38.6534.4937.9432.8139.2333.1138.0132.65
Table
Table 2. Mean PCT89, storm-top height, and reflectivity at 1 km above the SH for LPFs and NLPFs over SC, YHRB, NC and NEC during the summer for the time period 2014–2021.
Table 2. Mean PCT89, storm-top height, and reflectivity at 1 km above the SH for LPFs and NLPFs over SC, YHRB, NC and NEC during the summer for the time period 2014–2021.
SCYHRBNCNEC
LPFsNLPFsLPFsNLPFsLPFsNLPFsLPFsNLPFs
PCT89 (K)250268251268246265248265
storm-top height (km)7.85.87.65.67.55.76.95.1
Ze (dBZ)31.1527.2830.6425.7533.2427.2732.2926.78
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chen, F.; Zeng, M.; Yu, L.; Zhuge, X.; Huang, H. Regional Variability in Microphysical Characteristics of Precipitation Features with Lightning across China: Observations from GPM. Remote Sens. 2022, 14, 6072. https://doi.org/10.3390/rs14236072

AMA Style

Chen F, Zeng M, Yu L, Zhuge X, Huang H. Regional Variability in Microphysical Characteristics of Precipitation Features with Lightning across China: Observations from GPM. Remote Sensing. 2022; 14(23):6072. https://doi.org/10.3390/rs14236072

Chicago/Turabian Style

Chen, Fengjiao, Mingjian Zeng, Lu Yu, Xiaoyong Zhuge, and Hao Huang. 2022. "Regional Variability in Microphysical Characteristics of Precipitation Features with Lightning across China: Observations from GPM" Remote Sensing 14, no. 23: 6072. https://doi.org/10.3390/rs14236072

APA Style

Chen, F., Zeng, M., Yu, L., Zhuge, X., & Huang, H. (2022). Regional Variability in Microphysical Characteristics of Precipitation Features with Lightning across China: Observations from GPM. Remote Sensing, 14(23), 6072. https://doi.org/10.3390/rs14236072

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop