1. Introduction
Studies on the trends of extreme daily precipitation have found distinctive regional patterns of the trends in the Asian monsoon region including China (Wang and Zhou 2005; Zhai et al. 2005). The scattered positive trends over China, in contrast to the overall positive trends over North America and Europe (Min et al. 2011), are at least partially attributed to tropical cyclones over the western North Pacific, which bring rainfall with decreasing frequency and increasing intensity, and the “southern flood and northern drought” pattern that is a prominent interdecadal phenomenon of the East Asian summer monsoon rainfall (Chang et al. 2012).
With the availability of long-term hourly precipitation data based on rain gauge observations, several studies have focused on characteristics of extreme hourly precipitation in China. Zhang and Zhai (2011) examined diurnal variations and trends of extreme hourly precipitation (defined using the 95th percentile as local threshold) over eastern China in the warm season. Yu and Li (2012) classified late-summer rainfall over eastern contiguous China according to hourly intensity and analyzed the changes of moderate, intense, and extreme precipitation in response to variation of surface air temperature by comparing the periods 1966–85 and 1986–2005. Li et al. (2013) estimated the threshold values of hourly rainfall intensity for 5-yr return period and found significant regional differences of the threshold over eastern China. They further examined the average duration of extreme precipitation events and briefly discussed the earliest and latest month that the extreme rainfall was found in eastern China. Moreover, average warm-season or summer rainfall over central eastern China has been investigated with a focus on its diurnal variations (Yu et al. 2007a,b; Li et al. 2008; Yuan et al. 2012).
Precipitation over China is greatly influenced by the interactions between the summer monsoonal flow from the Bay of Bengal and the west Pacific Ocean and the cold air coming from the higher latitudes. Following the northward advance of the East Asian summer monsoon circulation, three major rainy seasons are formed sequentially: the early summer (May to mid-June) rainy season in southern China, the mei-yu season over the Yangtze and Huai river basins, and the late summer rainy season in northern China (Ding and Chan 2005). The synoptic weather systems contributing favorably to the occurrence of the heavy rainfall include 1) a low-level vortex that originates on the eastern flank of the Tibetan Plateau and is often steered out by an eastward-moving upper-level trough (Tao and Ding 1981), 2) a quasi-east–west-oriented shear line at 850 hPa and/or surface (i.e., mei-yu front) over east China that is characterized by weak temperature gradients but high equivalent potential temperature (θe) gradients (Ding and Chan 2005), 3) a southwesterly low-level jet (LLJ) originating in the tropical oceans and embedded in the anticyclonic circulation associated with the western Pacific subtropical high (Chen and Yu 1988; Chen et al. 2005, 2006), and 4) a synoptic-scale frontal boundary (usually oriented from northeast to southwest) with a corresponding trough aloft (Ding et al. 1980). Furthermore, heavy rainfall events in China are also closely associated with tropical cyclones (TCs) and their remnants (Chang et al. 2012; Su et al. 2016). However, studies on quantitative contributions to extreme rainfall in China by various synoptic systems, on hourly time scales in particular, are rather limited in the literature.
Using a newly released long series of hourly rainfall data based on densely distributed rain gauges, combined with weather maps and radar reflectivity data, this study further examines the characteristics of hourly precipitation over most of China (east of 96°E) with regard to two aspects. One is the seasonal variations of hourly precipitation (>0.1 mm h−1), including both their spatial distributions over the analysis domain and comparisons among nine selected subregions (areas A–I in Fig. 1). The more important one is synoptic situations associated with extreme hourly precipitation (defined in section 2), including a classification of the extreme hourly precipitation into four types based on the relevant synoptic features, and spatial, seasonal, and diurnal variations of each extreme hourly precipitation type. In section 2, the data and methods used in this study are presented. The seasonality of the hourly precipitation (>0.1 mm h−1) is described in section 3. The threshold, intensity, and occurrence frequency of the extreme hourly precipitation are represented in section 4. Section 5 documents the four types of precipitation extreme including their examples, spatial distributions, and seasonal and diurnal variations. A summary and conclusions are given in section 6.
Spatial distribution of rain gauge stations (dots) with three different colors/shapes representing different start/end months of the hourly rainfall data used in the present study (see text). Shadings represent surface elevation (km); black lines denote locations of the Yangtze, Huai, and Yellow Rivers. Outlined areas A–I denote the subregions that are used in Figs. 4, 13, and 14.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
2. Data and methods
The hourly precipitation dataset used in this study was obtained from the National Meteorological Information Center (NMIC) of the China Meteorological Administration (CMA). It consists of 2042 observation stations (dots/stars/triangles in Fig. 1) covering mainland China (east of 96°E) and Hainan Island from 1981 to 2015 and fewer stations in earlier years (e.g., less than 1000 stations in the 1950s). According to the start and end dates of the observation, the 2042 stations can be divided into three groups: 1) stations to the south of the Yangtze River and over the Sichuan basin, accounting for 44% of the total stations (the cyan dots in Fig. 1), have records throughout the year; 2) stations in central China and the North China Plain (35%; the tan stars in Fig. 1) have records from March/April to October/November; and 3) stations located from inner Mongolia to northeast China (21%; the green triangles in Fig. 1) have records from May to September, as these stations routinely stop taking rainfall measurement in the freezing conditions of the winter season under certain regulations. A quality control procedure has been applied to the dataset by NMIC, which consists of a climatological limit value test, a station extreme value test, and an internal consistency test. Additionally, these hourly data have been compared with manually checked daily rainfall data. The hourly records were replaced with missing values if the daily rainfall data are >5 mm per day (mm day−1) and the relative difference [(daily data − 24-h accumulated hourly data) / daily data] is >0.2, or the daily rainfall data are ≤5 mm day−1 and the absolute difference is >1 mm. All the quality-controlled records during 1981–2015 have been used in the present study in order to obtain more complete results in terms of the temporal and spatial coverages.
