Analysis of Paths and Sources of Moisture for the South China Rainfall during the Presummer Rainy Season of 1979–2014

+ Author Affiliations + Find other works by these authors
  • Corresponding author: Yali LUO, ylluo@cma.gov.cn
  • Funds:

    Supported by the National Natural Science Foundation of China (91437104 and 41775050) and Basic Research and Operational Practice Funds of the Chinese Academy of Meteorological Sciences (2017Z006)

  • doi: 10.1007/s13351-018-8069-7

PDF

  • The paths and sources of moisture supplied to South China during two periods of the presummer rainy season (April–June) of 1979–2014, i.e., before and after the onset of the summer monsoon over the South China Sea (SCS), are investigated by using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model. During the premonsoon-onset period, the moisture transport trajectories are clustered into 6 groups, with four ocean-originating paths providing 83.9% and two continent-originating paths (originating over Lake Baikal and the Persian Gulf) contributing the remaining 16.1% of the total moisture. The two Pacific-originating paths, from the western Pacific Ocean and the East China Sea, combined account for about 46%, the SCS-originating path contributes about 24.3%, while the Bay of Bengal-originating path accounts for 13.6% of the total moisture over South China. The trajectories during the postmonsoon-onset period are clustered into 4 groups, with three southwesterly paths (from the Arabian Sea, the central Indian Ocean, and the western Indian Ocean, respectively) accounting for more than 76% and the sole Pacific-originating path accounting for 23.8% of the total moisture. The formation of the moisture transport trajectories is substantially affected by the topography, especially the Tibetan Plateau and the Indian and Indo–China Peninsulas. The SCS region contributes the most moisture during both periods (35.3% and 31.1%). The Pacific Ocean is ranked second during the former period (about 21.0%) but its contribution is reduced to 5.0% during the latter period, while the contribution from the Bay of Bengal and the Indian Ocean combined increases from 17.1% to 43.2%.
  • 加载中
  • Fig. 1.  Schematic of an air-parcel trajectory passing through the grid cells. The green line denotes the trajectory and solid circles with an embedded cross denote the locations of the air parcel along this trajectory at eight sequential times (t1, t2, t3, t4, t5, t6, t7, and t8).

    Fig. 2.  Spatial distributions of the rainfall rate (shading, mm day–1) averaged over the selected rainfall days during (a) the premonsoon-onset period and (b) the postmonsoon-onset period.

    Fig. 3.  Spatial distributions of the 500-hPa geopotential height (blue solid contour lines, interval of 40 gpm), 850-hPa water vapor flux (shading, m s–1 kg kg–1), and horizontal wind (vectors, m s–1) averaged over the selected rainfall days during (a) the premonsoon-onset period and (b) the postmonsoon-onset period. The purple box in each panel indicates the target area around South China.

    Fig. 4.  Test log-likehood values as a function of the number of trajectory clusters in (a) the premonsoon-onset period and (b) the postmonsoon-onset period.

    Fig. 5.  Mean trajectories (bold lines) of (a) the six groups during the premonsoon-onset period and (b) the four groups during the postmonsoon-onset period. The shadings denote the trajectory frequency (%). The black circles denote the mean position of the air parcels in each group of the backward trajectories at the time of the first (–24 h), the fourth (–96 h), the seventh (–168 h), and the l0th day (–240 h) before arriving the target area. The percentages inside (outside) the brackets indicate the fractions of trajectory number of each group (the moisture contribution to the target area) in all trajectories.

    Fig. 6.  Topography (shaded) over East Eurasia and adjacent oceans, with the Tibetan Plateau, Indian Peninsula, and Indo–China Peninsula labeled. The claret-red lines and black contour, respectively, denote the group mean trajectories and the trajectory frequency of 0.5% during the (a) premonsoon-onset and (b) postmonsoon-onset periods. The white circles denote the mean position of the air parcels in each group of the backward trajectories at the time of the first (–24 h), the fourth (–96 h), the seventh (–168 h), and the l0th day (–240 h) before arriving in the target area. The black rectangle denotes the boundaries of the target area.

    Fig. 7.  Cluster mean (a) altitude and (b) specific humidity along the six groups of trajectories from 10 days prior (–240 h) to the moment that the air particles reach the target area (0 h) during the premonsoon-onset period.

    Fig. 8.  As in Fig. 7, but for the four groups of backward trajectories during the postmonsoon-onset period.

    Fig. 9.  Spatial distributions of the specific humidity changes (shading, g kg–1) along the trajectories of (a–f) groups 1–6 during the premonsoon-onset period. The red line indicates the mean trajectory of each group and the red circles denote the mean positions of the air parcels in each group at the time of the first (–24 h), the fourth (–96 h), the seventh (–168 h), and the l0th (–240 h) day before arriving in the target area. The blue contours indicate the trajectory frequencies (only the frequencies greater than 2% are shown). The red, dashed rectangle denotes the boundaries of the target area.

    Fig. 10.  As in Fig. 9, but for the four groups during the postmonsoon-onset period.

    Fig. 11.  (a) Distribution of the six regions over which moisture contribution to the target area is analyzed in the present study. (b) Contributions of the six regions to the total moisture of the South China rainfall during the premonsoon- (red columns) and postmonsoon-onset period (blue columns). The number above each column indicates the contribution of the corresponding region.

    Table 1.  Onset pentad of the South China Sea summer monsoon between 1979 and 2014

    YearOnset pentadYearOnset pentadYearOnset pentad
    197927199132200328
    198027199228200428
    198130199330200527
    198226199425200626
    198329199528200728
    198428199626200828
    198530199727200929
    198627199829201029
    198732199930201129
    198829200028201225
    198928200127201327
    199028200227201427
    Download: Download as CSV
  • [1]

    Brimelow, J. C., and G. W. Reuter, 2005: Transport of atmosphericmoisture during three extreme rainfall events over the Mackenzie River basin. J. Hydrometeorol., 6, 423–440. doi: 10.1175/JHM430.1.
    [2]

    Brubaker, K. L., D. Entekhabi, and P. S. Eagleson, 1993: Estimation of continental precipitation recycling. J. Climate, 6, 1077–1089. doi: 10.1175/1520-0442(1993)006<1077:EOCPR>2.0.CO;2.
    [3]

