Abstract

A quasi-stationary front, called the baiu front, often appears during the early-summer rainy season in East Asia (baiu in Japan). The present study examines how precipitation characteristics during the baiu season are determined by the large-scale environment, using satellite observation three-dimensional precipitation data. Emphasis is placed on the effect of subtropical jet (STJ) and lower-tropospheric convective instability (LCI).

A rainband appears together with a deep moisture convergence to the south of the STJ. Two types of mesoscale rainfall events (REs; contiguous rainfall areas), which are grouped by the stratiform precipitation ratio (SPR; stratiform precipitation over total precipitation), are identified: moderately stratiform REs (SPR of 0%–80%) representing tropical organized precipitation systems and highly stratiform REs (SPR of 80%–100%) representing midlatitude precipitation systems associated with extratropical cyclones. As the STJ becomes strong, rainfall from both types of mesoscale precipitation systems increases, with a distinct eastward extension of a midtropospheric moist region. In contrast, small systems appear regardless of the STJ, with high dependency on the LCI.

The results indicate that the STJ plays a role in moistening the midtroposphere owing to ascent associated with secondary circulation to the south of the STJ, producing environments favorable for organized precipitation systems in the southern part of the rainband. The horizontal moisture flux convergence may also contribute to precipitation just along the STJ. On the other hand, the LCI plays a role in generating shallow convection. In high-LCI conditions, deep convection can occur without the aid of mesoscale organization.

1. Introduction

The early-summer East Asian rainy season is called the baiu season in Japan and the mei-yu and changma seasons in China and South Korea, respectively. In this season, a quasi-stationary front, called the baiu front, often appears associated with a zonally elongated belt of cloud. The baiu front forms around the boundary between midlatitude and subtropical air masses and is characterized by a strong gradient of equivalent potential temperature θe (Ninomiya 1984; Ninomiya and Akiyama 1992). Additionally, the baiu front has a multiscale structure (Ninomiya and Akiyama 1992; Ninomiya and Shibagaki 2007). In one reported case, synoptic-scale disturbances and meso-α-scale disturbances coexisted, resulting in organized baiu frontal cloud systems having a length of ~5000 km (Akiyama 1990; Ninomiya and Akiyama 1992).

The baiu front is associated with not only appreciable precipitation but also a wide variety of precipitation characteristics (e.g., Akiyama 1978; Takayabu and Hikosaka 2009; Xu et al. 2009; Xu and Zipser 2011; Yokoyama et al. 2014; Park et al. 2014). Ninomiya (1978) investigated a case of heavy rainfall associated with a frontal depression in the baiu season. Using Tropical Rainfall Measuring Mission (TRMM) satellite-borne Precipitation Radar (PR) data, Takayabu and Hikosaka (2009) showed an increased contribution of tall convective precipitation to total rainfall immediately before the withdrawal of the baiu season, with an increased θe near the surface. Yokoyama et al. (2014) detected baiu fronts according to the large-scale environment and examined differences in precipitation characteristics between the southern and northern sides of the baiu front, which differ from one another in terms of atmospheric stratification. It was thus shown that precipitation characteristics drastically change in a meridional direction across the baiu front. Precipitation characteristics depend on what kinds of large-scale environments are dominant.

According to its characteristics and amount, precipitation associated with the baiu front strongly affects society in East Asia. There is thus an urgent need to estimate changes in precipitation during the baiu season in a future climate as accurately as possible. Current climate models have a poor ability to simulate detailed precipitation characteristics, which makes it difficult to predict future changes in precipitation characteristics. Meanwhile, large-scale environments, such as atmospheric circulation fields, are considered to be expressed by the models in a more appropriate manner than is the case for precipitation. The present study searches for physical relationships between large-scale environments and precipitation characteristics. If we assume that the physical relationships for the current climate do not significantly differ from those for a future climate, we can apply our findings to a future climate and thus estimate future changes in precipitation characteristics.

In terms of the environment, the importance of the low-level inflow of high-θe air to the baiu front and thus enhanced convective instability for active convection and heavy rainfall has been noted in many studies (e.g., Kodama 1992; Kato et al. 2003; Akiyama 1979). Meanwhile, studies have emphasized effects of the subtropical jet (STJ). Kodama (1993) indicated that the presence of the STJ in addition to a prevailing low-level northward flow along the western peripheries of the subtropical anticyclones is important for the formation of the baiu/mei-yu front. Sampe and Xie (2010) showed correspondence between mean ascending motion and warm horizontal temperature advection in the midtroposphere along the baiu/mei-yu rainband throughout summer, and they hypothesized that riding on the westerlies, warm air from the eastern flank of the Tibetan Plateau rises on the sloping isentropic surface and triggers convection.

Recent studies have shed light on the effects of mid- to upper-tropospheric circulations on East Asian precipitation. Horinouchi (2014) revealed that upper-tropospheric Rossby waves strongly affect the synoptic variability of lower-tropospheric moisture transport and precipitation over summertime East Asia and the northwestern Pacific. Employing Q vector analysis, he showed that a precipitation band appears where upwelling is dynamically forced by secondary circulation. A further study by Horinouchi and Hayashi (2017) revealed the relationship among these quantities in terms of the phases of upper-level disturbances. In an extreme rainfall event over summertime Japan, Hirota et al. (2016) showed that important factors were abundant moisture in the free troposphere, which was associated with an atmospheric river, and instability and dynamical ascents induced by an upper-tropospheric cutoff low.

Generally, lower-tropospheric convective instability (LCI) is an environment favorable for cumulus convection. However, it has not been clearly answered what roles LCI and other factors, such as the STJ, play in determining precipitation characteristics during the baiu season. Ninomiya (1989) found that major cloud systems that have deep convective clouds, which were identified with the nephanalysis chart of Geostationary Meteorological Satellite data, form under the coexistence of weak convectively unstable conditions and moderately strong baroclinicity in the lower troposphere during the baiu season. Ninomiya et al. (1981) showed that a meso-α-scale cloud cluster had characteristics that change with the environment during its passage from China to the Pacific area. Additionally, Kodama (1993) examined outgoing longwave radiation data in relation to the large-scale environment. However, detailed precipitation characteristics were not fully captured with these data. The present study directly examines how precipitation characteristics during the baiu season are determined by the large-scale environment, using three-dimensional precipitation data observed with the TRMM PR. We focus on effects of the STJ and LCI on the three-dimensional characteristics of precipitation.

2. Data and methodology

a. Data

We mainly use TRMM PR data (PR2A25, version 7; Iguchi et al. 2000, 2009) for the analysis of precipitation characteristics. The TRMM satellite observed precipitation between 35°S and 35°N over 17 yr beginning in November 1997. Three-dimensional observation provides us with various types of information, including the precipitation top height and precipitation type, such as stratiform, convective, and other precipitation.

We also use a rainfall-event (RE) database (Hamada et al. 2014), which is based on PR2A25. Each RE is defined as an area of contiguous rainfall pixels with near-surface precipitation rates exceeding 0.5 mm h−1. A filter developed by Hamada and Takayabu (2014) to remove suspicious extreme rainfall profiles is applied to this database. REs with at least four pixels are analyzed in the present study. With the use of this database, one can obtain characteristics of each RE, such as the area, accumulated total and stratiform near-surface precipitation rates, and maximum near-surface precipitation intensity. The position and observational time of an RE are defined as those at the pixel with the maximum near-surface precipitation intensity of the RE. Note that the method of determining the representative point of the RE does not affect our findings. To examine global distributions of precipitation systems beyond the TRMM observational region, we also analyze REs calculated using Global Precipitation Mission (GPM) core satellite on-board Ku-band radar data (2AKu, version 03B) between 65°N and 65°S.

