Abstract

This paper describes the study of the relationship between the thermal configuration of the Bay of Bengal (BOB)–Tibetan Plateau (TP) region and the precipitation anomaly in Yunnan, a province in China, in May using ERA-Interim data and precipitation data for May from 125 meteorological stations across Yunnan for 1979–2014. Results from the analysis indicate that the interannual variability of May precipitation in Yunnan is significantly modulated by the BOB–TP thermal configuration. Model runs with a linear baroclinic model suggest physical consistency. The thermal conditions over the BOB mainly impact the May precipitation anomaly in Yunnan via changes in water vapor transport from the eastern BOB northeastward to southwestern Yunnan. The second factor influencing precipitation anomalies relates to the characteristics and variability of cold air transport from the TP to northeastern Yunnan. When the BOB (the TP) is occupied by positive (negative) diabatic heating, a thermal gradient with a warmer (colder) center over the BOB and a colder (warmer) center over the TP is established, and more-than-normal (less than normal) precipitation in Yunnan will occur in May. This relationship can persist from April to the following May to some extent; therefore, the BOB–TP thermal configuration in April could be used to forecast May precipitation in Yunnan.

1. Introduction

Yunnan province, located in the southwestern China (21°8′32″–29°15′8″N, 97°31′39″–106°11′47″E) is bounded by the Tibetan Plateau (TP) to the northwest, the Bay of Bengal (BOB) to the southwest, and the South China Sea (SCS) to the southeast. The province is characterized by a mountainous and complex terrain with a pronounced northwest-to-southeast elevation gradient, creating seven distinct climate zones that range from northern tropical to southern temperate (Figs. 1a,b). The region experiences a typical monsoon climate with a rainy season in May–October followed by a dry season in November–April. The average May precipitation of Yunnan achieves 104 mm, which is almost twice as much as the average of the dry season but is nearly half as much as that averaged in June–September. May exhibits the largest variability in precipitation and is the key period for the transition from the dry season to the rainy season (e.g., Wang 1983a,b; Wang and LinHo 2002; Qin et al. 1997; Fig. 1c). Figure 1d also shows that there is a stronger relationship between the precipitation in May and the onset date of its rainy season (China Meteorological Administration 2013). Therefore, the precipitation in May directly influences industrial and agricultural production in Yunnan. For example, late rainy seasons related to less-than-normal precipitation from 2009 to 2012 in Yunnan led to economic losses exceeding CNY 50 billion (e.g., Barriopedro et al. 2012; Lü et al. 2012; Tao et al. 2013; Cao et al. 2014a,b).

Fig. 1.

The topography with (a) April and (b) May climatology horizontal winds at 850 hPa. (c) Long-term monthly means of precipitation averaged in Yunnan and its standard deviation. (d) The correlation coefficients between May precipitation and the onset date of the rainy season averaged in Yunnan. In (a) and (b) the altitude (m) is denoted (shaded area). In (c), the long-term monthly means of precipitation averaged in Yunnan (mmm; blue bars) and the corresponding standard deviation (mm; red lines). In (d), the critical value at the 95% confidence level (black line) and the critical value at the 99% confidence level (red line).

Fig. 1.

The topography with (a) April and (b) May climatology horizontal winds at 850 hPa. (c) Long-term monthly means of precipitation averaged in Yunnan and its standard deviation. (d) The correlation coefficients between May precipitation and the onset date of the rainy season averaged in Yunnan. In (a) and (b) the altitude (m) is denoted (shaded area). In (c), the long-term monthly means of precipitation averaged in Yunnan (mmm; blue bars) and the corresponding standard deviation (mm; red lines). In (d), the critical value at the 95% confidence level (black line) and the critical value at the 99% confidence level (red line).

Both the flow of water vapor and topography are important mechanisms regulating the characteristics of the wet season. Upon its establishment, the water vapor from the BOB and the SCS is carried by the southwest and southeast monsoons that merge in Yunnan and is transported to its downstream area, modulating the floods and droughts over East Asia (Xu et al. 2004; Cao et al. 2012; Day et al. 2015). Furthermore, Shi et al. (2017) found that the Yunnan–Guizhou Plateau weakens the Indian monsoon by enhancing the anomalous anticyclonic winds over the BOB–Indian subcontinent–Arabian Sea. Thus, studies on the May precipitation in Yunnan have practical value in terms of local socioeconomic development and scientific value in terms of understanding the flood–drought mechanism over Southeast Asia.

The Tibetan Plateau is an important source of cold-air mass that can affect the study region. Yan et al. (1995) reports that mechanical effects and thermal conditions from the TP can also impact May precipitation in Yunnan by modulating the Asian summer monsoon to some extent. The processes involved are similar to other studies that discuss the impact of the TP on the Asian monsoon (e.g., Broccoli and Manabe 1992; Yanai et al. 1992; Yanai and Li 1994; Boos and Kuang 2010; Wu et al. 2012). Movement of cold air from the TP into Yunnan is complex: southwesterly winds prevail in Yunnan during the rainy season and the East Asian winter monsoon is usually blocked by the Yunnan–Guizhou Plateau during the dry season. However, Qin et al. (1997) describe three pathways by which cold air can invade the Yunnan region. First, it may approach from the northeast after climbing over the Qinling Mountains. Alternately, it may invade south China and recirculate into Yunnan as a southeast flow. A third path is as a direct flow of cold air from the northwest originating in the TP (Xu et al. 2011).

In contrast with other studies that examine summer precipitation variability in East Asia [e.g., Chang (2004), and references cited therein; Wang (2006), and references cited therein], there has been relatively little work done on how May precipitation influences the onset of the rainy season in Yunnan. Some of the earlier studies link May precipitation to interactions between the southwest and southeast monsoons (Wang 1983a,b). Recent studies examine the sources and the transport of water vapor as important mechanisms for precipitation in the study region. They involve, for example, linking the onset and intensity of May precipitation to the onset times of the monsoon BOB (Yan et al. 2003; Jiang and Li 2011; Xing et al. 2016). Other relevant studies examine the intensity of water vapor flow over the Indian Ocean and BOB on May precipitation (Chen et al. 2006); the influence of La Niña/El Niño episodes in helping/hindering the westerly water vapor flows in turn affects May precipitation (Yang et al. 2011; Tang et al. 2013) and ENSO/European teleconnections (Deng et al. 2016).

Despite the progress in precipitation mechanisms as listed in the above paragraph, there are conflicting results linking the strength of the East Asian monsoon to precipitation in eastern China. Chen et al. (2000, 2013) found that a weak (strong) East Asian winter monsoon related to the El Niño–Southern Oscillation tends to favor flood (drought) in the Yangtze River valley during summer, and vice versa. However, Yan et al. (2007) suggest that a strong (weak) East Asian winter monsoon may increase (decrease) the summer precipitation in Yunnan. These studies may differ in their results, as they concentrate on a single sea state and water vapor condition. The key physical processes relating to the precipitation anomaly in May in Yunnan have remained unclear to date, as the precipitation is induced by a complicated interaction between the warm-wet and cold-dry air masses originating from the BOB and TP as important sources.

It is generally accepted that the monsoon is responsible for the seasonal changes in the land–sea thermal contrast (e.g., Flohn 1957; Krishnamurti 1985; Li and Yanai 1996; Meehl 1994; Webster et al. 1998; He et al. 2003; Krishnamurti et al. 2013). The TP is located to the northwest of Yunnan and the BOB to the southwest. Tamura et al. (2010) found that the latent heat released by the BOB can also influence the thermal conditions around the TP. This land–sea configuration implies that the thermal conditions are crucial for inducing the precipitation anomaly in May in Yunnan. The motivation of this study is to validate the abovementioned hypothesis and to reveal further the corresponding physical processes.

