1. Introduction
Persistent heavy rainfall over eastern China often results in hazardous climate events such as floods and droughts. For example, excessive rainfall over southern China caused serious floods in the lower Yangtze River valley (YRV; shown in Fig. 1) during the summers of 1991, 1995, 1996, 1998, and 1999, and deficient rainfall over northern China in 1997 caused a severe arid period of 226 days for the Yellow River valley (Zhai et al. 2005). Thus, it is important to understand the variations of rainfall over eastern China and their associated physical processes.
Many studies have revealed prominent long-term changes of summer rainfall over East Asia (e.g., Hu 1997; Ren et al. 2000; Xu 2001; Gong and Ho 2002). Weng et al. (1999) have showed an abrupt shift of rainfall regime over China in the late 1970s, associated with the quasi-in-phase reinforcement between the bidecadal and quadric-decadal signals. Ren et al. (2000) have demonstrated a wetting trend over the middle-lower YRV and a drying trend over the Yellow River valley during the past 40 years, and Xu (2001) has found a southward shift of the midsummer rain belt in the last two decades. By analyzing the long-term variations of the simulated summer climate over China, Hu et al. (2003) found a drying trend in northern China and a wetting trend in central China. Previous studies have also examined the trends in annual precipitation (Wang and Zhai 2003; Liu et al. 2005). Interestingly, when analyzing the positive–negative–positive pattern in the interdecadal variability of summer rainfall over eastern China, Zhao and Zhou (2006) showed that, when negative anomalies occurred over YRV, positive anomalies emerged over both southeastern and northern China. Ding et al. (2007) added that the interdecadal variability of summer rainfall over eastern China was associated with two meridional modes: a dipole pattern and a positive–negative–positive pattern.
The variability of summer rainfall over eastern China is strongly modulated by the East Asian monsoon, which is linked to the conditions of Eurasian and Tibetan snows and tropical sea surface temperature (SST), among others (e.g., Yang and Xu 1994; Zhao and Xu 2002; Hu et al. 2003; Zhang et al. 2004). On interannual time scales, wet anomalies usually appear over southern China during El Niño episodes (e.g., Huang and Wu 1989; Dai and Wigley 2000). On interdecadal time scales, the subtropical northwestern Pacific high and the tropical Pacific SST play important roles in the variations of rainfall over eastern China (Chang et al. 2000a,b). Gong and Ho (2002) also proposed that, since 1980, the variations of SST over the tropical eastern Pacific and the tropical Indian Ocean were primarily responsible for the shift in summer rainfall over eastern China through their effects on the subtropical northwestern Pacific high. Furthermore, Yang and Lau (2004) showed a positive trend of summer precipitation over central-eastern China and a negative trend over northern China in the past 50 years and linked these trends to the interdecadal variations of global SST.
Besides SSTs, Yu et al. (2004) attributed the pattern of rainfall changes over eastern China to the summer cooling at the upper troposphere over extratropical East Asia, which was assumedly associated with stratosphere–troposphere interactions. Changes in both land temperature and SST modify land–ocean temperature gradients. When the temperature gradients became smaller, the southwesterly monsoon flow weakened and the moisture transported to southern China decreased, causing the local drying trend (Cheng et al. 2005). More recently, Ding et al. (2007) showed that the significant weakening of the tropical upper-level easterly jet, which could also be a result of the reduced temperature gradients, provided a dominant mechanism for the weakening of the Asian summer monsoon over the past 40 years.
In summary, previous studies on the interdecadal variations of precipitation over China have focused mainly on the trend of precipitation and the associated physical explanations by the variability of tropical oceans and the subtropical northwestern Pacific high. However, several relevant issues remain unaddressed. First, the occurrence of the Asian monsoon rainfall anomalies is a consequence of the atmospheric response to the changes in the thermal contrast between Asia and the adjacent oceans. Associated with the global warming during the past decades, global land surface and ocean temperatures generally exhibit increasing trends (Solomon et al. 2007). Differences in these temperature trends between land and ocean may change the thermal contrast between the land and the oceans. Thus, it is important to examine the relationships between the long-term variation of the East Asian monsoon rainfall and the changes in global climate. Second, the elevated heating of the Tibetan Plateau (TP) plays an important role in establishing and maintaining the Asian summer monsoon circulation (e.g., Ye and Gao 1979; Li and Yanai 1996). Since the southern floods and northern droughts were closely associated with the anomalies of the East Asian monsoon circulation, it is necessary to investigate the role of TP in the anomalous monsoon rainfall pattern over eastern China.
In this study, we use the precipitation of rain gauge stations in China and the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis to investigate the link of the long-term variation of the durations of persistent rainfall over eastern China to the change in global climate. We also analyze the associated atmospheric circulation over East Asia and the North Pacific and the temperatures over land surface and oceans to explain the variations of rainy seasons and the pattern of southern floods and northern droughts.
2. Data
We use the monthly and daily data from the ECMWF reanalysis with a horizontal resolution of 2.5° latitude and longitude (Gibson et al. 1997) and the monthly SST from the Hadley Centre Sea Ice and Sea Surface Temperature dataset (HadISST) with a horizontal resolution of 1° latitude and longitude (Rayner et al. 1996) for 1958–2001. We also use the daily precipitation (from 1960 to 2001) of rain gauge stations in China whose locations are shown in Fig. 1.
