Excitation Mechanisms of the Teleconnection Patterns Affecting the July Precipitation in Northwest China

Guosen Chen Center of Monsoon System Research, Institute of Atmospheric Physics, Chinese Academy of Sciences, and Graduate University of Chinese Academy of Sciences, Beijing, China

Search for other papers by Guosen Chen in
Current site
Google Scholar
PubMed
Close
and
Ronghui Huang Center of Monsoon System Research, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China

Search for other papers by Ronghui Huang in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Using observational rainfall data and atmospheric reanalysis data, the precipitation variations in Northwest China during July and the corresponding atmospheric teleconnection patterns are studied. The results indicate that the leading modes of July precipitation variations in Northwest China are affected by the Silk Road pattern and the Europe–China (EC) pattern. The analysis suggests that the circumglobal teleconnection (CGT) could be considered as the interannual component of the Silk Road pattern.

To investigate the excitation mechanisms for the CGT pattern and EC pattern on interannual time scales, the singular value decomposition (SVD) analysis is performed between 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and tropical rainfall between (15°S and 30°N). The results suggest that the tropical heating anomalies most responsible for the CGT pattern are located over the North Indian Ocean, and the tropical heating anomalies most responsible for EC pattern are located over equatorial central Pacific, Indonesia, and tropical Atlantic. The tropical heating anomalies excite the CGT pattern and EC pattern by inducing divergent flow at the upper troposphere, and the advections of vorticity by the divergent component of the flow act as effective Rossby wave sources. Further analysis indicates that the tropical rainfall anomalies responsible for the CGT pattern and EC pattern are the leading modes of tropical rainfall variations, and these modes of tropical rainfall variations are related to the SST anomalies.

Corresponding author address: Guosen Chen, P. O. Box 2718, Bei-Er-Tiao 6, Zhong-Guan-Cun, Beijing 100190, China. E-mail: chgs@mail.iap.ac.cn

Abstract

Using observational rainfall data and atmospheric reanalysis data, the precipitation variations in Northwest China during July and the corresponding atmospheric teleconnection patterns are studied. The results indicate that the leading modes of July precipitation variations in Northwest China are affected by the Silk Road pattern and the Europe–China (EC) pattern. The analysis suggests that the circumglobal teleconnection (CGT) could be considered as the interannual component of the Silk Road pattern.

To investigate the excitation mechanisms for the CGT pattern and EC pattern on interannual time scales, the singular value decomposition (SVD) analysis is performed between 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and tropical rainfall between (15°S and 30°N). The results suggest that the tropical heating anomalies most responsible for the CGT pattern are located over the North Indian Ocean, and the tropical heating anomalies most responsible for EC pattern are located over equatorial central Pacific, Indonesia, and tropical Atlantic. The tropical heating anomalies excite the CGT pattern and EC pattern by inducing divergent flow at the upper troposphere, and the advections of vorticity by the divergent component of the flow act as effective Rossby wave sources. Further analysis indicates that the tropical rainfall anomalies responsible for the CGT pattern and EC pattern are the leading modes of tropical rainfall variations, and these modes of tropical rainfall variations are related to the SST anomalies.

Corresponding author address: Guosen Chen, P. O. Box 2718, Bei-Er-Tiao 6, Zhong-Guan-Cun, Beijing 100190, China. E-mail: chgs@mail.iap.ac.cn

1. Introduction

Northwest China is located in the center of the Eurasian continent, where it is far from a water source, and lies in the shadow of the Tibetan Plateau and Tianshan Mountain. These features render this area one of the most arid regions among the same latitude. However, there is significant interannual and interdecadal variability of summer precipitation in this region, although the annual precipitation is less than 100 mm in most of the region (Zhou and Huang 2008, 2010). Skillful seasonal prediction of the precipitation variations in this region would be of great value to mitigate the economic loss caused by the droughts.

The precipitation in Northwest China is concentrated in the summer season. The total rainfall in Northwest China is mainly contributed by the strong rain and the rainfall in strong intensity is increasing during the last 50 years (Chen and Dai 2009a). The studies of the spatial and temporal features of June–August (JJA) rainfall variations in Northwest China indicate a decreasing trend in eastern Northwest China but an increasing trend in western Northwest China (Li et al. 1997; Chen and Dai 2009b), which is consistent with the variations of water vapor amount in the atmosphere of Northwest China (Wang et al. 2004). Using observational weather station data, Zhou and Huang (2008) found that there are positive correlations between spring sensible heating in Northwest China and summer rainfall anomalies in western Northwest China, Northeast China, and Yangtze River basin, whereas negative correlations in eastern Northwest China. They suggested there might be some kind of atmospheric teleconnection between the monsoonal region of East China and the arid and semiarid region of Northwest China.

The key for skillful prediction of precipitation variations in arid and semiarid regions is to understand the controlling dynamics of these regions. Charney (1975) have investigated the effect of albedo to the growth of deserts. He proposed a well-known biosphere–albedo feedback mechanism that high albedo in subtropics could enhance the local Hadley circulation and leads to desert growth. Rodwell and Hoskins (1996) pointed out that a simple zonal-mean Hadley cell could not explain the existence of deserts because of the coexistence of deserts and monsoon regions around the same latitude and the relatively weak zonal mean vertical motion in subtropics during summer. They proposed a monsoon-desert mechanism that desertification could be forced by remote changes in monsoon strength through the decent motion to the west of monsoon heating. This mechanism could well explain the coexistence of deserts and monsoon areas in subtropics and it is theoretically based on the Sverdrup balance.

Although the monsoon–deserts mechanism could explain the interaction of monsoon and desert in the zonal direction in subtropics, it is less useful when it comes to explain the rainfall variations in the midlatitude arid and semiarid regions, such as Northwest China, which is due to the fact that the Sverdrup balance does not hold in midlatitudes and the advection effects could not be ignored. Consequently, there must be another mechanism to explain the summertime precipitation variations in Northwest China and the teleconnection between arid and semiarid area of Northwest China and monsoonal region of East China.

In the extratropics, the monthly atmospheric circulation is characterized by the stationary teleconnection patterns, and the local climate is largely influenced by these stationary teleconnection patterns. A teleconnection pattern induced by anomalous convective heating over the western North Pacific has been documented (Nitta 1987; Huang and Li 1988), which is known as Pacific–Japan (P–J) pattern or East Asia pattern (EAP). This teleconnection pattern has a great influence on climate of East Asian during boreal summer (Kosaka et al. 2011). Trenberth and Guillemot (1996) have studied the 1988 drought and 1993 floods in North America, and pointed out that the impacts of SST anomalies over Pacific on the North American summer circulation by forcing stationary waves. Considering the role of teleconnection patterns on the extratropical climate, it is reasonable to study the summer precipitation variations in Northwest China from the perspective of extratropical teleconnection patterns.

Based on one-point correlation analysis, Lu et al. (2002) identified the existence of a zonally oriented teleconnection pattern in July along the Asian jet at the upper troposphere over the Eurasian continent, which they suggested might have influence on the East Asian summer monsoon system. Later, Enomoto et al. (2003) suggested that the equivalent barotropic ridge near Japan in August is formed as a result of the propagation of quasi-stationary Rossby waves along the Asian jet in the upper troposphere, and they named it “the Silk Road pattern.” They have shown that the enhanced Asian jet in August is favorable for the propagation of quasi-stationary Rossby waves and that the regions of descent motions over the eastern Mediterranean Sea and the Caspian Sea induced by Indian monsoon heating act as two major wave sources. Sato and Takahashi (2006) have shown that the Silk Road pattern is self-strengthened through kinetic energy conversion near the entrance of the Asian jet over the Middle East. The analysis performed by Kosaka et al. (2009) suggests that the extraction of available potential energy from the baroclinic Asian jet is more efficient for its self-maintenance. Their analysis also indicates that the high sensitivity of its barotropic energy conversion to subtle zonal asymmetries of the Asian jet is a key factor for the geographically phase-locking of the wave pattern. Moreover, Ding and Wang (2005) found a boreal summer stationary circumglobal teleconnection (CGT) pattern over the Northern Hemisphere, which is also zonally oriented as the Silk Road pattern, but with circumglobal features. They suggested that CGT was related to the Indian summer monsoon. Using a nonlinear dry atmospheric model, Yasui and Watanabe (2010, hereafter YW10) pointed out that the origin of CGT lied in the internal dynamics, and they also suggested that the heating anomalies most responsible for the CGT-like steady response are located over the eastern Mediterranean region. However, using maximum covariance analysis, Ding et al. (2011, hereafter DWWB) suggested that the Indian monsoon heating anomalies are most responsible for the CGT, and they have also shown that CGT appeared preferentially in summers preceding the peak phase of ENSO cycle.