    Chang, Y., J. H. He, Y. Y. Liu, et al., 2006: Features of moisture transport of in pre-summmer flood season of drought and flood years over South China. Plateau Meteor., 25, 1064–1070. doi: 10.3321/j.issn:1000-0534.2006.06.013. (in Chinese)
    [4]

    Chen, X. C., Y. Xu, C. H. Xu, et al., 2014: Assessment of precipitation simulations in China by CMIP5 multi-models. Climatic Environ. Res., 10, 217–225. doi: 10.3969/j.issn.1673-1719.2014.03.011. (in Chinese)
    [5]

    Chi, Y. Z., J. H. He, and Z. W. Wu, 2005: Features analysis of the different precipitation periods in the pre-flood season in South China. J. Nanjing Inst. Meteor., 28, 163–171. doi: 10.3969/j.issn.1674-7097.2005.02.003. (in Chinese)
    [6]

    Chow, K. C., H. W. Tong, and J. C. L. Chan, 2008: Water vapor sources associated with the early summer precipitation over China. Climate Dyn., 30, 497–517. doi: 10.1007/s00382-007-0301-6.
    [7]

    Dee, D. P., S. M. Uppala, A. J. Simmons, et al., 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597. doi: 10.1002/qj.828.
    [8]

    Ding, Y. H., 1992: Summer monsoon rainfalls in China. J. Meteor. Soc. Japan, 70, 373–396. doi: 10.2151/jmsj1965.70.1B_373.
    [9]

    Ding, Y. H., 1994: Monsoons over China. Kluwer Academic Publishers, Dordrecht, 419 pp.
    [10]

    Ding, Y. H., 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.
    [11]

    Drumond, A., R. Nieto, L. Gimeno, et al., 2008: A Lagrangian identification of major sources of moisture over Central Brazil and La Plata Basin. J. Geophys. Res. Atmos., 113, D14128. doi: 10.1029/2007JD009547.
    [12]

    Drumond, A., R. Nieto, E. Hernandez, et al., 2011: A Lagrangian analysis of the variation in moisture sources related to drier and wetter conditions in regions around the Mediterranean Basin. Nat. Hazards Earth Syst. Sci., 11, 2307–2320. doi: 10.5194/nhess-11-2307-2011.
    [13]

    Eagleson, P. S., 1970: Dynamic Hydrology. McGraw-Hill, Inc., New York, 462 pp.
    [14]

    Gaffney, S., 2004: Probabilistic curve-aligned clustering and prediction with mixture models. Ph. D. dissertation, Department of Computer Science, University of California, Irvine, 281 pp.
    [15]

    Gaffney, S. J., A. W. Robertson, P. Smyth, et al., 2007: Probabilistic clustering of extratropical cyclones using regression mixture models. Climate Dyn., 29, 423–440. doi: 10.1007/s00382-007-0235-z.
    [16]

    Gao, X. J., M. L. Wang, and F. Giorgi, 2013: Climate change over China in the 21st century as simulated by BCC_CSM1.1-RegCM4.0. Atmos. Ocean. Sci. Lett., 6, 381–386. doi: 10.3878/j.issn.1674-2834.13.0029.
    [17]

    Gómez-Hernández, M., A. Drumond, L. Gimeno, et al., 2013: Variability of moisture sources in the Mediterranean region during the period 1980–2000. Water Resour. Res., 49, 6781–6794. doi: 10.1002/wrcr.20538.
    [18]

    Gustafsson, M., D. Rayner, and D. L. Chen, 2010: Extreme rainfall events in southern Sweden: Where does the moisture come from? Tellus, 62, 605–616. doi: 10.1111/j.1600-0870.2010.00456.x.
    [19]

    Huang, S. S., 1986: Heavy Rainfall over Southern China in the Pre-Summer Rainy Season. Guangdong Science and Technology Press, Guangzhou, 244 pp. (in Chinese)
    [20]

    Huang, Y. J., and X. P. Cui, 2015: Moisture sources of torrential rainfall events in the Sichuan Basin of China during summers of 2009–13. J. Hydrometeorol., 16, 1906–1917. doi: 10.1175/jhm-d-14-0220.1.
    [21]

    Izquierdo, R., A. Avila, and M. Alarcón, 2012: Trajectory statisticalanalysis of atmospheric transport patterns and trends in precipitation chemistry of a rural site in NE Spain in 1984–2009. Atmos. Environ., 61, 400–408. doi: 10.1016/j.atmosenv.2012.07.060.
    [22]

    Jiang, Z. H., S. Jiang, Y. Shi, et al., 2017: Impact of moisture source variation on decadal-scale changes of precipitation in North China from 1951 to 2010. J. Geophys. Res. Atmos., 122, 600–613. doi: 10.1002/2016JD025795.
    [23]

    Li, X. Z., W. Zhou, and Y. Q. Chen, 2016: Detecting the origins of moisture over southeast China: Seasonal variation and heavy Rainfall. Adv. Atmos. Sci., 33, 319–329. doi: 10.1007/s00376-015-4197-5.
    [24]

    Liu, B. Q., C. W. Zhu, Y. Yuan, et al., 2016: Two types of interannual variability of South China Sea summer monsoon onset related to the SST anomalies before and after 1993/94. J. Climate, 29, 6957–6971. doi: 10.1175/JCLI-D-16-0065.1.
    [25]

    Liu, R. X., J. H. Sun, J. Wei, et al., 2016: Classification of persistent heavy rainfall events over South China and associated moisture source analysis. J. Meteor. Res., 30, 678–693. doi: 10.1007/s13351-016-6042-x.
    [26]

    Liu, Y. J. and Y. H. Ding, 2000: Evolution of the atmospheric stratification and mixed layer before and after monsoon onset over the South China Sea. Climatic Environ. Res., 5, 459–468. doi: 10.3969/j.issn.1006-9585.2000.04.011. (in Chinese)
    [27]

    Liu, Y. J., Y. H. Ding, and Y. L. Song, 2005: The moisture transport and moisture budget over the South China Sea before and after the summer monsoon onset in 1998. J. Trop. Meteor., 21, 55–62. doi: 10.3969/j.issn.1004-4965.2005.01.006. (in Chinese)
    [28]

    Luo, Y. L., H. Wang, R. H. Zhang, et al., 2013: Comparison of rainfall characteristics and convective properties of monsoon precipitation systems over South China and the Yangtze and Huai River Basin. J. Climate, 26, 110–132. doi: 10.1175/JCLI-D-12-00100.1.
    [29]