To assess the robustness of the results obtained from analyzing the ECMWF surface and 500-mb temperatures, soil moisture content, and snow depth, we further analyze the monthly land surface air temperature (Fan and Van den Dool 2008) and soil moisture content (Fan and Van den Dool 2004) from the National Oceanic and Atmospheric Administration’s (NOAA) Climate Prediction Center (CPC), as well as the monthly snowfall, surface pressure, and 500-mb temperature observed at stations of China. The station data described above are archived at the National Meteorological Information Centre of the China Meteorological Administration. Moreover, we compare some results over TP between the ECMWF reanalysis and the National Centers for Environmental Prediction (NCEP) reanalysis.
3. Relationship between rainfall over eastern China and global climate
a. Cold and warm periods of global climate
Figure 2 shows the changes in the annual global-mean surface air temperature (GSAT). The ECMWF reanalysis exhibits a pronounced long-term change, with a negative polarity during the 1960s–70s and a positive polarity during the 1980s–90s. The temperature was the lowest in 1974 and the highest in 1998. The figure indicates a positive trend of the annual GSAT from the 1960s to the 1990s, similar to the features reported by the Intergovernmental Panel on Climate Change (Solomon et al. 2007). These varying features of temperature revealed by the ECMWF reanalysis also appear from two other independent datasets: the global surface air temperature over land from CPC and the SST data from HadISST. Shown by the dotted line in Fig. 2 is an index combining the two datasets through an area-weighted average.
We also note in Fig. 2 that the annual GSAT from the ECMWF reanalysis was lower before the late 1970s, but higher in the 1980s and 1990s, compared to the GSAT from the area-weighted average. The difference is mainly due to the lower temperature over land in the ECMWF reanalysis from the late 1960s to the mid-1970s and to the higher temperature over land in the 1980s and the 1990s and over oceans around 1980 and 1990. However, the difference between the two curves does not change their varying trends.
To understand the linkage between the variation of rainfall over eastern China and the change in global climate, we use the annual GSAT to measure the variability of global climate. We consider 1960–79 a relatively cold period and 1982–2001 a relatively warm period, and we analyze the differences in rainfall and associated atmospheric circulation between the two periods.
b. Long-term variation of rainfall
Figure 3a shows the climatological pattern of the annual accumulated rainfall over China. The rainfall exceeds 1600 mm over southern China, with a maximum of 2000 mm over YRV (Fig. 3a) where the standard deviation of the rainfall maximizes at 350 mm (Fig. 3b). Similar features are also observed for summer [June–August (JJA)], when the accumulated rainfall exceeds 600 mm over eastern China between 26° and 32°N (Fig. 3c), with a maximum standard deviation of 250 mm (Fig. 3d). Here, we define the period of persistent rainfall over the southern (northern) region of eastern China as the time span in which the daily rainfall exceeds 6.6 (4.6) mm day−1 and it is not continuously less than the value for more than 10 (7) days. The numbers of 6.6 and 4.6 used in this definition are based on the climatological daily means of JJA rainfall averaged over southern China (25°–30°N, 115°–120°E) and over northern China (35°–40°N, 115°–120°E), respectively.
Figure 4a shows the evolutions of daily rainfall averaged over the southern region of eastern China for the cold and warm periods. In the cold period, persistent rainfall exceeding 6.6 mm day−1 began on the 90th day (31 March) of the year and ended on the 180th day (29 June), with a persistent period of 91 days. In the warm period, however, the persistent rainfall commenced on the 75th day (16 March) and ended on the 188th day (7 July), with a persistent period of 114 days. As seen from the composite difference in the smoothed daily rainfall between the cold and warm periods (Fig. 4c), rainfall increased generally from the 75th day to the 89th day and from the 181st day to the 188th day. That is, over the southern region, the rainy season during the warm period was longer than the rainy season during the cold period by 23 days (significant at the 95% confidence level). Over the northern region (Fig. 4b), the persistent rainfall started on the 175th day (24 June) and stopped on the 234th day (22 August) during a cold period. It started on the 187th day (6 July) and ended on the 232nd day (20 August) in a warm period. It is observed from Fig. 4d that rainfall decreased generally from the 175th day to the 186th day. Over this region, the period of the persistent rainfall was shorter in the warm period (46 days) than in the cold period (60 days) by 14 days (significant at the 90% confidence level).
The differences in the length of rainy season over eastern China between the cold and warm periods caused long-term changes in the local rainfall. Figure 5a shows the composite difference in the accumulated rainfall for 16–30 March (representing the earlier beginning period of rainy season over southern China) between the warm and cold periods. In the figure, significant positive anomalies of the rainfall appeared over southeastern China, peaking at 50 mm near 25°N, indicating that the earlier start of the rainy season over southern China during the warm period resulted in an increase in rainfall. Similarly, Fig. 5b shows the composite difference in the accumulated rainfall for the time span from 30 June to 7 July (representing the later ending period of rainy season over southern China). Now, the positive anomalies exceeding 40 mm emerged over YRV near 30°N, with a maximum of 100 mm, indicating that the later ending of rainy season during the warm period resulted in a significant increase in the local rainfall. Compared to Fig. 5a, the largest positive anomalies in Fig. 5b shift northward. Climatologically, the rainy season over eastern China begins to the south of YRV during spring and advances to YRV during summer (e.g., Tao et al. 1958; He et al. 2008). Thus, this northward shift of the positive anomalies might be related to the seasonal variation of the East Asian rainfall.