The above-mentioned studies have not reached an agreement on the origins of Silk Road pattern. In this paper, we aimed to study the excitation mechanisms of two teleconnection patterns in the upper troposphere over Eurasian continent, which lead to the precipitation variations of Northwest China in July, one of them is the Silk Road pattern. Particularly, we will propose a different excitation mechanism for the Silk Road pattern. This paper is organized as follows. Section 2 will describe the data and methods used in this study. The features of the spatial and temporal variations of precipitation in Northwest China during July will be presented in section 3. In section 4, we will investigate two teleconnection patterns over the Eurasian continent during July, which could affect the precipitation variations in Northwest China, and their impacts on extratropical precipitation variations. The excitation mechanisms for these two teleconnection patterns on interannual time scale will be discussed in section 5. Section 6 gives a conclusion and discussion.

2. Data and methods

The following datasets are used in this study: 1) monthly observational precipitation data at 160 observational stations, provided by the National Meteorological Information Center of China, this dataset is operated by the Chinese Meteorological Administration, and meets the standards of the World Meteorological Organization; 2) the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40) atmospheric reanalysis monthly data (Uppala et al. 2005); 3) the National Oceanic and Atmospheric Administration (NOAA) Extended Reconstructed Sea Surface Temperature (SST) version 3 (Smith et al. 2008); 4) the NOAA global monthly precipitation reconstruction (PREC) dataset, on a horizontal resolution of 2.5° × 2.5° (Chen et al. 2004); and 5) the NOAA precipitation reconstruction data over land (PREC/L), on a horizontal resolution of 0.5° × 0.5° (Chen et al. 2002). All the datasets used in this paper are chosen from the period of 1958–2002. The station precipitation data are used to investigate the spatial and temporal variations of rainfall in Northwest China. The PREC/L data are used to investigate the impacts of extratropical teleconnection patterns on rainfall variations over land. The PREC data are used to identify coupled modes between extratropical teleconnection patterns and tropical thermal forcing.

The main statistic tools used in this study include singular value decomposition (SVD) analysis, empirical orthogonal functions (EOFs) analysis, rotated empirical orthogonal functions (REOF) analysis, regression analysis, and correlation analysis. We applied the EOF analysis to the 200-hPa meridional velocity over the Eurasian continent to identify the teleconnection patterns that are affecting the rainfall variations in Northwest China during July. We also perform EOF analysis for tropical rainfall to identify the principal modes of tropical rainfall variations which are coupled with the extratropical teleconnection patterns. The method proposed by North et al. (North et al. 1982) is used to assess statistical significance. The REOF analysis is used to study the features of spatial and temporal of July rainfall variations in Northwest China to overcome the limitation of EOF analysis. To find coupled patterns between teleconnection patterns and tropical heating, the SVD analysis is also used in this study.

In this study, the wave activity flux formulated by Takaya and Nakamura (2001) is used to study the propagation features of stationary Rossby wave. The formulation of the wave activity fluxes is given in (1), where u is the zonal wind velocity, υ is the meridional wind velocity, and is the streamfunction. The overbars represent the basic states and primes represent perturbations. In this analysis, we use the 45 year monthly climatology means as the basic states and monthly anomalies derived from the basic states as the perturbations:
e1

3. Features of rainfall variations and the corresponding atmosphere circulations

In this section, the spatial and temporal features of July precipitation variations in Northwest China and the corresponding atmospheric circulations will be studied. In this study, Northwest China is referred to the region to the west of 110°E and north of 32.5°. Thus, 33 observational stations are chosen from the complete data archive. The distribution of stations is shown in Fig. 1 by filled circles. Since the EOF analysis tends to depend on the size and shape of the data domain, and as the rainfall dataset used in this study is heterogeneous in spatial distribution and small in size, the REOF analysis is used to overcome these deficiencies. The data has been normalized and the first seven EOFs were rotated.

Fig. 1.
Fig. 1.

Spatial modes of the first two REOFs based on analysis of 45-yr (1958–2002) July precipitation in Northwest China: (a) REOF1 and (b) REOF2 (by contours). The contour interval is 0.05. The total variance fractions explained by each mode are 14.4% in (a) and 10.7% in (b). The filled circles represent the geographical distribution of the stations.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

The first two REOF modes explain about 14.4% and 10.7% of the total variance, respectively, and can be isolated from other modes. The spatial patterns of first two REOF modes are shown in Fig. 1 (by contours). The first REOF pattern shows positive values in the eastern Northwest China and northern Xingjiang province in the western Northwest China and negative values in southwestern Xingjiang province. The second REOF pattern exhibits uniform signs in the entire domain. These two modes show consensus with those obtained by other researchers (Yang and Zhang 2008a; Chen and Dai 2009b). The corresponding principal components (PCs) and their interdecadal variations are shown in Fig. 2. The interdecadal variation is obtained by removing the interannual variations with a period shorter than 9 years using Fourier harmonic analysis. PC1 shows abnormal high positive anomalies during mid-1970s, while negative anomalies during 1980s and positive anomalies during 1990s with smaller amplitudes. PC2 shows asymmetry between the positive events and negative events, with less and stronger negative events.

Fig. 2.
Fig. 2.

Normalized time series of the first two REOF modes based on analysis of 45-yr (1958–2002) July precipitation in Northwest China and their interdecadal variations: (a) PC1 and (b) PC2. The solid lines represent the normalized time series, and the dash lines represent the corresponding interdecadal variations by removing the interannual variations with a period shorter than 9 years.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

Since the rainfall variations in Northwest China are largely affected by the midlatitude atmospheric circulation (Yang and Zhang 2008b; Chen and Dai 2009b), the 200-hPa circulation anomalies are then studied. The regressions of 200-hPa geopotential height to first two PCs of the REOF analysis are calculated and results are shown in Fig. 3. Both Fig. 3a and Fig. 3b are showing wave-like teleconnection patterns. The pattern shown in Fig. 3a depicts a quasi-zonal wavelike pattern along Asian jet stream, with positive anomalous centers over the West Asia and East China, and negative anomalous centers over Central Europe and Middle Asia, respectively. This pattern resembles the Silk Road pattern (Lu et al. 2002; Enomoto et al. 2003). It is suggested from Fig. 3a that the rainfall correlations between Northwest China and Yangtze River basin identified by Zhou and Huang (2008) might be due to this wave-like pattern. In fact, this pattern is responsible for the leading mode of rainfall variations in the entire Northern China during summer (Huang et al. 2011). The second pattern shown in Fig. 3b exhibits a wave-like pattern from Northwest Europe to Southeast China, with positive anomalous centers over Western Europe and Northwestern China and negative anomalous centers over Eastern Europe and Southeastern China, respectively. This pattern will be defined in the following section. As a result, the analysis above shows that the leading modes of rainfall variations in Northwest China during July are affected by the atmospheric teleconnection patterns. Thus, to study the causes of these teleconnection patterns will improve our understanding of the July rainfall variations in Northwest China. The possible excitation mechanisms of these teleconnection patterns will be studied latter in this study.

Fig. 3.
Fig. 3.