    Luo, Y. L., R. H. Zhang, Q. L. Wan, et al., 2017: The Southern China monsoon rainfall experiment (SCMREX). Bull. Amer. Meteor. Soc., 98, 999–1013. doi: 10.1175/BAMS-D-15-00235.1.
    [30]

    Martius, O., H. Sodemann, H. Joos, et al., 2013: The role of upper-level dynamics and surface processes for the Pakistan flood of July 2010. Quart. J. Roy. Meteor. Soc., 139, 1780–1797. doi: 10.1002/qj.2082.
    [31]

    Nieto, R., L. Gimeno, and R. M. Trigo, 2006: A Lagrangian identification of major sources of Sahel moisture. Geophys. Res. Lett., 33, L18707. doi: 10.1029/2006GL027232.
    [32]

    Pfahl, S., and H. Wernli, 2008: Air parcel trajectory analysis of stable isotopes in water vapor in the eastern Mediterranean. J. Geophys. Res. Atmos., 113, D20104. doi: 10.1029/2008JD009839.
    [33]

    Salih, A. A. M., Q. Zhang, and M. Tjernström, 2015: Lagrangian tracing of Sahelian Sudan moisture sources. J. Geophys. Res. Atmos., 120, 6793–6808. doi: 10.1002/2015JD023238.
    [34]

    Sodemann, H., C. Schwierz, and H. Wernli, 2008: Interannual variability of Greenland winter precipitation sources: Lagrangian moisture diagnostic and North Atlantic Oscillation influence. J. Geophys. Res. Atmos., 113, D03107. doi: 10.1029/2007JD008503.
    [35]

    Sodemann, H., and E. Zubler, 2009: Seasonal and inter-annual variability of the moisture sources for Alpine precipitation during 1995–2002. Int. J. Climatol., 30, 947–961. doi: 10.1002/joc.1932.
    [36]

    Stein, A. F., R. R. Draxler, G. D. Rolph, et al., 2015: NOAA’s HYSPLIT atmospheric transport and dispersion modeling system. Bull. Amer. Meteor. Soc., 96, 2059–2077. doi: 10.1175/BAMS-D-14-00110.1.
    [37]

    Sun, B., and H. J. Wang, 2014: Moisture sources of semiarid grassland in China using the Lagrangian particle model FLEXPART. J. Climate, 27, 2457–2474. doi: 10.1175/JCLI-D-13-00517.1.
    [38]

    Sun, B., and H. J. Wang, 2015: Analysis of the major atmospheric moisture sources affecting three sub-regions of East China. Int. J. Climatol., 35, 2243–2257. doi: 10.1002/joc.4145.
    [39]

    Sun, J. H., H. J. Wang, J. Wei, et al., 2016: The sources and transportation of water vapor in persistent heavy rainfall events in the Yangtze–Huaihe River valley. Acta Meteor. Sinica, 74, 542–555. doi: 10.11676/qxxb2016.047. (in Chinese)
    [40]

    Tian, H., P. W. Guo, and W. S. Lu, 2004: Characteristics of vapor inflow corridors related to summer rainfall in China and impact factors. J. Trop. Meteor., 20, 401–408. doi: 10.3969/j.issn.1004-4965.2004.04.008. (in Chinese)
    [41]

    Trenberth, K. E., 1999: Atmospheric moisture recycling: Role of advection and local evaporation. J. Climate, 12, 1368–1381. doi: 10.1175/1520-0442(1999)012<1368:AMRROA>2.0.CO;2.
    [42]

    Trenberth, K. E., A. G. Dai, R. M. Rasmussen, et al., 2003: The changing character of precipitation. Bull. Amer. Meteor. Soc., 84, 1205–1216. doi: 10.1175/BAMS-84-9-1205.
    [43]

    Wang, J., J. H. He, X. F. Liu, et al., 2009: Interannual variability of the Meiyu onset over Yangtze–Huaihe River valley and analyses of its previous strong influence signal. Chinese Sci. Bull., 54, 687–695. doi: 10.1007/s11434-008-0534-8.
    [44]

    Wu, J., and X. J. Gao, 2013: A gridded daily observation dataset over China region and comparison with the other datasets. Chinese J. Geophys., 56, 1102–1111. doi: 10.6038/cjg20130406. (in Chinese)
    [45]

    Xie, A., Y. S. Chung, X. Liu, et al., 1998: The interannual variations of the summer monsoon onset over the South China Sea. Theor. Appl. Climatol., 59, 201–213. doi: 10.1007/s007040050024.
    [46]

    Zheng, B., J. Y. Liang, A. L. Lin, et al., 2006: Frontal rain and summer monsoon rain during pre-rainy season in South China. Part I: Determination of the division dates. Chinese J. Atmos. Sci., 30, 1207–1216. doi: 10.3878/j.issn.1006-9895.2006.06.15. (in Chinese)
    [47]

    Zhou, T. J., and R. C. Yu, 2005: Atmospheric water vapor transport associated with typical anomalous summer rainfall patterns in China. J. Geophys. Res. Atmos., 110, D08104. doi: 10.1029/2004JD005413.
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Analysis of Paths and Sources of Moisture for the South China Rainfall during the Presummer Rainy Season of 1979–2014

    Corresponding author: Yali LUO, ylluo@cma.gov.cn
  • 1. State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081
  • 2. University of Chinese Academy of Sciences, Beijing 100049
  • 3. Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing 210044
Funds: Supported by the National Natural Science Foundation of China (91437104 and 41775050) and Basic Research and Operational Practice Funds of the Chinese Academy of Meteorological Sciences (2017Z006)

Abstract: The paths and sources of moisture supplied to South China during two periods of the presummer rainy season (April–June) of 1979–2014, i.e., before and after the onset of the summer monsoon over the South China Sea (SCS), are investigated by using the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model. During the premonsoon-onset period, the moisture transport trajectories are clustered into 6 groups, with four ocean-originating paths providing 83.9% and two continent-originating paths (originating over Lake Baikal and the Persian Gulf) contributing the remaining 16.1% of the total moisture. The two Pacific-originating paths, from the western Pacific Ocean and the East China Sea, combined account for about 46%, the SCS-originating path contributes about 24.3%, while the Bay of Bengal-originating path accounts for 13.6% of the total moisture over South China. The trajectories during the postmonsoon-onset period are clustered into 4 groups, with three southwesterly paths (from the Arabian Sea, the central Indian Ocean, and the western Indian Ocean, respectively) accounting for more than 76% and the sole Pacific-originating path accounting for 23.8% of the total moisture. The formation of the moisture transport trajectories is substantially affected by the topography, especially the Tibetan Plateau and the Indian and Indo–China Peninsulas. The SCS region contributes the most moisture during both periods (35.3% and 31.1%). The Pacific Ocean is ranked second during the former period (about 21.0%) but its contribution is reduced to 5.0% during the latter period, while the contribution from the Bay of Bengal and the Indian Ocean combined increases from 17.1% to 43.2%.

