Preliminary Study of Land–Sea Microphysics Associated with the East Asian Summer Monsoon Rainband and Its Application to GPM DPR
1. Introduction
The East Asian summer monsoon (EASM) rainband, a zonally elongated monsoon rainband extending eastward from southwestern China to the northwestern Pacific (NWP) through the Japanese archipelago (Fig. 1a), is one of the most significant rainfall systems affecting the hydrological cycle in the East Asian region (Sampe and Xie 2010; Kuwano-Yoshida et al. 2013). Rainstorm associated with EASM is crucial for summertime water resources and sometimes results in severe natural disasters in East Asian countries (Ogura et al. 1985; Ninomiya 2000; Ninomiya and Shibagaki 2003; Yoshizaki et al. 2000; Kato and Aranami 2005; Kato 2006). Its extended also intimidates the security of naval ships on sea. Therefore, an urgent need to improve the performance of quantitative precipitation estimation (QPE) within the rainfall system for flood forecast, urban waterlogging prediction, and ocean storm monitoring is required.
(a) GPCP precipitation during mei-yu season (16 Jun–15 Jul) [contour interval: 2 mm day−1, with light (dark) shading for 4–8 (>8) mm day−1] and locations of two observational regions in EASM rainband. (b) Probability density function (PDF) of TBB during the rainy days in two parts of EASM rainband. The superimposed rectangles in (a) correspond to the assumed western part and eastern part of this rainband. The distribution of TBB in (b) are obtained from FY-2E data products.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
(a) GPCP precipitation during mei-yu season (16 Jun–15 Jul) [contour interval: 2 mm day−1, with light (dark) shading for 4–8 (>8) mm day−1] and locations of two observational regions in EASM rainband. (b) Probability density function (PDF) of TBB during the rainy days in two parts of EASM rainband. The superimposed rectangles in (a) correspond to the assumed western part and eastern part of this rainband. The distribution of TBB in (b) are obtained from FY-2E data products.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
(a) GPCP precipitation during mei-yu season (16 Jun–15 Jul) [contour interval: 2 mm day−1, with light (dark) shading for 4–8 (>8) mm day−1] and locations of two observational regions in EASM rainband. (b) Probability density function (PDF) of TBB during the rainy days in two parts of EASM rainband. The superimposed rectangles in (a) correspond to the assumed western part and eastern part of this rainband. The distribution of TBB in (b) are obtained from FY-2E data products.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Raindrop size distribution (DSD) is of great significance for the understanding of microphysical processes in generating rainfall particles and ultimately for improving radar QPE algorithms. Over the last decades, DSD characteristics in different continental and marine locations from tropics to midlatitudes along with different disdrometers have been well documented and are known to vary spatially and temporally (e.g., Rosenfeld and Ulbrich 2003; Kozu et al. 2005; Kirankumar et al. 2008; Chakravarty and Raj 2013; Williams et al. 2014; Wen et al. 2016; Zhang et al. 2017; Wu et al. 2019a). The entire EASM rainband locates from China to Japan and extends to NWP, which is an ideal precipitation system for studying microphysical properties of different parts within the same weather system, since it covers both continental and oceanic areas. Apart from the underlying surface variability in the EASM rainband, the land–sea atmospheric patterns also differ from each other with equivalent barotropy in its continental part whereas typical midlatitude baroclinicity in its oceanic part (Chen and Chang 1980). Although studies have been reported to address the DSD of EASM, they mainly focused on its continental part in East Asia (e.g., Wen et al. 2016, 2017; Chen et al. 2013; Bringi et al. 2006). DSDs of the monsoon rainfall over oceans have received little investigations.
Moreover, atmospheric aerosols, formed from natural and anthropogenic processes, can also have an effect on precipitation through aerosol–cloud interactions (ACI) (e.g., May et al. 2011; Ramanathan et al. 2001). It has been well documented in the literature that the aerosols can act as cloud condensation nuclei (CCN), increases the number of cloud drops and reduces the effective size of the cloud particles (e.g., Ramanathan et al. 2001; Breon et al. 2002). The aerosol effects on microphysics of EASM rainfall should not be neglected due to the rapidly expanding economic and industrial developments in East Asian countries, causing a prominent increase in aerosol loading over this area (Streets et al. 2008; Liu et al. 2011; Jiang et al. 2017; Zhang et al. 2019).
Various radar QPE algorithms, including the recent Global Precipitation Measurement (GPM) Dual-Frequency Precipitation Radar (DPR) highly depend on surface-based DSD observations (e.g., Liao et al. 2014; Radhakrishna et al. 2016; Chen et al. 2017; Liao and Meneghini 2019). GPM DPR satellite is expected to improve our knowledge of precipitation processes by providing greater dynamic range, more detailed information on microphysics, and better accuracy values in rainfall retrievals (Le and Chandrasekar 2013). Hydrological validation of DPR with surface DSD measurements helps to improve its physically based algorithms in EASM season.
Based on the above analyses, one specific subject herein is to obtain comprehensive understanding of DSDs in different parts of the EASM rainband including the aerosol effects, and ultimately improves the recent GPM DPR rainfall estimations in the EASM season. For a first attempt, measurements from surface Particle Size Velocity (Parsivel) disdrometers, along with satellite data such as Global Precipitation Climatology Project (GPCP), Fengyun-2E (FY-2E), Moderate Resolution Imaging Spectroradiometer (MODIS), and the European Centre for Medium-Range Weather Forecasts (ECMWF) interim reanalysis (ERA-Interim) data, in situ radiosonde data, are united to illustrate the dynamical and microphysical characteristics associated with the EASM rainband. And DSD-based relations are further derived to improve the accuracy of GPM DPR precipitation estimations in the EASM season.
The rest of this paper proceeds as follows: the data and methods are described in section 2, an outline of DSD and the possible reasons for the variations in overall DSD characteristics among two parts of the EASM rainband are analyzed in section 3, the relations of DSD parameters are obtained in section 4, GPM rainfall retrieval methods are discussed in section 5, and finally section 6 presents a summary and conclusions.
2. Data and methods
a. Observational sites and datasets
During June to July 2014, observations from in situ disdrometers and satellites are collected at Nanjing (32.0°N, 118.5°E; 15 m MSL), Chuzhou (32.3°N, 118.3°E; 18 m MSL) and NWP (27°–32°N, 143°–153°E; 22 m MSL) sites (Fig. 2). NWP is located in the southeast ocean area of Japan Sea, while Nanjing and Chuzhou are located in the East China continents. The thermal effect of land–sea contrast can contribute to the onset of EASM (Ding and Chan 2005). Due to different underlying surface properties and environmental conditions, the precipitation microphysical variability in different parts of the monsoon rainband could be of great significance. For further comparison, we separated the main EASM rainband and adopted Nanjing and Chuzhou data jointly as the western part of the EASM rainband (hereinafter the western part), while NWP data as the eastern part of this rainband (hereinafter the eastern part). And the multisource data products utilized in this work are presented as below:
Field view of (top right) the observational Chuzhou site (CZ) and Nanjing site (NJ), as well as (bottom) the northwestern Pacific site (NWP). (top left) The local topography is also presented.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Field view of (top right) the observational Chuzhou site (CZ) and Nanjing site (NJ), as well as (bottom) the northwestern Pacific site (NWP). (top left) The local topography is also presented.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Field view of (top right) the observational Chuzhou site (CZ) and Nanjing site (NJ), as well as (bottom) the northwestern Pacific site (NWP). (top left) The local topography is also presented.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
1) In situ Parsivel2 and radiosonde data
The DSD data selected for the analyses were measured by Parsivel2 disdrometers manufactured by OTT Hydromet, Germany (Tokay et al. 2014; Löffler-Mang and Joss 2000). Parsivel2 disdrometer herein is a second-generation optical disdrometer with a near-IR (780 nm) rectangular single beam of 27 mm wide, 180 mm long (that is a sensor horizontal area of 48.6 cm2), and 10 mm high. It archives equivalent drop diameters sorted into 32 classes of diameter from 0 to 25 mm and 32 classes of fall speeds from 0 to 22.4 m s−1. The time resolution can be selected and is set to 1 min in current study. The 1-min DSD data in NWP were measured over western Pacific by onboard Parsivel2 disdrometer during a marine survey (Zhang 2018). Meanwhile, two Parsivel2 disdrometers simultaneously measure the rainfall in Nanjing (in Jiangsu Province) and Chuzhou (in Anhui Province), China, respectively. Following the data quality control method in Wu et al. (2019a), as well as the definition of a rain event proposed by Tokay and Bashor (2010), 10 rain events incorporating 3688 one-minute effective DSD samples are identified for the western part, 11 rain events incorporating 975 samples are identified for the eastern part (Table 1). Meteorological variables (such as pressure, temperature, relative humidity, and horizontal winds) from in situ radiosonde (twice a day) are used.