Regressions of 200-hPa geopotential height to (a) PC1 and (b) PC2 obtained by REOF analysis based on 45-yr (1958–2002) July precipitation in Northwest China.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

4. The extratropical teleconnection patterns

Given the above relations between the precipitation variations and atmospheric circulations, one is led to think whether these wave-like patterns are prominent during July. If this is the case, it would improve the predictability of the July rainfall variations in Northwest China.

a. Midlatitude teleconnection patterns

To answer this question, the EOF analysis is performed on 200-hPa meridional wind velocity (V200) over the region of (30°–60°N, 30°–130°E). The reason to use meridional wind velocity is because Rossby wave is transverse, for a given streamfunction amplitude, its meridional velocity is maximized if the wave is propagating in zonal direction (Held 1983). The variance fractions explained by the first two EOFs are 30.6% and 18.9%, respectively, and can be isolated from other modes. Since the midlatitude atmospheric motions satisfy the geostrophic balance, the regressions of 200-hPa geopotential height anomalies to corresponding PCs (shown in Fig. 4a and Fig. 4b) will represent the spatial patterns of the EOF analysis. It appears that the spatial patterns in Fig. 4 obtained by EOF analysis on V200 are similar to those shown in Fig. 3. The corresponding spatial correlation coefficients between Fig. 3 and Fig. 4 are 0.88 and −0.85 [between the region of (30°–60°N, 30°–130°E)], and the temporal correlation coefficients between the corresponding PCs are 0.46 and −0.44, respectively. Thus, it is indicated that the teleconnection patterns affecting the July rainfall variations in Northwest China are the leading modes of V200 over the Eurasian continent. The first two PCs of EOF analysis to V200 and their interdecadal variations are shown in Fig. 5. The PC1 exhibits significant interdecadal variations with an increasing trend before the mid-1970s and a decreasing trend after mid-1970s. There is a decreasing trend since 1980 in PC2.

Fig. 4.
Fig. 4.

Regressions of 200-hPa geopotential height (by contours) to (a) PC1 and (b) PC2 obtained by EOF analysis on 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E). The vectors are the regressions of 200-hPa wave activity fluxes to the corresponding PCs. The shading indicates the 95% confidence level for the geopotential height.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

Fig. 5.
Fig. 5.

Normalized time series of the first two EOF modes based on the analysis of 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E), and their interdecadal variations: (a) PC1 and (b) PC2. The solid lines represent the nomalized time series, and the dash lines represent the corresponding interdecadal variations by removing the interannual variations with a period shorter than 9 years.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

The teleconnection pattern depicted in Fig. 4a is the leading modes of V200 variation over the Eurasian continent, and this pattern is the so-called Silk Road pattern. It starts from the Mediterranean Sea and the Caspian Sea at the entrance of the Asian jet and propagates zonally along the core of the Asian jet stream, with three significant anomalous centers over central Europe, west Asia, and middle Asia, and decays at the exit of the Asian jet. The regression of wave activity fluxes to PC1 (shown in Fig. 4a, by vectors) indicates that Silk Road pattern results from the propagation of stationary Rossby wave along the Asian jet, which is showing consensus with other researchers (Enomoto et al. 2003; Sato and Takahashi 2006; Kosaka et al. 2009).

The second EOF pattern shown in Fig. 4b exhibits a wave-like pattern from North Atlantic to China through an arc path over the Eurasian continent. There are three anomalous centers with significant amplitudes over the North Atlantic, western Europe, and eastern Europe, and other centers with weaker amplitudes over eastern North America, middle Asia, and South China. There is a dipole structure over North Atlantic with a stronger positive anomalous center to the north, which is analogous to blocking pattern. We define this pattern as Europe–China (EC) pattern. The associated wave activity fluxes (shown in Fig. 4b) also demonstrate that the EC pattern is the result of propagation of stationary Rossby wave, just as the Silk Road pattern. The only difference is that the EC pattern propagates through an arc path described by Hoskins and Karoly (1981), whereas the Silk Road pattern is trapped by the Asian Jet stream, which is a strong waveguide (Hoskins and Ambrizzi 1993). It can be seen that the Rossby wave is absorbed and reflected at the exit of the Asian Jet over East China, which might be due to the absorption and reflection effects of the critical line (Tung 1979; Held 1983; Held et al. 2002).

To study the vertical structures of the Silk Road pattern and EC pattern, the height–longitude cross section of regressions of geopotential height to PC1 and PC2 averaged between 30.0° and 50.0°N are calculated, and the results are shown in Fig. 6. It is clearly seen that vertical structures of Silk Road pattern and EC pattern are equivalent barotropic. The vertical structure of the Silk Road pattern and EC pattern are consistent with the theory of stationary external Rossby waves (Held et al. 1985). As a result, the Silk Road pattern and EC pattern are the results of propagations of stationary external Rossby waves. From this point of view, the dynamics of the Silk Road pattern and EC pattern could be understood qualitatively by using equivalent barotropic models.

Fig. 6.
Fig. 6.

Height–longitude cross section of geopotential height averaged between 30.0° and 50.0°N regressed onto (a) PC1 and (b) PC2 of EOF analysis based on the 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E).

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

To discuss the wave source regions of the Silk Road pattern and EC pattern, it is helpful to show the low-level circulations corresponding to Fig. 4. The 700-hPa circulations corresponding to Fig. 4 are shown in Fig. 7. As discussed by Held, the upper troposphere supports linear waves better than does the lower troposphere. Thus, the linear waves are able to propagate further away from the source region in upper troposphere than in lower troposphere. As a result, we would expect wave breaking and mixing in the lower level, except for the wave source region. In Fig. 7a, there is only one significant anomalous center over the eastern Mediterranean region and the Caspian Sea corresponding to the upper-level circulations. Thus, it is indicated that wave source region of Silk Road pattern is located over the Caspian Sea and the Mediterranean Sea. The case for the EC pattern is a little complicated. In Fig. 7b, there are three anomalous centers over North Atlantic, western Europe, and eastern Europe, corresponding to the anomalous wave centers in upper troposphere. Therefore, we would expect the source regions of EC are located in these regions. However, note that there is a significant divergence region of wave activity fluxes over Eastern Europe (between the center over Western Europe and the center over Eastern Europe) in Fig. 4b, for external stationary Rossby wave, this divergence region of wave activity flux indicates nonconservative source, and thus the Eastern Europe is at least one of the source regions for the EC pattern. As for the wave action center over the North Atlantic, the wave activity fluxes are weak compared to the region of Eastern Europe, so the signal of stationary wave propagation is not significant. For the source region of stationary Rossby wave, we would expect strong divergence of the wave activity fluxes, and which is not the case for the anomalous center over North Atlantic. Besides, in section 5, we will see that the anomalous center over North Atlantic may just reflect some interdecadal variations, and therefore would not be considered as the origin of the EC pattern. As a result, the analysis suggests that the origin of EC lies in the Eastern Europe.

Fig. 7.
Fig. 7.

Regressions of 700-hPa geopotential height to (a) PC1 and (b) PC2 obtained by EOF analysis on 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E).

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

In this section, we have used monthly data to identify teleconnection patterns. As discussed by Ding and Wang (2007), there may be strong contribution from intraseasonal variability in the monthly data, and there is evidence that the leading mode of 200-hPa geopotential height exists on an intraseasonal time scale. However, the analysis of daily data shows (not shown) that the teleconnection patterns in Fig. 4 are independent from each other and do not reflect some intraseasonal variability.

b. Impacts of the teleconnection patterns on rainfall anomalies

First, their impacts on rainfall in Northwest China will be examined. The correlation coefficients between the PCs of EOF analysis on monthly V200 and the observational precipitation in Northwest China are calculated, and the results are shown in Fig. 8. The precipitation anomalies in Fig. 8a depict an east–west pattern, which is consistent with Fig. 1a. The spatial distribution of rainfall anomalies in Fig. 8b is different from Fig. 1b. However, Fig. 8b exhibits uniform signs in the region of Northwest China as does in Fig. 1b, and thus is consistent with Fig. 1b from this perspective of view. Therefore, it is indicated that the result of REOF analysis and regression analysis in section 3 is robust. To interpret the results in Fig. 8, we resort to the Omega equation. As discussed by Holton (2004), the traditional Omega equation could be approximated by Eq. (2), where is the vertical motion in pressure coordinate, is the geostrophic wind velocity vector, and is the geopotential. Equation (2) can be further approximated by (3), where w is the vertical motion in Cartesian coordinate, and is the relative vorticity calculated by geostrophic wind. From the right-hand side of Eq. (3), the vertical motion is forced by the advection of absolute vorticity by thermal wind. In troposphere, the wind velocity is increasing. For the stationary external Rossby wave, the relative vorticity maximum gradient lags relative vorticity maximum about a quarter wavelengths, thus the relative vorticity advection is a maximum ahead of the trough. Therefore, for a short wave system where the advection of relative vorticity is larger than the planetary vorticity advection, there will be ascending motion ahead of the trough and descending motion ahead of the ridge. To better understand the dynamics, the corresponding regressions of 200-hPa geopotential height are also shown in Fig. 8, from which it is indicated that Omega equation could well explain how extratropical teleconnection patterns influence precipitation variations in Northwest China:
e2
e3
Fig. 8.
Fig. 8.

