Boreal Summer Intraseasonal Oscillation and Its Possible Impact on Precipitation over Southern China in 2019

Due to its location, China is strongly influenced by the East Asian summer monsoon (EASM) during boreal summer. The seasonal march of the EASM is accompanied by persistent heavy precipitation events that frequently occur over eastern China and often result in flooding in some regions. As one of the vital components of the EASM system, the intraseasonal oscillation (ISO) plays an important role in the occurrence and evolution of the large-scale circulation anomaly (Ding, 1993; Li et al., 2016), which leads to persistent heavy precipitation over eastern China (Tao and Xu, 1962; Zhu et al., 2003; Tao and Wei, 2006; Chen and Wei, 2012). The ISO in East Asia presents two distinct periods with 10–20 and 30–60 days (Lawrence and Webster, 2002), which modulate EASM activity with episodes of abundant precipitation in eastern China.

Many previous case studies have pointed out that the ISO periods differ between regions of eastern China and can depend on stages in the flood season (Lu and Ding, 1996; Chen et al., 2000; Xin et al., 2007; Wen et al., 2011; Chen et al., 2015; Li and Mao, 2019). Lu and Ding (1996) indicated that an 18-day ISO dominated the heavy rain over the Yangtze–Huaihe River basin during the summer of 1991. Chen et al. (2000) suggested that the summer precipitation in the Yangtze–Huaihe River basin exhibited a 40-day ISO feature in 1979, but it was 12–24 days in 1989. Wang and Ding (2008) showed that the period of the ISO of summer precipitation over the mid–lower reaches of the Yangtze River valley is relatively longer in the flood years, with the 30–60-day oscillation dominating. In the drought years, however, it displays a shorter period with a 10–30-day oscillation. Xin et al. (2007) demonstrated that precipitation in South China during the flood season of 1997 presented different low-frequency oscillation features in pre- and latter-flood seasons. Pre-flood season precipitation was mainly controlled by the 10–20-day oscillation, whereas precipitation in the latter-flood season showed no significant low-frequency oscillation. Zhang et al. (2012) analyzed the features of general circulation with ISO during the torrential rainstorms over southern China in 2010 and proposed that the climatological evolution of summer monsoon (first mode) coincided with ISO (second and third modes), ultimately leading to frequent rainstorms. However, it has been demonstrated that multi-scale ISO can exist simultaneously (Kajikawa and Yasunari, 2005; Chen et al., 2015) and jointly affect persistent precipitation (Kulkarni et al., 2011). The intensity of a quasi-biweekly oscillation in summer precipitation has significantly strengthened over southern China since 1993 (Chen et al., 2015). The influence of the westward or northwestward propagation of the 10–20-day oscillation on precipitation anomalies has received much more attention (Wen et al., 2011; Jia and Yang, 2013; Wang et al., 2017; Yan et al., 2019).

Precipitation in China during boreal summer 2019 had uneven spatial distribution and exhibited significant ISO (Ding and Gao, 2020). In July, in particular, seven regional torrential rain processes occurred, mostly in southern China, and extreme daily precipitation was measured at many stations (Xie and Zhou, 2019). Given that the ISO plays a key part in persistent heavy precipitation in southern China, the present study aims to investigate its impact on summer precipitation in 2019 and reveal the possible causes of abnormal precipitation over southern China. The remainder of the study is organized as follows. Section 2 describes the data and methods used in this study. The features of summer precipitation over China in 2019 and related atmospheric circulation anomalies are presented in Section 3. Section 4 analyzes the ISO features of southern China precipitation in summer. Section 5 explores the distinct ISO features of atmospheric circulation. The final section presents the conclusions and discussion.

The daily precipitation observation data from 2374 stations in China that are used in this study were provided by the National Meteorological Information Center. The atmospheric circulation dataset derives from the NCEP/NCAR Reanalysis 1 (NCEP-1; Kalnay et al., 1996), which includes geopotential height, wind, specific humidity, and vertical velocity, with a horizontal resolution of 2.5° × 2.5°. The daily outgoing longwave radiation (OLR) data with a horizontal resolution of 2.5° × 2.5° derive from the NOAA (Liebmann and Smith, 1996). The research covers the period from 1981 to 2019. Climatology represents the period from 1981 to 2010.