In addition, Japanese Geostationary Satellite (MTSAT1R) infrared data are used for a case study. For analysis of meteorological fields, Japanese 55-year Reanalysis (JRA-55) data (Kobayashi et al. 2015) are used. Daily mean data calculated from 6-hourly 1.25° latitude–longitude grid data are analyzed. The equivalent potential temperature is first calculated every 6 h and then averaged over a day. The moisture divergence and LCI are calculated from the daily mean data.

The analysis period is from June to July over a 13-yr period (2002–14) except for the GPM data, which are taken from June to July for 2014–15. Note that TRMM data only after 19 June are analyzed for 2009 because of data loss. The main analysis region is around Japan (122°–145°E). Composite analysis of precipitation characteristics relative to latitudes of the STJ is performed for the TRMM PR data observed between 0° and 35°N.

b. Precipitation characteristics analyzed in this study

This study examines various quantities representing precipitation characteristics. The stratiform precipitation ratio (SPR) is the ratio of the unconditional mean stratiform precipitation rate, which is affected by both the strength of stratiform precipitation and the fractional area, to unconditional mean total precipitation rate. Generally, stratiform precipitation has strength that is lower than that of convective precipitation but covers a larger area. According to accumulated knowledge on tropical–subtropical precipitation observed with the TRMM PR (e.g., Takayabu 2002; Schumacher and Houze 2003), mesoscale organized precipitation systems such as cloud clusters and squall lines, which consist of both an area of convective precipitation and a large area of stratiform precipitation (e.g., Rutledge et al. 1988; Houze et al. 1989), tend to have higher SPRs than isolated cumulonimbi. It is thus suggested that the SPR indicates how well precipitation systems are organized on a mesoscale. The precipitation top height, which also characterizes precipitation systems, is defined as the highest altitude at which the precipitation rate is higher than or equal to the threshold of 0.3 mm h−1 at pixels with the flag of “rain certain.”1 In the analysis of REs, we use four variables—namely, the area, SPR, maximum near-surface precipitation intensity, and maximum height of precipitation tops detected with the threshold of 0.5 mm h−1. Note that we use different thresholds for precipitation top heights between the analysis on the pixel basis and the analysis of REs, because the RE database does not provide information on precipitation top heights defined with a value of 0.3 mm h−1.

c. Definitions of the STJ and LCI

The STJ is detected according to the following procedure. First, the maximum and secondary maximum (if any) velocities of daily mean zonal winds are searched for at 200 hPa over 20°–50°N at each longitude. Latitudes with the maxima are recognized as candidates of the STJ. We then impose two conditions to remove the isolated or weak signals that are not like signals of the jet: 1) Signals should be zonally contiguous over a distance of at least 20 degrees, which roughly corresponds to the longitudinal width of the analysis region (122°–145°E). Here, we consider any two zonally adjacent grids to be contiguous if the meridional distance between the two grids is less than or equal to 3.75°. 2) The secondary maximum value should be greater than or equal to half the maximum value at a given longitude. We finally choose the southernmost among the remaining candidates and define it as the STJ. Figure 1 indicates an example of the detected STJ with black open circles. In this case, the jet splits into northern (green open circles) and southern branches over 100°–130°E. Note that zonal wind speeds at latitudes of the STJ are used in analyses of the strength of the STJ.

Fig. 1.

An example of detection of the STJ. Color shading indicates zonal winds at 200 hPa on 8 Jun 2003. Vertical red lines denote the longitudes of the analysis region. Open black circles indicate the detected STJ. In cases that two candidates for the STJ remain at a given longitude through the detection procedure, the southernmost one (open circles in black), not the northernmost one (open circles in green), is defined as the STJ. See the text for details on the detection method.

Fig. 1.

An example of detection of the STJ. Color shading indicates zonal winds at 200 hPa on 8 Jun 2003. Vertical red lines denote the longitudes of the analysis region. Open black circles indicate the detected STJ. In cases that two candidates for the STJ remain at a given longitude through the detection procedure, the southernmost one (open circles in black), not the northernmost one (open circles in green), is defined as the STJ. See the text for details on the detection method.

A convectively unstable atmosphere is characterized by the stratification structure where θe decreases with altitude and is favorable for the initiation of convection in general. The subtropical atmosphere is basically convectively unstable with minimum θe in the midtroposphere. Under this condition, when a layer is bodily lifted until it is saturated, the lower part becomes saturated more rapidly than the upper part and therefore experiences a smaller decrease in temperature than the upper part, resulting in destabilization of the layer. Because organized convective systems have a mechanism that lifts a layer, it is important to estimate destabilization due to lifting of a relatively thick layer. In the baiu season, Ninomiya and Akiyama (1992) indicated that intense precipitation of the baiu front occurs concurrently with the generation of convective instability.

In the present study, we define the lower-tropospheric convective instability (LCI) as the vertical gradient of θε between 1000 and 700 hPa or [()/300] (K hPa−1) to investigate its relationship with precipitation characteristics. Here, θe is defined as equal to θd exp(Lr/CpdT ), where θd is the potential temperature for dry air, r is the mixing ratio, T is temperature at the condensation level, L is the latent heat of condensation, and Cpd is the specific heat at constant pressure for dry air. Note that the LCI is calculated only where daily mean pressure at surface is greater than or equal to 1000 hPa.

d. Composites relative to the STJ

In the present study, we composite quantities representing precipitation characteristics relative to the reference latitudes of the STJ. The STJ tends to be close to the meridional boundary between stratospheric and tropospheric air masses at 200 hPa. Our composite is therefore close to that made by Horinouchi (2014) in which isentropic (350 K) constant-potential-vorticity (1.5 PVU; 1 PVU ≡ 10−6 K kg−1 m2 s−1) lines are used to define the relative latitudes. See also Horinouchi and Hayashi (2017) for the correspondence between these contours and the STJ at 200 hPa. Note that seasonal and zonal variations in precipitation and the large-scale environment, which are indicated by previous studies (Akiyama 1973; Hirasawa et al. 1995; Kato 1985, 1989; Kawamura and Murakami 1998; Ninomiya 1989; Ninomiya and Muraki 1986), are not shown in these composites.

The TRMM and GPM satellites observe more frequently at high latitudes than low latitudes. The data thus have a large meridional sampling variation. To adjust the difference in the sampling frequency among latitudes, we correct values with the sampling number in each latitude:

 
formula

Here wj is the inverse of the total number of pixels observed globally at the jth geographical latitude for the entire period of analysis, k indicates the individual pixel in the ith relative latitudinal bin and the jth geographical latitudinal bin, rijk is the value of the kth pixel in the ith relative latitudinal bin and jth geographical latitudinal bin, and Ri is a bias-corrected value composited in the ith relative latitudinal bin. This correction method is similar to that described in the appendix of the paper of Yokoyama et al. (2014), except that they adjusted differences in the sampling frequency with the value, which varies with not only geographical latitude but also relative latitude. To make composites of REs, we perform a similar adjustment for positions of REs instead of positions of individual pixels.

3. Results

a. Overview of precipitation and environments during the baiu season

We first examine precipitation and environments during the baiu season (June–July) in relation to the STJ. Analyses in this subsection are based on pixels, while characteristics of REs are investigated in the remaining subsections. Figure 2 shows climatological distributions for the precipitation rate, zonal wind at 200 hPa, and LCI for the analysis period. A band of precipitation appears from south China to the ocean east of Japan, with its maximum around 30°N and 130°E, along the south of the STJ, which is centered around 38°–40°N (Fig. 2a). Note that the TRMM PR may not typically observe the northern side of the STJ, because the northernmost observation (35°N) of the TRMM PR is to the south of the climatological STJ center. Using the Global Precipitation Climatology Project 1° daily precipitation (GPCP 1DD; Huffman et al. 2001; version 1.2), we confirmed that the center of the baiu precipitation band is on average located to the south of the STJ and that there is no appreciable precipitation to the north of the STJ except for the Korean Peninsula where precipitation straddles the STJ (not shown). Figure 2b shows that the LCI changes with latitude, especially over the ocean including Japan. The LCI increases toward the southwest in the precipitation band.