Precipitation events used for the present study in two parts of monsoon rainband.
2) GPM DPR level-2 products
GPM DPR consists of two precipitation radars (Ku band and Ka band), under cooperation of the National Aeronautics and Space Administration (NASA) and the Japan Aerospace Exploration Agency (JAXA). It has an orbital inclination angle of 65° and an altitude of approximately 407 km. In the present study, the attenuation-corrected radar reflectivities and rain rate at the estimated surface are obtained from level-2A products of DPR and are matched with the Parsivel disdrometer data, as is already described in Wu et al. (2019b). Overall, there are 4 (7) effective observations from 8 (11) instantaneous cases over the western part (eastern part) chosen for the comparison between DPR and Parsivel. Table 2 gives the DPR-observed rain events during the EASM period from June to July 2014.
Precipitation events observed by GPM DPR when it overpasses the two parts of monsoon rainband.
3) FY-2E radiation products and MODIS cloud products
Blackbody temperature (TBB) products from FY-2E satellite available at 0.1° × 0.1° spatial resolution are used with 1 h interval are used as in Wu et al. (2019a). Cloud and aerosol data, available at 1° × 1° spatial resolution, from level-3 daily global products of MODIS (MOD08_D3) are also used (Platnick et al. 2003). MOD08_D3 provides atmospheric parameters associated with aerosol particle properties, water vapor, as well as cloud optical and physical properties (Remer et al. 2005).
4) GPCP merged precipitation data and ERA-Interim data
Daily GPCP merged precipitation with 1° × 1° spatial resolution are used (Huffman et al. 2012). ERA-Interim data, available four times a day with a horizontal resolution of 0.125° × 0.125° and 38 pressure levels in the vertical direction, are used (Dee et al. 2011).
In this study, the data products from GPM, FY-2E, MODIS, GPCP, ERA-Interim, and in situ radiosonde are daily averaged to match with the Parsivel2 measured rainy days over two parts of the EASM rainband.
b. Parsivel2 disdrometer data processing
The DSDs are computed from the Parsivel2 disdrometer counts and the integral precipitation variables, such as the rain rate R (mm h−1), radar reflectivity Z (mm6 m−3), mass-weighted mean diameter Dm (mm), generalized intercept parameter Nw (mm−1 m−3), and standard deviation of the mass spectrum σM (mm) are further estimated from measured DSDs (e.g., Wen et al. 2016; Wu et al. 2019a).
So far, the most commonly used mathematical model in formulating and analyzing measured DSDs is the gamma distribution proposed by Ulbrich (1983):
where N(D) represents the particle concentration, three parameters N0, µ, and Λ represent the intercept, shape, and slope of DSDs, respectively. The truncated moment fitting (TMF) method proposed by Vivekanandan et al. (2004) is used to construct the gamma model from disdrometer data.
c. GPM DPR variables calculated from DSD
In terms of DSD observations, the effective radar reflectivity factor Ze for both Ku band and Ka band can be formulated as following:
where λ is the wavelength, N(D) is the DSD,
The difference between the Ku- and Ka-band reflectivity measurements can be well described by the dual-frequency ratio (DFR), which is defined as the ratio of Ze at Ku band (ZKu) to Ze at Ka band (ZKa). According to Seto et al. (2013), the DFR (dB) is uncorrelated with Nw and is only a function of Dm;
Note the calculations of Ze and DFR based on disdrometer observations are the same as in Wu et al. (2019b).
3. Outline of DSD
a. Drop size spectra
Before a detailed discussion of DSD property, we examined the average DSDs for the western part and the eastern part of the EASM rainband. To discern the precipitation differences between the two parts, the DSD samples are stratified into six classes on the basis of rain rate: R ≤ 2 mm h−1, 2 < R ≤ 5 mm h−1, 5 < R ≤ 10 mm h−1, 10 < R ≤ 20 mm h−1, 20 < R ≤ 40 mm h−1, and R > 40 mm h−1. Large (small) drops are assumed to be D > 4 (D < 1) mm.
Figure 3 presents the average drop spectra for two parts in terms of six rain-rate classes, along with the remained data samples N of each rain-rate class. Note that the DSDs exhibit similar peak structures, and the content of large drops has an increasing tendency with rain rate. Particularly, the drop size spectra of the western part is narrower than that of the eastern part when R ≤ 5 mm h−1, whereas it becomes significantly broader than that of the eastern part when R > 5 mm h−1. The concentration of small drops in the western part stays nearly constant when R ≤ 20 mm h−1, while it increased significantly when R > 20 mm h−1.
The average DSDs at the six indicated rain-rate classes over two parts of EASM rainband. The remaining data samples N of each rain-rate class are also presented.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The average DSDs at the six indicated rain-rate classes over two parts of EASM rainband. The remaining data samples N of each rain-rate class are also presented.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The average DSDs at the six indicated rain-rate classes over two parts of EASM rainband. The remaining data samples N of each rain-rate class are also presented.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
b. Analysis
To explain the unique characteristics of drop size spectra, some environmental variables are collected and analyzed. The distribution of TBB for the rainy days in two parts of the EASM rainband are described with probability density function (PDF) in Fig. 1b. CAPE values and surface temperature for the Parsivel2 measured rainy days over two parts are obtained from ERA-Interim, and are depicted with box-and-whisker plot in Figs. 4a and 4b. The average wind field, as well as relative humidity profile in Figs. 5a and 5b are all obtained from the in situ radiosonde, and are also drawn for the rainy days in two parts. The cloud structure, cloud effective radii (CER) of liquid and ice particles, and aerosol optical depth (AOD) for the rainy days over the observational regions are acquired from MODIS data products and are provided in Figs. 5a, 6a, and 6b.