As seen from the composite difference in the accumulated rainfall both from 24 June to 5 July (representing the later beginning period of rainy season over northern China) and from 21 to 22 August (representing the earlier ending period of rainy season over northern China) (Fig. 5c), the shorter rainy season over northern China during the warm period was linked to a decrease in rainfall as showed by the negative values below −20 mm over 34°–38°N. More analysis shows that the decrease in rainfall was mainly due to the later beginning of rainy season. This reduction in the local rainfall was accompanied by an increase in the rainfall over YRV. We also note that the pattern of Fig. 5c is similar to the pattern of Fig. 5b for eastern China.
Figure 5d shows the composite difference in JJA rainfall between the warm and cold periods. In this figure, positive values exceeding 150 mm appeared over YRV and large negative values were observed between 35° and 40°N, with a central value of −100 mm, indicating an increase in the summer rainfall over YRV and a decrease over northern China in the warm period relative to the cold period. This north–south stratification of JJA rainfall is similar to the rainfall pattern of southern floods and northern droughts documented previously (e.g., Hu 1997; Ren et al. 2000; Xu 2001; Gong and Ho 2002), showing that the pattern may be detected through a composite analysis of the global-mean surface temperature between the warm and cold periods.
We examine the relative contribution of the changes in the starting and ending dates of summer rainy season to the change in JJA total rainfall. Over YRV, the accumulated rainfall for a later ending time from 30 June to 7 July shown in Fig. 5b accounts for 20%–50% of the positive rainfall anomalies shown in Fig. 5d, contributing to the southern floods in the past decades. Similarly, over northern China, the accumulated rainfall for the time span both from 24 June to 5 July and from 21 to 22 August (shown in Fig. 5c) accounts for 20%–40% of the negative rainfall anomalies shown in Fig. 5d. Thus, the shorter rainy period over northern China contributed to the northern droughts in the warm period. Despite the fact that the pattern of southern floods and northern droughts has been studied extensively (e.g., Hu 1997; Ren et al. 2000; Xu 2001; Gong and Ho 2002), here we further reveal that the rainfall pattern is associated with the length of rainy season and linked to the change in global climate during the past decades.
We further assess the significance of the long-term changes in eastern China precipitation between the warm and cold periods. Figures 6a,b show the values of total rainy-season precipitation for individual years and their averages during 1960–79 and 1982–2001, respectively, for both southern and northern China. Over southern China (Fig. 6a), the mean values of rainy-season precipitation are 769 mm for the cold period and 933 mm for the warm period, with their difference of 164 mm. Our 100-time Monte Carlo simulations (Wilks 2005) show that the difference is significant at the 97% confidence level (Fig. 6c). Over northern China (Fig. 6b), the mean values of rainy-season precipitation for the cold and warm periods are 408 and 292 mm, respectively. Here, the decrease in precipitation (from the cold period to the warm period) by 116 mm is even more highly significant by exceeding the 99% confidence level (Fig. 6d).
During the cold period (Fig. 7a), the persistent rainfall exceeding 8 mm day−1 began on 1 June over southeastern China, 17 June over YRV, and 19 July over northern China (near 40°N), showing a pronounced advance from southeastern China through YRV to northern China. This is consistent with the general feature revealed by previous studies (e.g., Tao et al. 1958; Ding 2004; He et al. 2008). During the warm period (Fig. 7b), however, the persistent rainfall exceeding 8 mm day−1 appeared mainly over YRV and the rainfall over northern China was small. That is, during the warm period, the heavy rainband did not exhibit a pronounced northward advance to northern China.
4. Variations of atmospheric circulation between the warm and cold periods
a. Association with changes in rainfall durations
Figure 8a shows the composite patterns of 850-mb geopotential height averaged for 16–30 March in the cold and warm periods. In the cold period, the subtropical western Pacific high measured by the 1500-m contour appeared to the east of 140°E. In the warm period, however, it expanded westward to 120°E, indicating a more westward subtropical high. Significant positive anomalies were seen over the middle-lower latitudes of the western North Pacific while negative anomalies were observed over the midlatitudes of the East Asian continent (Fig. 8b). Accordingly (Fig. 8c), an anomalous anticyclonic circulation appeared over the western North Pacific, centering at 25°N, 140°E. Meanwhile, a smaller-scale anomalous cyclonic circulation occurred over central China, with a center at 30°N, 110°E. Anomalous southwesterly flow between the anomalous cyclonic and anticyclonic centers prevailed over southeastern China and its adjacent oceans. Since the springtime rainfall over southern China is associated with the strengthening of the southwesterly winds in front of the low near eastern TP and behind the subtropical high over the western North Pacific (Zhao et al. 2007a), the anomalous wind over southeastern China shown in Fig. 8c favored an earlier onset of the southwesterly winds over southeastern China. Thus, the analysis above indicates that the earlier appearance of rainfall over southern China during 16–30 March was associated with strengthening of the subtropical high over the western North Pacific and the low near eastern TP.
Figure 9a shows the composite patterns of 850-mb geopotential height averaged over the later ending period of rainy season (from 30 June to 7 July). Compared to the cold period, the subtropical high over the western North Pacific moved westward and the trough near eastern TP weakened in the warm period, with the contour of 1420 m near 115°E expanding northwestward and leading to an increase in the geopotential height over East Asia as seen in the significant positive anomalies (Fig. 9b). Correspondingly, an anomalous anticyclonic center appeared near 40°N, 120°E and anomalous easterly winds prevailed near YRV (Fig. 9c). They may strengthen the low-level convergence over YRV and lead to an increase in the local rainfall and a later ending of the rainy season. Meanwhile, the northern anomalous anticyclonic center may also decrease the rainfall over northern China.