The complicated regional features of the hourly precipitation (>0.1 mm h−1) over China (section 3) suggest that using a relative threshold value (i.e., a given percentile of the distribution function or a specific return frequency) is more appropriate for studying extreme precipitation in China than using a fixed criterion (e.g., 20 mm h−1). Therefore, in the present study, the extreme precipitation at each station is defined according to the cumulative density function of precipitation during a 33-yr period (1981–2013). Thresholds determined at the 99.9th, 99th, and 95th percentiles are examined and exhibit generally similar spatial patterns. However, the threshold values of the 99th percentile are generally smaller than the fixed threshold defined by the NMC for heavy hourly rainfall (20 mm h−1), and those of the 95th percentile are even smaller (not shown). Thus, the local thresholds determined at the 99.9th percentile are used to define the extreme precipitation in the present study.
Occurrence frequency of precipitation is defined as the number of hours (hours = records) with a rainfall amount ≥ 0.1 mm h−1. Occurrence frequency of extreme precipitation is defined as the number of hours when a rain gauge measured an hourly rainfall amount exceeding the local threshold, which is determined at the 99.9th percentile of hourly precipitation rates (>0.1 mm h−1). Monthly precipitation intensity is defined as the average rainfall amount among the hours with precipitation ≥ 0.1 mm h−1 during the month. Intensity of extreme hourly precipitation at a station is defined as the average of all extreme precipitation records at the station.
To classify extreme precipitation according to its associated synoptic weather systems (the TC or their remnants, the surface front, the low-level vortex/shear line) the western North Pacific (WNP) TC historical track data from the Shanghai Typhoon Institute of the CMA (http://tcdata.typhoon.gov.cn/zjljsjj_zlhq.html) and weather maps and composite radar reflectivity data from National Meteorological Center (NMC) of China are jointly used. The upper-air (500/700/850 hPa) synoptic analysis was available twice daily (0000/1200 UTC) and the surface analysis was four times per day (0000/0600/1200/1800 UTC). The digital mosaics of composite radar reflectivity (i.e., the horizontal distribution of maximum radar reflectivity in the vertical) have a pixel resolution of 4 km × 4 km and temporal resolution of 10 min. This dataset has been extensively used not only in operational weather forecasting but also in scientific researches to identify rainy systems (e.g., Meng et al. 2013; Zheng et al. 2013).
The TC-type extreme precipitation was first identified with an objective synoptic analysis technique (OSAT) (Ren et al. 2006, 2007, 2011) that uses the distance from TC center and the closeness and continuity between neighboring raining stations to trace TC-influenced rain belts that may extend from 500 to 1100 km away form a TC center. Second, the frontal type was identified based on some subjective judgment with quantitative criteria: 1) the extreme hourly rainfall was located within 50 km from a surface front (mostly a cold front) on the weather maps and 2) the associated rainband extended over more than ~500 km on the radar reflectivity images. Then, the vortex/shear line type was similarly identified using the following two criteria: 1) the extreme hourly rainfall was located inside the central circulation of a vortex or near (within 200-km distance) a shear line at low levels (850 hPa over the plains and 700/500 hPa over the plateaus) and 2) the associated precipitation region extended over ~500 km. Finally, the remaining cases were classified into the weak synoptic type. While the weak synoptic type was associated with weaker synoptic-scale forcing than the other three types, the convective organization was further classified into two subgroups: mesoscale convective systems (MCSs) and small-scale storms. Here an MCS has a contiguous area of strong reflectivity (>40 dBZ) that is longer than 100 km in at least one dimension when the extreme hourly rainfall occurs. This definition of MCSs is essentially consistent with those in previous studies (e.g., Schumacher and Johnson 2005).
The TC-type extreme rainfall was identified for 1981–2015 using a computer program to apply the OSAT. The other three types need manual identification that is much more time-consuming and thus only conducted for 2011–15. Representative examples of the four extreme rainfall types will be shown in section 5 to illustrate the major features of their synoptic situations. Apart from the three types of synoptic weather system discussed above, some other synoptic and mesoscale conditions are also favorable for extreme rainfall to occur, such as a southwesterly low-level jet (LLJ; Chen and Yu 1988), an outflow boundary (Maddox et al. 1979; Xu et al. 2012; Wang et al. 2014), a surface cold pool (Schumacher et al. 2013; Luo and Chen 2015), and mountain-associated lifting (Pontrelli et al. 1999; Soderholm et al. 2014). However, their effects are not considered in this study.
3. Seasonality of hourly rainfall intensity and frequency
Figure 2 shows the monthly averaged hourly precipitation intensity during 1981–2013, which suggests a close relationship between the seasonal cycle of the rainfall intensity and the East Asian monsoon. A center of strong rainfall (>2.5 mm h−1) first appears over the south coasts and Hainan Island in April. The rainfall enhances and expands northward during May–July, reaching the maximal intensity in July with two intense centers in the southern coastal areas and the North China Plain, respectively. The rainfall weakens substantially after the end of August but keeps its intensity until mid-autumn south of 22°N. During winter (December–February), the hourly precipitation data are available only to the south of the Yangtze River and over the Sichuan basin where the mean rainfall intensity is weak (mostly < 1 mm h−1).