Fig. 2.

(a) Climatology of precipitation rates observed by the TRMM PR (mm h−1; color shading) and JRA-55 zonal winds at 200 hPa (m s−1; contours) for June–July over a 13-yr period (2002–14). (b) As in (a), but for JRA-55 LCI (K hPa−1; color shading).

Fig. 2.

(a) Climatology of precipitation rates observed by the TRMM PR (mm h−1; color shading) and JRA-55 zonal winds at 200 hPa (m s−1; contours) for June–July over a 13-yr period (2002–14). (b) As in (a), but for JRA-55 LCI (K hPa−1; color shading).

We next perform composite analysis relative to the reference latitudes of the STJ. Figure 3a shows composite latitudinal–pressure cross sections of zonal wind, θe, and vertical pressure velocity. Note that positive (negative) relative latitudes in the composite figures correspond to the north (south) of the STJ. It is shown that both zonal winds and vertical velocity slightly tilt toward the north with altitude. There are upward velocities throughout the troposphere around 4° south of the STJ, while downward velocities are found to the north of the STJ. In addition, meridional changes in θe profiles are shown. LCI continuously increases across the STJ toward the south (Fig. 3c).

Fig. 3.

(a) Composite latitude–pressure cross sections of zonal winds (black contours with 5 m s−1 interval), θe (color shading, K), and vertical pressure velocity (white contours with 0.02 Pa s−1 interval). Composites are made with centers at the reference latitudes of the STJ. Positive (negative) latitudes are for north (south) of the reference latitude. (b) As in (a), but for moisture convergence (color shading, × 10−5 g kg−1 s−1). (c) Composite latitudinal distribution of LCI (K hPa−1); the red dashed line indicates an LCI of 0 K hPa−1.

Fig. 3.

(a) Composite latitude–pressure cross sections of zonal winds (black contours with 5 m s−1 interval), θe (color shading, K), and vertical pressure velocity (white contours with 0.02 Pa s−1 interval). Composites are made with centers at the reference latitudes of the STJ. Positive (negative) latitudes are for north (south) of the reference latitude. (b) As in (a), but for moisture convergence (color shading, × 10−5 g kg−1 s−1). (c) Composite latitudinal distribution of LCI (K hPa−1); the red dashed line indicates an LCI of 0 K hPa−1.

An obvious contrast between the south and north of the STJ is found in profiles of moisture convergence (Fig. 3b). Deep moisture convergence in the layer of 1000–400 hPa exists to the south of the STJ, while moisture convergence is weak and very shallow to the north. Moisture convergence sharply deepens from the reference latitude of the STJ toward −5°, where the deepest layer of moisture convergence appears. To the south of this relative latitude, moisture convergence remains relatively deep.

Similar composites for unconditional mean precipitation rates are shown in Fig. 4a. Both stratiform rainfall and convective rainfall are significant only to the south of the STJ, resulting in the total rainfall peaking 5° south of the STJ. Obviously, the precipitation band appears together with the deep moisture convergence to the south of the STJ, suggesting that the STJ affects the intensification of the baiu front.

Fig. 4.

(a) Composite latitude distributions of unconditional mean precipitation rates (mm h−1). Black, red, and blue lines indicate total, convective, and stratiform precipitation rates. Composites are made with centers at the reference latitudes of the STJ. Positive (negative) latitudes are for north (south) of the reference latitude. (b) Composite latitude distribution of frequencies of stratiform precipitation top height (%), which are normalized by the number of the total number of observational pixels for each relative latitudinal bin. (c) As in (b), but for convective precipitation top heights.

Fig. 4.

(a) Composite latitude distributions of unconditional mean precipitation rates (mm h−1). Black, red, and blue lines indicate total, convective, and stratiform precipitation rates. Composites are made with centers at the reference latitudes of the STJ. Positive (negative) latitudes are for north (south) of the reference latitude. (b) Composite latitude distribution of frequencies of stratiform precipitation top height (%), which are normalized by the number of the total number of observational pixels for each relative latitudinal bin. (c) As in (b), but for convective precipitation top heights.

In terms of frequencies of precipitation top heights, stratiform (convective) precipitation most frequently starts from 6 km (6–8 km) to the south of the STJ (Figs. 4b,c). In more detail, stratiform precipitation has a maximum frequency around the region from −5° to −2.5°, which is slightly shifted northward compared with the position where convective precipitation most frequently appears centered at around −5°. This meridional change in precipitation characteristics over the precipitation band is consistent with the findings of Yokoyama et al. (2014) but less obvious. This is because their composites were made to capture meridional differences in precipitation characteristics across the baiu front, while we pay more attention to changes in precipitation characteristics relative to the STJ.

It is worth noting that shallow convective precipitation increases across the STJ position toward the south. Remembering the southward increase in LCI shown in Fig. 2b, shallow precipitation seems to be generated under convective instability in the lower troposphere regardless of where the STJ exists. That differs from deeper precipitation, which mostly occurs only adjacent to the south of the STJ, where the atmosphere is relatively neutral. The neutral stability may be a result of the balance between the generation of convective instability due to large-scale motions and the release of the convective instability by deep convections, as indicated by Ninomiya (1984). It is also noted that deep precipitation is suppressed to the south of −8°, indicating a presence of the subtropical high to the south of the precipitation band.

Unconditional mean precipitation rates with respect to the STJ are then divided into LCI bins (Fig. 5a). Precipitation appears in a wide range of LCI, with its peak around an LCI of 0.02–0.04 K hPa−1. In addition, precipitation is found with higher LCI toward the south in the precipitation band, which is consistent with Fig. 2. It is remarkable that the SPR tends to decrease with increasing LCI in the precipitation band to the south of the STJ (Fig. 5b). Figure 5c is a scatterplot of the SPR against LCI for each LCI and relative latitude bin in Fig. 5b. There is negative correlation between these two parameters over the precipitation band (−10° to 0° relative to the STJ), with a correlation coefficient of −0.76, which is statistically significant at a significance level of 95%. Because mesoscale organized precipitation systems have a large stratiform region, the increase in SPR indicates that precipitation is more organized. To the north of the STJ, meanwhile, the SPR also decreases with increasing LCI, with slightly larger rates of change in the SPR with respect to LCI than to the north of the STJ. Relationships between the SPR and LCI are thus robust regardless of the position of the STJ.

Fig. 5.

(a) Composite precipitation rates, which are divided into LCI bins, with respect to the reference latitudes of the STJ. Positive (negative) latitudes are for the north (south) of the reference latitude. (b) As in (a), but for the SPR. Composite SPRs are obtained by dividing composite stratiform precipitation rates by composite precipitation rates. (c) Scatter diagram of the SPR against the LCI for each relative latitude and LCI bin to the south of the STJ (−10° to 0°) in (b). The coefficient of correlation between the two parameters is −0.76, and the regression line is indicated by the solid line.

Fig. 5.

(a) Composite precipitation rates, which are divided into LCI bins, with respect to the reference latitudes of the STJ. Positive (negative) latitudes are for the north (south) of the reference latitude. (b) As in (a), but for the SPR. Composite SPRs are obtained by dividing composite stratiform precipitation rates by composite precipitation rates. (c) Scatter diagram of the SPR against the LCI for each relative latitude and LCI bin to the south of the STJ (−10° to 0°) in (b). The coefficient of correlation between the two parameters is −0.76, and the regression line is indicated by the solid line.

b. Characteristics of REs composing the precipitation band

Before investigating effects of the LCI and STJ on precipitation characteristics in more detail with the use of REs, we first analyze characteristics of 3280 mesoscale (≥103.5 km2 in area) REs composing the precipitation band (−10° to 0° relative to the STJ). Figure 6a shows the relationship among the SPR, area, and maximum near-surface precipitation intensity of REs. Basically, the area tends to become large with increasing SPR. Larger REs tend to have stronger maximum near-surface precipitation intensities for a given SPR. Very high maximum precipitation intensities appear associated with REs having an SPR less than 60% and area of 104–105 km2.

