Abstract

The impacts of summer atmospheric heat source over the Tibetan Plateau (TP) on regional climate variation have attracted extensive attention. However, few studies have focused on possible causes of the interannual variation of atmospheric heat source over the TP. Total heat (TH) is generally composed of three components: surface sensible heat, latent heat release of condensation (LH), and radiative convergence. In this study, it is found that interannual variation of summer TH is dominated by LH in the central and eastern TP. The atmospheric circulation patterns associated with the TH over the TP in June are different from those in July and August. Large TH is accompanied by a cyclone centered over the South China Sea in June, which is replaced by an anticyclone in July and August. The interannual variation of July–August TH over the central and eastern TP is significantly affected by convection around the western Maritime Continent (WMC) that modulates the LH over the southeastern TP. Enhanced WMC convection induces an anticyclone to the south of the TP, which favors water vapor transport to the southeastern TP and thus an increase in precipitation. Enhanced convection over the southeastern TP may exert a positive feedback on local precipitation through pumping more water vapor from the southern boundary. Both observations and model simulations indicate that the enhanced WMC convection can induce the anticyclone to the south of the TP and convection–circulation is important for maintenance of the anticyclone.

1. Introduction

Atmospheric heat source over the Tibetan Plateau (TP) plays an important role in the evolution of the Asian summer monsoon (e.g., Wu and Zhang 1998; Wu et al. 2007, 2012; Duan and Wu 2008) and regional hydrological cycle on a wide range of time scales (e.g., Xu et al. 2008, 2010). Much effort has been devoted to estimating the magnitude, spatial pattern, and temporal variation of atmospheric heat source over the TP and their climate impacts (e.g., Ye and Gao 1979; Luo and Yanai 1984; Yanai et al. 1992; Zhao and Chen 2001a; Ueda et al. 2003; Yang et al. 2011). Seasonal evolution of the total heating (TH) over the TP is closely related to the seasonal evolution of the Asian summer monsoon, and the climatological spatial patterns and dominant components of TH are different between dry and rainy seasons (Luo and Yanai 1984; Yanai et al. 1992; Wu and Zhang 1998; Ueda et al. 2003; Xu et al. 2009; Tamura et al. 2010). Summer TH over the TP exerts significant impacts on East Asian summer climate on both interannual (e.g., Zhao and Chen 2001b; Hsu and Liu 2003) and interdecadal (e.g., Ding et al. 2009; Wu et al. 2010; Wu et al. 2012a) time scales. Strong (weak) TH favors above-normal precipitation over northern (southern) China (e.g., Zhao and Chen 2001b). Summer TH over the TP also modulates the relationship between the El Niño–Southern Oscillation and the East Asian summer monsoon (Wu et al. 2012b). Thus, a better understanding of the variations of TH provides insight into the variation and prediction of East Asian climate.

Previous studies have suggested that the snow in the preceding winter or spring over the TP could influence the TH in the following summer via its persistent effects on surface albedo and soil moisture (e.g., Ding et al. 2009). Thus, the winter and spring snow over the TP has been considered one of the predictors for the following summer climate over East Asia. However, Zhao et al. (2007) argued that the snow cover in the TP in April–May could affect the TH in June but not in July and August because of the short memory of soil moisture. Thus, there may be other factors that are responsible for the variation of summer TH over the TP, especially in July and August.

TH comprises three components: surface sensible heat (SH), latent heat release of condensation (LH), and radiative convergence (RC). Previous studies are mostly focused on the variations of TH rather than its individual components (e.g., Zhao and Chen 2001a; Ding et al. 2009). However, the various TH components are generally related to different land and atmospheric processes. A further analysis on variations of the three components could lead to a better understanding of the variation of TH. On the other hand, while the contributions of the different TH components to the seasonal evolution of TH have been investigated (e.g., Yang et al. 2011), related features on the interannual time scale are unclear.

A reliable estimation of TH is the basis for understanding its interannual variation. Previous studies have estimated the heat source based on analysis or reanalysis datasets in which very limited radiosonde data are assimilated or by using biased satellite products and oversimplified methods (Ye and Gao 1979; Yanai et al. 1992; Zhao and Chen 2001a; Ueda et al. 2003). Yang et al. (2011) concluded that these estimations caused biases in both peak month and long-term trend of TH. Based on high-accuracy experiment data, an updated land surface model, and carefully selected satellite data, Yang et al. (2011) developed a state-of-the-art estimate of the TH and its components over the TP. The new data provide an opportunity to investigate the contributions of various components to the interannual variation of atmospheric heat source over the TP.

Tropical precipitation-related latent heating plays an important role in determining the characteristics of atmospheric circulation and its variations (e.g., Nitta 1987; Nitta and Hu 1996; Hu and Nitta 1997; Wang et al. 2008a; Lu and Lin 2009). For example, the convection anomalies around the Philippine Sea (PS) and associated atmospheric circulation modulate the interannual variation of the East Asian summer monsoon (e.g., Nitta 1987; Wang et al. 2008a; Jiang et al. 2013a). Since state-of-the-art dynamical models have shown skills in the seasonal-to-interannual prediction of tropical convection (e.g., Jiang et al. 2013a,b,c), a decent operational prediction of extratropical climate may be achieved if the extratropical climate is closely linked to tropical convection and such a link can be identified.

The Maritime Continent (MC), characterized by unique geographical features, is within the most extensive rainy area on Earth (e.g., Qian 2008). As indicated by Ramage (1968), the release of latent heat over the MC provides a primary energy source for global atmospheric circulation. Neale and Slingo (2003) also pointed out that the MC convection anomaly exerted a significant impact on global climate anomaly. Strong convection over the MC favors a strong East Asian winter monsoon (e.g., Jiang et al. 2013c). Recently, Jiang et al. (2015) reported that the MC convection anomaly affected the summer rainfall over southeastern China and parts of the eastern TP, implying that the MC convection may affect summer TH over the TP by modulating LH.

Because of the important role of the TH over the TP in understanding the variation and prediction of East Asian climate, this study is aimed at revealing the interannual variation of TH over the TP and its possible causes, with a focus on the different contributions of LH, SH, and RC to TH and the link between the LH over the TP and the convection over MC. The remainder of this paper is organized as follows. A description of the data and methods applied is presented in section 2. We analyze the interannual variation of TH and the contribution of various components as well as the relationship of TH with precipitation and atmospheric circulation in section 3. Factors responsible for the interannual variation of July and August precipitation over the TP are discussed in section 4. The role of convection over the western MC (WMC) in the interannual variation of precipitation over the TP are presented in section 5. Finally, a discussion and summary are presented in sections 6 and 7, respectively.

2. Data and methods

The state-of-the-art estimations of SH, LH, and RH developed by Yang et al. (2011) are applied in this study. SH and surface radiation flux are estimated by the updated second-generation Simple Biosphere Model (SiB2; Sellers et al. 1996), which has been improved according to experimental data analyses and TP surface conditions (Yang et al. 2009). LH is calculated by corrected gauge-measured precipitation (Ye et al. 2004). The net radiation flux at the top of the atmosphere is the mean of the ISCCP flux data (FD; Zhang et al. 2004) and the GEWEX SRB (Stackhouse et al. 2004) products. More details about the data are provided in Yang et al. (2011). Because radiation estimation starts in January 1984 and ends in June 2005, the record length of June–July–August (JJA) mean atmospheric heat source is 21 years. The total number of stations is 77, and the locations of these stations are shown in Fig. 1d.