Spatial distribution of monthly mean of hourly rainfall intensity (mm h−1), averaged during 1981–2013, (a)–(i) from December to November. The gray dots denote stations with missing data. The thin and thick black lines represent terrain heights of 1 and 3 km, respectively. The blue lines denote locations of the Yangtze River and Yellow River.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
Unlike the seasonal features of hourly precipitation intensity described above, four regions with high occurrence frequency of hourly rainfall are found in different months throughout the year (Fig. 3). The most pronounced high-frequency region (>120 h month−1) appears as early as in March over southern China (approximately south of 30°N). Afterward, it gradually shrinks, moves toward the southeast coasts, and finally disappears in July. The second high-frequency region appears over the eastern flank of the Tibetan Plateau during summer. The third and fourth high-frequency regions are around the Sichuan basin and Hainan Island, respectively, during September and October. It is noticeable is that the rainfall intensities are relatively weak in these high-frequency regions (except for the south coasts), indicating that the large occurrence frequencies are contributed significantly by persistent stratiform precipitation with weak convection if there was any convection involved. Moreover, rainfall over northeast China occurs more frequently over the mountains than over the North China Plain.
As in Fig. 2, but for the rainfall probability (i.e., the number of hours with rain per day) in (a)–(i) each month from December to November averaged during 1981 to 2013.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
As described above, the seasonal variations of hourly precipitation intensity and occurrence frequency both vary regionally, but show different patterns over China. Thus nine representative subregions (areas A–I in Fig. 1) were chosen to further illustrate the regional differences and similarities. Figure 4 shows the normalized (divided by the annual mean) occurrence frequencies and statistics (top 0.1%, 0.5%, 1%, 5%, 10%) of local rainfall intensity in the nine subregions. Lower percentages correspond to moderate-to-weak rainfall intensities (<10 mm h−1) and are not shown.
Monthly variations of hourly rainfall in the nine subregions (areas A–I in Fig. 1). The upper part of each panel shows the number of hours with rain per day (h day−1) per station; the lower part shows the rainfall intensity with the middle of each bar representing the top 1% value, the top (bottom) of each bar indicating the top 0.5% (5%) value, and the top (bottom) dot showing the top 0.1% (10%) value.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
The subregions of the southeast coast (box A in Fig. 1) and south of the Yangtze River (box B in Fig. 1) are located in nearly the same latitudes and show a few similarities in the rainfall properties. They both have a peak of rainfall frequency with weak intensity in March. Rainfall intensity then increases monotonically with time, reaching a peak in August. The occurrence frequency of rainfall decreases sharply in July and remains low until February in both subregions. Precipitation on the south coast (box C in Fig. 1) during each month is more intense than that on the southeast coast, and remains at a strong level from April to September: the top 5% and 1% rainfall intensities exceed 10 and 20 mm h−1, respectively. Hainan Island (box D in Fig. 1) also shows intense precipitation during an even longer period from April to October; its rainfall frequency reaches the annual maximum in autumn. The Sichuan basin and the mountainous Yunnan subregions (circle E and box F in Fig. 1) are both located in southwest China. They present quite different rainfall properties, however, probably owing to their unique terrain features and different relationships with the East Asian summer monsoon; that is, precipitation in Sichuan is more closely related to the monsoon (Zhou et al. 2009). The precipitation frequency shows inconspicuous fluctuations in the Sichuan basin from May to October but a prominent single peak (in July) mode in Yunnan. As the rainfall intensity in the Sichuan basin exhibits substantial seasonal variations with a sharp peak in late summer (July–August), rainfall intensity in Yunnan shows little changes throughout the year and is the weakest among the nine subregions during the warm season (May–August). The remaining three subregions (boxes G, H, and I) similarly have a peak of rainfall occurrence in July and highest rainfall intensity in July and August.
4. Threshold, intensity, and frequency of extreme hourly precipitation
In the present study, the 99.9th percentile of hourly precipitation rates (>0.1 mm h−1) at a station during 1981–2013 is used to define the local threshold of extreme hourly rainfall for the station as described in section 2. Figure 5a shows the distribution of the 99.9th percentile value. The pattern is largely similar to the value for 5-yr return period (Li et al. 2013) and the 95th percentile for the warm season (Zhang and Zhai 2011). The largest threshold values are located at the south coasts, Hainan Island, and the North China Plain, with two secondary centers in the middle reaches of the Yangtze River and the western Sichuan basin. The maximal threshold value (68.8 mm h−1) is more than 8 times of the minimum (8 mm h−1). Compared with the four levels defined by the NMC for heavy hourly rainfall (20–30, 30–50, 50–80, and >80 mm h−1), there are 87.9%, 66.2%, and 7.8% of the stations having threshold values beyond 20, 30, and 50 mm h−1, respectively. Thus 50 or 80 mm h−1 seems to be a quite strict qualification to define the intense hourly precipitation for most stations over China. The distribution of the extreme hourly rainfall intensity (Fig. 5b) exhibits a similar pattern to that of the threshold (Fig. 5a), except for stronger contrasts between the plains (more intense) and mountains (weaker).