Box-and-whisker plots of (a) CAPE (kJ kg−1) and (b) surface temperature (°C) from ERA-Interim over the western (red box) and eastern (blue box) parts of EASM rainband. The center line of the box indicates the median, and the bottom and top lines of the box indicate the 25th and 75th percentiles, respectively. The bottom and top of the solid vertical lines indicate the minimum and maximum values, respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Box-and-whisker plots of (a) CAPE (kJ kg−1) and (b) surface temperature (°C) from ERA-Interim over the western (red box) and eastern (blue box) parts of EASM rainband. The center line of the box indicates the median, and the bottom and top lines of the box indicate the 25th and 75th percentiles, respectively. The bottom and top of the solid vertical lines indicate the minimum and maximum values, respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Box-and-whisker plots of (a) CAPE (kJ kg−1) and (b) surface temperature (°C) from ERA-Interim over the western (red box) and eastern (blue box) parts of EASM rainband. The center line of the box indicates the median, and the bottom and top lines of the box indicate the 25th and 75th percentiles, respectively. The bottom and top of the solid vertical lines indicate the minimum and maximum values, respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Vertical profiles of (a) cloud structure (heights of cloud base, level of 0°C, cloud top; all in km) and vertical wind field (m s−1) and (b) relative humidity (%) from MODIS and in situ radiosonde observations in two parts of EASM rainband. The green circles indicate cloud base, the red rectangles indicate the level of 0°C, and the blue triangles indicate cloud top. A full (half) wind barb represents a wind speed of 10 (5) m s−1.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Vertical profiles of (a) cloud structure (heights of cloud base, level of 0°C, cloud top; all in km) and vertical wind field (m s−1) and (b) relative humidity (%) from MODIS and in situ radiosonde observations in two parts of EASM rainband. The green circles indicate cloud base, the red rectangles indicate the level of 0°C, and the blue triangles indicate cloud top. A full (half) wind barb represents a wind speed of 10 (5) m s−1.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Vertical profiles of (a) cloud structure (heights of cloud base, level of 0°C, cloud top; all in km) and vertical wind field (m s−1) and (b) relative humidity (%) from MODIS and in situ radiosonde observations in two parts of EASM rainband. The green circles indicate cloud base, the red rectangles indicate the level of 0°C, and the blue triangles indicate cloud top. A full (half) wind barb represents a wind speed of 10 (5) m s−1.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
As in Fig. 4, but for (a) cloud effective radius (CER, μm) of liquid and ice particles and (b) aerosol optical depth (AOD) from MODIS.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
As in Fig. 4, but for (a) cloud effective radius (CER, μm) of liquid and ice particles and (b) aerosol optical depth (AOD) from MODIS.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
As in Fig. 4, but for (a) cloud effective radius (CER, μm) of liquid and ice particles and (b) aerosol optical depth (AOD) from MODIS.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
In general, an increase in rain rate could be associated with the enhancement of convection activity. TBB reflects the developing intensity of convection activity and is considered as an proof for convective development when TBB ≤ −32°C (241.15 K) (e.g., Maddox 1980). Figure 1b suggests more intense development of convections in terms of frequency of occurrence in the western part. Strong convective activity extends the clouds to deeper altitudes, which enhances cold rain processes. And this can be confirmed by the relatively larger CAPE value, higher surface temperature (Fig. 4b) and cloud top (Fig. 5a) in the western part. Deeper the extent of clouds, more ice/liquid particles grow to larger size. Yet we can notice larger CER of ice and water over observational site of the eastern part than the western part (Fig. 6a). This could be due to the higher AOD over the western part than the eastern part (Fig. 6b), causing an increase in the number of cloud drops and an decrease in the effective size of the cloud particles. Large size of cloud droplets are favorable for the warm rain process (collision–coalescence) when rain rate is low. Thus, the relatively lower concentration of CCN and cloud droplet over the eastern part could contribute to larger CER (Fig. 6a), which results in broader drop size spectra through the collision–coalescence process between cloud drops. Whereas the higher aerosol concentration over the western part (Fig. 6b) suppresses the collision–coalescence processes, increases the number of cloud droplets and decreases their effective radius (Fig. 6a), leading to narrow drop size spectra (Wallace and Hobbs 1977; Twomey 1977; Albrecht 1989; Rosenfeld 1999; Andreae et al. 2004).
As rain rate increases, small drops are carried to higher altitudes by strong updrafts. Thus, more CCN due to higher AOD in the western part enter the clouds above the melting layer. And this strengthens more freezing of cloud drops and associated latent heat release (Rosenfeld and Woodley 2000) through cold rain processes (Bergeron process, riming, aggregation), which can upturn the convective activity and bring more large droplets, contributing to broader drop size spectra in the western part. Based on the above analyses, the aerosols could help inhibit warm rain process while enhance cold rain process in the western part, which partly explains the finding that the drop size spectra of the western part could be narrower than that of the eastern part when R ≤ 5 mm h−1, whereas it becomes significantly broader than that of the eastern part when R > 5 mm h−1.
As the spectra broadens in the western part, more large drops would collect small drops, which is favorable for the collision–coalescence processes with a consumption of small drops. And this helps to explain the negligible variation of small drops in the western part with increasing rain rate when R ≤ 20 mm h−1. Compared to the eastern part, the stronger cold rain processes in the western part conduce to the collision–breakup processes in strong rainstorm. Besides, the eastern part shows lower humidity above 0°C layer (Fig. 5) and the melting behavior of graupel can be suppressed by such lower humidity (e.g., Rasmussen and Heymsfield 1987), while the moister conditions in the western part would help ice-phase particles melt, resulting in more efficient collision–breakup processes for raindrops. Accordingly, more small drops would be produced and that would help explain the large sum of small drops in the western part when R > 20 mm h−1. Therefore, the dominant microphysical processes in the western part vary with increasing rain rate.
The above findings suggest different microphysical mechanisms within the EASM rainband. To further understand the distinct variability of DSDs, the observation samples from the western part and the eastern part, according to the particle diameter, are stratified into six classes and the diameter values corresponding to each size classes are shown in Fig. 7.
Relative contribution of each size classes to (a) total drop concentration Nt (m−3) and (b) rain rate R (mm h−1) over two parts of the EASM rainband.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Relative contribution of each size classes to (a) total drop concentration Nt (m−3) and (b) rain rate R (mm h−1) over two parts of the EASM rainband.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Relative contribution of each size classes to (a) total drop concentration Nt (m−3) and (b) rain rate R (mm h−1) over two parts of the EASM rainband.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
It is obviously shown in Fig. 7 that the small drops dominate the number concentration (>72%) for the total dataset in the EASM rainband and contribute to about 27% of rain rate in the eastern part. Moreover, the small and midsize drops (D < 4 mm) contribute more than 99% of the number concentration, which indicates the small and midsize drops jointly dominate the precipitation during the monsoon. Compared to the eastern part, the contribution of midsize drops to number concentration as well as rain rate is both larger in the western part due to the significant coalescence and breakup processes, which has been explained in the above analysis of average drop spectra. Compared to the western part, the contribution of small drops to number concentration or rain rate is both larger in the eastern part, which indicates a pronounced effect of small drops on the total precipitation in the eastern part. This could be partly attributed to the stronger collision–breakup process caused by larger wind velocity (Fig. 5a) in the eastern part, which further reflected the typical baroclinicity in the eastern part (Chen and Chang 1980). Similar result was also demonstrated by Testik and Pei (2017) that the raindrop collisional breakup process is expected to intensify with an increase in wind speeds, which have a statistically significant effect of increasing the number of small drops and decreasing the number of large drops.
4. DSD parameters
a. Statistical properties
The DSD samples in two parts of the EASM rainband derived from Parsivel2 are further used to derive gamma model via the TMF method adopted by Vivekanandan et al. (2004). Table 3 presents the statistical values of obtained gamma parameters (N0, µ and Λ) in two parts, which indicates all larger mean value of three parameters in the western part. The larger µ in the western part could be attributed to the significant coalescence and breakup processes, both processes acting together increase the concentration of midsize drops in the western part (Fig. 7) and thereby result in an substantial increase of µ value (Rosenfeld and Ulbrich 2003). It is also reported by Rosenfeld and Ulbrich (2003) that stronger coalescence process decreases both N0 and Λ, while stronger breakup process increases both N0 and Λ. The larger N0 and Λ in the western part may indicate that both processes are significant with breakup process in predominance.
Statistical results of gamma model parameters in two parts of the EASM rainband. SD and SK represent standard deviation and skewness, respectively.