1.   Introduction
  • In general, moisture for precipitation over a region arises from three main sources: local evaporation, the moisture already presents in the atmosphere over the region, and the moisture transported into the region by atmospheric advection (Brubaker et al., 1993; Trenberth, 1999). On a long-term timescale, local evaporation accounts for only a small fraction of moisture for precipitation around the globe while moisture transport is the dominant contributor (Trenberth, 1999), especially for heavy rainfall events. During heavy rainfall events, rainfall-producing weather systems can gather moisture over the region approximately 3–5 times the radius of the precipitating area (Trenberth et al., 2003).

    The presummer rainy season, covering April to June over South China (SC), is the first monsoon rainy season over China (Ding, 1994). During this rainy season, SC often experiences flash floods due to the frequent occurrence of heavy rainfall events that produce large amounts of precipitation. The seasonal rainfall accumulation, on average, reaches up to about 1000 mm, accounting for 40%–50% of the annual total rainfall in SC (Huang, 1986; Luo et al., 2017). The transport of moisture in SC rainfall during this season is associated with the Indian summer monsoon flow, the South China Sea (SCS) summer monsoon (SCSSM) flow, and the subtropical monsoon flow (Tian et al., 2004; Zhou and Yu, 2005; Chow et al., 2008). Moreover, the prevailing winds over East and South Asia show abrupt changes around the onset of the SCSSM (Ding, 1992; Liu and Ding, 2000; Ding and Chan, 2005), which often occur in middle-to-late May (Xie et al., 1998; Wang et al., 2009; Luo et al., 2013). Therefore, characteristics of the water vapor transport to SC differ substantially during the two periods, i.e., before and after the onset of SCSSM (referred to as the premonsoon- and postmonsoon-onset periods hereinafter). Specifically, water vapor is mainly transported via the westerly winds across the Arabian Sea and the southwesterly winds along the southwest edge of the western Pacific subtropical high (WPSH) for the early period, while water vapor is mainly transported from the cross-equatorial flow over the Bay of Bengal and SCS for the late period (Chi et al., 2005; Liu et al., 2005; Chang et al., 2006). The earlier period is mainly characterized by frontal precipitation, probably due to the influence of cold air from the north. This is in contrast to the monsoonal precipitation seen during the later period (Zheng et al., 2006). However, these conclusions are mainly based on the analysis of water vapor flux in the frame of the Eulerian method, in which air parcels of a target precipitation region cannot be traced backward to the possible source regions. This means that the sources of moisture for the precipitation, and their trajectories, cannot be determined quantitatively.

    Recently, Lagrangian methods have been developed to detect the moisture sources of a target region and these have been successfully applied to study moisture sources and paths in many regions around the world, e.g., over the Americas (Brimelow and Reuter, 2005; Drumond et al., 2008), the continent of Europe (Gustafsson et al., 2010; Izquierdo et al., 2013), the Mediterranean region (Drumond et al., 2011; Gómez-Hernández et al., 2013), and Africa (Nieto et al., 2006, Salih et al., 2015). Lagrangian methods have also been used to analyze moisture sources and their variability over many sub-regions of China, e.g., East China (Sun and Wang, 2015; Sun et al., 2016), North China (Jiang et al., 2017), the Sichuan Basin (Huang and Cui, 2015), and the semiarid grasslands of China (Sun and Wang, 2014). For SC, Liu R. X. et al. (2016) analyzed the moisture sources of 16 non-typhoon influenced persistent heavy rainfall events that mainly affected two subregions within SC. Li et al. (2016) used one-year data to detect the sources which supply moisture to Southeast China in the summer and winter half year, respectively. They further analyzed the moisture sources of heavy rainfall events in Southeast China influenced by tropical storms, the WPSH and cold air in the summer, and by strong cold air in the winter. However, Langrangian analysis of the moisture sources and paths of the presummer rainfall over SC in the climatological mean state, looking at the differences between the pre- and postmonsoon-onset periods in particular, has received limited attention in the literature.

    The objective of this study is to present a 36-yr (1979–2014) climatological study of the transport paths and sources of moisture for the SC precipitation during the presummer rainy season. The study utilizes a Lagrangian model, named the Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) (Stein et al., 2015) (https://www.arl.noaa.gov/hysplit), with a focus on comparisons between the pre- and postmonsoon-onset periods. The rest of this paper is organized as follows: Section 2 describes the data and methodology, Section 3 illustrates the major features of rainfall patterns and associated circulation patterns for the two periods, Section 4 shows the moisture trajectories and sources of the SC rainfall during the two periods based on the quantitative analysis of the HYSPLIT results, and Section 5 provides the conclusions and discussion.

2.   Data and methodology
  • A gridded precipitation dataset CN05.1 (Wu and Gao, 2013) with a horizontal resolution of 0.25 degree is used in this study. The CN05.1 is based on the rain gauge records at more than 2000 surface weather stations over China mainland and thus can be regarded as observation. This dataset has been widely used to validate the regionalclimate model performance when simulating the precipitation over China (e.g., Gao et al., 2013; Chen et al., 2014). The atmospheric reanalysis dataset used here is the ECMWF Interim Re-Analysis dataset (ERA-Interim) (Dee et al., 2011). The ERA-Interim data is obtained at 0.75° horizontal and 6-h temporal resolutions. The variables used in the present study include terrestrial height, geopotential height, zonal, meridional, and vertical wind components, and specific humidity.