The remainder of the paper is arranged as follows. The data, the method, and the linear baroclinic model (LBM; Watanabe and Kimoto 2000) are described in section 2. Section 3 investigates the relationship between the TP–BOB thermal configuration and the May precipitation anomaly in Yunnan. Section 4 reveals the possible physical processes through which the TP–BOB thermal configuration influences the interannual variability of the May precipitation anomaly in Yunnan. The model results are shown in section 5, and are used to confirm the relationship and the associated physical processes revealed in sections 3 and 4. A summary and discussion are presented in section 6.

2. Data, method, and model

a. Data

We use daily ERA-Interim data from the European Centre for Medium-Range Weather Forecasts (ECMWF) (Simmons et al. 2004; Dee et al. 2011) for the period 1979–2014. The resolution of the ERA-Interim data is 2° in latitude and longitude, and there are 37 pressure levels from 1000 to 1 hPa. May precipitation data and the onset date of the rainy season are from the Climate Center of Yunnan Province, taken from 125 stations across Yunnan.

b. Method

The apparent heat source, , and the apparent moisture sink, , are estimated using the daily ERA-Interim data and the thermodynamic equations developed by Yanai et al. (1992), Yanai and Li (1994), and Li and Yanai (1996):

 
formula
 
formula

In Eqs. (1) and (2), is horizontal velocity, is potential temperature, is vertical velocity at the isobaric surface, is the mixing ratio of water vapor, is the isobaric gradient operator, is time, is pressure, hPa, and is the specific heat at a constant pressure of dry air. In Eq. (1), , where is the gas constant of dry air. In Eq. (2), is the latent heat of condensation. The column-integrated apparent heat source and apparent moisture sink are calculated using Eqs. (3) and (4):

 
formula
 
formula

In Eqs. (3) and (4), hPa and is the surface pressure.

c. LBM

Watanabe and Kimoto (2000) revealed the eddy vorticity feedback that tends to force the positive phase of the North Atlantic Oscillation uses the dry version of the LBM, which is driven by the associated diabatic heating. Watanabe and Jin (2003) adapt the dry version into the moisture LBM and further study the coupled dynamical convective response to El Niño. Lu and Lin (2009) suggested that the dry version of the LBM is more suitable than the moist version in analyzing the responses of regional diabatic heating. The model used in this study depends on primitive equations linearized about the May climatology calculated from the ERA-Interim for 1979–2014. Similar to previous studies, the diabatic heating patterns associated with the anomalous BOB–TP thermal configuration will illustrate its importance in driving the large-scale circulation affecting May precipitation in Yunnan. A horizontal resolution of T42 is adopted with 20 sigma levels in the vertical direction of this study. In the dry version of the LBM, the time scales of Rayleigh’s friction and Newtonian damping are 0.5 day−1 for , 1 day−1 for , and 30 day−1 for between 0.9 and 0.03. The circulation response reaching the steady state is approximately after day 15. A specified description of the dry version of the LBM can be found in Watanabe and Kimoto (2000, appendix B).

3. Observational analysis

a. Correlation between the BOB–TP thermal configuration and May precipitation

The correlation coefficients between the thermal configuration series and May precipitation are calculated, and the numbers of stations passing the significance test at the 95% and 99.9% confidence levels are further counted (Fig. 2a). The results are examined using a Monte Carlo test by generating 10 000 random shuffles of the May diabatic heating anomaly time series and correlating each with the observed time series of rainfall anomalies. There are two areas passing the Monte Carlo test at the 95% confidence level. One is located at the BOB and another at the TP. In addition, the maximum number of stations that pass the significance level with a center value exceeds 110 are located in Yunnan, suggesting that the rainfall measured at Yunnan Plateau meteorological stations is most highly correlated with diabatic heating on the Yunnan Plateau itself. In fact, these results fit the corresponding theory, because , where is the rate of condensation per unit mass of air and is the latent heat of condensation (Yanai and Li 1994).

Fig. 2.

(a) The number of stations passing the significance test above the 95% confidence level and (b) their variation over the period 1979–2014. The 95% confidence level (light shaded area; the critical value is 28) and the 99.9% confidence level (dark shaded area; the critical value is 91). In (b), May precipitation averaged in Yunnan (bars), the BOB– TP TCI (solid line), the thermal index over the BOB (dashed line), and the thermal index over the TP (dashed–dotted line).

Fig. 2.

(a) The number of stations passing the significance test above the 95% confidence level and (b) their variation over the period 1979–2014. The 95% confidence level (light shaded area; the critical value is 28) and the 99.9% confidence level (dark shaded area; the critical value is 91). In (b), May precipitation averaged in Yunnan (bars), the BOB– TP TCI (solid line), the thermal index over the BOB (dashed line), and the thermal index over the TP (dashed–dotted line).

As the results do not essentially change with rectangle width, here we choose two rectangles (8°–16°N, 84°–94°E and 30°–38°N, 84°–94°E) as the research domains. It is worth noting that the rectangle 8°–16°N, 84°–94°E mainly covers the BOB, and the rectangle 30°–38°N, 84°–94°E appears on the TP. To further confirm the relationship between the thermal configuration and May precipitation, we calculate the correlation coefficients between the precipitation averaged in Yunnan, the column-integrated apparent heat source averaged over the rectangle 8°–16°N, 84°–94°E; the column-integrated apparent heat source averaged over the rectangle 30°–38°N, 84°–94°E; and the time series of column-integrated apparent heat source averaged over rectangle 8°–16°N, 84°–94°E minus that averaged over rectangle 30°–38°N, 84°–94°E in May. The corresponding three time series are denoted as the BOB thermal index, the TP thermal index, and the BOB–TP thermal configuration index (TCI), respectively. The correlation coefficient associated with the column-integrated apparent heat source averaged over the rectangle 8°–16°N, 84°–94°E is 0.54, passing the significance test at the 99% confidence level, and that associated with the column-integrated apparent heat source averaged over the rectangle 30°–38°N, 84°–94°E is −0.50, also passing the significance test at the 99% confidence level. Note that the correlation coefficient associated with the BOB–TP TCI (0.70), which passes the significance test at the 99.9% confidence level, is higher than the previous two correlation coefficients (Fig. 2b). The distributions of the correlation coefficients all share the same feature, that is, the number of stations passing the significance test at the 95% confidence level, and the correlation intensity associated with the BOB–TP TCI is obviously higher than those associated with the individual BOB or TP thermal conditions (Fig. 3). This suggests that there is a close relationship between May precipitation and the current BOB–TP thermal configuration in the south–north direction. When the BOB thermal conditions are warmer (colder) than normal, the the TP thermal conditions are colder (warmer) than normal in May, that is, the BOB–TP TCI is higher (lower) than normal in May and the May precipitation in Yunnan is more (less) than normal.

Fig. 3.

The distribution of correlation coefficients between May precipitation in Yunnan and (a) the thermal index over the BOB, (b) the thermal index over the TP, and (c) the BOB–TP TCI. The 95% confidence level (light shaded areas) and the 99% confidence level (dark shaded areas).

Fig. 3.

The distribution of correlation coefficients between May precipitation in Yunnan and (a) the thermal index over the BOB, (b) the thermal index over the TP, and (c) the BOB–TP TCI. The 95% confidence level (light shaded areas) and the 99% confidence level (dark shaded areas).

We further calculate the correlation coefficients between May precipitation in Yunnan and the thermal index over the BOB, the thermal index over the TP, and the BOB–TP thermal configuration index averaged around April. Figure 4 shows that there are a few stations passing the significance test at the 90% confidence level (Figs. 4a,b). However, most stations over the midwestern regions of Yunnan pass the significance test above the 95% confidence level (Fig. 4c). Comparing Fig. 4c with Fig. 3c, it can be seen that the area passing the significance test tends to expand eastward in May. These results suggest that the May precipitation is closely related to the BOB–TP thermal configuration rather than the thermal condition of individual BOB or TP.