Figure 8 shows the coefficients µ and Λ with rain rate. The values of µ and Λ tend to decrease with increasing rain rate. While rain rate is larger than 25 mm h−1, the PDF indicates that µ and Λ hold steady at about 6.0 and 7.0 mm−1 for the western part and 5.0 and 6.0 mm−1 for the eastern part, respectively. Notably, the variation of µ and Λ are both marked once rain rate is lower than 10 mm h−1, which indicates the complicated microphysical mechanism of stratiform rain during EASM, where the cold cloud process, the warm cloud process, melting, and evaporation all have a significant impact on rain formation; gradually, the range of variation decreases with increasing rain rate and becomes more uniform at larger rain rates. The rain rate of 10 mm h−1 can be used as a reference for precipitation categorization. Tokay and Short (1996) noted that a rain rate greater than 10 mm h−1 is mostly determined by convective clouds. Therefore, the rainfall differences in rain rates below and above 10 mm h−1 may indicate the DSDs variability among different types of monsoon rain, which further helps us to categorize the precipitation into two types. Here, 1-min samples with R > 10 mm h−1 are considered convective rainfall (also in Wu et al. 2019a). As a result, the western part and the eastern part consist 83% (17%), 92% (8%) of stratiform (convective) rainfall samples, respectively.
Scatterplot of μ (dimensionless) and Λ (mm) vs R (mm h−1) for each 1-min DSD observed in two parts of the EASM rainband. The PDF of two parameters observed at two parts are given in each panel as well.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplot of μ (dimensionless) and Λ (mm) vs R (mm h−1) for each 1-min DSD observed in two parts of the EASM rainband. The PDF of two parameters observed at two parts are given in each panel as well.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplot of μ (dimensionless) and Λ (mm) vs R (mm h−1) for each 1-min DSD observed in two parts of the EASM rainband. The PDF of two parameters observed at two parts are given in each panel as well.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
b. Distribution of Dm and Nw
Previous studies (e.g., Bringi et al. 2003; Chen 2009; Chen et al. 2013) showed that the behavior of DSD parameters was distinctly different for the maritime and continental convective storms. Bringi et al. (2003) used a threshold of σR = 1.5 mm h−1 for the classification of stratiform and convective rain and found that maritime convective rain is characterized by larger log10(Nw) and smaller Dm, while continental convective rain is characterized by smaller log10(Nw) and larger Dm. To further discuss the DSD parameters for the EASM rainband, we also obtained log10(Nw) and Dm of two classified rain types. The results from two classification methods including the method proposed by Bringi et al. (2003) as well as the method used in section 4a are both presented in Table 4 for comparison. It is notable that the method of Bringi et al. (2003) yields larger Dm and smaller log10(Nw) in the western part, and slightly smaller log10(Nw) in the eastern part. Apparently, different rainfall classification methods could influence the DSD parameters to some extent.
Derived parameters from two different rainfall classification methods in Bringi et al. (2003) and Wu et al. (2019a).
The observation of convective rain in Bringi et al. (2003) are presented with two black boxes in Fig. 9. For comparison, the log10(Nw) and Dm obtained for the EASM rainband by using the same rainfall classification methods are also shown in Fig. 9. Note that the convective precipitation in the western part is identified closer to maritime-like than continental-like convective precipitation as defined by Bringi et al. (2003), while the convective precipitation in the eastern part is identified between maritime-like and continental-like.
Distribution of log10(Nw) and Dm for convective precipitation, stratiform precipitation, and total precipitation in two parts of EASM rainband (along with ±1 standard deviation). The black diamonds indicate comparative studied regions. The two outlined rectangles correspond to the maritime and continental convective clusters reported by Bringi et al. (2003). The black dashed line indicates the characteristics of stratiform rain (Bringi et al. 2003). The letters C, S, and T denote the results from convective precipitation, stratiform precipitation, and total precipitation, respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Distribution of log10(Nw) and Dm for convective precipitation, stratiform precipitation, and total precipitation in two parts of EASM rainband (along with ±1 standard deviation). The black diamonds indicate comparative studied regions. The two outlined rectangles correspond to the maritime and continental convective clusters reported by Bringi et al. (2003). The black dashed line indicates the characteristics of stratiform rain (Bringi et al. 2003). The letters C, S, and T denote the results from convective precipitation, stratiform precipitation, and total precipitation, respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Distribution of log10(Nw) and Dm for convective precipitation, stratiform precipitation, and total precipitation in two parts of EASM rainband (along with ±1 standard deviation). The black diamonds indicate comparative studied regions. The two outlined rectangles correspond to the maritime and continental convective clusters reported by Bringi et al. (2003). The black dashed line indicates the characteristics of stratiform rain (Bringi et al. 2003). The letters C, S, and T denote the results from convective precipitation, stratiform precipitation, and total precipitation, respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The convective precipitation in the western part (land area) shows an character of “maritime-like,” which could be related to the abundant moisture transported from tropical ocean during EASM. Figure 10 shows average horizontal moisture flux at 925-hPa calculated from reanalysis data during the EASM season. Moisture fluxes from the tropical Indian Ocean and North Pacific Ocean meet and turn northward around the South China Sea (SCS), then converge into the EASM rainband to sustain copious precipitation. Interestingly, it is noted that the Dm(Nw) is even smaller (larger) in the convective rain of the western part than the eastern part, which means the convective precipitation in the western part is closer to maritime-like than the eastern part based on the definition in Bringi et al. (2003). We believe this is because of the following two reasons: first, the vapor transport from tropical oceans to the western part of the rainband increases its humidity and thereby inhibits its evaporation process, causing more small drops being preserved, thus the Dm decreases; second, as we analyzed in section 4a, the predominant breakup process in heavy rain of the western part may lead to an decrease of large drops while an significant increase of small drops, thus the Dm decreases, too. Therefore, the Dm is even smaller in the convective rain of the western part than the eastern part. We think it a very interesting and meaningful finding, which indicates that the identification method of the “maritime-like” and “continental-like” precipitation in Bringi et al. (2003) may have some obvious limitations, especially when it comes to the precipitation of some typical weather systems like EASM.
The average horizontal moisture flux at 925 hPa (g kg m s−1) calculated from reanalysis data.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The average horizontal moisture flux at 925 hPa (g kg m s−1) calculated from reanalysis data.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The average horizontal moisture flux at 925 hPa (g kg m s−1) calculated from reanalysis data.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
On the other hand, the convective precipitation of the eastern part (subtropical ocean) is not identified “maritime-like,” but between “maritime-like” and “continental-like,” and is characterized by Dm = 1.95 and log10(Nw) = 3.70. Also, recent research work of oceanic DSD by Protat et al. (2019) suggested that the averaged Dm and Nw values (indicated by angle brackets) of convective rain over the global oceans are approximately characterized by ⟨Dm⟩ = 1.8–2.2 mm and log10⟨Nw⟩ = 3.0–4.0. Both of this research and the research by Protat et al. (2019) indicate that the definition of “maritime-like” precipitation in Bringi et al. (2003) shows obvious limitations due to the lack of rainfall observations at different latitude ocean areas. As is known that the definition of “maritime-like” convective precipitation proposed by Bringi et al. (2003) is based on the observations from tropical island and sea, and a few coastal areas in subtropics, which includes no subtropical ocean area or higher-latitude oceans. Whereas oceans at different latitudes could have significant DSD variability (Protat et al. 2019). The above analysis inspires us to propose another rainfall identification method that better represents the DSD variability globally, we leave that in our future works.
c. σM–Dm relations
Previous studies have shown that gamma function parameters (N0, µ and Λ) are not statistically independent but are correlated, with high Pearson correlation coefficients (Ulbrich 1983; Ulbrich and Atlas 1985; Chandrasekar and Bringi 1987; Illingworth and Blackman 2002; Moisseev and Chandrasekar 2007). Therefore, mathematical artifacts will appear in relationships once a DSD is assumed to follow a gamma function (Williams et al. 2014). Without an assumption of a gamma-shaped DSD a priori, raindrop mass-weighted mean diameter Dm and mass spectrum standard deviation σM are used to describe the DSD shape, and relationships between Dm and σM are investigated (Petersen and Jensen 2012; Williams et al. 2014; Thurai et al. 2014).