Spatial distributions of extreme hourly rainfall properties during 1981–2013: (a) value of 99.9th percentile (mm h−1), (b) mean intensity during the hours exceeding the 99.9th percentile (mm h−1), and (c) occurrence frequency (h a−1). The different shapes (dot, star, and triangle) represent the three groups of stations with data in different months: January–December, April–October, and May–September, respectively.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
Distribution of the extreme rainfall frequency (Fig. 5c) reflects the occurrence frequency of hourly precipitation in general because of the percentile threshold method used in the present study to define the extreme precipitation. The large value centers are located in three regions of different sizes. The most extensive one is oriented west–east to the south of the Yangtze River. The second one is over the Sichuan basin, especially in its southern portion, and the third one is around the southwest edge of Yunnan Province.
5. Synoptic situations associated with extreme hourly rainfall
During the five years 2011–15, 5797 extreme hourly rainfall records were found and their spatial distribution (not shown) is qualitatively consistent with that during 1981–2013 (Fig. 5c). Therefore, we believe that the 5-yr analysis results of the four types of weather patterns, including the spatial distribution and seasonal or diurnal variations, are reasonable. It is found that 8.0% of the extreme rainfall records (i.e., hours) were caused by TCs, 13.9% were associated with a surface front, 39.1% were related to a vortex or a shear line, and 39.0% occurred in an outwardly benign large-scale setting.
a. Examples
In this subsection, several representative examples of each extreme rainfall type will be described, with weather maps and composite radar reflectivity images, in order to demonstrate the major features of the synoptic patterns and the morphology of the rain systems.
1) TC type
Figure 6a shows two consecutive landfalling TCs, the twin Typhoons Saola (1209; the 9th typhoon of 2012) and Damrey (1210; the 10th typhoon of 2012), that caused hourly precipitation extremes in the cyclonic circulation around the storm center and near the coastal lines or the mountains, suggesting local intensification of rainfall and convection due to enhanced smaller-scale convergence related to the underlying inhomogeneous surface. Figure 6b shows a case of a landfalling TC encountering a midlatitude westerly trough. While the center of Supertyphoon Nanmadol (1111) stayed near the west of Taiwan Island on 29 August 2011, a north–south-oriented trough extended from northern China to the middle reach of the Yangtze River. One extreme rainfall record was found in the southeast coast near the typhoon center, whereas two extreme rainfall records were located near the bottom of the deep westerly trough. The latter were closely related to the moisture transport by the strong southeasterly flow in the northeastern periphery of the typhoon, which stretched the moisture channel toward the trough. Combined with the cyclonic vorticity, a thick, wet, and unstable layer was quite favorable for an increase in rainfall there (L. Chen et al. 2010).
Two examples of the TC type of extreme rainfall occurred on (a) 3 Aug 2012 and (b) 29 Aug 2011. The left half of each panel shows the synoptic circulation pattern at 500 hPa with the blue lines representing the geopotential height, the brown lines denoting the troughs or shear lines, along with red dots denoting locations of the extreme rainfall records. The gray shadings roughly represent topography with lighter denoting lower elevation. The right half of each panel shows the corresponding composite radar reflectivity (shading; unit: dBZ).
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
2) Frontal type
Figure 7a shows a cold front extending from northeast China to the Sichuan basin on 15 July 2013, with extreme rainfall occurring in both central and northern China along the front. The radar reflectivity distribution shows a narrow banded shape of precipitation along the front, reflecting an important role played by the frontal lifting of warm air in the rainfall production. In some cases, extreme precipitation occurred with a surface front extending from a low-level vortex center (e.g., on 2 May 2011; Fig. 7b). In such a case, the extreme precipitation records that were located inside the cyclonic circulation of the vortex belong to the vortex-shear line type, while those that were far away from the vortex center but near the surface front were classified as the frontal type.
As in Fig. 6, but for the frontal type of extreme rainfall that occurred on (a) 15 Jul 2013 and (b) 2 May 2011. The surface fronts are overlaid on both the (left) weather maps at 850 hPa and (right) radar reflectivity maps.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
3) Vortex/shear line type
Figure 8a shows a strong cyclone centered in central China dominating nearly the entire analysis domain and producing band-shaped precipitation in the south and rainy clusters in the north on 26 May 2013. The extreme rainfall mostly took place in the southern rainbands. This cyclone was weaker and smaller when it originated in the Sichuan basin the previous day. Such a vortex over the Sichuan basin, known as the southwest vortex in China, is a major rainfall-producing weather system over southwest China (Tao and Ding 1981). Formation of the southwest vortex is closely related to the low-level convergence of wind over the basin due to the blocking and deflecting effects exerted by the Tibetan Plateau (e.g., Wu and Chen 1985; Kuo et al. 1986; Chang et al. 1998). Figure 8b gives an example of extreme rainfall inside a cutoff low centered over northeast China on 2 July 2013. Such lows are known as northeast China cold vortices (usually identified at 500 hPa with a cold trough/center present) and exert important effects on the initiation and development of heavy rainfall and severe storms over the region during warm season (Sun 1997; Zhao and Sun 2007).
As in Fig. 6, but for the vortex/shear line type of extreme rainfall that occurred on (a) 26 May 2013 and (b) 2 Jul 2013. The weather map in (a) is at 850 hPa and in (b) is at 500 hPa. The surface cold fronts in (b) are overlaid on both the weather map at 500 hPa (at left) and radar reflectivity map (at right).