Fig. 1.

Patterns of mean (shading; W m−2) and variance (contours; W m−2) of JJA (a) TH, (b) SH, (c) LH, and (d) RC over the TP from 1984 to 2004. Surface observation stations whose data are used in this study are denoted by open cycles in (d). The black thick solid lines denote the topographic height of 2000 m.

Fig. 1.

Patterns of mean (shading; W m−2) and variance (contours; W m−2) of JJA (a) TH, (b) SH, (c) LH, and (d) RC over the TP from 1984 to 2004. Surface observation stations whose data are used in this study are denoted by open cycles in (d). The black thick solid lines denote the topographic height of 2000 m.

The datasets used in this study also include ERA-Interim with a horizontal resolution of 1.5° latitude × 1.5° longitude (Dee et al. 2011), monthly mean sea surface temperature (SST) from the NOAA Extended Reconstructed SST version 3b with a horizontal resolution of 2° in both latitude and longitude (Smith et al. 2008), the North Atlantic Oscillation (NAO) index from the Climate Prediction Center of NOAA (http://www.cpc.ncep.noaa.gov/products/precip/CWlink/pna/nao.shtml), and the precipitation data (1979–2013) from the latest version (V3) of surface climatological data compiled by the China National Meteorological Information Center. Monthly mean geopotential height, temperature, and winds from ERA-Interim are used. All statistical significance tests for linear correlation analysis are performed using the two-tailed Student’s t test.

Vertically integrated moisture flux can be expressed as follows:

 
formula

where is specific humidity, V is horizontal wind vector, is pressure, is surface pressure, and is acceleration due to gravity.

3. Dominant components of the interannual variation of atmospheric heat source over the TP

To reveal the different contributions of various components to the temporal variation of TH, we first analyze the variances of TH and its three components. Figure 1 shows the climatological mean JJA TH, SH, LH, RC, and their variances over the TP. It can be seen that the TH over the southeastern TP is higher than that over the northeastern and southwestern TP. The spatial pattern of TH variance resembles the climatological pattern, with a maximum over the southeastern TP. Comparisons among the patterns of mean TH, SH, LH, and RC as well as their variances indicate that the spatial variation of TH is overall dominated by the LH, which has the largest magnitude for both the mean and the variance over most of the TP. The variance of LH is mostly higher than 20 W m−2, but the variances of SH and RC only reach 5 and 10 W m−2, respectively. It is also noted that the spatial pattern of variance basically resembles that of the means for SH, LH, and RC, with maxima over the southeastern TP for LH and RC but the southwestern TP for SH.

Although the variance of TH is dominated by the variance of LH, the contribution of LH to TH shows large spatial inhomogeneity (Fig. 1). Thus, we further analyze the variances of TH and its three components in different regions. Following Yang et al. (2011), the TP is divided into three domains: western TP (WTP; west of 85°E), central TP (CTP; 85°–95°E), and eastern TP (ETP; east of 95°E), and there are 3, 20, and 54 stations in WTP, CTP, and ETP, respectively. Table 1 lists the variances of TH, SH, LH, and RC for the three domains and the entire TP. The variance of LH is the largest among the three components for all three domains. The variance of RC ranks second for CTP and ETP, while that of SH ranks second over WTP. The variance of TH over CTP and ETP is comparable, and it is about twice as large as that over WTP.

Table 1.

Variance of JJA mean atmospheric heat source over various parts of the TP from 1984 to 2004. Values in brackets are calculated from the detrended atmospheric heat source components.

Variance of JJA mean atmospheric heat source over various parts of the TP from 1984 to 2004. Values in brackets are calculated from the detrended atmospheric heat source components.
Variance of JJA mean atmospheric heat source over various parts of the TP from 1984 to 2004. Values in brackets are calculated from the detrended atmospheric heat source components.

Yang et al. (2011) have reported that there is a considerable linear trend in TH and its various components. As shown in Fig. 2, an obvious decreased trend also appears in the RC over CTP and ETP related to the warming and change in cloud height over the TP (see Wu et al. 2015). This trend apparently increases the variance of RC and TH over CTP and ETP (Table 1). For example, the variance of detrended RC is just about half of the variance of the raw RC over ETP. The variance of detrended LH is apparently higher than that of detrended RC and SH, indicating that the year-to-year variation of TH is dominated by LH.

Fig. 2.

Temporal variations of JJA (a) TH, (b) SH, (c) LH, and (d) RC for various regions of the TP.

Fig. 2.

Temporal variations of JJA (a) TH, (b) SH, (c) LH, and (d) RC for various regions of the TP.

We further analyze the temporal correlation among TH, SH, LH, and RC on the interannual scale (Table 2). For the whole TP, the variation of raw TH is dominated by RC because of its linear trend. After removing the linear trend, the year-to-year variation of TH is dominated by LH. In spite of its small variance, SH is highly correlated with TH because it is strongly related to LH (or precipitation). Since we focus on year-to-year variations, the following discussions are merely based on detrended time series. SH is negatively and highly correlated with LH over the whole TP, but both SH and LH are not significantly correlated with RC. LH is highly correlated with TH over both the CTP and ETP, but TH is not significantly correlated with any component over the WTP, meaning that the interannual variation of TH over the WTP is not dominated by any particular component.

Table 2.

Coefficients of correlation among detrended JJA mean SH, LH, RC, and TH for various regions of the TP from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

Coefficients of correlation among detrended JJA mean SH, LH, RC, and TH for various regions of the TP from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.
Coefficients of correlation among detrended JJA mean SH, LH, RC, and TH for various regions of the TP from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

Different components of TH have different features and contributions from the WTP to the ETP. What are the relationships of these components among the various domains on an interannual time scale? As shown in Table 3, TH and its components are all significantly and positively correlated between the CTP and ETP. However, they are insignificantly correlated with those over the WTP except for LH.

Table 3.

Coefficients of correlation among various regions of the TP for detrended JJA mean total heating and its three components from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

Coefficients of correlation among various regions of the TP for detrended JJA mean total heating and its three components from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.
Coefficients of correlation among various regions of the TP for detrended JJA mean total heating and its three components from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

The above analyses indicate that the year-to-year variation of JJA mean TH is dominated by LH in both magnitude and phase, which is mainly contributed by the in-phase variation of LH between the CTP and ETP and their dominant roles in the variation of TH. The three components of TH also have large subseasonal variations (e.g., Yang et al. 2011). Does the contribution of LH to TH also exhibit a subseasonal variation? Table 4 lists coefficients of correlations between the three components and TH in June, July, and August. The correlation between LH and TH increases from June to August. The contribution of RC to TH in June is more important than that in July and August, especially for July when RC is not correlated with TH. However, the correlation between SH and TH in June is obviously lower than that in July and August. Given the significant and negative correlation between SH and LH, these subseasonal variations of correlations indicate that LH plays a more important role in the variation of TH in July and August. This subseasonal variation may be caused by the subseasonal variation of rainfall, which is less heavy in June than in July and August since June is a transition month from dry season to rainy season in most of the TP (figures not shown).

Table 4.

