Interdecadal Variability of Summer Precipitation in Northwest China and Associated Atmospheric Circulation Changes

中国西北地区夏季降水的年代际变化及相关大气环流成因分析

+ Author Affiliations + Find other works by these authors
Funds: 
Supported by the National Key Research and Development Program of China (2017YFA0605004) and National Natural Science Foundation of China (41775082 and 41975100)

PDF

  • Daily precipitation data from 149 rain gauge stations in China and NCEP–NCAR reanalysis data during 1961–2018 are used to investigate the interdecadal variability of summer precipitation in Northwest China and related causes. The results suggest that, on the interdecadal timescale, Northwest China shifts into a rainy period from the year 1987, with an increase in the precipitation amount and intensity; an increase in the probability of moderate rain, heavy rain, torrential rain, and extremely heavy rain; and a decrease in the probability of light rain. More than 60% of the increase in precipitation can be attributed to rainfall with intensity above the grade of heavy rain. The associated interdecadal variability of atmospheric circulations over midlatitude Eurasia in summer is examined and it is found that the interdecadal variability is mainly characterized by the Silk Road pattern (SRP), with a cyclonic circulation anomaly and an anticyclonic circulation anomaly over central Asia and Mongolia, respectively; enhanced ascending motion and atmospheric instability in Northwest China; and strengthened easterly winds caused by the Mongolian anticyclonic anomaly along the northern boundary of the Tibetan Plateau. On the south side of the Mongolian anticyclone, the water vapor transported from the Pacific and Indian Oceans as well as the South China Sea to Northwest China by easterly winds increases significantly, providing the main water vapor source for the increase in precipitation in Northwest China on the interdecadal timescale. The transition of the Atlantic multidecadal oscillation to a positive phase may be the main cause of the interdecadal transition of the SRP to a positive phase, resulting in the interdecadal increase in summer precipitation in Northwest China.
    利用1961–2018年中国西北地区共149个雨量站日降水量资料和NCEP–NCAR再分析资料,研究了中国西北地区夏季降水的年代际变化特征及其成因。发现夏季降水量于1987年发生突变并进入多雨期后,降水量增加、降水强度增强,小雨概率减少,中雨、大雨、暴雨、大暴雨、特大暴雨概率均增加,降水量增加的60%以上是由大雨以上的强降雨带来的。夏季大气环流的年代际变化的主要特征是:中纬度欧亚地区上空存在丝绸之路波列(SRP),中亚和蒙古分别是气旋环流异常和反气旋环流异常,西北地区大气上升运动及层结不稳定性增强,蒙古反气旋异常增强了青藏高原北缘的东风。蒙古反气旋南侧,通过东风输送到西北地区的太平洋、印度洋及南海的水汽显著增加,并成为降水量年代际增加的主要水汽来源。大西洋多年代际振荡(AMO)向正位相的年代际过渡可能是SRP向正位相年代际过渡的主要原因,从而导致西北地区降水年代际增加。
  • Northwest China, located in central Asia, covers a vast territory with complex topography, including mountains, plateaus, basins, and deserts. Due to its long distance from the ocean and the blocking of humid airflow by plateaus and mountains, this region receives relatively low precipitation overall, making it a typical arid/semiarid region (Yang et al., 2003). In addition, Northwest China is an essential section of the ancient Silk Road and has a high political and economic status under the “One Belt and One Road” strategy of China. Since the lack of water resources seriously restricts regional socioeconomic development and ecological construction, the variations of precipitation in Northwest China have long been a concern of governments at various levels and research communities far and near.

    In the early 21st century, Chinese scientists took the lead in proposing that, since the 1980s, Northwest China has experienced a climate transition from warm–dry to warm–wet (Shi et al., 2002, 2003). Since then, many scholars have carried out a series of studies on the dry–wet variability in Northwest China. Precipitation plays a crucial role in the dry–wet evolution, and there are large-scale, accurate observations of precipitation in Northwest China. On this basis, numerous studies have been conducted to analyze the trend of the variation in precipitation over Northwest China, which have revealed that, annually and seasonally, precipitation has shown an increasing trend since the mid-1980s (Li et al., 2011; Geng et al., 2016; Li et al., 2016; Ren et al., 2016; Wu et al., 2019; Li and Liu, 2020; Su et al., 2020; Wang C. H. et al., 2021). Also, several studies have indicated that the trends of variation in precipitation are different in the eastern and western parts of Northwest China, i.e., an increasing trend in the west and a decreasing trend in the east (Chen and Dai, 2009; Geng et al., 2016; Liu et al., 2017; Zhang et al., 2019).

    As an essential condition for precipitation, water vapor is a basic physical factor for analyzing precipitation variations. At present, there are different viewpoints on the water vapor sources for the interdecadal increase in precipitation. Studies indicate that there is a westerly climate zone to the west of 102.5°E (Lu et al., 2021), and that after 1987 the transport of water vapor by the westerly wind belt to Northwest China increased, resulting in an increase in precipitation (Dai and Zhang, 2012). However, correlation analysis shows that the increase in precipitation over Xinjiang after 1987 may be related to the stronger intensity and the westward location of the East Asia–Pacific teleconnection, and a significant enhancement of anomalous water vapor transport from the Northwest Pacific to the inland areas of China (Chen et al., 2012). Nevertheless, according to recent studies, the transport of water vapor from the Indian Ocean and the South China Sea to the arid regions of central Asia by easterly winds along the northern edge of the Tibetan Plateau has been the main source of the increase in precipitation since the mid-1980s, and that there is no transport of water vapor from the Pacific Ocean (Chen et al., 2021). Therefore, whether water vapor from the Pacific Ocean can be transported to Northwest China in summer and how it varies remains an open question. In other words, determining if water vapor from the Pacific Ocean is related to the interdecadal increase in precipitation over Northwest China is worthy of continued study.

    Precipitation variability is closely related to the evolution of atmospheric circulations. The Silk Road pattern (SRP), also known as the global teleconnection pattern, is the predominant teleconnecting wave train in Eurasia during summer (Krishnan and Sugi, 2001; Enomoto, 2004; Hong et al., 2017; Wang et al., 2017). It propagates eastward along the westerly jet and can affect the spatial distribution of summer precipitation on the interdecadal timescale in midlatitude Eurasia (Huang et al., 2015a). Correlation analysis shows that the interannual variations of both the July precipitation in Northwest China and the summer precipitation in the Tarim Basin are affected by the SRP (Chen et al., 2012; Huang et al., 2015b). Nonetheless, few studies have been conducted on the impact of the SRP on the interdecadal variation of precipitation in Northwest China, and little attention has been paid to the correlation between heavy precipitation weather events and the SRP.

    Presently, there is no agreement on the SRP’s generation and maintenance mechanisms. One viewpoint is that the SRP is a result of the internal dynamics of the atmosphere, whilst another perspective is that the SRP is caused by external forcing factors such as the non-adiabatic heating in the eastern Mediterranean, the Indian summer monsoon region (Wang et al., 2017), the northern Indian Ocean (Chen and Huang, 2012), and the tropical Atlantic (Sun et al., 2019). It has also been suggested that Atlantic multidecadal oscillation (AMO) might regulate the interdecadal variation of the SRP, which causes zonal heterogeneous heating over Eurasia (Hong et al., 2017) and affects the climate in Northeast Asia (Si et al., 2021). Indeed, it has been reported that the AMO is very likely a driver of the interdecadal variations of the SRP, although the linear relationship between indices of the SRP and AMO is weak on the interdecadal timescale (Wang et al., 2017).

    Based on the above, the aim of this paper is to analyze the impact of Pacific water vapor transport on the interdecadal increase of precipitation in Northwest China, as well as the influence of the SRP on the interdecadal variations of precipitation and heavy precipitation events in this region, and the interdecadal relationship between the AMO and SRP. Our hope is that this study will provide a useful reference for water resources planning and management in Northwest China.

    The remainder of the paper is organized as follows. Section 2 introduces the study area, data, and methods used. In Section 3, the interdecadal variability of summer precipitation in Northwest China is analyzed in terms of the regional average, spatial distribution, and intensity of precipitation. The possible causes of the interdecadal variations of summer precipitation in Northwest China are illustrated in Section 4, including an analysis of the interdecadal variations of related atmospheric circulations and associated water vapor flux, the effects of the SRP on the interdecadal changes in precipitation, and the links with the SRP and AMO. Finally, Section 5 summarizes our main conclusions and provides some further discussion.

