Previous studies reported that the summer western Pacific subtropical high (WPSH) has extended westward since the late 1970s and the change has affected summer rainfall over China and tropical cyclone prevailing tracks in the western North Pacific. The authors show that the 500-hPa geopotential height in the midlatitudes of the Northern Hemisphere has trended upward in the warming climate and the westward extension of the WPSH quantified with the 500-hPa geopotential height is mainly a manifestation of the global rising trend. That is, the summer 500-hPa WPSH has not remarkably extended westward since the late 1970s when the global trend is removed. It is suggested that the index that indicates the west–east shift of the summer 500-hPa WPSH should be redefined and that further investigation is needed to understand the observed climate change in the summer rainfall over China and tropical cyclone prevailing tracks in the western North Pacific.
About one-third of the global population is affected by the East Asian summer monsoon, which carries moist air from the Indian Ocean and Pacific Ocean to East Asia, including China, Japan, the Korea Peninsula, and the surrounding areas. The western Pacific subtropical high (WPSH) is an important component of the East Asian summer monsoon system (Rodwell and Hoskins 2001) and may intensify in a warming climate (Li et al. 2012). The WPSH is closely associated with the timing and spatial distribution of summer rainfall in East Asia (Tao and Chen 1987; Ding 1994), as well as prevailing tropical cyclone tracks over the western North Pacific (Ho et al. 2004; Wu and Wang 2004; Wu et al. 2005). Thus variations in the position, shape, and strength of the WPSH are of great importance in understanding of the climate change in the East Asian summer monsoon, summer rainfall in East Asia, and tropical cyclone activity in the western North Pacific basin.
Gong and Ho (2002) and He and Gong (2002) were among the first who investigated the interdecadal change of the WPSH and the associated climate effects. Gong and Ho (2002) found that the WPSH has enlarged, intensified, and shifted southwestward since 1980. These changes were linked to the shifts of the summer rain belts over China and tropical cyclone prevailing tracks and frequency in the western North Pacific that occurred around the late 1970s (Gong and Ho 2002; Ho et al. 2004; Liu and Chan 2008, 2013). Zhou et al. (2009) suggested that the westward extension of the WPSH since the late 1970s was a result of the negative heating in the central and eastern tropical Pacific and enhanced convective heating in the equatorial Indian Ocean and Maritime Continent.
This study was motivated by Wu et al. (2005), which showed an increasing trend in tropical cyclone influence over subtropical East Asia during the period 1965–2003, while the tropical cyclone influence considerably decreased in the South China Sea. In the western North Pacific, the prevailing tropical cyclone tracks are intimately linked to the mean large-scale circulation patterns (Wu and Wang 2004). If westward extension and southward expansion of the WPSH occurs, the prevailing tropical cyclone tracks should shift southward. However, the observed track changes indicate a northward shift (Wu et al. 2005), which is in conflict with the westward extension and southward expansion of the WPSH (He and Gong 2002; Gong and Ho 2002; Ho et al. 2004; Zhou et al. 2009; Liu and Chan 2008, 2013). The future tropical cyclone track change based on climate models also suggested that such a shift will continue in a warming climate (Wang et al. 2011; Wang and Wu 2015).
Many studies identified increases in tropopause height over the past several decades (Highwood and Hoskins 1998; Randel et al. 2000; Seidel et al. 2001; Santer et al. 2003a,b). Numerical simulations suggest that the tropospheric warming is one of the major drivers in the rise of tropopause height (Santer et al. 2003a). As the tropopause height increases, the height of the 500-hPa pressure level should also rise as a result of the tropospheric warming. In this study, we show that the reported westward extension and southward expansion of the WPSH quantified with geopotential height at 500 hPa are mainly a manifestation of the global-scale rise of the geopotential height of pressure levels in the midlatitudes of the Northern Hemisphere.
In previous studies, the westward extension and southward expansion of the WPSH have been quantified with the geopotential height at 500 hPa (Hu 1997; Gong and Ho 2002; He and Gong 2002; Ho et al. 2004; Zhou et al. 2009). In addition to geopotential height, relative vorticity at 500 hPa is also used in this study because of the anticyclonic circulation of the WPSH. Two groups of datasets are used in this study. Three relatively long reanalysis datasets are the National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research (NCAR) reanalysis (2.5° latitude × 2.5° longitude with 17 vertical levels, 1948–2012; Kalnay et al. 1996); the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40; 2.5° latitude × 2.5° longitude with 23 vertical levels, 1958–2001; Uppala et al. 2005); and the National Oceanic and Atmospheric Administration (NOAA) Earth System Research Laboratory (ESRL) Twentieth Century Reanalysis, version 2 (20CRv2; 2° latitude × 2° longitude with 24 vertical levels, 1948–2010; Compo et al. 2011). Four modern reanalysis datasets are the ECMWF interim reanalysis (ERA-Interim; 1.5° latitude × 1.5° longitude with 37 vertical levels, 1979–2012; Simmons et al. 2007), the Japanese 25-year Reanalysis Project (JRA-25; 1.25° latitude × 1.25° longitude with 23 vertical levels, 1979–2012; Onogi et al. 2005), the National Aeronautics and Space Administration (NASA) Modern-Era Retrospective Analysis for Research and Applications (MERRA; ½° latitude × ⅔° longitude with 42 vertical levels, 1979–2012; Rienecker et al. 2011), and the NCEP Climate Forecast System Reanalysis (CFSR; 0.5° latitude × 0.5° longitude with 64 levels, 1979–2009; Saha et al. 2010).