Fig. 4.

The distribution of correlation coefficients between May precipitation in Yunnan and (a) the thermal index over the BOB, (b) the thermal index over the TP, and (c) the BOB–TP thermal configuration index averaged around April. The 95% confidence level (light shaded areas) and the 99% confidence level (dark shaded areas).

Fig. 4.

The distribution of correlation coefficients between May precipitation in Yunnan and (a) the thermal index over the BOB, (b) the thermal index over the TP, and (c) the BOB–TP thermal configuration index averaged around April. The 95% confidence level (light shaded areas) and the 99% confidence level (dark shaded areas).

b. Anomalous BOB–TP thermal configuration

To reveal the physical process through which the May precipitation anomalies in Yunnan are related to the variability of the BOB–TP thermal configuration, we used the TCI time series and a criterion of ±0.7 standard deviation (Fig. 2b) to identify 9 (10) out of the 36 years as positive (negative) BOB–TP TCI years. The nine positive BOB–TP TCI years are 1981, 1988, 1990, 1999, 2000, 2001, 2002, 2004, and 2006. The 10 negative BOB–TP TCI years are 1979, 1986, 1987, 1991, 1996, 1997, 2003, 2005, 2008, and 2012.

Figure 5 displays apparent anomalous heating in positive and negative composite years (Figs. 5a,b). Apparent moisture sinks are likewise shown for each composite (Figs. 5c,d). Anomalous heating is further decomposed into the anomalous surface sensible heat flux (Figs. 5e,f) and the anomalous surface latent heat flux (Figs. 5g,h). The results obtained above agree well with each other, suggesting that anomalous diabatic heating corresponds to increased evaporation and latent heating in the BOB and Arabian Sea, whereas on the Tibetan Plateau heating anomalies correspond to changes in sensible heating (Figs. 5e,f) and radiation heating (Figs. 6e,f), since the plateau is very arid during May (Taniguchi and Koike 2008).

Fig. 5.

Apparent heat source anomalies in May in (a) positive and (b) negative BOB–TP TCI years, the apparent moisture sink anomalies in May in (c) positive and (d) negative BOB–TP TCI years, the anomalies of surface sensible heat flux in May in (e) positive and (f) negative BOB–TP TCI years, and the anomalies of surface latent heat flux in May in (g) positive and (h) negative BOB–TP TCI years (W m−2). The areas passing the significance test at the 90%, 95%, and 99% confidence levels (shaded areas from light to dark).

Fig. 5.

Apparent heat source anomalies in May in (a) positive and (b) negative BOB–TP TCI years, the apparent moisture sink anomalies in May in (c) positive and (d) negative BOB–TP TCI years, the anomalies of surface sensible heat flux in May in (e) positive and (f) negative BOB–TP TCI years, and the anomalies of surface latent heat flux in May in (g) positive and (h) negative BOB–TP TCI years (W m−2). The areas passing the significance test at the 90%, 95%, and 99% confidence levels (shaded areas from light to dark).

Fig. 6.

Horizontal wind anomalies at 700 hPa in May in (a) positive BOB–TP TCI years and (b) negative BOB–TP TCI years, and those at 500 hPa in May in (c) positive BOB–TP TCI years and (d) negative BOB–TP TCI years (m s−1). The geopotential height anomalies at 500 hPa in the (e) positive BOB–TP TCI years and (f) negative BOB–TP TCI years (geopotential meter). The areas passing the significance test at the 90%, 95%, and 99% confidence levels (shaded areas from light to dark).

Fig. 6.

Horizontal wind anomalies at 700 hPa in May in (a) positive BOB–TP TCI years and (b) negative BOB–TP TCI years, and those at 500 hPa in May in (c) positive BOB–TP TCI years and (d) negative BOB–TP TCI years (m s−1). The geopotential height anomalies at 500 hPa in the (e) positive BOB–TP TCI years and (f) negative BOB–TP TCI years (geopotential meter). The areas passing the significance test at the 90%, 95%, and 99% confidence levels (shaded areas from light to dark).

c. Anomalous circulation patterns

The atmospheric circulation corresponds to the anomalous BOB–TP thermal configuration and presents some pronounced anomalies. Figure 6 shows the anomalous horizontal winds at 700 and 500 hPa and the anomalous geopotential height at 500 hPa in May. It can be seen from Fig. 6a that a cyclonic anomaly dominates the northern BOB and an anticyclonic anomaly controls the TP in the positive BOB–TP TCI years. The southwesterly anomalies located east of the cyclonic anomaly run from the BOB to the northern Indochina Peninsula and then turn westward into Yunnan. Meanwhile, the northeasterly anomalies that occur east of the anticyclonic anomaly run into Yunnan. The anomalous pattern of horizontal winds at 500 hPa (Fig. 6c) resembles the anomalous pattern at 700 hPa (Fig. 6a). The 500-hPa anomalous geopotential heights are higher than normal from 20°N, 60°E and 40°N, 60°E northeastward to 40°N, 120°E and 50°N, 120°E, but they are lower than normal from 10°N, 60°E and 20°N, 60°E northeastward to 20°N, 120°E and 40°N, 120°E (Fig. 6e). The anomalous pattern of geopotential height is very consistent with the anomalous patterns of horizontal winds at 700 and 500 hPa.

These results suggest that the warm-wet air mass from the BOB converges with the cold air mass from the east edge of the TP in positive BOB–TP TCI years and that it will result in a heavier-than-normal May precipitation in Yunnan. The anomalous horizontal winds at 700 and 500 hPa and the anomalous geopotential height at 500 hPa in negative BOB–TP TCI years are similar to those in positive BOB–TP TCI years with opposite signs to a larger degree, except that the areas of horizontal winds and geopotential height passing the significance test shrink over the Indian Ocean at 700 and 500 hPa to some extent. Obviously, the warm-wet airflow from the BOB and the cold airflow from the east edge of the TP are weaker—that is, Yunnan is dominated by an anomalous divergent airflow in negative BOB–TP TCI years—and will result in less-than-normal May precipitation in Yunnan.

To confirm this interpretation associated with horizontal winds and geopotential height, we further calculated the column-integrated water flux (integrated from the surface to 300 hPa) and its divergence. In positive BOB–TP TCI years, the water vapor flux is characterized by a cyclonic anomaly over the BOB. Significant northeastward water vapor flux anomalies appear over the whole BOB and the Indochina Peninsula (Fig. 7a), and turn northwest around southwestern Yunnan, which is similar to the anomalous wind pattern (Figs. 6a,c). A second important feature of positive BOB-TC TCI years is an inflow in the water vapor flux into Yunnan, both from the northeast and southwest (Fig. 7a), which is also linked to a pronounced anomalous water vapor convergence region extending from the BOB to the Indochina Peninsula (Fig. 7c). In contrast, during negative BOB–TP TCI years, an anticyclonic anomaly occurs over the BOB. Clear northeasterly anomalies associated with this anticyclone are found in the whole BOB and the Indochina Peninsula (Fig. 7b). The anomalous southwestward transport of the water vapor flux over southwestern Yunnan and the northeasterly anomalies over northeastern Yunnan are almost opposite of those in positive BOB–TP TCI years, producing an anomalous water vapor divergence in the region (Figs. 7b,d).

Fig. 7.