    Based on the trend of variation in summer precipitation from 1961 to 2018 from 259 rain gauge stations in the region west of 110°E and north of 32.5°N in China (figure omitted), it is found that the precipitation at most of the stations to the west of 103°E shows an increasing trend, whereas a decreasing trend is apparent to the east of 103°E. Therefore, the area to the west of 103°E is selected as the main study area (32.5°–50°N, 72.5°–103°E). This study area includes Xinjiang, Qinghai, the Hexi Corridor in Gansu Province, and West Inner Mongolia (Fig. 1). From north to south, the study area includes high-altitude mountains (such as the Altai Mountains, Tianshan Mountains, Kunlun Mountains, and Qilian Mountains) and basins (such as Junggar Basin, Tarim Basin, and Qaidam Basin), together with deserts, Gobi, lakes, oases, and other natural landscapes.

    Fig  1.  Study area and locations of the rain gauge stations (red circles). Shading indicates the altitude (m).

    The daily precipitation data during 1961–2018 used in this study are from the Homogenized Precipitation Dataset of China National Surface Weather Stations (V1.0) (Yang and Li, 2014). According to the Meteorological Industry Standard of the People’s Republic of China, i.e., “Specifications for surface meteorological observation. Part 21: Processed of missing data and statistic of half-baked data” (QX/T 65-2007), more than seven days of missing records of precipitation on a daily basis should be considered as a missing observation for that month. Therefore, the effect of months with missing observations is excluded in this study, meaning that ultimately a total of 149 rain gauge stations (Fig. 1) are selected to analyze the interdecadal variations of precipitation in Northwest China.

    The monthly and daily meteorological data used in this study are from the NCEP–NCAR reanalysis datasets for the period January 1961–December 2018 (Kalnay et al., 1996), including geopotential height, u- and v-component winds, specific humidity, surface pressure, air temperature, and vertical velocity, at a spatial resolution of 2.5° × 2.5°. Based on these data, the atmospheric circulation anomalies and their relationship with the interdecadal variations of precipitation are analyzed. The monthly AMO index (AMOI) for the period 1961–2018 from the NOAA Physical Sciences Laboratory (https://psl.noaa.gov/data/timeseries/AMO) is used to analyze the relationship between the AMO and atmospheric circulation anomalies. The SRP index (SRPI) is the time coefficient corresponding to the first eigenvector field of an empirical orthogonal function (EOF) analysis of the 200-hPa meridional wind in the midlatitude Northern Hemisphere (20°–60°N, 30°–130°E) in summer 1961–2018 (Kosaka et al., 2009; Wang et al., 2017).

    The Mann–Kendall method is adopted to identify the timing of the abrupt change in the precipitation field, thereby enabling the delineation of the rainy and less rainy period. The mean values of meteorological variables in these two sub-periods are calculated, and the t test is used to verify whether their changes are statistically significant. The interdecadal change is divided into four categories (significant decrease, decrease, increase, and significant increase), among which a significant decrease and significant increase mean that the interdecadal change passes the t test at the 0.05 significance level. Linear regression is used to analyze the relationship between atmospheric circulations and the AMOI. Composite analysis is adopted to diagnose the characteristics of vertically integrated water vapor flux variations. The t test is also used to examine the significance of correlation coefficients. The climatic mean is the average value from 1961 to 2018, and the summer season is defined as June–July–August.

    Water vapor flux is a physical quantity that indicates the intensity of water vapor transport. Usually, the vertically integrated water vapor flux is calculated (Huang et al., 2015a; Chen et al., 2021), as shown in Eq. (1):

    $$ {\boldsymbol{Q}}=-\frac{1}{g}{\int }_{{p}_{{\rm{s}}}}^{{p}}q{\boldsymbol{V}}\mathrm{d}p , $$ (1)

    where ${\boldsymbol{V}}$, $ q $, g, ps, p, and ${\boldsymbol{Q}}$ represent the horizontal wind speed vector, specific humidity, gravitational acceleration, surface pressure, top-layer pressure, and water vapor flux, respectively.

    The wave-activity flux proposed by Takaya and Nakamura (2001) is calculated and used to analyze the characteristics of Rossby wave propagation related to the SRP. The horizontal component of this flux ($\boldsymbol{W}$) can be expressed in the pressure coordinates as follows:

    $$ \boldsymbol{W}=\frac{1}{2\left|{\boldsymbol{V}}\right|}\left(\genfrac{}{}{0pt}{}{u\left({\varPsi }_{x}^{'2}-{\varPsi }'{\varPsi }_{xx}'\right)+v\left({\varPsi }_{x}'{\varPsi }_{y}'-{\varPsi }'{\varPsi }_{xy}'\right)}{u\left({\varPsi }_{x}'{\varPsi }_{y}'-{\varPsi }'{\varPsi }_{xy}'\right)+v\left({\varPsi }_{{y}}^{{'}2}-{\varPsi }'{\varPsi }_{yy}'\right)}\right) , $$ (2)

    where ${\boldsymbol{V}}$ denotes the climatic average of the horizontal wind vector, u is the zonal component of ${\boldsymbol{V}}$, v is the meridional component of ${\boldsymbol{V}}$, and ${\varPsi }'$ is the disturbance stream function.

    The average summer precipitation in Northwest China from 1961 to 2018 is 115.1 mm, accounting for 51.3% of the annual precipitation. The results from the Mann–Kendall test indicate that there is an abrupt change in summer precipitation in 1987, i.e., the precipitation anomalies change mainly from negative to positive (figure omitted). Therefore, 1961–1986 is defined as the less rainy period (P1), 1987–2018 as the rainy period (P2), and the difference between the two as the interdecadal variation.

    The average summer precipitation anomaly in P1 and P2 is −6.0% and 5.0%, respectively, showing an interdecadal increase that is significant at the 0.05 level (Fig. 2a). The average number of summer precipitation days in P2 is 0.77 days more than in P1, but the interdecadal increase is not significant (Fig. 2b). However, the average anomaly of precipitation intensity (total precipitation amount divided by the number of days) is −0.25 mm day−1 in P1, 0.20 mm day−1 in P2, and the interdecadal increase is 0.45 mm day−1, which is statistically significant at the 0.05 level (Fig. 2c).

    Fig  2.  Time series of the anomalies of the (a) precipitation amount (%), (b) number of precipitation days, and (c) precipitation intensity (mm day−1) averaged regionally in Northwest China during 1961–2018. The straight lines indicate the average of P1 (1961–1986) and P2 (1987–2018).

    The interdecadal increase in the precipitation amount is apparent at most stations in Northwest China, but is especially significant in the Yili River valley, the surrounding area of the Tianshan Mountains, the northern margins of the Kunlun Mountains, and the Qilian Mountains region (Fig. 3a). The number of precipitation days shows an interdecadal increase in most areas, but is particularly significant in southern Xinjiang. In contrast, an interdecadal decrease in the number of precipitation days can be seen in some areas of the Hexi Corridor and eastern Qinghai (Fig. 3b). The precipitation intensity increases in most areas, but is especially significant in the Yili River valley, the surrounding areas of the Tianshan Mountains, and in the Qilian Mountains region (Fig. 3c). Overall, it can be seen that the interdecadal increase in rainfall in Xinjiang and western Qinghai is the result of a simultaneous interdecadal increase in the number of days and intensity of precipitation, while in eastern Qinghai and the Hexi Corridor, the main reason for the interdecadal increase in rainfall is the increase in precipitation intensity.

    Fig  3.  Interdecadal variations of the spatial distributions of the (a) amount, (b) number of days, and (c) intensity of summer precipitation at rain gauge stations in Northwest China.

    According to the statistics, the total number of summer precipitation days across the 149 rain gauge stations in P1 and P2 is 108,804 and 137,587, respectively, with an average of 0.77 more precipitation days in P2 per station each summer. Dividing the daily precipitation into different classes every 3 mm and then calculating the probability of the different classes of precipitation in P1 and P2, it is found that the probability of precipitation below 3 mm decreases by 2.435 × 10−2, but the probability of precipitation between 3 and 72 mm increases. Interestingly, the probability of precipitation between 72 and 96 mm decreases by 4 × 10−5 but the probability of precipitation above 96 mm increases by 2 × 10−5 (Figs. 4a, b).