Fig. 6.

(a) Scatter diagram representing the relationship among the SPR (%; abscissa), maximum near-surface precipitation intensity (mm h−1; ordinate), and area [log10(km2); sizes and colors of circles] of REs composing the precipitation band (−10° to 0° relative to the STJ). Maximum near-surface precipitation intensities are indicated on a logarithmic scale. The black line indicates maximum near-surface precipitation intensities averaged with weights of areas for each SPR bin. (b) Frequencies (%) of REs, which are normalized by the total number of REs with areas > 103.5 km2, in terms of SPRs and maximum near-surface precipitation intensities.

Fig. 6.

(a) Scatter diagram representing the relationship among the SPR (%; abscissa), maximum near-surface precipitation intensity (mm h−1; ordinate), and area [log10(km2); sizes and colors of circles] of REs composing the precipitation band (−10° to 0° relative to the STJ). Maximum near-surface precipitation intensities are indicated on a logarithmic scale. The black line indicates maximum near-surface precipitation intensities averaged with weights of areas for each SPR bin. (b) Frequencies (%) of REs, which are normalized by the total number of REs with areas > 103.5 km2, in terms of SPRs and maximum near-surface precipitation intensities.

More interestingly, the relationship among the three parameters drastically changes around an SPR of 80%. For REs with an SPR less than 80%, on average, the maximum near-surface precipitation intensity is nearly constant around 101.9–102 mm h−1 except for the lowest SPR bin, where the sampling number of REs is very small (black line in Fig. 6a). Once the SPR exceeds 80%, the maximum near-surface precipitation intensity significantly decreases with increasing SPR, although the area is still large. These changes in relationship among characteristics of REs suggest that the precipitation band consists of two different groups of mesoscale precipitation systems.

Figure 6b shows frequencies of REs, which are normalized by the total number of large (≥103.5 km2) REs, in terms of the maximum precipitation intensity and SPR. There are two obvious peaks of the frequency of REs around SPRs of 60%–70% and 90%–100%, with a constriction around an SPR of 80%, suggesting that the precipitation band of the baiu front consists of two different types of REs. Hereafter, we refer to REs with an SPR of at least 80% and those with an SPR less than 80% as highly stratiform REs and moderately stratiform REs, respectively.

Figure 7 shows a case of highly stratiform REs and moderately stratiform REs detected on 17 June 2006. According to a surface weather chart provided by the Japan Meteorological Agency and the horizontal distribution of infrared brightness temperature, highly stratiform REs appear associated with a broad area of cold brightness temperature along a warm conveyor belt in an extratropical cyclone. Moderately stratiform REs, meanwhile, appear associated with mesoscale aggregations of tall clouds along a stationary front connecting with the cyclone. In this case, the STJ averaged around Japan (122.5°–145°E) is relatively strong with a zonal wind speed of ~44 m s−1. Subjectively, similar cases are found on days when the STJ is strong around Japan, suggesting that the generation of tall large precipitation systems such as moderately stratiform REs and highly stratiform REs may be affected by the strength of the STJ.

Fig. 7.

Case of 17 Jun 2006. (a) Infrared brightness temperature (color shading, K) at 0330 UTC and positions of highly stratiform REs (circles) and moderately stratiform REs (crosses) observed from 0300 to 0400 UTC. (b) Zonal winds at 200 hPa (color shading, m s−1) and positions of the STJ (circles). The zonal wind speed of the STJ averaged over 122.5°–145°E (indicated by the two vertical red lines) is ~44 m s−1. (c) Surface weather chart at 0000 UTC provided by the Japan Meteorological Agency.

Fig. 7.

Case of 17 Jun 2006. (a) Infrared brightness temperature (color shading, K) at 0330 UTC and positions of highly stratiform REs (circles) and moderately stratiform REs (crosses) observed from 0300 to 0400 UTC. (b) Zonal winds at 200 hPa (color shading, m s−1) and positions of the STJ (circles). The zonal wind speed of the STJ averaged over 122.5°–145°E (indicated by the two vertical red lines) is ~44 m s−1. (c) Surface weather chart at 0000 UTC provided by the Japan Meteorological Agency.

To confirm the characteristic separations of highly stratiform REs and moderately stratiform REs, we examine rainfall distributions of REs based on the GPM Ku-band data for June–July in 2014–15 (Fig. 8). Note that the color scale is different between Figs. 8a and 8b. Pink lines in Fig. 8 indicate mean potential vorticity of 2.5 PVU at the 350-K isentropic surface. Potential vorticity of 1.5–3 PVU roughly corresponds to the position of the STJ. In a composite of potential vorticity at the 350-K isentropic surface relative to the STJ, the potential vorticity significantly changes in a meridional direction across the STJ, with 2.5 PVU of potential vorticity over the STJ (not shown). The STJ and corresponding value of potential vorticity are therefore considered to be the boundary between the midlatitude and subtropics–tropics. In terms of this dynamical boundary, distributions of highly stratiform REs are separated from those of moderately stratiform REs except for the region around Japan where both types of REs are found to the south of the STJ and contribute to large rainfall there. Highly stratiform REs are distributed northeastward across the STJ over the North Pacific, indicating that they appear along the storm track at midlatitudes. In contrast, moderately stratiform REs appear over tropical regions, such as the warm pool and the intertropical convergence zone, where mesoscale organized precipitation systems are dominant.

Fig. 8.

Rainfall from (a) moderately stratiform REs and (b) highly stratiform REs, based on the GPM Ku-band data for June–July in 2014–15. The pink line indicates a mean potential vorticity of 2.5 PVU at the 350-K isentropic surface. Note that the color scales of (a) and (b) are different.

Fig. 8.

Rainfall from (a) moderately stratiform REs and (b) highly stratiform REs, based on the GPM Ku-band data for June–July in 2014–15. The pink line indicates a mean potential vorticity of 2.5 PVU at the 350-K isentropic surface. Note that the color scales of (a) and (b) are different.

As already mentioned, the baiu front appears around the boundary between subtropical and midlatitude air masses. The results of this subsection has thus shown that moderately stratiform REs represent tropical mesoscale organized precipitation systems, while highly stratiform REs represent midlatitude precipitation systems associated with extratropical cyclones. Moderately stratiform REs with larger SPRs are assumed to be better organized than those with smaller SPRs. It is worth noting that some REs with very high (≥80%) SPRs may represent tropical organized precipitation systems at the decay stage because most tropical organized precipitation systems have increasing stratiform fractions as they approach the end of their life cycle. However, the frequency of REs has two separated peaks of highly stratiform REs and moderately stratiform REs rather than a continuous change with the SPR (Fig. 6b), and highly stratiform REs appear along a storm track in a very different manner from moderately stratiform REs (Fig. 8). It is thus reasonable that most highly stratiform REs are associated with extratropical cyclones.

c. LCI effects on precipitation characteristics

As defined in section 2c, the LCI is an indicator of convective instability in the lower troposphere, which is hypothesized to initiate convection. In section 3a, negative correlation between the LCI and SPR is shown on a pixel basis, indicating precipitation is more organized with decreasing LCI. Here, we examine how the LCI affects characteristics of REs.

Figure 9 shows frequency distributions of REs against LCI, categorized by their size, height, and SPR. Note that frequencies are normalized by the number of REs for each category. Characteristics of precipitation systems change with the LCI. REs with areas of at least 105 km2 most frequently appear around an LCI of 0.03 K hPa−1, while the peak for REs with areas smaller than 103.5 km2 is found at an LCI of 0.04 K hPa−1. Focusing on frequencies of relatively small (<103.5 km2) REs, deep (≥8 km) systems are found in the region of higher LCI than shallow (<8 km) systems (Fig. 9b). In terms of the SPR, more stratiform REs tend to exist in an environment of lower LCI (Fig. 9c). These results indicate that in an environment of lower LCI, convection initiation is more difficult without the aid of mesoscale organizations. Note that the occurrence of organized precipitation systems (REs with large areas and large SPRs) in an environment of relatively low LCI may also be partly because these precipitation systems decrease LCI through cold pools and downward advection of low θe from the midtroposphere. There may also be a possibility that the result represents the consumption of the LCI mean state by organized precipitation systems.