Composite anomalies of May column-integrated water vapor flux in (a) positive BOB–TP TCI years and (b) negative BOB–TP TCI years (kg m−1 s−1), and the water vapor flux divergence in (c) positive BOB–TP TCI years and (d) negative BOB–TP TCI years (contour interval = 10−4 kg m−2 s−1) in positive BOB–TP TCI years. The areas passing the significance test at the 90%, 95%, and 99% confidence levels (shaded areas from light to dark), respectively.

Fig. 7.

Composite anomalies of May column-integrated water vapor flux in (a) positive BOB–TP TCI years and (b) negative BOB–TP TCI years (kg m−1 s−1), and the water vapor flux divergence in (c) positive BOB–TP TCI years and (d) negative BOB–TP TCI years (contour interval = 10−4 kg m−2 s−1) in positive BOB–TP TCI years. The areas passing the significance test at the 90%, 95%, and 99% confidence levels (shaded areas from light to dark), respectively.

d. May precipitation anomalies in Yunnan

The covariance between the BOB–TP thermal configuration anomalies and the anomalous circulation patterns induces the May precipitation anomaly in Yunnan. In positive BOB–TP TCI years, this covariance causes the May precipitation, especially over western and southern Yunnan, to increase significantly through enhancement of the southwest monsoon from the BOB, the cold airflow at the east edge of the TP, and the water vapor convergence around Yunnan. The precipitation anomalies over most of Yunnan pass the significance test at the 99% confidence level (Fig. 8a). In negative BOB–TP TCI years, this covariance causes the May precipitation to decrease obviously in Yunnan through weakening of the southwest monsoon, enhancement of the southeasterly anomaly around northeastern Yunnan, and the water vapor divergence around Yunnan. However, the area passing the significance test is smaller compared with equivalent areas during positive BOB–TP TCI years (Fig. 8b).

Fig. 8.

May precipitation anomalies in Yunnan in (a) positive BOB–TP TCI years and (b) negative BOB–TP TCI years (mm). The 95% confidence level (light shaded areas) and the 99% confidence level (dark shaded areas).

Fig. 8.

May precipitation anomalies in Yunnan in (a) positive BOB–TP TCI years and (b) negative BOB–TP TCI years (mm). The 95% confidence level (light shaded areas) and the 99% confidence level (dark shaded areas).

e. Persistence of the correlation between BOB–TP thermal configuration and May precipitation

To verify whether the BOB–TP thermal configuration could be used as a precursor for the May precipitation anomaly in Yunnan, we illustrate the temporal evolution of the correlation coefficient between May precipitation averaged in Yunnan and the 30-day running mean of the column-integrated apparent heat source averaged in 84°–94°E (Fig. 9). The significantly negative correlation coefficients between the apparent heat source and May precipitation averaged in Yunnan occur around 35°N (the TP) and the significantly positive correlation coefficients occur around 20°N (the BOB) during May. The apparent heat source at 6°N from the end of winter to the end of April presents a significantly positive correlation with May precipitation averaged in Yunnan. The significantly positive correlation between the BOB–TP 30-day running-average TCI and May precipitation averaged in Yunnan can persist over the whole of May, once the positive correlation established in early April (Fig. 9b). These results are consistent with Fig. 4, suggesting that the high positive correlation between BOB–TP TCI and May precipitation shown in Fig. 9b are reliable. It is important to note that despite the significant correlation between May precipitation and individual thermal components of TP and BOB, these do not establish sufficient and necessary conditions for a high correlation between May precipitation and BOB–TP TCI. The evolution of the correlation coefficients between the 30-day running-average BOB–TP TCI and May BOB–TP TCI and those between the 30-day running-average BOB–TP TCI and May precipitation averaged in Yunnan show a shorter persistence time and a similar pattern to the dashed curve in Fig. 9b (30-day running-average BOB–TP TCI vs May BOB–TCI). Finally, there are significantly positive correlation coefficients between the May precipitation averaged in Yunnan and the apparent moisture sink in the BOB–TP averaged around May (Fig. 9c). The significantly positive correlation appears later than that associated with an apparent heat source to some extent, and the correlation intensity is also weaker (Fig. 9d).

Fig. 9.

(a) Correlation coefficients between the 30-day running-average apparent heat source averaged in 84°–94°E and May precipitation averaged in Yunnan. (b) Correlation coefficients between the 30-day running-average BOB–TP TCI, May BOB–TP TCI, and May precipitation averaged in Yunnan. (c) Correlation coefficients between the 30-day running-average apparent moisture sink and May precipitation averaged in Yunnan. (d) Correlation coefficients between May precipitation averaged in Yunnan, the 30-day running-average BOB–TP TCI, and May BOB–TP TCI associated with the apparent moisture sink. Correlation coefficients passing the significance test at the 95% and 99% confidence levels (areas shaded from light to dark). In (a) and (c), the negative correlation coefficient (blue lines), the positive correlation coefficient (red lines), and the meridional range (green lines) of two key regions.

Fig. 9.

(a) Correlation coefficients between the 30-day running-average apparent heat source averaged in 84°–94°E and May precipitation averaged in Yunnan. (b) Correlation coefficients between the 30-day running-average BOB–TP TCI, May BOB–TP TCI, and May precipitation averaged in Yunnan. (c) Correlation coefficients between the 30-day running-average apparent moisture sink and May precipitation averaged in Yunnan. (d) Correlation coefficients between May precipitation averaged in Yunnan, the 30-day running-average BOB–TP TCI, and May BOB–TP TCI associated with the apparent moisture sink. Correlation coefficients passing the significance test at the 95% and 99% confidence levels (areas shaded from light to dark). In (a) and (c), the negative correlation coefficient (blue lines), the positive correlation coefficient (red lines), and the meridional range (green lines) of two key regions.

The persistence obtained above suggests that a positive feedback mechanism associated with the release of latent heat plays a critical role over the BOB, and that in the TP these mechanisms relate to sensible and radiation heating. The anomalous latent heating over the BOB will force relatively weak anomalous easterly winds over the SCS (east of the heating center) and significant westerlies from the southern Arabian Sea to the BOB (west of the heating center) at mid- to low tropospheric levels (Figs. 6a,c). These anomalous winds are superimposed onto the mean easterlies from the BOB to the SCS (Fig. 1a). The winds will accelerate to the east of the heating center over the BOB and decelerate to the west, which further increases the evaporation over the SCS and decreases the evaporation over the western BOB. The evaporation anomaly will in turn influence the latent heating anomaly, which favors more-than-normal May precipitation in Yunnan. The phenomenon that the area passing the significance test tends to expand eastward can be also explained using the east-propagating modes induced by the feedback (Figs. 3c, 4c). In the meantime, the anomalous anticyclonic circulation forces subsidence anomalies and transports cold air into the TP through the northerly anomalies over the northern portion of the anomalous anticyclone. The stronger-than-normal radiation cooling and the lower-than-normal level of sensible heat in turn enhance the anomalous anticyclone through forcing subsidence and northerly anomalies over the TP. When a positive latent heat anomaly occurs in the BOB, and a negative sensible heat anomaly and a negative radiation heat anomaly occur over the TP, a thermal gradient is established with a warmer center over the BOB and a colder center over the TP, and it remains from early April to the end of May through the two positive feedbacks. The southwesterly anomalies caused by the persistently positive thermal configuration of the BOB–TP extend from the BOB and the western Indochina Peninsula northeastward to southwestern Yunnan, and the northeasterly anomalies induced by the same thermal configuration prevail from the west edge of the TP southwestward to the northeastern Yunnan (Figs. 6a,e). The persistent convergence zone continuously occupies Yunnan and leads to more-than-normal May precipitation in Yunnan (Fig. 8a). By contrast, less-than-normal precipitation occurs in Yunnan when a thermal gradient with a colder center over the BOB and a warmer center over the TP is established and persists from April to the end of May (Fig. 8b).