For the 14-day period both from 24 June to 5 July and during 21–22 August, the composite differences in the 850-mb geopotential height and winds between the warm and cold periods (figures not shown) are similar to those shown in Fig. 9. Namely, significant positive anomalies in geopotential height appeared over East Asia, with an anomalous anticyclonic pattern over East Asia, and anomalous northerly winds prevailed over eastern China. These variations may be responsible for the later beginning and earlier ending of the rainy season over northern China.
b. Association with changes in summer rainfall
Figure 10 shows the composite differences in JJA sea level pressure (SLP) and 850-mb winds between the warm and cold periods. In Fig. 10a, positive anomalies of SLP exceeding 0.5 mb were found over the middle-lower latitudes of Asia, with the maximum values above 1 mb near TP, indicating an increase in SLP during the warm period. A further analysis from the monthly surface pressure at stations over China indicates that positive anomalies of JJA surface pressure appeared over TP and its adjacent areas (figure not shown), consistent with the feature shown in Fig. 10a (for the ECMWF reanalysis). Because climatologically a large-scale surface low pressure system, namely the Asian continent low, is located over Asia during summer (figure not shown), the positive anomalies of SLP indeed indicate a weaker Asian continent low in the warm period. Previous studies have emphasized effects of the subtropical high over the western North Pacific on the southern floods and northern droughts over eastern China (e.g., Chang et al. 2000a,b; Gong and Ho 2002). Here, we demonstrate that there exist more pronounced differences (between the warm and cold periods measured by the annual GSAT) in the pressure systems over Asia than over the western North Pacific.
Corresponding to the increase in SLP over eastern China, an anomalous anticyclonic circulation at 850 mb appeared to the northeast of TP, with anomalous northeasterly or northwesterly winds over eastern China to the south of 40°N (Fig. 10b). Climatologically, the summer monsoon circulation over eastern China is characterized by the prevalence of low-level southwesterly winds (e.g., Chen et al. 1991; Ding 2004; Zhao et al. 2007a). When the monsoon circulation is weak (strong), the southwesterly winds weaken (strengthen) and stagnate over southern (expand to northern) China. Thus, the anomalous northeasterly or northwesterly winds over eastern China indicate weaker summer monsoon circulation over East Asia in the warm period compared to the cold period.
The mei-yu front forms between the dry and cold air masses over northern China and the warm and wet air masses over southern China during summer and its position fluctuates meridionally around YRV (Tao et al. 1958). Relatively denser contours (meaning a larger meridional gradient) of pseudoequivalent potential temperature (θse) may indicate the location of the mei-yu front (Zhao et al. 2004). The anomalous northerly winds associated with a weaker monsoon circulation over eastern China favor the northern dry and cold air masses to invade southward into YRV, affecting the mei-yu front. It is seen from Fig. 11a that the significant negative anomalies of JJA θse appeared at the lower troposphere between 28° and 40°N, indicating a drier and colder air mass associated with the local anomalous northerly winds. Meanwhile, significant positive anomalies of θse mainly appeared to the south of 25°N, indicating a warmer and wetter air mass. More analysis shows that the mei-yu front appeared at surface between 34° and 40°N in the cold period and stayed between 31° and 39°N in the warm period (figures not shown). That is, the mei-yu front was located more southward in the warm period compared to the cold period.
Figure 11b shows the composite difference in JJA vertical p velocity. Anomalous upward motion at the middle-lower troposphere over YRV is seen in front of the negative anomalous center of θse (between 30° and 35°N), indicating a stronger upward motion. Anomalous downward motion, or weaker upward motion, appeared between 33° and 40°N. These variations in vertical motions may be associated with an increase in the summer rainfall over YRV and a decrease in rainfall over northern China, as shown in Fig. 5d. Thus, corresponding to a weaker monsoon circulation over eastern China, the rainy belt stagnated over YRV for a longer time before advancing northward to northern China. This led to an increase in summer rainfall over YRV and a decrease in rainfall over northern China in the warm period.
The foregoing analyses show that the East Asian summer monsoon circulation became weaker in the warm period, relative to the cold period. What, then, are the physical processes responsible for the weakening of the monsoon circulation during the warm period?
5. Explanation of weaker East Asian summer monsoon circulation
a. Thermal contrast between the Asian continent and adjacent oceans
Since the Asian monsoon varies strongly with the thermal contrast between the Asian continent and the adjacent oceans such as the Indian Ocean and the North Pacific (e.g., Webster and Yang 1992; Li and Yanai 1996; Zhao et al. 2007b), we examine how the temperatures over the continent and the oceans changed between the warm and cold periods.
Figure 12a shows the composite difference in JJA land surface temperature. Negative anomalies were clearly seen in the subtropics–extratropics near TP. They first appeared in spring (March–May; figures not shown). These indicate a cooling land in subtropical–extratropical East Asia during spring and summer in the warm period, which is different from a general increase in global land temperature under the recent global warming. Figure 12b shows the composite difference in JJA 500-mb temperature and it reveals negative anomalies over TP and its adjacent areas, with a central value of −0.6°C over eastern TP. An examination of the radiosonde data at Xining station (see Table 1) where the negative anomalous center over eastern TP was located (shown in Fig. 12b) also indicates that the JJA 500-mb temperature decreased in the warm period relative to the cold period (figure not shown), confirming the result shown by the ECMWF reanalysis.