    The presummer rainy season (April–June) is separated into the pre- and postmonsoon-onset periods by the onset date of the SCSSM. The onset date is defined as the first pentad when the area-averaged meridional temperature gradient (MTG) in the mid-to-upper troposphere (500–200 hPa) over the South China Sea (10°–20°N, 110°–120°E) changes from negative to positive phase after pentad 21 and remains positive for at least three pentads (Liu B. Q. et al., 2016). The MGT is monitored in real-time by the National Climate Center (NCC) of China (http://cmdp.ncc-cma.net/Monitoring/monsoon_ mtg.php) and has been used as an indicator of the SCSSM onset. Table 1 lists the onset pentad of SCSSM for each year. The pre- and postmonsoon-onset periods between 1979 and 2014, respectively, contain 1617 and 1479 days. The onset pentad of SCSSM is not included in either period. Only moderate and heavy rainfall events are included when analyzing moisture sources and transport paths, as light rainfall events are, in general, characterized by less evident moisture transport. In this analysis, the days with an area-mean daily rainfall greater than 10 mm day–1 over the target area (21°–26°N, 106°–117°E), are selected as the moderate-and-heavy rainfall days. In total, 379 and 499 rainy days are selected for the pre- and postmonsoon-onset periods, respectively, which account for about 23.4% and 33.7% of the total days in each period.

    YearOnset pentadYearOnset pentadYearOnset pentad
    197927199132200328
    198027199228200428
    198130199330200527
    198226199425200626
    198329199528200728
    198428199626200828
    198530199727200929
    198627199829201029
    198732199930201129
    198829200028201225
    198928200127201327
    199028200227201427

    Table 1.  Onset pentad of the South China Sea summer monsoon between 1979 and 2014

    As the moisture transport is mainly contained in the lower troposphere, air parcels released over the target area with 7 × 15 horizontal coordinates on 3 vertical levels (925, 850, and 700 hPa) are tracked backwards every 6 h to obtain a good representation of the moisture transport pathways into the target region. Every backward-trajectory calculation is run for 10 days, which is the approximate time for the moisture to remain in the atmosphere (Eagleson, 1970). For each vertical column of the model grid, the geopotentials of the model pressure levels are converted to heights above the surface by using Eq. (1),

    $${\rm{AGL = }}\frac{{{\varPhi _p}}}{g} - z,$$ (1)

    where ${\varPhi _p}$ is the geopotential height for pressure level p, g stands for the gravity acceleration, and z is the ground elevation, which can be directly obtained from the ERA-Interim dataset. The model output includes the three-dimensional location (latitude, longitude, altitude) and the specific humidity for the air parcels at a 1-h interval. Since not all backward trajectories serve as moisture sources for the SC rainfall, only effective trajectories with specific humidity decreases over the target area are taken into consideration in the following analysis.

    In total, 206,555 and 320,305 effective trajectories are obtained for the 379 and 499 rainy days during the pre- and postmonsoon-onset periods, respectively. In order to gain a better understanding of the water vapor paths for the SC rainfall, the trajectories are clustered with the Curve Clustering Toolbox (Gaffney, 2004) to obtain several group mean trajectories, following Li et al. (2016). In the clustering process, the number of clusters (k) is determined by the in-sample log-likelihood value. This value, akin to a goodness-of-fit metric for probabilistic models, is expressed as the log-probability of the origi-nal data under the model (Gaffney, 2004). As the number of clusters k increases, the log-likelihood values increase and show diminishing improvement in fit for k higher than a certain value, which leads to an optimal range of k choices (Gaffney et al., 2007). After clustering of the trajectories, the water vapor contribution ratio of each cluster to the total moisture transported to the target area is calculated as follows:

    $${\rm Contrubution} = \bigg(\sum\limits_1^m {\Delta {q_ {\rm{sc}}}} /\sum\limits_1^n {\Delta {q_ {\rm{sc}}}}\bigg),$$ (2)

    where $\Delta {q_{\rm sc}}$is the net specific humidity change of a trajectory over the target area, m is the number of effective trajectories in the individual cluster, and n is the number of effective trajectories in all clusters.

    In order to examine the moisture changes (i.e., net evaporation) along the trajectories in greater detail, the grid cell mean moisture change is estimated under a three step procedure. Firstly, the moisture change between any two adjacent output moments (i.e., time t and time t – 1) along each trajectory is calculated by

    $${\rm d}{q_t} = {q_t} - {q_{t - 1}}, $$ (3)

    where qt and qt – 1 are the air specific humidity at the time t and time t – 1, respectively. Then, for each trajectory, the moisture change in each 0.75-degree grid cell is obtained by adding up all ${\rm d}{q_t}$ within the grid cell. For example, in Fig. 1, the green line denotes one trajectory, and the solid circles with crosses denote the locations of the air parcel along this trajectory at the times t1, t2, t3, t4, t5, t6, t7, and t8. The moisture change of the air parcel along this trajectory, as it passes through grid cells A (dqA), B (dqB), C (dqC), and D (dqD), is calculated as,

    Figure 1.  Schematic of an air-parcel trajectory passing through the grid cells. The green line denotes the trajectory and solid circles with an embedded cross denote the locations of the air parcel along this trajectory at eight sequential times (t1, t2, t3, t4, t5, t6, t7, and t8).

    $${\rm d}{q_{\rm A}} = {\rm d}{q_2} + {\rm d}{q_3}, $$ (4)
    $${\rm d}{q_{\rm B}} = {\rm d}{q_4}, \qquad \,\,$$ (5)
    $${\rm d}{q_{\rm C}} = {\rm d}{q_5} + {\rm d}{q_6}, $$ (6)
    $${\rm d}{q_{\rm D}} = {\rm d}{q_7} + {\rm d}{q_8},$$ (7)

    where ${\rm d}{q_2}$, ${\rm d}{q_3}$, ${\rm d}{q_4}$, ${\rm d}{q_5}$, ${\rm d}{q_6}$, ${\rm d}{q_7}$, and ${\rm d}{q_8}$ are the moisture changes at the times t2, t3, t4, t5, t6, t7, and t8.. Finally, the moisture changes of all trajectories passing through each grid cell are averaged to obtain the grid cell mean moisture change.

3.   Rainfall and circulation patterns
  • The distributions of the average rainfall amount during the selected rainy days in the pre- and postmonsoon-onset periods are shown in Fig. 2. For the premonsoon-onset period, a single rainfall maximum, where the daily accumulated rainfall exceeds 25 mm day–1, is located in central Guangdong Province. This rainfall center still exists but tends to shift southeastward after the onset of SCSSM. Another rainfall center appears over northern Guangxi Region during the postmonsoon-onset period.