A relation of
While for the eastern part is given by
The σM–Dm two-dimensional PDF distributions and fitting results based on Parsivel disdrometer observations in two parts of the EASM rainband. The resolution of PDF is 0.1 mm × 0.1 mm and the size of the color mark represents the percentage. The dashed lines in corresponding color represent the empirical σM–Dm relations from Petersen and Jensen (2012), Thurai et al. (2014), and Williams et al. (2014), respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The σM–Dm two-dimensional PDF distributions and fitting results based on Parsivel disdrometer observations in two parts of the EASM rainband. The resolution of PDF is 0.1 mm × 0.1 mm and the size of the color mark represents the percentage. The dashed lines in corresponding color represent the empirical σM–Dm relations from Petersen and Jensen (2012), Thurai et al. (2014), and Williams et al. (2014), respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
The σM–Dm two-dimensional PDF distributions and fitting results based on Parsivel disdrometer observations in two parts of the EASM rainband. The resolution of PDF is 0.1 mm × 0.1 mm and the size of the color mark represents the percentage. The dashed lines in corresponding color represent the empirical σM–Dm relations from Petersen and Jensen (2012), Thurai et al. (2014), and Williams et al. (2014), respectively.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Compared to Oklahoma and Alabama on the same latitude (Petersen and Jensen 2012; Thurai et al. 2014; Williams et al. 2014), both the western part and the eastern part have smaller exponents for σM–Dm relations, causing smaller σM with an increase of Dm. And due to that, the spectral width of intense convective precipitation (usually having large Dm) would be narrower in monsoon rainfall than the other regions on the same latitude, which reflects the microphysical variability of precipitation in different climate regimes. Comparing the western part with the eastern part of the same EASM rainband, the western part has larger exponent for σM–Dm relations, resulting in broader spectral width with more sufficient large droplets of the convective precipitation in the western part. And this further implies the microphysics variability within the EASM rainband.
5. GPM–Parsivel comparison
a. Validation of GPM DPR products
With its onboard DPR and the Microwave Imager (GMI), the GPM Core Observatory satellite provides a global surveillance of precipitation since 2014 (e.g., Skofronick-Jackson et al. 2017). As a successor of Tropical Rainfall Measuring Mission (TRMM), GPM satellite faces its challenge in well behaving not only at tropics but more importantly over midlatitudes. It is therefore significant to evaluate how accurately GPM can determine precipitation over midlatitude EASM regions. To assess and improve the ability of the GPM satellite, we have compared concurrent DPR measurements with Parsivel observations in terms of attenuation-corrected radar reflectivity and rain rate at estimated surface.
Following Wu et al. (2019b), we choose DPR_MS product to obtain the satellite-observational variable of rain rate, while we choose Ku-PR and Ka-HS products to obtain the effective reflectivity factor at both Ku and Ka bands due to the advantages of each product. For validation, the Parsivel2 observational data are used to calculate the Ku-band (Ka-band) effective radar reflectivity factor ZKu(ZKa) as well as rain rate over two parts of the EASM rainband using the method in section 2. Accordingly, the DFR can also be obtained via Eq. (3). The relationship between DFR and rainfall helps improve our understanding of microphysics regarding precipitation retrieval algorithm (Zhang and Fu 2018; Wu et al. 2019b). Thus, the statistical relationship between DFR and rain rate R in both the western part and the eastern part of the EASM rainband is analyzed and compared given that the results from GPM and Parsivel2 observations are close to each other both spatially and temporally.
Figure 12 shows the distribution of DFR and rain rate at estimated surface from both GPM DPR measurements (Fig. 12, left) and disdrometer derivations (Fig. 12, right) for the western part (Fig. 12, red) and the eastern part (Fig. 12, blue). Similar to the results in Wu et al. (2019b), when high rain rates occur, the DFR may achieve a state of equilibrium with the effective reflectivity of two bands correlated linearly. Specifically, the DFR values from DPR measurements and DSD derivations approach to approximately 0.0 and 0.5 dB in the western part, while 0.1 and 0.5 dB in the eastern part, respectively. The DPR–Parsivel normalized mean bias (BIAS) is estimated with the following equation:
where DPR means GPM DPR observational data and DSD means Parsivel measurement values. The comparison results (BIAS = 1 for western, BIAS = 0.8 for eastern) indicate that GPM underestimate the DFR more in the western part than in the eastern part, which suggests that GPM may have better performance over sea than land.
Scatterplots of DFR and rain rate from (left) DPR observations and (right) DSD derivations over two parts of the EASM rainband. The PDF of DFR value are also given in each panel.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplots of DFR and rain rate from (left) DPR observations and (right) DSD derivations over two parts of the EASM rainband. The PDF of DFR value are also given in each panel.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplots of DFR and rain rate from (left) DPR observations and (right) DSD derivations over two parts of the EASM rainband. The PDF of DFR value are also given in each panel.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
It is additionally suggested in Fig. 12 that compared with Parsivel observations, GPM DPR measurements show distinct lower-frequency ratio of strong precipitation samples. Besides the attenuation caused by instrumental effect or multiple scattering, the possible reason causing such discrepancy could also come from two important aspects: one is the influence of nonuniform beam filling (NUBF) on GPM DPR estimates of path attenuation by surface reference technique (SRT), which can be clearly noticed for heavy precipitation (Seto et al. 2015). And because of the comparatively stronger convective activity and broader DSD spectral width over the western part than the eastern part, DPR would underestimate more in the western part. The other can be related to the relatively complicated topography over land area, causing a poorer performance of SRT on land than sea.
b. Improvement of GPM DPR retrieval algorithm
The development and refinement of physically based GPM DPR algorithm requires an improved understanding of precipitation microphysics. To fulfill the need for better converting the measurements of GPM DPR to rainfall amounts in midlatitude EASM region, Parsivel2 disdrometers are preliminarily employed over this region to develop the optimized constraints for GPM DPR retrieval algorithm.
In the latest GPM DPR product, DSD parameters (i.e., mass-weighted diameter Dm and generalized intercept parameter Nw) are retrieved at each range bin by utilizing a dual-frequency algorithm, where DFR are computed and expressed as a function of Dm (Iguchi et al. 2018). It is, however, difficult to retrieve Dm uniquely due to the well-known “dual value” problem already reported by many researches (e.g., Meagher and Haddad 2006; Wu et al. 2019a,b). That is, at a fixed µ, DFR increases monotonically with Dm only given Dm > 1 mm, whereas there exit dual solutions for Dm when DFR is a negative value (Meagher and Haddad 2006). This is because the DFR of a single drop is not a monotonic increasing function of drop size, but decreases first at small drop diameter before increasing at larger drop diameters (Munchak and Tokay 2008). Thus, a DSD consisting of small drops may have the same DFR value as a DSD including larger drops. To prevent the dual-value phenomenon, we followed the method in Wu et al. (2019b), by utilizing simply the effective radar reflectivities of Ka and Ku bands estimated from disdrometers to obtain their statistical relationship with Dm.
The scatterplots and fitted curves of Dm–ZKu (ZKa) and log10(Nw)–Dm are obtained from disdrometer observations in two parts of the EASM rainband and depicted in Fig. 13. Similar to the findings in Wu et al. (2019b), we also notice a highly correlated monotonic increasing relationship between Dm and ZKu (ZKa) and a nonlinear inverse relationship between Dm and log10(Nw) from Fig. 13. Using a least squares method, the quadratic polynomial relationships of Dm–ZKu (ZKa) and log10(Nw)–Dm are fitted and shown in Table 5. Based on these relationships, Dm can be first derived with given ZKu or ZKa, then Nw can be estimated by substituting Dm into log10(Nw)–Dm relationship. According to the normalized gamma distribution function already presented in section 2b, the DSD can be determined from the derived Dm and Nw after constraining µ value, and once DSD is derived, the corresponding precipitation variables like rain rate can be calculated eventually. Note that the µ value should be constrained by different rain types, herein µ = 5.8 (5.0) for convective rain and µ = 5.9 (4.7) for stratiform rain in the western (eastern) part as shown in Table 3.