Correlation coefficients between (a) PC1 and (b) PC2 of EOF analysis on 200-hPa meridional wind velocity and rainfall anomalies of 33 observational stations in Northwest China (by shadings), and the regressions of 200-hPa geopotential height to the corresponding PCs (by contours). The contour intervals are 10 m in (a) and 5 m in (b).

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

To reveal the impacts exerted by the Silk Road pattern and EC pattern on extratropical rainfall anomalies over Eurasian continent, the correlation coefficients between the PREC/L precipitation and first two PCs of EOF analysis on monthly V200 have been calculated, and the results are shown in Fig. 9. The anomalous rainfall pattern associated with Silk Road pattern exhibits quasi-zonal wave-like distribution from the Mediterranean Sea to East Asia along the Asian jet stream with alternate positive and negative centers. The correlation pattern associated with EC pattern reveals northwest–southeast wavelike distribution from eastern Europe to southeast China. The corresponding regressions of 200-hPa geopotential height are also shown in Fig. 9. It can be inferred from Fig. 9 that the results can be well explained by the Omega equation. There is no surprise to see the good relation between the precipitation and the two teleconnection patterns identified in this section since they are the leading two modes of V200 over the Eurasian continent, and thus will make contributions to the extratropical climate. The quasi-zonal feature of the rainfall pattern shown in Fig. 9a and its relation to the Silk Road pattern suggest that the monsoon–deserts connection between Northwest China and East China proposed by Zhou and Huang (2008) might result from the Silk Road teleconnection pattern and do not result from the monsoon–desert mechanism proposed by Rodwell and Hoskins (Rodwell and Hoskins 1996).

Fig. 9.
Fig. 9.

Correlation coefficients between (a) PC1 and (b) PC2 of EOF analysis on July V200 and land precipitation (PREC/L) anomalies (by shadings), and the regressions of 200-hPa geopotential height to the corresponding PCs (by contours).

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

5. Excitation mechanisms for the extratropical teleconnection patterns

We have discussed the Silk Road pattern and EC pattern and their influences on extratropical rainfall variations over the Eurasian continent. In this section we will discuss the excitation mechanisms of these teleconnection patterns. As can be seen in Fig. 5a, the time series of Silk Road pattern reveals significant signals of interdecadal variability. Considering there are a variety of factors affecting the interdecadal variations of atmosphere, thus in this section, we will focus our attention on the interannual variability of Silk Road pattern and EC pattern and their excitation mechanisms. Hence, the datasets used in this section have been filtered by using Fourier harmonic analysis, the long-term trend and decadal variations with a period longer than 9 years are removed.

a. Interannual variability

Before discussing the excitation mechanisms, the interannual variability of the Silk Road pattern and EC pattern will be examined. The EOF analysis is performed on the filtered V200 over the same region as does in section 4. The variance fractions explained by the first two EOFs are 25.4% and 20.0%, respectively, and can be isolated from other modes. The variance fraction explained by the first EOF mode decreases and that explained by the second EOF mode increases, comparing to the EOF analysis in section 4. The combined variance fraction explained by first two EOF modes remains almost invariant. Thus, on the interannual time scale, the Silk Road pattern appears to be relatively less prominent and the EC pattern appears to be relatively more prominent.

The regressions of 200-hPa geopotential height to the first two PCs of EOF analysis on filtered V200 are shown in Fig. 10a and Fig. 10b, respectively. The first pattern shown in Fig. 10a exhibits circumglobal features and resembles the CGT pattern. The anomalous centers over the Mediterranean region to middle Asia shown in Fig. 10a are almost identical to those shown in Fig. 4a, except for the weaker amplitudes. The circumglobal features and the enhancing of anomalous centers at the exit of the Asian jet are the major differences between filtered and unfiltered patterns. These differences indicate the interdecadal variations of EOF1 are not just manifested themselves in the time series (Fig. 5a), but also in spatial distributions of the teleconnection patterns. The correlation coefficient between PC1 of EOF analysis in this section and CGTI (defined by DW) is 0.75. Therefore, it is suggested that CGT could be considered as the interannual component of the Silk Road pattern. Hereafter, we refer it to CGT when discussing the interannual variations of the Silk Road pattern. The second mode shown in Fig. 10b is almost identical to the pattern shown in Fig. 4b, except that the anomalous dipole structure over the North Atlantic is not significant, which indicates that the anomalous center over North Atlantic may just reflects some interdecadal variations. The associated wave-activity fluxes shown in Fig. 10 indicate that CGT and EC are the results of propagations of stationary Rossby waves.

Fig. 10.
Fig. 10.

As in Fig. 4, but for data on interannual time scale.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

b. Excitation mechanisms

As shown in Fig. 9a, the rainfall anomalies over the Middle Asia and western Northwest China are in opposite signs with the rainfall anomalies over India, which agrees with the results of Yang et al. (2009). Since the Silk Road pattern does not appear to affect the rainfall over India directly, it is suggested that the Indian monsoon heating might be responsible for the excitation of Silk Road pattern (Enomoto et al. 2003; Ding and Wang 2005; DWWB). However, Yasui and Watanabe (2010) suggested that the heating anomalies most responsible for CGT are located over the eastern Mediterranean region, and the monsoon heating is not significantly related to the CGT. One possible reason for discrepancy between DWWB and YW10 is that the different forcing domain they are chosen. In DWWB, the authors considered CGT as extratropical response to tropical forcing, while in YW10, the extratropical heating including transient eddy fluxes were also taken into considerations. As shown in Fig. 9a, the Silk Road pattern and EC pattern could affect local precipitation by inducing vertical motions, and thus the latent heat released by precipitation could be coupled with these teleconnection patterns. Besides, the extratropical heating field can itself be influenced by tropical heating, orography, and the effects of transient eddy fluxes may be secondary (Held et al. 2002). Therefore, it is reasonable to consider CGT and EC as the responses to tropical forcing.

To reveal the relation between tropical thermal forcing and extratropical teleconnection patterns, the SVD analysis is performed between V200 over the region of (30°–60°N, 30°–130°E) and PREC precipitation between tropical region of (15°S–30°N). The main difference between DWWB and the analysis in this analysis is that we use meridional wind velocity instead of geopotential height, so the influences of ENSO may be reduced (YW10). Besides, we choose regional domain for the meridional wind velocity instead of the whole Northern Hemisphere. Another difference is that we study July as a case instead of the whole boreal summer, which turns out to be crucial to the results (as will be discussed in section 6). We use V200 as the left field and the tropical precipitation as the right field, and denote the expansion coefficients of the leading mode to the left field as U1, and V1 for the expansion coefficients of the leading mode to the right field, the rest can be done in the same manner.

The first two SVD modes (denoted by S1 and S2) explain about 37.4% and 21% of the total covariance between the two fields. The correlation coefficients between the associated expansion coefficient time series are 0.71 and 0.65, which are inferring good correlations between the couple modes. The homogenous regressions of 200-hPa geopotential height to the corresponding expansion coefficients are shown in Fig. 11 (by contours). The first pattern shown in Fig. 11a resembles the CGT pattern and the second pattern shown in Fig. 11b resembles the EC pattern. The spatial correlation coefficients between Fig. 11 and Fig. 10 are 0.94 and 0.92 [between the region of (30°–60°N, 30°–130°E)], and the temporal correlation coefficients between the corresponding time series are 0.93 and 0.82. Thus, it is indicated that the first two SVD modes are the CGT pattern and EC pattern, and the corresponding tropical thermal forcing.

Fig. 11.
Fig. 11.

Leading two SVD modes based on the analysis of July 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and PREC tropical rainfall between (15°S and 30°N) during 1958–2002: (a) S1 and (b) S2. Contours are the homogeneous regressions of 200-hPa geopotential height to the corresponding expansion coefficients, and shadings are the heterogeneous correlation coefficients between tropical rainfall and the corresponding expansion coefficients.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

To reveal the tropical heating anomalies that are responsible for CGT and EC, the heterogeneous correlation coefficients between the PREC tropical rainfall and the corresponding expansion coefficients are calculated, and the results are shown in Fig. 11 (by shadings). The anomalous tropical rainfall pattern corresponding to CGT pattern (Fig. 11a) exhibits a significant positive anomalous center over the North Indian Ocean, and anomalies over equatorial Pacific with weaker correlation coefficients. The anomalous tropical rainfall pattern corresponding to EC pattern (Fig. 11b) depicts a negative anomalous center over the equatorial central Pacific and positive anomalous centers over the equatorial Atlantic and Indonesia. Since the heterogeneous correlation coefficients of the SVD analysis indicate the key regions of the one field to the other, it is indicated that the tropical heating anomalies responsible for the CGT pattern are located over the North Indian Ocean and equatorial Pacific, while the tropical heating anomalies responsible for the EC pattern are located over the equatorial central Pacific, Indonesia and equatorial Atlantic.