    Fig  4.  (a) Probability density curve and (b) cumulative probability density curve of different intensities of summer precipitation, and (c) variations of summer precipitation at different intensity grades per station in Northwest China.

    According to the local standard characteristics of arid and semiarid areas in Northwest China, i.e., precipitation grades (DB65/T 3273-2011), the precipitation can be classified into light rain (0.1–6 mm day−1), moderate rain (6.1–12 mm day−1), heavy rain (12.1–24 mm day−1), rainstorm (24.1–48 mm day−1), large rainstorm (48.1–96 mm day−1), and extraordinary rainstorm (above 96.1 mm day−1). The results show that the probability of light rain decreases by 2.160 × 10−2, while the probability of moderate rain, heavy rain, rainstorm, large rainstorm, and extraordinary rainstorm increases by 1.046 × 10−2, 7.73 × 10−3, 3.17 × 10−3, 2.4 × 10−4, and 2 × 10−5, respectively. The maximum value of extraordinary rainstorm increases from 120.4 mm in P1 to 131.7 mm in P2 (Figs. 4a, b). The average summer precipitation increases by 13.2 mm per station, within which light rain, moderate rain, heavy rain, and precipitation above the rainstorm grade increase by 1.5, 3.6, 4.5, and 3.6 mm, respectively (Fig. 4c). Although the probability of light rain decreases, the intensity of light rain increases, so its amount also increases. In addition, the contribution of precipitation above the grade of heavy rain to the total increase in precipitation reaches 61.4%. This finding is similar to the results reported by Wang et al. (2020), who concluded that more than half of the increase in precipitation in Xinjiang during 1986–2019 compared with 1961–1985 was contributed by extreme precipitation. The suggestion, therefore, is that extreme precipitation is increasing with global warming (Papalexiou and Montanari, 2019; Lu et al., 2021).

    The difference in the atmospheric circulation field between P2 and P1 is referred to as the abnormal field of atmospheric circulation. The geopotential height anomalies at 500 hPa are presented in Fig. 5a, and the results show that at the middle latitudes (30°–60°N) there is an apparent positive abnormal zone of geopotential height from Europe to the western Pacific Ocean, with positive abnormal centers exceeding 30 and 40 gpm over eastern Europe and Mongolia, respectively. Although the geopotential height anomalies over central Asia are positive, they are weaker. This distribution indicates that high pressure ridges over the Caspian Sea and Mongolia are stronger during the rainy period, and the trough over central Asia is relatively deep. The significantly stronger Mongolian high pressure ridge is conducive to a slowing down of the movement of the central Asian weather system and an increase in the lag time. The distribution of wind vector anomalies (Fig. 5a) shows that circulation anomalies with alternating anticyclones and cyclones at middle latitudes occur from eastern Europe, to central Asia, to Mongolia, and to East China–Japan. In the far east of Asia, the anomalous cyclonic circulation over Japan brings the southeasterly airflow from the Pacific Ocean to Northeast China, and this southeasterly flow turns to a northeasterly flow along the east side of the Mongolian anomalous anticyclone, which moves air southward to the northern edge of the Tibetan Plateau. Subsequently, it turns to an easterly or southeasterly airflow to affect Northwest China (Fig. 5a).

    Fig  5.  Difference in geopotential height (contours; gpm) and wind (arrows; m s−1) between P2 and P1 at (a) 500 hPa and (b) 200 hPa, along with altitude–longitude profiles of the difference in (c) vertical velocity (10−2 Pa s−1) and (d) air temperature (°C) averaged in the range of 32.5°–50°N between P2 and P1 in summer. Values exceeding the 95% confidence level are shaded, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    Compared with the situation at 500 hPa, we find that the anomalies of geopotential height and wind in the rainy period at 200 hPa are characterized by stronger positive geopotential height anomalies over the high-pressure ridge areas of the Caspian Sea and Mongolia than over the low-trough area of central Asia, resulting in a more significant deepening of the central Asian trough. In middle latitudes, circulation anomalies with alternating anticyclones and cyclones can be found from eastern Europe, to central Asia, to Mongolia, and to Japan, within which the cyclonic anomaly over central Asia is significantly enhanced, and the wave train pattern of wind anomalies is more obvious. A significant cyclonic circulation anomaly occurs over central Asia on the north side of the South Asian high, indicating that the westerly circulation at the bottom of the central Asian low trough is stronger during the rainy period in summer compared with that in the less rainy period, which is conducive to the enhancement of convergence and ascending motion in the middle and lower troposphere (Fig. 5b).

    Further analyzing the variations of the anomalies of the average vertical velocity and air temperature by longitude and altitude in the range 32.5°–50°N during the rainy period, we find significant negative anomalies of vertical velocity in Northwest China at 700–200 hPa, suggesting that the enhanced ascending motion is favorable for water vapor condensation and the development of precipitation weather. At 150 hPa, the ascending motion is obviously weakened (Fig. 5c). In terms of the air temperature during the rainy period, positive anomalies appear in the middle and lower troposphere in Northwest China, while apparent negative anomalies occur above 250 hPa, indicating that the instability of the atmospheric stratification is enhanced, which is favorable for convective weather formation (Fig. 5d).

    Overall, the main characteristics of the atmospheric circulation anomalies during the rainy period are that circulation anomalies with alternating anticyclones and cyclones occur at the middle latitudes from eastern Europe, to central Asia, to Mongolia, and to Japan. The anomalous cyclones over central Asia strengthen the central Asian trough, meaning weather systems are more likely to generate, develop, and move eastward to affect Northwest China. In addition, the variations of vertical velocity and temperature stratification are all conducive to precipitation weather in Northwest China.

    The water vapor flux anomalies during the rainy period (Fig. 6) indicate that the wave train pattern is clear at middle latitudes. In addition, the circulation anomalies with alternating anticyclones and cyclones appear from eastern Europe, to central Asia, to Northwest China–Mongolia, and to East China–Japan, which is similar to the 500-hPa circulation anomalies (Fig. 5a). The airflow from the anomalous cyclone over Japan carrying water vapor from the Pacific Ocean to Northeast China turns into a northeasterly airflow. Then, this northeasterly airflow moves southward to the northern edge of the Tibetan Plateau along the east side of the Mongolian anomalous anticyclone, and subsequently turns into easterly or southeasterly airflow to affect Northwest China. Also, there is pronounced water vapor convergence in Xinjiang, the Hexi Corridor, and eastern Qinghai. The anomalous cyclone over central Asia transports water vapor from the Arabian Sea and Indian Ocean northward to central Asia along the eastern side of the Iranian Plateau; however, little water vapor is transported to Northwest China. Therefore, the water vapor resulting in the increase in precipitation in Northwest China derives primarily from the Pacific Ocean, transported by easterly winds (Fig. 6). This is different to the conclusion reached by Chen et al. (2021), which is that the increase in precipitation over central Asia since the mid-1980s is mainly due to the increase in water vapor from the Indian Ocean and the South China Sea transported by easterly winds caused by the weakening of the East Asian summer monsoon, the westward extension of the western Pacific subtropical high, and the strengthening of the Mongolian anticyclone.

    Fig  6.  Difference in vertically integrated water vapor flux (arrows; kg m−1 s−1) and water vapor flux divergence (shadings; 10−6 kg m−2 s−1) between P2 and P1 in summer. The green dotted area and the arrows exceed the 95% confidence level, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    The average vertically integrated water vapor flux divergence from the surface to 300 hPa in summer in Northwest China decreases from 0.004 × 10−5 kg m−2 s−1 in P1 to −0.389 × 10−5 kg m−2 s−1 in P2, indicating an increase in moisture convergence in P2, which is a statistically significant difference at the 0.05 level (Fig. 7a). Ascending motion facilitates the condensation of water vapor and rainfall. The average vertical velocity at 700 hPa in summer in Northwest China decreases from −3.5 × 10−2 Pa s−1 in P1 to −4.0 × 10−2 Pa s−1 in P2, which again is a statistically significant difference at the 0.05 level. That is, ascending motion enhances significantly in P2, which is conducive to the increase in precipitation (Fig. 7b). The interdecadal variability signals can be characterized by the time series being processed with a 9-yr running mean (Zhou and Huang, 2010; Sun et al., 2019). On this basis, the interdecadal correlation coefficients between the precipitation and the average vertically integrated water vapor flux divergence and average vertical velocity in Northwest China during 1961–2018 are −0.79 and −0.73, respectively, both of which are statistically significant at the 0.01 level.