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
4) Weak-synoptic forcing type
Figure 9a shows two extreme precipitation records in southeast China on 15 May 2013, with prevailing southwesterly low-level flow that transports warm, moist air over the region. Occurrence of the extreme rainfall under this situation is probably associated with topographically triggered convection. Figure 9b shows another two examples of this extreme precipitation type in the mountainous southwest area of China, which occurred without pronounced features on the weather maps and inside nonorganized precipitating patches on the radar reflectivity image.
As in Fig. 6, but for the weak-synoptic type of extreme rainfall that occurred on (a) 15 May 2013 and (b) 26 Jul 2013. The weather map in (a) is at 850 hPa and in (b) is at 700 hPa.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
b. Spatial distribution
Figure 10 shows the spatial distribution of each extreme rainfall type during 2011–15. The majority of the TC-type extreme rainfall occurred in the southeast and south coasts and the occurrence frequency decreases progressively toward inland China (Fig. 10a). This is attributed to the fact that tropical cyclones tend to weaken after making landfalling due to decreased energy supply from the underlying surface and increased frictional effects. The TC type accounts for more than 30% of the total cases over the southeast coast and 20% in the south coastal regions. The 35-yr (1981–2015) distribution (Fig. 11) is generally consistent with the analysis of the five years regardless of many more samples being collected.
Spatial distributions of the occurrence frequency of the four extreme rainfall types (h a−1) during 2011–15: (a) TC, (b) frontal, (c) vortex/shear line, and (d) weak-synoptic. The colored symbols denote the numbers of annual occurrence while the black lines represent the fractional contribution (%) to the annual total occurrence of all the types. The different shapes (dot, star, and triangle) represent the three groups of stations with data in different months: January–December, April–October, and May–September, respectively.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
As in Fig. 10, but for the TC-type extreme precipitation averaged during 1981–2015 and with the monthly fractional contributions (%) to the total occurrences of the TC-type during 1981–2015 being shown in the top left corner.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
The frontal type is distributed relatively evenly east of 104°E without obvious high-frequency centers (Fig. 10b), as contrasts between warmer air and colder air brought about by northerly flows are found throughout the eastern China mainland along with the subseasonal progress and retreat of the Asian summer monsoon (Ding 1994). The vortex/shear line type shows a prominent frequency center around the Sichuan basin, where up to 70% of the extreme hourly rainfall is associated with a vortex/shear line (Fig. 10c). From this center, two high-frequency bands extend eastward and northeastward, respectively, to southeastern and northern China. The southwest vortex and associated shear line have been known as important rain producers for not only the local region, but also for the eastern, southern, and even northern China more broadly after they move out of the basin (Tao and Ding 1981). The results here suggest that they are also major contributors to the extreme hourly rainfall over these regions, even being a dominant one for the basin region. Furthermore, this type of extreme rainfall accounts for up to 50% in northeast China, reflecting its close association with the cold vortex activities there.
The weak-synoptic type has four high-frequency regions in southeast, southwest, and northern China and in the easternmost portion of northeast China, respectively (Fig. 10d), accounting for 40%–50% or even more than 50% of the total occurrence of the precipitation extremes at these regions. Occurrences of the weak-synoptic type extreme rainfall are expected to be associated with smaller-scale forcing (e.g., local topographical features) and local favorable atmospheric conditions (e.g., abundant moisture and sufficient convective available potential energy). However, different strengths of the forcing and conditions directly determine the diversity in convection structure. Compared with a smaller and less organized storm, an MCS may exert more extensive influence due to its larger size and longer duration. While a thorough comparison between the two subgroups is beyond the scope of this study, a preliminary analysis of their proportions and spatial distributions is presented here. The results show that the two subgroups are comparable in terms of the contributions to the weak-synoptic type occurrence [i.e., 50.4% (MCS) vs 49.6% (smaller storms)]. Their spatial distributions exhibit some notable differences although they both are quite widely distributed over the domain. The extreme-rain-producing MCSs prone to occur over the lower altitudes including the hilly lands in southeast China and the North China Plain (Fig. 12a), while the smaller storms tend to occur more frequently in the mountains over the eastern flank of the Tibetan Plateau and Yunnan (Fig. 12b).
As in Fig. 10, but for the two subgroups of weak-synoptic type extreme rainfall (yr−1) averaged during 2011–15: (a) MCSs and (b) smaller storms.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
c. Seasonal and diurnal variations
The monthly contributions (%) of the four synoptic types to the annual total hours of the extreme rainfall in the nine subregions averaged during 2011–15 are shown in Fig. 13. The seasonal variations exhibit distinct features among the subregions. In the southeast coast (subregion A), the extreme rainfall starts in March and ends in November, with a pronounced peak in August for the total occurrence and the weak-synoptic and TC types as well. The extreme hourly rainfall in the south of Yangtze River (subregion B) also starts in March and ends in November but peaks earlier (in June) than the southeast coast. The south coast (subregion C) peaks in May, dominated by the weak-synoptic type, reflecting the importance of the warm-sector heavy rainfall in the presummer rainy season (Ding 1994; Ding and Sikka 2006), whereas Hainan Island (subregion D) peaks in July. These two subregions have more contributions in March and November than the other subregions, due to occurrence of the weak-synoptic and vortex/shear line types in March and the TC type in November. The extreme rainfall in the Sichuan basin (subregion E), dominated by the vortex/shear line type, starts in April and ends in October with a pronounced peak in July. The extreme hourly precipitation in Yunnan (subregion F) mostly occurs in summer and early autumn with comparable monthly occurrence frequency from June to September. The three subregions in the north (the middle Huai River, the North China Plain, and the northeastern plains) have the extreme hourly precipitation concentrated in summer especially July and August with about 80% from the vortex/shear line and weak-synoptic types.