Coefficients of correlation of detrended TH with detrended SH, LH, and RC for June, July, and August from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

Coefficients of correlation of detrended TH with detrended SH, LH, and RC for June, July, and August from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.
Coefficients of correlation of detrended TH with detrended SH, LH, and RC for June, July, and August from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

Table 5 shows coefficients of correlations among June, July, and August for detrended SH, LH, RC, and TH from 1984 to 2004. Correlations between July and August are generally higher than those for June–July and June–August for TH and its three components, indicating more coherent interannual variations in July and August. Both SH and RC show more coherent variations in July and August compared to LH. The subseasonal variations of interannual variation of TH and the contribution of the three components to TH imply that the factors responsible for the year-to-year variation of June TH may be different from those for July–August TH.

Table 5.

Coefficients of correlation among June, July, and August for detrended SH, LH, RC, and TH from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

Coefficients of correlation among June, July, and August for detrended SH, LH, RC, and TH from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.
Coefficients of correlation among June, July, and August for detrended SH, LH, RC, and TH from 1984 to 2004. Values in italic and bold fonts exceed the confidence levels of 95% and 99%, respectively.

4. Precipitation and atmospheric circulations associated with interannual variation of TP TH

a. JJA mean

To identify the possible factors affecting the interannual variation of TH, we first analyze its relation to precipitation and atmospheric circulation. Figure 3 shows the patterns of regression of precipitation and circulation against the detrended TH. Strong TH over the TP is associated with above-normal precipitation from the central TP to the east of Japan and over the MC, especially the vicinity of WMC, and below-normal precipitation from northern India to the PS, with centers over the northeastern Bay of Bengal (BOB), southeastern China, and the PS (Fig. 3a). In the lower troposphere, there is anticyclonic circulation to the south of the TP and over the western North Pacific, cyclonic circulation around the Sea of Japan, and a pair of cyclonic circulations over the tropical Indian Ocean (IO) with anomalous westerlies over the equator (Fig. 3a). These anomalies of atmospheric circulation develop in the midtroposphere (figures not shown). There is a convergent pattern of 500-hPa winds over the eastern TP, with significant southwesterlies over the southern TP (figures not shown). In the upper troposphere, there is a strong anticyclone centered over the southeastern TP, accompanied by two cyclones to the northeast and northwest, respectively (Fig. 3b).

Fig. 3.

Patterns of regression of JJA mean winds (vectors; m s−1) at various levels and rainfall (shading; mm day−1) against the detrended JJA total heat source over the TP. (a) 850-hPa winds and rainfall, (b) 200-hPa winds and their divergence (shading; ×10−5 s−1). Only values exceeding the 90% confidence level are shown for rainfall. Red vectors denote the values of either zonal or meridional wind component exceeding the 90% confidence level. The black dashed lines denote the topographic height of 1500 m in (a) and 3000 m in (b).

Fig. 3.

Patterns of regression of JJA mean winds (vectors; m s−1) at various levels and rainfall (shading; mm day−1) against the detrended JJA total heat source over the TP. (a) 850-hPa winds and rainfall, (b) 200-hPa winds and their divergence (shading; ×10−5 s−1). Only values exceeding the 90% confidence level are shown for rainfall. Red vectors denote the values of either zonal or meridional wind component exceeding the 90% confidence level. The black dashed lines denote the topographic height of 1500 m in (a) and 3000 m in (b).

The anomalies of atmospheric circulation and precipitation are well coupled dynamically. The above-normal precipitation from the southeastern TP to the east of Japan is accompanied by lower-tropospheric convergence and upper-tropospheric divergence (Fig. 3b). The enhanced convection over the tropical WMC favors the pair of cyclones over the IO and the aloft divergent wind pattern (e.g., Gill 1980; Fig. 3b). The suppressed convection from the northern BOB to the east of the PS is accompanied by an anticyclonic circulation at the lower–midtroposphere but a convergent wind pattern at the upper troposphere (Fig. 3b).

b. Subseasonal variations

The contributions of various components to TH exhibit considerable subseasonal variability. As shown in Fig. 4, the precipitation and atmospheric circulation associated with TH are also different from June to August. The anomalous patterns of precipitation and atmospheric circulation in July and August basically resemble those of the JJA mean (Figs. 4b,c) but are different from those in June (Fig. 4a). There are also differences in precipitation and atmospheric circulation between July and August. The anomalous anticyclone over the western North Pacific and the associated precipitation anomalies to its north shift northward from July to August. The anomalous precipitation over the PS becomes weaker from July to August. The convection over the WMC and the associated cyclone over northern IO are strong in July but become weaker in August (Figs. 4b,c).

Fig. 4.

Patterns of regression of 850-hPa winds (vectors; m s−1) and rainfall (shading; mm day−1) against the detrended simultaneous total heat source over the TP for (a) June, (b) July, (c) August, and (d) mean of July and August. For rainfall, only the values exceeding the 90% confidence level are shown. Red vectors denote values of either zonal or meridional wind component exceeding the 90% confidence level. The black dashed lines denote the topographic height of 1500 m.

Fig. 4.

Patterns of regression of 850-hPa winds (vectors; m s−1) and rainfall (shading; mm day−1) against the detrended simultaneous total heat source over the TP for (a) June, (b) July, (c) August, and (d) mean of July and August. For rainfall, only the values exceeding the 90% confidence level are shown. Red vectors denote values of either zonal or meridional wind component exceeding the 90% confidence level. The black dashed lines denote the topographic height of 1500 m.

The significant anomalies of precipitation and atmospheric circulation are less extensive in June than in July and August (Figs. 4a–c). Precipitation shows a significant increase only over a small portion of the southeastern TP and the northern South China Sea. The small extent of significant precipitation over the southern TP in June is consistent with the above result that LH plays a more important role in the variation of TH in July and August. To the south of the TP the anomalous cyclonic circulation in June is replaced by an anomalous anticyclonic circulation in July–August. The weak anomalous cyclone centered over the northern SCS in June is replaced by an anomalous anticyclone over the western North Pacific in July and August. Previous studies have also reported that a strong JJA TH over the TP is associated with strong southwesterlies over Southeast Asia and enhanced mei-yu, baiu, and changma over eastern China, Japan, and Korea (e.g., Zhao and Chen 2001b). However, the above analyses indicate that these features appear only in July and August.

Both theoretical and numerical studies have reported that heating over the TP induces a lower-tropospheric cyclonic circulation around the TP (e.g., Wu et al. 2007; Wang et al. 2008b). Thus, the cyclonic circulation along the southern and southeastern TP in the lower troposphere may be a response to the TP heating (Figs. 4b,c). In addition, TH is significantly correlated with precipitation and atmospheric circulation over several remote regions, even the southern tropical IO (Figs. 4b,c). However, numerical simulations indicate that TP heating, especially that over the main body of the plateau, does not exert a significant impact on atmospheric circulation over remote regions (Boos and Kuang 2010; Wu et al. 2012). Because the circulation anomalies analyzed above, especially the southwesterlies to the south of the TP, favor enhanced precipitation over the southeastern TP and because TH is dominated by LH, the precipitation and atmospheric circulation over the remote regions may exert a remote impact on the variation of TH over the TP through the modulation of LH.