    Fig  7.  Variations of atmospheric circulation factors during 1961–2018: (a) average integrated water vapor flux divergence from the surface to 300 hPa, (b) average vertical velocity at 700 hPa, (c) easterly water vapor transport days (EWDs; black) and strong EWDs (SEWDs; red) from the surface to 700 hPa, (d) easterly water vapor transport flux (EWVTF; black) and strong EWVTF (SEWVTF; red) from the surface to 700 hPa, (e) EWDs (black) and SEWDs (red) from the surface to 300 hPa, and (f) EWVTF (black) and SEWVTF (red) from the surface to 300 hPa.

    In terms of the interdecadal variability of water vapor flux across the four borders of Northwest China and its net water vapor flux, the water vapor input across the western border increases from 8.25 × 107 kg s−1 in P1 to 8.31 × 107 kg s−1 in P2, i.e., by 0.06 × 107 kg s−1. Meanwhile, the water vapor input across the southern border increases by 0.5 × 107 kg s−1 from 0.72 × 107 kg s−1 in P1 to 1.22 × 107 kg s−1 in P2; the water vapor input across the northern border decreases by 2.8 × 107 kg s−1 from 2.99 × 107 kg s−1 in P1 to 0.19 × 107 kg s−1 in P2; and the water vapor output across the eastern border decreases by 3.44 × 107 kg s−1 from 12.39 × 107 kg s−1 in P1 to 8.95 × 107 kg s−1 in P2. This is similar to the conclusion reached by Ren et al. (2016) that the input of water vapor from the western, northern, and southern borders has not increased significantly since 1979 through Northwest China, while the output of water vapor from the eastern border has decreased significantly. The net water vapor flux in the region increases by 1.2 × 107 kg s−1 from a deficit of 0.43 × 107 kg s−1 in P1 to a surplus of 0.77 × 107 kg s−1 in P2. The easterly water vapor input across the eastern border increases by 3.44 × 107 kg s−1, accounting for 86% of the total increase in water vapor, and is therefore the main source of water vapor for the interdecadal increase in summer precipitation (figure omitted).

    The daily integrated water vapor fluxes from the surface to 700 hPa and from the surface to 300 hPa in summer from 1961 to 2018 are calculated to investigate the interdecadal variability of the easterly transport of water vapor from the Pacific Ocean, Indian Ocean, and South China Sea to Northwest China. Our study (Fig. 6) and Chen et al. (2021) both find that the transport of water vapor to central Asia by easterly winds takes place mainly along a narrow area in the northern margins of the Tibetan Plateau. Thus, two types of events are defined: easterly water vapor transport and strong easterly water vapor transport. An easterly water vapor transport day (EWD) is defined when the average zonal water vapor flux at five grid points at the eastern border (32.5°–42.5°N, 102.5°E) in Northwest China is less than 0. The easterly water vapor transport flux (EWVTF) is defined as the sum of the water vapor fluxes transported by the easterly wind at the eastern border on all EWDs each summer. Likewise, a strong easterly water vapor transport day (SEWD) is defined when the average zonal water vapor flux is less than 0 at five grid points at the border (32.5°–42.5°N, 102.5°E), and meanwhile the zonal water vapor flux at eight grid points (35°N, 102.5°E to 35°N, 120°E) are all less than 0, i.e., a day when water vapor from the Pacific Ocean can be transported to Northwest China. The strong easterly water vapor transport flux (SEWVTF) is the sum of the water vapor fluxes transported by the easterly wind at the eastern border on all SEWDs each summer.

    Next, we examine the easterly water vapor transport from the surface to 700 hPa. According to our calculation criteria, there are 2315 EWDs and 604 SEWDs in the summers of 1961–2018. Composite analysis of the 2315 EWDs’ water vapor flux fields shows that some of the monsoon water vapor from the Indian Ocean and the South China Sea is transported northward along the west side of the western Pacific subtropical high to the south side of the Mongolian anticyclone, and then westward along the northern margin of the Tibetan Plateau to Northwest China (Fig. 8a). This is similar to the vapor flux field composited from 2481 easterly days determined according to the 700-hPa zonal winds during 1958–2019 (Chen et al., 2021). However, according to the composite analysis of the 604 SEWDs’ water vapor flux fields, it can be seen that some of the monsoon water vapor from the Pacific Ocean, Indian Ocean, and South China Sea is transported northward to the south side of the Mongolian anticyclone, and then westward along the northern margin of the Tibetan Plateau to Northwest China. At the same time, the monsoon is obviously suppressed, and the geographical location of the northern edge of the monsoon water vapor reaches 42.5°N in Northeast China (Fig. 8b). In terms of interdecadal variations, the number of EWDs increases significantly from 27.4 in P1 to 44.8 in P2 each summer, which is an increase of 63.5%. The number of SEWDs, meanwhile, increases significantly from 6.3 in P1 to 13.7 in P2 each summer, which is an increase of 117.5%, both of which pass the 0.01 reliability test (Fig. 7c). EWVTF increases from 36.2 × 107 kg s−1 in P1 to 67.3 × 107 kg s−1 in P2 each summer, which is a significant increase of 85.9%. SEWVTF, meanwhile, also increases significantly each summer, from 9.3 × 107 kg s−1 in P1 to 22.9 × 107 kg s−1 in P2, which is an increase of 146.2%, both of which pass the 0.01 reliability test (Fig. 7d). The interdecadal correlation coefficients between summer precipitation and EWVTF/SEWVTF during 1961–2018 in Northwest China are 0.79/0.84, indicating that precipitation is closely related to the interdecadal increase in EWVTF and SEWVTF.

    Fig  8.  Composite analysis of daily integrated water vapor flux (arrows; kg m−1 s−1) and divergence of water vapor flux (shadings; 10−5 kg m−2 s−1) from the surface to 700 hPa on (a) EWDs and (b) SEWDs, and from the surface to 300 hPa on (c) EWDs and (d) SEWDs, in summer 1961–2018. The green dotted area and the arrows exceed the 95% confidence level, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    In terms of the easterly water vapor transport from the surface to 300 hPa, according to our calculation criteria, there are 706 EWDs and 133 SEWDs in the summers of 1961–2018. Composite analysis of the 706 EWDs’ water vapor flux fields clearly shows that some of the monsoon water vapor from the Indian Ocean and the South China Sea is transported northward along the west side of the western Pacific subtropical high to the south side of the Mongolian anticyclone, and then westward along the northern margin of the Tibetan Plateau to Northwest China. Meanwhile, the northerly flow to the east side of the Mongolian anticyclone prevents the northward advancement of the monsoon water vapor (Fig. 8c). After compositing the water vapor flux fields of the 133 SEWDs, it is clear that some of the monsoon water vapor from the Pacific Ocean, Indian Ocean, and South China Sea is transported northward to the south side of the Mongolian anticyclones, and then westward along the northern margin of the Tibetan Plateau to Northwest China. Due to the strong Mongolian anticyclone, the monsoon water vapor is suppressed and cannot affect Northeast China (Fig. 8d). Analyzing the interdecadal variation, we find that the number of EWDs increases significantly from 6.5 in P1 to 16.8 in P2 each summer, which is an increase of 158.5%. The number of SEWDs, meanwhile, increases from 0.9 in P1 to 3.4 in P2 each summer, which is an increase of 277.8%, and both pass the 0.01 reliability test (Fig. 7e). EWVTF increases by 199.3% from 14.6 × 107 kg s−1 in P1 to 43.7 × 107 kg s−1 in P2 each summer, and SEWVTF increases by 455.0% from 2.2 × 107 kg s−1 in P1 to 12.3 × 107 kg s−1 in P2, both of which pass the 0.01 reliability test (Fig. 7f). The interdecadal correlation coefficients between summer precipitation and EWVTF/SEWVTF from 1961 to 2018 in Northwest China are 0.61/0.75, respectively, both of which pass the 0.01 reliability test. This indicates that the interdecadal variation of precipitation is also closely related to EWVTF and SEWVTF.