Fig. 9.

Frequencies of REs with respect to the LCIs in the region from 122.5° to 145°E and from −10° to 7.5° relative to the STJ latitudes. (a) Frequencies of REs in each area range. The red solid line, yellow line with circles, blue line with triangles, and purple line with squares are for REs with areas of 102–103.5, 103.5–104.5, 104.5–105, and ≥105 km2, respectively. (b) Frequencies of small (<103.5 km2) REs. The red solid line and black dashed line are for REs with maximum heights of precipitation top ≥8 km and <8 km, respectively. (c) Frequencies of REs in each SPR range. The red solid line, orange dashed line, yellow line with circles, blue line with triangles, and purple line with squares are for REs with SPRs of 0%–20%, 20%–40%, 40%–60%, 60%–80%, and 80%–100%, respectively. Frequencies are normalized by the total number of REs for each distribution.

Fig. 9.

Frequencies of REs with respect to the LCIs in the region from 122.5° to 145°E and from −10° to 7.5° relative to the STJ latitudes. (a) Frequencies of REs in each area range. The red solid line, yellow line with circles, blue line with triangles, and purple line with squares are for REs with areas of 102–103.5, 103.5–104.5, 104.5–105, and ≥105 km2, respectively. (b) Frequencies of small (<103.5 km2) REs. The red solid line and black dashed line are for REs with maximum heights of precipitation top ≥8 km and <8 km, respectively. (c) Frequencies of REs in each SPR range. The red solid line, orange dashed line, yellow line with circles, blue line with triangles, and purple line with squares are for REs with SPRs of 0%–20%, 20%–40%, 40%–60%, 60%–80%, and 80%–100%, respectively. Frequencies are normalized by the total number of REs for each distribution.

While precipitation characteristics change with LCI to some degree, Fig. 9 shows that precipitation systems with a variety of characteristics coexist in the region with LCI of 0.02–0.04 K hPa−1, where precipitation rates peak (Fig. 5a). Characteristics of precipitation systems during the baiu season thus cannot be determined only with LCI. In particular, questions remain unanswered with regard to environments favorable for mesoscale precipitation systems, which are a major contributor to rainfall amounts. In the next subsection, we examine effects of the STJ on precipitation characteristics.

d. STJ effects on precipitation characteristics

The case study presented in section 3b suggested that the strength of the STJ may affect precipitation during the baiu season. In elucidating statistical relationships between the STJ strength and precipitation, we classify days according to daily strengths of the STJ averaged around Japan (122.5°–145°E) to examine the mean horizontal maps of environments and precipitation. It is remarkable that specific humidity in the midtroposphere (600 hPa) changes associated with the strength of the STJ (Fig. 10). When the STJ is weak, a broad region with midtropospheric specific humidity greater than 4 g kg−1 is found to the southwest of Japan. As the STJ strengthens, a band-shaped region with high midtropospheric specific humidity extends toward the east along the south of the STJ. Note that there is no similar relationship between lower-tropospheric specific humidity and the strength of the STJ (not shown).

Fig. 10.

Horizontal maps of specific humidity at 600 hPa (color shading, g kg−1), zonal winds at 200 hPa (black contours with an interval of 5 m s−1 starting at 20 m s−1), and the LCI (purple contours drawn at intervals of 0.02, 0.04, and 0.06 K hPa−1), which are averaged for days when mean strengths of the STJ over 122.5°–145°E are (a) 20–30, (b) 30–40, (c) 40–50, and (d) 50–60 m s−1.

Fig. 10.

Horizontal maps of specific humidity at 600 hPa (color shading, g kg−1), zonal winds at 200 hPa (black contours with an interval of 5 m s−1 starting at 20 m s−1), and the LCI (purple contours drawn at intervals of 0.02, 0.04, and 0.06 K hPa−1), which are averaged for days when mean strengths of the STJ over 122.5°–145°E are (a) 20–30, (b) 30–40, (c) 40–50, and (d) 50–60 m s−1.

To diagnose forcing of vertical motion, the Q vector (Hoskins et al. 1978) is daily calculated on the β plane with a reference latitude of 35°:

 
formula

where R, p, Vg, T, and ∇H are the gas constant, pressure, geostrophic wind velocity, temperature, and horizontal gradient operator, respectively. Convergences (divergences) of the Q vector indicate ascent (descent), which is associated with secondary circulation forced by the geostrophic winds. Figure 11 shows composite divergences of the Q vector averaged between 600 and 400 hPa, which are classified according to strengths of the STJ in a manner similar to the classification of Fig. 10.

Fig. 11.

As in Fig. 10, but for divergence of the Q vector averaged between 600 and 400 hPa (kg−1 m s−1; color shading) and zonal winds at 200 hPa (m s−1; black contours).

Fig. 11.

As in Fig. 10, but for divergence of the Q vector averaged between 600 and 400 hPa (kg−1 m s−1; color shading) and zonal winds at 200 hPa (m s−1; black contours).

Figure 11 shows that the Q vector tends to converge to the south of the STJ in all categories of the STJ strength. Large convergence appears as the STJ becomes strong, which is similar to the behavior of specific humidity at 600 hPa. It is also interesting to note that convergence of the Q vector is strong especially around the jet entrance when the STJ is strong (Figs. 11c,d). Therefore, ascent is significantly induced by secondary circulation around the jet entrance and the midtropospheric humidity is enhanced there.

Figure 12 shows composite maps of horizontal and vertical moisture flux convergence at 600 hPa for days when the STJ speed is 50–60 m s−1. The vertical moisture flux convergence is strong where specific humidity at 600 hPa is high (Fig. 12a), indicating that it largely contributes to the distribution of specific humidity at this level. Ascent associated with secondary circulations of the STJ thus plays a major role in moistening the midtroposphere, although cumulus convection there may also feed back to the large-scale environment through the enhancement of ascending motion and midtropospheric moisture. Meanwhile, the horizontal moisture flux convergence at 600 hPa is found just along the STJ and to the north of the region where the vertical moisture flux convergence is strong (Fig. 12b), contributing to precipitation there. It seems that the moisture is increased associated with ascent around the jet entrance, and then transported further northeastward. Note that horizontal moisture flux convergence is stronger in lower levels. At 800 hPa, northeastward moisture flux largely converges over a wide region to the south of the STJ (not shown).

Fig. 12.

Composite maps of (a) vertical and (b) horizontal moisture flux convergence (color shading, 10−5 g kg−1 s−1) at 600 hPa for days when the STJ speed is 50–60 m s−1. Gray vectors indicate horizontal moisture flux at 600 hPa. Black and purple contours indicate zonal winds at 200 hPa and specific humidity at 600 hPa, respectively. Contours for zonal winds at 200 hPa are drawn with intervals of 10 m s−1, starting at 20 m s−1. Contours for specific humidity are drawn with intervals of 0.5 g kg−1, starting at 3.5 g kg−1.

Fig. 12.

Composite maps of (a) vertical and (b) horizontal moisture flux convergence (color shading, 10−5 g kg−1 s−1) at 600 hPa for days when the STJ speed is 50–60 m s−1. Gray vectors indicate horizontal moisture flux at 600 hPa. Black and purple contours indicate zonal winds at 200 hPa and specific humidity at 600 hPa, respectively. Contours for zonal winds at 200 hPa are drawn with intervals of 10 m s−1, starting at 20 m s−1. Contours for specific humidity are drawn with intervals of 0.5 g kg−1, starting at 3.5 g kg−1.