On the other hand, significant positive anomalies of SST occurred over the tropical Indian and western Pacific Oceans, with central values exceeding 0.5°C during summer (Fig. 12c). The increase in the Indo-Pacific SST is consistent with the rise of global-mean SST documented in Solomon et al. (2007).
Corresponding to the decrease in surface temperature over TP and its adjacent areas and the increase in the Indo-Pacific SST, temperature anomalies occurred within much of the troposphere. Negative temperature differences appeared over TP during both spring (figures not shown) and summer (Fig. 13). Along 70°–110°E (Fig. 13a), negative anomalies were found at the entire troposphere between 30° and 40°N, while positive anomalies were observed at the lower troposphere below 600 mb over the tropical oceans. Along 32.5°N (Fig. 13b), there existed significant negative differences at the whole troposphere over Asia and there were positive differences between 850 and 200 mb over the central-eastern Pacific and at the lower troposphere over the western Pacific (near 130°E).
Therefore, the decrease in the land surface and tropospheric temperatures over TP and its adjacent areas and the increase in the sea surface and low-level tropospheric temperatures over the tropical Indian and western Pacific Oceans reduced the low-level meridional and zonal thermal contrasts between East Asia and the surrounding oceans. These changes in thermal contrasts were associated with the weakening of the Asian low and the East Asian summer monsoon circulation (see Fig. 10).
Moreover, significant positive anomalies of land surface temperature occurred to the south of 25°N, extending vertically to a thin layer (below 700 mb), and significant negative anomalies appeared to the north of 30°N (Fig. 12a). These changes in surface temperature strengthen the meridional thermal contrast between southern and northern China, favoring the mei-yu front southward under a weaker summer monsoon circulation.
b. Reasons for the cooling of Tibetan land
What is responsible for the decrease in the Tibetan temperature during spring and summer? Previous studies, both observational analyses and model simulations, have documented that the cold-season snow cover over TP exerts a strong influence on the variability of the local atmospheric temperature. When the Tibetan snow cover increases in spring, the local tropospheric temperature decreases during spring and subsequent summer, and vice versa (e.g., Ose 1996; Sankar-Rao et al. 1996; Zhang et al. 2004). Because of the limitation in surface stations over western TP before 1973, we examine the snow depth in the ECMWF reanalysis and the snowfall at some Tibetan stations where continuous snowfall records are available during 1960–2001.
During the past 40 years, the variability of Tibetan snow was unique, exhibiting an increasing trend in both winter and spring. Figure 14a presents the composite difference in winter–spring (January–May) snow depth between the warm and cold periods. It is found that positive anomalies appeared over the central southwestern TP, with largest values over southern TP, indicating an increase in snow cover in the warm period. We select the surface stations of southwestern TP (see Table 1) where significant positive anomalies of snow depth were located (shown in Fig. 14a). Figure 15 shows a positive linear trend of the winter–spring snowfall at the stations (exceeding the 90% confidence level) during 1960–2001, indicating more snowfall over southwestern TP in the warm period relative to the cold period. These results are in general consistent with the finding by Zhang et al. (2004), who analyzed the eastern Tibetan snow depth during 1962–93 and documented an increasing trend of Tibetan spring snow depth after the late 1970s due to an increase in snowfall.
Figure 14b shows the composite difference in the ECMWF spring–summer (from March through August) soil (0–7 cm) moisture content. There existed positive anomalies of soil moisture content over central and western TP and its neighboring areas, indicating an increase in soil moisture content in the warm period. Overall, positive anomalies of soil moisture content corresponded to an increase in snow depth (see Fig. 14a) and a decrease in temperature (Figs. 12a,b). Figure 14c further shows the composite difference in soil moisture content from the CPC dataset. Here, positive anomalies appeared over most of TP, similar to those seen from Fig. 14b. This analysis further confirms the result obtained using the ECMWF soil moisture content.
Therefore, as the annual GSAT increased over the past decades, the winter–spring snowfall over southwestern TP exhibited a positive trend, which may lead to an increase in the local soil moisture content. Consequently, the land surface and tropospheric temperatures over these regions decreased during spring and summer.
6. Summary and discussion
We have used the precipitation from rain gauge stations over China to investigate the long-term variations of the durations of persistent rainfall under global warming during the past 40 years. It is found that the global-mean surface air temperature experienced a relatively cold period in the 1960s–70s and a relatively warm period in the 1980s–90s, and different features in the long-term variations of rainfall over eastern China appeared between the two periods. Compared to the cold period, the persistent rainfall over the southern portion of eastern China began earlier and ended later during the warm period, lengthening the rainy season by 23 days. Over the northern portion, the persistent rainfall during the warm period began later, with a shorter rainy season by 14 days. These changes in rainfall led to an increase of rainfall over the south and a decrease over the north, contributing to the pattern of southern floods and northern droughts over eastern China. Moreover, heavy summer rainfall advanced northward from southeastern China through YRV to northern China during the cold period. On the contrary, the heavy rain belt experienced little northward shift during the warm period.