    In summary, the interdecadal variations of water vapor flux also show the Rossby wave train pattern at the middle latitudes in Eurasia. The vertical ascending motion and vertically integrated water vapor convergence increase significantly along the northern margin of the Tibetan Plateau, meaning the transport of water vapor by easterly winds also increases significantly, which is closely related to the interdecadal increase in precipitation in Northwest China. The water vapor transported to Northwest China by the easterly winds includes water vapor from the Indian Ocean and the South China Sea, as well as from the Pacific Ocean, which differs to the findings of Chen et al. (2021). The enhancement of the easterly transport of water vapor is related to the easterly flow in the south of the Mongolian anticyclone. According to the analysis in Section 4.1, the anomalous Mongolian anticyclone of the SRP is beneficial to the strengthening of the Mongolian anticyclone and the easterly transport of water vapor. Meanwhile, the northerly flow to the east side of the Mongolian anticyclone prevents the northward advancement of the East Asian summer monsoon (Zhu et al., 2018), and this indicates that the SRP has an important influence on the East Asian summer monsoon (Chowdary et al., 2019; Wang L. et al., 2021). Some of the northward monsoon water vapor reaches the southern side of the Mongolian anticyclone and then turns to the west to Northwest China. Thus, the interdecadal increase in monsoon water vapor transported from the Pacific Ocean, Indian Ocean, and South China Sea to Northwest China is related to depression of the East Asian summer monsoon. These findings extend the range of water vapor transported westward by the East Asian summer monsoon.

    During summer in the rainy period, both the atmospheric circulation anomalies and water vapor flux anomalies indicate the existence of a teleconnecting wave train from eastern Europe to Japan at middle latitudes, and its distribution pattern is similar to that of the SRP. Based on the definition of the SRP index (SRPI) in Section 2.2, an EOF decomposition of the 200-hPa meridional wind field during 1961–2018 is carried out. The first mode (Fig. 9a) presents a wave train structure in which the meridional wind direction turns northerly–southerly–northerly every 30° of longitude from west to east, passing North’s significance test (North et al., 1982). In contrast, the interdecadal difference fields of the 200-hPa meridional wind and vorticity during summer in Northwest China (Fig. 9b) show that the midlatitude meridional wind field also presents a northerly–southerly–northerly turning of the meridional wind direction every 30° of longitude, and that there are positive–negative–positive vorticity centers over central Asia, Mongolia, and eastern China–Japan. The wave train structure is quite similar to that of the SRP. The interdecadal correlation coefficient between the summer precipitation and the summer SRPI from 1961 to 2018 reaches 0.54, passing the 0.05 significance test level, which indicates that the interdecadal variation of summer precipitation in Northwest China is associated with the SRP.

    Fig  9.  (a) First mode of the EOF decomposition of the 200-hPa meridional wind field in the range of 20°–60°N, 30°–130°E during summer from 1961 to 2018, and (b) the difference fields of the meridional wind (shadings; m s−1) and vorticity (contours; 10−6 s−1) at 200 hPa between P2 and P1 in summer, in which values exceeding the 95% confidence level are stippled. The box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    According to the results reported in Section 3.3, more than 60% of the increase in precipitation is contributed by rainfall above the grade of heavy rain, and there is an interdecadal increase in the probability of heavy precipitation. Therefore, the relationship between the atmospheric circulation anomalies and the SRP during heavy precipitation events in summer in Northwest China is further analyzed. The multiyear average from 1961 to 2018 is defined as the climatic average to calculate the daily anomalies of 200-hPa geopotential height in summer (1 June–31 August). The day with the largest domain-averaged precipitation per year in Northwest China is defined as day 0. The geopotential height anomalies for 9 days before and after day 0 are composited and analyzed, and the respective wave activity fluxes are calculated.

    As shown in Fig. 10a, a teleconnecting wave train can be seen at 200 hPa over Eurasia at high latitudes 9 days before the precipitation, but the wave train is not clear at middle latitudes. In addition, the negative geopotential height anomaly centers over the northeastern Atlantic Ocean and West Asia are connected, and the wave activity flux from the northeastern Atlantic Ocean does not propagate significantly to the high-latitude wave train; rather, it mainly propagates to the entrance of the midlatitude westerly jet. On the 8th day before the precipitation (Fig. 10b), the negative geopotential height anomaly at the location of the subtropical westerly jet over the Atlantic Ocean (Hong et al., 2018) strengthens. Additionally, the wave activity flux propagates to the eastern Atlantic, causing the positive anomaly center to enhance. Simultaneously, the wave activity flux propagating to the northeastern Atlantic increases, and the negative geopotential height anomaly centers over the northeastern Atlantic and West Asia become separated and move southward. On the 6th day before the precipitation (Fig. 10c), the negative geopotential height anomaly center over the northeastern Atlantic moves southward to 40°N near eastern Europe, and the wave activity flux propagating to West Asia increases. In the following 11 days, the maximum amplitude of the geopotential height anomaly and the wave activity flux gradually propagate eastward along a line near 40°N. On the 4th day before the precipitation (Fig. 10d), the positive geopotential height anomaly center in West Asia strengthens, the wave activity flux propagating to the east also strengthens, and the negative geopotential height anomaly center over central Asia is generated. On the second day before the precipitation (Fig. 10e), the negative geopotential height anomaly center over central Asia is significantly enhanced, and the wave activity flux propagating to the positive geopotential height anomaly center over Mongolia strengthens significantly. On the day of the precipitation (Fig. 10f), the negative geopotential height anomaly center over central Asia moves to Northwest China, the Mongolian positive geopotential height anomaly center develops to its strongest level, and the wave activity flux propagating to the east reaches a maximum. In this situation, the negative geopotential height anomaly center over Japan is generated. Moreover, a similar SRP structure of the six activity centers in Eurasia is quite clear. Thus, it can be inferred that the southeasterly wind from the north side of the Japanese anomalous cyclone transports the water vapor from the Pacific Ocean to Northeast China, and then turns into a northeasterly wind and moves southward along the east side of the anomalous anticyclone over Mongolia to the northern edge of the Tibetan Plateau. Subsequently, this airflow turns into an easterly or southeasterly wind and transports water vapor to Northwest China. This finding is consistent with the channel of water vapor flux anomalies during the rainy period (Fig. 6). On the second day after the precipitation (Fig. 10g), the negative geopotential height anomaly center over Northwest China weakens, while the negative geopotential height anomaly center over Japan strengthens. On the 4th day after the precipitation (Fig. 10h), all the geopotential height anomaly centers are further weakened, and the SRP-like structure begins to vanish. Similar characteristics of the geopotential height anomalies and evolution of the wave activity flux field are found at 500 hPa (figure omitted).

    Fig  10.  Composite analysis of geopotential height anomalies (contours; gpm) and wave activity fluxes (arrows; m2 s−2) at 200 hPa on the (a) 9th, (b) 8th, (c) 6th, (d) 4th, and (e) 2nd days before the precipitation; (f) on the day of the precipitation; and on the (g) 2nd and (h) 4th days after the precipitation in terms of 58 heavy precipitation events in Northwest China during 1961–2018. Values exceeding the 95% confidence level are shaded.

    Samples of heavy precipitation events are added for analysis, and the domain-averaged daily precipitation in the summers of 1961–2018 in Northwest China is arranged from smallest to largest. A total of 522 heavy precipitation days based on the 90th percentile are selected to analyze the 200-hPa geopotential height anomalies on the heavy precipitation days, and the results reveal the existence of a similar SRP anomaly in Eurasia (figure omitted), like the result in Fig. 10f. Therefore, heavy precipitation events in Northwest China are related to the SRP, and, in view of the interdecadal increase in heavy precipitation events, the SRP also has an important influence on the interdecadal increase in precipitation. Indeed, recent research found that, for 36 regional extreme precipitation events in central Asia, a similar eastward-propagating wave train existed at 250 hPa from 10 days before the precipitation event to 4 days after the precipitation event (Xu et al., 2022).