Figures 13 and 14 show distributions for precipitation from moderately stratiform REs and highly stratiform REs, respectively. Consistent with Fig. 8, distributions of highly stratiform REs differ from those of moderately stratiform REs. Most of the precipitation from highly stratiform REs is found along the south of the STJ in a low-LCI region. With some overlaps, meanwhile, moderately stratiform REs are dominant in a high-LCI region to the southwest of the region where highly stratiform REs are dominant.

Fig. 13.

As in Fig. 10, but for rainfall from moderately stratiform REs (color shading, mm h−1), which are observed by the TRMM PR. In this figure, contours for zonal winds at 200 hPa are drawn with intervals of 10 m s−1, starting at 20 m s−1.

Fig. 13.

As in Fig. 10, but for rainfall from moderately stratiform REs (color shading, mm h−1), which are observed by the TRMM PR. In this figure, contours for zonal winds at 200 hPa are drawn with intervals of 10 m s−1, starting at 20 m s−1.

Fig. 14.

As in Fig. 13, but for rainfall from highly stratiform REs.

Fig. 14.

As in Fig. 13, but for rainfall from highly stratiform REs.

As the STJ strengthens, precipitation from both moderately stratiform REs and highly stratiform REs increases (Figs. 13 and 14). Precipitation from moderately stratiform REs is enhanced in the midtropospheric moist region to the southwest of Japan when the STJ is strong (Figs. 13c,d). Figure 14 shows that although there is little precipitation from highly stratiform REs when the STJ is weak (Figs. 14a,b), it appears in a band shape extending eastward from ~125°E when the STJ is 40 m s−1 or higher (Fig. 14c). The band of precipitation from highly stratiform REs is further enhanced for the STJ of 50–60 m s−1 (Fig. 14d). These changes in spatial distributions of precipitation from highly stratiform REs associated with strengths of the STJ are similar to those of midtropospheric specific humidity. Note that the band of precipitation from highly stratiform REs is not completely in accordance with the midtropospheric moisture band but located in its northern half. Because mesoscale REs including both moderately stratiform REs and highly stratiform REs largely contribute to the total rainfall, the total rainfall tends to increase with increasing STJ speed (not shown).

In summary, under the influence of the STJ, highly stratiform REs and moderately stratiform REs are found in low- and high-LCI regions, respectively. Because precipitation from mesoscale systems makes a large contribution to the total rainfall, the STJ plays a role in the formation of the precipitation band during the baiu season. It is noted that there is less precipitation from moderately stratiform REs when the STJ is absent, even though LCI is relatively high (Figs. 14a,b). We can therefore emphasize the STJ in addition to the LCI as a factor of heavy precipitation during the baiu season.

e. Relationship among precipitation characteristics, LCI, and the strength of the STJ

Figure 15 summarizes the relationship among precipitation characteristics, LCI, and the strength of the STJ. Contributions of rainfall at different LCIs and STJ speeds to the total rainfall are shown for small REs (<103.5 km2; black contours in Fig. 15a), moderately stratiform REs (red contours in Fig. 15a), and highly stratiform REs (Fig. 15b). For each type of RE, the values are normalized by its total rainfall over the region from 110°E to 170°E and from −10° to 0° relative to the STJ.

Fig. 15.

(a) Rainfall contributions (%) from small (<103.5 km2) REs (black contours) and moderately stratiform REs (red contours) under different LCI and jet speed conditions during June–July 2002–14. For each type of REs, values are normalized by its total rainfall from 110° to 170°E and from −10° to 0° relative to the STJ. (b) As in (b), but for rainfall contributions (%) from highly stratiform REs. (c) Difference in rainfall contribution (%) between small REs and moderately stratiform REs [values indicated with red contours minus values indicated with black contours in (a)]. (d) Number of occurrences for each bin of STJ speeds. Total rainfall from small REs, moderately stratiform REs, and highly stratiform REs over the region for the period is 0.031, 0.13, and 0.050 mm h−1, respectively.

Fig. 15.

(a) Rainfall contributions (%) from small (<103.5 km2) REs (black contours) and moderately stratiform REs (red contours) under different LCI and jet speed conditions during June–July 2002–14. For each type of REs, values are normalized by its total rainfall from 110° to 170°E and from −10° to 0° relative to the STJ. (b) As in (b), but for rainfall contributions (%) from highly stratiform REs. (c) Difference in rainfall contribution (%) between small REs and moderately stratiform REs [values indicated with red contours minus values indicated with black contours in (a)]. (d) Number of occurrences for each bin of STJ speeds. Total rainfall from small REs, moderately stratiform REs, and highly stratiform REs over the region for the period is 0.031, 0.13, and 0.050 mm h−1, respectively.

Figure 15c shows the difference in the rainfall contribution between small REs and moderately stratiform REs (values indicated with red contours minus values indicated with black contours in Fig. 15a). It is shown that precipitation from small REs tends to appear in higher-LCI regions compared with that from moderately stratiform REs and is almost independent of the STJ speed with high dependency on the LCI. Moderately stratiform REs can exist even in a region with relatively low LCI if the STJ is strong, indicating that the large-scale environments associated with a strong STJ, such as an environment with enhanced midtropospheric specific humidity, are favorable for organization of cumulus convection in the baiu season.

Meanwhile, rainfall contributions of highly stratiform REs are large in lower-LCI regions than other types and tend to increase as the STJ becomes strong (Fig. 15b). Note that rainfall from highly stratiform REs is largest when the STJ is 50–60 m s−1 (Fig. 14), while the number of days with such strong STJ speeds is relatively small (Fig. 15d). As a result, the peak of rainfall contributions does not appear for the STJ of 50–60 m s−1 but appears for the STJ of 40–50 m s−1 (Fig. 15b). As mentioned in section 3b, highly stratiform REs are mostly associated with extratropical cyclones and bring precipitation via a mechanism different from that for small REs and moderately stratiform REs representing tropical organized precipitation systems. Other environmental fields, such as enhanced baroclinicity, may also be considered in addition to the enhanced midtropospheric specific humidity. Further studies are needed to elucidate the mechanism for the relationship between the STJ speed and highly stratiform REs.

4. Discussion and summary

In this study, we examine how the subtropical jet (STJ) and lower-tropospheric convective instability (LCI) affect precipitation characteristics in the baiu season, using three-dimensional precipitation data observed with the TRMM PR as well as other datasets.

It is shown that the precipitation band along the baiu front appears to the south of the STJ, which is consistent with the findings of Horinouchi (2014) and Yokoyama et al. (2014). The precipitation band is found to be associated with a deep moisture convergence. The stratiform precipitation ratio (SPR) has negative correlation with the LCI, which increases across the precipitation band toward the south.

Two types of mesoscale REs composing the precipitation band are identified: that is, moderately stratiform REs with SPRs of 0%–80% and highly stratiform REs with SPRs of 80%–100%. With some overlaps, moderately stratiform REs tend to be dominant to the southwest of the region where highly stratiform REs are dominant along the south of the STJ. From a case study and a comparison of distributions between these two types of mesoscale REs, moderately stratiform REs are found to represent tropical mesoscale organized precipitation systems such as mesoscale convective systems (herein referred to as tropical organized precipitation systems), whereas highly stratiform REs are found to represent midlatitude precipitation systems associated with extratropical cyclones (herein referred to as midlatitude precipitation systems). These two types of precipitation systems have a different relationship among the SPR, area, and maximum precipitation intensity. Tropical organized precipitation systems are characterized by high intensity and the coexistence of stratiform and convective precipitation, while midlatitude precipitation systems are characterized by moderate intensity and the predominance of stratiform precipitation. The coexistence of these two types of mesoscale precipitation systems is consistent with the fact that the baiu front forms the boundary between subtropical and midlatitude air masses.