Over southern China, the earlier start of spring rainfall during the warm period was linked to a more westward subtropical high over the western North Pacific and a stronger trough near eastern TP. The later ending of summer rainfall was associated with the westward stretch of the subtropical high and the weakening of the trough. Over northern China, the shorter rainy season in summer was also associated with these changes in the trough and the subtropical high.
Over TP, the winter–spring snow cover exhibited an increasing trend in the past decades, which is different from the decreasing trend of snow cover over the plains and lower mountains of the Northern Hemisphere as a whole (Solomon et al. 2007). This increase in snow cover over TP increased the local soil moisture content, accompanied by a decreasing trend of land and tropospheric temperatures over TP during spring and summer. Meanwhile, as the global-mean temperature increased, the summer SST over the tropical Indian Ocean and the western North Pacific also increased. The decrease in temperature over TP and adjacent areas and the increase in temperature over the oceans weakened the thermal contrasts between Asia and its adjacent oceans during summer. The low pressure system over Asia weakened and anomalous northerly winds prevailed over eastern China. These features indicate a weak monsoon circulation over eastern China, consistent with an overall weakening of the global land precipitation (Wang and Ding 2006).
Corresponding to the weak summer monsoon circulation over eastern China, the summertime mei-yu front failed to advance northward significantly but stagnated in YRV, with upward motions strengthened over YRV and weakened over northern China. Meanwhile, rainfall increased over YRV but decreased over northern China, explaining the pattern of southern floods and northern droughts over eastern China.
Previous studies have attempted to explain the pattern of southern floods and northern droughts by different aspects of atmospheric circulation including the changes in the subtropical high over the western North Pacific, the tropical upper-tropospheric easterly jet stream, and the tropospheric cooling over East Asia (e.g., Chang et al. 2000a,b; Gong and Ho 2002; Yu et al. 2004; Ding et al. 2007). Here, we have emphasized the link of this pattern to the low pressure system over Asia and demonstrated the important role of the Tibetan Plateau.
According to Zhang et al. (2004), the India–Burma trough over southern TP influences the anomalies of snowfall over TP. When the trough deepens, the subtropical westerly jet and the ascending motion over TP are stronger, contributing to more snowfall. Moisture supply coming from the Bay of Bengal and the Indian Ocean is also a factor for the Tibetan snowfall anomalies. However, the features of long-term changes in these factors and their effects on winter–spring snowfall are unclear at present. Moreover, different from spring and summer, the wintertime (January–February) temperature averaged over TP increased in the warm period compared to the cold period (Fig. 16), which was consistent with the increasing trend in the global-mean temperature. However, the role of this winter warming over TP in the increase of the local snowfall also remains unclear and should be addressed by future studies.
Acknowledgments
We thank Dr. Y. Fan at the NOAA’s Climate Prediction Center for providing the monthly data of surface air temperature over land and soil moisture content. We also thank the European Centre for Medium-Range Weather Forecasts, NOAA’s Climate Diagnostic Center, and the Hadley Centre, Met Office for providing the reanalysis datasets available on the Internet. Finally, we are grateful to the editor, A. J. Pitman, and the anonymous reviewers for their helpful and careful reviews that greatly improved the manuscript. This work was jointly sponsored by the National Natural Science Foundation of China (40625014; 40890053) and the National Key Basic Research Project of China (2009CB421404).
REFERENCES
Chang, C-P., Y. Zhang, and T. Li, 2000a: Interannual and interdecadal variations of the East Asian summer monsoon and tropical Pacific SSTs. Part I: Roles of the subtropical ridge. J. Climate, 13 , 4310–4325.
Chang, C-P., Y. Zhang, and T. Li, 2000b: Interannual and interdecadal variations of the East Asian summer monsoon and tropical Pacific SSTs. Part II: Meridional structure of the monsoon. J. Climate, 13 , 4326–4340.
Chen, L-X., Q-G. Zhu, and H-B. Luo, 1991: East Asian Monsoons. Chinese Meteorology Press, 362 pp.
Cheng, Y., U. Lohmann, J. Zhang, Y. Luo, Z. Liu, and G. Lesins, 2005: Contribution of changes in sea surface temperature and aerosol loading to the decreasing precipitation trend in Southern China. J. Climate, 18 , 1381–1390.
Dai, A., and M. L. Wigley, 2000: Global patterns of ENSO-induced precipitation. Geophys. Res. Lett., 27 , 1283–1286.
Ding, Y., 2004: Seasonal march of the East-Asian summer monsoon. East Asian Monsoon, C.-P. Chang, Ed., World Scientific, 3–53.
Ding, Y., Z. Wang, and Y. Sun, 2007: Inter-decadal variation of the summer precipitation in East China and its association with decreasing Asian summer monsoon. Part I: Observed evidences. Int. J. Climatol., 28 , 1139–1161. doi:10.1002/joc.1615.
Fan, Y., and H. Van den Dool, 2004: Climate Prediction Center global monthly soil moisture data set at 0.5° resolution for 1948 to present. J. Geophys. Res., 109 , D10102. doi:10.1029/2003JD004345.
Fan, Y., and H. Van den Dool, 2008: A global monthly land surface air temperature analysis for 1948–present. J. Geophys. Res., 113 , D01103. doi:10.1029/2007JD008470.
Gibson, J. K., P. W. Kållberg, S. Uppala, A. Hernandez, A. Nomura, and E. Serrano, 1997: ERA description. ECWMF Re-Analysis Project Report Series 1, 72 pp.