    The above analysis indicates that disturbance of the geopotential height over the Atlantic Ocean can trigger an SRP anomaly over Eurasia. Furthermore, regression analysis of the summer AMOI on the 200-hPa wind field shows that a northerly–southerly–northerly turning of the meridional wind direction occurs every 30° of longitude at middle latitudes (30°–130°E), with cyclonic circulation observed over both central Asia and East China to Japan, and anticyclonic circulation over Mongolia (Fig. 11a), similar to the structure of the SRP (Fig. 9a). Moreover, regression of the spring AMOI on the 200-hPa wind field in summer yields similar results (Fig. 11b), indicating that the AMO is related to the summer SRP.

    Fig  11.  Regression of (a) summer AMOI and (b) spring AMOI onto the 200-hPa summer wind field (arrows; m s−1), meridional wind field (shadings; m s−1), and vorticity field (contours; 10–6 s−1). Values exceeding the 95% confidence level are stippled, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E). Panel (c) plots the AMOI curve, and (d) presents the normalization curves of the SRPI in summer during 1961–2018.

    Interdecadal correlation analysis between the AMO and SRP reveals the average AMOI increases from −0.16 in P1 to 0.13 in P2 and the average SRPI increases from −0.38 in P1 to 0.31 in P2, both passing the test for significance difference at the 0.05 level (Figs. 11c, d). These results indicate significant interdecadal variation of AMOI and SRPI. The interdecadal correlation coefficients between the spring/summer AMOI and summer SRPI during 1961–2018 are 0.84/0.82 (Table 1). Interestingly, previous research found that the interdecadal correlation coefficients between the AMOI and SRPI in the summers of 1920–2010 were not significant (Wang et al., 2017). According to the analysis of the variation curve of the interdecadal SRPI in their study, it is possible that the interdecadal cycle of the SRPI was shorter during 1920–1960, while the interdecadal cycle of the AMOI was relatively longer. The overall interdecadal correlation between the two was weak. After 1961, the interdecadal cycle of the SRPI lengthens, hence the obvious interdecadal correlation between the AMOI and SRPI from 1961 to 2018.

    Table  1.  Interdecadal correlation coefficients between spring/summer AMOI and the precipitation, SRPI, and atmospheric circulation factors (EWDs, EWVTF, SEWDs, and SEWVTF) determined from the vertically integrated vapor flux (values not in brackets: surface to 700 hPa; values in brackets: surface to 300 hPa) during 1961–2018. All values are statistically significant at the 0.01 level
    SRPIEWDEWVTFSEWDSEWVTFPrecipitation
    Spring AMOI0.840.80 (0.79)0.85 (0.83)0.73 (0.74)0.80 (0.83)0.64
    Summer AMOI0.820.91 (0.87)0.94 (0.87)0.87 (0.85)0.92 (0.91)0.79
     | Show Table
    DownLoad: CSV

    To judge the interdecadal effects of the AMO on easterly water vapor transport and precipitation, the interdecadal correlation coefficients between the spring/summer AMOI and EWDs, EWVTF, SEWDs, SEWVTF, and precipitation from 1961 to 2018 are calculated, and the results show significant correlations (Table 1). This indicates that the AMO has an important influence on the decadal variation of easterly water vapor transport and precipitation.

    The above analysis demonstrates that the interdecadal transition of the AMO to a positive phase is conducive to the occurrence of the positive phase of the SRP anomaly. The anomalous cyclone over central Asia enhances the ascending motion in Northwest China, and the anomalous anticyclone over Mongolia strengthens the easterly winds along the northern margin of the Tibetan Plateau. The water vapor transported from the Pacific, Indian Ocean, and South China Sea to Northwest China via the easterly winds increases, resulting in excessive interdecadal precipitation in Northwest China. The Atlantic Ocean may be the energy source of the interdecadal evolution of the SRP and atmospheric circulation factors (Chen et al., 2019), while the Pacific Ocean is the key water vapor source behind the interdecadal increase in precipitation in Northwest China. This finding is similar to other recent reports that the AMO and Pacific decadal oscillation synergistically affect the atmospheric heat source in the Tibetan Plateau and East Asian monsoon region, thereby influencing the atmospheric circulation of the Northern Hemisphere during summer (Sun et al., 2021) and significantly affecting the summer precipitation in Northeast Asia (Si et al., 2021).

    Daily precipitation data from 149 rain gauge stations in Northwest China and NCEP–NCAR reanalysis data from 1961 to 2018 are used in this study to analyze the interdecadal variability of summer precipitation in Northwest China and the underlying mechanisms involved. The main conclusions can be summarized as follows.

    After shifting from a less rainy to a rainy period in 1987, the precipitation recorded at most rain gauge stations in Northwest China increases. The precipitation increase in Xinjiang and west Qinghai is the result of the combined effect of an increase in the number of precipitation days and an enhanced precipitation intensity, while in east Qinghai and the Hexi Corridor of Gansu, it is mainly due to a significant enhancement in precipitation intensity. The probability of light rain decreases, but the probabilities of moderate rain, heavy rain, rainstorms, large rainstorms, and extraordinary rainstorms all increase, with more than 60% of the increase in precipitation attributable to rainfall above the grade of heavy rain.

    During summer in the rainy period, the atmospheric circulation anomalies in the middle and upper troposphere show a Rossby wave train pattern similar to the positive phase of the SRP, in which the cyclonic anomaly over central Asia is favorable for the generation of precipitation weather systems and eastward movement that affects Northwest China, and the vertical ascending motion and stratification instability in Northwest China are obviously enhanced.

    The transport of water vapor by the easterly winds is the main source of the increase in precipitation. On the south side of the Mongolian anticyclone, along the northern margin of the Tibetan Plateau, water vapor over the Pacific, Indian Ocean, and South China Sea is transported to Northwest China by easterly winds, which is significantly correlated with the interdecadal variation of precipitation. The Mongolian anticyclone affects the transport of water vapor via the easterly winds and may weaken the East Asian summer monsoon. The Mongolian anticyclone anomaly of the positive-phase SRP may play an important role in this interdecadal enhancement of the Mongolian anticyclone.

    The interdecadal increase in precipitation is closely related to the SRP anomaly. Heavy precipitation events in summer in Northwest China are obviously influenced by the SRP anomaly. Interdecadal correlation analysis shows that the AMO affects the evolution of the SRP anomaly and atmospheric circulation factors, and the interdecadal transition of the AMO to a positive phase may be the main cause of the interdecadal transition of the SRP to a positive phase and the interdecadal increase in precipitation in Northwest China.

    To estimate the contribution rate of the AMO to the interdecadal increase in precipitation in Northwest China, the contribution rate of the AMOI variation to the precipitation increase during the rainy period (CAMO) is calculated as follows: CAMO = IAMO × PAMO/Pobs × 100% (Wang et al., 2017), where IAMO (1.38) denotes the standardized AMOI variation in the rainy period, PAMO is the increase in precipitation in Northwest China (5.06 mm) based on linear regression when the AMOI changes by one standard deviation, and Pobs indicates the actual precipitation increase of 13.2 mm in P2 compared with that in P1 (Section 3.3). On this basis, the contribution rate of the AMOI to the precipitation increase in P2 is 53%. Therefore, the AMO is the dominant factor affecting the precipitation increase in Northwest China during summer. Since the AMOI variation has a cycle of 65–80 yr (Hong et al., 2017) and the AMOI has been in a positive phase since the 1990s (Fig. 11c), the AMOI may enter a negative phase around 2030, i.e., the summer precipitation in Northwest China could enter a less rainy interdecadal period after that year. It is necessary to further study the physical mechanism underlying the AMO’s effect on the SRP and to use climate models to predict the AMO variation and its impact on precipitation in Northwest China against the background of global warming.