These mesoscale precipitation systems, either tropical organized precipitation systems or midlatitude precipitation systems, are affected by the presence of the STJ. Tropical organized precipitation systems tend to appear with higher LCI than midlatitude precipitation systems, but they can also exist even in a region of relatively low LCI if the STJ is strong. In contrast to mesoscale precipitation systems, small precipitation systems including both shallow and deep systems appear regardless of the STJ, with high dependency on the LCI.

As the STJ intensifies, precipitation from mesoscale precipitation systems increases along the south of the STJ, with a distinct eastward extension of midtropospheric moist region. At the same time, the Q vector converges to the south of the STJ, indicating that ascent associated with the secondary circulation moistens the midtroposphere. Convergence of the Q vector tends to be strong especially around the jet entrance, where secondary circulation is expected to be forced by mechanisms including the confluence and shear deformation of geostrophic flows. Further studies are needed to reveal the mechanism that forces the realistic secondary circulation. The ascent induced by secondary circulation around the jet entrance enhances the midtropospheric humidity there. The moisture may be advected further eastward by westerly winds in the lower-mid troposphere, which are located to the south of the STJ.

We next discuss how the STJ and LCI affect cumulus convection. Previous studies suggest that the LCI determines the generation of cumulus congestus but does not guarantee the development of deep cumulus convection, which is controlled by humidity in the midtroposphere through entrainment. Using observational data, Takayabu et al. (2010) showed that diabatic heating associated with cumulus congestus linearly correlates with sea surface temperature (SST), but heating associated with deep convection is shown to be suppressed by large-scale subsidence, which is accompanied by dry air, even under a high-SST condition. Note that the LCI can be considered to have a similar effect to SST because increasing SST leads to an increase in the LCI. Kuang and Bretherton (2006) conducted numerical simulations to show that the development of deep convection does not occur even with a substantial convectively available potential energy when the midtroposphere is relatively dry.

This study shows two pathways for the development of deep cumulus convection. Under high-LCI conditions, congestus can develop to deep cumulus convection without the aid of mesoscale organization. In conditions of relatively low LCI, meanwhile, moistening of the midtroposphere due to ascent associated with secondary circulations of the STJ is important for development to deep cumulus convection. Mesoscale convective systems have deeper convergence than young convective features, which have strong near-surface convergence (Mapes and Houze 1995). Organized precipitation systems may therefore more efficiently collect water vapor in the mid- to lower troposphere, resulting in the dominance of these systems in a region with relatively low LCI and strong STJ during the baiu season.

From the above results and discussion, we conclude that the STJ and LCI play roles in determining precipitation characteristics around the baiu front as follows. The STJ moistens the midtroposphere owing to the ascent associated with secondary circulations to the south of the STJ, producing an environment favorable for tropical organized precipitation systems in the southern part of the baiu rainband. Horizontal moisture flux convergence may also contribute to precipitation just along the STJ. As for midlatitude precipitation systems associated with extratropical cyclones, other environmental fields, such as enhanced baroclinicity, should also be considered in addition to the enhanced midtropospheric specific humidity. Further studies are needed in regard to this point. Meanwhile, the LCI is considered to be important in generating shallow precipitation systems. In high-LCI conditions, deep convection can occur without the aid of mesoscale organization.

Recent studies underline the importance of upper-tropospheric circulations in generating precipitation (Horinouchi 2014; Hirota et al. 2016). It has also been pointed out that moisture in the free troposphere is important for deep convection and organized precipitation systems (e.g., Takayabu et al. 2010; Hamada et al. 2015). Most recently, Hirota et al. (2016) pointed out that the free-tropospheric moisture, which was ample along an atmospheric river, and an upper-tropospheric cutoff low were important factors in an extreme rainfall event in Japan. Our results show that in a statistical manner, combined effects of upper-tropospheric circulations and midtropospheric moisture play an important role in the generation of organized precipitation systems, which often bring heavy rainfall, during the baiu season.

Acknowledgments

This study is supported by the 8th RA of the Japan Aerospace Exploration Agency (JAXA) Precipitation Measuring Mission (PMM) science, the Environment Research and Technology Development Fund (2-1503) of the Environmental Restoration and Conservation Agency, Japan, and JSPS KAKENHI Grant 15H02132. The authors would like to express their gratitude to Prof. Edward J. Zipser and two anonymous reviewers for their very helpful comments. The authors also thank JAXA and National Aeronautics and Space Administration (NASA) for providing the TRMM and GPM data. The GPM RE data were calculated by Ms. Marika Ono. The JRA-55 data used in this study were provided by the Japan Meteorological Agency (JMA). The JMA surface weather chart was obtained from TENKI (the Bulletin Journal of the Meteorological Society of Japan in Japanese). The MTSAT-1R data used in this study were received by the JMA; Weathernews Inc.; the Earthquake Research Institute, the University of Tokyo; and Takeuchi Laboratory, the Institute of Industrial Science, the University of Tokyo, and they were processed and provided by the Center for Environmental Remote Sensing (CEReS), Chiba University.