    Figure 2.  Spatial distributions of the rainfall rate (shading, mm day–1) averaged over the selected rainfall days during (a) the premonsoon-onset period and (b) the postmonsoon-onset period.

    Figure 3 gives circulation features during the selected rainfall days of the two periods, showing the magnitude of water vapor flux and the wind vectors at 850 hPa and the 500-hPa geopotential height. During the premonsoon-onset period, three water-vapor channels for SC can be observed, encompassing a westerly channel originating from the north Arabian Sea, a southwesterly channel from the Indian Ocean, and a southeasterly channel along the edge of the WPSH from the Pacific Ocean. After the onset of SCSSM, significant cross-equatorial flow brings water vapor from the Arabian Sea, Indian Ocean, and SCS into the target area. The southeasterly channel in the earlier period is no longer evident during this later perioddue to the significant influence of the cross-equatorial flow and the northward jump of WPSH from roughly 18°N to 25°N. These findings are qualitatively consistent with previous studies (Chi et al., 2005; Chang et al., 2006).

    Figure 3.  Spatial distributions of the 500-hPa geopotential height (blue solid contour lines, interval of 40 gpm), 850-hPa water vapor flux (shading, m s–1 kg kg–1), and horizontal wind (vectors, m s–1) averaged over the selected rainfall days during (a) the premonsoon-onset period and (b) the postmonsoon-onset period. The purple box in each panel indicates the target area around South China.

4.   Moisture paths and sources for the SC rainfall detected by HYSPLIT
  • The curve in Fig. 4a shows diminishing returns in terms of improvement in fit beyond k = 6, suggesting that this is a reasonable stopping point for the cluster analysisof the trajectories during the premonsoon-onset period. Figure 5a illustrates the six group mean moisture transport paths and the distribution of trajectory frequency. The trajectory frequency is expressed as the ratio of the number of trajectories that fall within each grid cell to the total number of trajectories. The grid resolution for the trajectory frequency is 0.75 degree × 0.75 degree. The group mean trajectories roughly pass through the areas with high trajectory frequency (Fig. 5a), demonstrating the effectiveness of the clustering method and the rationality of the number of clusters (k = 6).

    Figure 4.  Test log-likehood values as a function of the number of trajectory clusters in (a) the premonsoon-onset period and (b) the postmonsoon-onset period.

    Figure 5.  Mean trajectories (bold lines) of (a) the six groups during the premonsoon-onset period and (b) the four groups during the postmonsoon-onset period. The shadings denote the trajectory frequency (%). The black circles denote the mean position of the air parcels in each group of the backward trajectories at the time of the first (–24 h), the fourth (–96 h), the seventh (–168 h), and the l0th day (–240 h) before arriving the target area. The percentages inside (outside) the brackets indicate the fractions of trajectory number of each group (the moisture contribution to the target area) in all trajectories.

    Overall, the four oceanic moisture paths, namely, the southwesterly path (group 1) originating from the Bay of Bengal, the shorter southeasterly path (group 2) from the SCS, the longer southeasterly path (group 3) from the western North Pacific, and the northeasterly path (group 4) from the East China Sea, combine to contribute more than 80% of the moisture for the target area during the premonsoon-onset period (Fig. 5a). The two Pacific-originating paths (group 3 and group 4) jointly contribute 46% of the moisture, suggesting that the Pacific Ocean, including the adjacent seas of China, is the important moisture source for South China rainfall during the premonsoon-onset period. The longer southeasterly path (group 3) reaches the target area by crossing the Philippines and the SCS, while the northeasterly path (group 4) flows parallel to the coastline of China and turns westward over the SCS before arriving in the target area. For the other two oceanic paths, the shorter southeasterly path (group 2), passing by the SCS and the east edge of the Indo–China Peninsula before reaching the target area, contributes 24.3% of the total moisture; while the southwesterly path (group 1), passing through the middle of the Indo–China Peninsula, accounts for 13.6% of the total moisture.

    The remaining two paths have continental origins and jointly contribute approximately 16.1% of the total moisture. The northwesterly path (group 5), which starts in the vicinity of Lake Baikal, and passes southeastward through Siberia, before traversing the East China coast, contributes 9.0% to the total moisture. The westerly path (group 6) that originates in the Persian Gulf and passes over northern Indian Peninsula and northern Indo–China Peninsula, accounts for 7.1% of the total moisture transported to the target area. Moreover, due to the detouring flow around the Tibetan Plateau, most of the trajectories in groups 5 and 6 run along the northeast and south edges of the Tibetan Plateau (Fig. 6a).

    Figure 6.  Topography (shaded) over East Eurasia and adjacent oceans, with the Tibetan Plateau, Indian Peninsula, and Indo–China Peninsula labeled. The claret-red lines and black contour, respectively, denote the group mean trajectories and the trajectory frequency of 0.5% during the (a) premonsoon-onset and (b) postmonsoon-onset periods. The white circles denote the mean position of the air parcels in each group of the backward trajectories at the time of the first (–24 h), the fourth (–96 h), the seventh (–168 h), and the l0th day (–240 h) before arriving in the target area. The black rectangle denotes the boundaries of the target area.

    Of interest is the change of moisture along the trajectories. The moisture might increase due to underlying surface evaporation but, on the other hand, decrease due to condensation/precipitation; i.e., it results from the net evaporation. Condensation is usually associated with upwards flowing air that can be reflected by an increase in the altitude of air parcels along the trajectories. Therefore, Fig. 7 shows the group averaged altitude and specifichumidity for the air parcels during the 10 days of backward tracking. It is not surprising to find that the altitudes of all paths increase during the first day of backward tracking (Fig. 7a), suggesting impacts of the upward motion associated with the SC rainfall. There is a moisture increase by net evaporation along the ocean-originating trajectories during the fifth to tenth day of the backward tracking, suggesting the contribution of moisture from the oceans (Fig. 7b). The altitudes of the four ocean-originating paths (light and dark green lines, blue line, and purple line in Fig. 7a) commence at approximately 1.0–1.6 km AGL and remain under for approximately 1.0 km during the second to seventh days of the backward tracking, indicating the importance of air flows in the planetary boundary layer (PBL) for moisture transport. One exception is the southwesterly path (group 1; light green line in Fig. 7a), which begins to ascend around the fourth day when it arrives over the Indo–China Peninsula. This is probably associated with topographical lifting (Fig. 6a).