    The SRP’s wavelength is variable. Enomoto (2004) pointed out that the SRP is a wave train in which the meridional wind direction turns from northerly to southerly in intervals of 2000–3000 km in the upper troposphere at middle latitudes. Since the interval distance of the meridional wind’s turning is not fixed, the SRP wavelength varies. The summer meridional wind anomaly field at 200 hPa during the rainy period shows that a northerly–southerly turning of the wind direction appears every 30° of longitude in Eurasia (Figs. 5b, 9b), which is similar to the structure of the SRP defined by Kosaka et al. (2009) and Wang et al. (2017), i.e., a northerly–southerly turning of the wind direction every 30° of longitudes (Fig. 9a) in which there are five meridional wind anomaly centers in the range of 0°–150°E over Eurasia. Recent research reports interdecadal variations in the SRP after the 1990s, with six meridional wind anomaly centers occurring in the range of 0°–150°E over Eurasia during summer (Liu et al., 2020). For the daily 200-hPa geopotential height anomalies before and after heavy precipitation events in Northwest China, the SRP has six abnormal geopotential height disturbance centers in the range of 0°–150°E over Eurasia (Fig. 10). This finding is basically consistent with the SRP generated by the 200-hPa daily geopotential height disturbance in August 1985 (Enomoto, 2004). The regression of the SRPI onto the 200-hPa vorticity field also shows six anomaly centers (Kosaka et al., 2009). The above analysis verifies that the wavelength of the SRP is variable.

  • Fig.  1.   Study area and locations of the rain gauge stations (red circles). Shading indicates the altitude (m).

    Fig.  2.   Time series of the anomalies of the (a) precipitation amount (%), (b) number of precipitation days, and (c) precipitation intensity (mm day−1) averaged regionally in Northwest China during 1961–2018. The straight lines indicate the average of P1 (1961–1986) and P2 (1987–2018).

    Fig.  3.   Interdecadal variations of the spatial distributions of the (a) amount, (b) number of days, and (c) intensity of summer precipitation at rain gauge stations in Northwest China.

    Fig.  4.   (a) Probability density curve and (b) cumulative probability density curve of different intensities of summer precipitation, and (c) variations of summer precipitation at different intensity grades per station in Northwest China.

    Fig.  5.   Difference in geopotential height (contours; gpm) and wind (arrows; m s−1) between P2 and P1 at (a) 500 hPa and (b) 200 hPa, along with altitude–longitude profiles of the difference in (c) vertical velocity (10−2 Pa s−1) and (d) air temperature (°C) averaged in the range of 32.5°–50°N between P2 and P1 in summer. Values exceeding the 95% confidence level are shaded, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    Fig.  6.   Difference in vertically integrated water vapor flux (arrows; kg m−1 s−1) and water vapor flux divergence (shadings; 10−6 kg m−2 s−1) between P2 and P1 in summer. The green dotted area and the arrows exceed the 95% confidence level, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    Fig.  7.   Variations of atmospheric circulation factors during 1961–2018: (a) average integrated water vapor flux divergence from the surface to 300 hPa, (b) average vertical velocity at 700 hPa, (c) easterly water vapor transport days (EWDs; black) and strong EWDs (SEWDs; red) from the surface to 700 hPa, (d) easterly water vapor transport flux (EWVTF; black) and strong EWVTF (SEWVTF; red) from the surface to 700 hPa, (e) EWDs (black) and SEWDs (red) from the surface to 300 hPa, and (f) EWVTF (black) and SEWVTF (red) from the surface to 300 hPa.

    Fig.  8.   Composite analysis of daily integrated water vapor flux (arrows; kg m−1 s−1) and divergence of water vapor flux (shadings; 10−5 kg m−2 s−1) from the surface to 700 hPa on (a) EWDs and (b) SEWDs, and from the surface to 300 hPa on (c) EWDs and (d) SEWDs, in summer 1961–2018. The green dotted area and the arrows exceed the 95% confidence level, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    Fig.  9.   (a) First mode of the EOF decomposition of the 200-hPa meridional wind field in the range of 20°–60°N, 30°–130°E during summer from 1961 to 2018, and (b) the difference fields of the meridional wind (shadings; m s−1) and vorticity (contours; 10−6 s−1) at 200 hPa between P2 and P1 in summer, in which values exceeding the 95% confidence level are stippled. The box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E).

    Fig.  10.   Composite analysis of geopotential height anomalies (contours; gpm) and wave activity fluxes (arrows; m2 s−2) at 200 hPa on the (a) 9th, (b) 8th, (c) 6th, (d) 4th, and (e) 2nd days before the precipitation; (f) on the day of the precipitation; and on the (g) 2nd and (h) 4th days after the precipitation in terms of 58 heavy precipitation events in Northwest China during 1961–2018. Values exceeding the 95% confidence level are shaded.

    Fig.  11.   Regression of (a) summer AMOI and (b) spring AMOI onto the 200-hPa summer wind field (arrows; m s−1), meridional wind field (shadings; m s−1), and vorticity field (contours; 10–6 s−1). Values exceeding the 95% confidence level are stippled, and the box indicates the range of Northwest China (32.5°–50°N, 72.5°–103°E). Panel (c) plots the AMOI curve, and (d) presents the normalization curves of the SRPI in summer during 1961–2018.

    Table  1   Interdecadal correlation coefficients between spring/summer AMOI and the precipitation, SRPI, and atmospheric circulation factors (EWDs, EWVTF, SEWDs, and SEWVTF) determined from the vertically integrated vapor flux (values not in brackets: surface to 700 hPa; values in brackets: surface to 300 hPa) during 1961–2018. All values are statistically significant at the 0.01 level