The SVD analysis indicates that CGT pattern and EC pattern are closely related to the tropical heating anomalies. The question is, how do these tropical heating anomalies excite the extratropical teleconnection patterns? As discussed by Held et al. (Held et al. 2002), in tropics, the diabatic heating is mainly balance by the adiabatic cooling. The divergent flow associated with the vertical motion induced by heating can be thought to determine the rotational component of the flow by the stretching term . One might think that the divergence induced by the tropical heating becomes weaker in midlatitudes, so the effect of vorticity generation by the divergence would not be significant in midlatitudes. However, as the latitude is increased, the planetary vorticity is also increased, thus the stretching term increases in magnitude, the effect of the stretching term would still be significant as it is in the tropical heating region. Besides, the advection of vorticity by the divergent component of the flow (, where is the divergent wind velocity) can also be considered as the Rossby wave sources (Sardeshmukh and Hoskins 1988), which would displace the wave sources further poleward.

To verify the theory given above, the statistical analysis is performed. First, to reveal the divergent circulations induced by the tropical heating, the correlation coefficients between the expansion coefficients to tropical rainfall of SVD analysis and 200-hPa velocity potential (by contours), divergent flow (by vectors) are calculated, and the results are shown in Fig. 12. The homogeneous correlation between PREC tropical precipitation and the corresponding expansion coefficients of SVD analysis are also shown in Fig. 12. Second, to better understand the dynamics, the correlation coefficients between 200-hPa Rossby wave sources (defined as , by Sardeshmukh and Hoskins 1988) and the expansion coefficients to V200 of SVD analysis are calculated, and the results are shown in Fig. 13. For the sake of discussion, the teleconnection patterns in Fig. 11 and divergent circulations in Fig. 12 are also shown in Fig. 13.

Fig. 12.
Fig. 12.

Correlation coefficients between 200-hPa velocity potential (by contours), divergent flow (by vectors) and (a) V1 and (b) V2 of SVD analysis based on July 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and PREC tropical rainfall between (15°S and 30°N) during 1958–2002. The homogeneous correlation coefficients PREC tropical rainfall and the corresponding expansion coefficients are also shown by shadings. The contour interval is 0.1. The contour levels above 95% confidence level are plotted.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

Fig. 13.
Fig. 13.

Correlation coefficients of Rossby wave sources (defined as , by Sardeshmukh and Hoskins, 1988) and (a) U1 and (b) U2 of SVD analysis (by shadings), and homogeneous regressions of 200-hPa geopotential height on to the corresponding expansion coefficients (by colorful contours), and correlation coefficients between 200-hPa velocity potential (by black contours), divergent flow (by vectors) and (a) V1 and (b) V2 of SVD analysis. The contour interval for correlations of velocity potential is 0.1. The contour levels of velocity potential above 95% confidence level are plotted.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

It is seen from Fig. 12a that the latent heat released by the anomalous rainfall over the North Indian Ocean leads to divergent circulation at the upper troposphere. The Rossby wave sources over the Caspian Sea and Mediterranean Sea (shown in Fig. 13a, by shadings) indicate the role of advection of vorticity by divergent flow in exiting the CGT pattern. The 200-hPa velocity potential pattern (shown in Fig. 12b, by contours) corresponding for EC pattern exhibits an east–west seesaw structure with positive correlations over the east of 180° and negative correlations over the west of 180°. This velocity potential pattern is caused by the anomalous tropical heating (shown in Fig. 12b, by shadings). Because of this kind of configuration of velocity potential, the advection of vorticity by the divergent flow over the Eastern European Plain acts as effective Rossby wave sources for the EC pattern, which can be inferred from the correlation coefficients between the Rossby wave sources and U2 (as shown in Fig. 13b, by shadings). Therefore, the analysis has verified the theory proposed above. As a result, the anomalous tropical forcing could excite the CGT pattern and EC pattern by inducing upper level divergent flow, and the advections of vorticity by the divergent component of the flow place the wave sources further poleward.

Since the CGT pattern and EC pattern are the leading modes of V200 variations over the Eurasian continent, it is reasonable to assume that the tropical heating anomalies forcing these teleconnection patterns are also prominent. To verify this hypothesis, we perform EOF analysis on precipitation over the tropical region of (15°S–30°N). The data have been normalized and the spatial patterns are shown in Fig. 14. The total variance fractions explained by each mode are 21.5% and 13.8%, respectively. The leading tropical rainfall mode resembles the second SVD tropical rainfall mode and the correlation coefficient between the corresponding time series is 0.86. The second tropical rainfall mode resembles the first SVD tropical rainfall mode, and the correlation coefficient between the corresponding time series is 0.63. As a result, it is indicated that the CGT pattern and EC pattern are excited by the leading modes of tropical thermal forcing, which would improve the predictability of the CGT pattern and EC pattern, and thus the predictability of rainfall variations in Northwest China during July.

Fig. 14.
Fig. 14.

Spatial modes of first two EOFs based on analysis of PREC tropical rainfall between 15°S and 30°N: (a) EOF1 and (b) EOF2. The total variance fractions explained by each mode are 21.5% in (a) and 13.8% in (b).

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

As the year-to-year tropical heating anomalies are largely controlled by the SST anomalies, it is worth it to examine whether these anomalous tropical heating patterns are related to the SST anomalies. The correlation coefficients between the first two PCs of EOF on tropical rainfall and SST anomalies are calculated, and the results are shown in Fig. 15. The correlation patterns shown in Fig. 15 suggest that the tropical heating anomalies responsible for CGT pattern and EC pattern are related to the SST anomalies. The SST anomalies in North Indian Ocean are responsible for the rainfall anomalies over the region (Fig. 15b), and are thus responsible for exciting CGT pattern. The SST anomalies associated with EC pattern appear to be related to the ENSO cycles (Fig. 15a). The study of the causes of these SST anomalies, especially the role of ENSO cycles, is beyond the scope of this study.

Fig. 15.
Fig. 15.

Correlation coefficients between SST anomalies and (a) PC1 and (b) PC2 of EOF analysis on PREC tropical rainfall. The contour interval is 0.1. The shading indicates the 95% confidence level.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

Although there are three anomalous rainfall centers in Fig. 11a, in the authors’ opinion, the tropical heating anomalies most responsible for the CGT pattern are located over the North Indian Ocean. To prove this, first look at the heating anomalies over equatorial western Pacific, it does not force significant velocity potential anomalies (Fig. 12a), thus it could not force CGT directly. Besides, there are no significant SST anomalies in equatorial western Pacific (Fig. 15b), which suggests that the heating anomalies over equatorial western Pacific may be causing by the heating anomalies over the North Indian Ocean through divergence flow. As for the heating anomalies over equatorial eastern Pacific, the velocity potential (Fig. 12a) also suggests that it would not force CGT directly. Therefore, it is suggested that the heating anomalies most responsible for CGT are located over the North Indian Ocean.

6. Conclusions and discussion

In this paper, we have studied the features of the precipitation variation in Northwest China during July, and the corresponding anomalous atmospheric circulations. The results show that the Silk Road pattern and EC pattern are responsible for the leading modes of precipitation variations in Northwest China during July. The analysis suggests that the CGT pattern could be considered as the interannual component of the Silk Road pattern. The SVD analysis is used to investigate the excitation mechanisms for the CGT pattern and EC pattern on interannual time scale. The results suggest that the CGT pattern and EC pattern are extratropical responses to the tropical heating anomalies. The tropical heating anomalies most responsible for the CGT pattern are located over the North Indian Ocean, and the tropical heating anomalies most responsible for EC pattern are located over equatorial central Pacific, Indonesia, and tropical Atlantic. The tropical heating anomalies excite the CGT and EC by inducing divergent flow at the upper troposphere, and the advections of vorticity by the divergent component of the flow act as effective Rossby wave sources. Further analysis shows that the tropical rainfall anomalies responsible for the CGT pattern and EC pattern are the leading modes of tropical rainfall variations, and these modes of anomalous tropical forcing are related to the SST anomalies, which would improve the predictability of CGT pattern and EC pattern. As a result, the SST anomalies are affecting the leading modes of rainfall variations in Northwest China during July by forcing extratropical teleconnection patterns, and thus make it possible to improve the seasonal prediction of July rainfall variability in Northwest China.