    Figure 7.  Cluster mean (a) altitude and (b) specific humidity along the six groups of trajectories from 10 days prior (–240 h) to the moment that the air particles reach the target area (0 h) during the premonsoon-onset period.

    The continent-originating air parcels of the westerly (claret-red lines in Fig. 7) and the northwesterly (black lines in Fig. 7) paths (groups 5 and 6) commence at approximately 3–4 km AGL with a smaller amount of moisture (about 1/3 to 1/2) compared to the ocean-originating air parcels. The northwesterly path (group 5) experiences a rapid increase in the amount of moisture during the second to seventh days of the backward tracking, which can be attributed to evaporation over the Chinese Bohai Sea, the Yellow Sea, and the East China Sea (Fig. 5a). The westerly path (group 6) also gains moisture continuously, but with a small magnitude of moisture increase. Further details about the moisture gain/loss along the trajectories will be presented in Section 4.3, whilst the contributions of six relevant regions to the total moisture over the target area will be discussed in Section 4.4.

  • Four clusters of moisture transport paths into the target area are identified during the postmonsoon-onset period (Fig. 5b), based on the evolution of the test log-likelihood values (Fig. 4b). All four group mean paths (Fig. 5b) are oceanic, with three southwesterly paths (groups 1, 2, 3) and one easterly path (group 4). Almost all of the trajectories of the three southwesterly paths are located over the southern edge of the Tibetan Plateau, suggesting the Plateau creates a blocking effect on the monsoon flow during this period (Fig. 6b). Two of the three southwesterly paths originate over the Arabian Sea (group 1) and the central Indian Ocean (group 2). They cross the Bay of Bengal and the Indo–China Peninsula sequentially before arriving in the target area. The third southwesterly path (group 3) originates over the eastern Indian Ocean near Sumatra and crosses the southeastern Indo–China Peninsula and SCS before arriving in the target area. The sole easterly path (group 4) originates over the Northwest Pacific Ocean and reaches the target area without passing over any major land areas. Amongst all four paths, group 3 contributes the largest fraction (about 31.7%) of the total moisture, while the other three paths make roughly equal contributions (group 1, 21.9%; group 2, 22.8%; group 4, 23.8%). The East China Sea-originating path and the two continent-originating paths seen during the premonsoon-onset period are no longer observed during the postmonsoon-onset period. The missing of the two northerly paths during the postmonsoon-onset period indicates that the association between the SC rainfall and cold air in the latter period is perhaps less significant than that in the former period. Moisture contributions from the southwest increase substantially during the latter period (76.2%) compared with the former period (13.6%), meaning that the southwesterly moisture paths become the key moisture channels for the target area after the onset of the SCSSM. The group-averaged altitude and specific humidity along the trajectories during the postmonsoon-onsest period are shown in Fig. 8. All the group mean trajectories are largely confined to the PBL (below about 1.5 km), which is similar to the ocean-originating trajectories in the premonsoon-onset period (cf. Figs. 7a, 6a). The trajectory from the eastern Indian Ocean (group 3; blue line in Fig. 8a) is limited to less than 600 m, and is characterized by the highest moisture content (blue line in Fig. 8b). This is consistent with its largest contribution to the total moisture over the target area. The Northwest Pacific-originating path (group 4) gains moisture continuously along the trajectory until it arrives in the target area, with the specific humidity increasing from 12 g kg–1 to approximately 15 g kg–1 (purple blue line in Fig. 8b). A significant increase in moisture is also observed in the group 2 trajectories (black line in Fig. 8b) when they spend roughly a week over the Indian Ocean and Bay of Bengal moving northeastward. The group 1 and group 2 trajectories (green and black lines, respectively, in Fig. 8) both experience notable height increases and moisture decreases during the second to third day (Figs. 5, 8) when they pass over the Indo–China Peninsula (Fig. 5b). This is likely due to the topographical lifting in that region as the group 1 trajectories in the premonsoon-onset period experience the same (Fig. 6).

    Figure 8.  As in Fig. 7, but for the four groups of backward trajectories during the postmonsoon-onset period.

  • Figure 9 shows details of the spatial distribution of moisture changes along each group of trajectories for the premonsoon-onset period. The trajectories of the southwesterly path (Fig. 9a) are distributed widely over the Bay of Bengal where they obtain abundant moisture. Later on, these trajectories tend to converge towards the target area and lose a significant amount of moisture when passing over the Indo–China Peninsula. This is associated with the topographical precipitation in that region (Fig. 6a). The southeasterly trajectories are distributed to the south of the target area, covering the SCS, the Indo–China Peninsula, and the Philippine Islands (Fig. 9b). These trajectories also experience some moisture loss over the Indo–China Peninsula, although with a lower amplitude than that of the southwesterly path (Fig. 9a). The two groups of Pacific-originating trajectories (Figs. 9c, d) are similarly characterized by persistent moisture increase over the western North Pacific and the adjacent Chinese seas before reaching the target area. One of the two groups originating further eastward over the Pacific, moves westward toward the SCS, and reaches the target area mostly over its southern boundary (Fig. 9c). Originating to the northeast of the target area, the other group reaches the target area mostly across its southern and eastern boundaries, but also its northern boundary (Fig. 9d). The northwesterly trajectory moves towards the southeast after originating around Lake Baikal, changing to a southwestward flow over the East China Sea, before finally entering the target area mostly over its southern and eastern boundaries (Fig. 9e). These trajectories begin to gain moisture when the air parcels are located close to the East China coast, with the moisture increase continuing until the trajectories arrive in the target area (Fig. 9e). Most of the westerly trajectories are distributed close to the group mean, and experience multiple cycles of moisture loss and gain before entering the target area through its western and southern boundaries (Fig. 9f).

    Figure 9.  Spatial distributions of the specific humidity changes (shading, g kg–1) along the trajectories of (a–f) groups 1–6 during the premonsoon-onset period. The red line indicates the mean trajectory of each group and the red circles denote the mean positions of the air parcels in each group at the time of the first (–24 h), the fourth (–96 h), the seventh (–168 h), and the l0th (–240 h) day before arriving in the target area. The blue contours indicate the trajectory frequencies (only the frequencies greater than 2% are shown). The red, dashed rectangle denotes the boundaries of the target area.