Gong, D-Y., and C-H. Ho, 2002: Shift in the summer rainfall over the Yangtze River valley in the late 1970s. Geophys. Res. Lett., 29 , 1436. doi:10.1029/2001GL014523.
He, J-H., P. Zhao, C-W. Zhu, R-H. Zhang, X. Tang, L-X. Chen, and X-J. Zhou, 2008: Discussion of some problems as to the East Asian subtropical monsoon. Acta Meteor. Sin., 22 , 419–434.
Hu, Z-Z., 1997: Interdecadal variability of summer climate over East Asia and its association with 500 hPa height and global sea surface temperature. J. Geophys. Res., 102 , (D16). 19403–19412.
Hu, Z-Z., S. Yang, and R. Wu, 2003: Long-term climate variations in China and global warming. J. Geophys. Res., 108 , 4614. doi:10.1029/2003JD003651.
Huang, R., and Y. Wu, 1989: The influence of ENSO on the summer climate change in China and its mechanism. Adv. Atmos. Sci., 6 , 21–33.
Li, C., and M. Yanai, 1996: The onset and interannual variability of the Asian summer monsoon in relation to land-sea thermal contrast. J. Climate, 9 , 358–375.
Liu, B., M. Xu, M. Henderson, and Y. Qi, 2005: Observed trends of precipitation amount, frequency, and intensity in China, 1960–2000. J. Geophys. Res., 110 , D08103. doi:10.1029/2004JD004864.
Ose, T., 1996: The comparison of the simulated response to the regional snow mass anomalies over Tibet, eastern Europe, and Siberia. J. Meteor. Soc. Japan, 74 , 845–866.
Rayner, N. A., E. B. Horton, and D. E. Parker, 1996: Version 2.2 of the global sea-ice and sea surface temperature data set, 1903–1994. Hadley Centre Climate Research Tech. Note 74, 1–21.
Ren, G., H. Wu, and Z. Chen, 2000: Spatial patterns of change trend in rainfall of China. Quart. J. Appl. Meteor., 11 , 322–330.
Sankar-Rao, M., K-M. Lau, and S. Yang, 1996: On the relationship between Eurasian snow cover and the Asian summer monsoon. Int. J. Climatol., 16 , 605–616.
Solomon, S., D. Qin, M. Manning, M. Marquis, K. Averyt, M. M. B. Tignor, H. L. Miller Jr., and Z. Chen, Eds.,. 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.
Tao, S-Y., Y-J. Zhao, and X-M. Chen, 1958: The relationship between Mey-yü in Far East and the behavior of circulation over Asia. Acta Meteor. Sin., 29 , 119–134.
Wang, B., and Q-H. Ding, 2006: Changes in global monsoon precipitation over the past 56 years. Geophys. Res. Lett., 33 , L06711. doi:10.1029/2005GL025347.
Wang, Z., and P. Zhai, 2003: Variation of drought over northern China during 1950–2000. J. Geogr. Sci., 13 , 480–487.
Webster, P. J., and S. Yang, 1992: Monsoon and ENSO: Selectively interactive systems. Quart. J. Roy. Meteor. Soc., 118 , 877–926.
Weng, H., K-M. Lau, and Y. Xue, 1999: Multi-scale summer rainfall variability over China and its long-term link to global sea surface temperature variability. J. Meteor. Soc. Japan, 77 , 845–857.
Wilks, D., 2005: Statistical Methods in the Atmospheric Sciences. 2nd ed. Academic Press, 592 pp.
Xu, Q., 2001: Abrupt change of the mid-summer climate in central east China by the influence of atmospheric pollution. Atmos. Environ., 35 , 5029–5040.
Yang, F., and K-M. Lau, 2004: Trend and variability of China precipitation in spring and summer: Linkage to sea-surface temperatures. Int. J. Climatol., 24 , 1625–1644.
Yang, S., and L. Xu, 1994: Linkage between Eurasian winter snow cover and regional Chinese summer rainfall. Int. J. Climatol., 14 , 739–750.
Ye, T-Z., and Y-X. Gao, 1979: The Meteorology of the Qinghai-Xizang (Tibet) Plateau. Science Press, 278 pp.
Yu, R., B. Wang, and T. Zhou, 2004: Tropospheric cooling and summer monsoon weakening trend over East Asia. Geophys. Res. Lett., 31 , L22212. doi:10.1029/2004GL021270.
Zhai, P., X. Zhang, and H. Wan, 2005: Trends in total precipitation and frequency of daily precipitation extremes over China. J. Climate, 18 , 1096–1108.
Zhang, Y., T. Li, and B. Wang, 2004: Decadal change of the spring snow depth over the Tibetan Plateau: The associated circulation and influence on the East Asian summer monsoon. J. Climate, 17 , 2780–2793.
Zhao, P., and X-J. Zhou, 2006: Decadal variability of rainfall persistence time and rainbelt shift over eastern China in recent 40 years. J. Appl. Meteor. Sci., 17 , 548–556.
Zhao, P., X-D. Zhang, X-J. Zhou, M. Ikeda, and Y-H. Yin, 2004: The sea ice extent anomaly in the North Pacific and its impact on the East Asian summer monsoon rainfall. J. Climate, 17 , 3434–3447.