4. Model results

The complicated topography in Yunnan significantly influences the precipitation environment of the region (Qin et al. 1997; Xu et al. 2011). However, the interannual variability of May precipitation in Yunnan may not be significantly impacted by the complicated topography in Yunnan (Cao et al. 2014a,b), as there is spatial coherence in the correlation coefficients (Fig. 3) and the composite results (Figs. 8a,b). Although the T42 resolution of the model employed most likely does not resolve the topography of the Yunnan Plateau, it is suitable for studying the relationship between the thermal configuration of the BOB–TP region and the May precipitation anomaly in Yunnan to some extent.

The apparent heat source patterns associated with the variability of May precipitation in Yunnan in normal BOB–TP TCI years, positive BOB–TP TCI years, and negative BOB–TP TCI years are adopted as the prescribed forcing for the dry version of the LBM. Figure 10a shows the prescribed vertical profiles of the apparent heat source in positive BOB–TP TCI years (red lines) and normal BOB–TP TCI years (black lines) in the sigma coordinate system. The center intensities of the apparent heat source are governed by the prescribed vertical profiles, and the corresponding spatial distributions exhibited in Figs. 10b,c resemble those shown in Fig. 4a. As the atmospheric circulation response achieves the steady state after day 10, the differences between the positive BOB–TP TCI years and the normal BOB–TP TCI years at day 15 are shown as the steady-state response. Figure 10d shows the simulated horizontal wind differences between the positive BOB–TP TCI years and the normal BOB–TP TCI years after imposing only the apparent heat source in the BOB. As the intensity of the apparent heat source in positive BOB–TP TCI years is higher than in normal BOB–TP TCI years over the BOB, the positive anomaly of the apparent heat source will result in a cyclonic anomaly at 700 hPa. Stronger-than-normal southwesterlies prevail over the southeastern flank of the cyclonic anomaly. The anomalous northerly winds, which may benefit the cold air moving southward toward Yunnan, are also forced over the eastern TP, but their intensities are very weak. The circulation anomalies, responding to the anomalous apparent heat source, resemble the composite anomalies of the May circulation. We further calculate the explained variances for the zonal and meridional winds using the method adopted by Weng et al. (2011). It can be seen from Table 1 that the modeling results can explain more than 15% of the variances for the observational winds over the key region (4°–40°N, 82°–110°E). This implies that the positive anomaly of the apparent heat source over the BOB has a key role in transporting more-than-normal water vapor into Yunnan and that it causes stronger-than-normal May precipitation in Yunnan. Figure 10e shows the atmospheric circulation responses to the apparent heat source imposed on only the TP. Because the intensity of the apparent heat source in positive BOB–TP TCI years is lower than in normal BOB–TP TCI years over the TP, an anticyclonic anomaly occupies the TP with anomalous northerly winds around its eastern flank and anomalous easterly winds around its southern flank at 700 hPa. The anomalous northerly winds will favor the cold air moving southwestward. However, the anomalous southwesterly winds are very weak in the eastern BOB. The explained variance for the zonal wind is relatively high but is only 9% for meridional winds (Table 1). The contribution of the apparent heat source over the TP to the circulation responses associated with the variability of May precipitation in Yunnan mainly focuses on the cold air transport. When the apparent heat sources in positive BOB–TP TCI years are imposed on both the BOB and the TP, a significant cyclonic anomaly dominates the BOB and a significant anticyclonic anomaly controls the TP at 700 hPa (Fig. 10f). In contrast to Figs. 10d,e, the anomalous circulation responding to the joint influences of the apparent heat sources over the BOB and the TP (Fig. 10f) best resembles the observed anomalies shown in Figs. 6a, 7a. In fact, the corresponding explained variances for the zonal and meridional winds are the largest among the three numerical experiments in the positive BOB–TP TCI years. The relatively large values of the total atmospheric circulation variances explained suggest that the positive BOB–TP thermal configuration induces a stronger-than-normal May precipitation in Yunnan through significantly converging the warm-wet airflow from the BOB with the cold airflow from the eastern TP in Yunnan.

Fig. 10.

(a) Vertical profile and (b),(c) horizontal distribution of the prescribed apparent heat source. The units are °C day−1 in (a), and the contour interval is 1.0 K d −1 in (b) and (c). (d)–(f) The 700-hPa horizontal wind anomalies (m s−1) difference between positive anomaly years and normal BOB–TP TCI years in May, forced by the apparent heat source over the BOB, the TP, and both, respectively. (g)–(i) As in (d)–(f), but at 500 hPa. In (a), the vertical profile in positive anomaly (red solid line) and normal (black solid line) years over the BOB, the vertical profile in positive anomaly (red dashed line) and normal (black dashed line) years over the TP, and zero heating (black dashed–dotted line).

Fig. 10.

(a) Vertical profile and (b),(c) horizontal distribution of the prescribed apparent heat source. The units are °C day−1 in (a), and the contour interval is 1.0 K d −1 in (b) and (c). (d)–(f) The 700-hPa horizontal wind anomalies (m s−1) difference between positive anomaly years and normal BOB–TP TCI years in May, forced by the apparent heat source over the BOB, the TP, and both, respectively. (g)–(i) As in (d)–(f), but at 500 hPa. In (a), the vertical profile in positive anomaly (red solid line) and normal (black solid line) years over the BOB, the vertical profile in positive anomaly (red dashed line) and normal (black dashed line) years over the TP, and zero heating (black dashed–dotted line).

Table 1.

The explained variance (%) of the circulation anomalies over 4°–40°N, 82°–110°E. The numbers in brackets are the effective sample size (Bretherton et al. 1999).

The explained variance (%) of the circulation anomalies over 4°–40°N, 82°–110°E. The numbers in brackets are the effective sample size (Bretherton et al. 1999).
The explained variance (%) of the circulation anomalies over 4°–40°N, 82°–110°E. The numbers in brackets are the effective sample size (Bretherton et al. 1999).

Figure 11a is the same as Fig. 10a but with the prescribed vertical profiles of the apparent heat source in negative BOB–TP TCI years (blue lines) and normal BOB–TP TCI years (black lines) in the sigma coordinate system. The center intensities of the apparent heat source are also given by the prescribed vertical profiles, and the corresponding spatial distributions in Figs. 11b,c resemble those shown in Fig. 2a. We calculate the differences between the negative BOB–TP TCI years and the normal BOB–TP TCI years at the integrated-day 15. When imposing only the apparent heat source on the BOB, an anticyclonic anomaly at 700 hPa will result from the negative anomaly of the apparent heat source. Stronger-than-normal northeasterlies around the southeastern flank of the anticyclonic anomaly weaken the water vapor transport northeastward to Yunnan. A very weak easterly anomaly can be observed over the southeastern TP. When imposing only the apparent heat source on the TP, a cyclonic anomaly occurs with southerly anomalies around the eastern TP and westerly anomalies around the southern TP at 700 hPa, which does not favor the cold air moving southwestward. The northeasterly anomalies are very weak in the eastern BOB. When the apparent heat sources are imposed on both the BOB and the TP, we find that a significant anticyclonic anomaly dominates the BOB and a significant cyclonic anomaly controls the TP at 700 hPa (Fig. 11f). Compared with Fig. 11d,e, the anomalous circulation responding to the anomalous apparent heat sources over the BOB–TP (Fig. 11f) also best resembles the observed anomalies shown in Figs. 6b, 7b. In fact, under the condition that the apparent heat sources are imposed on both the BOB and the TP, the explained variances with 44.7% for the zonal winds and 34.9% for the meridional winds are the largest in comparison with those in relation to two other numerical experiments (Table 1).

Fig. 11.