For verification, we further analyze the atmospheric circulation responsible for the variation of precipitation over the TP. Because of the longer data record, the following calculations related to precipitation are based on the period of 1979–2013, and LH is constructed as the mean precipitation at the 77 stations as shown in Fig. 1d. The TP LH is shown in Fig. 5, and patterns of regression of atmospheric circulation and precipitation on the LH are shown in Fig. 6. It can be seen that overall the anomalous patterns of precipitation and 850-hPa winds related to LH resemble those related to TH, although the time periods are different, further confirming the dominant role of LH in the year-to-year variation of TP TH (Figs. 4 and 6). However, differences also can be seen. The precipitation anomalies related to the LH over western MC become stronger in August compared to those related to TH. The cyclone pattern related to the LH in June extends from the South China Sea to the PS, while the cyclone related to the TH covers only the South China Sea.

Fig. 5.

Year-to-year variations of normalized July–August mean rainfalls over the TP (black lines with plus) and the western Maritime Continent (7.5°S–5.0°N, 85°–115°E; red lines with open circles).

Fig. 5.

Year-to-year variations of normalized July–August mean rainfalls over the TP (black lines with plus) and the western Maritime Continent (7.5°S–5.0°N, 85°–115°E; red lines with open circles).

Fig. 6.

As in Fig. 4, but detrended TH is replaced by the average rainfall over the TP from 1979 to 2013. The black boxes denote the region over 7.5°S–5.0°N, 85°–115°E, where rainfall is used to construct the convection index over the western Maritime Continent.

Fig. 6.

As in Fig. 4, but detrended TH is replaced by the average rainfall over the TP from 1979 to 2013. The black boxes denote the region over 7.5°S–5.0°N, 85°–115°E, where rainfall is used to construct the convection index over the western Maritime Continent.

How do these circulation anomalies affect the precipitation over the TP? Because of the large difference in anomalous precipitation and 850-hPa winds between June and July–August and the relatively smaller contribution of LH to TH in June, the following analyses focus on the July and August mean. Figure 7 shows composite patterns of anomalous water vapor transport in July and August for the years with normalized TP precipitation above one standard deviation (1980, 1993, and 1998) or below one standard deviation (1994, 1997, 2006, 2011, and 2013) (see Fig. 5). To quantitatively reveal the role of water vapor in the precipitation anomalies, water vapor budget (Table 6) is calculated over the central and eastern TP based on the pattern of precipitation anomalies related to TH. Results indicate that above- (below-)normal precipitation corresponds to more (less) net water vapor over the TP. Above-normal precipitation is accompanied by more tropical water vapor transported to the TP through the southern boundary by the anomalous southwesterlies associated with the anticyclonic circulation to the south of the TP (Figs. 6d and 7a). For below-normal precipitation, the features are nearly opposite (Fig. 7b). There is anomalous westerly (easterly) water vapor transport across the southern TP, and thus the increase (decrease) in inflow through the western boundary is almost balanced by the increase (decrease) in outflow through the eastern boundary (Table 5). The anomalies of water vapor through the northern boundary are small. Because the inflow of water vapor is mainly from the southern boundary, the increase (decrease) in net water vapor can be attributed to the increase (decrease) in inflow through the southern boundary (Table 5). Therefore, the July–August precipitation over the TP should be affected by the anomalous anticyclone to its south via modulating water vapor transportation. We will discuss the possible causes for the formation of the anticyclone in the following section.

Fig. 7.

Composite patterns of vertically integrated moisture transport (vectors; kg m−1 s−1) in July and August for normalized precipitation over the TP for (a) greater than one standard deviation and (b) less than one negative standard deviation. Values exceeding the 95% confidence level are shaded. The red boxes denote the boundary of central and eastern TP (27°–34.5°N, 85.5°–102°E).

Fig. 7.

Composite patterns of vertically integrated moisture transport (vectors; kg m−1 s−1) in July and August for normalized precipitation over the TP for (a) greater than one standard deviation and (b) less than one negative standard deviation. Values exceeding the 95% confidence level are shaded. The red boxes denote the boundary of central and eastern TP (27°–34.5°N, 85.5°–102°E).

Table 6.

July–August mean total column moisture transport (106 kg s−1) across the four boundaries of 85.5°–102°E, 27°–34.5°N and corresponding atmospheric water budget over the central and eastern TP.

July–August mean total column moisture transport (106 kg s−1) across the four boundaries of 85.5°–102°E, 27°–34.5°N and corresponding atmospheric water budget over the central and eastern TP.
July–August mean total column moisture transport (106 kg s−1) across the four boundaries of 85.5°–102°E, 27°–34.5°N and corresponding atmospheric water budget over the central and eastern TP.

5. Role of convection over the western Maritime Continent

a. Observations

Jiang et al. (2015) recently indicated that the out-of-phase variation of precipitation over southeastern China and the eastern edge of the TP was induced by the enhanced convection over the MC. The strong LH is also accompanied by significantly enhanced convection over WMC. Then, what is the possible link between them?

We define a WMC convection index (WMCCI; Fig. 5) as the average precipitation over 7.5°S–5.0°N, 85°–115°E to understand the role of convection over the MC. Figure 8 presents the patterns of regressions of 850-hPa winds and precipitation against the WMCCI. Strong WMC convection is accompanied by below-normal precipitation over the PS, the equatorial western Pacific, and the northern BOB; anticyclonic circulation over the western North Pacific; and cyclonic circulation over the northwestern Indian Ocean. Anomalies of precipitation and winds to the south of the TP exhibit substantial subseasonal variation. Significant westerly and southerly anomalies to the south of the TP associated with an anticyclone over the BOB are only observed in July and August, accompanied by significant suppressed precipitation over a larger region over the northern BOB compared to that in June. Wind anomalies to the south of the TP are weak in June. Enhanced WMC convection is also accompanied by above-normal precipitation over the southeastern TP in only July and August. The coefficients of correlation between WMCCI and precipitation averaged in the TP are 0.53 and 0.44 for July and August, respectively, exceeding the 99% confidence level. Thus, the anticyclone circulation to the south of the TP may be the medium by which the WMC convection affects the precipitation over the southern TP in July and August. The magnitude of the anticyclone may be affected by the magnitude of precipitation anomaly over the northern BOB.

Fig. 8.

As in Fig. 4, but detrended TH is replaced by the average rainfall from 1979 to 2013 over the western Maritime Continent (7.5°S–5.0°N, 85°–115°E), denoted by the black boxes.

Fig. 8.

As in Fig. 4, but detrended TH is replaced by the average rainfall from 1979 to 2013 over the western Maritime Continent (7.5°S–5.0°N, 85°–115°E), denoted by the black boxes.

Figure 9 shows the July–August atmospheric circulation associated with WMC convection. In the upper troposphere, as a response to the anomalous latent heating below, there is a pair of anticyclonic circulations over the western IO. The southwesterlies associated with the cyclonic circulation over the northwestern IO meet the northeasterlies associated with the anticyclone centered over the southeastern TP, resulting in convergence there. The anomalies of meridional circulation indicate that strong convection over the WMC alters the local Hadley circulation; the air rises around 5°S, moves northward in the upper troposphere, and descends around 15°N, where below-normal precipitation is observed. That is, the WMC convection could suppress precipitation over the northern BOB by changing the local Hadley circulation.

Fig. 9.

Patterns of regression of July–August mean (a) 200-hPa winds (vectors; m s−1) and their divergence (shading; ×10−5 s−1) (shadings; gpm) and (b) meridional circulation averaged over 85°–100°E (units of V wind and vertical velocity are m s−1 and −100 Pa s−1, respectively) against the simultaneous rainfall over the western Maritime Continent. Red vectors denote the values of either zonal or meridional wind component exceeding the 90% confidence level. The black dashed lines denote the topographic height of 3000 m in (a).