    Figure 10 is similar to Fig. 9 except for the four ocean-originating groups of trajectories during the postmonsoon-onset period. The three groups from the southwest experience moisture gain from the ocean and moisture loss from the land especially when passing over large terrain such as the Indian Peninsula and the Indo–China Peninsula (Figs. 10ac, 7b). The fourth group of trajectories originates over an area extending from the East China Sea to the west North Pacific, although its group mean shows that the original location is between Taiwan Island and the Philippine Islands (Fig. 10d). These trajectories are marked by persistent moisture increase over the western North Pacific and the adjacent seas of China, without obvious moisture loss, before reaching the target area.

    Figure 10.  As in Fig. 9, but for the four groups during the postmonsoon-onset period.

  • The above results suggest that during the 10-day backward tracking, the air parcels may undergo multiple cycles of evaporation and precipitation, i.e., an increase and decrease of specific humidity along the trajectory before reaching the target area. Therefore, earlier moisture source regions may contribute less and less to the precipitation over the target area due to the moisture changes caused by precipitation/evaporation along the moisture transport way. Here, we use the source attribution method proposed by Sodemann et al. (2008) to calculate the contribution from six relevant regions to the water vapor in the target area during the presummer rainy season, similar to previous studies (Pfahl and Wernli, 2008; Sodemann and Zubler, 2009; Martius et al., 2013). The six regions defined include the Indian Ocean, the Bay of Bengal (BoB), the South China Sea (SCS), the Pacific Ocean, eastern China (EC), and Eurasia (Fig. 11a). For both the pre- and postmonsoon-onset periods, the SCS region accounts for the largest fraction of the total moisture supply for the target area (about 35.3% and 31.1% in the former and latter periods, respectively). The Pacific Ocean ranks second during the premonsoon-onset period contributing about 21.0% to the total moisture; however its contribution significantly reduces to about 5.0% during the postmonsoon-onset period. In contrast, the BoB and the Indian Ocean experience an apparent increase in the contribution to the total moisture after the onset of the SCSSM and become the second (24.7%) and third (18.5%) largest contributors, respectively, during the postmonsoon-onset period. As previously analyses mentioned in Section 3, the significant cross-equatorial flow after the SCSSM onset brings abundant moisture from the Indian Ocean to SC, while the moisture from the Pacific Ocean is reduced, probably related to the northward jump of the WPSH (Fig. 3b). With the northward relocation of the WPSH, moisture from the Pacific Ocean tends to be transported to central East China along the edge of the WPSH during the latter period. EC makes comparable contributions to the total moisture during the two periods (18.1% and 15.9%, respectively), while Eurasia contributes only 8.5% and 4.8% of the total moisture during the former and latter periods.

    Figure 11.  (a) Distribution of the six regions over which moisture contribution to the target area is analyzed in the present study. (b) Contributions of the six regions to the total moisture of the South China rainfall during the premonsoon- (red columns) and postmonsoon-onset period (blue columns). The number above each column indicates the contribution of the corresponding region.

5.   Conclusions and discussion
  • The paths and sources of moisture for rainfall over South China (SC) during the presummer rainy season (April–June) between 1979 and 2014 are investigated by using the HYSPLIT model to perform backward tracking of air parcels released over SC, with a focus on comparisons between two periods, i.e., the premonsoon-onset period and the postmonsoon-onset period (before/after the onset of SCSSM). The major conclusions are as follows.

    (1) During the premonsoon-onset period, the trajectories are clustered into six groups. Four of the paths originate over the Bay of Bengal, the South China Sea, the western North Pacific, and the East China Sea. These ocean-originating paths account for 83.9% of the total moisture over SC. The remaining two paths originate over Lake Baikal and the Persian Gulf. The trajectories during the postmonsoon-onset period are clustered into four groups that originate over the Arabian Sea, the central Indian Ocean, the eastern Indian Ocean near Sumatra, and the Northwest Pacific Ocean. After the SCSSM onset, the contribution to the total moisture amount over SC by the Pacific-originating trajectories decreases from about 46.0% to 23.8%, while the contribution from the southwest trajectories increases from 15.1% to 76.1%.

    (2) The formation of moisture transport trajectories is greatly affected by topography. During the premonsoon-onset period, formation of the westerly and northerly trajectories is closely associated with the detouring effect of the Tibetan Plateau. During the postmonsoon-onset period,the Indian Ocean-originating trajectories turn eastward to converge at the target area at least partially due to the blocking effect of the Tibetan Plateau. Moreover, air parcels along the trajectories often gain moisture over the oceans but lose moisture over the land especially terrains of the Indian and Indo–China Peninsulas, while the spatial distribution of moisture change along the trajectories exhibits distinctive features among the groups during each period.

    (3) Moisture contribution from six regions that are of relevance to the moisture contribution for the SC rainfall are quantitatively examined, namely, the Indian Ocean, the Bay of Bengal (BoB), the South China Sea (SCS), the Pacific Ocean, eastern China (EC), and Eurasia. The SCS region contributes the most (about one third) in both periods. After the SCSSM onset, the moisture contribution from the BoB and Indian Ocean substantially increases from 17.1% to 43.2% while that of the Pacific Ocean decreases from 21.0% to 5.0%.

    Caution is needed when applying the source attribution method. As stated in Sodemann et al. (2008), the sources of only part of the moisture transported to the target area (SC in our study) can be considered with this method as the sources at the end locations of the backward-tracking trajectories cannot be determined. In this study, about 69.3% and 64.5% of the total moisture transported to SC can be accounted for during the pre- and postmonsoon-onset periods, respectively. Both fractions are comparable or larger than those obtained by previous studies (e.g., Pfahl and Wernli, 2008; Sodemann and Zubler, 2009). Therefore, the quantitative analysis of the moisture paths and sources for the SC rainfall during the presummer rainy season using the Lagrangian trajectory model is meaningful. By comparing the difference between the two periods of the presummer rainy season, i.e., the pre- and postmonsoon-onset periods, this study adds to our knowledge of the changes in moisture transport and source of the SC rainfall around the SCSSM onset.

  • Acknowledgments The ERA-Interim data was downloaded from http://apps.ecmwf.int/datasets. The Lagran-gian model named the Hybrid Single-Particle Lagran-gian Integrated Trajectory (HYSPLIT) was obtained from https://www.arl.noaa.gov/hysplit.

Reference (47)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return