As in Fig. 10, but with (a) vertical profile and (b),(c) horizontal distribution of the prescribed apparent heat source in negative BOB–TP TCI years. The contour interval is 1.0 K day −1 in (b) and (c). (d)–(f) The 700-hPa horizontal wind anomalies (m s−1) difference between negative-anomaly years and normal BOB–TP TCI years in May, forced by the apparent heat source over the BOB, the TP, and both, respectively. (g)–(i) As in (d)–(f), but at 500 hPa. In (a), the vertical profile in negative anomaly (blue solid line) and normal (black solid line) years over the BOB, the vertical profile in negative anomaly (blue dashed line) and normal (black dashed line) years over the TP, and zero heating (black dashed–dotted line).

Fig. 11.

As in Fig. 10, but with (a) vertical profile and (b),(c) horizontal distribution of the prescribed apparent heat source in negative BOB–TP TCI years. The contour interval is 1.0 K day −1 in (b) and (c). (d)–(f) The 700-hPa horizontal wind anomalies (m s−1) difference between negative-anomaly years and normal BOB–TP TCI years in May, forced by the apparent heat source over the BOB, the TP, and both, respectively. (g)–(i) As in (d)–(f), but at 500 hPa. In (a), the vertical profile in negative anomaly (blue solid line) and normal (black solid line) years over the BOB, the vertical profile in negative anomaly (blue dashed line) and normal (black dashed line) years over the TP, and zero heating (black dashed–dotted line).

The 500-hPa horizontal wind responses to the apparent heat source over the BOB, the TP, and both are similar to those at 700 hPa. It is noteworthy that the 500-hPa horizontal winds over the TP become stronger than the 700-hPa horizontal winds over the same region (Figs. 10g–h, 11g–h), which results from the maximal value of the apparent heat source given around 500 hPa over the TP to a large degree. These modeling results suggest that the negative BOB–TP thermal configuration causes less-than-normal May precipitation in Yunnan through significant weakening of the warm-wet airflow from the BOB and the cold airflow from the eastern TP.

5. Conclusions and discussion

A significant relationship was demonstrated between average May precipitation in Yunnan and the thermal configuration of the Bay of Bengal–Tibetan Plateau (BOB–TP) region. For the 1979–2014 period, the correlation between the BOB–TP thermal configuration index (TCI), calculated from ERA-Interim data and the May precipitation in Yunnan, based on 125 stations, was 0.70, which passed the significance test at the 99.9% confidence level. In the BOB–TP thermal configuration, the diabatic heating over the BOB is positively related to May precipitation in Yunnan, but it is negatively correlated to precipitation over the TP.

The physical processes associated with the covariation of the BOB–TP thermal configuration and the May precipitation anomalies in Yunnan were further investigated with composite analysis. The corresponding results illustrate that for a positive BOB–TP TCI in May—that is, diabatic heating anomalies above (below) normal over the BOB (the TP)—an anomalous thermal gradient between the BOB and the TP builds up. This anomalous thermal gradient induces the development of an anomalous cyclone (anticyclone) over the BOB (the TP), accompanied by the appearance of southwesterly (northeasterly) anomalies at the eastern flank of the anomalous cyclone (anticyclone). The southwesterly anomalies carrying more water vapor from the BOB when they converge with the northeasterly anomalies bringing more cold air from the TP. The anomalous convergence zone is located precisely in Yunnan and leads to more-than-normal precipitation in May. For a negative BOB–TP TCI in May, almost opposite conditions occur, that is, diabatic heating and sensible heat anomalies below (above) normal over the BOB (the TP). These conditions cause less-than-normal precipitation in May.

Both the April and May BOB–TP thermal configurations are significantly associated with May precipitation in Yunnan. A positive (negative) April TCI tends to be followed by more-than-normal (less than normal) May precipitation in Yunnan. The BOB–TP thermal configuration in April has a relatively high prediction potential for May precipitation in Yunnan and could therefore be used in forecasting applications.

The modeling results obtained from the LBM substantiate the dominant physical processes mentioned above. Furthermore, these modeling results can explain the persistence of the correlation between the BOB–TP thermal configuration and May precipitation through two positive feedback mechanisms behind the physical processes. One mechanism is associated with the evaporation–wind feedback, which may explain the persistence of the thermal conditions over the BOB (e.g., Neelin et al. 1987; Emanuel et al. 1987, 1994; Privé and Plumb 2007; Boos and Storelvmo 2016). The eastward-propagating modes induced by the evaporation–wind feedback may further explain an early retreat of the western Pacific subtropical high that leads to the summer monsoon onset over the SCS (Xing et al. 2016). As the South Asia high does not appear over the TP and the latent heat release is not established over the TP before May, other mechanisms must be responsible for the persistence of thermal conditions over the TP. A likely candidate is feedbacks from snow albedo over the high-altitude plateau (e.g., Budyko 1969; Sellers 1969; Cess et al. 1991; Wu and Liu 2003; Hall 2004; Duan and Wu 2008; Wu et al. 2009). In addition, forced by BOB–TP thermal configuration, the explained variances for the zonal winds are always the maxima for both experiments.

Wu et al. (2012) suggested that the warmer TP will induce a stronger-than-normal northern branch of Indian summer monsoon. Recently, Shi et al. (2017) suggested that the Yunnan–Guizhou Plateau will impede the northern branch of southwesterly winds and force it to become an anomalous northerly. Subsequently, anomalous anticyclonic circulations associated with northerly anomalies tend to reduce precipitation over the BOB, the Indian subcontinent, and the Arabian Sea in April–August. These results agree with our findings, indicating that the thermal configuration associated with topography may play one of the crucial roles in the variability of Yunnan precipitation by modulating the northern branch of southwesterly winds that are located between the two key regions in this study (Fig. 2a).

Tamura et al. (2010) suggested that the atmosphere over the TP is heated by the diabatic heating from the surface and that the adiabatic subsidence heating in the upper troposphere is caused by the divergent flow from the BOB. Their results may provide a negative feedback that prevents the thermal gradient between BOB and TP from continuously increasing. However, we found that the diabatic heating over the BOB is basically independent of the diabatic heating over the TP, as the correlation coefficient between them is very small (−0.10). The two inconsistent results imply that the relationship of diabatic heating over the BOB with that over the TP may be a time lag rather than a synchronous relationship at interannual time scales. The complicated time-lag relationship between them and its impact on the precipitation in Yunnan during the rainy season are worth examining in future studies.

Xing et al. (2015, 2016) found that the early (late) onset of the BOB summer monsoon in relation to the warm sea surface temperature appearing in early April over the central BOB can enhance (reduce) May precipitation in the southern Indian Peninsula, the Indochina Peninsula, southwest China, and the South China Sea. By contrast, it reduces (enhances) May precipitation in south China by driving the subtropical high away from the BOB, enhancing the local convection, and further strengthening the southwesterlies in the BOB. The results associated with the anomalous diabatic heating over the BOB in this study are also consistent with their results on the relationship between the summer monsoon onset in the BOB and May precipitation in southwest China, but with higher correlation intensity and a larger area passing the significance test at the 90% confidence level. This implies that the BOB–TP thermal configuration may exert its effects on the May precipitation over a much broader domain than that listed in Xing et al. (2016). Therefore, the corresponding relationship between them is also worthy of investigation in future studies.

In addition, Figs. 5a,c,g, 6a–d, and Figs. 7a–c show that there are some correlations between the May precipitation of Yunnan and states of the South China Sea and mei-yu front. In future studies we will also examine the key physical processes by which the states of the South China Sea and mei-yu front influence the May precipitation of Yunnan.