The three-dimensional structure of the time-evolving state in the 20CR dataset is numerically produced by using only observed sea surface temperature, sea ice, and surface pressure observations (Compo et al. 2006, 2008). Because of a relative lack of observational constraints above the surface compared to the other reanalysis datasets, the resulting time series of the atmospheric state may be relatively free of strong biases in the dataset due to relatively homogeneous observations of pressure, sea ice, and sea surface temperature through the whole period (Emanuel 2010). Although the temporal ranges of the three reanalyses are different, they cover the reported westward extension and southward expansion of the WPSH since the late 1970s well. In the following analysis, the ERA-40 dataset has been extended using the ERA-Interim dataset since the ERA-40 dataset is only available by 2001.
3. Rise of the 500-hPa pressure level
In a warming climate, theory and numerical modeling predict a tropical tropospheric warming that increases with height and reaches its maximum around 200 hPa (Manabe and Wetherald 1975; Meehl et al. 2007). Increasing observational evidence supports the projected warming trend (Fu et al. 2004; Santer et al. 2005; Allen and Sherwood 2008; Thorne et al. 2011). Based on the hydrostatic balance of large-scale circulation, the 500-hPa pressure level should be elevated as a result of the tropospheric warming.
Figure 1a shows the spatial distribution of linear trends in 500-hPa geopotential height during the period 1948–2012 from the NCEP–NCAR reanalysis. We can see the significant positive trends of geopotential height in the tropics and subtropics in summer (June–August). In the region of the WPSH, the trends are generally moderate, compared to northern Asia with the largest trend and an area with relatively large trends to the north of the WPSH. The global increasing trends suggest that the rise of the summer 500-hPa geopotential height is not a regional feature. Figure 1b further shows the summer 500-hPa geopotential height averaged over the latitudinal band (0°–40°N) from the NCEP–NCAR reanalysis, ERA-40, and 20CR, indicating the significant increasing trends of 3.7, 2.3, and 1.6 m decade−1, respectively. Close inspection finds that the upward trends can be largely accounted for by an abrupt increase in the late 1970s. This is consistent with interdecadal-scale transition of East Asian climate since the late 1970s (Hu 1997; Wang 2001; Hu et al. 2003; Yu et al. 2004; Li et al. 2005; Xin et al. 2006; Yu and Zhou 2007).
In Fig. 1b, we also show the time series of the mean height that is derived from the hydrostatic balance equation with the mean NCEP–NCAR reanalysis temperature profile over the latitudinal band (0°–40°N). The calculated mean height perfectly matches the mean geopotential height in the reanalysis dataset, clearly suggesting that the rises of the 500-hPa pressure level in the reanalyses are mainly a result of the tropospheric warming, in agreement with the rise of tropopause height (Santer et al. 2003a).
4. Changes in the WPSH
In previous studies, the change of the WPSH is indicated by the shift of the geopotential height contour of 5870 gpm at 500 hPa (e.g., Gong and Ho 2002; Ho et al. 2004; Zhou et al. 2009; Liu and Chan 2013). Following these previous studies, Fig. 2a shows the westward extension and southward expansion of the WPSH by comparing the 5870-gpm contour before and after 1979. The post-1979 mean geopotential height is calculated over the periods 1980–2012, 1980–2012, and 1980–2010 for the NCEP–NCAR reanalysis, ERA-40, and 20CR, respectively. Although the degrees of the expansion of the 5870-gpm contour are different in individual reanalysis datasets, as documented in previous studies, the figure clearly shows the westward extension and southward expansion of the contour since the late 1970s.