Morlet wavelet period analysis (Torrence and Compo, 1998), as a functional technique for time–frequency decomposition of a time series, is applied to analyze the time series of summer precipitation in southern China to identify the dominant ISO periods. The Lanczos band-pass filter (Duchon, 1979) is used to extract 10–20- and 30–60-day periods.

Monthly and total precipitation anomaly percentages over China in summer (June–August) 2019 are shown in Fig. 1. Eastern China exhibits an anomaly pattern that is positive in the south and negative in the north, with a significant precipitation anomaly observed to the south of the Yangtze River basin (Fig. 1d). The monthly precipitation anomalies indicate that the above-normal precipitation in southern China is the most significant in July, and the precipitation amount is more than twice the climatology (Fig. 1b). The spatial distribution characteristics of the precipitation anomaly in July are also similar to those of the entire summer. Above-normal precipitation in southern China can also be observed in June (Fig. 1a), but the precipitation anomaly magnitude is apparently less than that in July. Precipitation over southern China turns into a deficient phase in August. Therefore, the summer precipitation anomaly over southern China in 2019 is mainly concentrated in July. An investigation into the cause of the anomalous precipitation over southern China in July is the key to understanding the summer precipitation anomaly in 2019. The domain with the most significant positive precipitation anomaly in July (shown by the blue dotted stations in Fig. 2a) is selected to represent southern China. A standardized precipitation index in southern China is calculated by the area-averaging precipitation anomaly in July (Fig. 2b). The precipitation index in 2019 exceeds twice the standard deviation, and reaches the peak since 1981.

Figure 1.  Precipitation anomaly percentages in China in (a) June, (b) July, (c) August, and (d) summer 2019.

Figure 2.  (a) The station distributions in southern China marked by blue dots and (b) the yearly variation of standardized precipitation anomaly index over southern China in July.

Figure 3a shows the wind anomalies at 850 hPa and vertical velocity anomalies at 500 hPa in July 2019. There is a significant low-level anomalous cyclone located from eastern China to Japan. The western North Pacific subtropical high (WNPSH) is apparently stronger than normal, as well as extending farther westward and southward. The water vapor transport from the Bay of Bengal, South China Sea, and western North Pacific results in an apparent convergence anomaly over southern China (Fig. 3b). Consistent with the enhanced convergence in the lower troposphere, a more active vertical ascending motion can be found in the middle troposphere, which is conducive to more precipitation in southern China.

Figure 3.  (a) Wind anomalies (vectors; m s⁻¹) at 850 hPa and vertical velocity anomalies (shading; Pa s⁻¹) at 500 hPa in July 2019. Contours denote 5880-gpm geopotential height at 500 hPa. The red dashed and solid lines denote climatology and July 2019, respectively. (b) Vertically integrated water vapor flux anomalies (vectors; kg s⁻¹ m⁻¹) and moisture divergence anomalies (shading; 10⁻⁵ kg s⁻¹ m⁻²) in July 2019.

The heavy precipitation in the EASM region is closely linked to the atmospheric low-frequency oscillation. The ISO in flood years is obviously stronger than that in drought years (Wang and Ding, 2008). A wavelet spectrum analysis reveals distinct ISO activities of summer precipitation in southern China (Fig. 4). The 10–20-day ISO from June to mid-July is significant, and the 30–60-day period from mid-June to mid-July also reaches the 90% confidence level (Fig. 4a). Figure 4b shows that the 10–20-day periodic oscillation of summer precipitation passes the significance test. The explained variances of the 10–20- and 30–60-day ISO are 30.7% and 19.3%, respectively, which indicate that the 10–20-day ISO is the more significant during the summer of 2019.

Figure 4.  (a) Wavelet analysis of precipitation (Pre) over southern China. The bold dashed lines indicate the 90% confidence level for a red-noise process and the bold solid line denotes the cone of influence outside which the edge effects become important. (b) Power spectrum of precipitation over southern China. The dashed line indicates the 90% confidence level for a red-noise process.

To investigate the effect of different oscillation periods on summer precipitation over southern China in 2019, the daily precipitation is filtered into different periods in Fig. 5. As shown in Fig. 5, heavy precipitation occurs frequently in southern China from early June to mid-July, but the strongest precipitation process with the longest duration occurs in the first half of July. The amplitude of the 10–20-day oscillation of precipitation in southern China is larger than that of the 30–60-day oscillation, particularly in early summer. The daily precipitation in the first half of July shows the correspondence of the wet phase of the 30–60-day oscillation with an entire cycle of the 10–20-day oscillation. Moreover, the peak of the 10–20-day oscillation coincides with the peak of the 30–60-day oscillation. The explained variances of the 10–20- and 30–60-day ISO are respectively 31.5% and 27.7% during the first half of July 2019, when both periods are important with dual-mode feature presented in Mao and Chan (2005). Thus, the ISO with both 10–20- and 30–60-day periods could exert a synergistic effect on the persistent heavy precipitation over southern China in July. Therefore, the following analysis will focus on the influence of the associated atmospheric circulation evolution with the two different periods on the anomalous precipitation over southern China in July 2019.