Fig. 9.

Patterns of regression of July–August mean (a) 200-hPa winds (vectors; m s−1) and their divergence (shading; ×10−5 s−1) (shadings; gpm) and (b) meridional circulation averaged over 85°–100°E (units of V wind and vertical velocity are m s−1 and −100 Pa s−1, respectively) against the simultaneous rainfall over the western Maritime Continent. Red vectors denote the values of either zonal or meridional wind component exceeding the 90% confidence level. The black dashed lines denote the topographic height of 3000 m in (a).

It is also noted that the anticyclone to the south of the TP is accompanied by decreased precipitation over the northern BOB (Figs. 3, 4, 6, and 8). The diabatic cooling associated with suppressed convection could induce an anticyclone to the northwest (e.g., Gill 1980; Jiang et al. 2013a). Thus, enhanced WMC convection may cause an anticyclone to the south of the TP by suppressing convection over the northern BOB.

b. LBM simulation

The linear baroclinic model (LBM) used in this study was developed by Watanabe and Kimoto (2000) and has been used for various purposes. Here, we use a dry version with a horizontal resolution of T42 and 20 sigma levels in vertical. The method of time integrations is adopted to obtain the linear atmospheric response to specific forcing. The mean flow used in the model is the climatological July–August mean circulation derived from ERA-40 (Uppala et al. 2005).

To understand the impact of the MC convection, the model is forced by a prescribed heat source over WMC, as shown in Figs. 10a,b. In the vertical, the heating has a sinusoidal profile with a maximum at sigma of 0.45 (about 400 hPa) to mimic the condensational heat released from deep convection. The heating has a cosine-squared profile in an elliptical region, with a center at 1.5°S, 100°E . The radiuses of the region in latitudinal and longitudinal directions are 15° and 7.5°, respectively. The maximum heating around 400 hPa is 1 K day−1, which is approximately equal to an anomalous precipitation rate of 2 mm day−1. The model is integrated for 30 days, and the response of atmospheric circulation to heating approaches a steady state after day 20. The mean results from day 20 to 29 are analyzed in this study.

Fig. 10.

The western Maritime Continent heat source in the linear baroclinic model. (a) The vertical profile of the specific heat source (K day−1) around the horizontal maximum heating center and (b) the spatial pattern of the specific heat source (K day−1) at the sigma level of 0.45. The heat source induces (c) 850-hPa winds (m s−1) and its divergence (shadings; ×10−6 m s−2) as well as (d) latitudinal (80°–100°E) mean meridional circulation (units of V wind and vertical velocity are m s−1 and −100 Pa s−1, respectively). Only winds with wind speed higher than 0.02 m s−1 are plotted in (c); the black dashed lines denote the topographic height of 1500 m in (c).

Fig. 10.

The western Maritime Continent heat source in the linear baroclinic model. (a) The vertical profile of the specific heat source (K day−1) around the horizontal maximum heating center and (b) the spatial pattern of the specific heat source (K day−1) at the sigma level of 0.45. The heat source induces (c) 850-hPa winds (m s−1) and its divergence (shadings; ×10−6 m s−2) as well as (d) latitudinal (80°–100°E) mean meridional circulation (units of V wind and vertical velocity are m s−1 and −100 Pa s−1, respectively). Only winds with wind speed higher than 0.02 m s−1 are plotted in (c); the black dashed lines denote the topographic height of 1500 m in (c).

Figures 10c,d present the atmospheric circulation induced by WMC heating. The model captures the observed lower-tropospheric cyclone over the western IO and the anticyclone to the south of the TP and the SCS, which resulted in divergence from the southern South China Sea to the northern BOB. It also simulates the observed increase in 500-hPa geopotential height to the south of the TP and the western North Pacific, the decrease in 500-hPa geopotential height over the IO, and the anticyclone to the south of the TP (figures not shown). The simulated meridional circulation resembles the observation, but there are no northerlies below the 700-hPa level. These results indicate that enhanced WMC convection induces many features responsible for the above-normal precipitation over the TP, especially the anticyclone to the south of the TP.

Because the experiments with the LBM do not include the feedback of moisture processes, the possible role of suppressed convection over the northern BOB in the anticyclone to the south of the TP is not reflected in the results shown in Figs. 10c,d. To reveal the possible role of reduced precipitation in the formation of the anticyclone, another experiment is conducted as the WMC heating experiment except that the heating is replaced by cooling over the northern BOB. Output from the simulation indicates that the diabatic cooling associated with the reduced precipitation indeed can induce an anticyclone to the northwest (figures not shown). Thus, convection–circulation feedback is important for the maintenance of the anticyclone.

c. General circulation model simulation

To further reveal the influence of WMC convection on rainfall over the TP and the important role of the convection–circulation feedback in the maintenance of the anomalous anticyclone over the northern BOB, we have conducted another two experiments by the Community Atmosphere Model version 5.30 (CAM5; Neale et al. 2010). The control simulation is integrated for 25 years using the finite-volume dynamical core at a horizontal resolution of 0.9° × 1.25° with 30 vertical levels. There is no chemical process during the integration. The model is forced with observed monthly climatologies of SST and sea ice. The control simulation is designated as CTL. Enhanced convection over the WMC is accompanied by local warm SST in observation, with a correlation coefficient of 0.48 for detrended data. Thus, a sensitivity experiment is performed that is identical to the CTL except that the SST is raised by 0.5°C uniformly in WMC (7.5°S–5.0°N, 85°–115°E) in July and August, which is referred to as WMC_warm.

The differences in July–August mean precipitation and 850-hPa winds of the last 20 years of the 25-year integrations between the WMC_warm and the CTL are shown in Fig. 11. Warm SST in the WMC causes an apparent increase in local precipitation in the simulation (Fig. 11a). Thus, the difference between the WMC_warm and the CTL can be regarded as a response to enhanced convection over the WMC.

Fig. 11.

Differences in (a) precipitation, (b) 850-hPa winds and their divergence (shading; ×10−5 s−1), (c) 200-hPa winds and their divergence (shading; ×10−5 s−1), (d) meridional circulation (units of V wind and vertical velocity are m s−1 and −100 Pa s−1, respectively) averaged in 80°–100°E between the WMC_warm and the CTL.

Fig. 11.

Differences in (a) precipitation, (b) 850-hPa winds and their divergence (shading; ×10−5 s−1), (c) 200-hPa winds and their divergence (shading; ×10−5 s−1), (d) meridional circulation (units of V wind and vertical velocity are m s−1 and −100 Pa s−1, respectively) averaged in 80°–100°E between the WMC_warm and the CTL.

Enhanced convection over the WMC can indeed cause an increase in precipitation over the southern and eastern TP and a strong anticyclone to the south of the TP. It also causes lower-tropospheric convergence and upper-tropospheric divergence over the WMC. The simulated meridional circulation indicates that air rises around the WMC, moves northward at the upper troposphere, and descends around the northern BOB. Thus, enhanced WMC convection suppresses rainfall over the northern BOB by inducing descending motion there. Comparison of the output between the CAM5 and the LBM indicates that the CAM5 simulates a stronger anticyclone to the south of the TP, highlighting the importance of the convection–circulation feedback in the formation of the anticyclone.