Fractional contributions (%) from the four synoptic-pattern types in each month to the annual total hours of extreme hourly precipitation in the nine subregions, with red, blue, purple, and green representing the TC, frontal, vortex/shear line, and weak-synoptic types, respectively. The locations of the nine subregions are shown in Fig. 1. The annual contributions (%) to the total hours of extreme precipitation from individual types are labeled in the upper-left corner of each panel.
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
Figure 14 shows the diurnal variations of normalized occurrence frequency (divided by the daily total) of the four synoptic pattern types in the nine subregions during 2011–15. The southeast coast, Hainan Island, and middle Huai River have a single afternoon peak of the total extreme rainfall occurrence at 1500–1800 local solar time (LST), with the peak in Hainan Island being the most pronounced under the weak-synoptic forcing. The Sichuan basin has a single nocturnal peak at 0000–0300 LST and a minimum around noon, largely determined by the diurnal variation of the vortex/shear line type. The other five subregions have double peaks, one in late afternoon (1500–1800) and the other around midnight (0000–0300 in the North China Plain and northeastern plains) or in early morning (0600–0900 south of the Yangtze River and at the south coast, and 0300–0600 in Yunnan). The midnight peaks in the Sichuan basin, North China Plain, and northeastern plains are closely related to the occurrences of the two synoptic-forcing types, which is probably contributed by the joint effects of the nocturnal LLJs associated with the synoptic weather systems and the mountain–plains solenoid (He and Zhang 2010; H. Chen et al. 2010). The weak-synoptic type tends to occur more often in late afternoon than other hours in most of the subregions (Figs. 14a–d,g,h). It is related to the diurnal variation of solar heating, which results in maximum low-level atmospheric instability and helps local intensification of rainfall and convection in the afternoon (Dai et al. 1999).
Diurnal variations of normalized occurrence frequency (divided by the daily total) of the four extreme-precipitation types in the nine subregions (shown in Fig. 1) during 2011–15, with red, blue, purple, and green represent the TC, frontal, vortex/shear line, and weak-synoptic types, respectively. Numbers at the bottom of each panel represent the hours (e.g., “00-03” means the hours from 0000 to 0300 LST, “21-00” from 2100 to 0000 LST).
Citation: Journal of Climate 29, 24; 10.1175/JCLI-D-16-0057.1
6. Summary and conclusions
The present study investigates hourly precipitation characteristics over China (east of 96°E). The seasonality of the hourly rainfall (>0.1 mm h−1) is analyzed using a quality-controlled dataset based on rain gauge observations over China during 1981–2015. The extreme hourly rainfall (defined using the 99.9th percentile as the local threshold) in 2011–15 is then classified into four types according to the synoptic situations under which they occurred by applying the objective synoptic analysis technique (Ren et al. 2011) and examining the weather maps and composite radar reflectivity data. The spatial distribution and seasonal/diurnal variations of each extreme-precipitation type are further investigated. The major conclusions are summarized below.
Seasonality of the hourly precipitation shows a close relationship of the rainfall intensity with the fluctuations of the East Asian summer monsoon and reaches its annual maximal strength in July. The rainfall occurrence shows four high-frequency regions: the most pronounced one over southern China during spring, the second one over the eastern flank of the Tibetan Plateau during summer, and the third and fourth ones around the Sichuan basin and Hainan Island, respectively, during September and October. Analyses for nine representative subregions further illustrate the complicated regional differences and similarities.
The extreme precipitation presents remarkable spatial features with two intense (>60 mm h−1) regions over the south coastal areas (including Hainan Island) and the North China Plain, and two secondary ones (45–50 mm h−1) in the middle reach of the Yangtze River and west Sichuan basin. Occurrence of the extreme precipitation shows substantial regional and seasonal variations, with 77% of the total in summer and a peak in July (30.4%).
There are 5797 extreme hourly rainfall records in 2011–15, among which 39.1% are related to a low-level vortex/shear line, 13.9% are associated with a surface front, 8.0% are caused by TC, and 39.0% occur under largely quiescent synoptic conditions. Each type of extreme precipitation presents distinctive characteristics in terms of regional distribution and seasonal or diurnal variations. Images of the composite radar reflectivity indicate comparable contributions to the total occurrence of the weak-synoptic type from MCSs and small-scale storms with notable differences in their preferred locations.
These results advance the understanding of hourly precipitation characteristics, especially the extreme rainfall properties, over China. They are also indispensable for validation and improvement of numerical models for weather forecast and climate prediction, as well as to help forecasters in their operational applications. In the future, a variety of properties of extreme rainfall events over China deserve further examination due to their close association with the resultant social impacts, such as the response of extreme precipitation to temperature, the relationship between probability of rainfall intensity and duration, the linkage between precipitation extremes and El Niño–Southern Oscillation (ENSO), and the initiation/propagation/organization of extreme-rain-producing storms under major types of synoptic pattern.
Acknowledgments
This study was supported by the National Natural Science Foundation of China (Projects 91437104 and 41221064), the National Basic Research Program of China (973 Program; 2012CB417202), and the National Key Technology Research and Development Program of China (2012BAC22B03). The weather maps and composite radar reflectivity data were obtained from the National Meteorological Center of China Meteorological Administration. We thank graduate student Ruoyun Ma at Chinese Academy of Meteorological Sciences for help partitioning the weak-synoptic type extreme precipitation into the mesoscale convective systems and smaller-scale storms.