Figure 5.  Time series of daily rainfall anomalies over southern China (histogram; mm day⁻¹) and the 10–20- and 30–60-day filtered anomalies during summer 2019.

As mentioned above, the 10–20-day oscillation of precipitation in southern China was more significant than the 30–60-day oscillation during the summer of 2019. Therefore, we first focus on the evolution and propagation characteristics of the atmospheric circulation in the 10–20-day period. An entire cycle of quasi-biweekly oscillation can be observed from June 30 to July 16 (Fig. 5). The valley of the dry phase occurs on June 30, and July 4 is the transition period from dry to wet phases. After reaching the crest of the wet phase on July 8, the oscillation turns into the dry phase on July 12 and reaches another valley of dry phase on July 16. The eight different phases are synthesized separately following the definition of Chan et al. (2002). Phase 1 (5) refers to the driest (wettest) phase of precipitation in southern China, and Phase 3 (7) refers to the phase when the oscillation changes from dry (wet) to wet (dry) episodes with a value of 0. Phases 2, 4, 6, and 8 are the phases when the oscillation reaches half of the crest or valley.

Figure 6 shows the evolution of 850-hPa wind field and OLR anomalies after 10–20-day filtering in each phase. In Phase 1, an anomalous low-level cyclone accompanied by active convection obviously exists in the South China Sea, and inactive convection presents in southern China (Fig. 6a). From Phases 2 to 3, the easterly wind in the western equatorial Pacific strengthens remarkably and an anomalous low-level anticyclone occurs over the western North Pacific (Figs. 6b, c). Meanwhile, convection weakens in the South China Sea, and the anomalous cyclone over the South China Sea attenuates. From Phases 4 to 6 (Figs. 6d–f), the anomalous anticyclone develops northward and westward, accompanied by the easterly wind anomalies in the Philippine Sea propagating westward and northward. The WNPSH also strengthens and extends westward and southward, which enhances water vapor transport from the western North Pacific to southern China during Phases 4–6 (Figs. 6d–f). The feature of the WNPSH extending westward agrees with the results observed by Mao et al. (2010) and Ren et al. (2013). As a result, convection is further suppressed near the Philippine Sea and the South China Sea during these same phases. Subsequently, from Phases 7 to 8, the westerly wind develops in the western equatorial Pacific and the WNPSH weakens and retreats northward and eastward. Convection becomes active in the Philippine Sea and the South China Sea, with an anomalous low-level cyclone occurring there. Water vapor transport and its convergence are important for precipitation. Following the method of Ren et al. (2020), we calculated the vertically integrated water vapor flux and divergence anomalies from Phases 1 to 8. The well-organized moisture divergence (convergence) is consistent with the suppressed (enhanced) convection in Fig. 6 (omitted).

Figure 6.  Composite 10–20-day filtered 850-hPa wind (vectors; m s⁻¹), OLR (shading; W m⁻²) anomalies from 30 June to 16 July 2019. Contours denote 5880-gpm geopotential height at 500 hPa. Green and blue solid lines denote 2019 and climatology, respectively. (a)–(h) represent Phases 1–8, respectively.

The enhanced ascending (descending) motion plays an important dynamic role for abundant (deficient) precipitation. Thus, the vertical features of 10–20-day filtered circulation anomalies are analyzed. Figure 7 shows the vertical motion and divergence anomalies averaged between 110° and 120°E. In Phase 1, a clear vertical–meridional circulation cell presents with an ascending branch south of 20°N and a descending branch over southern China. Significant low-level convergence and upper-level divergence anomalies are associated with the ascending branch over the South China Sea, while the descending branch results in upper-level convergence and low-level divergence anomalies over southern China (Fig. 7a). In Phase 2, the intensity of the descending branch over southern China sharply weakens, and ascending motion anomalies occur in the north of southern China (Fig. 7b). Subsequently, during the wet phase (Phases 3–6), the anomalous divergence in the upper troposphere is replaced by anomalous strong convergence, and anomalous divergence dominates in the lower troposphere over the South China Sea (Figs. 7c–f). The induced strong descending motion over the South China Sea, together with the ascending motion located over southern China, constitutes a reverse vertical–meridional circulation cell of Phase 1. From Phases 7 to 8, the convergence anomalies intensify in the upper troposphere over southern China where descending motion prevails (Figs. 7g–h).