Scatterplots of (a),(b) Dm (mm) and ZKu (dBZ), (c),(d) Dm (mm) and ZKa (dBZ), and (e),(f) log10(Nw) (mm−1 m−3) and Dm (mm) for EASM rainband for (left) the western part and (right)the eastern part. The overlaid red lines represent the fitted curves.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplots of (a),(b) Dm (mm) and ZKu (dBZ), (c),(d) Dm (mm) and ZKa (dBZ), and (e),(f) log10(Nw) (mm−1 m−3) and Dm (mm) for EASM rainband for (left) the western part and (right)the eastern part. The overlaid red lines represent the fitted curves.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplots of (a),(b) Dm (mm) and ZKu (dBZ), (c),(d) Dm (mm) and ZKa (dBZ), and (e),(f) log10(Nw) (mm−1 m−3) and Dm (mm) for EASM rainband for (left) the western part and (right)the eastern part. The overlaid red lines represent the fitted curves.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Second-degree polynomial relations of Dm–ZKu, Dm–ZKa, and log10(Nw)–Dm derived in two parts of the EASM rainband.
The performance of the derived new relations is evaluated by plotting the rain-rate values from both DPR and Parsivel together in two parts of the EASM rainband as shown in Fig. 14. It is notable that the green dots fit well with black line at low rain rates, while the dots gradually deviate from the line with an increase in rain rate. That indicates Ka band shows better performance at lower rain rates. Quite different from that, Ku band (orange dots) has better performance at higher rain rates. Thus, instead of using DFR for rain retrieval, we propose to combine two bands by estimating light rains with Ka-band relation (Dm–ZKa) while heavy rains with Ku-band relation (Dm–ZKu), which is preliminarily shown to have an advantage without any “dual value” problem. Meanwhile, since the performance of Ka and Ku differentiated by low and high rainfall intensity and rainfall microphysics shows large differences between convective and stratiform rain types. The relations between Dm and ZKu(ZKa) are also derived for different rain types as shown in Table 6. Similarly, it is recommended to use Dm–ZKu for the retrieval of convective rain while Dm–ZKa for the retrieval of stratiform rain in both parts of the EASM rainband.
Scatterplots of GPM estimated RDPR vs Parsivel-observed RDSD for two parts of EASM rainband. The green crosses indicate results from Ka band. The orange circles indicate results from Ku band.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplots of GPM estimated RDPR vs Parsivel-observed RDSD for two parts of EASM rainband. The green crosses indicate results from Ka band. The orange circles indicate results from Ku band.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
Scatterplots of GPM estimated RDPR vs Parsivel-observed RDSD for two parts of EASM rainband. The green crosses indicate results from Ka band. The orange circles indicate results from Ku band.
Citation: Journal of Atmospheric and Oceanic Technology 37, 7; 10.1175/JTECH-D-19-0059.1
6. Summary and conclusions
Utilizing measurements from OTT Parsivel2 disdrometers, as well as multisource satellite observations, we compared the DSD characteristics in the western and eastern parts of the EASM rainband and discussed the recent GPM rainfall retrievals in the EASM season. The study mainly reveals:
Notably, the seasonal migration of the EASM rainband was not investigated but worth our future research with effective data collected. Besides, the GPM validation results are obtained during the EASM season, more observations are required to achieve robust validations beyond a specific meteorological regime. Although it is yet to be investigated, the performance of the other used instruments and products (FY-2E, MODIS, GPCP, and ERA-Interim) under different surface condition are likely to be significant as well. In addition, more detailed evaluation work on the performance of GPM precipitation product will be carried out in future research, like in terms of different rain types, or even different rain-rate classes. The aerosol effects on DSDs also need further investigation, since the effect of the other factors cannot be excluded. From the rainfall identification results in both the western part and the eastern part, it should be noted that the classification method of Bringi et al. (2003) is easily affected by climate conditions (humidity, air pollution, etc.) and geophysical locations (tropics, midlatitudes, etc.). Hence an improved identification method of “maritime-like” and “continental-like” convective precipitation is required with various climate conditions under consideration as well as more marine precipitation data collected.
Acknowledgments
We are thankful for the support from the National Key Research and Development Program of China (2018YFC1507304), the National Nature Science Foundation of China (41975066, 41865009) and the Beijing Open Research Fund for Nanjing Joint Center of Atmospheric Research (NJCAR2018ZD03). Thanks also due to the editors and three anonymous reviewers for their critical and constructive comments, as well as the photographer Jinfeng Ding for providing the photo at sea. We acknowledge FY-2E (CMA), ECMWF, GPCP, GPM and MODIS (NASA) for providing the data. Parsivel2 disdrometers and in situ radiosonde data for the monsoon rainband research can be obtained online (https://doi.org/10.5281/zenodo.2549596).
REFERENCES
Albrecht, B. A., 1989: Aerosols, cloud microphysics and fractional cloudiness. Science, 245, 1227–1230, https://doi.org/10.1126/science.245.4923.1227.
Andreae, M. O., D. Rosenfeld, P. Artaxo, A. A. Costa, G. P. Frank, K. M. Longo, and M. A. F. Silva-Dias, 2004: Smoking rain clouds over the Amazon. Science, 303, 1337–1342, https://doi.org/10.1126/science.1092779.
Breon, F. M., D. Tanre, and S. Generoso, 2002: Aerosol effect on cloud droplet size monitored from satellite. Science, 295, 834–838, https://doi.org/10.1126/science.1066434.
Bringi, V. N., V. Chandrasekar, J. Hubbert, E. Gorgucci, W. L. Randeu, and M. Schoenhuber, 2003: Raindrop size distribution in different climatic regimes from disdrometer and dual-polarized radar analysis. J. Atmos. Sci., 60, 354–365, https://doi.org/10.1175/1520-0469(2003)060<0354:RSDIDC>2.0.CO;2.
Bringi, V. N., M. Thurai, K. Nakagawa, G. J. Huang, T. Kobayashi, A. Adachi, H. Hanado, and S. Sekizawa, 2006: Rainfall estimation from C-band polarimetric radar in Okinawa, Japan: Comparisons with 2D-video disdrometer and 400 MHz wind profiler. J. Meteor. Soc. Japan, 84, 705–724, https://doi.org/10.2151/jmsj.84.705.
Chakravarty, K., and P. E. Raj, 2013: Raindrop size distributions and their association with characteristics of clouds and precipitation during monsoon and post-monsoon periods over a tropical Indian station. Atmos. Res., 124, 181–189, https://doi.org/10.1016/j.atmosres.2013.01.005.
Chandrasekar, V., and V. N. Bringi, 1987: Simulation of radar reflectivity and surface measurements of rainfall. J. Atmos. Oceanic Technol., 4, 464–478, https://doi.org/10.1175/1520-0426(1987)004<0464:sorras>2.0.co;2.
Chen, B., J. Yang, and J. Pu, 2013: Statistical characteristics of raindrop size distribution in the meiyu season observed in eastern China. J. Meteor. Soc. Japan, 91, 215–227, https://doi.org/10.2151/jmsj.2013-208.
Chen, G., and Coauthors, 2017: Improving polarimetric C-band radar rainfall estimation with two-dimensional video disdrometer observations in eastern China. J. Hydrometeor., 18, 1375–1391, https://doi.org/10.1175/JHM-D-16-0215.1.
Chen, T.-J. G., and C.-P. Chang, 1980: The structure and vorticity budget of an early summer monsoon trough (mei-yu) over southeastern China and Japan. Mon. Wea. Rev., 108, 942–953, https://doi.org/10.1175/1520-0493(1980)108<0942:TSAVBO>2.0.CO;2.