6. Discussion

Given that the features of regression of precipitation and atmospheric circulation are based on the precipitation over the WMC or TP and that the precipitation over the TP is highly correlated with the precipitation over the WMC (r = 0.56), many of the features associated with TP precipitation may be a response to the latent heating over the TP instead of that over WMC. To reveal the effect of TP heating on atmospheric circulation, we also conduct an experiment using the LBM, which is forced by a prescribed heat source over the central-eastern TP (see Figs. 12a,b). In the vertical, the heating has a sinusoidal profile with a maximum of 1 K day−1 at sigma of 0.55 (about 400 hPa). The heating has a cosine-squared profile in an elliptical region, with a center at 30°N, 95°E. The radiuses of the region in latitudinal and longitudinal directions are 10° and 5°, respectively.

Fig. 12.

As in Fig. 10, but for heat source over the central and eastern TP and meridional circulation averaged over 85°–105°E.

Fig. 12.

As in Fig. 10, but for heat source over the central and eastern TP and meridional circulation averaged over 85°–105°E.

The atmospheric circulation induced by TP heating is shown in Figs. 12c,d. The TP heating causes cyclonic circulation in the lower troposphere along the TP, with strong westerlies or southwesterlies to the south and southeast of the TP. Apparent 850-hPa wind anomalies are mainly located to the north of 10°N, consistent with the results of previous numerical studies (Boos and Kuang 2010; Wu et al. 2012). The simulated meridional circulation also indicates that the TP heating does not cause an apparent circulation anomaly to the south of 15°N. Compared to the atmospheric circulation induced by the heating over WMC, the lower-tropospheric circulation means that the heating of WMC is responsible for most of the features associated with the precipitation over the TP. It seems that the rainfall over the WMC is not affected by heating over the TP. However, it does not mean that the role of TP heating can be neglected. Because the southwesterlies induced by WMC heating are weak, the TP heating may play an important role in pumping water vapor to TP because it induces strong southwesterlies to the south of the TP.

As shown in Figs. 6 and 8, enhanced precipitation over both the TP and the WMC is accompanied by suppressed precipitation over the PS, which could excite an anticyclone to the northwest according to the Gill model and previous observation analyses (e.g., Gill 1980; Jiang et al. 2013a), and is the basis of Pacific–Japan teleconnection pattern (Nitta 1987; Lu 2004). Then, is it possible that the suppressed precipitation contributes to the anticyclone to the south of the TP? To answer this question, we conduct another numerical experiment by the LBM, which is forced by a prescribed cooling over the PS (see Figs. 13a,b). The simulated results indicate that the cooling over the PS also contributes to formation of the anticyclone to the south of the TP in July and August but only limited to the lower troposphere (Figs. 13c,d). However, the cooling-excited anticyclone over southeastern Asia extends from the lower troposphere to midtroposphere.

Fig. 13.

The PS diabatic cooling in the linear baroclinic model. (a) The vertical profile of the specific heat source (K day−1) around the horizontal maximum heating center and (b) the spatial pattern of the specific cooling (K day−1) at the sigma level of 0.45. The diabatic cooling induces (c) 850-hPa winds (m s−1) and its divergence (shadings; ×10−5 m s−2) as well as (d) 500-hPa winds (m s−1) and geopotential height (contours; m). Only winds with wind speed higher than 0.02 m s−1 are plotted in (c) and (d); the black dashed lines denote the topographic height of 1500 m.

Fig. 13.

The PS diabatic cooling in the linear baroclinic model. (a) The vertical profile of the specific heat source (K day−1) around the horizontal maximum heating center and (b) the spatial pattern of the specific cooling (K day−1) at the sigma level of 0.45. The diabatic cooling induces (c) 850-hPa winds (m s−1) and its divergence (shadings; ×10−5 m s−2) as well as (d) 500-hPa winds (m s−1) and geopotential height (contours; m). Only winds with wind speed higher than 0.02 m s−1 are plotted in (c) and (d); the black dashed lines denote the topographic height of 1500 m.

The simulated results by the CAM5 indicate that enhanced convection over the WMC causes a weak decrease in precipitation over the PS. But the simulation results reported by Jiang et al. (2015) and Neale and Slingo (2003) indicated that enhanced convection over the entire Maritime Continent can suppress precipitation over the PS. This difference implies that the suppressed precipitation over the PS may be caused by enhanced convection over the eastern Maritime Continent. That the observed enhanced WMC convection is accompanied by suppressed precipitation over the PS may be ascribed to the fact that the enhanced WMC convection is also accompanied by suppressed precipitation over the eastern Maritime Continent (Fig. 8).

Previous studies reported that the NAO can significantly modulate JJA precipitation over the TP (e.g., Liu and Yin 2001; Liu et al. 2015). The pattern of regression of July–August precipitation on detrended NAO shows that the NAO is negatively correlated with precipitation over the southeastern TP (figure not shown). The correlation between detrended NAO and simultaneous precipitation over the TP is −0.41. A positive NAO is not accompanied by significant circulation anomalies around the TP in the lower troposphere but accompanied by an anomalous cyclone over the TP in the upper troposphere (Liu et al. 2015). The anomalous cyclone is a part of a wave train originating from the North Atlantic. The wave train causes significant upper-tropospheric convergence over the TP, suppressing precipitation over the southeastern TP (figure not shown). Thus, the NAO may affect the TP precipitation dynamically by upper-tropospheric circulation. On the other hand, the WMC convection affects the TP precipitation by modulating the water vapor transport in the mid- and lower troposphere. However, the TP precipitation shows a higher correlation with the WMC convection compared to the NAO, suggesting that the WMC convection may play a more important role in the year-to-year variation of TP precipitation.

Because of the close link between the TP LH and the WMC convection, the former may be predicted if the latter is well predicted. Correlation between precipitation over the WMC and simultaneous SST shows that precipitation over the WMC is positively correlated with local SST but negatively correlated with SST in the central and eastern Pacific (figures not shown). The coefficient of correlation between the WMCCI and Niño-3.4 is −0.63. Because precipitation over the WMC is closed linked to both local and remote SST, it can be well predicted by the dynamical model and persistence of SST (Zhang et al. 2015). Because of the important role of the TP TH in variation of East Asian summer climate, predictability of TP TH requires further studies.

7. Summary

Based on the state-of-the-art estimation developed by Yang et al. (2011), we analyze the interannual variations of summer TH and its various components over the TP. Results show that the interannual variation of the TH is dominated by LH in the central and eastern TP. The contribution of LH to TH exhibits a considerable subseasonal variation, with a larger contribution in July and August than in June.

Previous studies have emphasized the persistent impact of the anomalies of winter–spring snow cover/depth and spring sensible heat on the anomalies of atmospheric heat source in summer (e.g., Wu et al. 2010; Wang et al. 2014). Here, we focus on the simultaneous impact of tropical convection on the interannual variation of atmospheric heat source over the TP. The WMC convection exerts a significant impact on the interannual variation of TP TH by modulating the LH over the southeastern TP. Enhanced WMC convection excites an anticyclone to the south of the TP, which transports moist air to the southeastern TP, favoring above-normal regional precipitation. Enhanced WMC convection induces the anticyclone to the south of the TP by causing significant descending motion over the northern BOB. The descending motion suppresses precipitation over the northern BOB, which further strengthens the anticyclone by convection–circulation feedback.