Figure 7.  The vertical–meridional cross-section of 10–20-day filtered vertical velocity (10⁻² Pa s⁻¹) and meridional wind (vectors; m s⁻¹) and divergence (shading; 10⁻⁶ m s⁻²) anomalies zonally averaged between 110° and 120°E from 30 June to 16 July 2019. (a)–(h) represent Phases 1–8, respectively. The area between the green dashed lines indicates the latitude scope of southern China.

The above analyses illustrate that convective activity and low-level wind fields over the South China Sea played an important role in precipitation over southern China in July 2019. It is notable that convection activity in the tropics exhibits significant quasi-biweekly oscillation that closely links to the quasi-biweekly oscillation of summer monsoon precipitation in eastern China (Zhan et al., 2008). Figure 8a depicts the meridional propagation of 850-hPa zonal wind anomalies averaged between 110° and 120°E filtered by 10–20-day period. In late June and early July, the low-level wind field over the South China Sea (10°N) is dominated by a cyclone anomaly (Fig. 6a). The zonal wind shows a dipole with westerly wind anomalies in the west of the Philippine Sea and easterly wind anomalies in southern China (25°N; Fig. 8a). This feature has undergone a phase shift in the first dekad of July. The westerly wind anomalies in southern China enhance water vapor transport, which is conducive to more precipitation. In addition, the zonal wind presents apparent southward propagation before July, while northward propagation is dominant after early July (Fig. 8a). The zonal propagation of 850-hPa zonal wind averaged between the equator and 10°N filtered by 10–20-day period is shown in Fig. 8b. The tropical zonal wind exhibits distinct westward propagation, which strengthens remarkably from late June to mid-July. Therefore, from July, the 10–20-day ISO of zonal wind and its accompanying convection anomalies in southern China originate from the quasi-biweekly oscillation in the tropical western Pacific propagating westward and northward.

Figure 8.  Latitude–time cross-section of 10–20-day ISO of 850-hPa zonal wind (m s⁻¹) anomalies averaged (a) between 110° and 120°E, and (b) as in (a), but for longitude–time cross-section between the equator and 10°N.

It has been shown that the 30–60-day oscillation of precipitation in southern China is in its wet phase in the first half of July (Fig. 5). Therefore, the circulation characteristics of the 30–60-day oscillation and its influence on precipitation are further investigated. Figure 9 shows its evolution, including an entire cycle of 850-hPa wind field and OLR anomalies, filtered by 30–60-day period from 21 June to 28 July 2019. From Phases 1 to 2, an anomalous cyclone with active convection is located in the South China Sea (Figs. 9a, b). From Phases 3 to 5, the anomalous cyclone shifts northward and dominates in southern China with active convection, when the WNPSH is southward (Figs. 9c–e). This feature is similar to the result of Zhang et al. (2012) regarding the low-frequency oscillation of wind and precipitation of persistent rainstorms in southern China in 2010. This suggests that the 30–60-day period oscillation provides favorable background circulation for persistent precipitation in southern China during the first half of July. However, the anomalous cyclone over southern China disappears in Phase 6 (Fig. 9f). At the same time, the anomalous anticyclone develops over the South China Sea and shifts northward to southern China, associated with the WNPSH shifting northward and extending westward in Phases 7 and 8 (Figs. 9g–h), when convection in southern China is suppressed during late July 2019.

Figure 9.  As in Fig. 6, but for the 30–60-day period from 21 June to 28 July 2019.