The excitation mechanism of the CGT (or the Silk Road) pattern proposed in this study is different from DWWB, where they used seasonal mean data and emphasized the role of Indian monsoon heating in the excitation of the CGT pattern. To explain the discrepancy, we perform SVD analysis on data of June, August, and the summer season (JJA) separately, and the results are shown in Fig. 16. Both the leading SVD mode of June and August are the CGT pattern, but the related tropical heating anomalies are different. The anomalous tropical rainfall pattern of June in SVD analysis (shown in Fig. 16a) suggests that the descent motion over the Mediterranean region induced by the Indian summer monsoon heating accounts for the excitation of CGT pattern, which agrees with other researchers (Enomoto et al. 2003; Ding and Wang 2005; DWWB). The anomalous tropical rainfall pattern of August (shown in Fig. 16b) is similar to July, which indicates the heating anomalies over the North Indian Ocean accounts for the excitation of CGT pattern. The percentages of the total covariance explain by the fist SVD modes of June, July, and August are 46.9%, 37.4%, and 32.2%, respectively. Therefore, it might explain why features of leading SVD mode on seasonal mean data (shown in Fig. 16c, which explain 35.9% of the total covariance) have more features of June, since the covariance fraction explain by June is larger than all other summer month. As a result, even though the leading mode of precipitation variation in Northwest China is affected by the Silk Road pattern (or CGT) during summer (results not shown), the tropical heating responsible for the Silk Road pattern (or CGT) is different in different summer month. Therefore, the successful prediction of precipitation variation in Northwest China during summer lies in the complete understanding of the tropical heating responsible for the Silk Road pattern (or CGT) in different summer months, especially the relation between these heating anomalies during different summer months.

Fig. 16.
Fig. 16.

Leading mode of SVD analysis based on 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and PREC tropical rainfall between 15°S and 30°N during 1958–2002: (a) June, (b) August, and (c) summer. Contours are the homogeneous regressions of 200-hPa geopotential height to the corresponding expansion coefficients, and shadings are the heterogeneous correlation coefficients between tropical rainfall and the corresponding expansion coefficients.

Citation: Journal of Climate 25, 22; 10.1175/JCLI-D-11-00684.1

In this study, we have focused our attention on the role of tropical heating anomalies in exciting the extratropical teleconnection patterns. However, the causes of low-frequency variability of atmosphere are diverse, such as the forcing by high-frequency transients. Many studies have shown that the internal dynamics might play an important role in shaping and maintaining the Silk Road (or CGT) pattern (Sato and Takahashi 2006; Kosaka et al. 2009; YW10; DWWB). In fact, the eddy sensible heat flux convergence may be considered as a part of heating in extratropics (Held et al. 2002), and much of the low-frequency variability is associated with disturbances which derive their energy from the basic state through barotropic instability (Simmons et al. 1983). However, the internal dynamics of CGT is not jet fully understood, especially the effects of transient eddies. Thus, it appears necessary to study the effects of transient eddies to the stationary teleconnection patterns. The study of the effects of transients is currently under way and will be reported soon.

Acknowledgments

This study was supported by the National Basic Research Program of China (Grant 2010CB950403), the National Nature Science Foundation of China (Grant 41175055 and 40905027), and the Special Scientific Research Project for Public Interest (Grant GYHY2D1006021).

REFERENCES

  • Charney, J. G., 1975: Dynamics of deserts and drought in the Sahel. Quart. J. Roy. Meteor. Soc., 101, 193202.

  • Chen, D., and Y. Dai, 2009a: Characteristics of Northwest China rainfall intensity in recent 50 years. Chin. J. Atmos. Sci., 33, 923935.

    • Search Google Scholar
    • Export Citation
  • Chen, D., and Y. Dai, 2009b: Characteristics and analysis of typical anomalous summer rainfall patterns in northwest China over the last 50 years. Chin. J. Atmos. Sci., 33, 12471258.

    • Search Google Scholar
    • Export Citation
  • Chen, M., P. Xie, J. E. Janowiak, and P. A. Arkin, 2002: Global land precipitation: A 50-yr monthly analysis based on gauge observations. J. Hydrometeor., 3, 249266.

    • Search Google Scholar
    • Export Citation
  • Chen, M., P. Xie, J. E. Janowiak, P. A. Arkin, and T. M. Smith, 2004: Verifying the reanalysis and climate models outputs using a 56-year data set of reconstructed global precipitation. Preprints, 14th Conf. on Applied Climatology, Seattle, WA, Amer. Meteor. Soc., J6. 1. [Available online at https://ams.confex.com/ams/84Annual/techprogram/paper_70083.htm.]

  • Ding, Q., and B. Wang, 2005: Circumglobal teleconnection in the Northern Hemisphere summer. J. Climate, 18, 34833505.

  • Ding, Q., and B. Wang, 2007: Intraseasonal teleconnection between the summer Eurasian wave train and the Indian Monsoon. J. Climate, 20, 37513767.

    • Search Google Scholar
    • Export Citation
  • Ding, Q., B. Wang, J. M. Wallace, and G. Branstator, 2011: Tropical–extratropical teleconnections in boreal summer: Observed interannual variability. J. Climate, 24, 18781896.

    • Search Google Scholar
    • Export Citation
  • Enomoto, T., B. J. Hoskins, and Y. Matsuda, 2003: The formation mechanism of the Bonin high in August. Quart. J. Roy. Meteor. Soc., 129, 157178.

    • Search Google Scholar
    • Export Citation
  • Held, I. M., 1983: Stationary and quasi-stationary eddies in the extratropical troposphere: Theory. Large-Scale Dynamical Processes in The Atmosphere, B. J. Hoskins and R. P. Pearce, Eds., Academic Press, 127–168.

  • Held, I. M., R. T. Pierrehumbert, and R. L. Panetta, 1985: Stationary external Rossby waves in vertical shear. J. Atmos. Sci., 42, 865883.

    • Search Google Scholar
    • Export Citation
  • Held, I. M., M. Ting, and H. Wang, 2002: Northern winter stationary waves: Theory and modeling. J. Climate, 15, 21252144.

  • Holton, J. R., 2004: An Introduction to Dynamic Meteorology. Academic Press, 529 pp.

  • Hoskins, B. J., and D. J. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38, 11791196.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci., 50, 16611661.

    • Search Google Scholar
    • Export Citation
  • Huang, G., Y. Liu, and R. Huang, 2011: The interannual variability of summer rainfall in the arid and semiarid regions of Northern China and its association with the northern hemisphere circumglobal teleconnection. Adv. Atmos. Sci., 28, 257268.

    • Search Google Scholar
    • Export Citation
  • Huang, R., and W. Li, 1988: Influence of heat source anomaly over the western tropical Pacific on the subtropical high over East Asia and its physical mechanism. Chin. J. Atmos. Sci., 12, 107116.

    • Search Google Scholar
    • Export Citation
  • Kosaka, Y., H. Nakamura, M. Watanabe, and M. Kimoto, 2009: Analysis on the dynamics of a wave-like teleconnection pattern along the summertime Asian jet based on a reanalysis dataset and climate model simulations. J. Meteor. Soc. Japan, 87, 561580.

    • Search Google Scholar
    • Export Citation
  • Kosaka, Y., S.-P. Xie, and H. Nakamura, 2011: Dynamics of interannual variability in summer precipitation over East Asia. J. Climate, 24, 54355453.

    • Search Google Scholar
    • Export Citation
  • Li, D., J. Xie, and W. Wang, 1997: A study of summer precipitation features and anomaly in Northwest China. Chin. J. Atmos. Sci., 21, 331340.

    • Search Google Scholar
    • Export Citation
  • Lu, R.-Y., J.-H. Oh, and B.-J. Kim, 2002: A teleconnection pattern in upper-level meridional wind over the North African and Eurasian continent in summer. Tellus, 54A, 4455.