It should be noted that the atmospheric circulation induced by enhanced WMC convection just provides a favorable large-scale condition in the lower and midtroposphere for above-normal precipitation over the southeastern TP, which is also affected by a circulation anomaly in the upper troposphere excited by the NAO. The enhanced convection over the southeastern TP may also exert a positive feedback on precipitation by pumping more water vapor from the southern boundary since the latent heating over the southeastern TP could induce lower-tropospheric southwesterly flow to the southeast of the TP. Nevertheless, the variations of convection over the WMC and the precipitation over the southeastern TP are not in phase for some particular years. Thus, further studies are needed to understand how the convection over the WMC works together with other factors (e.g., spring sensible heat and extratropical atmospheric circulation) to affect the atmospheric heat source over the TP.

The TH-related atmospheric circulation patterns exhibit a strong subseasonal variation, especially from June to July. It is interesting that the TH-related cyclone centering over the northern South China Sea in June (Fig. 4a) exhibits different features from previous results that were based on the June–July–August mean data (e.g., Zhao and Chen 2001b) in which a summer enhanced atmospheric heat source over the TP was accompanied by southwesterly flow over southern China. In this study, we have just examined the atmospheric circulation patterns related to the July–August TH and LH over the TP. The annual cycle of SH and LH over the TP shows that the LH (SH) over the TP increases (decreases) rapidly in June, when transition of dry to rainy season usually occurs (figures not shown). In addition, SH is significantly and negatively correlated with LH over the TP (Table 2). Thus, the interannual variation of LH (SH) in June may be affected by a rainy season onset anomaly over the TP. Zhao et al. (2007) reported that spring snow cover in the TP exerts significant impact on TH in the following June on interannual time scale via its persistent effect on soil moisture. Therefore, the TH over the TP in June may be affected by both snow cover in previous spring and onset of rainy season. How these factors affect TH in June deserves further investigation.

It should be noted that the atmospheric heat source data are only 21 years, and just 3 stations are located in WTP. There may also be some biases in the data. These may affect the robustness of the conclusions.

Acknowledgments

We thank Prof. M. Watanabe for providing the LBM and the three anonymous reviewers for their constructive comments, which improved the overall quality of the paper. This study was jointly supported by the National Natural Science Foundation of China (Grants 91337107, 91337215, and 41375081), the National Basic Research Program of China (Grants 2012CB417202 and 2014CB953900), the Basic Research and Operation Program of the CMA Institute of Plateau Meteorology (BROP 201514), and Special Funds of Guangdong Province of China (“Thousand-Talent Plan” Fund YCJ2013-196).