The features of the 30–60-day filtered vertical motion and divergence anomalies are shown in Fig. 10. In Phase 1, an apparent descending branch presents over southern China with anomalous convergence and divergence in the upper and lower troposphere, respectively, while an ascending branch south of 20°N is located over the South China Sea. A distinct vertical–meridional circulation cell occurs over southern China and its north side (Fig. 10a). In Phase 2, the vertical–meridional circulation cell over southern China and its north side weakens (Fig. 10b). From Phases 3 to 5, the descending branch near 10°N strengthens and shifts northward, and vertical motion turns into ascending anomalies and maintains over southern China where anomalous divergence and convergence occur in the upper and lower troposphere, respectively. Meanwhile, a reverse vertical–meridional circulation cell of Phase 1 presents over southern China and its north side (Figs. 10c–e). It should be noted that ascending motion and convergence anomalies in the lower troposphere over southern China are significantly stronger than those in the 10–20-day wet phase. In Phase 6, ascending anomalies weaken (Fig. 10f) and then turn into descending anomalies over southern China during Phases 7 to 8 (Figs. 10g–h).

Figure 10.  As in Fig. 7, but for the 30–60-day period from 21 June to 28 July 2019.

The 10–20-day oscillation of atmospheric circulation induces the appearance and maintenance of anomalous low-level anticyclone in the South China Sea with an intensified and southwestward-extended WNPSH. This enhances water vapor transport from the northwest periphery of the WNPSH and favors more precipitation in southern China (Figs. 6d–f). The anomalous anticyclone associated with suppressed convection in the South China Sea induces an anomalous local vertical–meridional circulation cell to intensify the anomalous ascending motion over southern China (Figs. 7c–f). During the first half of July, a significant anomalous low-level cyclone, induced by the 30–60-day oscillation, is maintained over southern China (Figs. 9c–e). This is conducive to the southward position of the WNPSH and enhances the convergence of water vapor in southern China with an ascending branch center. These lead to anomalous water vapor convergence and dynamic ascending in southern China and favor the occurrence of persistent heavy precipitation.

Summer precipitation over southern China exhibited an apparent intraseasonal variation in 2019 with abundant precipitation in early summer and deficient precipitation in late summer. The standardized precipitation index over southern China in July reached its maximum since 1981. The most significantly persistent heavy precipitation occurred mainly in the first half of July. The ISO of summer precipitation over southern China presents apparent 10–20- and 30–60-day characteristics, with the amplitude of the 10–20-day ISO larger than that of the 30–60-day event in summer 2019.

The occurrence of persistent heavy precipitation in southern China during July 2019 is ascribed mainly to the synergistic effect of both 10–20- and 30–60-day ISO in the first half of July. The precipitation in the first half of July corresponds to an active wet phase of the 30–60-day ISO and a simultaneous entire cycle of the 10–20-day oscillation. During the wet phase of the 10–20-day oscillation, an anomalous low-level anticyclone is observed over the South China Sea, which contributes to a strengthened and southwestward-extending WNPSH accompanied by enhanced water vapor transport in southern China. During the wet phase of the 30–60-day ISO, an anomalous low-level cyclone maintains over southern China and intensifies water vapor convergence in the lower troposphere with strong ascending motion anomalies centered in southern China. The low-level anticyclone during the wet phase of the 10–20-day oscillation, associated with suppressed convection in the South China Sea, induces an anomalous local vertical–meridional circulation cell to strengthen the anomalous ascending motion over southern China. The overlapping peaks of the two oscillation periods in the wet phase provide favorable atmospheric conditions for the occurrence of persistent heavy precipitation in southern China.

The features of this distinct ISO period in boreal summer 2019 and its possible influence on precipitation over southern China are revealed in this study. The westward and northward propagation of the 10–20-day ISO plays an important part in persistent heavy precipitation over southern China in the first half of July. However, the reasons for the remarkably intensified westward propagation of the 10–20-day ISO during the period from late June to mid-July, and an apparent phase shift of the northward propagation in the first dekad of July, merit further study. In addition, the interannual variations of ISO are linked to the evolution of sea surface temperature in the tropical Pacific (Li and Zhou, 1994; Qi et al., 2008; Wu and Song, 2018). It has been demonstrated that the Indian Ocean dipole (IOD) can significantly affect the intensity and propagation of ISO in South Asia (Ajayamohan et al., 2008, 2009). A significant positive phase of the IOD is observed during summer 2019, and its possible impact on the propagation of ISO needs further investigation.

Acknowledgments. The authors thank the Innovation Team of Climate Prediction Theory and Application of the China Meteorological Administration, the Innovation Team of Subseasonal to Seasonal Climate Prediction of the Sichuan Meteorological Bureau, and the Climate Science for Service Partnership (CSSP) for their support. We also appreciate the two anonymous reviewers whose comments and suggestions greatly helped to improve the paper.