REFERENCES
Chang, C.-P., S. C. Hou, H. C. Kuo, and G. T. J. Chen, 1998: The development of an intense East Asian summer monsoon disturbance with strong vertical coupling. Mon. Wea. Rev., 126, 2692–2712, doi:10.1175/1520-0493(1998)126<2692:TDOAIE>2.0.CO;2.
Chang, C.-P., Y. Lei, C.-H. Sui, X. Lin, and F. Ren, 2012: Tropical cyclone and extreme rainfall trends in East Asian summer monsoon since mid-20th century. Geophys. Res. Lett., 39, L18702, doi:10.1029/2012GL052945.
Chen, G. T.-J., and C.-C. Yu, 1988: Study of low-level jet and extremely heavy rainfall over northern Taiwan in the mei-yu season. Mon. Wea. Rev., 116, 884–891, doi:10.1175/1520-0493(1988)116<0884:SOLLJA>2.0.CO;2.
Chen, G. T.-J., C.-C. Wang, and D. T.-W. Lin, 2005: Characteristics of low-level jets over northern Taiwan in mei-yu season and their relationship to heavy rain events. Mon. Wea. Rev., 133, 20–43, doi:10.1175/MWR-2813.1.
Chen, G. T.-J., C.-C. Wang, and L.-F. Lin, 2006: A diagnostic study of a retreating mei-yu front and the accompanying low-level jet formation and intensification. Mon. Wea. Rev., 134, 874–896, doi:10.1175/MWR3099.1.
Chen, H., R. Yu, J. Li, W. Yuan, and T. Zhou, 2010: Why nocturnal long-duration rainfall presents an eastward-delayed diurnal phase of rainfall down the Yangtze River valley. J. Climate, 23, 905–917, doi:10.1175/2009JCLI3187.1.
Chen, L. S., Y. Li, and Z. Q. Cheng, 2010: An overview of research and forecasting on rainfall associated with landfalling tropical cyclones. Adv. Atmos. Sci., 27, 967–976, doi:10.1007/s00376-010-8171-y.
Dai, A., F. Giorgi, and K. E. Trenberth, 1999: Observed and model‐simulated diurnal cycles of precipitation over the contiguous United States. J. Geophys. Res., 104, 6377–6402, doi:10.1029/98JD02720.
Ding, Y., 1994: Monsoons over China. Kluwer Academic, 419 pp.
Ding, Y., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117–142, doi:10.1007/s00703-005-0125-z.
Ding, Y., and D. R. Sikka, 2006: Synoptic systems and weather. The Asian Monsoon, B. Wang, Ed., Praxis, 131–201.
Ding, Y., H. Li, Z. Cai, and J. Li, 1980: On the physical conditions of occurrence of heavy rainfall and severe convective weather. Proc. Eighth Conf. on Weather Forecasting and Analysis, Denver, CO, Amer. Meteor. Soc., 371–377.
He, H., and F. Zhang, 2010: Diurnal variations of warm-season precipitation over northern China. Mon. Wea. Rev., 138, 1017–1025, doi:10.1175/2010MWR3356.1.
Kuo, Y.-H., L. Cheng, and R. A. Anthes, 1986: Mesoscale analyses of the Sichuan flood catastrophe, 11–15 July 1981. Mon. Wea. Rev., 114, 1984–2003, doi:10.1175/1520-0493(1986)114<1984:MAOTSF>2.0.CO;2.
Li, J., R. Yu, and T. Zhou, 2008: Seasonal variation of the diurnal cycle of rainfall in southern contiguous China. J. Climate, 21, 6036–6043, doi:10.1175/2008JCLI2188.1.
Li, J., R. Yu, and W. Sun, 2013: Duration and seasonality of hourly extreme rainfall in the central eastern China. Acta Meteor. Sin., 27, 799–807, doi:10.1007/s13351-013-0604-y.
Luo, Y., and Y. Chen, 2015: Investigation of the predictability and physical mechanisms of an extreme-rainfall-producing mesoscale convective system along the Meiyu front in East China: An ensemble approach. J. Geophys. Res. Atmos., 120, 10 593–10 618, doi:10.1002/2015JD023584.
Maddox, R. A., C. F. Chappell, and L. R. Hoxit, 1979: Synoptic and meso-α scale aspects of flash flood events. Bull. Amer. Meteor. Soc., 60, 115–123, doi:10.1175/1520-0477-60.2.115.
Meng, Z., D. Yan, and Y. Zhang, 2013: General features of squall lines in east China. Mon. Wea. Rev., 141, 1629–1647, doi:10.1175/MWR-D-12-00208.1.
Min, S.-K., X. Zhang, F. W. Zwiers, and G. C. Hegerl, 2011: Human contribution to more-intense precipitation extremes. Nature, 470, 378–381, doi:10.1038/nature09763.
Pontrelli, M. D., G. Bryan, and J. M. Fritsch, 1999: The Madison County, Virginia, flash flood of 27 June 1995. Wea. Forecasting, 14, 384–404, doi:10.1175/1520-0434(1999)014<0384:TMCVFF>2.0.CO;2.
Ren, F., G. Wu, W. Dong, X. Wang, Y. Wang, W. Ai, and W. Li, 2006: Changes in tropical cyclone precipitation over China. Geophys. Res. Lett., 33, L20702, doi:10.1029/2006GL02795.