Acknowledgments

We thank Prof. M. Watanabe for providing the linear baroclinic model and the three anonymous reviewers for their valuable comments, which lead to improving the manuscript. This work was supported by the National Key Research and Development Program of China (2016YFA0601600), and the National Natural Science Foundation of China (U1502233 and 41565002).

REFERENCES

REFERENCES
Barriopedro
,
D.
,
C. M.
Gouveia
,
R. M.
Trigo
, and
L.
Wang
,
2012
:
The 2009/10 drought in China: Possible causes and impacts on vegetation
.
J. Hydrometeor.
,
13
,
1251
1267
, doi:.
Boos
,
W. R.
, and
Z.
Kuang
,
2010
:
Dominant control of the South Asian monsoon by orographic insulation versus plateau heating
.
Nature
,
463
,
218
222
, doi:.
Boos
,
W. R.
, and
T.
Storelvmo
,
2016
:
Reply to Levermann et al.: Linear scaling for monsoons based on well-verified balance between adiabatic cooling and latent heat release
.
Proc. Natl. Acad. Sci. USA
,
113
,
E2350
E2351
, doi:.
Bretherton
,
S. C.
,
M.
Widmann
,
V. P.
Dymnikov
,
J. M.
Wallace
, and
I.
Blade
,
1999
:
The effective number of spatial degrees of freedom of a time-varying field
.
J. Climate
,
12
,
1990
2009
, doi:.
Broccoli
,
A. J.
, and
S.
Manabe
,
1992
:
The effects of orography on midlatitude Northern Hemisphere dry climates
.
J. Climate
,
5
,
1181
1201
, doi:.
Budyko
,
M. I.
,
1969
:
The effect of solar radiation variations on the climate of the earth
.
Tellus
,
21
,
611
619
, doi:.
Cao
,
J.
,
J. M.
Hu
, and
Y.
Tao
,
2012
:
An index of the interface between the Indian summer monsoon and the East Asian summer monsoon
.
J. Geophys. Res.
,
117
, D18108, doi:.
Cao
,
J.
,
P.
Yao
,
L.
Wang
, and
K.
Liu
,
2014a
:
Summer rainfall variability in low-latitude highlands of China and subtropical Indian Ocean dipole
.
J. Climate
,
27
,
880
892
, doi:.
Cao
,
J.
,
Y.
Tao
,
R. W.
Yang
, and
J. M.
Hu
,
2014b
: Impact of Indo-Pacific Air-Sea Interaction on Rainfall and Geological Disasters in Low Latitude Highland of China (in Chinese). Science Press, 255 pp.
Cess
,
R. D.
, and Coauthors
,
1991
:
Interpretation of snow-climate feedback as produced by 17 general circulation models
.
Science
,
253
,
888
892
, doi:.
Chang
,
C.-P.
, Ed.,
2004
: East Asian Monsoon. World Scientific Series on Meteorology of East Asia, Vol. 2, World Scientific, 572 pp.
Chen
,
W.
,
H. F.
Graf
, and
R. H.
Huang
,
2000
:
The interannual variability of East Asian winter monsoon and its relation to the summer monsoon
.
Adv. Atmos. Sci.
,
17
,
48
60
, doi:.
Chen
,
W.
,
J.
Feng
, and
R. G.
Wu
,
2013
:
Roles of ENSO and PDO in the link of the East Asian winter monsoon to the following summer monsoon
.
J. Climate
,
26
,
622
635
, doi:.
Chen
,
Y.
,
Y. H.
Ding
, and
Z. N.
Xiao
,
2006
:
The impact of water vapor transport on the summer monsoon onset and abnormal rainfall in Yunnan Province in May (in Chinese)
.
Chin. J. Atmos. Sci.
,
30
,
25
37
.
China Meteorological Administration
,
2013
: Criterion for determining the onset of the rainy season in southwest China. 5 pp.
Day
,
J. A.
,
I.
Fung
, and
C.
Risi
,
2015
:
Coupling of South and East Asian monsoon precipitation in July–August
.
J. Climate
,
28
,
4330
4356
, doi:.
Dee
,
D. P.
, and Coauthors
,
2011
:
The ERA‐Interim reanalysis: Configuration and performance of the data assimilation system
.
Quart.J. Roy. Meteor. Soc.
,
137
,
553
597
, doi:.
Deng
,
M. Y.
,
R. Y.
Lu
,
W.
Chen
, and
J.
Cao
,
2016
:
Interannual variability of precipitation in May over the South Asian monsoonal region
.
Int. J. Climatol.
,
36
,
1724
1732
, doi:.
Duan
,
A.
, and
G. X.
Wu
,
2008
:
Weakening trend in the atmospheric heat source over the Tibetan Plateau during recent decades. Part I: Observations
.
J. Climate
,
21
,
3149
3164
, doi:.
Emanuel
,
K. A.
,
1987
:
An air–sea interaction model of intraseasonal oscillations in the tropics
.
J. Atmos. Sci.
,
44
,
2324
2340
, doi:.
Emanuel
,
K. A.
,
J. D.
Neelin
, and
C. S.
Bretherton
,
1994
:
On large-scale circulations in convecting atmospheres
.
Quart. J. Roy. Meteor. Soc.
,
120
,
1111
1143
, doi:.
Flohn
,
H.
,
1957
:
Large-scale aspects of the “summer monsoon” in South and East Asia
.
J. Meteor. Soc. Japan
,
35A
,
180
186
, doi:.
Hall
,
A.
,
2004
:
The role of surface albedo feedback in climate
.
J. Climate
,
17
,
1550
1568
, doi:.
He
,
H. Y.
,
C.-H.
Sui
,
M. Q.
Jian
,
Z. P.
Wen
, and
G. D.
Lan
,
2003
:
The evolution of tropospheric temperature field and its relationship with the onset of Asian summer monsoon
.
J. Meteor. Soc. Japan
,
81
,
1201
1223
, doi:.
Jiang
,
X. W.
, and
J. P.
Li
,
2011
:
Influence of the annual cycle of sea surface temperature on the monsoon onset
.
J. Geophys. Res.
,
116
,
D10105
, doi:.
Krishnamurti
,
T. N.
,
1985
:
Summer monsoon experiment—A review
.
Mon. Wea. Rev.
,
113
,
1590
1626
, doi:.
Krishnamurti
,
T. N.
,
L.
Stefanova
, and
V.
Misra
,
2013
: Tropical Meteorology: An Introduction. Springer Science+Business Media, 424 pp., doi:.
Li
,
C. P.
, and
M.
Yanai
,
1996
:
The onset and interannual variability of the Asian summer monsoon in relation to land–sea thermal contrast
.
J. Climate
,
9
,
358
375
, doi:.
Lu
,
R. Y.
, and
Z. D.
Lin
,
2009
:
Role of subtropical precipitation anomalies in maintaining the summertime meridional teleconnection over the western North Pacific and East Asia
.
J. Climate
,
22
,
2058
2072
, doi:.
,
J. M.
,
J. H.
Ju
,
J. Z.
Ren
, and
W. W.
Gan
,
2012
:
The influence of the Madden-Julian oscillation activity anomalies on Yunnan extreme drought of 2009–2010
.
Sci. China Earth Sci.
,
55
,
98
112
, doi:.
Meehl
,
G. A.
,
1994
:
Coupled land-ocean-atmosphere processes and South Asian monsoon variability
.
Science
,
266
,
263
267
, doi:.
Neelin
,
J. D.
,
I. M.
Held
, and
K. H.
Cook
,