Zhao, P., R-H. Zhang, J-P. Liu, X-J. Zhou, and J-H. He, 2007a: Onset of southwesterly wind over eastern China and associated atmospheric circulation and rainfall. Climate Dyn., 28 , 797–811.
Zhao, P., Y-N. Zhu, and R-H. Zhang, 2007b: An Asia-Pacific teleconnection in summer tropospheric temperature and associated Asian climate variability. Climate Dyn., 29 , 293–303.
Zhao, Z., and L. Xu, 2002: Potential impact on El Niño events on the circulation and climate variation in China. Wea. Climate, 1 , 109–118.
Spatial distribution of 693 rain gauge stations over China.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Anomalies of the annual GSAT from the ECMWF reanalysis (solid line) and the annual area-weighted temperature (HadISST SST over oceans and CPC 2-m air temperature over land; dotted line) for 1958–2001 (units: °C).
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
(a) Climatology of the annual accumulated rainfall (mm) during 1960–2001 and (b) its standard deviation (mm). (c) As in (a), but for the JJA accumulated rainfall. (d) As in (b) but for the JJA accumulated rainfall.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Composite daily rainfall (mm day−1) for the cold (green bars) and warm (red line) periods averaged over (a) southern China (25°–30°N, 115°–120°E) and (b) northern China (35°–40°N, 115°–120°E), and composite differences in daily rainfall (mm day−1) over (c) southern and (d) northern China between the warm and cold periods (warm minus cold).
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Composite differences in the accumulated rainfall (mm) between the warm and cold periods. (a) The period from 16 to 30 Mar (representing the earlier beginning period of rainy season in southern China); (b) from 30 Jun to 7 Jul (representing the later ending period of rainy season in southern China); (c) from 24 Jun to 5 Jul (representing the later beginning period of rainy season in northern China) and for 21–22 Aug (representing the earlier ending period of rainy season in northern China); and (d) for JJA. Light and heavy shadings represent the negative and positive anomalies exceeding the 90% confidence level, respectively.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Time series of total rainy-season precipitation (mm; solid line) and their averages (dashed line) over 1960–79 and 1982–2001, respectively for (a) southern China and (b) northern China. Absolute values of rainy-season precipitation difference (mm) between the warm and cold periods for the observation (dashed line) and 100-time Monte Carlo simulations (solid line) are shown for (c) southern China and (d) northern China.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Time–latitude cross sections of daily rainfall (mm day−1) along 115°–120°E for (a) cold and (b) warm periods.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
(a) Composite pattern of the ECMWF 850-mb geopotential height (×10 m) averaged for 16–30 Mar (representing the earlier beginning period of rainy season in southern China) in the cold (green lines) and warm (red lines) periods. (b) As in (a), but for the difference between the warm and cold periods, in which light and heavy shadings represent the negative and positive anomalies exceeding the 90% confidence level, respectively. (c) As in (b), but for the composite difference in 850-mb winds (m s−1), in which thick vectors (m s−1) exceed the 90% confidence level and “A” and “C” indicate the positions of anomalous anticyclonic and cyclonic centers, respectively. The thick dash lines measure the topography of 1500 m.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
As in Fig. 8, but for the time span from 30 Jun to 7 Jul (representing the later ending period of rainy season in southern China).
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Composite differences in (a) JJA SLP (mb) and (b) 850-mb winds (m s−1) between the warm and cold periods for the ECMWF reanalysis. In (a), light and heavy shadings represent the negative and positive anomalies exceeding the 90% confidence level, respectively. In (b), thick vectors (m s−1) exceed the 90% confidence level and “A” indicates the position of anomalous anticyclonic center.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
(a) Latitude–height cross section of composite difference in the ECMWF JJA θse (K) along 115°–120°E between the warm and cold periods; (b) as in (a), but for JJA vertical p velocity (×10−2 Pa s−1) along 115°–120°E. The black shadings denote topography and the light and heavy shadings represent the negative and positive anomalies exceeding the 90% confidence level, respectively.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
(a) Composite difference in the ECMWF JJA land surface temperature (°C) between the warm and cold periods. (b) As in (a), but for JJA 500-mb temperature (°C). (c) As in (a), but for the HadISST JJA SST (°C). Light and heavy shadings represent the negative and positive anomalies exceeding the 90% confidence level, respectively.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
As in Fig. 12a, but for (a) latitude–height cross section of JJA temperature along 70°–100°E, and (b) longitude–height cross section of JJA temperature along 32.5°N for the ECMWF reanalysis. The black shadings in (a) and (b) denote topography.
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
As in Fig. 12a, but for (a) the ECMWF winter–spring snow depth (×10 mm), (b) ECMWF spring–summer soil (0–7 cm) moisture content (×0.01, unit: fraction), and (c) CPC spring–summer soil moisture content (mm).
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Curves of the winter–spring accumulated snowfall (mm; solid lines) and its linear trend (dashed lines) at Wuqiatuoyun (51701), Naqu (55299), Lhasa (55591), Yadong (55773), and Suoxian (56106).
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Curves of anomalies of vertically (500–200 mb) and regionally (over TP where the altitude is above 1500 m) averaged January–February temperature (°C) for 1960–2001 (Solid line: ECMWF reanalysis; dotted line: NCEP reanalysis).
Citation: Journal of Climate 23, 6; 10.1175/2009JCLI2660.1
Identification numbers (ID), names, latitudes, longitudes, and elevations (m) above the sea level for stations over TP.