Chen, Y., 2009: The characteristic of drop size distribution during SoWMEX (in Chinese). M.S. thesis, Dept. of Atmospheric Sciences, National Central University, 68 pp.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, https://doi.org/10.1002/qj.828.
Ding, Y., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117–142, https://doi.org/10.1007/s00703-005-0125-z.
Huffman, G. J., D. T. Bolvin, and R. F. Adler, 2012: GPCP version 1.2 1-degree daily (1DD) precipitation data set. NCDC, accessed 12 January 2019, https://rda.ucar.edu/datasets/ds728.3.
Iguchi, T., and Coauthors, 2018: GPM/DPR level-2. Algorithm Theoretical Basis Doc., 127 pp., https://gpm.nasa.gov/sites/default/files/document_files/ATBD_DPR_201811_with_Appendix3b_0.pdf.
Illingworth, A. J., and T. M. Blackman, 2002: The need to represent raindrop size spectra as normalized gamma distributions for the interpretation of polarization radar observations. J. Appl. Meteor., 41, 286–297, https://doi.org/10.1175/1520-0450(2002)041<0286:tntrrs>2.0.co;2.
Jiang, Z., F. Huo, H. Ma, J. Song, and A. Dai, 2017: Impact of Chinese urbanization and aerosol emissions on the East Asian summer monsoon. J. Climate, 30, 1019–1039, https://doi.org/10.1175/JCLI-D-15-0593.1.
Kato, T., 2006: Structure of the band-shaped precipitation system inducing the heavy rainfall observed over northern Kyushu, Japan on 29 June 1999. J. Meteor. Soc. Japan, 84, 129–153, https://doi.org/10.2151/jmsj.84.129.
Kato, T., and K. Aranami, 2005: Formation factors of 2004 Niigata-Fukushima and Fukui heavy rainfalls and problems in the predictions using a cloud-resolving model. SOLA, 1, 1–4, https://doi.org/10.2151/sola.2005-001.
Kirankumar, N. V., T. N. Rao, B. Radhakrishna, and D. N. Rao, 2008: Statistical characteristics of raindrop size distribution in southwest monsoon season. J. Appl. Meteor. Climatol., 47, 576–590, https://doi.org/10.1175/2007JAMC1610.1.
Kozu, T., T. Shimomai, Z. Akramin, Marzuki, Y. Shibagaki, H. Hashiguchi, 2005: Intraseasonal variation of raindrop size distribution at Koto Tabang, West Sumatra, Indonesia. Geophys. Res. Lett., 32, L07803, https://doi.org/10.1029/2004GL022340.
Kuwano-Yoshida, A., B. Taguchi, and S. P. Xie, 2013: Baiu rainband termination in atmospheric and coupled atmosphere–ocean models. J. Climate, 26, 10 111–10 124, https://doi.org/10.1175/JCLI-D-13-00231.1.
Le, M., and V. Chandrasekar, 2013: Hydrometeor profile characterization method for dual-frequency precipitation radar onboard the GPM. IEEE Trans. Geosci. Remote Sens., 51, 3648–3658, https://doi.org/10.1109/TGRS.2012.2224352.
Liao, L., and R. Meneghini, 2019: Physical evaluation of GPM DPR single- and dual-wavelength algorithms. J. Atmos. Oceanic Technol., 36, 883–902, https://doi.org/10.1175/JTECH-D-18-0210.1.
Liao, L., R. Meneghini, and A. Tokay, 2014: Uncertainties of GPM DPR rain estimates caused by DSD parameterizations. J. Appl. Meteor. Climatol., 53, 2524–2537, https://doi.org/10.1175/JAMC-D-14-0003.1.
Liu, J., Y. Zheng, Z. Li, and M. Cribb, 2011: Analysis of cloud condensation nuclei properties at a polluted site in southeastern China during the AMF-China Campaign. J. Geophys. Res., 116, D00K35, https://doi.org/10.1029/2011JD016395.
Löffler-Mang, M., and J. Joss, 2000: An optical disdrometer for measuring size and velocity of hydrometeors. J. Atmos. Oceanic Technol., 17, 130–139, https://doi.org/10.1175/1520-0426(2000)017<0130:AODFMS>2.0.CO;2.
Maddox, R. A., 1980: Mesoscale convective complexes. Bull. Amer. Meteor. Soc., 61, 1374–1387, https://doi.org/10.1175/1520-0477(1980)061<1374:MCC>2.0.CO;2.
May, P. T., V. N. Bringi, and M. Thurai, 2011: Do we observe aerosol impacts on DSDs in strongly forced tropical thunderstorms? J. Atmos. Sci., 68, 1902–1910, https://doi.org/10.1175/2011JAS3617.1.
Meagher, J. P., and Z. S. Haddad, 2006: To what extent can raindrop size be determined by a multiple-frequency radar? J. Appl. Meteor. Climatol., 45, 529–536, https://doi.org/10.1175/JAM2344.1.
Moisseev, D. N., and V. Chandrasekar, 2007: Examination of the μ–Λ relation suggested for drop size distribution parameter. J. Atmos. Oceanic Technol., 24, 847–855, https://doi.org/10.1175/JTECH2010.1.
Munchak, S. J., and A. Tokay, 2008: Retrieval of raindrop size distribution from simulated dual-frequency radar measurements. J. Appl. Meteor. Climatol., 47, 223–239, https://doi.org/10.1175/2007JAMC1524.1.
Ninomiya, K., 2000: Large- and meso-α-scale characteristics of meiyu/baiu front associated with intense rainfalls in 1–0 July 1991. J. Meteor. Soc. Japan, 78, 141–157, https://doi.org/10.2151/jmsj1965.78.2_141.
Ninomiya, K., and Y. Shibagaki, 2003: Cloud system families in the meiyu-baiu front observed during 1-10 July 1991. J. Meteor. Soc. Japan, 81, 193–209, https://doi.org/10.2151/jmsj.81.193.
Ogura, Y., T. Asai, and K. Dohi, 1985: A case study of a heavy precipitation event along the baiu front in northern Kyushu, 23 July 1982: Nagasaki heavy rainfall. J. Meteor. Soc. Japan, 63, 883–900, https://doi.org/10.2151/jmsj1965.63.5_883.
Petersen, W., and M. Jensen, 2012: The NASA-GPM and DOE-ARM Midlatitude Continental Convective Clouds Experiment (MC3E). Int. J. Appl. Earth Obs., 24, 12–18.
Platnick, S., M. D. King, S. A. Ackerman, W. P. Menzel, B. A. Baum, J. C. Riedl, and R. A. Frey, 2003: The MODIS cloud products: Algorithms and examples from Terra. IEEE Trans. Geosci. Remote Sens., 41, 459–473, https://doi.org/10.1109/TGRS.2002.808301.
Protat, A., C. Klepp, V. Louf, W. A. Petersen, S. P. Alexander, A. Barros, J. Leinonen, and G. G. Mace, 2019: The latitudinal variability of oceanic rainfall properties and its implication for satellite retrievals: 1. Drop size distribution properties. J. Geophys. Res. Atmos., 124, 13 291–13 311, https://doi.org/10.1029/2019jd031010.
Radhakrishna, B., S. Satheesh, T. Narayana Rao, K. Saikranthi, and K. Sunilkumar, 2016: Assessment of DSDs of GPM-DPR with ground-based disdrometer at seasonal scale over Gadanki, India. J. Geophys. Res. Atmos., 121, 11 792–11 802, https://doi.org/10.1002/2015jd024628.
Ramanathan, V., P. J. Crutzen, J. T. Kiehl, and D. Rosenfeld, 2001: Aerosols, climate, and the hydrological cycle. Science, 294, 2119–2124, https://doi.org/10.1126/science.1064034.