    • Search Google Scholar
    • Export Citation
  • Nitta, T., 1987: Convective activities in the tropical western Pacific and their impact on the Northern Hemisphere summer circulation. J. Meteor. Soc. Japan, 65, 373390.

    • Search Google Scholar
    • Export Citation
  • North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110, 699706.

    • Search Google Scholar
    • Export Citation
  • Rodwell, M. J., and B. J. Hoskins, 1996: Monsoons and the dynamics of deserts. Quart. J. Roy. Meteor. Soc., 122, 13851404.

  • Sardeshmukh, P. D., and B. J. Hoskins, 1988: The generation of global rotational flow by steady idealized tropical divergence. J. Atmos. Sci., 45, 12281251.

    • Search Google Scholar
    • Export Citation
  • Sato, N., and M. Takahashi, 2006: Dynamical processes related to the appearance of quasi-stationary waves on the subtropical jet in the midsummer Northern Hemisphere. J. Climate, 19, 15311544.

    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., J. M. Wallace, and G. W. Branstator, 1983: Barotropic wave propagation and instability, and atmospheric teleconnection patterns. J. Atmos. Sci., 40, 13631392.

    • Search Google Scholar
    • Export Citation
  • Smith, T. M., R. W. Reynolds, T. C. Peterson, and J. Lawrimore, 2008: Improvements to NOAA’s historical merged land–ocean surface temperature analysis (1880–2006). J. Climate, 21, 22832296.

    • Search Google Scholar
    • Export Citation
  • Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608627.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., and C. J. Guillemot, 1996: Physical processes involved in the 1988 drought and 1993 floods in North America. J. Climate, 9, 12881298.

    • Search Google Scholar
    • Export Citation
  • Tung, K. K., 1979: A theory of stationary long waves. Part III: Quasi-normal modes in a singular waveguide. Mon. Wea. Rev., 107, 751774.

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 29613012.

  • Wang, B., Y. Huang, J. He, and L. Wang, 2004: Relation between vapor transportation in the period of East Asian summer monsoon and drought in Northwest China. Plateau Meteor., 23, 912918.

    • Search Google Scholar
    • Export Citation
  • Yang, L., and Q. Zhang, 2008a: Effects of the North Atlantic Oscillation on the summer rainfall anomalies in Xinjiang. Chin. J. Atmos. Sci., 32, 11871196.

    • Search Google Scholar
    • Export Citation
  • Yang, L., and Q. Zhang, 2008b: Interannual variation of summer precipitation in Xinjiang and Asian subtropical westerly jet stream. J. Appl. Meteor. Sci., 19, 171179.

    • Search Google Scholar
    • Export Citation
  • Yang, L., X. Duolaite, and Q. Zhang, 2009: Relationships between rainfall anomalies in Xinjiang summer and Indian rainfall. Plateau Meteor., 28, 564572.

    • Search Google Scholar
    • Export Citation
  • Yasui, S., and M. Watanabe, 2010: Forcing processes of the summertime circumglobal teleconnection pattern in a dry AGCM. J. Climate, 23, 20932114.

    • Search Google Scholar
    • Export Citation
  • Zhou, L., and R. Huang, 2008: Interdecadal variability of sensible heat in arid and semi-arid regions of Northwest China and its relation to summer precipitation in China. Chin. J. Atmos. Sci., 32, 12761288.

    • Search Google Scholar
    • Export Citation
  • Zhou, L., and R. H. Huang, 2010: Interdecadal variability of summer rainfall in Northwest China and its possible causes. Int. J. Climatol., 30, 549557.

    • Search Google Scholar
    • Export Citation
Save
  • Charney, J. G., 1975: Dynamics of deserts and drought in the Sahel. Quart. J. Roy. Meteor. Soc., 101, 193202.

  • Chen, D., and Y. Dai, 2009a: Characteristics of Northwest China rainfall intensity in recent 50 years. Chin. J. Atmos. Sci., 33, 923935.

    • Search Google Scholar
    • Export Citation
  • Chen, D., and Y. Dai, 2009b: Characteristics and analysis of typical anomalous summer rainfall patterns in northwest China over the last 50 years. Chin. J. Atmos. Sci., 33, 12471258.

    • Search Google Scholar
    • Export Citation
  • Chen, M., P. Xie, J. E. Janowiak, and P. A. Arkin, 2002: Global land precipitation: A 50-yr monthly analysis based on gauge observations. J. Hydrometeor., 3, 249266.

    • Search Google Scholar
    • Export Citation
  • Chen, M., P. Xie, J. E. Janowiak, P. A. Arkin, and T. M. Smith, 2004: Verifying the reanalysis and climate models outputs using a 56-year data set of reconstructed global precipitation. Preprints, 14th Conf. on Applied Climatology, Seattle, WA, Amer. Meteor. Soc., J6. 1. [Available online at https://ams.confex.com/ams/84Annual/techprogram/paper_70083.htm.]

  • Ding, Q., and B. Wang, 2005: Circumglobal teleconnection in the Northern Hemisphere summer. J. Climate, 18, 34833505.

  • Ding, Q., and B. Wang, 2007: Intraseasonal teleconnection between the summer Eurasian wave train and the Indian Monsoon. J. Climate, 20, 37513767.

    • Search Google Scholar
    • Export Citation
  • Ding, Q., B. Wang, J. M. Wallace, and G. Branstator, 2011: Tropical–extratropical teleconnections in boreal summer: Observed interannual variability. J. Climate, 24, 18781896.

    • Search Google Scholar
    • Export Citation
  • Enomoto, T., B. J. Hoskins, and Y. Matsuda, 2003: The formation mechanism of the Bonin high in August. Quart. J. Roy. Meteor. Soc., 129, 157178.

    • Search Google Scholar
    • Export Citation
  • Held, I. M., 1983: Stationary and quasi-stationary eddies in the extratropical troposphere: Theory. Large-Scale Dynamical Processes in The Atmosphere, B. J. Hoskins and R. P. Pearce, Eds., Academic Press, 127–168.

  • Held, I. M., R. T. Pierrehumbert, and R. L. Panetta, 1985: Stationary external Rossby waves in vertical shear. J. Atmos. Sci., 42, 865883.

    • Search Google Scholar
    • Export Citation
  • Held, I. M., M. Ting, and H. Wang, 2002: Northern winter stationary waves: Theory and modeling. J. Climate, 15, 21252144.

  • Holton, J. R., 2004: An Introduction to Dynamic Meteorology. Academic Press, 529 pp.

  • Hoskins, B. J., and D. J. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38, 11791196.

    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and T. Ambrizzi, 1993: Rossby wave propagation on a realistic longitudinally varying flow. J. Atmos. Sci., 50, 16611661.

    • Search Google Scholar
    • Export Citation
  • Huang, G., Y. Liu, and R. Huang, 2011: The interannual variability of summer rainfall in the arid and semiarid regions of Northern China and its association with the northern hemisphere circumglobal teleconnection. Adv. Atmos. Sci., 28, 257268.

    • Search Google Scholar
    • Export Citation
  • Huang, R., and W. Li, 1988: Influence of heat source anomaly over the western tropical Pacific on the subtropical high over East Asia and its physical mechanism. Chin. J. Atmos. Sci., 12, 107116.

    • Search Google Scholar
    • Export Citation
  • Kosaka, Y., H. Nakamura, M. Watanabe, and M. Kimoto, 2009: Analysis on the dynamics of a wave-like teleconnection pattern along the summertime Asian jet based on a reanalysis dataset and climate model simulations. J. Meteor. Soc. Japan, 87, 561580.

    • Search Google Scholar
    • Export Citation
  • Kosaka, Y., S.-P. Xie, and H. Nakamura, 2011: Dynamics of interannual variability in summer precipitation over East Asia. J. Climate, 24, 54355453.

    • Search Google Scholar
    • Export Citation
  • Li, D., J. Xie, and W. Wang, 1997: A study of summer precipitation features and anomaly in Northwest China. Chin. J. Atmos. Sci., 21, 331340.

    • Search Google Scholar
    • Export Citation
  • Lu, R.-Y., J.-H. Oh, and B.-J. Kim, 2002: A teleconnection pattern in upper-level meridional wind over the North African and Eurasian continent in summer. Tellus, 54A, 4455.