    SRPIEWDEWVTFSEWDSEWVTFPrecipitation
    Spring AMOI0.840.80 (0.79)0.85 (0.83)0.73 (0.74)0.80 (0.83)0.64
    Summer AMOI0.820.91 (0.87)0.94 (0.87)0.87 (0.85)0.92 (0.91)0.79
    Download: Download as CSV
  • Chen, C. Z., X. J. Zhang, H. Y. Lu, et al., 2021: Increasing summer precipitation in arid Central Asia linked to the weakening of the East Asian summer monsoon in the recent decades. Int. J. Climatol., 41, 1024–1038. doi: 10.1002/joc.6727
    Chen, D. D., and Y. J. Dai, 2009: Characteristics of Northwest China rainfall intensity in recent 50 years. Chinese J. Atmos. Sci., 33, 923–935. (in Chinese) doi: 10.3878/j.issn.1006-9895.2009.05.04
    Chen, G. S., and R. H. Huang, 2012: Excitation mechanisms of the teleconnection patterns affecting the July precipitation in Northwest China. J. Climate, 25, 7834–7851. doi: 10.1175/JCLI-D-11-00684.1
    Chen, H.-P., J.-Q. Sun, and K. Fan, 2012: Possible mechanism for the interdecadal change of Xinjiang summer precipitation. Chinese J. Geophys., 55, 1844–1851. (in Chinese)
    Chen, W., L. Wang, J. Feng, et al., 2019: Recent progress in studies of the variabilities and mechanisms of the East Asian monsoon in a changing climate. Adv. Atmos. Sci., 36, 887–901. doi: 10.1007/s00376-019-8230-y
    Chowdary, J. S., K. M. Hu, G. Srinivas, et al., 2019: The Eurasian jet streams as conduits for East Asian monsoon variability. Curr. Climate Change Rep., 5, 233–244. doi: 10.1007/s40641-019-00134-x
    Dai, X.-G., and K.-J. Zhang, 2012: A mechanism analysis of the interdecadal precipitation change in Northwestern China for the last 30 years of twentieth century. Acta Phys. Sinica, 61, 199201. (in Chinese) doi: 10.7498/aps.61.199201
    Enomoto, T., 2004: Interannual variability of the Bonin high associated with the propagation of Rossby waves along the Asian jet. J. Meteor. Soc. Japan, 82, 1019–1034. doi: 10.2151/jmsj.2004.1019
    Geng, Q. L., P. T. Wu, and X. N. Zhao, 2016: Spatial and temporal trends in climatic variables in arid areas of Northwest China. Int. J. Climatol., 36, 4118–4129. doi: 10.1002/joc.4621
    Hong, X. W., R. Y. Lu, and S. L. Li, 2017: Amplified summer warming in Europe–West Asia and Northeast Asia after the mid-1990s. Environ. Res. Lett., 12, 094007. doi: 10.1088/1748-9326/aa7909
    Hong, X.-W., S.-H. Xue, R.-Y. Lu, et al., 2018: Comparison between the interannual and decadal components of the Silk Road pattern. Atmos. Ocean. Sci. Lett., 11, 270–274. doi: 10.1080/16742834.2018.1439661
    Huang, W., J. H. Chen, X. J. Zhang, et al., 2015a: Definition of the core zone of the “westerlies-dominated climatic regime”, and its controlling factors during the instrumental period. Sci. China Earth Sci., 58, 676–684. doi: 10.1007/s11430-015-5057-y
    Huang, W., S. Feng, J. H. Chen, et al., 2015b: Physical mechanisms of summer precipitation variations in the Tarim basin in northwestern China. J. Climate, 28, 3579–3591. doi: 10.1175/JCLI-D-14-00395.1
    Kalnay, E., M. Kanamitsu, R. Kistler, et al., 1996: The NCEP/NCAR 40-year reanalysis project. Bull. Amer. Meteor. Soc., 77, 437–472. doi: 10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2
    Kosaka, Y., H. Nakamura, M. Watanabe, et al., 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, 561–580. doi: 10.2151/jmsj.87.561
    Krishnan, R., and M. Sugi, 2001: Baiu rainfall variability and associated monsoon teleconnections. J. Meteor. Soc. Japan, 79, 851–860. doi: 10.2151/jmsj.79.851
    Li, B. F., Y. N. Chen, Z. S. Chen, et al., 2016: Why does precipitation in northwest China show a significant increasing trend from 1960 to 2010? Atmos. Res., 167, 275–284. doi: 10.1016/j.atmosres.2015.08.017
    Li, Q. H., Y. N. Chen, Y. J. Shen, et al., 2011: Spatial and temporal trends of climate change in Xinjiang, China. J. Geogr. Sci., 21, 1007–1018. doi: 10.1007/s11442-011-0896-8
    Li, Z., and H. H. Liu, 2020: Temporal and spatial variations of precipitation change from Southeast to Northwest China during the period 1961–2017. Water, 12, 2622. doi: 10.3390/w12092622
    Liu, W. C., Q. Zhang, and Z. Fu, 2017: Variation characteristics of precipitation and its affecting factors in Northwest China over the past 55 years. Plateau Meteor., 36, 1533–1545. (in Chinese) doi: 10.7522/j.issn.1000-0534.2017.00081
    Liu, Y., W. Zhou, X. Qu, et al., 2020: An interdecadal change of the boreal summer Silk Road pattern around the late 1990s. J. Climate, 33, 7083–7100. doi: 10.1175/JCLI-D-19-0795.1
    Lu, S., Z. Y. Hu, H. P. Yu, et al., 2021: Changes of extreme precipitation and its associated mechanisms in Northwest China. Adv. Atmos. Sci., 38, 1665–1681. doi: 10.1007/s00376-021-0409-3
    North, G. R., T. L. Bell, R. F. Cahalan, et al., 1982: Sampling errors in the estimation of empirical orthogonal functions. Mon. Wea. Rev., 110, 699–706. doi: 10.1175/1520-0493(1982)110<0699:SEITEO>2.0.CO;2
    Papalexiou, S. M., and A. Montanari, 2019: Global and regional increase of precipitation extremes under global warming. Water Resour. Res., 55, 4901–4914. doi: 10.1029/2018WR024067
    Ren, G.-Y., Y.-J. Yuan, Y.-J. Liu, et al., 2016: Changes in precipitation over Northwest China. Arid Zone Res., 33, 1–19. (in Chinese) doi: 10.13866/j.azr.2016.01.01
    Shi, Y.-F., Y.-P. Shen, and R.-J. Hu, 2002: Preliminary study on signal, impact and foreground of climatic shift from warm-dry to warm-humid in Northwest China. J. Glaciol. Geocryol., 24, 219–226. (in Chinese) doi: 10.3969/j.issn.1000-0240.2002.03.001
    Shi, Y. F., Y. P. Shen, D. L. Li, et al., 2003: Discussion on the present climate change from warm-dry to warm-wet in Northwest China. Quat. Sci., 23, 152–164. (in Chinese) doi: 10.3321/j.issn:1001-7410.2003.02.005
    Si, D., D. B. Jiang, and Y. H. Ding, 2021: Synergistic impacts of the Atlantic and Pacific Oceans on interdecadal variations of summer rainfall in Northeast Asia. J. Meteor. Res., 35, 844–856. doi: 10.1007/s13351-021-0191-2
    Su, Y. F., C. F. Zhao, Y. Wang, et al., 2020: Spatiotemporal variations of precipitation in China using surface gauge observations from 1961 to 2016. Atmosphere, 11, 303. doi: 10.3390/atmos11030303
    Sun, X. Q., S. L. Li, X. W. Hong, et al., 2019: Simulated influence of the Atlantic Multidecadal Oscillation on summer Eurasian nonuniform warming since the mid-1990s. Adv. Atmos. Sci., 36, 811–822. doi: 10.1007/s00376-019-8169-z
    Sun, X. T., Y. H. Ding, and Q. Q. Li, 2021: Interdecadal variation of the atmospheric heat source over the Tibetan Plateau and surrounding Asian monsoon region: Impact on the Northern Hemisphere summer circulation. J. Meteor. Res., 35, 238–257. doi: 10.1007/s13351-021-0101-7
    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, 608–627. doi: 10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2
    Wang, C. H., S. N. Zhang, K. C. Li, et al., 2021: Change characteristics of precipitation in Northwest China from 1961 to 2018. Chinese J. Atmos. Sci., 45, 713–724. (in Chinese) doi: 10.3878/j.issn.1006-9895.2101.20216
    Wang, L., P. Q. Xu, W. Chen, et al., 2017: Interdecadal variations of the Silk Road pattern. J. Climate, 30, 9915–9932. doi: 10.1175/JCLI-D-17-0340.1
    Wang, L., P. Q. Xu, and J. S. Chowdary, 2021: Chapter 14—Teleconnection along the Asian jet stream and its association with the Asian summer monsoon. Indian Summer Monsoon Variability: El Niño-Teleconnections and Beyond, Elsevier, Amsterdam, 287–298, doi: 10.1016/B978-0-12-822402-1.00009-0.
    Wang, Q., P.-M. Zhai, and D.-H. Qin, 2020: New perspectives on ‘warming–wetting’ trend in Xinjiang, China. Adv. Climate Change Res., 11, 252–260. doi: 10.1016/j.accre.2020.09.004
    Wu, P., Y. H. Ding, Y. J. Liu, et al., 2019: The characteristics of moisture recycling and its impact on regional precipitation against the background of climate warming over Northwest China. Int. J. Climatol., 39, 5241–5255. doi: 10.1002/joc.6136
    Xu, P. Q., L. Wang, and J. Ming, 2022: Central Asian precipitation extremes affected by an intraseasonal planetary wave pattern. J. Climate, 35, 2603–2616. doi: 10.1175/JCLI-D-21-0657.1
    Yang, J. P., Y. J. Ding, R. S. Chen, et al., 2003: The fluctuation of dry and wet climate boundary and its causal analyses in China. Acta Meteor. Sinica, 61, 364–373. (in Chinese) doi: 10.3321/j.issn:0577-6619.2003.03.011
    Yang, S., and Q. X. Li, 2014: Improvement in homogeneity analysis method and update of China precipitation data. Progressus Inquisitiones De Mutatione Climatis, 10, 276–281. (in Chinese) doi: 10.3969/j.issn.1673-1719.2014.04.008
    Zhang, Q., J. J. Lin, W. C. Liu, et al., 2019: Precipitation seesaw phenomenon and its formation mechanism in the eastern and western parts of Northwest China during the flood season. Sci. China Earth Sci., 62, 2083–2098. doi: 10.1007/s11430-018-9357-y
    Zhou, L.-T., and R.-H. Huang, 2010: Interdecadal variability of summer rainfall in Northwest China and its possible causes. Int. J. Climatol., 30, 549–557. doi: 10.1002/joc.1923
    Zhu, X.-C., Y.-Y. Guo, H.-Y. Zhang, et al., 2018: A southward withdrawal of the northern edge of the East Asian summer monsoon around the early 1990s. Atmos. Ocean. Sci. Lett., 11, 136–142. doi: 10.1080/16742834.2018.1410058
  • Other Related Supplements

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return