1987
:
Evaporation–wind feedback and low-frequency variability in the tropical atmosphere
.
J. Atmos. Sci.
,
44
,
2341
2348
, doi:.
Privé
,
N. C.
, and
R. A.
Plumb
,
2007
:
Monsoon dynamics with interactive forcing. Part I: Axisymmetric studies
.
J. Atmos. Sci.
,
64
,
1417
1430
, doi:.
Qin
,
J.
,
J. H.
Ju
, and
M. E.
Xie
,
1997
: Weather and Climate in Low Latitudes Plateau (in Chinese). China Meteorology Press, 210 pp.
Sellers
,
W. D.
,
1969
:
A global climatic model based on the energy balance of the earth–atmosphere system
.
J. Appl. Meteor.
,
8
,
392
400
, doi:.
Shi
,
Z. G.
,
Y. Y.
Sha
, and
X. D.
Liu
,
2017
:
Effect of Yunnan–Guizhou topography at the southeastern Tibetan Plateau on the Indian monsoon
.
J. Climate
,
30
,
1259
1272
, doi:.
Simmons
,
A. J.
, and Coauthors
,
2004
:
Comparison of trends and low-frequency variability in CRU, ERA-40, and NCEP/NCAR analyses of surface air temperature
.
J. Geophys. Res.
,
109
,
D24115
, doi:.
Tamura
,
T.
,
K.
Taniguchi
, and
T.
Koike
,
2010
:
Mechanism of upper tropospheric warming around the Tibetan Plateau at the onset phase of the Asian summer monsoon
.
J. Geophys. Res.
,
115
,
D02106
, doi:.
Tang
,
Y.
,
R. W.
Yang
,
J. M.
Hu
, and
W.
Huang
,
2013
:
ESVD analyses of the anomaly of sea surface temperature and the precipitation in May over low-latitude plateau of China (in Chinese)
.
J. Trop. Meteor.
,
29
,
449
457
.
Taniguchi
,
K.
, and
T.
Koike
,
2008
:
Seasonal variation of cloud activity and atmospheric profiles over the eastern part of the Tibetan Plateau
.
J. Geophys. Res.
,
113
,
D10104
, doi:.
Tao
,
Y.
,
J.
Cao
,
J. M.
Hu
, and
Z. C.
Dai
,
2013
:
A cusp catastrophe model of mid–long-term landslide evolution over low latitude highlands of China
.
Geomorphology
,
187
,
80
85
, doi:.
Wang
,
B.
,
2006
: The Asian Monsoon. Environmental Sciences, Springer-Verlag, 845 pp., doi:.
Wang
,
B.
, and
LinHo
,
2002
:
Rainy season of the Asian–Pacific summer monsoon
.
J. Climate
,
15
,
386
398
, doi:.
Wang
,
C. Y.
,
1983a
: Weather analysis of the impact of the beginning of the rainy season in Yunnan in May. The National Symposium Collections on the Tropical Summer Monsoon (in Chinese), People’s Press Yunnan Province, 189–197.
Wang
,
C. Y.
,
1983b
: Effects of monsoon anomalies on the early summer drought in Yunnan and its long term forecast. The National Symposium Collections on the Tropical Summer Monsoon (in Chinese), People’s Press Yunnan Province, 174–183.
Watanabe
,
M.
, and
M.
Kimoto
,
2000
:
Atmosphere-ocean thermal coupling in the North Atlantic: A positive feedback
.
Quart. J. Roy. Meteor. Soc.
,
126
,
3343
3369
, doi:.
Watanabe
,
M.
, and
F. F.
Jin
,
2003
:
A moist linear baroclinic model: Coupled dynamical convective response to El Niño
.
J. Climate
,
16
,
1121
1140
, doi:.
Webster
,
P. J.
,
V. O.
Magaña
,
T. N.
Palmer
,
J.
Shukla
,
R. A.
Tomas
,
M.
Yanai
, and
T.
Yasuari
,
1998
:
Monsoons: Processes, predictability and the prospects for prediction
.
J. Geophys. Res.
,
103
,
14 451
14 510
, doi:.
Weng
,
H. Y.
,
G. X.
Wu
,
Y. M.
Liu
,
K. B.
Swadhin
, and
T.
Yamagata
,
2011
:
Anomalous summer climate in China influenced by the tropical Indo-Pacific Oceans
.
Climate Dyn.
,
36
,
769
782
, doi:.
Wu
,
G. X.
, and
Y.
Liu
,
2003
:
Summertime quadruplet heating pattern in the subtropics and the associated atmospheric circulation
.
Geophys. Res. Lett.
,
30
,
1201
, doi:.
Wu
,
G. X.
,
Y.
Liu
,
X.
Zhu
,
W.
Li
,
R.
Ren
,
A.
Duan
, and
X.
Liang
,
2009
:
Multi-scale forcing and the formation of subtropical desert and monsoon
.
Ann. Geophys.
,
27
,
3631
3644
, doi:.
Wu
,
G. X.
,
Y. M.
Liu
, B. He,
Q.
Bao
,
A. M.
Duan
, and
F.-F.
Jin
,
2012
:
Thermal controls on the Asian summer monsoon
.
Sci. Rep.
,
2
, 404, doi:.
Xing
,
N.
,
J. P.
Li
, and
L. N.
Wang
,
2016
:
Effect of the early and late onset of summer monsoon over the Bay of Bengal on Asian precipitation in May
.
Climate Dyn.
,
47
,
1961
1970
, doi:.
Xing
,
N.
,
J. P.
Li
,
X. W.
Jiang
, and
L. N.
Wang
,
2016
:
Local oceanic precursors for the summer monsoon onset over the Bay of Bengal and the underlying processes
.
J. Climate
,
29
,
8455
8470
, doi:.
Xu
,
M. L.
,
X.
Duan
, and
M. H.
Qi
,
2011
:
Cold wave
. Weather Forecaster Handbook of Yunnan Province (in Chinese), M. L. Xu, X. Duan, and M. H. Qi, Eds., Meteorological Press, 183–194.
Xu
,
X. D.
,
L. S.
Chen
,
X. R.
Wang
,
Q. J.
Miao
, and
S. Y.
Tao
,
2004
:
Moisture transport source/sink structure of the Meiyu rain belt along the Yangtze River valley
.
Chin. Sci. Bull.
,
49
,
181
188
, doi:.
Yan
,
H. M.
,
Z. N.
Xiao
, and
L.
Wang
,
2003
:
Activities of Bay of Bengal monsoon and beginning date of rain season in Yunnan (in Chinese)
.
Plateau Meteor.
,
22
,
624
630
.
Yan
,
H. M.
,
L.
Wang
,
G. L.
Zhou
, and
Y. B.
Lu
,
2007
:
The relationship between rainfall in Yunnan summer and the circulation in preceding winter (in Chinese)
.
J. Appl. Meteor. Sci.
,
18
,
340
349
.
Yan
,
H. S.
,
Y. B.
Lu
, and
W. H.
You
,
1995
:
A study of weather and climate features of precipitation in May in Yunnan (in Chinese)
.
Quart. J. Appl. Meteor.
,
6
,
124
128
.
Yanai
,
M.
, and
C.
Li
,
1994
:
Mechanism of heating and the boundary layer over the Tibetan Plateau
.
Mon. Wea. Rev.
,
122
,
305
323
, doi:.
Yanai
,
M.
,
C.
Li
, and
Z.
Song
,
1992
:
Seasonal heating of the Tibetan Plateau and its effects on the evolution of the Asian summer monsoon
.
J. Meteor. Soc. Japan
,
70
,
319
351
, doi:.
Yang
,
Y. L.
,
Y.
Du
,
H. S.
Chen
, and
Y. S.
Zhang
,
2011
:
Influence of ENSO event on rainfall anomaly in Yunnan Province and its neighboring regions during late spring-early summer (in Chinese)
.
Chin. J. Atmos. Sci.
,
35
,
729
738
.

Footnotes

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