    • Search Google Scholar
    • Export Citation
  • Nitta, T., 1987: Convective activities in the tropical western Pacific and their impact on the Northern Hemisphere summer circulation. J. Meteor. Soc. Japan, 65, 373390.

    • Search Google Scholar
    • Export Citation
  • North, G. R., T. L. Bell, R. F. Cahalan, and F. J. Moeng, 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110, 699706.

    • Search Google Scholar
    • Export Citation
  • Rodwell, M. J., and B. J. Hoskins, 1996: Monsoons and the dynamics of deserts. Quart. J. Roy. Meteor. Soc., 122, 13851404.

  • Sardeshmukh, P. D., and B. J. Hoskins, 1988: The generation of global rotational flow by steady idealized tropical divergence. J. Atmos. Sci., 45, 12281251.

    • Search Google Scholar
    • Export Citation
  • Sato, N., and M. Takahashi, 2006: Dynamical processes related to the appearance of quasi-stationary waves on the subtropical jet in the midsummer Northern Hemisphere. J. Climate, 19, 15311544.

    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., J. M. Wallace, and G. W. Branstator, 1983: Barotropic wave propagation and instability, and atmospheric teleconnection patterns. J. Atmos. Sci., 40, 13631392.

    • Search Google Scholar
    • Export Citation
  • Smith, T. M., R. W. Reynolds, T. C. Peterson, and J. Lawrimore, 2008: Improvements to NOAA’s historical merged land–ocean surface temperature analysis (1880–2006). J. Climate, 21, 22832296.

    • Search Google Scholar
    • Export Citation
  • Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608627.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., and C. J. Guillemot, 1996: Physical processes involved in the 1988 drought and 1993 floods in North America. J. Climate, 9, 12881298.

    • Search Google Scholar
    • Export Citation
  • Tung, K. K., 1979: A theory of stationary long waves. Part III: Quasi-normal modes in a singular waveguide. Mon. Wea. Rev., 107, 751774.

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 29613012.

  • Wang, B., Y. Huang, J. He, and L. Wang, 2004: Relation between vapor transportation in the period of East Asian summer monsoon and drought in Northwest China. Plateau Meteor., 23, 912918.

    • Search Google Scholar
    • Export Citation
  • Yang, L., and Q. Zhang, 2008a: Effects of the North Atlantic Oscillation on the summer rainfall anomalies in Xinjiang. Chin. J. Atmos. Sci., 32, 11871196.

    • Search Google Scholar
    • Export Citation
  • Yang, L., and Q. Zhang, 2008b: Interannual variation of summer precipitation in Xinjiang and Asian subtropical westerly jet stream. J. Appl. Meteor. Sci., 19, 171179.

    • Search Google Scholar
    • Export Citation
  • Yang, L., X. Duolaite, and Q. Zhang, 2009: Relationships between rainfall anomalies in Xinjiang summer and Indian rainfall. Plateau Meteor., 28, 564572.

    • Search Google Scholar
    • Export Citation
  • Yasui, S., and M. Watanabe, 2010: Forcing processes of the summertime circumglobal teleconnection pattern in a dry AGCM. J. Climate, 23, 20932114.

    • Search Google Scholar
    • Export Citation
  • Zhou, L., and R. Huang, 2008: Interdecadal variability of sensible heat in arid and semi-arid regions of Northwest China and its relation to summer precipitation in China. Chin. J. Atmos. Sci., 32, 12761288.

    • Search Google Scholar
    • Export Citation
  • Zhou, L., and R. H. Huang, 2010: Interdecadal variability of summer rainfall in Northwest China and its possible causes. Int. J. Climatol., 30, 549557.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Spatial modes of the first two REOFs based on analysis of 45-yr (1958–2002) July precipitation in Northwest China: (a) REOF1 and (b) REOF2 (by contours). The contour interval is 0.05. The total variance fractions explained by each mode are 14.4% in (a) and 10.7% in (b). The filled circles represent the geographical distribution of the stations.

  • Fig. 2.

    Normalized time series of the first two REOF modes based on analysis of 45-yr (1958–2002) July precipitation in Northwest China and their interdecadal variations: (a) PC1 and (b) PC2. The solid lines represent the normalized time series, and the dash lines represent the corresponding interdecadal variations by removing the interannual variations with a period shorter than 9 years.

  • Fig. 3.

    Regressions of 200-hPa geopotential height to (a) PC1 and (b) PC2 obtained by REOF analysis based on 45-yr (1958–2002) July precipitation in Northwest China.

  • Fig. 4.

    Regressions of 200-hPa geopotential height (by contours) to (a) PC1 and (b) PC2 obtained by EOF analysis on 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E). The vectors are the regressions of 200-hPa wave activity fluxes to the corresponding PCs. The shading indicates the 95% confidence level for the geopotential height.

  • Fig. 5.

    Normalized time series of the first two EOF modes based on the analysis of 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E), and their interdecadal variations: (a) PC1 and (b) PC2. The solid lines represent the nomalized time series, and the dash lines represent the corresponding interdecadal variations by removing the interannual variations with a period shorter than 9 years.

  • Fig. 6.

    Height–longitude cross section of geopotential height averaged between 30.0° and 50.0°N regressed onto (a) PC1 and (b) PC2 of EOF analysis based on the 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E).

  • Fig. 7.

    Regressions of 700-hPa geopotential height to (a) PC1 and (b) PC2 obtained by EOF analysis on 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E).

  • Fig. 8.

    Correlation coefficients between (a) PC1 and (b) PC2 of EOF analysis on 200-hPa meridional wind velocity and rainfall anomalies of 33 observational stations in Northwest China (by shadings), and the regressions of 200-hPa geopotential height to the corresponding PCs (by contours). The contour intervals are 10 m in (a) and 5 m in (b).

  • Fig. 9.

    Correlation coefficients between (a) PC1 and (b) PC2 of EOF analysis on July V200 and land precipitation (PREC/L) anomalies (by shadings), and the regressions of 200-hPa geopotential height to the corresponding PCs (by contours).

  • Fig. 10.

    As in Fig. 4, but for data on interannual time scale.

  • Fig. 11.

    Leading two SVD modes based on the analysis of July 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and PREC tropical rainfall between (15°S and 30°N) during 1958–2002: (a) S1 and (b) S2. Contours are the homogeneous regressions of 200-hPa geopotential height to the corresponding expansion coefficients, and shadings are the heterogeneous correlation coefficients between tropical rainfall and the corresponding expansion coefficients.

  • Fig. 12.

    Correlation coefficients between 200-hPa velocity potential (by contours), divergent flow (by vectors) and (a) V1 and (b) V2 of SVD analysis based on July 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and PREC tropical rainfall between (15°S and 30°N) during 1958–2002. The homogeneous correlation coefficients PREC tropical rainfall and the corresponding expansion coefficients are also shown by shadings. The contour interval is 0.1. The contour levels above 95% confidence level are plotted.

  • Fig. 13.

    Correlation coefficients of Rossby wave sources (defined as , by Sardeshmukh and Hoskins, 1988) and (a) U1 and (b) U2 of SVD analysis (by shadings), and homogeneous regressions of 200-hPa geopotential height on to the corresponding expansion coefficients (by colorful contours), and correlation coefficients between 200-hPa velocity potential (by black contours), divergent flow (by vectors) and (a) V1 and (b) V2 of SVD analysis. The contour interval for correlations of velocity potential is 0.1. The contour levels of velocity potential above 95% confidence level are plotted.

  • Fig. 14.

    Spatial modes of first two EOFs based on analysis of PREC tropical rainfall between 15°S and 30°N: (a) EOF1 and (b) EOF2. The total variance fractions explained by each mode are 21.5% in (a) and 13.8% in (b).

  • Fig. 15.

    Correlation coefficients between SST anomalies and (a) PC1 and (b) PC2 of EOF analysis on PREC tropical rainfall. The contour interval is 0.1. The shading indicates the 95% confidence level.

  • Fig. 16.

    Leading mode of SVD analysis based on 200-hPa meridional wind velocity over the region of (30°–60°N, 30°–130°E) and PREC tropical rainfall between 15°S and 30°N during 1958–2002: (a) June, (b) August, and (c) summer. Contours are the homogeneous regressions of 200-hPa geopotential height to the corresponding expansion coefficients, and shadings are the heterogeneous correlation coefficients between tropical rainfall and the corresponding expansion coefficients.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1385 357 38
PDF Downloads 1225 319 28