Rasmussen, R. M., and A. J. Heymsfield, 1987: Melting and shedding of graupel and hail. Part II: Sensitivity study. J. Atmos. Sci., 44, 2764–2782, https://doi.org/10.1175/1520-0469(1987)044<2764:MASOGA>2.0.CO;2.
Remer, L. A., and Coauthors, 2005: The MODIS aerosol algorithm, products and validation. J. Atmos. Sci., 62, 947–973, https://doi.org/10.1175/JAS3385.1.
Rosenfeld, D., 1999: TRMM observed first direct evidence of smoke from forest fires inhibiting rainfall. Geophys. Res. Lett., 26, 3105–3108, https://doi.org/10.1029/1999GL006066.
Rosenfeld, D., and W. L. Woodley, 2000: Deep convective clouds with sustained supercooled liquid water down to −37.5°C. Nature, 405, 440–442, https://doi.org/10.1038/35013030.
Rosenfeld, D., and C. W. Ulbrich, 2003: Cloud microphysical properties, processes, and rainfall estimation opportunities. Radar and Atmospheric Science: A Collection of Essays in Honor of David Atlas, Meteor. Monogr., No. 52, Amer. Meteor. Soc., 237–258.
Sampe, T., and S. P. Xie, 2010: Large-scale dynamics of the meiyu-baiu rainband: Environmental forcing by the westerly jet. J. Climate, 23, 113–134, https://doi.org/10.1175/2009JCLI3128.1.
Seto, S., T. Iguchi, and T. Oki, 2013: 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., 51, 5239–5251, https://doi.org/10.1109/TGRS.2012.2231686.
Seto, S., T. Iguchi, T. Shimozuma, and S. Hayashi, 2015: NUBF correction methods for the GPM/DPR level-2 algorithms. 2015 Int. Geoscience and Remote Sensing Symp., Milan, Italy, IEEE, 2612–2614, https://doi.org/10.1109/IGARSS.2015.7326347.
Skofronick-Jackson, G., and Coauthors, 2017: The Global Precipitation Measurement (GPM) mission for science and society. Bull. Amer. Meteor. Soc., 98, 1679–1695, https://doi.org/10.1175/BAMS-D-15-00306.1.
Streets, D. G., C. Yu, Y. Wu, M. Chin, Z. Zhao, T. Hayasaka, and G. Shi, 2008: Aerosol trends over China, 1980–2000. Atmos. Res., 88, 174–182, https://doi.org/10.1016/j.atmosres.2007.10.016.
Testik, F. Y., and B. Pei, 2017: Wind effects on the shape of raindrop size distribution. J. Hydrometeor., 18, 1285–1303, https://doi.org/10.1175/JHM-D-16-0211.1.
Thurai, M., C. R. William, and V. N. Bring, 2014: Examining the correlations between drop size distribution parameters using data from two side-by-side 2D-video disdrometers. Atmos. Res., 144, 95–110, https://doi.org/10.1016/j.atmosres.2014.01.002.
Tokay, A., and D. A. Short, 1996: Evidence from tropical raindrop spectra of the origin of rain from stratiform versus convective clouds. J. Appl. Meteor., 35, 355–371, https://doi.org/10.1175/1520-0450(1996)035<0355:EFTRSO>2.0.CO;2.
Tokay, A., and P. G. Bashor, 2010: An experimental study of small-scale variability of raindrop size distribution. J. Appl. Meteor. Climatol., 49, 2348–2365, https://doi.org/10.1175/2010JAMC2269.1.
Tokay, A., D. B. Wolff, and W. A. Petersen, 2014: Evaluation of the new version of the laser-optical disdrometer, OTT Parsivel2. J. Atmos. Oceanic Technol., 31, 1276–1288, https://doi.org/10.1175/JTECH-D-13-00174.1.
Twomey, S., 1977: The influence of pollution on the shortwave albedo of clouds. J. Atmos. Sci., 34, 1149–1152, https://doi.org/10.1175/1520-0469(1977)034<1149:TIOPOT>2.0.CO;2.
Ulbrich, C. W., 1983: Natural variations in the analytical form of the drop size distribution. J. Climate Appl. Meteor., 22, 1764–1775, https://doi.org/10.1175/1520-0450(1983)022<1764:NVITAF>2.0.CO;2.
Ulbrich, C. W., and D. Atlas, 1985: Extinction of visible and infrared radiation in rain: Comparison of theory and experiment. J. Atmos. Oceanic Technol., 2, 331–339, https://doi.org/10.1175/1520-0426(1985)002<0331:EOVAIR>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, https://doi.org/10.1175/1520-0450(2004)043<0217:PREBOA>2.0.CO;2.
Wallace, J. M., and P. V. Hobbs, 1977: Atmospheric Science: An Introductory Survey. Academic Press, 467 pp.
Wen, L., K. Zhao, G. Zhang, M. Xue, B. Zhou, S. Liu, and X. Chen, 2016: Statistical characteristics of raindrop size distributions observed in East China during the Asian summer monsoon season using 2-D video disdrometer and Micro Rain Radar data. J. Geophys. Res. Atmos., 121, 2265–2282, https://doi.org/10.1002/2015jd024160.
Wen, L., K. Zhao, G. Zhang, S. Liu, and G. Chen, 2017: Impacts of instrument limitations on estimated raindrop size distribution, radar parameters and model microphysics during mei-yu season in East China. J. Atmos. Oceanic Technol., 34, 1021–1037, https://doi.org/10.1175/JTECH-D-16-0225.1.
Williams, C. R., and Coauthors, 2014: Describing the shape of raindrop size distributions using uncorrelated raindrop mass spectrum parameters. J. Appl. Meteor. Climatol., 53, 1282–1296, https://doi.org/10.1175/JAMC-D-13-076.1.
Wu, Z., Y. Zhang, L. Zhang, H. Lei, Y. Xie, L. Wen, and J. Yang, 2019a: Characteristics of summer season raindrop size distribution in three typical regions of western Pacific. J. Geophys. Res. Atmos., 124, 4054–4073, https://doi.org/10.1029/2018jd029194.
Wu, Z., Y. Zhang, L. Zhang, X. Hao, H. Lei, and H. Zheng, 2019b: Validation of GPM precipitation products by comparison with ground-based Parsivel disdrometers over Jianghuai region. Water, 11, 1260, https://doi.org/10.3390/w11061260.
Yoshizaki, M., T. Kato, Y. Tanaka, H. Takayama, Y. Shoji, and H. Seko, 2000: Analytical and numerical study of the 26 June 1998 orographic rainband observed in western Kyushu, Japan. J. Meteor. Soc. Japan, 78, 835–856, https://doi.org/10.2151/jmsj1965.78.6_835.
Zhang, A., and Y. Fu, 2018: The structural characteristics of precipitation cases detected by dual-frequency radar of GPM satellite (in Chinese). Chin. J. Atmos. Sci., 42, 33–51, https://doi.org/10.3878/j.issn.1006-9895.1705.16220.
Zhang, H., Y. Zhang, H. He, Y. Xie, and Q. Zeng, 2017: Comparison of raindrop size distributions in a midlatitude continental squall line during different stages as measured by Parsivel over East China. J. Appl. Meteor. Climatol., 56, 2097–2111, https://doi.org/10.1175/JAMC-D-16-0336.1.
Zhang, Y., 2018: Characteristics of summer season raindrop size distribution in western Pacific (data source). Zenodo, accessed 18 January 2019, http://doi.org/10.5281/zenodo.1465082.
Zhang, Y., Z. Wu, L. Zhang, Y. Xie, H. Lei, H. Zheng, and X. Ma, 2019: Modelling the effects of aerosol on mei-yu frontal precipitation and physical processes. Appl. Sci., 9, 3802, https://doi.org/10.3390/app9183802.