The area mean value of the geopotential height at 500 hPa was calculated within the region 15°–30°N, 120°–140°E. As shown in Fig. 2a, the selected area mainly covers the extension of the 5870-gpm contour since the late 1970s. Sui et al. (2007) found that a large standard deviation of the geopotential height on the interannual time scale also occurs in this area. Figure 2b shows the time series from the three reanalysis datasets, all showing a significant increasing trend. The trends are 3.5, 1.7, and 1.5 m decade−1 in the NCEP–NCAR reanalysis, ERA-40, and 20CR, respectively. The linear trends are actually less than the global averages over the latitudinal band (0°–40°N). That is, the increases of geopotential height in the selected area are mainly a result of the global-scale rise of the geopotential height over the tropics and subtropics of the Northern Hemisphere. Further examination indicates that no significant westward shifts can be identified in the center of the WPSH (figure not shown). Moreover, it seems that the strength of the WPSH since 1979 in Fig. 2b is weaker.
The influence of the rising 500-hPa pressure level can be clearly demonstrated with the geopotential height relative to the global mean averaged over 0°–40°N (Fig. 3a). For clarity, we used the 8-gpm contour to represent the WPSH since it is close to the climatologic position of the 5870-gpm contour, which has been extensively used in previous studies. We use the band mean as a reference because the WPSH is identified at 500 hPa by comparing with the surrounding geopotential height. The westward extension of the WPSH shown in Fig. 2a disappears although the spatial coverage of the WPSH is different in the individual datasets. We also used the mean geopotential height averaged over 15°–30°N instead of the above latitudinal band and found little change in the westward extension of the WPSH. Figure 3b is similar to Fig. 2b, but the global-scale rise of the geopotential height was removed from the time series. After the zonal mean is subtracted, no trends are statistically significant in the three reanalysis datasets.
We further calculated the mean relative vorticity in the selected area (Fig. 3c). Note that the global mean averaged over 0°–40°N was not removed in this figure because the global rise of the pressure level should have little influence on the relative vorticity of the WPSH. Because of the nature of anticyclonic circulation, the selected area (15°–30°N, 120°–140°E) is generally controlled by negative relative vorticity (Fig. 3c). The only significant trend in relative vorticity is found in the 20CR dataset, in fact indicating the weakening of the WPSH in the selected area. Therefore, we can conclude that the WPSH has not extended westward since the late 1970s because the westward extension of the geopotential height contour of 5870 gpm in previous studies was a manifestation of the rising 500-hPa pressure level.
We also examined the changes of the 500-hPa geopotential height averaged over the region 15°–30°N, 120°–140°E in the modern reanalysis datasets during the period 1979–2012 (Fig. 4). As we know, there is a significant increasing trend in land and ocean temperature since 1979. Figure 4a shows an insignificantly increasing trend in these datasets during this period. After the zonal mean is removed, insignificant decreasing trends can be seen (Fig. 4b), indicating that no increasing trends are found during the period 1979–2012.
The westward extension of the summer WPSH since the late 1970s has been reported in previous studies (Gong and Ho 2002; Ho et al. 2004; Zhou et al. 2009). In this study, we demonstrate that the westward extension quantified with the 500-hPa geopotential height is mainly a manifestation of the global-scale rise of the geopotential height of the pressure level in the midlatitudes of the Northern Hemisphere. It is suggested that the index for the west–east shift of the summer 500-hPa WPSH should be redefined. Since the WPSH has not extended westward since the late 1970s, this study suggests that further investigation is needed to understand the climate change of the rainband over China and tropical cyclone prevailing tracks in the western North Pacific basin.
Huang et al. (2015) also examined the west–east extension of the summer 850-hPa WPSH on the interdecadal time scale and found that the WPSH recessed eastward during 1979–2009 relative to 1948–78. They argued that the water vapor transport from the Indian summer monsoon played a more important role in the summer precipitation over the middle and lower reaches of the Yangtze River valley during 1979–2009 relative to 1948–78. Their results agree with our analysis, although they focused on the lower pressure level.
Previous studies have shown that the mechanisms for the formation and maintenance of the WPSH are very complicated. The change of the WPSH can result from the impact of the Tibetan Plateau (Ye and Gao 1979; Ye and Wu 1998; Wang et al 2008), diabatic heating associated with the monsoon (Ting 1994; Hoskins 1996; Rodwell and Hoskins 2001), land–ocean heating contrast (Wu and Liu 2003; Miyasaka and Nakamura 2005), air–sea interaction (Seager et al. 2003), and sea surface temperature changes (Lu and Dong 2001; Zhou and Wang 2006; Wang et al. 2013). It is likely that the individual impacts of these mechanisms can cancel each other out, resulting in little net impact on the WPSH. Further investigation is thus needed to understanding of the change of the WPSH.
This research was jointly supported by the National Basic Research Program of China (2013CB430103 and 2015CB452803), the National Natural Science Foundation of China (Grant 41275093), and the project of the specially appointed professorship of Jiangsu Province.