REFERENCES

REFERENCES
Boos
,
W. R.
, and
Z. M.
Kuang
,
2010
:
Dominant control of the South Asian monsoon by orographic insulation versus plateau heating
.
Nature
,
463
,
218
223
, 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:.
Ding
,
Y.-H.
,
Y.
Sun
,
Z.-Y.
Wang
,
Y.-X.
Zhu
, and
Y.-F.
Song
,
2009
:
Inter-decadal variation of the summer precipitation in China and its association with decreasing Asian summer monsoons. Part II: Possible causes
.
Int. J. Climatol.
,
29
,
1926
1944
, doi:.
Duan
,
A.
, and
G.
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:.
Gill
,
A.
,
1980
:
Some simple solutions for heat-induced tropical circulation
.
Quart. J. Roy. Meteor. Soc.
,
106
,
447
462
, doi:.
Hsu
,
H.-H.
, and
X.
Liu
,
2003
:
Relationship between the Tibetan Plateau heating and East Asian summer monsoon rainfall
.
Geophys. Res. Lett.
,
30
,
2066
, doi:.
Hu
,
Z.
, and
T.
Nitta
,
1997
:
Seasonality of the interaction between convection over the western Pacific and general circulation in the Northern Hemisphere
.
Adv. Atmos. Sci.
,
14
,
539
553
, doi:.
Jiang
,
X.
,
S.
Yang
,
J.
Li
,
Y.
Li
,
H.
Hu
, and
Y.
Lian
,
2013a
:
Variability of the Indian Ocean SST and its possible impact on summer western North Pacific anticyclone in the NCEP Climate Forecast System
.
Climate Dyn.
,
41
,
2199
2212
, doi:.
Jiang
,
X.
,
S.
Yang
,
Y.
Li
,
A.
Kumar
,
X.
Liu
,
Z.
Zuo
, and
B.
Jha
,
2013b
:
Seasonal-to-interannual prediction of the Asian summer monsoon in the NCEP Climate Forecast System version 2
.
J. Climate
,
26
,
3708
3727
, doi:.
Jiang
,
X.
,
S.
Yang
,
Y.
Li
,
A.
Kumar
,
W.
Wang
, and
Z.
Gao
,
2013c
:
Dynamical prediction of the East Asian winter monsoon by the NCEP climate forecast system
.
J. Geophys. Res. Atmos.
,
118
,
1312
1328
, doi:.
Jiang
,
X.
,
Y.
Li
,
S.
Yang
,
J.
Shu
,
G.
He
,
2015
: Interannual variation of mid-summer heavy rainfall in the eastern edge of the Tibetan Plateau. Climate Dyn.,
45
,
3091
3102
, doi:.
Liu
,
H.
,
K.
Duan
,
M.
Li
,
P.
Shi
,
J.
Yang
,
X.
Zhang
, and
J.
Sun
,
2015
:
Impact of the North Atlantic Oscillation on the Dipole Oscillation of summer precipitation over the central and eastern Tibetan Plateau
.
Int. J. Climatol.
, doi:, in press.
Liu
,
X.
, and
Z.-Y.
Yin
,
2001
:
Spatial and temporal variation of summer precipitation over the eastern Tibetan Plateau and the North Atlantic Oscillation
.
J. Climate
,
14
,
2896
2909
, doi:.
Lu
,
R.
,
2004
:
Associations among the components of the East Asian summer monsoon system in the meridional direction
.
J. Meteor. Soc. Japan
,
82
,
155
165
, doi:.
Lu
,
R.
, and
Z.
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:.
Luo
,
H.
, and
M.
Yanai
,
1984
:
The large-scale circulation and heat sources over the Tibetan and surrounding areas during the early summer of 1979. Part II: Heat and moisture budgets
.
Mon. Wea. Rev.
,
112
,
966
989
, doi:.
Neale
,
R.
, and
J.
Slingo
,
2003
:
The Maritime Continent and its role in the global climate: A GCM study
.
J. Climate
,
16
,
834
848
, doi:.
Neale
,
R.
, and Coauthors
,
2010
: Description of the NCAR Community Atmosphere Model (CAM 5.0). NCAR Tech. Note NCAR/TN-486+STR, 274 pp. [Available online at http://www.cesm.ucar.edu/models/cesm1.0/cam/docs/description/cam5_desc.pdf.]
Nitta
,
T.
,
1987
:
Convective activities in the tropical western Pacific and their impact on the Northern Hemisphere summer circulation
.
J. Meteor. Soc. Japan
,
65
,
373
390
.
Nitta
,
T.
, and
Z.
Hu
,
1996
:
Summer climate variability in China and its association with 500 hPa height and tropical convection
.
J. Meteor. Soc. Japan
,
74
,
425
445
.
Qian
,
J. H.
,
2008
:
Why precipitation is mostly concentrated over islands in the Maritime Continent
.
J. Atmos. Sci.
,
65
,
1428
1441
, doi:.
Ramage
,
C. S.
,
1968
:
Role of a tropical “Maritime Continent” in the atmospheric circulation
.
Mon. Wea. Rev.
,
96
,
365
369
, doi:.
Sellers
,
P. J.
, and Coauthors
,
1996
:
A revised land surface parameterization (SiB2) for atmospheric GCMs. Part I: Model formulation
.
J. Climate
,
9
,
676
705
, doi:.
Smith
,
T. M.
,
R. W.
Reynolds
,
C. P.
Thomas
, and
J.
Lawrimore
,
2008
:
Improvements to NOAA’s historical merged land–ocean surface temperature analysis (1880–2006)
.
J. Climate
,
21
,
2283
2296
, doi:.
Stackhouse
,
P. W.
,
S. K.
Gupta
,
S. J.
Cox
,
J. C.
Mikovitz
,
T.
Zhang
, and
M.
Chiacchio
,
2004
: 12-year surface radiation budget data set. GEWEX News, No. 14, International GEWEX Project Office, Silver Spring, MD, 10–12.
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:.
Ueda
,
H.
,
H.
Kamahori
, and
N.
Yamazaki
,
2003
:
Seasonal contrasting features of heat and moisture budgets between the eastern and western Tibetan Plateau during the GAME IOP
.
J. Climate
,
16
,
2309
2324
, doi:.
Uppala
,
S. M.
, and Coauthors
,
2005
:
The ERA-40 Re-Analysis
.
Quart. J. Roy. Meteor. Soc.
,
131
,
2961
3012
, doi:.
Wang
,
B.
,
Z.
Wu
,
J.
Li
,
J.
Liu
,
C.
Chang
,
Y.
Ding
, and
G.
Wu
,
2008a
:
How to measure the strength of the East Asian summer monsoon?
J. Climate
,
21
,
4449
4463
, doi:.
Wang
,
B.
,
Q.
Bao
,
B.
Hoskins
,
G.
Wu
, and
Y.
Liu
,
2008b
:
Tibetan Plateau warming and precipitation changes in East Asia
.
Geophys. Res. Lett.
,
35
,
L14702
, doi:.
Wang
,
Z.
,
A.
Duan
, and
G.
Wu
,
2014
:
Time-lagged impact of spring sensible heat over the Tibetan Plateau on the summer rainfall anomaly in East China: Case studies using the WRF model
.
Climate Dyn.
,
42
,
2885
2898
, doi:.
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:; Corrigendum,
127
,
733
734
, doi:.
Wu
,
G.
, and
Y.
Zhang
,
1998
:
Tibetan Plateau forcing and timing of the monsoon onset over South Asia and the South China Sea
.
Mon. Wea. Rev.
,
126
,
913
927
, doi:.
Wu
,
G.
, and Coauthors
,
2007
:
The influence of mechanical and thermal forcing by the Tibetan Plateau on Asian climate
.
J. Hydrometeor.
,
8
,
770
789
, doi:.
Wu
,
G.
,
Y.
Liu
,
B.
He
, and Coauthors
,
2012
:
Thermal controls on the Asian summer monsoon
.
Sci. Rep.
,
2
, 404, doi:.
Wu
,
H.
,
K.
Yang
,
X.
Niu
, and
Y.
Chen
,
2015
:
The role of cloud height and warming in the decadal weakening of atmospheric heat source over the Tibetan Plateau
. Sci. China,
58D
, 395–403, doi:.
Wu
,
R.
,
Z.
Wen
,
S.
Yang
, and
Y.
Li
,
2010
:
An interdecadal change in southern China summer rainfall around 1992/93
.
J. Climate
,
23
,
2389
2403
, doi:.
Wu
,
Z.
,
Z.
Jiang
,
J.
Li
,
S.
Zhong
, and
L.
Wang
,
2012a
:
Possible association of the western Tibetan Plateau snow cover with the decadal to interdecadal variations of northern China heatwave frequency
.
Climate Dyn.
,
39
,
2393
2402
, doi:.
Wu
,
Z.
,
J.
Li
,
Z.
Jiang
, and
T.
Ma
,
2012b
:
Modulation of the Tibetan Plateau snow cover on the ENSO teleconnections: From the East Asian summer monsoon perspective
.
J. Climate
,
25
,
2481
2489
, doi:.
Xu
,
L.
,
H.
Gao
, and
Y.
Li
,
2009
:
Sensible heating over the Tibetan Plateau linked to the onset of Asian monsoon
.
Atmos. Ocean. Sci. Lett.
,
2
,
350
356
, doi:.
Xu
,
X.
,
C.
Lu
,
X.
Shi
, and
S.
Gao
,
2008
:
World water tower: An atmospheric perspective
.
Geophys. Res. Lett.
,
35
, L20815, doi:.
Xu
,
X.
,
C.
Lu
,
X.
Shi
, and
Y.
Ding
,
2010
:
Large-scale topography of China: A factor for the seasonal progression of the meiyu rainband?
J. Geophys. Res.
,
115
,
D02110
, 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
.
Yang
,
K.
,
Y.
Chen
, and
J.
Qin
,
2009
:
Some practical notes on the land surface modeling in the Tibetan Plateau
.
Hydrol. Earth Syst. Sci.
,
13
,
687
701
, doi:.
Yang
,
K.
,
X.
Guo
,
J.
He
,
J.
Qin
, and
T.
Koike
,
2011
:
On the climatology and trend of the atmospheric heat source over the Tibetan Plateau: An experiments-supported revisit
.
J. Climate
,
24
,
1525
1541
, doi:.
Ye
,
B.
,
D.
Yang
,
Y.
Ding
,
T.
Han
, and
T.
Koike
,
2004
:
A bias-corrected precipitation climatology for China
.
J. Hydrometeor.
,
5
,
1147
1160
, doi:.
Ye
,
D.
, and
Y.
Gao
,
1979
: The Meteorology of the Qinghai-Xizang (Tibet) Plateau (in Chinese). Science Press, 278 pp.
Zhang
,
T.
,
S.
Yang
,
X.
Jiang
, and
P.
Zhao
,
2015
: Seasonal–interannual variation and prediction of wet and dry season rainfall over the Maritime Continent: Roles of ENSO and monsoon circulation. J. Climate, doi:, in press.
Zhang
,
Y.
,
W. B.
Rossow
,
A. A.
Lacis
,
V.
Oinas
, and
M. I.
Mishchenko
,
2004
:
Calculation of radiative fluxes from the surface to top of atmosphere based on ISCCP and other global data sets: Refinements of the radiative transfermodel and the input data
.
J. Geophys. Res.
,
109
,
D19105
, doi:.
Zhao
,
P.
, and
L.
Chen
,
2001a
:
Climatic features of atmospheric heat source/sink over the Qinghai-Xizang Plateau in 35 years and its relation to rainfall in China
.
Sci. China
,
44D
,
858
864
, doi:.
Zhao
,
P.
, and
L.
Chen
,
2001b
:
Interannual variability of the atmospheric heat source/sink over the Qinghai–Xizang (Tibetan) Plateau and its relation to circulation
.
Adv. Atmos. Sci.
,
18
,
106
116
, doi:.
Zhao
,
P.
,
Z.
Zhou
, and
J.
Liu
,
2007
:
Variability of Tibetan spring snow and its associations with the hemispheric extratropical circulation and East Asian summer monsoon rainfall: An observational investigation
.
J. Climate
,
20
,
3942
3955
, doi:.