Ren, F., Y. Wang, X. Wang, and W. Li, 2007: Estimating tropical cyclone precipitation from station observations. Adv. Atmos. Sci., 24, 700–711, doi:10.1007/s00376-007-0700-y.
Ren, F., J. Liang, G. Wu, W. Dong, and X. Yang, 2011: Reliability analysis of climate change of tropical cyclone activity over the western North Pacific. J. Climate, 24, 5887–5898, doi:10.1175/2011JCLI3996.1.
Schumacher, R. S., and R. H. Johnson, 2005: Organization and environmental properties of extreme-rain-producing mesoscale convective systems. Mon. Wea. Rev., 133, 961–976, doi:10.1175/MWR2899.1.
Schumacher, R. S., A. J. Clark, M. Xue, and F. Kong, 2013: Factors influencing the development and maintenance of nocturnal heavy-rain-producing convective systems in a storm-scale ensemble. Mon. Wea. Rev., 141, 2778–2801, doi:10.1175/MWR-D-12-00239.1.
Soderholm, B., B. Ronalds, and D. J. Kirshbaum, 2014: The evolution of convective storms initiated by an isolated mountain ridge. Mon. Wea. Rev., 142, 1430–1451, doi:10.1175/MWR-D-13-00280.1.
Su, Z., F. Ren, J. Wei, X. Lin, S. Shi, and X. Zhou, 2016: Changes in monsoon and tropical cyclone extreme precipitation in southeast China from 1960 to 2012. Trop. Cyclone Res. Rev., 4, 12–17, doi:10.6057/2015TCRR01.02.
Sun, L., 1997: A study of the persistence activity of northeast cold vortex in China (in Chinese). Sci. Atmos. Sin., 21, 297–307.
Tao, S.-Y., and Y.-H. Ding, 1981: Observational evidence of the influence of the Qinghai-Xizang (Tibet) plateau on the occurrence of heavy rain and severe convective storms on China. Bull. Amer. Meteor. Soc., 62, 23–30, doi:10.1175/1520-0477(1981)062<0023:OEOTIO>2.0.CO;2.
Wang, H., Y. Luo, and B. J.-D. Jou, 2014: Initiation, maintenance, and properties of convection in an extreme rainfall event during SCMREX: Observational analysis. J. Geophys. Res. Atmos., 119, 13 206–13 232, doi:10.1002/2014JD022339.
Wang, Y., and L. Zhou, 2005: Observed trends in extreme precipitation events in China during 1961–2001 and the associated changes in large-scale circulation. Geophys. Res. Lett., 32, L09707, doi:10.1029/2005GL022574.
Wu, G.-X., and S.-J. Chen, 1985: The effect of mechanical forcing on the formation of a mesoscale vortex. Quart. J. Roy. Meteor. Soc., 111, 1049–1070, doi:10.1002/qj.49711147009.
Xu, W., E. J. Zipser, Y.-L. Chen, C. Liu, Y.-C. Liou, W. C. Lee, and B. Jou, 2012: An orography-associated extreme rainfall event during TiMREX: Initiation, storm evolution, and maintenance. Mon. Wea. Rev., 140, 2555–2574, doi:10.1175/MWR-D-11-00208.1.
Yu, R., and J. Li, 2012: Hourly rainfall changes in response to surface air temperature over eastern contiguous China. J. Climate, 25, 6851–6861, doi:10.1175/JCLI-D-11-00656.1.
Yu, R., Y. Xu, T. Zhou, and J. Li, 2007a: Relation between rainfall duration and diurnal variation in the warm season precipitation over central eastern China. Geophys. Res. Lett., 34, L13703, doi:10.1029/2007GL030315.
Yu, R., T. Zhou, A. Xiong, Y. Zhu, and J. Li, 2007b: Diurnal variations of summer precipitation over contiguous China. Geophys. Res. Lett., 34, L01704, doi:10.1029/2006GL028129.
Yuan, W., R. Yu, M. Zhang, W. Lin, H. Chen, and J. Li, 2012: Regimes of diurnal variation of summer rainfall over subtropical East Asia. J. Climate, 25, 3307–3320, doi:10.1175/JCLI-D-11-00288.1.
Zhai, P., X. Zhang, H. Wan, and X. Pan, 2005: Trends on total precipitation and frequency of daily precipitation extremes over China. J. Climate, 18, 1096–1108, doi:10.1175/JCLI-3318.1.
Zhang, H., and P. Zhai, 2011: Temporal and spatial characteristics of extreme hourly precipitation over eastern China in the warm season. Adv. Atmos. Sci., 28, 1177–1183, doi:10.1007/s00376-011-0020-0.
Zhao, S., and J. Sun, 2007: Study on cut-off low-pressure systems with floods over northeast Asia. Meteor. Atmos. Phys., 96, 159–180, doi:10.1007/s00703-006-0226-3.
Zheng, L., J. Sun, X. Zhang, and C. Liu, 2013: Organizational modes of mesoscale convective systems over central East China. Wea. Forecasting, 28, 1081–1098, doi:10.1175/WAF-D-12-00088.1.
Zhou, T., D. Gong, J. Li, and B. Li, 2009: Detecting and understanding the multi-decadal variability of the East Asian summer monsoon—Recent progress and state of affairs. Meteor. Z., 18, 455–467, doi:10.1127/0941-2948/2009/0396.