REFERENCES

REFERENCES
Akiyama
,
T.
,
1973
:
The large-scale aspects of the characteristic features of the baiu front—with special emphasis on the relation among the ageostrophic low-level jet stream, moist tongue, convective warming, convergence zone within baiu front and heavy rainfall
.
Pap. Meteor. Geophys.
,
24
,
157
188
, doi:.
Akiyama
,
T.
,
1978
:
Mesoscale pulsation of convective rain in medium-scale disturbances developed in baiu front
.
J. Meteor. Soc. Japan
,
56
,
267
283
, doi:.
Akiyama
,
T.
,
1979
:
Thermal stratification in baiu frontal medium-scale disturbances with heavy rainfalls
.
J. Meteor. Soc. Japan
,
57
,
587
598
, doi:.
Akiyama
,
T.
,
1990
:
Large, synoptic and meso scale variations of the baiu front during July 1982. Part II: Frontal structure and disturbances
.
J. Meteor. Soc. Japan
,
68
,
557
574
, doi:.
Hamada
,
A.
, and
Y. N.
Takayabu
,
2014
:
A removal filter for suspicious extreme rainfall profiles in TRMMPR 2A25 version-7 data
.
J. Appl. Meteor. Climatol.
,
53
,
1252
1271
, doi:.
Hamada
,
A.
,
T.
Murayama
, and
Y. N.
Takayabu
,
2014
:
Regional characteristics of extreme rainfall extracted from TRMM PR measurements
.
J. Climate
,
27
,
8151
8169
, doi:.
Hamada
,
A.
,
Y. N.
Takayabu
,
C.
Liu
, and
E. J.
Zipser
,
2015
:
Weak linkage between the heaviest rainfall and tallest storms
.
Nat. Commun.
,
6
,
6213
, doi:.
Hirasawa
,
N.
,
K.
Kato
, and
T.
Takeda
,
1995
:
Abrupt change in the characteristics of the cloud zone in subtropical East Asia around the middle of May
.
J. Meteor. Soc. Japan
,
73
,
221
239
, doi:.
Hirota
,
N.
,
Y. N.
Takayabu
,
M.
Kato
, and
S.
Arakane
,
2016
:
Roles of an atmospheric river and a cutoff low in the extreme precipitation event in Hiroshima on 19 August 2014
.
Mon. Wea. Rev.
,
144
,
1145
1160
, doi:.
Horinouchi
,
T.
,
2014
:
Influence of upper tropospheric disturbances on the synoptic variability of precipitation and moisture transport over summertime East Asia and the northwestern Pacific
.
J. Meteor. Soc. Japan
,
92
,
519
541
, doi:.
Horinouchi
,
T.
, and
A.
Hayashi
,
2017
:
Meandering subtropical jet and precipitation over summertime East Asia and the northwestern Pacific
.
J. Atmos. Sci.
,
74
,
1233
1247
, doi:.
Hoskins
,
B. J.
,
I.
Draghici
, and
H. C.
Davies
,
1978
:
A new look at the ω-equation
.
Quart. J. Roy. Meteor. Soc.
,
104
,
31
38
, doi:.
Houze
,
R. A.
, Jr.
,
S. A.
Rutledge
,
M. I.
Biggerstaff
, and
B. F.
Smull
,
1989
:
Interpretation of Doppler weather radar displays of midlatitude mesoscale convective systems
.
Bull. Amer. Meteor. Soc.
,
70
,
608
619
, doi:.
Huffman
,
G. J.
,
R. F.
Adler
,
M. M.
Morrissey
,
D. T.
Bolvin
,
S.
Curtis
,
R.
Joyce
,
B.
McGavock
, and
J.
Susskind
,
2001
:
Global precipitation at one-degree daily resolution from multisatellite observations
.
J. Hydrometeor.
,
2
,
36
50
, doi:.
Iguchi
,
T.
,
T.
Kozu
,
R.
Meneghini
,
J.
Awaka
, and
K.
Okamoto
,
2000
:
Rain-profiling algorithm for the TRMM Precipitation Radar
.
J. Appl. Meteor.
,
39
,
2038
2052
, doi:.
Iguchi
,
T.
,
T.
Kozu
,
J.
Kwiatkowski
,
R.
Meneghini
,
J.
Awaka
, and
K.
Okamoto
,
2009
:
Uncertainties in the rain profiling algorithm for the TRMM Precipitation Radar
.
J. Meteor. Soc. Japan
,
87A
,
1
30
, doi:.
Kato
,
K.
,
1985
:
On the abrupt change in the structure of the baiu front over the China continent in late May of 1979
.
J. Meteor. Soc. Japan
,
63
,
20
36
, doi:.
Kato
,
K.
,
1989
:
Seasonal transition of the lower-level circulation systems around the baiu front in China in 1979 and its relation to the northern summer monsoon
.
J. Meteor. Soc. Japan
,
67
,
249
265
, doi:.
Kato
,
T.
, and Coauthors
,
2003
:
Reason for the failure of the simulation of heavy rainfall during X-BAIU-01—importance of a vertical profile of water vapor for numerical simulations
.
J. Meteor. Soc. Japan
,
81
,
993
1013
, doi:.
Kawamura
,
R.
, and
T.
Murakami
,
1998
:
Baiu near Japan and its relation to summer monsoons over Southeast Asia and the western North Pacific
.
J. Meteor. Soc. Japan
,
76
,
619
639
, doi:.
Kobayashi
,
S.
, and Coauthors
,
2015
:
The JRA-55 reanalysis: General specifications and basic characteristics
.
J. Meteor. Soc. Japan
,
93
,
5
48
, doi:.
Kodama
,
Y.-M.
,
1992
:
Large-scale common features of subtropical precipitation zones (the baiu frontal zone, the SPCZ, and the SACZ). Part I: Characteristics of subtropical frontal zones
.
J. Meteor. Soc. Japan
,
70
,
813
836
, doi:.
Kodama
,
Y.-M.
,
1993
:
Large-scale common features of sub-tropical precipitation zones (the baiu frontal zone, the SPCZ, and the SACZ). Part II: Conditions of the circulations for generating the STCZs
.
J. Meteor. Soc. Japan
,
71
,
581
610
, doi:.
Kuang
,
Z.
, and
C. S.
Bretherton
,
2006
:
A mass-flux scheme view of a high-resolution simulation of a transition from shallow to deep cumulus convection
.
J. Atmos. Sci.
,
63
,
1895
1909
, doi:.
Mapes
,
B. E.
, and
R. A.
Houze
Jr.
,
1995
:
Diabatic divergence profiles in western Pacific mesoscale convective systems
.
J. Atmos. Sci.
,
52
,
1807
1828
, doi:.
Ninomiya
,
K.
,
1978
:
Heavy rainfalls associated with frontal depression in Asian subtropical humid region
.
J. Meteor. Soc. Japan
,
56
,
253
266
, doi:.
Ninomiya
,
K.
,
1984
:
Characteristics of baiu front as a predominant subtropical front in the summer Northern Hemisphere
.
J. Meteor. Soc. Japan
,
62
,
880
894
, doi:.
Ninomiya
,
K.
,
1989
:
Cloud distribution over East Asia during baiu period in 1979
.
J. Meteor. Soc. Japan
,
67
,
639
658
, doi:.
Ninomiya
,
K.
, and
H.
Muraki
,
1986
:
Large-scale circulations over East Asia during baiu period of 1979
.
J. Meteor. Soc. Japan
,
64
,
409
429
, doi:.
Ninomiya
,
K.
, and
T.
Akiyama
,
1992
:
Multi-scale features of baiu, the summer monsoon over Japan and the East Asia
.
J. Meteor. Soc. Japan
,
70
,
467
495
, doi:.
Ninomiya
,
K.
, and
Y.
Shibagaki
,
2007
:
Multi-scale features of the meiyu-baiu front and associated precipitation systems
.
J. Meteor. Soc. Japan
,
85B
,
103
122
, doi:.
Ninomiya
,
K.
,
M.
Ikawa
, and
T.
Akiyama
,
1981
:
Long-lived medium-scale cumulonimbus cluster in Asian subtropical humid region
.
J. Meteor. Soc. Japan
,
59
,
564
577
, doi:.
Park
,
M.-S.
,
M.-I.
Lee
,
H.
Kim
,
J.
Im
, and
J.-M.
Yoo
,
2014
:
Spatial and diurnal variations of storm heights in the East Asia summer monsoon: Storm heights regimes and large-scale diurnal modulation
.
Climate Dyn.
,
46
,
745
763
, doi:.
Rutledge
,
S. A.
,
R. A.
Houze
Jr.
,
M. I.
Biggerstaff
, and
T.
Matejka
,
1988
:
The Oklahoma–Kansas mesoscale convective system of 10–11 June 1985: Precipitation structure and single-Doppler radar analysis
.
Mon. Wea. Rev.
,
116
,
1409
1430
, doi:.
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
, doi:.
Schumacher
,
C.
, and
R. A.
Houze
Jr.
,
2003
:
Stratiform rain in the tropics as seen by the TRMM Precipitation Radar
.
J. Climate
,
16
,
1739
1756
, doi:.
Takayabu
,
Y. N.
,
2002
:
Spectral representation of rain profiles and diurnal variations observed with TRMM PR over the equatorial area
.
Geophys. Res. Lett.
,
29
,
1584
, doi:.
Takayabu
,
Y. N.
, and
K.
Hikosaka
,
2009
:
Statistical analysis of oceanic rainfall characteristics in the baiu season utilizing TRMM PR data
.
J. Meteor. Soc. Japan
,
87A
,
339
352
, doi:.
Takayabu
,
Y. N.
,
S.
Shige
,
W.-K.
Tao
, and
N.
Hirota
,
2010
:
Shallow and deep latent heating modes over tropical oceans observed with TRMM PR spectral latent heating data
.
J. Climate
,
23
,
2030
2046
, doi:.
Xu
,
W.
, and
E. J.
Zipser
,
2011
:
Diurnal variations of precipitation, deep convection, and lightning over east of the eastern Tibetan Plateau
.
J. Climate
,
24
,
448
465
, doi:.
Xu
,
W.
,
E. J.
Zipser
, and
C.
Liu
,
2009
:
Rainfall characteristics and convective properties of mei-yu precipitation systems over south China, Taiwan, and the South China Sea. Part I: TRMM observations
.
Mon. Wea. Rev.
,
137
,
4261
4275
, doi:.
Yokoyama
,
C.
,
S.
Kanada
, and
Y. N.
Takayabu
,
2014
:
A contrast in precipitation characteristics across the baiu front near Japan. Part I: TRMM PR observation
.
J. Climate
,
27
,
5872
5890
, doi:.

Footnotes

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

1

Over 122°–145°E and 0°–35°N during June–July for 2002–14, the mode values of PR reflectivity corresponding to 0.3 mm h−1 with the flag of “rain certain” are ~15 and ~15.8 dBZ for convective and stratiform precipitation, respectively.